5.6 MiB
5.6 MiB
,name,file,line,type,comment
0,UserInputError,tensorflow/configure.py,74,class,
1,is_windows,tensorflow/configure.py,78,function,
2,is_linux,tensorflow/configure.py,82,function,
3,is_macos,tensorflow/configure.py,86,function,
4,is_ppc64le,tensorflow/configure.py,90,function,
5,is_cygwin,tensorflow/configure.py,94,function,
6,get_input,tensorflow/configure.py,98,function,
7,symlink_force,tensorflow/configure.py,109,function,"Force symlink, equivalent of 'ln -sf'.
Args:
target: items to link to.
link_name: name of the link."
8,sed_in_place,tensorflow/configure.py,126,function,"Replace old string with new string in file.
Args:
filename: string for filename.
old: string to replace.
new: new string to replace to."
9,write_to_bazelrc,tensorflow/configure.py,141,function,
10,write_action_env_to_bazelrc,tensorflow/configure.py,146,function,
11,run_shell,tensorflow/configure.py,150,function,
12,cygpath,tensorflow/configure.py,163,function,Convert path from posix to windows.
13,get_python_path,tensorflow/configure.py,168,function,Get the python site package paths.
14,get_python_major_version,tensorflow/configure.py,198,function,Get the python major version.
15,setup_python,tensorflow/configure.py,203,function,Setup python related env variables.
16,reset_tf_configure_bazelrc,tensorflow/configure.py,273,function,Reset file that contains customized config settings.
17,cleanup_makefile,tensorflow/configure.py,278,function,"Delete any leftover BUILD files from the Makefile build.
These files could interfere with Bazel parsing."
18,get_var,tensorflow/configure.py,292,function,"Get boolean input from user.
If var_name is not set in env, ask user to enable query_item or not. If the
response is empty, use the default.
Args:
environ_cp: copy of the os.environ.
var_name: string for name of environment variable, e.g. ""TF_NEED_CUDA"".
query_item: string for feature related to the variable, e.g. ""CUDA for
Nvidia GPUs"".
enabled_by_default: boolean for default behavior.
question: optional string for how to ask for user input.
yes_reply: optional string for reply when feature is enabled.
no_reply: optional string for reply when feature is disabled.
Returns:
boolean value of the variable.
Raises:
UserInputError: if an environment variable is set, but it cannot be
interpreted as a boolean indicator, assume that the user has made a
scripting error, and will continue to provide invalid input.
Raise the error to avoid infinitely looping."
19,set_build_var,tensorflow/configure.py,377,function,"Set if query_item will be enabled for the build.
Ask user if query_item will be enabled. Default is used if no input is given.
Set subprocess environment variable and write to .bazelrc if enabled.
Args:
environ_cp: copy of the os.environ.
var_name: string for name of environment variable, e.g. ""TF_NEED_CUDA"".
query_item: string for feature related to the variable, e.g. ""CUDA for
Nvidia GPUs"".
option_name: string for option to define in .bazelrc.
enabled_by_default: boolean for default behavior.
bazel_config_name: Name for Bazel --config argument to enable build feature."
20,set_action_env_var,tensorflow/configure.py,411,function,"Set boolean action_env variable.
Ask user if query_item will be enabled. Default is used if no input is given.
Set environment variable and write to .bazelrc.
Args:
environ_cp: copy of the os.environ.
var_name: string for name of environment variable, e.g. ""TF_NEED_CUDA"".
query_item: string for feature related to the variable, e.g. ""CUDA for
Nvidia GPUs"".
enabled_by_default: boolean for default behavior.
question: optional string for how to ask for user input.
yes_reply: optional string for reply when feature is enabled.
no_reply: optional string for reply when feature is disabled.
bazel_config_name: adding config to .bazelrc instead of action_env."
21,convert_version_to_int,tensorflow/configure.py,446,function,"Convert a version number to a integer that can be used to compare.
Version strings of the form X.YZ and X.Y.Z-xxxxx are supported. The
'xxxxx' part, for instance 'homebrew' on OS/X, is ignored.
Args:
version: a version to be converted
Returns:
An integer if converted successfully, otherwise return None."
22,check_bazel_version,tensorflow/configure.py,471,function,"Check installed bazel version is between min_version and max_version.
Args:
min_version: string for minimum bazel version (must exist!).
max_version: string for maximum bazel version (must exist!).
Returns:
The bazel version detected."
23,set_cc_opt_flags,tensorflow/configure.py,518,function,"Set up architecture-dependent optimization flags.
Also append CC optimization flags to bazel.rc..
Args:
environ_cp: copy of the os.environ."
24,set_tf_cuda_clang,tensorflow/configure.py,546,function,"set TF_CUDA_CLANG action_env.
Args:
environ_cp: copy of the os.environ."
25,set_tf_download_clang,tensorflow/configure.py,566,function,Set TF_DOWNLOAD_CLANG action_env.
26,get_from_env_or_user_or_default,tensorflow/configure.py,582,function,"Get var_name either from env, or user or default.
If var_name has been set as environment variable, use the preset value, else
ask for user input. If no input is provided, the default is used.
Args:
environ_cp: copy of the os.environ.
var_name: string for name of environment variable, e.g. ""TF_NEED_CUDA"".
ask_for_var: string for how to ask for user input.
var_default: default value string.
Returns:
string value for var_name"
27,set_clang_cuda_compiler_path,tensorflow/configure.py,607,function,Set CLANG_CUDA_COMPILER_PATH.
28,prompt_loop_or_load_from_env,tensorflow/configure.py,630,function,"Loop over user prompts for an ENV param until receiving a valid response.
For the env param var_name, read from the environment or verify user input
until receiving valid input. When done, set var_name in the environ_cp to its
new value.
Args:
environ_cp: (Dict) copy of the os.environ.
var_name: (String) string for name of environment variable, e.g. ""TF_MYVAR"".
var_default: (String) default value string.
ask_for_var: (String) string for how to ask for user input.
check_success: (Function) function that takes one argument and returns a
boolean. Should return True if the value provided is considered valid. May
contain a complex error message if error_msg does not provide enough
information. In that case, set suppress_default_error to True.
error_msg: (String) String with one and only one '%s'. Formatted with each
invalid response upon check_success(input) failure.
suppress_default_error: (Bool) Suppress the above error message in favor of
one from the check_success function.
resolve_symlinks: (Bool) Translate symbolic links into the real filepath.
n_ask_attempts: (Integer) Number of times to query for valid input before
raising an error and quitting.
Returns:
[String] The value of var_name after querying for input.
Raises:
UserInputError: if a query has been attempted n_ask_attempts times without
success, assume that the user has made a scripting error, and will
continue to provide invalid input. Raise the error to avoid infinitely
looping."
29,create_android_ndk_rule,tensorflow/configure.py,696,function,Set ANDROID_NDK_HOME and write Android NDK WORKSPACE rule.
30,create_android_sdk_rule,tensorflow/configure.py,724,function,Set Android variables and write Android SDK WORKSPACE rule.
31,get_ndk_api_level,tensorflow/configure.py,788,function,Gets the appropriate NDK API level to use for the provided Android NDK path.
32,set_gcc_host_compiler_path,tensorflow/configure.py,836,function,Set GCC_HOST_COMPILER_PATH.
33,reformat_version_sequence,tensorflow/configure.py,858,function,"Reformat the version string to have the given number of sequences.
For example:
Given (7, 2) -> 7.0
(7.0.1, 2) -> 7.0
(5, 1) -> 5
(5.0.3.2, 1) -> 5
Args:
version_str: String, the version string.
sequence_count: int, an integer.
Returns:
string, reformatted version string."
34,set_tf_cuda_paths,tensorflow/configure.py,881,function,Set TF_CUDA_PATHS.
35,set_tf_cuda_version,tensorflow/configure.py,892,function,Set TF_CUDA_VERSION.
36,set_tf_cudnn_version,tensorflow/configure.py,904,function,Set TF_CUDNN_VERSION.
37,is_cuda_compatible,tensorflow/configure.py,916,function,Check compatibility between given library and cudnn/cudart libraries.
38,set_tf_tensorrt_version,tensorflow/configure.py,945,function,Set TF_TENSORRT_VERSION.
39,set_tf_nccl_version,tensorflow/configure.py,962,function,Set TF_NCCL_VERSION.
40,get_native_cuda_compute_capabilities,tensorflow/configure.py,979,function,"Get native cuda compute capabilities.
Args:
environ_cp: copy of the os.environ.
Returns:
string of native cuda compute capabilities, separated by comma."
41,set_tf_cuda_compute_capabilities,tensorflow/configure.py,1003,function,Set TF_CUDA_COMPUTE_CAPABILITIES.
42,set_other_cuda_vars,tensorflow/configure.py,1074,function,Set other CUDA related variables.
43,set_host_cxx_compiler,tensorflow/configure.py,1083,function,Set HOST_CXX_COMPILER.
44,set_host_c_compiler,tensorflow/configure.py,1100,function,Set HOST_C_COMPILER.
45,set_computecpp_toolkit_path,tensorflow/configure.py,1117,function,Set COMPUTECPP_TOOLKIT_PATH.
46,set_trisycl_include_dir,tensorflow/configure.py,1149,function,Set TRISYCL_INCLUDE_DIR.
47,set_system_libs_flag,tensorflow/configure.py,1216,function,
48,is_reduced_optimize_huge_functions_available,tensorflow/configure.py,1233,function,"Check to see if the system supports /d2ReducedOptimizeHugeFunctions.
The above compiler flag is a new compiler flag introduced to the Visual Studio
compiler in version 16.4 (available in Visual Studio 2019, Preview edition
only, as of 2019-11-19). TensorFlow needs this flag to massively reduce
compile times, but until 16.4 is officially released, we can't depend on it.
See also
https://groups.google.com/a/tensorflow.org/d/topic/build/SsW98Eo7l3o/discussion
Because it's very annoying to check this manually (to check the MSVC installed
versions, you need to use the registry, and it's not clear if Bazel will be
using that install version anyway), we expect enviroments who know they may
use this flag to export TF_VC_VERSION=16.4
TODO(angerson, gunan): Remove this function when TensorFlow's minimum VS
version is upgraded to 16.4.
Arguments:
environ_cp: Environment of the current execution
Returns:
boolean, whether or not /d2ReducedOptimizeHugeFunctions is available on this
machine."
49,set_windows_build_flags,tensorflow/configure.py,1262,function,Set Windows specific build options.
50,config_info_line,tensorflow/configure.py,1283,function,Helper function to print formatted help text for Bazel config options.
51,configure_ios,tensorflow/configure.py,1288,function,"Configures TensorFlow for iOS builds.
This function will only be executed if `is_macos()` is true."
52,validate_cuda_config,tensorflow/configure.py,1305,function,"Run find_cuda_config.py and return cuda_toolkit_path, or None."
53,VarsAndArithmeticObjectGraph,tensorflow/tensorflow/cc/saved_model/testdata/generate_saved_models.py,37,class,Three vars (one in a sub-module) and compute method.
54,compute,tensorflow/tensorflow/cc/saved_model/testdata/generate_saved_models.py,51,method,
55,ReferencesParent,tensorflow/tensorflow/cc/saved_model/testdata/generate_saved_models.py,55,class,
56,CyclicModule,tensorflow/tensorflow/cc/saved_model/testdata/generate_saved_models.py,64,class,
57,tfadd,tensorflow/tensorflow/compiler/aot/tests/make_test_graphs.py,48,function,
58,tfadd_with_ckpt,tensorflow/tensorflow/compiler/aot/tests/make_test_graphs.py,54,function,
59,tfadd_with_ckpt_saver,tensorflow/tensorflow/compiler/aot/tests/make_test_graphs.py,69,function,
60,tfassert_eq,tensorflow/tensorflow/compiler/aot/tests/make_test_graphs.py,88,function,
61,tfcond,tensorflow/tensorflow/compiler/aot/tests/make_test_graphs.py,96,function,
62,tfgather,tensorflow/tensorflow/compiler/aot/tests/make_test_graphs.py,104,function,
63,tfmatmul,tensorflow/tensorflow/compiler/aot/tests/make_test_graphs.py,110,function,
64,tfmatmulandadd,tensorflow/tensorflow/compiler/aot/tests/make_test_graphs.py,116,function,
65,tffunction,tensorflow/tensorflow/compiler/aot/tests/make_test_graphs.py,124,function,
66,tfsplits,tensorflow/tensorflow/compiler/aot/tests/make_test_graphs.py,135,function,"A more complex graph, including splits."
67,tftop_k,tensorflow/tensorflow/compiler/aot/tests/make_test_graphs.py,152,function,
68,tfvariable_readonly,tensorflow/tensorflow/compiler/aot/tests/make_test_graphs.py,158,function,
69,tfvariable,tensorflow/tensorflow/compiler/aot/tests/make_test_graphs.py,169,function,
70,tfvariable_sequential_updates,tensorflow/tensorflow/compiler/aot/tests/make_test_graphs.py,177,function,
71,export_debug_info,tensorflow/tensorflow/compiler/aot/tests/make_test_graphs.py,189,function,"Exports debug information from a graph.
Args:
exported_graph: A Graph that has been created by tracing a saveable view.
Returns:
Corresponding GraphDebugInfo with traces for all ops in exported_graph."
72,write_graph,tensorflow/tensorflow/compiler/aot/tests/make_test_graphs.py,204,function,Build a graph using build_graph and write it out.
73,set_tf_options,tensorflow/tensorflow/compiler/mlir/tensorflow/tests/tf_saved_model/common_v1.py,38,function,
74,ReferencesParent,tensorflow/tensorflow/compiler/mlir/tensorflow/tests/tf_saved_model/cyclic_object_graph.py,27,class,
75,Child,tensorflow/tensorflow/compiler/mlir/tensorflow/tests/tf_saved_model/dag_object_graph.py,27,class,
76,plus,tensorflow/tensorflow/compiler/mlir/tensorflow/tests/tf_saved_model/defun_export.py,29,function,
77,write_vocabulary_file,tensorflow/tensorflow/compiler/mlir/tensorflow/tests/tf_saved_model/hash_table_asset_v1.py,39,function,Write temporary vocab file for module construction.
78,mnist_model,tensorflow/tensorflow/compiler/mlir/tensorflow/tests/tf_saved_model/keras.py,27,function,Creates a MNIST model.
79,adam_update_numpy,tensorflow/tensorflow/compiler/tests/adam_test.py,34,function,
80,GetRunMetadataLabels,tensorflow/tensorflow/compiler/tests/dense_layer_test.py,36,function,Returns all labels in run_metadata.
81,InLabels,tensorflow/tensorflow/compiler/tests/dense_layer_test.py,45,function,Returns true iff one of the labels contains substr.
82,ReferenceDepthwiseConv2D,tensorflow/tensorflow/compiler/tests/depthwise_conv_op_test.py,35,function,
83,multiple_tpus,tensorflow/tensorflow/compiler/tests/eager_test.py,772,function,
84,ExtractImagePatches,tensorflow/tensorflow/compiler/tests/extract_image_patches_op_test.py,29,class,Functional tests for ExtractImagePatches op.
85,pick_10,tensorflow/tensorflow/compiler/tests/fft_test.py,38,function,
86,to_32bit,tensorflow/tensorflow/compiler/tests/fft_test.py,45,function,
87,GatherBenchmark,tensorflow/tensorflow/compiler/tests/gather_test.py,158,class,Microbenchmarks for the gather op.
88,benchmarkSliceGatherAxis0,tensorflow/tensorflow/compiler/tests/gather_test.py,183,method,
89,benchmarkSliceGatherAxis0XLA,tensorflow/tensorflow/compiler/tests/gather_test.py,186,method,
90,benchmarkSliceGatherAxis1,tensorflow/tensorflow/compiler/tests/gather_test.py,189,method,
91,benchmarkSliceGatherAxis1XLA,tensorflow/tensorflow/compiler/tests/gather_test.py,192,method,
92,benchmarkSliceGatherAxis4,tensorflow/tensorflow/compiler/tests/gather_test.py,195,method,
93,benchmarkSliceGatherAxis4XLA,tensorflow/tensorflow/compiler/tests/gather_test.py,198,method,
94,benchmarkNontrivialGatherAxis0,tensorflow/tensorflow/compiler/tests/gather_test.py,201,method,
95,benchmarkNontrivialGatherAxis0XLA,tensorflow/tensorflow/compiler/tests/gather_test.py,204,method,
96,benchmarkNontrivialGatherAxis1,tensorflow/tensorflow/compiler/tests/gather_test.py,207,method,
97,benchmarkNontrivialGatherAxis1XLA,tensorflow/tensorflow/compiler/tests/gather_test.py,210,method,
98,benchmarkNontrivialGatherAxis4,tensorflow/tensorflow/compiler/tests/gather_test.py,213,method,
99,benchmarkNontrivialGatherAxis4XLA,tensorflow/tensorflow/compiler/tests/gather_test.py,216,method,
100,BuilderFn,tensorflow/tensorflow/compiler/tests/gather_test.py,163,method,
101,NoRewriteSessionConfig,tensorflow/tensorflow/compiler/tests/jit_test.py,46,function,
102,CompiledKernel,tensorflow/tensorflow/compiler/tests/jit_test.py,56,function,"Execute 'fn' as a compiled XLA kernel, with 'inputs'."
103,RunMetadataLabels,tensorflow/tensorflow/compiler/tests/jit_test.py,68,function,Returns all labels in run_metadata.
104,InLabels,tensorflow/tensorflow/compiler/tests/jit_test.py,77,function,Returns true iff one of the labels contains substr.
105,MetadataHasXlaRunOp,tensorflow/tensorflow/compiler/tests/jit_test.py,82,function,Returns true if there are XlaRun kernels in run_metadata's timeline.
106,Clip,tensorflow/tensorflow/compiler/tests/lstm.py,38,function,"Clips x to the range [-1., 1.]."
107,LSTMCellWeightsShape,tensorflow/tensorflow/compiler/tests/lstm.py,43,function,Returns the shape of the weights for a single LSTM cell.
108,LSTMCell,tensorflow/tensorflow/compiler/tests/lstm.py,50,function,"Unrolls a single LSTM cell with clipped activations forward by one step.
Args:
weights: Weight matrix with shape LSTMCellWeightsShape.
m_prev: Previous m states with shape [batch_size, num_nodes].
c_prev: Previous c states with shape [batch_size, num_nodes].
x: Input with shape [batch_size, num_inputs].
pad: Padding with shape [batch_size, 1]. Each padding value is either
0 or 1, where 1 indicates padding; i.e. the input is shorter than the
sequence length, and the (m, c) states should simply be passed through
from the previous states.
Returns:
The next (m, c) states, each with shape [batch_size, num_nodes]."
109,LSTMLayer,tensorflow/tensorflow/compiler/tests/lstm.py,88,function,"Unrolls a layer of LSTM cells forward by the sequence length.
The sequence length is determined by the length of x_seq and pad_seq, which
must be the same.
Args:
cell_name: Base name of each cell.
weights: Weight matrix with shape LSTMCellWeightsShape.
m: Initial m states with shape [batch_size, num_nodes].
c: Initial c states with shape [batch_size, num_nodes].
x_seq: List of inputs, each with shape [batch_size, num_inputs].
The length of the list is the sequence length.
pad_seq: List of paddings, each with shape [batch_size, 1].
The length of the list is the sequence length.
Each padding value is either 0 or 1, where 1 indicates padding;
i.e. the input is shorter than the sequence length.
Returns:
List of per-sequence-step outputs, each with shape [batch_size, num_nodes].
Raises:
ValueError: If len(x_seq) != len(pad_seq)."
110,RandomVar,tensorflow/tensorflow/compiler/tests/lstm.py,121,function,Returns a variable of the given shape initialized to random values.
111,RandomInputs,tensorflow/tensorflow/compiler/tests/lstm.py,127,function,"Returns randomly initialized (x_seq, pad_seq) sequences."
112,BuildLSTMLayer,tensorflow/tensorflow/compiler/tests/lstm.py,140,function,"Builds a single LSTM layer with random weights and inputs.
Args:
batch_size: Inputs are fed in batches of this size.
seq_length: The sequence length to unroll the LSTM layer.
num_inputs: Dimension of inputs that are fed into each LSTM cell.
num_nodes: The number of nodes in each LSTM cell.
Returns:
(out_seq, weights) pair. The out_seq is a list of per-sequence-step
outputs, each with shape [batch_size, num_nodes]. The weights are a list of
weight variables that may be trained."
113,LSTMBenchmark,tensorflow/tensorflow/compiler/tests/lstm_test.py,238,class,Mcro-benchmarks for a single layer of LSTM cells.
114,benchmarkLayerInference,tensorflow/tensorflow/compiler/tests/lstm_test.py,256,method,
115,benchmarkLayerInferenceXLA,tensorflow/tensorflow/compiler/tests/lstm_test.py,260,method,
116,benchmarkLayerTraining,tensorflow/tensorflow/compiler/tests/lstm_test.py,264,method,
117,benchmarkLayerTrainingXLA,tensorflow/tensorflow/compiler/tests/lstm_test.py,268,method,
118,zip_to_first_list_length,tensorflow/tensorflow/compiler/tests/matrix_diag_ops_test.py,32,function,
119,repack_diagonals,tensorflow/tensorflow/compiler/tests/matrix_diag_ops_test.py,40,function,
120,square_cases,tensorflow/tensorflow/compiler/tests/matrix_diag_ops_test.py,95,function,
121,tall_cases,tensorflow/tensorflow/compiler/tests/matrix_diag_ops_test.py,173,function,
122,fat_cases,tensorflow/tensorflow/compiler/tests/matrix_diag_ops_test.py,261,function,
123,MakePlaceholder,tensorflow/tensorflow/compiler/tests/matrix_triangular_solve_op_test.py,36,function,
124,NHWCToNCHW,tensorflow/tensorflow/compiler/tests/pooling_ops_test.py,33,function,"Convert the input from NHWC format to NCHW.
Args:
input_tensor: a 4-D tensor, or a 4-element array representing the same.
Returns:
the converted tensor or a shape array"
125,NCHWToNHWC,tensorflow/tensorflow/compiler/tests/pooling_ops_test.py,48,function,"Convert the input from NCHW format to NHWC.
Args:
input_tensor: a 4-D tensor, or a 4-element array representing the same.
Returns:
the converted tensor or a shape array"
126,numpy_reverse,tensorflow/tensorflow/compiler/tests/scan_ops_test.py,32,function,
127,handle_options,tensorflow/tensorflow/compiler/tests/scan_ops_test.py,43,function,Adds tf options to numpy scan ops.
128,space_to_batch_direct,tensorflow/tensorflow/compiler/tests/spacetobatch_op_test.py,30,function,"Direct Python implementation of space-to-batch conversion.
This is used for tests only.
Args:
input_array: N-D array
block_shape: 1-D array of shape [num_block_dims].
paddings: 2-D array of shape [num_block_dims, 2].
Returns:
Converted tensor."
129,implicit_reparameterization_grad,tensorflow/tensorflow/compiler/tests/special_math_test.py,58,function,
130,xla_device,tensorflow/tensorflow/compiler/tests/stateful_random_ops_test.py,41,function,
131,xla_device_name,tensorflow/tensorflow/compiler/tests/stateful_random_ops_test.py,55,function,
132,StatelessRandomOpsBenchmark,tensorflow/tensorflow/compiler/tests/stateless_random_ops_test.py,136,class,Microbenchmarks for the stateless random ops.
133,benchmarkUniformF32,tensorflow/tensorflow/compiler/tests/stateless_random_ops_test.py,152,method,
134,benchmarkUniformF64,tensorflow/tensorflow/compiler/tests/stateless_random_ops_test.py,156,method,
135,benchmarkUniformF32XLA,tensorflow/tensorflow/compiler/tests/stateless_random_ops_test.py,160,method,
136,benchmarkUniformF64XLA,tensorflow/tensorflow/compiler/tests/stateless_random_ops_test.py,164,method,
137,BuilderFn,tensorflow/tensorflow/compiler/tests/stateless_random_ops_test.py,141,method,
138,ConvertBetweenDataFormats,tensorflow/tensorflow/compiler/tests/test_utils.py,26,function,Converts 4D tensor between data formats.
139,PermuteDimsBetweenDataFormats,tensorflow/tensorflow/compiler/tests/test_utils.py,47,function,Get new shape for converting between data formats.
140,RunWithWarmup,tensorflow/tensorflow/compiler/tests/test_utils.py,71,function,Runs a graph a few times to ensure that its clusters are compiled.
141,nhwc_to_format,tensorflow/tensorflow/compiler/tests/unary_ops_test.py,37,function,Converts a numpy array from NHWC format to `data_format`.
142,StridedSliceAssignChecker,tensorflow/tensorflow/compiler/tests/variable_ops_test.py,422,class,Compares the results of a slice assignment using Tensorflow and numpy.
143,is_compile_on_demand,tensorflow/tensorflow/compiler/tests/while_test.py,260,function,
144,parse_disabled_manifest,tensorflow/tensorflow/compiler/tests/xla_test.py,55,function,
145,Benchmark,tensorflow/tensorflow/compiler/tests/xla_test.py,250,function,"Build a graph and run benchmarks against it, with or without XLA.
Args:
tf_bench: An instance of tf.test.Benchmark, used to run the benchmark.
builder_fn: A function that builds a graph when invoked, and returns
(name, fetches), where name is the name of the test, and fetches
is a list of tensors to fetch as output.
use_xla_jit: If true compile with the XLA JIT, otherwise use regular TF.
device: The tensorflow device to run on, e.g. ""cpu"", ""gpu"".
separate_compiled_gradients: If true put each gradient subgraph into a
separate compilation scope. This gives fine-grained control over which
portions of the graph will be compiled as a single unit. Compiling
gradients separately may yield better performance for some graphs.
The scope is named based on the scope of the forward computation as well
as the name of the gradients. As a result, the gradients will be compiled
in a scope that is separate from both the forward computation, and from
other gradients."
146,broadcast,tensorflow/tensorflow/compiler/tf2xla/python/xla.py,226,function,
147,clamp,tensorflow/tensorflow/compiler/tf2xla/python/xla.py,234,function,
148,conv,tensorflow/tensorflow/compiler/tf2xla/python/xla.py,241,function,"Wraps the XLA ConvGeneralDilated operator.
ConvGeneralDilated is the most general form of XLA convolution and is
documented at
https://www.tensorflow.org/performance/xla/operation_semantics#conv_convolution
Args:
lhs: the input tensor
rhs: the kernel tensor
window_strides: the inter-window strides
padding: the padding to apply at the start and end of each input dimensions
lhs_dilation: dilation to apply between input elements
rhs_dilation: dilation to apply between kernel elements
dimension_numbers: a `ConvolutionDimensionNumbers` proto.
feature_group_count: number of feature groups for grouped convolution.
precision_config: a `xla.PrecisionConfig` proto.
name: an optional name for the operator
Returns:
A tensor representing the output of the convolution."
149,dot,tensorflow/tensorflow/compiler/tf2xla/python/xla.py,291,function,
150,dot_general,tensorflow/tensorflow/compiler/tf2xla/python/xla.py,295,function,
151,self_adjoint_eig,tensorflow/tensorflow/compiler/tf2xla/python/xla.py,307,function,
152,svd,tensorflow/tensorflow/compiler/tf2xla/python/xla.py,311,function,
153,random_normal,tensorflow/tensorflow/compiler/tf2xla/python/xla.py,327,function,
154,random_uniform,tensorflow/tensorflow/compiler/tf2xla/python/xla.py,333,function,
155,reduce_window,tensorflow/tensorflow/compiler/tf2xla/python/xla.py,343,function,"Wraps the XLA ReduceWindow operator.
ReduceWindow is documented at
https://www.tensorflow.org/performance/xla/operation_semantics#reducewindow .
Args:
operand: the input tensor
init: a scalar tensor representing the initial value for the reduction
reducer: a reduction function that combines a pair of scalars.
window_dimensions: shape of the window, as a list of integers
window_strides: inter-window strides, as a list of integers. Optional; if
omitted, defaults to strides of 1.
padding: padding to apply to 'operand'. List of (low, high) pairs of
integers that specify the padding to apply before and after each
dimension. Optional; if omitted, defaults to no padding.
name: the operator name, or None.
Returns:
A tensor that represents the output of the reduce_window operator."
156,reshape,tensorflow/tensorflow/compiler/tf2xla/python/xla.py,391,function,
157,select,tensorflow/tensorflow/compiler/tf2xla/python/xla.py,398,function,
158,slice,tensorflow/tensorflow/compiler/tf2xla/python/xla.py,406,function,
159,gather,tensorflow/tensorflow/compiler/tf2xla/python/xla.py,452,function,
160,scatter,tensorflow/tensorflow/compiler/tf2xla/python/xla.py,463,function,
161,Sharding,tensorflow/tensorflow/compiler/xla/experimental/xla_sharding/xla_sharding.py,28,class,"A class to support adding sharding attributes to Ops.
Use the factory constructors and then call apply_to_tensor:
Sharding.replicate().apply_to_tensor(tensor)"
162,replicate,tensorflow/tensorflow/compiler/xla/experimental/xla_sharding/xla_sharding.py,40,method,"Returns a replicated sharding attribute.
This causes an op to be computed in its entirety independently on all
cores in the XLA device."
163,assign_device,tensorflow/tensorflow/compiler/xla/experimental/xla_sharding/xla_sharding.py,50,method,"Returns an AssignDevice sharding attribute.
This causes an op to be computed in its entirety only on one core in
the XLA device.
Args:
core: The core to assign this Op to."
164,tile,tensorflow/tensorflow/compiler/xla/experimental/xla_sharding/xla_sharding.py,65,method,"Returns a Tiled sharding attribute.
This causes an op to be partially computed on multiple cores in the
XLA device.
Args:
tile_assignment: An np.ndarray describing the topology of the tiling and
which device will compute which part of the topology.
Raises:
TypeError: tile_assignment was not of np.array type.
TODO(jmolloy): This concept is nefarious and is not
something we really want to expose to users (especially as the
contract for tile_assignment is very strict)."
165,split,tensorflow/tensorflow/compiler/xla/experimental/xla_sharding/xla_sharding.py,93,method,"Returns a Sharding that splits a tensor across a dimension.
This creates a Tiled attribute, similar to tile(), but easier to use for the
common case of tiling a tensor N ways in one dimension.
Args:
tensor: A tf.Tensor to split.
split_dimension: The dimension number to split.
num_devices: The number of cores to split `tensor` over.
input_shape: The shape of the original tensor.
Raises:
ValueError: The tensor to split was smaller in the split dimension than
the number of devices to split over."
166,apply_to_tensor,tensorflow/tensorflow/compiler/xla/experimental/xla_sharding/xla_sharding.py,128,method,"Applies this Sharding attribute to `tensor`.
Args:
tensor: A tf.Tensor to split.
assign_tuple_sharding: If the sharding type should be a tuple."
167,proto,tensorflow/tensorflow/compiler/xla/experimental/xla_sharding/xla_sharding.py,152,method,Return the sharding protobuf of type xla_data_pb2.OpSharding.
168,replicate,tensorflow/tensorflow/compiler/xla/experimental/xla_sharding/xla_sharding.py,179,function,
169,assign_device,tensorflow/tensorflow/compiler/xla/experimental/xla_sharding/xla_sharding.py,188,function,Returns a tensor that has AssignDevice sharding attribute.
170,tile,tensorflow/tensorflow/compiler/xla/experimental/xla_sharding/xla_sharding.py,202,function,"Returns a tensor that has tiled sharding.
Args:
tensor: A tf.Tensor to shard.
tile_assignment: An np.ndarray describing the topology of the tiling and
which device will compute which part of the topology.
assign_tuple_sharding: If the sharding type should be a tuple.
use_sharding_op: If true, adds a sharding op to set the sharding."
171,split,tensorflow/tensorflow/compiler/xla/experimental/xla_sharding/xla_sharding.py,224,function,"Returns a tensor that is split along the given dimension.
Args:
tensor: A tf.Tensor to split.
split_dimension: The dimension to split.
num_devices: The number of devices to partition the dimension.
assign_tuple_sharding: If the sharding type should be a tuple.
use_sharding_op: If true, adds a sharding op to set the sharding.
input_shape: The full shape of the input tensor."
172,get_op_sharding,tensorflow/tensorflow/compiler/xla/experimental/xla_sharding/xla_sharding.py,248,function,"Returns sharding attribute of an op.
Args:
op: a TensorFlow op.
Returns:
The attribute representing XLA sharding on this op."
173,auto_to_manual_spmd_partition,tensorflow/tensorflow/compiler/xla/experimental/xla_sharding/xla_sharding.py,260,function,"Switches from automatic SPMD partitioning to manual partitioning.
Converts a full-shaped tensor (to be automatically partitioned by SPMD
partitioner) to a shard-shaped tensor to be consumed by manually partitioned
ops.
Args:
tensor: A tf.Tensor in full shape.
manual_sharding: a serialized string of OpSharding to be used in manual
partitioning.
Returns:
A shard-shaped tensor to be consumed by manually partitioned ops."
174,manual_to_auto_spmd_partition,tensorflow/tensorflow/compiler/xla/experimental/xla_sharding/xla_sharding.py,279,function,"Switches from manual partitioning to automatic SPMD partitioning.
Converts a shard-shaped tensor (manually partitioned in SPMD-style) to a
full-shaped tensor to be partitioned automatically by the SPMD partitioner.
Args:
tensor: A tf.Tensor in shard shape.
manual_sharding: a serialized string of OpSharding to be used in manual
partitioning.
full_shape: the shape of tensor before partitioning.
Returns:
A full-shaped tensor to be partitioned automatically by the SPMD
partitioner."
175,numpy_assert_allclose,tensorflow/tensorflow/compiler/xla/python/bfloat16_test.py,35,function,
176,register_local_backend_factory,tensorflow/tensorflow/compiler/xla/python/xla_client.py,101,function,
177,get_local_backend,tensorflow/tensorflow/compiler/xla/python/xla_client.py,131,function,"Returns a local backend.
Args:
name: the backend name. If `None`, a default local backend is returned,
typically `gpu` if one is present, or `cpu` if not. If a string, the named
backend is returned or an exception raised.
Returns:
A LocalBackend object."
178,OpMetadata,tensorflow/tensorflow/compiler/xla/python/xla_client.py,152,class,Python representation of a xla.OpMetadata protobuf.
179,CurrentSourceInfoMetadata,tensorflow/tensorflow/compiler/xla/python/xla_client.py,163,function,Helper for use in source mapping that returns an OpMetadata object.
180,dtype_to_etype,tensorflow/tensorflow/compiler/xla/python/xla_client.py,206,function,Convenience function for reading DTYPE_TO_XLA_ELEMENT_TYPE.
181,shape_from_pyval,tensorflow/tensorflow/compiler/xla/python/xla_client.py,272,function,Returns a Shape that describes a tuple-tree of Numpy arrays.
182,execute_with_python_values,tensorflow/tensorflow/compiler/xla/python/xla_client.py,334,function,Execute on one replica with Python values as arguments and output.
183,execute_with_python_values_replicated,tensorflow/tensorflow/compiler/xla/python/xla_client.py,345,function,"Execute on many replicas with Python values as arguments and output.
Arguments:
executable: the program to run.
arguments: a list of lists of Python values indexed by `[replica][arg_num]`
to pass as inputs.
backend: the backend we are targeting.
Returns:
A list of python values, one per replica."
184,PaddingType,tensorflow/tensorflow/compiler/xla/python/xla_client.py,374,class,
185,window_padding_type_to_pad_values,tensorflow/tensorflow/compiler/xla/python/xla_client.py,379,function,Maps PaddingType or string to pad values (list of pairs of ints).
186,register_custom_call_target,tensorflow/tensorflow/compiler/xla/python/xla_client.py,418,function,"Registers a custom call target.
Args:
name: bytes containing the name of the function.
fn: a PyCapsule object containing the function pointer.
platform: the target platform."
187,PaddingConfigDimension,tensorflow/tensorflow/compiler/xla/python/xla_client.py,433,class,Python representation of a xla.PaddingConfigDimension protobuf.
188,PaddingConfig,tensorflow/tensorflow/compiler/xla/python/xla_client.py,443,class,Python representation of a xla.PaddingConfig protobuf.
189,make_padding_config,tensorflow/tensorflow/compiler/xla/python/xla_client.py,451,function,"Create PaddingConfig proto from list of triples of integers.
Args:
padding_config: either a PaddingConfig or a list of integer triples
(edge_padding_low, edge_padding_high, interior_padding) representing the
configuration of the padding operation.
Returns:
A `PaddingConfig` object."
190,DotDimensionNumbers,tensorflow/tensorflow/compiler/xla/python/xla_client.py,476,class,Python representation of a xla.DotDimensionNumbers protobuf.
191,make_dot_dimension_numbers,tensorflow/tensorflow/compiler/xla/python/xla_client.py,488,function,"Builds a DotDimensionNumbers object from a specification.
Args:
dimension_numbers: either a `DotDimensionNumbers` or a nested tuple
`((lhs_contract, rhs_contract), (lhs_batch, rhs_batch))` of lists of
integers representing the dimensions to treat as contracting dimensions
and batch dimensions on each input operand.
Returns:
A `DotDimensionNumbers` object."
192,ConvolutionDimensionNumbers,tensorflow/tensorflow/compiler/xla/python/xla_client.py,516,class,Python representation of a xla.ConvolutionDimensionNumbers protobuf.
193,make_convolution_dimension_numbers,tensorflow/tensorflow/compiler/xla/python/xla_client.py,536,function,"Builds a ConvolutionDimensionNumbers object from a specification.
Args:
dimension_numbers: optional, either a ConvolutionDimensionNumbers object or
a tuple (lhs_spec, rhs_spec, out_spec). Each element is a string of
length N+2 identifying by position: (1) batch dimensions in lhs, rhs, and
the output with the character 'N', (2) feature dimensions in lhs and the
output with the character 'C', (3) input and output feature dimensions
in rhs with the characters 'I' and 'O' respectively, and (4) spatial
dimension correspondences between lhs, rhs, and the output using any
distinct characters. For example, to indicate dimension numbers
consistent with the Conv operation with two spatial dimensions, one
could use ('NCHW', 'OIHW', 'NCHW'). As another example, to indicate
dimension numbers consistent with the TensorFlow Conv2D operation, one
could use ('NHWC', 'HWIO', 'NHWC'). When using the latter form of
convolution dimension specification, window strides are associated with
spatial dimension character labels according to the order in which the
labels appear in the rhs_spec string, so that window_strides[0] is
matched with the dimension corresponding to the first character
appearing in rhs_spec that is not 'I' or 'O'. By default, use the same
dimension numbering as Conv and ConvWithGeneralPadding.
num_spatial_dimensions: the number of spatial dimensions.
Returns:
A `ConvolutionDimensionNumbers` object."
194,OpSharding,tensorflow/tensorflow/compiler/xla/python/xla_client.py,600,class,Python representation of a xla.OpSharding protobuf.
195,PrecisionConfig,tensorflow/tensorflow/compiler/xla/python/xla_client.py,614,class,Python representation of a xla.PrecisionConfig protobuf.
196,GatherDimensionNumbers,tensorflow/tensorflow/compiler/xla/python/xla_client.py,624,class,Python representation of a xla.GatherDimensionNumbers protobuf.
197,ScatterDimensionNumbers,tensorflow/tensorflow/compiler/xla/python/xla_client.py,636,class,Python representation of a xla.ScatterDimensionNumbers protobuf.
198,ReplicaGroup,tensorflow/tensorflow/compiler/xla/python/xla_client.py,648,class,Python representation of a xla.ReplicaGroup protobuf.
199,make_replica_groups,tensorflow/tensorflow/compiler/xla/python/xla_client.py,662,function,
200,tracebacks,tensorflow/tensorflow/compiler/xla/python/xla_client.py,677,function,Context manager that enables or disables traceback collection.
201,heap_profile,tensorflow/tensorflow/compiler/xla/python/xla_client.py,687,function,Returns a gzipped pprof protocol buffer containing a heap profile.
202,TpuBackend,tensorflow/tensorflow/compiler/xla/python/tpu_driver/client/tpu_client.py,29,class,XLA backend implemented using the Tpu driver API.
203,create,tensorflow/tensorflow/compiler/xla/python/tpu_driver/client/tpu_client.py,36,method,Constructs a Cloud TPU backend.
204,ConvertLiteralToNumpyArray,tensorflow/tensorflow/compiler/xla/python_api/xla_literal.py,28,function,Converts a XLA literal to a Numpy array.
205,ConvertNumpyArrayToLiteral,tensorflow/tensorflow/compiler/xla/python_api/xla_literal.py,85,function,Converts a Numpy array or a nested tuple thereof to an XLA literal.
206,Shape,tensorflow/tensorflow/compiler/xla/python_api/xla_shape.py,29,class,"Wraps a xla_data_pb2.ShapeProto message with a convenient Python type.
Provides direct access to the underlying xla_data_pb2.ShapeProto message in
the
message attribute, along with accessor wrappers to the message's fields.
Avoid direct access to .message unless interacting directly with protobuf APIs
like CopyFrom. In other words, prefer hauling the shape around in a Shape, and
only access .message when strictly required by the protobuf API."
207,element_type,tensorflow/tensorflow/compiler/xla/python_api/xla_shape.py,71,method,
208,is_tuple,tensorflow/tensorflow/compiler/xla/python_api/xla_shape.py,74,method,
209,dimensions,tensorflow/tensorflow/compiler/xla/python_api/xla_shape.py,77,method,
210,tuple_shapes,tensorflow/tensorflow/compiler/xla/python_api/xla_shape.py,82,method,"If this is a tuple, returns its sequence of constituent Shape objects.
Returns:
Tuple sub-shapes.
Raises:
ValueError: if this is not a tuple."
211,layout,tensorflow/tensorflow/compiler/xla/python_api/xla_shape.py,95,method,
212,from_pyval,tensorflow/tensorflow/compiler/xla/python_api/xla_shape.py,99,method,
213,CreateShapeFromNumpy,tensorflow/tensorflow/compiler/xla/python_api/xla_shape.py,129,function,"Create a Shape from a Numpy array or a nested tuple structure thereof.
Args:
value: Numpy array or (possibly nested) tuple structure that bottoms out in
Numpy arrays.
Returns:
A Shape object."
214,CreateShapeFromDtypeAndTuple,tensorflow/tensorflow/compiler/xla/python_api/xla_shape.py,147,function,"Create a shape from a Numpy dtype and a sequence of nonnegative integers.
Args:
dtype: a numpy dtype, e.g. np.dtype('int32').
shape_tuple: a sequence of nonnegative integers.
Returns:
A Shape object."
215,load_graph,tensorflow/tensorflow/examples/label_image/label_image.py,26,function,
216,read_tensor_from_image_file,tensorflow/tensorflow/examples/label_image/label_image.py,38,function,
217,load_labels,tensorflow/tensorflow/examples/label_image/label_image.py,65,function,
218,MaybeDistributionScope,tensorflow/tensorflow/examples/saved_model/integration_tests/distribution_strategy_utils.py,48,class,Provides a context allowing no distribution strategy.
219,from_name,tensorflow/tensorflow/examples/saved_model/integration_tests/distribution_strategy_utils.py,52,method,
220,make_feature_extractor,tensorflow/tensorflow/examples/saved_model/integration_tests/export_mnist_cnn.py,56,function,Returns a Keras Model to compute a feature vector from MNIST images.
221,set_feature_extractor_hparams,tensorflow/tensorflow/examples/saved_model/integration_tests/export_mnist_cnn.py,72,function,
222,make_classifier,tensorflow/tensorflow/examples/saved_model/integration_tests/export_mnist_cnn.py,76,function,Returns a Keras Model to classify MNIST using feature_extractor.
223,wrap_keras_model_for_export,tensorflow/tensorflow/examples/saved_model/integration_tests/export_mnist_cnn.py,87,function,Wraps `model` for saving and loading as SavedModel.
224,write_vocabulary_file,tensorflow/tensorflow/examples/saved_model/integration_tests/export_simple_text_embedding.py,34,function,Write temporary vocab file for module construction.
225,TextEmbeddingModel,tensorflow/tensorflow/examples/saved_model/integration_tests/export_simple_text_embedding.py,44,class,"Text embedding model.
A text embeddings model that takes a sentences on input and outputs the
sentence embedding."
226,TextRnnModel,tensorflow/tensorflow/examples/saved_model/integration_tests/export_text_rnn_model.py,31,class,"Text RNN model.
A full generative text RNN model that can train and decode sentences from a
starting word."
227,train,tensorflow/tensorflow/examples/saved_model/integration_tests/export_text_rnn_model.py,81,method,
228,decode_greedy,tensorflow/tensorflow/examples/saved_model/integration_tests/export_text_rnn_model.py,143,method,
229,MaybeRunScriptInstead,tensorflow/tensorflow/examples/saved_model/integration_tests/integration_scripts.py,62,function,
230,load_reshaped_data,tensorflow/tensorflow/examples/saved_model/integration_tests/mnist_util.py,34,function,Returns MNIST or Fashion MNIST or fake train and test data.
231,make_feature_extractor,tensorflow/tensorflow/examples/saved_model/integration_tests/use_mnist_cnn.py,72,function,Load a pre-trained feature extractor and wrap it for use in Keras.
232,make_classifier,tensorflow/tensorflow/examples/saved_model/integration_tests/use_mnist_cnn.py,100,function,Returns a Keras Model to classify MNIST using feature_extractor.
233,train,tensorflow/tensorflow/examples/saved_model/integration_tests/use_model_in_sequential_keras.py,35,function,Build a Keras model and train with mock data.
234,train,tensorflow/tensorflow/examples/saved_model/integration_tests/use_text_embedding_in_dataset.py,34,function,Build a Keras model and train with mock data.
235,StreamingAccuracyStats,tensorflow/tensorflow/examples/speech_commands/accuracy_utils.py,24,class,"Get streaming accuracy statistics every time a new command is founded.
Attributes:
_how_many_gt: How many ground truths.
_how_many_gt_matched: How many ground truths have been matched.
_how_many_fp: How many commands have been fired as false positive.
_how_many_c: How many commands have been fired correctly.
_how_many_w: How many commands have been fired wrongly.
_gt_occurrence: A list to record which commands and when it occurs in the
input audio stream.
_previous_c: A variable to record the last status of _how_many_c.
_previous_w: A variable to record the last status of _how_many_w.
_previous_fp: A variable to record the last status of _how_many_fp."
236,read_ground_truth_file,tensorflow/tensorflow/examples/speech_commands/accuracy_utils.py,52,method,Load ground truth and timestamp pairs and store it in time order.
237,delta,tensorflow/tensorflow/examples/speech_commands/accuracy_utils.py,64,method,Compute delta of StreamingAccuracyStats against last status.
238,calculate_accuracy_stats,tensorflow/tensorflow/examples/speech_commands/accuracy_utils.py,83,method,"Calculate accuracy statistics when a new commands is founded.
Given ground truth and corresponding predictions founded by
model, figure out how many were correct. Take a tolerance time, so that only
predictions up to a point in time are considered.
Args:
found_words: A list of all founded commands up to now.
up_to_time_ms: End timestamp of this audio piece.
time_tolerance_ms: The tolerance milliseconds before and after
up_to_time_ms to match a ground truth."
239,print_accuracy_stats,tensorflow/tensorflow/examples/speech_commands/accuracy_utils.py,137,method,Write a human-readable description of the statistics to stdout.
240,create_inference_graph,tensorflow/tensorflow/examples/speech_commands/freeze.py,63,function,"Creates an audio model with the nodes needed for inference.
Uses the supplied arguments to create a model, and inserts the input and
output nodes that are needed to use the graph for inference.
Args:
wanted_words: Comma-separated list of the words we're trying to recognize.
sample_rate: How many samples per second are in the input audio files.
clip_duration_ms: How many samples to analyze for the audio pattern.
clip_stride_ms: How often to run recognition. Useful for models with cache.
window_size_ms: Time slice duration to estimate frequencies from.
window_stride_ms: How far apart time slices should be.
feature_bin_count: Number of frequency bands to analyze.
model_architecture: Name of the kind of model to generate.
preprocess: How the spectrogram is processed to produce features, for
example 'mfcc', 'average', or 'micro'.
Returns:
Input and output tensor objects.
Raises:
Exception: If the preprocessing mode isn't recognized."
241,save_graph_def,tensorflow/tensorflow/examples/speech_commands/freeze.py,161,function,"Writes a graph def file out to disk.
Args:
file_name: Where to save the file.
frozen_graph_def: GraphDef proto object to save."
242,save_saved_model,tensorflow/tensorflow/examples/speech_commands/freeze.py,176,function,"Writes a SavedModel out to disk.
Args:
file_name: Where to save the file.
sess: TensorFlow session containing the graph.
input_tensor: Tensor object defining the input's properties.
output_tensor: Tensor object defining the output's properties."
243,mix_in_audio_sample,tensorflow/tensorflow/examples/speech_commands/generate_streaming_test_wav.py,55,function,"Mixes the sample data into the main track at the specified offset.
Args:
track_data: Numpy array holding main audio data. Modified in-place.
track_offset: Where to mix the sample into the main track.
sample_data: Numpy array of audio data to mix into the main track.
sample_offset: Where to start in the audio sample.
clip_duration: How long the sample segment is.
sample_volume: Loudness to mix the sample in at.
ramp_in: Length in samples of volume increase stage.
ramp_out: Length in samples of volume decrease stage."
244,prepare_words_list,tensorflow/tensorflow/examples/speech_commands/input_data.py,58,function,"Prepends common tokens to the custom word list.
Args:
wanted_words: List of strings containing the custom words.
Returns:
List with the standard silence and unknown tokens added."
245,which_set,tensorflow/tensorflow/examples/speech_commands/input_data.py,70,function,"Determines which data partition the file should belong to.
We want to keep files in the same training, validation, or testing sets even
if new ones are added over time. This makes it less likely that testing
samples will accidentally be reused in training when long runs are restarted
for example. To keep this stability, a hash of the filename is taken and used
to determine which set it should belong to. This determination only depends on
the name and the set proportions, so it won't change as other files are added.
It's also useful to associate particular files as related (for example words
spoken by the same person), so anything after '_nohash_' in a filename is
ignored for set determination. This ensures that 'bobby_nohash_0.wav' and
'bobby_nohash_1.wav' are always in the same set, for example.
Args:
filename: File path of the data sample.
validation_percentage: How much of the data set to use for validation.
testing_percentage: How much of the data set to use for testing.
Returns:
String, one of 'training', 'validation', or 'testing'."
246,load_wav_file,tensorflow/tensorflow/examples/speech_commands/input_data.py,118,function,"Loads an audio file and returns a float PCM-encoded array of samples.
Args:
filename: Path to the .wav file to load.
Returns:
Numpy array holding the sample data as floats between -1.0 and 1.0."
247,save_wav_file,tensorflow/tensorflow/examples/speech_commands/input_data.py,136,function,"Saves audio sample data to a .wav audio file.
Args:
filename: Path to save the file to.
wav_data: 2D array of float PCM-encoded audio data.
sample_rate: Samples per second to encode in the file."
248,get_features_range,tensorflow/tensorflow/examples/speech_commands/input_data.py,160,function,"Returns the expected min/max for generated features.
Args:
model_settings: Information about the current model being trained.
Returns:
Min/max float pair holding the range of features.
Raises:
Exception: If preprocessing mode isn't recognized."
249,AudioProcessor,tensorflow/tensorflow/examples/speech_commands/input_data.py,190,class,"Handles loading, partitioning, and preparing audio training data."
250,maybe_download_and_extract_dataset,tensorflow/tensorflow/examples/speech_commands/input_data.py,205,method,"Download and extract data set tar file.
If the data set we're using doesn't already exist, this function
downloads it from the TensorFlow.org website and unpacks it into a
directory.
If the data_url is none, don't download anything and expect the data
directory to contain the correct files already.
Args:
data_url: Web location of the tar file containing the data set.
dest_directory: File path to extract data to."
251,prepare_data_index,tensorflow/tensorflow/examples/speech_commands/input_data.py,247,method,"Prepares a list of the samples organized by set and label.
The training loop needs a list of all the available data, organized by
which partition it should belong to, and with ground truth labels attached.
This function analyzes the folders below the `data_dir`, figures out the
right
labels for each file based on the name of the subdirectory it belongs to,
and uses a stable hash to assign it to a data set partition.
Args:
silence_percentage: How much of the resulting data should be background.
unknown_percentage: How much should be audio outside the wanted classes.
wanted_words: Labels of the classes we want to be able to recognize.
validation_percentage: How much of the data set to use for validation.
testing_percentage: How much of the data set to use for testing.
Returns:
Dictionary containing a list of file information for each set partition,
and a lookup map for each class to determine its numeric index.
Raises:
Exception: If expected files are not found."
252,prepare_background_data,tensorflow/tensorflow/examples/speech_commands/input_data.py,333,method,"Searches a folder for background noise audio, and loads it into memory.
It's expected that the background audio samples will be in a subdirectory
named '_background_noise_' inside the 'data_dir' folder, as .wavs that match
the sample rate of the training data, but can be much longer in duration.
If the '_background_noise_' folder doesn't exist at all, this isn't an
error, it's just taken to mean that no background noise augmentation should
be used. If the folder does exist, but it's empty, that's treated as an
error.
Returns:
List of raw PCM-encoded audio samples of background noise.
Raises:
Exception: If files aren't found in the folder."
253,prepare_processing_graph,tensorflow/tensorflow/examples/speech_commands/input_data.py,369,method,"Builds a TensorFlow graph to apply the input distortions.
Creates a graph that loads a WAVE file, decodes it, scales the volume,
shifts it in time, adds in background noise, calculates a spectrogram, and
then builds an MFCC fingerprint from that.
This must be called with an active TensorFlow session running, and it
creates multiple placeholder inputs, and one output:
- wav_filename_placeholder_: Filename of the WAV to load.
- foreground_volume_placeholder_: How loud the main clip should be.
- time_shift_padding_placeholder_: Where to pad the clip.
- time_shift_offset_placeholder_: How much to move the clip in time.
- background_data_placeholder_: PCM sample data for background noise.
- background_volume_placeholder_: Loudness of mixed-in background.
- output_: Output 2D fingerprint of processed audio.
Args:
model_settings: Information about the current model being trained.
summaries_dir: Path to save training summary information to.
Raises:
ValueError: If the preprocessing mode isn't recognized.
Exception: If the preprocessor wasn't compiled in."
254,set_size,tensorflow/tensorflow/examples/speech_commands/input_data.py,498,method,"Calculates the number of samples in the dataset partition.
Args:
mode: Which partition, must be 'training', 'validation', or 'testing'.
Returns:
Number of samples in the partition."
255,get_data,tensorflow/tensorflow/examples/speech_commands/input_data.py,509,method,"Gather samples from the data set, applying transformations as needed.
When the mode is 'training', a random selection of samples will be returned,
otherwise the first N clips in the partition will be used. This ensures that
validation always uses the same samples, reducing noise in the metrics.
Args:
how_many: Desired number of samples to return. -1 means the entire
contents of this partition.
offset: Where to start when fetching deterministically.
model_settings: Information about the current model being trained.
background_frequency: How many clips will have background noise, 0.0 to
1.0.
background_volume_range: How loud the background noise will be.
time_shift: How much to randomly shift the clips by in time.
mode: Which partition to use, must be 'training', 'validation', or
'testing'.
sess: TensorFlow session that was active when processor was created.
Returns:
List of sample data for the transformed samples, and list of label indexes
Raises:
ValueError: If background samples are too short."
256,get_features_for_wav,tensorflow/tensorflow/examples/speech_commands/input_data.py,612,method,"Applies the feature transformation process to the input_wav.
Runs the feature generation process (generally producing a spectrogram from
the input samples) on the WAV file. This can be useful for testing and
verifying implementations being run on other platforms.
Args:
wav_filename: The path to the input audio file.
model_settings: Information about the current model being trained.
sess: TensorFlow session that was active when processor was created.
Returns:
Numpy data array containing the generated features."
257,get_unprocessed_data,tensorflow/tensorflow/examples/speech_commands/input_data.py,640,method,"Retrieve sample data for the given partition, with no transformations.
Args:
how_many: Desired number of samples to return. -1 means the entire
contents of this partition.
model_settings: Information about the current model being trained.
mode: Which partition to use, must be 'training', 'validation', or
'testing'.
Returns:
List of sample data for the samples, and list of labels in one-hot form."
258,load_graph,tensorflow/tensorflow/examples/speech_commands/label_wav.py,43,function,Unpersists graph from file as default graph.
259,load_labels,tensorflow/tensorflow/examples/speech_commands/label_wav.py,51,function,"Read in labels, one label per line."
260,run_graph,tensorflow/tensorflow/examples/speech_commands/label_wav.py,56,function,Runs the audio data through the graph and prints predictions.
261,label_wav,tensorflow/tensorflow/examples/speech_commands/label_wav.py,77,function,"Loads the model and labels, and runs the inference to print predictions."
262,load_graph,tensorflow/tensorflow/examples/speech_commands/label_wav_dir.py,44,function,Unpersists graph from file as default graph.
263,load_labels,tensorflow/tensorflow/examples/speech_commands/label_wav_dir.py,52,function,"Read in labels, one label per line."
264,run_graph,tensorflow/tensorflow/examples/speech_commands/label_wav_dir.py,57,function,Runs the audio data through the graph and prints predictions.
265,label_wav,tensorflow/tensorflow/examples/speech_commands/label_wav_dir.py,85,function,"Loads the model and labels, and runs the inference to print predictions."
266,prepare_model_settings,tensorflow/tensorflow/examples/speech_commands/models.py,39,function,"Calculates common settings needed for all models.
Args:
label_count: How many classes are to be recognized.
sample_rate: Number of audio samples per second.
clip_duration_ms: Length of each audio clip to be analyzed.
window_size_ms: Duration of frequency analysis window.
window_stride_ms: How far to move in time between frequency windows.
feature_bin_count: Number of frequency bins to use for analysis.
preprocess: How the spectrogram is processed to produce features.
Returns:
Dictionary containing common settings.
Raises:
ValueError: If the preprocessing mode isn't recognized."
267,create_model,tensorflow/tensorflow/examples/speech_commands/models.py,95,function,"Builds a model of the requested architecture compatible with the settings.
There are many possible ways of deriving predictions from a spectrogram
input, so this function provides an abstract interface for creating different
kinds of models in a black-box way. You need to pass in a TensorFlow node as
the 'fingerprint' input, and this should output a batch of 1D features that
describe the audio. Typically this will be derived from a spectrogram that's
been run through an MFCC, but in theory it can be any feature vector of the
size specified in model_settings['fingerprint_size'].
The function will build the graph it needs in the current TensorFlow graph,
and return the tensorflow output that will contain the 'logits' input to the
softmax prediction process. If training flag is on, it will also return a
placeholder node that can be used to control the dropout amount.
See the implementations below for the possible model architectures that can be
requested.
Args:
fingerprint_input: TensorFlow node that will output audio feature vectors.
model_settings: Dictionary of information about the model.
model_architecture: String specifying which kind of model to create.
is_training: Whether the model is going to be used for training.
runtime_settings: Dictionary of information about the runtime.
Returns:
TensorFlow node outputting logits results, and optionally a dropout
placeholder.
Raises:
Exception: If the architecture type isn't recognized."
268,load_variables_from_checkpoint,tensorflow/tensorflow/examples/speech_commands/models.py,153,function,"Utility function to centralize checkpoint restoration.
Args:
sess: TensorFlow session.
start_checkpoint: Path to saved checkpoint on disk."
269,create_single_fc_model,tensorflow/tensorflow/examples/speech_commands/models.py,164,function,"Builds a model with a single hidden fully-connected layer.
This is a very simple model with just one matmul and bias layer. As you'd
expect, it doesn't produce very accurate results, but it is very fast and
simple, so it's useful for sanity testing.
Here's the layout of the graph:
(fingerprint_input)
v
[MatMul]<-(weights)
v
[BiasAdd]<-(bias)
v
Args:
fingerprint_input: TensorFlow node that will output audio feature vectors.
model_settings: Dictionary of information about the model.
is_training: Whether the model is going to be used for training.
Returns:
TensorFlow node outputting logits results, and optionally a dropout
placeholder."
270,create_conv_model,tensorflow/tensorflow/examples/speech_commands/models.py,207,function,"Builds a standard convolutional model.
This is roughly the network labeled as 'cnn-trad-fpool3' in the
'Convolutional Neural Networks for Small-footprint Keyword Spotting' paper:
http://www.isca-speech.org/archive/interspeech_2015/papers/i15_1478.pdf
Here's the layout of the graph:
(fingerprint_input)
v
[Conv2D]<-(weights)
v
[BiasAdd]<-(bias)
v
[Relu]
v
[MaxPool]
v
[Conv2D]<-(weights)
v
[BiasAdd]<-(bias)
v
[Relu]
v
[MaxPool]
v
[MatMul]<-(weights)
v
[BiasAdd]<-(bias)
v
This produces fairly good quality results, but can involve a large number of
weight parameters and computations. For a cheaper alternative from the same
paper with slightly less accuracy, see 'low_latency_conv' below.
During training, dropout nodes are introduced after each relu, controlled by a
placeholder.
Args:
fingerprint_input: TensorFlow node that will output audio feature vectors.
model_settings: Dictionary of information about the model.
is_training: Whether the model is going to be used for training.
Returns:
TensorFlow node outputting logits results, and optionally a dropout
placeholder."
271,create_low_latency_conv_model,tensorflow/tensorflow/examples/speech_commands/models.py,333,function,"Builds a convolutional model with low compute requirements.
This is roughly the network labeled as 'cnn-one-fstride4' in the
'Convolutional Neural Networks for Small-footprint Keyword Spotting' paper:
http://www.isca-speech.org/archive/interspeech_2015/papers/i15_1478.pdf
Here's the layout of the graph:
(fingerprint_input)
v
[Conv2D]<-(weights)
v
[BiasAdd]<-(bias)
v
[Relu]
v
[MatMul]<-(weights)
v
[BiasAdd]<-(bias)
v
[MatMul]<-(weights)
v
[BiasAdd]<-(bias)
v
[MatMul]<-(weights)
v
[BiasAdd]<-(bias)
v
This produces slightly lower quality results than the 'conv' model, but needs
fewer weight parameters and computations.
During training, dropout nodes are introduced after the relu, controlled by a
placeholder.
Args:
fingerprint_input: TensorFlow node that will output audio feature vectors.
model_settings: Dictionary of information about the model.
is_training: Whether the model is going to be used for training.
Returns:
TensorFlow node outputting logits results, and optionally a dropout
placeholder."
272,create_low_latency_svdf_model,tensorflow/tensorflow/examples/speech_commands/models.py,462,function,"Builds an SVDF model with low compute requirements.
This is based in the topology presented in the 'Compressing Deep Neural
Networks using a Rank-Constrained Topology' paper:
https://static.googleusercontent.com/media/research.google.com/en//pubs/archive/43813.pdf
Here's the layout of the graph:
(fingerprint_input)
v
[SVDF]<-(weights)
v
[BiasAdd]<-(bias)
v
[Relu]
v
[MatMul]<-(weights)
v
[BiasAdd]<-(bias)
v
[MatMul]<-(weights)
v
[BiasAdd]<-(bias)
v
[MatMul]<-(weights)
v
[BiasAdd]<-(bias)
v
This model produces lower recognition accuracy than the 'conv' model above,
but requires fewer weight parameters and, significantly fewer computations.
During training, dropout nodes are introduced after the relu, controlled by a
placeholder.
Args:
fingerprint_input: TensorFlow node that will output audio feature vectors.
The node is expected to produce a 2D Tensor of shape:
[batch, model_settings['fingerprint_width'] *
model_settings['spectrogram_length']]
with the features corresponding to the same time slot arranged contiguously,
and the oldest slot at index [:, 0], and newest at [:, -1].
model_settings: Dictionary of information about the model.
is_training: Whether the model is going to be used for training.
runtime_settings: Dictionary of information about the runtime.
Returns:
TensorFlow node outputting logits results, and optionally a dropout
placeholder.
Raises:
ValueError: If the inputs tensor is incorrectly shaped."
273,create_tiny_conv_model,tensorflow/tensorflow/examples/speech_commands/models.py,673,function,"Builds a convolutional model aimed at microcontrollers.
Devices like DSPs and microcontrollers can have very small amounts of
memory and limited processing power. This model is designed to use less
than 20KB of working RAM, and fit within 32KB of read-only (flash) memory.
Here's the layout of the graph:
(fingerprint_input)
v
[Conv2D]<-(weights)
v
[BiasAdd]<-(bias)
v
[Relu]
v
[MatMul]<-(weights)
v
[BiasAdd]<-(bias)
v
This doesn't produce particularly accurate results, but it's designed to be
used as the first stage of a pipeline, running on a low-energy piece of
hardware that can always be on, and then wake higher-power chips when a
possible utterance has been found, so that more accurate analysis can be done.
During training, a dropout node is introduced after the relu, controlled by a
placeholder.
Args:
fingerprint_input: TensorFlow node that will output audio feature vectors.
model_settings: Dictionary of information about the model.
is_training: Whether the model is going to be used for training.
Returns:
TensorFlow node outputting logits results, and optionally a dropout
placeholder."
274,create_tiny_embedding_conv_model,tensorflow/tensorflow/examples/speech_commands/models.py,765,function,"Builds a convolutional model aimed at microcontrollers.
Devices like DSPs and microcontrollers can have very small amounts of
memory and limited processing power. This model is designed to use less
than 20KB of working RAM, and fit within 32KB of read-only (flash) memory.
Here's the layout of the graph:
(fingerprint_input)
v
[Conv2D]<-(weights)
v
[BiasAdd]<-(bias)
v
[Relu]
v
[Conv2D]<-(weights)
v
[BiasAdd]<-(bias)
v
[Relu]
v
[Conv2D]<-(weights)
v
[BiasAdd]<-(bias)
v
[Relu]
v
[MatMul]<-(weights)
v
[BiasAdd]<-(bias)
v
This doesn't produce particularly accurate results, but it's designed to be
used as the first stage of a pipeline, running on a low-energy piece of
hardware that can always be on, and then wake higher-power chips when a
possible utterance has been found, so that more accurate analysis can be done.
During training, a dropout node is introduced after the relu, controlled by a
placeholder.
Args:
fingerprint_input: TensorFlow node that will output audio feature vectors.
model_settings: Dictionary of information about the model.
is_training: Whether the model is going to be used for training.
Returns:
TensorFlow node outputting logits results, and optionally a dropout
placeholder."
275,RecognizeResult,tensorflow/tensorflow/examples/speech_commands/recognize_commands.py,25,class,"Save recognition result temporarily.
Attributes:
founded_command: A string indicating the word just founded. Default value
is '_silence_'
score: An float representing the confidence of founded word. Default
value is zero.
is_new_command: A boolean indicating if the founded command is a new one
against the last one. Default value is False."
276,founded_command,tensorflow/tensorflow/examples/speech_commands/recognize_commands.py,43,method,
277,founded_command,tensorflow/tensorflow/examples/speech_commands/recognize_commands.py,47,method,
278,score,tensorflow/tensorflow/examples/speech_commands/recognize_commands.py,51,method,
279,score,tensorflow/tensorflow/examples/speech_commands/recognize_commands.py,55,method,
280,is_new_command,tensorflow/tensorflow/examples/speech_commands/recognize_commands.py,59,method,
281,is_new_command,tensorflow/tensorflow/examples/speech_commands/recognize_commands.py,63,method,
282,RecognizeCommands,tensorflow/tensorflow/examples/speech_commands/recognize_commands.py,67,class,"Smooth the inference results by using average window.
Maintain a slide window over the audio stream, which adds new result(a pair of
the 1.confidences of all classes and 2.the start timestamp of input audio
clip) directly the inference produces one and removes the most previous one
and other abnormal values. Then it smooth the results in the window to get
the most reliable command in this period.
Attributes:
_label: A list containing commands at corresponding lines.
_average_window_duration: The length of average window.
_detection_threshold: A confidence threshold for filtering out unreliable
command.
_suppression_ms: Milliseconds every two reliable founded commands should
apart.
_minimum_count: An integer count indicating the minimum results the average
window should cover.
_previous_results: A deque to store previous results.
_label_count: The length of label list.
_previous_top_label: Last founded command. Initial value is '_silence_'.
_previous_top_time: The timestamp of _previous results. Default is -np.inf."
283,load_graph,tensorflow/tensorflow/examples/speech_commands/test_streaming_accuracy.py,80,function,"Read a tensorflow model, and creates a default graph object."
284,read_label_file,tensorflow/tensorflow/examples/speech_commands/test_streaming_accuracy.py,92,function,Load a list of label.
285,read_wav_file,tensorflow/tensorflow/examples/speech_commands/test_streaming_accuracy.py,101,function,Load a wav file and return sample_rate and numpy data of float64 type.
286,verbosity_arg,tensorflow/tensorflow/examples/speech_commands/train.py,480,function,"Parses verbosity argument.
Args:
value: A member of tf.logging.
Raises:
ArgumentTypeError: Not an expected value."
287,requires_contrib,tensorflow/tensorflow/examples/speech_commands/train_test.py,32,function,
288,DictStruct,tensorflow/tensorflow/examples/speech_commands/train_test.py,44,class,
289,wav_to_features,tensorflow/tensorflow/examples/speech_commands/wav_to_features.py,47,function,"Converts an audio file into its corresponding feature map.
Args:
sample_rate: Expected sample rate of the wavs.
clip_duration_ms: Expected duration in milliseconds of the wavs.
window_size_ms: How long each spectrogram timeslice is.
window_stride_ms: How far to move in time between spectrogram timeslices.
feature_bin_count: How many bins to use for the feature fingerprint.
quantize: Whether to train the model for eight-bit deployment.
preprocess: Spectrogram processing mode; ""mfcc"", ""average"" or ""micro"".
input_wav: Path to the audio WAV file to read.
output_c_file: Where to save the generated C source file."
290,create_model,tensorflow/tensorflow/examples/tf2_showcase/mnist.py,69,function,"Model to recognize digits in the MNIST dataset.
Network structure is equivalent to:
https://github.com/tensorflow/tensorflow/blob/r1.5/tensorflow/examples/tutorials/mnist/mnist_deep.py
and
https://github.com/tensorflow/models/blob/master/tutorials/image/mnist/convolutional.py
But uses the tf.keras API.
Returns:
A tf.keras.Model."
291,mnist_datasets,tensorflow/tensorflow/examples/tf2_showcase/mnist.py,115,function,
292,loss,tensorflow/tensorflow/examples/tf2_showcase/mnist.py,125,function,
293,compute_accuracy,tensorflow/tensorflow/examples/tf2_showcase/mnist.py,131,function,
294,train,tensorflow/tensorflow/examples/tf2_showcase/mnist.py,140,function,Trains model on `dataset` using `optimizer`.
295,train_and_export,tensorflow/tensorflow/examples/tf2_showcase/mnist.py,184,function,"Run MNIST training and eval loop in eager mode.
Args:
flags_obj: An object containing parsed flag values."
296,import_and_eval,tensorflow/tensorflow/examples/tf2_showcase/mnist.py,237,function,
297,apply_clean,tensorflow/tensorflow/examples/tf2_showcase/mnist.py,247,function,
298,placeholder_inputs,tensorflow/tensorflow/examples/tutorials/mnist/fully_connected_feed.py,37,function,"Generate placeholder variables to represent the input tensors.
These placeholders are used as inputs by the rest of the model building
code and will be fed from the downloaded data in the .run() loop, below.
Args:
batch_size: The batch size will be baked into both placeholders.
Returns:
images_placeholder: Images placeholder.
labels_placeholder: Labels placeholder."
299,fill_feed_dict,tensorflow/tensorflow/examples/tutorials/mnist/fully_connected_feed.py,59,function,"Fills the feed_dict for training the given step.
A feed_dict takes the form of:
feed_dict = {
<placeholder>: <tensor of values to be passed for placeholder>,
....
}
Args:
data_set: The set of images and labels, from input_data.read_data_sets()
images_pl: The images placeholder, from placeholder_inputs().
labels_pl: The labels placeholder, from placeholder_inputs().
Returns:
feed_dict: The feed dictionary mapping from placeholders to values."
300,do_eval,tensorflow/tensorflow/examples/tutorials/mnist/fully_connected_feed.py,87,function,"Runs one evaluation against the full epoch of data.
Args:
sess: The session in which the model has been trained.
eval_correct: The Tensor that returns the number of correct predictions.
images_placeholder: The images placeholder.
labels_placeholder: The labels placeholder.
data_set: The set of images and labels to evaluate, from
input_data.read_data_sets()."
301,run_training,tensorflow/tensorflow/examples/tutorials/mnist/fully_connected_feed.py,116,function,Train MNIST for a number of steps.
302,read_data_sets,tensorflow/tensorflow/examples/tutorials/mnist/input_data.py,266,function,
303,inference,tensorflow/tensorflow/examples/tutorials/mnist/mnist.py,45,function,"Build the MNIST model up to where it may be used for inference.
Args:
images: Images placeholder, from inputs().
hidden1_units: Size of the first hidden layer.
hidden2_units: Size of the second hidden layer.
Returns:
softmax_linear: Output tensor with the computed logits."
304,loss,tensorflow/tensorflow/examples/tutorials/mnist/mnist.py,86,function,"Calculates the loss from the logits and the labels.
Args:
logits: Logits tensor, float - [batch_size, NUM_CLASSES].
labels: Labels tensor, int32 - [batch_size].
Returns:
loss: Loss tensor of type float."
305,training,tensorflow/tensorflow/examples/tutorials/mnist/mnist.py,101,function,"Sets up the training Ops.
Creates a summarizer to track the loss over time in TensorBoard.
Creates an optimizer and applies the gradients to all trainable variables.
The Op returned by this function is what must be passed to the
`sess.run()` call to cause the model to train.
Args:
loss: Loss tensor, from loss().
learning_rate: The learning rate to use for gradient descent.
Returns:
train_op: The Op for training."
306,evaluation,tensorflow/tensorflow/examples/tutorials/mnist/mnist.py,130,function,"Evaluate the quality of the logits at predicting the label.
Args:
logits: Logits tensor, float - [batch_size, NUM_CLASSES].
labels: Labels tensor, int32 - [batch_size], with values in the
range [0, NUM_CLASSES).
Returns:
A scalar int32 tensor with the number of examples (out of batch_size)
that were predicted correctly."
307,train,tensorflow/tensorflow/examples/tutorials/mnist/mnist_with_summaries.py,38,function,
308,word2vec_basic,tensorflow/tensorflow/examples/tutorials/word2vec/word2vec_basic.py,49,function,"Example of building, training and visualizing a word2vec model."
309,suppress_exception,tensorflow/tensorflow/lite/examples/experimental_new_converter/stack_trace_example.py,37,function,
310,load_labels,tensorflow/tensorflow/lite/examples/python/label_image.py,29,function,
311,dynamic_rnn,tensorflow/tensorflow/lite/experimental/examples/lstm/rnn.py,42,function,"Creates a recurrent neural network specified by RNNCell `cell`.
Performs fully dynamic unrolling of `inputs`.
Example:
```python
# create a BasicRNNCell
rnn_cell = tf.compat.v1.nn.rnn_cell.BasicRNNCell(hidden_size)
# 'outputs' is a tensor of shape [batch_size, max_time, cell_state_size]
# defining initial state
initial_state = rnn_cell.zero_state(batch_size, dtype=tf.float32)
# 'state' is a tensor of shape [batch_size, cell_state_size]
outputs, state = tf.compat.v1.nn.dynamic_rnn(rnn_cell, input_data,
initial_state=initial_state,
dtype=tf.float32)
```
```python
# create 2 LSTMCells
rnn_layers = [tf.compat.v1.nn.rnn_cell.LSTMCell(size) for size in [128, 256]]
# create a RNN cell composed sequentially of a number of RNNCells
multi_rnn_cell = tf.compat.v1.nn.rnn_cell.MultiRNNCell(rnn_layers)
# 'outputs' is a tensor of shape [batch_size, max_time, 256]
# 'state' is a N-tuple where N is the number of LSTMCells containing a
# tf.nn.rnn_cell.LSTMStateTuple for each cell
outputs, state = tf.compat.v1.nn.dynamic_rnn(cell=multi_rnn_cell,
inputs=data,
dtype=tf.float32)
```
Args:
cell: An instance of RNNCell.
inputs: The RNN inputs.
If `time_major == False` (default), this must be a `Tensor` of shape:
`[batch_size, max_time, ...]`, or a nested tuple of such elements.
If `time_major == True`, this must be a `Tensor` of shape: `[max_time,
batch_size, ...]`, or a nested tuple of such elements. This may also be
a (possibly nested) tuple of Tensors satisfying this property. The
first two dimensions must match across all the inputs, but otherwise the
ranks and other shape components may differ. In this case, input to
`cell` at each time-step will replicate the structure of these tuples,
except for the time dimension (from which the time is taken). The input
to `cell` at each time step will be a `Tensor` or (possibly nested)
tuple of Tensors each with dimensions `[batch_size, ...]`.
sequence_length: (optional) An int32/int64 vector sized `[batch_size]`. Used
to copy-through state and zero-out outputs when past a batch element's
sequence length. So it's more for performance than correctness.
initial_state: (optional) An initial state for the RNN. If `cell.state_size`
is an integer, this must be a `Tensor` of appropriate type and shape
`[batch_size, cell.state_size]`. If `cell.state_size` is a tuple, this
should be a tuple of tensors having shapes `[batch_size, s] for s in
cell.state_size`.
dtype: (optional) The data type for the initial state and expected output.
Required if initial_state is not provided or RNN state has a heterogeneous
dtype.
parallel_iterations: (Default: 32). The number of iterations to run in
parallel. Those operations which do not have any temporal dependency and
can be run in parallel, will be. This parameter trades off time for
space. Values >> 1 use more memory but take less time, while smaller
values use less memory but computations take longer.
swap_memory: Transparently swap the tensors produced in forward inference
but needed for back prop from GPU to CPU. This allows training RNNs which
would typically not fit on a single GPU, with very minimal (or no)
performance penalty.
time_major: The shape format of the `inputs` and `outputs` Tensors. If true,
these `Tensors` must be shaped `[max_time, batch_size, depth]`. If false,
these `Tensors` must be shaped `[batch_size, max_time, depth]`. Using
`time_major = True` is a bit more efficient because it avoids transposes
at the beginning and end of the RNN calculation. However, most TensorFlow
data is batch-major, so by default this function accepts input and emits
output in batch-major form.
scope: VariableScope for the created subgraph; defaults to ""rnn"".
Returns:
A pair (outputs, state) where:
outputs: The RNN output `Tensor`.
If time_major == False (default), this will be a `Tensor` shaped:
`[batch_size, max_time, cell.output_size]`.
If time_major == True, this will be a `Tensor` shaped:
`[max_time, batch_size, cell.output_size]`.
Note, if `cell.output_size` is a (possibly nested) tuple of integers
or `TensorShape` objects, then `outputs` will be a tuple having the
same structure as `cell.output_size`, containing Tensors having shapes
corresponding to the shape data in `cell.output_size`.
state: The final state. If `cell.state_size` is an int, this
will be shaped `[batch_size, cell.state_size]`. If it is a
`TensorShape`, this will be shaped `[batch_size] + cell.state_size`.
If it is a (possibly nested) tuple of ints or `TensorShape`, this will
be a tuple having the corresponding shapes. If cells are `LSTMCells`
`state` will be a tuple containing a `LSTMStateTuple` for each cell.
Raises:
TypeError: If `cell` is not an instance of RNNCell.
ValueError: If inputs is None or an empty list.
RuntimeError: If not using control flow v2."
312,bidirectional_dynamic_rnn,tensorflow/tensorflow/lite/experimental/examples/lstm/rnn.py,279,function,"Creates a dynamic version of bidirectional recurrent neural network.
Takes input and builds independent forward and backward RNNs. The input_size
of forward and backward cell must match. The initial state for both directions
is zero by default (but can be set optionally) and no intermediate states are
ever returned -- the network is fully unrolled for the given (passed in)
length(s) of the sequence(s) or completely unrolled if length(s) is not
given.
Args:
cell_fw: An instance of RNNCell, to be used for forward direction.
cell_bw: An instance of RNNCell, to be used for backward direction.
inputs: The RNN inputs.
If time_major == False (default), this must be a tensor of shape:
`[batch_size, max_time, ...]`, or a nested tuple of such elements.
If time_major == True, this must be a tensor of shape: `[max_time,
batch_size, ...]`, or a nested tuple of such elements.
sequence_length: (optional) An int32/int64 vector, size `[batch_size]`,
containing the actual lengths for each of the sequences in the batch. If
not provided, all batch entries are assumed to be full sequences; and time
reversal is applied from time `0` to `max_time` for each sequence.
initial_state_fw: (optional) An initial state for the forward RNN. This must
be a tensor of appropriate type and shape `[batch_size,
cell_fw.state_size]`. If `cell_fw.state_size` is a tuple, this should be a
tuple of tensors having shapes `[batch_size, s] for s in
cell_fw.state_size`.
initial_state_bw: (optional) Same as for `initial_state_fw`, but using the
corresponding properties of `cell_bw`.
dtype: (optional) The data type for the initial states and expected output.
Required if initial_states are not provided or RNN states have a
heterogeneous dtype.
parallel_iterations: (Default: 32). The number of iterations to run in
parallel. Those operations which do not have any temporal dependency and
can be run in parallel, will be. This parameter trades off time for
space. Values >> 1 use more memory but take less time, while smaller
values use less memory but computations take longer.
swap_memory: Transparently swap the tensors produced in forward inference
but needed for back prop from GPU to CPU. This allows training RNNs which
would typically not fit on a single GPU, with very minimal (or no)
performance penalty.
time_major: The shape format of the `inputs` and `outputs` Tensors. If true,
these `Tensors` must be shaped `[max_time, batch_size, depth]`. If false,
these `Tensors` must be shaped `[batch_size, max_time, depth]`. Using
`time_major = True` is a bit more efficient because it avoids transposes
at the beginning and end of the RNN calculation. However, most TensorFlow
data is batch-major, so by default this function accepts input and emits
output in batch-major form.
scope: VariableScope for the created subgraph; defaults to
""bidirectional_rnn""
Returns:
A tuple (outputs, output_states) where:
outputs: A tuple (output_fw, output_bw) containing the forward and
the backward rnn output `Tensor`.
If time_major == False (default),
output_fw will be a `Tensor` shaped:
`[batch_size, max_time, cell_fw.output_size]`
and output_bw will be a `Tensor` shaped:
`[batch_size, max_time, cell_bw.output_size]`.
If time_major == True,
output_fw will be a `Tensor` shaped:
`[max_time, batch_size, cell_fw.output_size]`
and output_bw will be a `Tensor` shaped:
`[max_time, batch_size, cell_bw.output_size]`.
It returns a tuple instead of a single concatenated `Tensor`, unlike
in the `bidirectional_rnn`. If the concatenated one is preferred,
the forward and backward outputs can be concatenated as
`tf.concat(outputs, 2)`.
output_states: A tuple (output_state_fw, output_state_bw) containing
the forward and the backward final states of bidirectional rnn.
Raises:
TypeError: If `cell_fw` or `cell_bw` is not an instance of `RNNCell`."
313,TfLiteRNNCell,tensorflow/tensorflow/lite/experimental/examples/lstm/rnn_cell.py,39,class,"The most basic RNN cell.
This is used only for TfLite, it provides hints and it also makes the
variables in the desired for the tflite ops."
314,state_size,tensorflow/tensorflow/lite/experimental/examples/lstm/rnn_cell.py,88,method,
315,output_size,tensorflow/tensorflow/lite/experimental/examples/lstm/rnn_cell.py,92,method,
316,build,tensorflow/tensorflow/lite/experimental/examples/lstm/rnn_cell.py,95,method,"Builds the RNN cell.
Args:
inputs_shape: Rnn input tensor shape.
Raises:
ValueError: If last dimension of the input shape is not known."
317,call,tensorflow/tensorflow/lite/experimental/examples/lstm/rnn_cell.py,127,method,Most basic RNN: output = new_state = act(W * input + U * state + B).
318,get_config,tensorflow/tensorflow/lite/experimental/examples/lstm/rnn_cell.py,150,method,
319,add_variable_wrapped,tensorflow/tensorflow/lite/experimental/examples/lstm/rnn_cell.py,110,method,
320,TFLiteLSTMCell,tensorflow/tensorflow/lite/experimental/examples/lstm/rnn_cell.py,162,class,"Long short-term memory unit (LSTM) recurrent network cell.
This is used only for TfLite, it provides hints and it also makes the
variables in the desired for the tflite ops (transposed and separated).
The default non-peephole implementation is based on:
https://pdfs.semanticscholar.org/1154/0131eae85b2e11d53df7f1360eeb6476e7f4.pdf
Felix Gers, Jurgen Schmidhuber, and Fred Cummins.
""Learning to forget: Continual prediction with LSTM."" IET, 850-855, 1999.
The peephole implementation is based on:
https://research.google.com/pubs/archive/43905.pdf
Hasim Sak, Andrew Senior, and Francoise Beaufays.
""Long short-term memory recurrent neural network architectures for
large scale acoustic modeling."" INTERSPEECH, 2014.
The class uses optional peep-hole connections, optional cell clipping, and
an optional projection layer.
Note that this cell is not optimized for performance. Please use
`tf.contrib.cudnn_rnn.CudnnLSTM` for better performance on GPU, or
`tf.contrib.rnn.LSTMBlockCell` and `tf.contrib.rnn.LSTMBlockFusedCell` for
better performance on CPU."
321,state_size,tensorflow/tensorflow/lite/experimental/examples/lstm/rnn_cell.py,284,method,
322,output_size,tensorflow/tensorflow/lite/experimental/examples/lstm/rnn_cell.py,288,method,
323,build,tensorflow/tensorflow/lite/experimental/examples/lstm/rnn_cell.py,291,method,"Build TfLite LSTM cell graph.
Args:
inputs_shape: The inputs_shape must be known, and is [batch_size,
input_size] shape.
Raises:
ValueError: if the inputs_shape is invalid."
324,call,tensorflow/tensorflow/lite/experimental/examples/lstm/rnn_cell.py,392,method,"Run one step of LSTM.
Args:
inputs: input Tensor, 2D, `[batch, num_units]`.
state: if `state_is_tuple` is False, this must be a state Tensor, `2-D,
[batch, state_size]`. If `state_is_tuple` is True, this must be a tuple
of state Tensors, both `2-D`, with column sizes `c_state` and `m_state`.
Returns:
A tuple containing:
- A `2-D, [batch, output_dim]`, Tensor representing the output of the
LSTM after reading `inputs` when previous state was `state`.
Here output_dim is:
num_proj if num_proj was set,
num_units otherwise.
- Tensor(s) representing the new state of LSTM after reading `inputs` when
the previous state was `state`. Same type and shape(s) as `state`.
Raises:
ValueError: If input size cannot be inferred from inputs via
static shape inference."
325,get_config,tensorflow/tensorflow/lite/experimental/examples/lstm/rnn_cell.py,519,method,
326,add_variable_wrapped,tensorflow/tensorflow/lite/experimental/examples/lstm/rnn_cell.py,317,method,
327,audio_microfrontend,tensorflow/tensorflow/lite/experimental/microfrontend/python/ops/audio_microfrontend_op.py,34,function,"Audio Microfrontend Op.
This Op converts a sequence of audio data into one or more
feature vectors containing filterbanks of the input. The
conversion process uses a lightweight library to perform:
1. A slicing window function
2. Short-time FFTs
3. Filterbank calculations
4. Noise reduction
5. PCAN Auto Gain Control
6. Logarithmic scaling
Args:
audio: 1D Tensor, int16 audio data in temporal ordering.
sample_rate: Integer, the sample rate of the audio in Hz.
window_size: Integer, length of desired time frames in ms.
window_step: Integer, length of step size for the next frame in ms.
num_channels: Integer, the number of filterbank channels to use.
upper_band_limit: Float, the highest frequency included in the filterbanks.
lower_band_limit: Float, the lowest frequency included in the filterbanks.
smoothing_bits: Int, scale up signal by 2^(smoothing_bits) before reduction.
even_smoothing: Float, smoothing coefficient for even-numbered channels.
odd_smoothing: Float, smoothing coefficient for odd-numbered channels.
min_signal_remaining: Float, fraction of signal to preserve in smoothing.
enable_pcan: Bool, enable PCAN auto gain control.
pcan_strength: Float, gain normalization exponent.
pcan_offset: Float, positive value added in the normalization denominator.
gain_bits: Int, number of fractional bits in the gain.
enable_log: Bool, enable logarithmic scaling of filterbanks.
scale_shift: Integer, scale filterbanks by 2^(scale_shift).
left_context: Integer, number of preceding frames to attach to each frame.
right_context: Integer, number of preceding frames to attach to each frame.
frame_stride: Integer, M frames to skip over, where output[n] = frame[n*M].
zero_padding: Bool, if left/right context is out-of-bounds, attach frame of
zeroes. Otherwise, frame[0] or frame[size-1] will be copied.
out_scale: Integer, divide all filterbanks by this number.
out_type: DType, type of the output Tensor, defaults to UINT16.
Returns:
filterbanks: 2D Tensor, each row is a time frame, each column is a channel.
Raises:
ValueError: If the audio tensor is not explicitly a vector."
328,SupportedOp,tensorflow/tensorflow/lite/experimental/tensorboard/ops_util.py,26,class,"Spec of supported ops.
Args:
op: string of op name."
329,get_potentially_supported_ops,tensorflow/tensorflow/lite/experimental/tensorboard/ops_util.py,35,function,"Returns operations potentially supported by TensorFlow Lite.
The potentially support list contains a list of ops that are partially or
fully supported, which is derived by simply scanning op names to check whether
they can be handled without real conversion and specific parameters.
Given that some ops may be partially supported, the optimal way to determine
if a model's operations are supported is by converting using the TensorFlow
Lite converter.
Returns:
A list of SupportedOp."
330,time_wrapping,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/data_augmentation.py,29,function,Generate (molecule/denominator)x speed data.
331,augment_data,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/data_augmentation.py,43,function,Perform data augmentation.
332,DataLoader,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/data_load.py,35,class,Loads data and prepares for training.
333,get_data_file,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/data_load.py,50,method,"Get train, valid and test data from files."
334,pad,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/data_load.py,66,method,Get neighbour padding.
335,format_support_func,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/data_load.py,81,method,"Support function for format.(Helps format train, valid and test.)"
336,format,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/data_load.py,98,method,"Format data(including padding, etc.) and get the dataset for the model."
337,prepare_original_data,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/data_prepare.py,46,function,Read collected data from files.
338,generate_negative_data,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/data_prepare.py,86,function,Generate negative data labeled as 'negative6~8'.
339,write_data,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/data_prepare.py,143,function,
340,read_data,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/data_split.py,40,function,
341,split_data,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/data_split.py,51,function,"Splits data into train, validation and test according to ratio."
342,person_split,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/data_split_person.py,41,function,Split data by person.
343,reshape_function,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/train.py,37,function,
344,calculate_model_size,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/train.py,42,function,
345,build_cnn,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/train.py,51,function,Builds a convolutional neural network in Keras.
346,build_lstm,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/train.py,78,function,Builds an LSTM in Keras.
347,load_data,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/train.py,93,function,
348,build_net,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/train.py,101,function,
349,train_net,tensorflow/tensorflow/lite/micro/examples/magic_wand/train/train.py,111,function,Trains the model.
350,to_cc,tensorflow/tensorflow/lite/micro/examples/micro_speech/CMSIS/create_constants.py,26,function,Writes table values to a C++ source file.
351,to_h,tensorflow/tensorflow/lite/micro/examples/micro_speech/CMSIS/create_constants.py,44,function,Writes a header file for the table values.
352,new_data_to_array,tensorflow/tensorflow/lite/micro/examples/micro_speech/apollo3/captured_data_to_wav.py,28,function,
353,new_data_to_array,tensorflow/tensorflow/lite/micro/examples/micro_speech/apollo3/compare_1k.py,29,function,Converts file information to an in-memory array.
354,to_float,tensorflow/tensorflow/lite/micro/examples/micro_speech/apollo3/compare_1k.py,63,function,
355,check_file_existence,tensorflow/tensorflow/lite/micro/examples/person_detection/utils/raw_to_bitmap.py,52,function,
356,show_and_save_bitmaps,tensorflow/tensorflow/lite/micro/examples/person_detection/utils/raw_to_bitmap.py,60,function,"Display and save a list of bitmaps.
Args:
input_file: input file name
bitmap_list: list of numpy arrays to represent bitmap images
channels: color channel count"
357,reshape_bitmaps,tensorflow/tensorflow/lite/micro/examples/person_detection/utils/raw_to_bitmap.py,87,function,"Reshape flat integer arrays.
Args:
frame_list: list of 1-D arrays to represent raw image data
width: image width in pixels
height: image height in pixels
channels: color channel count
Returns:
list of numpy arrays to represent bitmap images"
358,parse_file,tensorflow/tensorflow/lite/micro/examples/person_detection/utils/raw_to_bitmap.py,109,function,"Convert log file to array of pixels.
Args:
inputfile: log file to parse
width: image width in pixels
height: image height in pixels
channels: color channel count
Returns:
list 1-D arrays to represent raw image data."
359,generate_conv_model,tensorflow/tensorflow/lite/micro/testing/generate_test_models.py,34,function,"Creates a basic Keras model and converts to tflite.
This model does not make any relevant classifications. It only exists to
generate a model that is designed to run on embedded devices."
360,rename_example_subfolder_files,tensorflow/tensorflow/lite/micro/tools/make/fix_arduino_subfolders.py,29,function,Moves source files in example subfolders to equivalents at root.
361,move_person_data,tensorflow/tensorflow/lite/micro/tools/make/fix_arduino_subfolders.py,41,function,Moves the downloaded person model into the examples folder.
362,move_person_data_experimental,tensorflow/tensorflow/lite/micro/tools/make/fix_arduino_subfolders.py,61,function,Moves the downloaded person model into the examples folder.
363,move_image_data_experimental,tensorflow/tensorflow/lite/micro/tools/make/fix_arduino_subfolders.py,83,function,Moves the downloaded image detection model into the examples folder.
364,parse_args,tensorflow/tensorflow/lite/micro/tools/make/fix_arduino_subfolders.py,124,function,Converts the raw arguments into accessible flags.
365,sanitize_xml,tensorflow/tensorflow/lite/micro/tools/make/generate_keil_project.py,29,function,Uses a allowlist to avoid generating bad XML.
366,parse_args,tensorflow/tensorflow/lite/micro/tools/make/generate_keil_project.py,82,function,Converts the raw arguments into accessible flags.
367,parse_args,tensorflow/tensorflow/lite/micro/tools/make/merge_arduino_zips.py,39,function,Converts the raw arguments into accessible flags.
368,replace_includes,tensorflow/tensorflow/lite/micro/tools/make/transform_arduino_source.py,29,function,Updates any includes to reference the new Arduino library paths.
369,check_ino_functions,tensorflow/tensorflow/lite/micro/tools/make/transform_arduino_source.py,51,function,Ensures the required functions exist.
370,add_example_ino_library_include,tensorflow/tensorflow/lite/micro/tools/make/transform_arduino_source.py,65,function,Makes sure the example includes the header that loads the library.
371,replace_example_includes,tensorflow/tensorflow/lite/micro/tools/make/transform_arduino_source.py,71,function,Updates any includes for local example files.
372,parse_args,tensorflow/tensorflow/lite/micro/tools/make/transform_arduino_source.py,108,function,Converts the raw arguments into accessible flags.
373,replace_arduino_includes,tensorflow/tensorflow/lite/micro/tools/make/transform_source.py,36,function,Updates any includes to reference the new Arduino library paths.
374,check_ino_functions,tensorflow/tensorflow/lite/micro/tools/make/transform_source.py,58,function,Ensures the required functions exist.
375,add_example_ino_library_include,tensorflow/tensorflow/lite/micro/tools/make/transform_source.py,72,function,Makes sure the example includes the header that loads the library.
376,replace_arduino_example_includes,tensorflow/tensorflow/lite/micro/tools/make/transform_source.py,78,function,Updates any includes for local example files.
377,replace_esp_example_includes,tensorflow/tensorflow/lite/micro/tools/make/transform_source.py,92,function,Updates any includes for local example files.
378,transform_arduino_sources,tensorflow/tensorflow/lite/micro/tools/make/transform_source.py,109,function,"Transform sources for the Arduino platform.
Args:
input_lines: A sequence of lines from the input file to process.
flags: Flags indicating which transformation(s) to apply.
Returns:
The transformed output as a string."
379,transform_esp_sources,tensorflow/tensorflow/lite/micro/tools/make/transform_source.py,138,function,"Transform sources for the ESP-IDF platform.
Args:
input_lines: A sequence of lines from the input file to process.
flags: Flags indicating which transformation(s) to apply.
Returns:
The transformed output as a string."
380,parse_args,tensorflow/tensorflow/lite/micro/tools/make/transform_source.py,171,function,Converts the raw arguments into accessible flags.
381,OpsSet,tensorflow/tensorflow/lite/python/convert.py,80,class,"Enum class defining the sets of ops available to generate TFLite models.
WARNING: Experimental interface, subject to change."
382,get_options,tensorflow/tensorflow/lite/python/convert.py,115,method,Returns a list of OpsSet options as a list of strings.
383,ConverterError,tensorflow/tensorflow/lite/python/convert.py,120,class,Raised when an error occurs during model conversion.
384,mlir_quantize,tensorflow/tensorflow/lite/python/convert.py,125,function,"Quantize `input_data_str` with calibration results.
Args:
input_data_str: Input data in serialized form (e.g. a TFLITE model with
calibration results).
disable_per_channel: Bool indicating whether to do per-channel or per-tensor
quantization
fully_quantize: Bool indicating whether to fully quantize the model. Besides
model body, the input/output will be quantized as well.
inference_type: Data type for the activations. The default value is int8.
Returns:
Quantized model in serialized form (e.g. a TFLITE model) with floating-point
inputs and outputs."
385,mlir_sparsify,tensorflow/tensorflow/lite/python/convert.py,150,function,"Sparsify `input_data_str` to encode sparse tensor with proper format.
Args:
input_data_str: Input data in serialized form (e.g. a TFLITE model).
Returns:
Sparsified model in serialized form (e.g. a TFLITE model)."
386,toco_convert_protos,tensorflow/tensorflow/lite/python/convert.py,162,function,"Convert `input_data_str` according to model and toco parameters.
Unless you know what you are doing consider using
the more friendly `tf.compat.v1.lite.toco_convert`.
Args:
model_flags_str: Serialized proto describing model properties, see
`toco/model_flags.proto`.
toco_flags_str: Serialized proto describing conversion properties, see
`toco/toco_flags.proto`.
input_data_str: Input data in serialized form (e.g. a graphdef is common)
debug_info_str: Serialized `GraphDebugInfo` proto describing logging
information. (default None)
enable_mlir_converter: Enables MLIR-based conversion instead of the default
TOCO conversion. (default False)
Returns:
Converted model in serialized form (e.g. a TFLITE model is common).
Raises:
ConverterError: When conversion fails in TFLiteConverter, usually due to
ops not being supported.
RuntimeError: When conversion fails, an exception is raised with the error
message embedded."
387,build_toco_convert_protos,tensorflow/tensorflow/lite/python/convert.py,291,function,"Builds protocol buffers describing a conversion of a model using TOCO.
Typically this is to convert from TensorFlow GraphDef to TFLite, in which
case the default `input_format` and `output_format` are sufficient.
Args:
input_tensors: List of input tensors. Type and shape are computed using
`foo.shape` and `foo.dtype`.
output_tensors: List of output tensors (only .name is used from this).
inference_type: Target data type of real-number arrays in the output file.
Must be `{tf.float32, tf.uint8, tf.int8}`. (default tf.float32)
inference_input_type: Target data type of real-number input arrays. Allows
for a different type for input arrays in the case of quantization. Must be
`{tf.float32, tf.uint8, tf.int8}`. (default `inference_type`)
input_format: Type of data to read Currently must be
`{TENSORFLOW_GRAPHDEF}`. (default TENSORFLOW_GRAPHDEF)
input_shapes: Input array shape. It needs to be a list of the same length as
`input_tensors`, or None. (default None)
output_format: Output file format. Currently must be `{TFLITE,
GRAPHVIZ_DOT}`. (default TFLITE)
quantized_input_stats: List of tuples of floats representing the mean and
standard deviation. Each tuple maps to the corresponding input tensor.
Only need if `inference_input_type` is `QUANTIZED_UINT8` or `INT8`.
real_input_value = (quantized_input_value - mean_value) / std_dev_value.
(default None)
default_ranges_stats: Tuple of integers representing (min, max) range values
for all arrays without a specified range. Intended for experimenting with
quantization via ""dummy quantization"". (default None)
drop_control_dependency: Boolean indicating whether to drop control
dependencies silently. This is due to TFLite not supporting control
dependencies. (default True)
reorder_across_fake_quant: Boolean indicating whether to reorder FakeQuant
nodes in unexpected locations. Used when the location of the FakeQuant
nodes is preventing graph transformations necessary to convert the graph.
Results in a graph that differs from the quantized training graph,
potentially causing differing arithmetic behavior. (default False)
allow_custom_ops: Boolean indicating whether to allow custom operations.
When false any unknown operation is an error. When true, custom ops are
created for any op that is unknown. The developer will need to provide
these to the TensorFlow Lite runtime with a custom resolver. (default
False)
custom_opdefs: List of strings representing custom ops OpDefs that are
included in the GraphDef. Required when using custom operations with the
MLIR-based converter. (default None)
change_concat_input_ranges: Boolean to change behavior of min/max ranges for
inputs and outputs of the concat operator for quantized models. Changes
the ranges of concat operator overlap when true. (default False)
post_training_quantize: Boolean indicating whether to quantize the weights
of the converted float model. Model size will be reduced and there will be
latency improvements (at the cost of accuracy). (default False)
quantize_to_float16: Boolean indicating whether to convert float buffers to
float16. (default False)
dump_graphviz_dir: Full filepath of folder to dump the graphs at various
stages of processing GraphViz .dot files. Preferred over
--output_format=GRAPHVIZ_DOT in order to keep the requirements of the
output file. (default None)
dump_graphviz_video: Boolean indicating whether to dump the graph after
every graph transformation. (default False)
target_ops: Experimental flag, subject to change. Set of OpsSet options
indicating which converter to use. (default set([OpsSet.TFLITE_BUILTINS]))
allow_nonexistent_arrays: Allow specifying array names that don't exist or
are unused in the final graph. (default False)
debug_info: `GraphDebugInfo` proto containing the stack traces for the
original nodes referred by the converted graph.
conversion_summary_dir: A string, the path to the generated conversion logs.
saved_model_dir: Filepath of the saved model to be converted. This value
will be non-empty only when the saved model import path will be used.
Otherwises, the graph def-based conversion will be processed.
saved_model_version: SavedModel file format version of The saved model file
to be converted. This value will be set only when the SavedModel import
path will be used.
saved_model_tags: Set of string saved model tags, formatted in the
comma-separated value. This value will be set only when the SavedModel
import path will be used.
saved_model_exported_names: Names to be exported (default: export all) when
the saved model import path is on. This value will be set only when the
SavedModel import path will be used.
Returns:
model_flags, toco_flags, debug_info: three protocol buffers describing the
conversion process and debug information.
Raises:
ValueError:
If the input tensor type is unknown
Missing mean_values or std_dev_values
RuntimeError: If TOCO fails to convert (in which case the runtime error's
error text will contain the TOCO error log)"
388,toco_convert_graph_def,tensorflow/tensorflow/lite/python/convert.py,485,function,"""Convert a model using TOCO.
This function is used to convert GraphDefs that cannot be loaded into
TensorFlow to TFLite. Conversion can be customized by providing arguments
that are forwarded to `build_toco_convert_protos` (see documentation for
details).
Args:
input_data: Input data (i.e. often `sess.graph_def`),
input_arrays_with_shape: Tuple of strings representing input tensor names
and list of integers representing input shapes
(e.g., [(""foo"" : [1, 16, 16, 3])]). Use only when graph cannot be loaded
into TensorFlow and when `input_tensors` is None. (default None)
output_arrays: List of output tensors to freeze graph with. Use only when
graph cannot be loaded into TensorFlow and when `output_tensors` is None.
(default None)
enable_mlir_converter: Enables MLIR-based conversion instead of TOCO
conversion.
*args: See `build_toco_convert_protos`,
**kwargs: See `build_toco_convert_protos`.
Returns:
The converted data. For example if TFLite was the destination, then
this will be a tflite flatbuffer in a bytes array.
Raises:
Defined in `build_toco_convert_protos`."
389,toco_convert_impl,tensorflow/tensorflow/lite/python/convert.py,541,function,"""Convert a model using TOCO.
Typically this function is used to convert from TensorFlow GraphDef to TFLite.
Conversion can be customized by providing arguments that are forwarded to
`build_toco_convert_protos` (see documentation for details).
Args:
input_data: Input data (i.e. often `sess.graph_def`),
input_tensors: List of input tensors. Type and shape are computed using
`foo.shape` and `foo.dtype`.
output_tensors: List of output tensors (only .name is used from this).
enable_mlir_converter: Enables MLIR-based conversion instead of TOCO
conversion.
*args: See `build_toco_convert_protos`,
**kwargs: See `build_toco_convert_protos`.
Returns:
The converted data. For example if TFLite was the destination, then
this will be a tflite flatbuffer in a bytes array.
Raises:
Defined in `build_toco_convert_protos`."
390,toco_convert,tensorflow/tensorflow/lite/python/convert.py,580,function,"Convert a model using TOCO.
Typically this function is used to convert from TensorFlow GraphDef to TFLite.
Conversion can be customized by providing arguments that are forwarded to
`build_toco_convert_protos` (see documentation for details). This function has
been deprecated. Please use `lite.TFLiteConverter` instead.
Args:
input_data: Input data (i.e. often `sess.graph_def`),
input_tensors: List of input tensors. Type and shape are computed using
`foo.shape` and `foo.dtype`.
output_tensors: List of output tensors (only .name is used from this).
*args: See `build_toco_convert_protos`,
**kwargs: See `build_toco_convert_protos`.
Returns:
The converted data. For example if TFLite was the destination, then
this will be a tflite flatbuffer in a bytes array.
Raises:
Defined in `build_toco_convert_protos`."
391,get_meta_graph_def,tensorflow/tensorflow/lite/python/convert_saved_model.py,46,function,"Validate saved_model and extract MetaGraphDef.
Args:
saved_model_dir: saved_model path to convert.
tag_set: Set of tag(s) of the MetaGraphDef to load.
Returns:
The meta_graph_def used for tflite conversion.
Raises:
ValueError: No valid MetaGraphDef for given tag_set."
392,get_signature_def,tensorflow/tensorflow/lite/python/convert_saved_model.py,63,function,"Get the signature def from meta_graph with given signature_key.
Args:
meta_graph: meta_graph_def.
signature_key: signature_def in the meta_graph_def.
Returns:
The signature_def used for tflite conversion.
Raises:
ValueError: Given signature_key is not valid for this meta_graph."
393,get_inputs_outputs,tensorflow/tensorflow/lite/python/convert_saved_model.py,88,function,"Get inputs and outputs from SignatureDef.
Args:
signature_def: SignatureDef in the meta_graph_def for conversion.
Returns:
The inputs and outputs in the graph for conversion."
394,freeze_saved_model,tensorflow/tensorflow/lite/python/convert_saved_model.py,155,function,"Converts a SavedModel to a frozen graph.
Args:
saved_model_dir: SavedModel directory to convert.
input_arrays: List of input tensors to freeze graph with. Uses input arrays
from SignatureDef when none are provided.
input_shapes: Dict of strings representing input tensor names to list of
integers representing input shapes (e.g., {""foo"": : [1, 16, 16, 3]}).
Automatically determined when input shapes is None (e.g., {""foo"" : None}).
output_arrays: List of output tensors to freeze graph with. Uses output
arrays from SignatureDef when none are provided.
tag_set: Set of tags identifying the MetaGraphDef within the SavedModel to
analyze. All tags in the tag set must be present.
signature_key: Key identifying SignatureDef containing inputs and outputs.
Returns:
frozen_graph_def: Frozen GraphDef.
in_tensors: List of input tensors for the graph.
out_tensors: List of output tensors for the graph.
graph: `Graph` object.
Raises:
ValueError:
SavedModel doesn't contain a MetaGraphDef identified by tag_set.
signature_key is not in the MetaGraphDef.
assets/ directory is in the MetaGraphDef.
input_shapes does not match the length of input_arrays.
input_arrays or output_arrays are not valid."
395,Delegate,tensorflow/tensorflow/lite/python/interpreter.py,42,class,"Python wrapper class to manage TfLiteDelegate objects.
The shared library is expected to have two functions:
TfLiteDelegate* tflite_plugin_create_delegate(
char**, char**, size_t, void (*report_error)(const char *))
void tflite_plugin_destroy_delegate(TfLiteDelegate*)
The first one creates a delegate object. It may return NULL to indicate an
error (with a suitable error message reported by calling report_error()).
The second one destroys delegate object and must be called for every
created delegate object. Passing NULL as argument value is allowed, i.e.
tflite_plugin_destroy_delegate(tflite_plugin_create_delegate(...))
always works."
396,report,tensorflow/tensorflow/lite/python/interpreter.py,101,method,
397,load_delegate,tensorflow/tensorflow/lite/python/interpreter.py,132,function,"Returns loaded Delegate object.
Args:
library: Name of shared library containing the
[TfLiteDelegate](https://www.tensorflow.org/lite/performance/delegates).
options: Dictionary of options that are required to load the delegate. All
keys and values in the dictionary should be convertible to str. Consult
the documentation of the specific delegate for required and legal options.
(default None)
Returns:
Delegate object.
Raises:
ValueError: Delegate failed to load.
RuntimeError: If delegate loading is used on unsupported platform."
398,Interpreter,tensorflow/tensorflow/lite/python/interpreter.py,159,class,"Interpreter interface for TensorFlow Lite Models.
This makes the TensorFlow Lite interpreter accessible in Python.
It is possible to use this interpreter in a multithreaded Python environment,
but you must be sure to call functions of a particular instance from only
one thread at a time. So if you want to have 4 threads running different
inferences simultaneously, create an interpreter for each one as thread-local
data. Similarly, if you are calling invoke() in one thread on a single
interpreter but you want to use tensor() on another thread once it is done,
you must use a synchronization primitive between the threads to ensure invoke
has returned before calling tensor()."
399,allocate_tensors,tensorflow/tensorflow/lite/python/interpreter.py,242,method,
400,get_tensor_details,tensorflow/tensorflow/lite/python/interpreter.py,365,method,"Gets tensor details for every tensor with valid tensor details.
Tensors where required information about the tensor is not found are not
added to the list. This includes temporary tensors without a name.
Returns:
A list of dictionaries containing tensor information."
401,get_input_details,tensorflow/tensorflow/lite/python/interpreter.py,382,method,"Gets model input details.
Returns:
A list of input details."
402,set_tensor,tensorflow/tensorflow/lite/python/interpreter.py,392,method,"Sets the value of the input tensor.
Note this copies data in `value`.
If you want to avoid copying, you can use the `tensor()` function to get a
numpy buffer pointing to the input buffer in the tflite interpreter.
Args:
tensor_index: Tensor index of tensor to set. This value can be gotten from
the 'index' field in get_input_details.
value: Value of tensor to set.
Raises:
ValueError: If the interpreter could not set the tensor."
403,resize_tensor_input,tensorflow/tensorflow/lite/python/interpreter.py,410,method,"Resizes an input tensor.
```
interpreter = Interpreter(model_content=tflite_model)
interpreter.resize_tensor_input(0, [1, 224, 224, 3], strict=True)
interpreter.allocate_tensors()
interpreter.invoke()
```
Args:
input_index: Tensor index of input to set. This value can be gotten from
the 'index' field in get_input_details.
tensor_size: The tensor_shape to resize the input to.
strict: Only unknown dimensions can be resized when `strict` is True.
Unknown dimensions are indicated as `-1` in the `shape_signature`
attribute of a given tensor. (default False)
Raises:
ValueError: If the interpreter could not resize the input tensor."
404,get_output_details,tensorflow/tensorflow/lite/python/interpreter.py,437,method,"Gets model output details.
Returns:
A list of output details."
405,get_tensor,tensorflow/tensorflow/lite/python/interpreter.py,447,method,"Gets the value of the input tensor (get a copy).
If you wish to avoid the copy, use `tensor()`. This function cannot be used
to read intermediate results.
Args:
tensor_index: Tensor index of tensor to get. This value can be gotten from
the 'index' field in get_output_details.
Returns:
a numpy array."
406,tensor,tensorflow/tensorflow/lite/python/interpreter.py,462,method,"Returns function that gives a numpy view of the current tensor buffer.
This allows reading and writing to this tensors w/o copies. This more
closely mirrors the C++ Interpreter class interface's tensor() member, hence
the name. Be careful to not hold these output references through calls
to `allocate_tensors()` and `invoke()`. This function cannot be used to read
intermediate results.
Usage:
```
interpreter.allocate_tensors()
input = interpreter.tensor(interpreter.get_input_details()[0][""index""])
output = interpreter.tensor(interpreter.get_output_details()[0][""index""])
for i in range(10):
input().fill(3.)
interpreter.invoke()
print(""inference %s"" % output())
```
Notice how this function avoids making a numpy array directly. This is
because it is important to not hold actual numpy views to the data longer
than necessary. If you do, then the interpreter can no longer be invoked,
because it is possible the interpreter would resize and invalidate the
referenced tensors. The NumPy API doesn't allow any mutability of the
the underlying buffers.
WRONG:
```
input = interpreter.tensor(interpreter.get_input_details()[0][""index""])()
output = interpreter.tensor(interpreter.get_output_details()[0][""index""])()
interpreter.allocate_tensors() # This will throw RuntimeError
for i in range(10):
input.fill(3.)
interpreter.invoke() # this will throw RuntimeError since input,output
```
Args:
tensor_index: Tensor index of tensor to get. This value can be gotten from
the 'index' field in get_output_details.
Returns:
A function that can return a new numpy array pointing to the internal
TFLite tensor state at any point. It is safe to hold the function forever,
but it is not safe to hold the numpy array forever."
407,invoke,tensorflow/tensorflow/lite/python/interpreter.py,512,method,"Invoke the interpreter.
Be sure to set the input sizes, allocate tensors and fill values before
calling this. Also, note that this function releases the GIL so heavy
computation can be done in the background while the Python interpreter
continues. No other function on this object should be called while the
invoke() call has not finished.
Raises:
ValueError: When the underlying interpreter fails raise ValueError."
408,reset_all_variables,tensorflow/tensorflow/lite/python/interpreter.py,527,method,
409,InterpreterWithCustomOps,tensorflow/tensorflow/lite/python/interpreter.py,552,class,"Interpreter interface for TensorFlow Lite Models that accepts custom ops.
The interface provided by this class is experimental and therefore not exposed
as part of the public API.
Wraps the tf.lite.Interpreter class and adds the ability to load custom ops
by providing the names of functions that take a pointer to a BuiltinOpResolver
and add a custom op."
410,Optimize,tensorflow/tensorflow/lite/python/lite.py,88,class,"Enum defining the optimizations to apply when generating tflite graphs.
Some optimizations may come at the cost of accuracy.
DEFAULT
Default optimization strategy.
Converter will do its best to improve size and latency based on the
information provided.
Enhanced optimizations are gained by providing a representative_dataset.
This is recommended, and is currently equivalent to the modes below.
Currently, weights will be quantized and if representative_dataset is
provided, activations for quantizable operations will also be quantized.
OPTIMIZE_FOR_SIZE
Deprecated. Does the same as DEFAULT.
OPTIMIZE_FOR_LATENCY
Deprecated. Does the same as DEFAULT."
411,RepresentativeDataset,tensorflow/tensorflow/lite/python/lite.py,131,class,"Representative dataset to evaluate optimizations.
A representative dataset that can be used to evaluate optimizations by the
converter. E.g. converter can use these examples to estimate (min, max) ranges
by calibrating the model on inputs. This can allow converter to quantize a
converted floating point model."
412,TargetSpec,tensorflow/tensorflow/lite/python/lite.py,153,class,"Specification of target device.
Details about target device. Converter optimizes the generated model for
specific device.
Attributes:
supported_ops: Experimental flag, subject to change. Set of OpsSet options
supported by the device. (default set([OpsSet.TFLITE_BUILTINS]))
supported_types: List of types for constant values on the target device.
Supported values are types exported by lite.constants. Frequently, an
optimization choice is driven by the most compact (i.e. smallest) type in
this list (default [constants.FLOAT])"
413,QuantizationMode,tensorflow/tensorflow/lite/python/lite.py,177,class,QuantizationMode determines the quantized conversion from user options.
414,post_training_int8_no_float,tensorflow/tensorflow/lite/python/lite.py,189,method,"Post training int8 quantize, disallow float fallback."
415,post_training_int8_allow_float,tensorflow/tensorflow/lite/python/lite.py,195,method,"Post training int8 quantize, allow float fallback."
416,is_post_training_integer_quantize,tensorflow/tensorflow/lite/python/lite.py,202,method,Post training integer quantization.
417,training_time_int8_allow_float,tensorflow/tensorflow/lite/python/lite.py,207,method,"Training-time int8 quantize, allow float fallback."
418,post_training_int16x8_no_float,tensorflow/tensorflow/lite/python/lite.py,213,method,"Post training int16x8 quantize, disallow float fallback."
419,post_training_int16x8_allow_float,tensorflow/tensorflow/lite/python/lite.py,220,method,"Post training int16x8 quantize, allow float fallback."
420,post_training_dynamic_range_int8,tensorflow/tensorflow/lite/python/lite.py,224,method,"Post training int8 const, on-the-fly int8 quantize of dynamic tensors."
421,post_training_fp16,tensorflow/tensorflow/lite/python/lite.py,233,method,Post training fp16 quantize.
422,fp32_execution,tensorflow/tensorflow/lite/python/lite.py,238,method,If none of the above are true.
423,activations_type,tensorflow/tensorflow/lite/python/lite.py,248,method,
424,converter_flags,tensorflow/tensorflow/lite/python/lite.py,252,method,Flags to the converter.
425,quantizer_flags,tensorflow/tensorflow/lite/python/lite.py,292,method,Default flags to the TFMOT quantizer.
426,contains_training_quant_op,tensorflow/tensorflow/lite/python/lite.py,371,method,Checks if the graph contains any training-time quantization ops.
427,TFLiteConverterBase,tensorflow/tensorflow/lite/python/lite.py,384,class,Converter subclass to share functionality between V1 and V2 converters.
428,TFLiteConverterBaseV2,tensorflow/tensorflow/lite/python/lite.py,522,class,"Converter subclass to share functionality between V2 converters.
Attributes:
allow_custom_ops: Boolean indicating whether to allow custom operations.
When False, any unknown operation is an error. When True, custom ops are
created for any op that is unknown. The developer needs to provide these
to the TensorFlow Lite runtime with a custom resolver. (default False)
optimizations: Experimental flag, subject to change. A list of optimizations
to apply when converting the model. E.g. `[Optimize.DEFAULT]`
representative_dataset: A representative dataset that can be used to
generate input and output samples for the model. The converter can use the
dataset to evaluate different optimizations. Note that this is an optional
attribute but it is necessary if INT8 is the only support builtin ops in
target ops.
target_spec: Experimental flag, subject to change. Specification of target
device.
inference_input_type: Data type of the input layer. Note that integer types
(tf.int8 and tf.uint8) are currently only supported for post training
integer quantization. (default tf.float32, must be in {tf.float32,
tf.int8, tf.uint8})
inference_output_type: Data type of the output layer. Note that integer
types (tf.int8 and tf.uint8) are currently only supported for post
training integer quantization. (default tf.float32, must be in
{tf.float32, tf.int8, tf.uint8})
experimental_new_converter: Experimental flag, subject to change. Enables
MLIR-based conversion instead of TOCO conversion. (default True)"
429,convert,tensorflow/tensorflow/lite/python/lite.py,574,method,"Converts a TensorFlow GraphDef based on instance variables.
Args:
graph_def: Frozen TensorFlow GraphDef.
input_tensors: List of input tensors. Type and shape are computed using
`foo.shape` and `foo.dtype`.
output_tensors: List of output tensors (only .name is used from this).
Returns:
The converted data in serialized format.
Raises:
ValueError:
No concrete functions is specified.
Multiple concrete functions are specified.
Input shape is not specified.
Invalid quantization parameters."
430,TFLiteSavedModelConverterV2,tensorflow/tensorflow/lite/python/lite.py,652,class,"Converts the given SavedModel into TensorFlow Lite model.
Attributes:
saved_model_dir: Directory of the SavedModel."
431,convert,tensorflow/tensorflow/lite/python/lite.py,686,method,"Converts a TensorFlow GraphDef based on instance variables.
Returns:
The converted data in serialized format.
Raises:
ValueError:
No concrete functions is specified.
Multiple concrete functions are specified.
Input shape is not specified.
Invalid quantization parameters."
432,TFLiteKerasModelConverterV2,tensorflow/tensorflow/lite/python/lite.py,719,class,Converts the given Keras model into TensorFlow Lite model.
433,convert,tensorflow/tensorflow/lite/python/lite.py,781,method,"Converts a keras model based on instance variables.
Returns:
The converted data in serialized format.
Raises:
ValueError:
Multiple concrete functions are specified.
Input shape is not specified.
Invalid quantization parameters."
434,TFLiteFrozenGraphConverterV2,tensorflow/tensorflow/lite/python/lite.py,840,class,Converts the given frozen graph into TensorFlow Lite model.
435,convert,tensorflow/tensorflow/lite/python/lite.py,859,method,"Converts a TensorFlow GraphDef based on instance variables.
Returns:
The converted data in serialized format.
Raises:
ValueError:
No concrete functions is specified.
Multiple concrete functions are specified.
Input shape is not specified.
Invalid quantization parameters."
436,TFLiteConverterV2,tensorflow/tensorflow/lite/python/lite.py,910,class,"Converts a TensorFlow model into TensorFlow Lite model.
Attributes:
allow_custom_ops: Boolean indicating whether to allow custom operations.
When False, any unknown operation is an error. When True, custom ops are
created for any op that is unknown. The developer needs to provide these
to the TensorFlow Lite runtime with a custom resolver. (default False)
optimizations: Experimental flag, subject to change. A list of optimizations
to apply when converting the model. E.g. `[Optimize.DEFAULT]`
representative_dataset: A representative dataset that can be used to
generate input and output samples for the model. The converter can use the
dataset to evaluate different optimizations. Note that this is an optional
attribute but it is necessary if INT8 is the only support builtin ops in
target ops.
target_spec: Experimental flag, subject to change. Specification of target
device.
inference_input_type: Data type of the input layer. Note that integer types
(tf.int8 and tf.uint8) are currently only supported for post training
integer quantization. (default tf.float32, must be in {tf.float32,
tf.int8, tf.uint8})
inference_output_type: Data type of the output layer. Note that integer
types (tf.int8 and tf.uint8) are currently only supported for post
training integer quantization. (default tf.float32, must be in
{tf.float32, tf.int8, tf.uint8})
experimental_new_converter: Experimental flag, subject to change. Enables
MLIR-based conversion instead of TOCO conversion. (default True)
Example usage:
```python
# Converting a SavedModel to a TensorFlow Lite model.
converter = tf.lite.TFLiteConverter.from_saved_model(saved_model_dir)
tflite_model = converter.convert()
# Converting a tf.Keras model to a TensorFlow Lite model.
converter = tf.lite.TFLiteConverter.from_keras_model(model)
tflite_model = converter.convert()
# Converting ConcreteFunctions to a TensorFlow Lite model.
converter = tf.lite.TFLiteConverter.from_concrete_functions([func])
tflite_model = converter.convert()
```"
437,from_concrete_functions,tensorflow/tensorflow/lite/python/lite.py,971,method,"Creates a TFLiteConverter object from ConcreteFunctions.
Args:
funcs: List of TensorFlow ConcreteFunctions. The list should not contain
duplicate elements. Currently converter can only convert a single
ConcreteFunction. Converting multiple functions is under development.
Returns:
TFLiteConverter object.
Raises:
Invalid input type."
438,from_saved_model,tensorflow/tensorflow/lite/python/lite.py,995,method,"Creates a TFLiteConverter object from a SavedModel directory.
Args:
saved_model_dir: SavedModel directory to convert.
signature_keys: List of keys identifying SignatureDef containing inputs
and outputs. Elements should not be duplicated. By default the
`signatures` attribute of the MetaGraphdef is used. (default
saved_model.signatures)
tags: Set of tags identifying the MetaGraphDef within the SavedModel to
analyze. All tags in the tag set must be present. (default set(SERVING))
Returns:
TFLiteConverter object.
Raises:
Invalid signature keys."
439,from_keras_model,tensorflow/tensorflow/lite/python/lite.py,1057,method,"Creates a TFLiteConverter object from a Keras model.
Args:
model: tf.Keras.Model
Returns:
TFLiteConverter object."
440,convert,tensorflow/tensorflow/lite/python/lite.py,1069,method,"Converts a TensorFlow GraphDef based on instance variables.
Returns:
The converted data in serialized format.
Raises:
ValueError:
No concrete functions is specified.
Multiple concrete functions are specified.
Input shape is not specified.
Invalid quantization parameters."
441,TFLiteConverterBaseV1,tensorflow/tensorflow/lite/python/lite.py,1085,class,"Converter subclass to share functionality between V1 converters.
Attributes:
inference_type: Target data type of real-number arrays in the output file.
Must be `{tf.float32, tf.uint8}`. If `optimzations` are provided, this
parameter is ignored. (default tf.float32)
inference_input_type: Target data type of real-number input arrays. Allows
for a different type for input arrays. If an integer type is provided and
`optimizations` are not used, `quantized_inputs_stats` must be provided.
If `inference_type` is tf.uint8, signaling conversion to a fully quantized
model from a quantization-aware trained input model, then
`inference_input_type` defaults to tf.uint8. In all other cases,
`inference_input_type` defaults to tf.float32. Must be `{tf.float32,
tf.uint8, tf.int8}`
inference_output_type: Target data type of real-number output arrays. Allows
for a different type for output arrays. If `inference_type` is tf.uint8,
signaling conversion to a fully quantized model from a quantization-aware
trained output model, then `inference_output_type` defaults to tf.uint8.
In all other cases, `inference_output_type` must be tf.float32, an error
will be thrown otherwise. Must be `{tf.float32, tf.uint8, tf.int8}`
output_format: Output file format. Currently must be `{TFLITE,
GRAPHVIZ_DOT}`. (default TFLITE)
quantized_input_stats: Dict of strings representing input tensor names
mapped to tuple of floats representing the mean and standard deviation
of the training data (e.g., {""foo"" : (0., 1.)}). Only need if
`inference_input_type` is `QUANTIZED_UINT8`. real_input_value =
(quantized_input_value - mean_value) / std_dev_value. (default {})
default_ranges_stats: Tuple of integers representing (min, max) range values
for all arrays without a specified range. Intended for experimenting with
quantization via ""dummy quantization"". (default None)
drop_control_dependency: Boolean indicating whether to drop control
dependencies silently. This is due to TFLite not supporting control
dependencies. (default True)
reorder_across_fake_quant: Boolean indicating whether to reorder FakeQuant
nodes in unexpected locations. Used when the location of the FakeQuant
nodes is preventing graph transformations necessary to convert the graph.
Results in a graph that differs from the quantized training graph,
potentially causing differing arithmetic behavior. (default False)
change_concat_input_ranges: Boolean to change behavior of min/max ranges for
inputs and outputs of the concat operator for quantized models. Changes
the ranges of concat operator overlap when true. (default False)
allow_custom_ops: Boolean indicating whether to allow custom operations.
When false any unknown operation is an error. When true, custom ops are
created for any op that is unknown. The developer will need to provide
these to the TensorFlow Lite runtime with a custom resolver. (default
False)
post_training_quantize: Deprecated. Please specify `[Optimize.DEFAULT]` for
`optimizations` instead. Boolean indicating whether to quantize the
weights of the converted float model. Model size will be reduced and
there will be latency improvements (at the cost of accuracy). (default
False)
dump_graphviz_dir: Full filepath of folder to dump the graphs at various
stages of processing GraphViz .dot files. Preferred over
--output_format=GRAPHVIZ_DOT in order to keep the requirements of the
output file. (default None)
dump_graphviz_video: Boolean indicating whether to dump the graph after
every graph transformation. (default False)
conversion_summary_dir: A string indicating the path to the generated
conversion logs.
target_ops: Deprecated. Please specify `target_spec.supported_ops` instead.
Set of OpsSet options indicating which converter to use. (default
set([OpsSet.TFLITE_BUILTINS]))
target_spec: Experimental flag, subject to change. Specification of target
device.
optimizations: Experimental flag, subject to change. A list of optimizations
to apply when converting the model. E.g. `[Optimize.DEFAULT]`
representative_dataset: A representative dataset that can be used to
generate input and output samples for the model. The converter can use the
dataset to evaluate different optimizations.
experimental_new_converter: Experimental flag, subject to change. Enables
MLIR-based conversion instead of TOCO conversion. (default True)"
442,convert,tensorflow/tensorflow/lite/python/lite.py,1226,method,"Converts a TensorFlow GraphDef based on instance variables.
Returns:
The converted data in serialized format. Either a TFLite Flatbuffer or a
Graphviz graph depending on value in `output_format`.
Raises:
ValueError:
Input shape is not specified.
None value for dimension in input_tensor."
443,get_input_arrays,tensorflow/tensorflow/lite/python/lite.py,1352,method,"Returns a list of the names of the input tensors.
Returns:
List of strings."
444,TFLiteSavedModelConverter,tensorflow/tensorflow/lite/python/lite.py,1410,class,"Converts the given SavedModel into TensorFlow Lite model.
Attributes:
saved_model_dir: Directory of the SavedModel."
445,TFLiteKerasModelConverter,tensorflow/tensorflow/lite/python/lite.py,1458,class,Converts the given SavedModel into TensorFlow Lite model.
446,convert,tensorflow/tensorflow/lite/python/lite.py,1567,method,"Converts a Keras model based on instance variables.
Returns:
The converted data in serialized format. Either a TFLite Flatbuffer or a
Graphviz graph depending on value in `output_format`.
Raises:
ValueError:
Input shape is not specified.
None value for dimension in input_tensor."
447,TFLiteFrozenGraphConverter,tensorflow/tensorflow/lite/python/lite.py,1586,class,Converts the given frozen graph def into TensorFlow Lite model.
448,TFLiteConverter,tensorflow/tensorflow/lite/python/lite.py,1634,class,"Convert a TensorFlow model into `output_format`.
This is used to convert from a TensorFlow GraphDef, SavedModel or tf.keras
model into either a TFLite FlatBuffer or graph visualization.
Attributes:
inference_type: Target data type of real-number arrays in the output file.
Must be `{tf.float32, tf.uint8}`. If `optimzations` are provided, this
parameter is ignored. (default tf.float32)
inference_input_type: Target data type of real-number input arrays. Allows
for a different type for input arrays.
If an integer type is provided and `optimizations` are not used,
`quantized_inputs_stats` must be provided.
If `inference_type` is tf.uint8, signaling conversion to a fully quantized
model from a quantization-aware trained input model, then
`inference_input_type` defaults to tf.uint8.
In all other cases, `inference_input_type` defaults to tf.float32.
Must be `{tf.float32, tf.uint8, tf.int8}`
inference_output_type: Target data type of real-number output arrays. Allows
for a different type for output arrays.
If `inference_type` is tf.uint8, signaling conversion to a fully quantized
model from a quantization-aware trained output model, then
`inference_output_type` defaults to tf.uint8.
In all other cases, `inference_output_type` must be tf.float32, an error
will be thrown otherwise.
Must be `{tf.float32, tf.uint8, tf.int8}`
output_format: Output file format. Currently must be `{TFLITE,
GRAPHVIZ_DOT}`. (default TFLITE)
quantized_input_stats: Dict of strings representing input tensor names
mapped to tuple of floats representing the mean and standard deviation
of the training data (e.g., {""foo"" : (0., 1.)}). Only need if
`inference_input_type` is `QUANTIZED_UINT8`.
real_input_value = (quantized_input_value - mean_value) / std_dev_value.
(default {})
default_ranges_stats: Tuple of integers representing (min, max) range values
for all arrays without a specified range. Intended for experimenting with
quantization via ""dummy quantization"". (default None)
drop_control_dependency: Boolean indicating whether to drop control
dependencies silently. This is due to TFLite not supporting control
dependencies. (default True)
reorder_across_fake_quant: Boolean indicating whether to reorder FakeQuant
nodes in unexpected locations. Used when the location of the FakeQuant
nodes is preventing graph transformations necessary to convert the graph.
Results in a graph that differs from the quantized training graph,
potentially causing differing arithmetic behavior. (default False)
change_concat_input_ranges: Boolean to change behavior of min/max ranges for
inputs and outputs of the concat operator for quantized models. Changes
the ranges of concat operator overlap when true. (default False)
allow_custom_ops: Boolean indicating whether to allow custom operations.
When false any unknown operation is an error. When true, custom ops are
created for any op that is unknown. The developer will need to provide
these to the TensorFlow Lite runtime with a custom resolver.
(default False)
post_training_quantize: Deprecated. Please specify `[Optimize.DEFAULT]` for
`optimizations` instead. Boolean indicating whether to quantize the
weights of the converted float model. Model size will be reduced and
there will be latency improvements (at the cost of accuracy).
(default False)
dump_graphviz_dir: Full filepath of folder to dump the graphs at various
stages of processing GraphViz .dot files. Preferred over
--output_format=GRAPHVIZ_DOT in order to keep the requirements of the
output file. (default None)
dump_graphviz_video: Boolean indicating whether to dump the graph after
every graph transformation. (default False)
conversion_summary_dir: A string indicating the path to the generated
conversion logs.
target_ops: Deprecated. Please specify `target_spec.supported_ops` instead.
Set of OpsSet options indicating which converter to use.
(default set([OpsSet.TFLITE_BUILTINS]))
target_spec: Experimental flag, subject to change. Specification of target
device.
optimizations: Experimental flag, subject to change. A list of optimizations
to apply when converting the model. E.g. `[Optimize.DEFAULT]`
representative_dataset: A representative dataset that can be used to
generate input and output samples for the model. The converter can use
the dataset to evaluate different optimizations.
experimental_new_converter: Experimental flag, subject to change.
Enables MLIR-based conversion instead of TOCO conversion. (default True)
Example usage:
```python
# Converting a GraphDef from session.
converter = tf.compat.v1.TFLiteConverter.from_session(
sess, in_tensors, out_tensors)
tflite_model = converter.convert()
open(""converted_model.tflite"", ""wb"").write(tflite_model)
# Converting a GraphDef from file.
converter = tf.compat.v1.TFLiteConverter.from_frozen_graph(
graph_def_file, input_arrays, output_arrays)
tflite_model = converter.convert()
open(""converted_model.tflite"", ""wb"").write(tflite_model)
# Converting a SavedModel.
converter = tf.compat.v1.TFLiteConverter.from_saved_model(saved_model_dir)
tflite_model = converter.convert()
open(""converted_model.tflite"", ""wb"").write(tflite_model)
# Converting a tf.keras model.
converter = tf.compat.v1.TFLiteConverter.from_keras_model_file(keras_model)
tflite_model = converter.convert()
open(""converted_model.tflite"", ""wb"").write(tflite_model)
```"
449,from_session,tensorflow/tensorflow/lite/python/lite.py,1776,method,"Creates a TFLiteConverter class from a TensorFlow Session.
Args:
sess: TensorFlow Session.
input_tensors: List of input tensors. Type and shape are computed using
`foo.shape` and `foo.dtype`.
output_tensors: List of output tensors (only .name is used from this).
Returns:
TFLiteConverter class."
450,from_frozen_graph,tensorflow/tensorflow/lite/python/lite.py,1796,method,"Creates a TFLiteConverter class from a file containing a frozen GraphDef.
Args:
graph_def_file: Full filepath of file containing frozen GraphDef.
input_arrays: List of input tensors to freeze graph with.
output_arrays: List of output tensors to freeze graph with.
input_shapes: Dict of strings representing input tensor names to list of
integers representing input shapes (e.g., {""foo"" : [1, 16, 16, 3]}).
Automatically determined when input shapes is None (e.g., {""foo"" :
None}). (default None)
Returns:
TFLiteConverter class.
Raises:
IOError:
File not found.
Unable to parse input file.
ValueError:
The graph is not frozen.
input_arrays or output_arrays contains an invalid tensor name.
input_shapes is not correctly defined when required"
451,from_saved_model,tensorflow/tensorflow/lite/python/lite.py,1889,method,"Creates a TFLiteConverter class from a SavedModel.
Args:
saved_model_dir: SavedModel directory to convert.
input_arrays: List of input tensors to freeze graph with. Uses input
arrays from SignatureDef when none are provided. (default None)
input_shapes: Dict of strings representing input tensor names to list of
integers representing input shapes (e.g., {""foo"" : [1, 16, 16, 3]}).
Automatically determined when input shapes is None (e.g., {""foo"" :
None}). (default None)
output_arrays: List of output tensors to freeze graph with. Uses output
arrays from SignatureDef when none are provided. (default None)
tag_set: Set of tags identifying the MetaGraphDef within the SavedModel to
analyze. All tags in the tag set must be present. (default set(""serve""))
signature_key: Key identifying SignatureDef containing inputs and outputs.
(default DEFAULT_SERVING_SIGNATURE_DEF_KEY)
Returns:
TFLiteConverter class."
452,from_keras_model_file,tensorflow/tensorflow/lite/python/lite.py,1936,method,"Creates a TFLiteConverter class from a tf.keras model file.
Args:
model_file: Full filepath of HDF5 file containing the tf.keras model.
input_arrays: List of input tensors to freeze graph with. Uses input
arrays from SignatureDef when none are provided. (default None)
input_shapes: Dict of strings representing input tensor names to list of
integers representing input shapes (e.g., {""foo"" : [1, 16, 16, 3]}).
Automatically determined when input shapes is None (e.g., {""foo"" :
None}). (default None)
output_arrays: List of output tensors to freeze graph with. Uses output
arrays from SignatureDef when none are provided. (default None)
custom_objects: Dict mapping names (strings) to custom classes or
functions to be considered during model deserialization. (default None)
Returns:
TFLiteConverter class."
453,convert,tensorflow/tensorflow/lite/python/lite.py,1964,method,"Converts a TensorFlow GraphDef based on instance variables.
Returns:
The converted data in serialized format. Either a TFLite Flatbuffer or a
Graphviz graph depending on value in `output_format`.
Raises:
ValueError:
Input shape is not specified.
None value for dimension in input_tensor."
454,TocoConverter,tensorflow/tensorflow/lite/python/lite.py,1979,class,"Convert a TensorFlow model into `output_format` using TOCO.
This class has been deprecated. Please use `lite.TFLiteConverter` instead."
455,from_session,tensorflow/tensorflow/lite/python/lite.py,1988,method,Creates a TocoConverter class from a TensorFlow Session.
456,from_frozen_graph,tensorflow/tensorflow/lite/python/lite.py,1995,method,Creates a TocoConverter class from a file containing a frozen graph.
457,from_saved_model,tensorflow/tensorflow/lite/python/lite.py,2007,method,Creates a TocoConverter class from a SavedModel.
458,from_keras_model_file,tensorflow/tensorflow/lite/python/lite.py,2022,method,Creates a TocoConverter class from a tf.keras model file.
459,FromConstructor,tensorflow/tensorflow/lite/python/lite_test.py,76,class,
460,FromFrozenGraphFile,tensorflow/tensorflow/lite/python/lite_test.py,1464,class,
461,FromFrozenGraphObjectDetection,tensorflow/tensorflow/lite/python/lite_test.py,1650,class,
462,MyAddLayer,tensorflow/tensorflow/lite/python/lite_test.py,1897,class,
463,call,tensorflow/tensorflow/lite/python/lite_test.py,1903,method,
464,get_config,tensorflow/tensorflow/lite/python/lite_test.py,1906,method,
465,FromKerasFile,tensorflow/tensorflow/lite/python/lite_test.py,1912,class,
466,setUp,tensorflow/tensorflow/lite/python/lite_test.py,1914,method,
467,tearDown,tensorflow/tensorflow/lite/python/lite_test.py,1921,method,
468,UnknownShapes,tensorflow/tensorflow/lite/python/lite_v2_test.py,1047,class,
469,model,tensorflow/tensorflow/lite/python/lite_v2_test.py,1056,method,
470,model,tensorflow/tensorflow/lite/python/lite_v2_test.py,1077,method,
471,calibration_gen,tensorflow/tensorflow/lite/python/lite_v2_test.py,1093,method,
472,model,tensorflow/tensorflow/lite/python/lite_v2_test.py,1157,method,
473,model,tensorflow/tensorflow/lite/python/lite_v2_test.py,1178,method,
474,OpHint,tensorflow/tensorflow/lite/python/op_hint.py,97,class,"A class that helps build tflite function invocations.
It allows you to take a bunch of TensorFlow ops and annotate the construction
such that toco knows how to convert it to tflite. This embeds a pseudo
function in a TensorFlow graph. This allows embedding high-level API usage
information in a lower level TensorFlow implementation so that an alternative
implementation can be substituted later.
Essentially, any ""input"" into this pseudo op is fed into an identity, and
attributes are added to that input before being used by the constituent ops
that make up the pseudo op. A similar process is done to any output that
is to be exported from the current op."
475,add_input,tensorflow/tensorflow/lite/python/op_hint.py,388,method,"Add a wrapped input argument to the hint.
Args:
*args: The input tensor.
**kwargs:
""name"" label
""tag"" a tag to group multiple arguments that will be aggregated. I.e.
a string like 'cool_input'. Basically multiple inputs can be added
to the same hint for parallel operations that will eventually be
combined. An example would be static_rnn which creates multiple copies
of state or inputs.
""aggregate"" aggregation strategy that is valid only for tag non None.
Acceptable values are OpHint.AGGREGATE_FIRST, OpHint.AGGREGATE_LAST,
and OpHint.AGGREGATE_STACK.
""index_override"" The global index to use. This corresponds to the
argument order in the final stub that will be generated.
Returns:
The wrapped input tensor."
476,add_output,tensorflow/tensorflow/lite/python/op_hint.py,410,method,"Add a wrapped output argument to the hint.
Args:
*args: The output tensor.
**kwargs:
""name"" label
""tag"" a tag to group multiple arguments that will be aggregated. I.e.
a string like 'cool_input'. Basically multiple inputs can be added
to the same hint for parallel operations that will eventually be
combined. An example would be static_rnn which creates multiple copies
of state or inputs.
""aggregate"" aggregation strategy that is valid only for tag non None.
Acceptable values are OpHint.AGGREGATE_FIRST, OpHint.AGGREGATE_LAST,
and OpHint.AGGREGATE_STACK.
""index_override"" The global index to use. This corresponds to the
argument order in the final stub that will be generated.
Returns:
The wrapped output tensor."
477,add_inputs,tensorflow/tensorflow/lite/python/op_hint.py,432,method,"Add a sequence of inputs to the function invocation.
Args:
*args: List of inputs to be converted (should be Tf.Tensor).
**kwargs: This allows 'names' which should be a list of names.
Returns:
Wrapped inputs (identity standins that have additional metadata). These
are also are also tf.Tensor's."
478,add_outputs,tensorflow/tensorflow/lite/python/op_hint.py,451,method,"Add a sequence of outputs to the function invocation.
Args:
*args: List of outputs to be converted (should be tf.Tensor).
**kwargs: See
Returns:
Wrapped outputs (identity standins that have additional metadata). These
are also tf.Tensor's."
479,add,tensorflow/tensorflow/lite/python/op_hint.py,229,method,"Return a wrapped tensor of an input tensor as an argument.
Args:
arg: A TensorFlow tensor that should be considered an argument.
tag: String tag to identify arguments that should be packed.
name: Name of argument. This is included in the Identity hint op names.
aggregate: Strategy to aggregate.
Acceptable values are OpHint.AGGREGATE_FIRST, OpHint.AGGREGATE_LAST,
and OpHint.AGGREGATE_STACK.
Note, aggregate is only valid if tag is specified.
index_override: Specify what input/output index should this be in the
final stub. i.e. add(arg0, index=1); add(arg1, index=0) will make the
final stub be as stub_func(inputs[arg1, arg0], outputs=[]) rather than
the default call order based ordering.
Returns:
A tensor representing the wrapped argument.
Raises:
ValueError: When indices are not consistent."
480,assert_dictlist_has_keys,tensorflow/tensorflow/lite/python/op_hint.py,365,method,
481,find_all_hinted_output_nodes,tensorflow/tensorflow/lite/python/op_hint.py,1257,function,"Find all Ophints output nodes in the graph.
This is used to get all the output nodes those are ophinted, it is important
for operation like convert_variables_to_constants keep all ophints structure.
Note: only one of session or graph_def should be used, not both.
Why this can be useful? Some TensorFlow ops (e.g. bidirectional rnn), can
generate multiple outputs for unfused subgraph. If not all output nodes are
consumed, graph optimization can potentially drop the unused nodes and cause
ophints in an invalid states (due to missing ophinted output nodes). So it's
important for us to find all those hinted output nodes and make sure they're
not discarded away.
Args:
session: A TensorFlow session that contains the graph to convert.
graph_def: A graph def that we should convert.
Returns:
A list of OpHints output nodes.
Raises:
ValueError: If both session and graph_def are provided."
482,is_ophint_converted,tensorflow/tensorflow/lite/python/op_hint.py,1292,function,
483,convert_op_hints_to_stubs,tensorflow/tensorflow/lite/python/op_hint.py,1305,function,"Converts a graphdef with LiteOp hints into stub operations.
This is used to prepare for toco conversion of complex intrinsic usages.
Note: only one of session or graph_def should be used, not both.
Args:
session: A TensorFlow session that contains the graph to convert.
graph_def: A graph def that we should convert.
write_callback: A function pointer that can be used to write intermediate
steps of graph transformation (optional).
Returns:
A new graphdef with all ops contained in OpHints being replaced by
a single op call with the right parameters.
Raises:
ValueError: If both session and graph_def are provided."
484,convert_dtype_to_tflite_type,tensorflow/tensorflow/lite/python/util.py,59,function,"Converts tf.dtype to TFLite proto type.
Args:
tf_dtype: tf.dtype
Raises:
ValueError: Unsupported tf.dtype.
Returns:
types_flag_pb2."
485,get_tensor_name,tensorflow/tensorflow/lite/python/util.py,77,function,"Returns name of the input tensor.
Args:
tensor: tf.Tensor
Returns:
str"
486,get_tensors_from_tensor_names,tensorflow/tensorflow/lite/python/util.py,98,function,"Gets the Tensors associated with the `tensor_names` in the provided graph.
Args:
graph: TensorFlow Graph.
tensor_names: List of strings that represent names of tensors in the graph.
Returns:
A list of Tensor objects in the same order the names are provided.
Raises:
ValueError:
tensor_names contains an invalid tensor name."
487,set_tensor_shapes,tensorflow/tensorflow/lite/python/util.py,141,function,"Sets Tensor shape for each tensor if the shape is defined.
Args:
tensors: TensorFlow ops.Tensor.
shapes: Dict of strings representing input tensor names to list of
integers representing input shapes (e.g., {""foo"": : [1, 16, 16, 3]}).
Raises:
ValueError:
`shapes` contains an invalid tensor.
`shapes` contains an invalid shape for a valid tensor."
488,get_grappler_config,tensorflow/tensorflow/lite/python/util.py,172,function,"Creates a tf.compat.v1.ConfigProto for configuring Grappler.
Args:
optimizers_list: List of strings that represents the list of optimizers.
Returns:
tf.ConfigProto."
489,run_graph_optimizations,tensorflow/tensorflow/lite/python/util.py,188,function,"Apply standard TensorFlow optimizations to the graph_def.
Args:
graph_def: Frozen GraphDef to be optimized.
input_arrays: List of arrays that are considered inputs of the graph.
output_arrays: List of arrays that are considered outputs of the graph.
config: tf.ConfigProto.
graph: TensorFlow Graph. Required when Eager mode is enabled. (default None)
Returns:
A new, optimized GraphDef."
490,freeze_graph,tensorflow/tensorflow/lite/python/util.py,241,function,"Returns a frozen GraphDef.
Runs a Grappler pass and freezes a graph with Variables in it. Otherwise the
existing GraphDef is returned. The Grappler pass is only run on models that
are frozen in order to inline the functions in the graph.
If OpHints is present, it will try to convert the OpHint graph.
Args:
sess: TensorFlow Session.
input_tensors: List of input tensors.
output_tensors: List of output tensors (only .name is used from this).
Returns:
Frozen GraphDef."
491,is_frozen_graph,tensorflow/tensorflow/lite/python/util.py,281,function,"Determines if the graph is frozen.
Determines if a graph has previously been frozen by checking for any
operations of type Variable*. If variables are found, the graph is not frozen.
Args:
sess: TensorFlow Session.
Returns:
Bool."
492,build_debug_info_func,tensorflow/tensorflow/lite/python/util.py,300,function,"Returns a method to retrieve the `GraphDebugInfo` from the original graph.
Args:
original_graph: The original `Graph` containing all the op stack traces.
Returns:
A function which retrieves the stack traces from the original graph and
converts them to a `GraphDebugInfo` for a given set of nodes."
493,convert_debug_info_func,tensorflow/tensorflow/lite/python/util.py,339,function,"Returns a method to retrieve the `GraphDebugInfo` from the original graph.
Args:
saved_debug_info: The `GraphDebugInfo` containing all the debug info.
Returns:
A function which retrieves the stack traces from the original graph and
converts them to a `GraphDebugInfo` for a given set of nodes."
494,get_debug_info,tensorflow/tensorflow/lite/python/util.py,368,function,"Returns the debug info for the original nodes in the `converted_graph`.
Args:
nodes_to_debug_info_func: The method to collect the op debug info for the
nodes.
converted_graph: A `GraphDef` after optimization and transformation.
Returns:
`GraphDebugInfo` for all the original nodes in `converted_graph`."
495,convert_bytes_to_c_source,tensorflow/tensorflow/lite/python/util.py,399,function,"Returns strings representing a C constant array containing `data`.
Args:
data: Byte array that will be converted into a C constant.
array_name: String to use as the variable name for the constant array.
max_line_width: The longest line length, for formatting purposes.
include_guard: Name to use for the include guard macro definition.
include_path: Optional path to include in the source file.
use_tensorflow_license: Whether to include the standard TensorFlow Apache2
license in the generated files.
Returns:
Text that can be compiled as a C source file to link in the data as a
literal array of values.
Text that can be used as a C header file to reference the literal array."
496,wrapped_toco_convert,tensorflow/tensorflow/lite/python/wrap_toco.py,29,function,Wraps TocoConvert with lazy loader.
497,wrapped_get_potentially_supported_ops,tensorflow/tensorflow/lite/python/wrap_toco.py,41,function,Wraps TocoGetPotentiallySupportedOps with lazy loader.
498,wrapped_experimental_mlir_quantize,tensorflow/tensorflow/lite/python/wrap_toco.py,46,function,Wraps experimental mlir quantize model.
499,wrapped_experimental_mlir_sparsify,tensorflow/tensorflow/lite/python/wrap_toco.py,55,function,Wraps experimental mlir sparsify model.
500,Calibrator,tensorflow/tensorflow/lite/python/optimize/calibrator.py,33,class,"Calibrates a floating point model and then quantizes it.
This is an internal class, not a public interface."
501,calibrate_and_quantize,tensorflow/tensorflow/lite/python/optimize/calibrator.py,58,method,"Calibrates the model with specified generator and then quantizes it.
The input shapes of the calibrator are resized with the calibration data if
`resize_input` is set.
Returns:
A quantized model.
Args:
dataset_gen: A generator that generates calibration samples.
input_type: A tf.dtype representing the desired real-value input type.
output_type: A tf.dtype representing the desired real-value output type.
allow_float: A boolean. False if the resulting model cannot perform float
computation, useful when targeting an integer-only backend.
If False, an error will be thrown if an operation cannot be
quantized, otherwise the model will fallback to float ops.
activations_type: A tf.dtype representing the desired type for
activations.
resize_input: A boolean. True if the shape of the sample data is different
from the input."
502,calibrate_and_quantize_single,tensorflow/tensorflow/lite/python/optimize/calibrator.py,100,method,"Calibrates the model with specified generator and then quantizes it.
Only the single op with output op_output_name will be quantized.
The input shapes of the calibrator are resized with the calibration data.
Returns:
A quantized model.
Args:
dataset_gen: A generator that generates calibration samples.
input_type: A tf.dtype representing the desired real-value input type.
output_type: A tf.dtype representing the desired real-value output type.
allow_float: A boolean. False if the resulting model cannot perform float
computation, useful when targeting an integer-only backend. If False, an
error will be thrown if an operation cannot be quantized, otherwise the
model will fallback to float ops.
op_output_name: A string, only this op will be quantized.
resize_input: A boolean. True if the shape of the sample data is different
from the input."
503,calibrate,tensorflow/tensorflow/lite/python/optimize/calibrator.py,140,method,"Calibrates the model with specified generator.
Returns:
A model with min and max calibration stats.
Args:
dataset_gen: A generator that generates calibration samples."
504,TemporaryDirectoryResource,tensorflow/tensorflow/lite/schema/upgrade_schema.py,57,function,
505,Converter,tensorflow/tensorflow/lite/schema/upgrade_schema.py,65,class,"Converts TensorFlow flatbuffer models from old to new version of schema.
This can convert between any version to the latest version. It uses
an incremental upgrade strategy to go from version to version.
Usage:
converter = Converter()
converter.Convert(""a.tflite"", ""a.json"")
converter.Convert(""b.json"", ""b.tflite"")"
506,Convert,tensorflow/tensorflow/lite/schema/upgrade_schema.py,309,method,"Perform schema conversion from input_file to output_file.
Args:
input_file: Filename of TensorFlow Lite data to convert from. Must
be `.json` or `.bin` extension files for JSON or Binary forms of
the TensorFlow FlatBuffer schema.
output_file: Filename to write to. Extension also must be `.json`
or `.bin`.
Raises:
RuntimeError: Generated when none of the upgrader supported schemas
matche the `input_file` data."
507,FindSchema,tensorflow/tensorflow/lite/schema/upgrade_schema.py,88,method,
508,RemapOperator,tensorflow/tensorflow/lite/schema/upgrade_schema.py,209,method,"Go from old schema op name to new schema op name.
Args:
opcode_name: String representing the ops (see :schema.fbs).
Returns:
Converted opcode_name from V1 to V2."
509,RemapOperatorType,tensorflow/tensorflow/lite/schema/upgrade_schema.py,232,method,"Remap operator structs from old names to new names.
Args:
operator_type: String representing the builtin operator data type
string.
(see :schema.fbs).
Raises:
ValueError: When the model has consistency problems.
Returns:
Upgraded builtin operator data type as a string."
510,JsonDumpAndFlush,tensorflow/tensorflow/lite/schema/upgrade_schema_test.py,242,function,"Write the dictionary `data` to a JSON file `fp` (and flush).
Args:
data: in a dictionary that is JSON serializable.
fp: File-like object"
511,MultiGenState,tensorflow/tensorflow/lite/testing/generate_examples_lib.py,176,class,"State of multiple set generation process.
This state class stores the information needed when generating the examples
for multiple test set. The stored informations are open archive object to be
shared, information on test target for current iteration of generation,
accumulated generation results."
512,Options,tensorflow/tensorflow/lite/testing/generate_examples_lib.py,203,class,All options for example generation.
513,generate_examples,tensorflow/tensorflow/lite/testing/generate_examples_lib.py,256,function,"Generate examples for a test set.
Args:
options: Options containing information to generate examples.
Raises:
RuntimeError: if the test function cannot be found."
514,generate_multi_set_examples,tensorflow/tensorflow/lite/testing/generate_examples_lib.py,294,function,"Generate examples for test sets.
Args:
options: Options containing information to generate examples.
test_sets: List of the name of test sets to generate examples."
515,make_report_table,tensorflow/tensorflow/lite/testing/generate_examples_report.py,32,function,"Make an HTML report of the success/failure reports.
Args:
fp: File-like object in which to put the html.
title: ""Title of the zip file this pertains to.""
reports: a list of conversion attempts. (report_args, report_vals) i.e.
({""shape"": [1,2,3], ""type"": ""tf.float32""},
{""tf"": ""SUCCESS"", ""toco"": ""FAILURE"", ""toco_log"": ""Unsupported type."",
""tf_log"": """"})"
516,toco_options,tensorflow/tensorflow/lite/testing/toco_convert.py,31,function,"Create TOCO options to process a model.
Args:
data_types: input and inference types used by TOCO.
input_arrays: names of the input tensors
output_arrays: name of the output tensors
shapes: shapes of the input tensors
extra_toco_options: additional toco options
Returns:
the options in a string."
517,toco_convert,tensorflow/tensorflow/lite/testing/toco_convert.py,78,function,"Convert a model's graph def into a tflite model.
NOTE: this currently shells out to the toco binary, but we would like
convert to Python API tooling in the future.
Args:
options: An Options instance.
graph_def: A GraphDef object.
input_tensors: List of input tensor tuples `(name, shape, type)`.
output_tensors: List of output tensors (names).
**kwargs: Extra options to be passed.
Returns:
output tflite model, log_txt from conversion
or None, log_txt if it did not convert properly."
518,ExtraTocoOptions,tensorflow/tensorflow/lite/testing/zip_test_utils.py,88,class,"Additional toco options besides input, output, shape."
519,create_tensor_data,tensorflow/tensorflow/lite/testing/zip_test_utils.py,106,function,"Build tensor data spreading the range [min_value, max_value)."
520,create_scalar_data,tensorflow/tensorflow/lite/testing/zip_test_utils.py,126,function,Build scalar tensor data range from min_value to max_value exclusively.
521,freeze_graph,tensorflow/tensorflow/lite/testing/zip_test_utils.py,144,function,"Freeze the current graph.
Args:
session: Tensorflow sessions containing the graph
outputs: List of output tensors
Returns:
The frozen graph_def."
522,format_result,tensorflow/tensorflow/lite/testing/zip_test_utils.py,158,function,Convert a tensor to a format that can be used in test specs.
523,write_examples,tensorflow/tensorflow/lite/testing/zip_test_utils.py,168,function,"Given a list `examples`, write a text format representation.
The file format is csv like with a simple repeated pattern. We would ike
to use proto here, but we can't yet due to interfacing with the Android
team using this format.
Args:
fp: File-like object to write to.
examples: Example dictionary consisting of keys ""inputs"" and ""outputs"""
524,get_input_shapes_map,tensorflow/tensorflow/lite/testing/zip_test_utils.py,225,function,"Gets a map of input names to shapes.
Args:
input_tensors: List of input tensor tuples `(name, shape, type)`.
Returns:
{string : list of integers}."
525,get_filepath,tensorflow/tensorflow/lite/testing/model_coverage/model_coverage_lib.py,47,function,"Returns the full path of the filename.
Args:
filename: Subdirectory and name of the model file.
base_dir: Base directory containing model file.
Returns:
str."
526,get_image,tensorflow/tensorflow/lite/testing/model_coverage/model_coverage_lib.py,63,function,"Returns an image loaded into an np.ndarray with dims [1, size, size, 3].
Args:
size: Size of image.
Returns:
np.ndarray."
527,evaluate_frozen_graph,tensorflow/tensorflow/lite/testing/model_coverage/model_coverage_lib.py,222,function,"Returns a function that evaluates the frozen graph on input data.
Args:
filename: Full filepath of file containing frozen GraphDef.
input_arrays: List of input tensors to freeze graph with.
output_arrays: List of output tensors to freeze graph with.
Returns:
Lambda function ([np.ndarray data] : [np.ndarray result])."
528,evaluate_saved_model,tensorflow/tensorflow/lite/testing/model_coverage/model_coverage_lib.py,260,function,"Returns a function that evaluates the SavedModel on input data.
Args:
directory: SavedModel directory to convert.
tag_set: Set of tags identifying the MetaGraphDef within the SavedModel to
analyze. All tags in the tag set must be present.
signature_key: Key identifying SignatureDef containing inputs and outputs.
Returns:
Lambda function ([np.ndarray data] : [np.ndarray result])."
529,evaluate_keras_model,tensorflow/tensorflow/lite/testing/model_coverage/model_coverage_lib.py,286,function,"Returns a function that evaluates the tf.keras model on input data.
Args:
filename: Full filepath of HDF5 file containing the tf.keras model.
Returns:
Lambda function ([np.ndarray data] : [np.ndarray result])."
530,compare_models,tensorflow/tensorflow/lite/testing/model_coverage/model_coverage_lib.py,299,function,"Compares TensorFlow and TFLite models.
Unless the input data is provided, the models are compared with random data.
Args:
tflite_model: Serialized TensorFlow Lite model.
tf_eval_func: Lambda function that takes in input data and outputs the
results of the TensorFlow model ([np.ndarray data] : [np.ndarray result]).
input_shapes_resize: A map where the key is the input tensor name and the
value is the shape of the input tensor. This resize happens after model
conversion, prior to calling allocate tensors. (default None)
input_data: np.ndarray to pass into models during inference. (default None)
input_data_range: A map where the key is the input tensor name and
the value is a tuple (min_val, max_val) which specifies the value range of
the corresponding input tensor. For example, '{'input1': (1, 5)}' means to
generate a random value for tensor `input1` within range [1.0, 5.0)
(half-inclusive). (default None)
tolerance: Decimal place to check accuracy to. (default 5)."
531,compare_models_v2,tensorflow/tensorflow/lite/testing/model_coverage/model_coverage_lib.py,336,function,"Compares TensorFlow and TFLite models for TensorFlow 2.0.
Unless the input data is provided, the models are compared with random data.
Currently only 1 input and 1 output are supported by this function.
Args:
tflite_model: Serialized TensorFlow Lite model.
tf_eval_func: Function to evaluate TensorFlow model. Either a lambda
function that takes in input data and outputs the results or a TensorFlow
ConcreteFunction.
input_data: np.ndarray to pass into models during inference. (default None).
input_data_range: A map where the key is the input tensor name and
the value is a tuple (min_val, max_val) which specifies the value range of
the corresponding input tensor. For example, '{'input1': (1, 5)}' means to
generate a random value for tensor `input1` within range [1.0, 5.0)
(half-inclusive). (default None)
tolerance: Decimal place to check accuracy to. (default 5)"
532,EvaluateFrozenGraph,tensorflow/tensorflow/lite/testing/model_coverage/model_coverage_lib_test.py,42,class,
533,plus_placeholder,tensorflow/tensorflow/lite/testing/model_coverage/model_coverage_lib_test.py,92,method,
534,EvaluateSavedModel,tensorflow/tensorflow/lite/testing/model_coverage/model_coverage_lib_test.py,142,class,
535,EvaluateKerasModel,tensorflow/tensorflow/lite/testing/model_coverage/model_coverage_lib_test.py,160,class,
536,relu1,tensorflow/tensorflow/lite/testing/op_tests/conv_activation.py,143,function,
537,make_l2_pool,tensorflow/tensorflow/lite/testing/op_tests/pool.py,119,function,Given an input perform a sequence of TensorFlow ops to produce l2pool.
538,html_escape,tensorflow/tensorflow/lite/toco/logging/gen_html.py,37,function,
539,get_input_type_from_signature,tensorflow/tensorflow/lite/toco/logging/gen_html.py,41,function,"Parses op_signature and returns a string denoting the input tensor type.
Args:
op_signature: a string specifying the signature of a particular operator.
The signature of an operator contains the input tensor's shape and type,
output tensor's shape and type, operator's name and its version. It has
the following schema:
INPUT:input_1_shape::input_1_type::input_2_shape::input_2_type::..
::OUTPUT:output_1_shape::output_1_type::output_2_shape::output_2_type::
..::NAME:operator_name ::VERSION:operator_version
An example of an operator signature is:
INPUT:[1,73,73,160]::float::[64,1,1,160]::float::[64]::float::
OUTPUT:[1,73,73,64]::float::NAME:Conv::VERSION:1
Returns:
A string denoting the input tensors' type. In the form of shape/type
separated
by comma. For example:
shape:[1,73,73,160],type:float,shape:[64,1,1,160],type:float,shape:[64],
type:float"
540,get_operator_type,tensorflow/tensorflow/lite/toco/logging/gen_html.py,78,function,
541,HTMLGenerator,tensorflow/tensorflow/lite/toco/logging/gen_html.py,87,class,Utility class to generate an HTML report.
542,generate,tensorflow/tensorflow/lite/toco/logging/gen_html.py,111,method,"Generates the HTML report and writes it to local directory.
This function uses the fields in `toco_conversion_log_before` and
`toco_conversion_log_after` to populate the HTML content. Certain markers
(placeholders) in the HTML template are then substituted with the fields
from the protos. Once finished it will write the HTML file to the specified
local file path.
Args:
toco_conversion_log_before: A `TocoConversionLog` protobuf generated
before the model is converted by TOCO.
toco_conversion_log_after: A `TocoConversionLog` protobuf generated after
the model is converted by TOCO.
post_training_quant_enabled: A boolean, whether post-training quantization
is enabled.
dot_before: A string, the dot representation of the model
before the conversion.
dot_after: A string, the dot representation of the model after
the conversion.
toco_err_log: A string, the logs emitted by TOCO during conversion. Caller
need to ensure that this string is properly anonymized (any kind of
user data should be eliminated).
tflite_graph_path: A string, the filepath to the converted TFLite model.
Raises:
RuntimeError: When error occurs while generating the template."
543,gen_conversion_log_html,tensorflow/tensorflow/lite/toco/logging/gen_html.py,208,function,"Generates an HTML report about the conversion process.
Args:
conversion_log_dir: A string specifying the file directory of the conversion
logs. It's required that before calling this function, the
`conversion_log_dir`
already contains the following files: `toco_log_before.pb`,
`toco_log_after.pb`, `toco_tf_graph.dot`,
`toco_tflite_graph.dot`.
quantization_enabled: A boolean, passed from the tflite converter to
indicate whether post-training quantization is enabled during conversion.
tflite_graph_path: A string, the filepath to the converted TFLite model.
Raises:
IOError: When any of the required files doesn't exist."
544,execute,tensorflow/tensorflow/lite/toco/python/toco_from_protos.py,32,function,Runs the converter.
545,TensorName,tensorflow/tensorflow/lite/toco/python/toco_from_protos_test.py,30,function,Get the canonical (non foo:0 name).
546,get_image,tensorflow/tensorflow/lite/tools/convert_image_to_csv.py,41,function,"Returns an image loaded into an np.ndarray with dims [height, width, (3 or 1)].
Args:
width: Width to rescale the image to.
height: Height to rescale the image to.
want_grayscale: Whether the result should be converted to grayscale.
filepath: Path of the image file..
Returns:
np.ndarray of shape (height, width, channels) where channels is 1 if
want_grayscale is true, otherwise 3."
547,array_to_int_csv,tensorflow/tensorflow/lite/tools/convert_image_to_csv.py,65,function,"Converts all elements in a numerical array to a comma-separated string.
Args:
array_data: Numerical array to convert.
Returns:
String containing array values as integers, separated by commas."
548,convert_bytearray_to_object,tensorflow/tensorflow/lite/tools/flatbuffer_utils.py,38,function,Converts a tflite model from a bytearray to an object for parsing.
549,read_model,tensorflow/tensorflow/lite/tools/flatbuffer_utils.py,44,function,"Reads a tflite model as a python object.
Args:
input_tflite_file: Full path name to the input tflite file
Raises:
RuntimeError: If input_tflite_file path is invalid.
IOError: If input_tflite_file cannot be opened.
Returns:
A python object corresponding to the input tflite file."
550,read_model_with_mutable_tensors,tensorflow/tensorflow/lite/tools/flatbuffer_utils.py,64,function,"Reads a tflite model as a python object with mutable tensors.
Similar to read_model() with the addition that the returned object has
mutable tensors (read_model() returns an object with immutable tensors).
Args:
input_tflite_file: Full path name to the input tflite file
Raises:
RuntimeError: If input_tflite_file path is invalid.
IOError: If input_tflite_file cannot be opened.
Returns:
A mutable python object corresponding to the input tflite file."
551,convert_object_to_bytearray,tensorflow/tensorflow/lite/tools/flatbuffer_utils.py,83,function,Converts a tflite model from an object to a bytearray.
552,write_model,tensorflow/tensorflow/lite/tools/flatbuffer_utils.py,93,function,"Writes the tflite model, a python object, into the output file.
Args:
model_object: A tflite model as a python object
output_tflite_file: Full path name to the output tflite file.
Raises:
IOError: If output_tflite_file path is invalid or cannot be opened."
553,strip_strings,tensorflow/tensorflow/lite/tools/flatbuffer_utils.py,108,function,"Strips all nonessential strings from the model to reduce model size.
We remove the following strings:
(find strings by searching "":string"" in the tensorflow lite flatbuffer schema)
1. Model description
2. SubGraph name
3. Tensor names
We retain OperatorCode custom_code and Metadata name.
Args:
model: The model from which to remove nonessential strings."
554,randomize_weights,tensorflow/tensorflow/lite/tools/flatbuffer_utils.py,130,function,"Randomize weights in a model.
Args:
model: The model in which to randomize weights.
random_seed: The input to the random number generator (default value is 0)."
555,build_mock_flatbuffer_model,tensorflow/tensorflow/lite/tools/test_utils.py,30,function,Creates a flatbuffer containing an example model.
556,load_model_from_flatbuffer,tensorflow/tensorflow/lite/tools/test_utils.py,211,function,Loads a model as a python object from a flatbuffer model.
557,build_mock_model,tensorflow/tensorflow/lite/tools/test_utils.py,218,function,Creates an object containing an example model.
558,TensorTypeToName,tensorflow/tensorflow/lite/tools/visualize.py,202,function,Converts a numerical enum to a readable tensor type.
559,BuiltinCodeToName,tensorflow/tensorflow/lite/tools/visualize.py,210,function,Converts a builtin op code enum to a readable name.
560,NameListToString,tensorflow/tensorflow/lite/tools/visualize.py,218,function,Converts a list of integers to the equivalent ASCII string.
561,OpCodeMapper,tensorflow/tensorflow/lite/tools/visualize.py,229,class,Maps an opcode index to an op name.
562,DataSizeMapper,tensorflow/tensorflow/lite/tools/visualize.py,245,class,"For buffers, report the number of bytes."
563,TensorMapper,tensorflow/tensorflow/lite/tools/visualize.py,255,class,Maps a list of tensor indices to a tooltip hoverable indicator of more.
564,GenerateGraph,tensorflow/tensorflow/lite/tools/visualize.py,278,function,Produces the HTML required to have a d3 visualization of the dag.
565,GenerateTableHtml,tensorflow/tensorflow/lite/tools/visualize.py,337,function,"Given a list of object values and keys to print, make an HTML table.
Args:
items: Items to print an array of dicts.
keys_to_print: (key, display_fn). `key` is a key in the object. i.e.
items[0][key] should exist. display_fn is the mapping function on display.
i.e. the displayed html cell will have the string returned by
`mapping_fn(items[0][key])`.
display_index: add a column which is the index of each row in `items`.
Returns:
An html table."
566,CamelCaseToSnakeCase,tensorflow/tensorflow/lite/tools/visualize.py,375,function,Converts an identifier in CamelCase to snake_case.
567,FlatbufferToDict,tensorflow/tensorflow/lite/tools/visualize.py,381,function,"Converts a hierarchy of FB objects into a nested dict.
We avoid transforming big parts of the flat buffer into python arrays. This
speeds conversion from ten minutes to a few seconds on big graphs.
Args:
fb: a flat buffer structure. (i.e. ModelT)
preserve_as_numpy: true if all downstream np.arrays should be preserved.
false if all downstream np.array should become python arrays
Returns:
A dictionary representing the flatbuffer rather than a flatbuffer object."
568,CreateDictFromFlatbuffer,tensorflow/tensorflow/lite/tools/visualize.py,413,function,
569,CreateHtmlFile,tensorflow/tensorflow/lite/tools/visualize.py,419,function,"Given a tflite model in `tflite_input` file, produce html description."
570,modify_model_interface,tensorflow/tensorflow/lite/tools/optimize/python/modify_model_interface_lib.py,52,function,"Modify a quantized model's interface (input/output) from float to integer.
Args:
input_file: Full path name to the input tflite file.
output_file: Full path name to the output tflite file.
input_type: Final input interface type.
output_type: Final output interface type.
Raises:
RuntimeError: If the modification of the model interface was unsuccessful.
ValueError: If the input_type or output_type is unsupported."
571,build_tflite_model_with_full_integer_quantization,tensorflow/tensorflow/lite/tools/optimize/python/modify_model_interface_lib_test.py,31,function,
572,get_build_cpus,tensorflow/tensorflow/lite/tools/pip_package/setup.py,72,function,
573,make_args,tensorflow/tensorflow/lite/tools/pip_package/setup.py,80,function,Construct make command line.
574,make_output,tensorflow/tensorflow/lite/tools/pip_package/setup.py,94,function,Invoke make on the target and return output.
575,make,tensorflow/tensorflow/lite/tools/pip_package/setup.py,99,function,"Invoke make to build tflite C++ sources.
Build dependencies:
apt-get install swig libjpeg-dev zlib1g-dev python3-dev python3-nump"
576,download_dependencies,tensorflow/tensorflow/lite/tools/pip_package/setup.py,108,function,Download build dependencies if haven't done yet.
577,CustomBuildExt,tensorflow/tensorflow/lite/tools/pip_package/setup.py,114,class,Customized build extension.
578,get_ext_filename,tensorflow/tensorflow/lite/tools/pip_package/setup.py,117,method,
579,run,tensorflow/tensorflow/lite/tools/pip_package/setup.py,123,method,
580,CustomBuildPy,tensorflow/tensorflow/lite/tools/pip_package/setup.py,130,class,
581,run,tensorflow/tensorflow/lite/tools/pip_package/setup.py,132,method,
582,get_pybind_include,tensorflow/tensorflow/lite/tools/pip_package/setup.py,137,function,"pybind11 include directory is not correctly resolved.
This fixes include directory to /usr/local/pythonX.X
Returns:
include directories to find pybind11"
583,set_signature_defs,tensorflow/tensorflow/lite/tools/signature/signature_def_utils.py,25,function,"Sets SignatureDefs to the Metadata of a TfLite flatbuffer buffer.
Args:
tflite_model: Binary TFLite model (bytes or bytes-like object) to which to
add signature_def.
signature_def_map: dict containing SignatureDefs to store in metadata.
Returns:
buffer: A TFLite model binary identical to model buffer with
metadata field containing SignatureDef.
Raises:
ValueError:
tflite_model buffer does not contain a valid TFLite model.
signature_def_map is empty or does not contain a SignatureDef."
584,get_signature_defs,tensorflow/tensorflow/lite/tools/signature/signature_def_utils.py,51,function,"Get SignatureDef dict from the Metadata of a TfLite flatbuffer buffer.
Args:
tflite_model: TFLite model buffer to get the signature_def.
Returns:
dict containing serving names to SignatureDefs if exists, otherwise, empty
dict.
Raises:
ValueError:
tflite_model buffer does not contain a valid TFLite model.
DecodeError:
SignatureDef cannot be parsed from TfLite SignatureDef metadata."
585,clear_signature_defs,tensorflow/tensorflow/lite/tools/signature/signature_def_utils.py,78,function,"Clears SignatureDefs from the Metadata of a TfLite flatbuffer buffer.
Args:
tflite_model: TFLite model buffer to remove signature_defs.
Returns:
buffer: A TFLite model binary identical to model buffer with
no SignatureDef metadata.
Raises:
ValueError:
tflite_model buffer does not contain a valid TFLite model."
586,read32,tensorflow/tensorflow/lite/tutorials/dataset.py,35,function,Read 4 bytes from bytestream as an unsigned 32-bit integer.
587,check_image_file_header,tensorflow/tensorflow/lite/tutorials/dataset.py,41,function,Validate that filename corresponds to images for the MNIST dataset.
588,check_labels_file_header,tensorflow/tensorflow/lite/tutorials/dataset.py,57,function,Validate that filename corresponds to labels for the MNIST dataset.
589,download,tensorflow/tensorflow/lite/tutorials/dataset.py,67,function,Download (and unzip) a file from the MNIST dataset if not already done.
590,dataset,tensorflow/tensorflow/lite/tutorials/dataset.py,86,function,Download and parse MNIST dataset.
591,train,tensorflow/tensorflow/lite/tutorials/dataset.py,114,function,tf.data.Dataset object for MNIST training data.
592,run_eval,tensorflow/tensorflow/lite/tutorials/mnist_tflite.py,47,function,"Performs evaluation for input image over specified model.
Args:
interpreter: TFLite interpreter initialized with model to execute.
input_image: Image input to the model.
Returns:
output: output tensor of model being executed."
593,set_dlopen_flags,tensorflow/tensorflow/python/pywrap_dlopen_global_flags.py,43,function,
594,reset_dlopen_flags,tensorflow/tensorflow/python/pywrap_dlopen_global_flags.py,48,function,
595,import_graphdef,tensorflow/tensorflow/python/pywrap_mlir.py,26,function,
596,experimental_convert_saved_model_to_mlir,tensorflow/tensorflow/python/pywrap_mlir.py,32,function,
597,experimental_convert_saved_model_v1_to_mlir,tensorflow/tensorflow/python/pywrap_mlir.py,39,function,
598,experimental_run_pass_pipeline,tensorflow/tensorflow/python/pywrap_mlir.py,48,function,
599,enable,tensorflow/tensorflow/python/tf2.py,30,function,
600,disable,tensorflow/tensorflow/python/tf2.py,36,function,
601,enabled,tensorflow/tensorflow/python/tf2.py,42,function,
602,AssertTransformer,tensorflow/tensorflow/python/autograph/converters/asserts.py,27,class,Transforms Assert nodes to Call so they can be handled as functions.
603,visit_Assert,tensorflow/tensorflow/python/autograph/converters/asserts.py,30,method,
604,transform,tensorflow/tensorflow/python/autograph/converters/asserts.py,50,function,
605,BreakTransformer,tensorflow/tensorflow/python/autograph/converters/break_statements.py,39,class,Canonicalizes break statements into additional conditionals.
606,visit_Break,tensorflow/tensorflow/python/autograph/converters/break_statements.py,42,method,
607,visit_While,tensorflow/tensorflow/python/autograph/converters/break_statements.py,75,method,
608,visit_For,tensorflow/tensorflow/python/autograph/converters/break_statements.py,121,method,
609,transform,tensorflow/tensorflow/python/autograph/converters/break_statements.py,183,function,
610,CallTreeTransformer,tensorflow/tensorflow/python/autograph/converters/call_trees.py,96,class,Transforms the call tree by renaming transformed symbols.
611,visit_Lambda,tensorflow/tensorflow/python/autograph/converters/call_trees.py,99,method,
612,visit_FunctionDef,tensorflow/tensorflow/python/autograph/converters/call_trees.py,108,method,
613,visit_With,tensorflow/tensorflow/python/autograph/converters/call_trees.py,126,method,
614,visit_Call,tensorflow/tensorflow/python/autograph/converters/call_trees.py,160,method,
615,transform,tensorflow/tensorflow/python/autograph/converters/call_trees.py,211,function,"Transform function call to the compiled counterparts.
Args:
node: AST
ctx: EntityContext
Returns:
A tuple (node, new_names):
node: The transformed AST
new_names: set(string), containing any newly-generated names"
616,MockConvertedCall,tensorflow/tensorflow/python/autograph/converters/call_trees_test.py,30,class,
617,ConditionalExpressionTransformer,tensorflow/tensorflow/python/autograph/converters/conditional_expressions.py,28,class,Converts conditional expressions to functional form.
618,visit_IfExp,tensorflow/tensorflow/python/autograph/converters/conditional_expressions.py,31,method,
619,transform,tensorflow/tensorflow/python/autograph/converters/conditional_expressions.py,48,function,
620,ContinueCanonicalizationTransformer,tensorflow/tensorflow/python/autograph/converters/continue_statements.py,60,class,Canonicalizes continue statements into additional conditionals.
621,visit_Continue,tensorflow/tensorflow/python/autograph/converters/continue_statements.py,63,method,
622,visit_While,tensorflow/tensorflow/python/autograph/converters/continue_statements.py,125,method,
623,visit_For,tensorflow/tensorflow/python/autograph/converters/continue_statements.py,132,method,
624,visit_If,tensorflow/tensorflow/python/autograph/converters/continue_statements.py,140,method,
625,visit_With,tensorflow/tensorflow/python/autograph/converters/continue_statements.py,145,method,
626,visit_Try,tensorflow/tensorflow/python/autograph/converters/continue_statements.py,150,method,
627,visit_ExceptHandler,tensorflow/tensorflow/python/autograph/converters/continue_statements.py,158,method,
628,transform,tensorflow/tensorflow/python/autograph/converters/continue_statements.py,163,function,
629,ControlFlowTransformer,tensorflow/tensorflow/python/autograph/converters/control_flow.py,46,class,Transforms control flow structures like loops an conditionals.
630,visit_Lambda,tensorflow/tensorflow/python/autograph/converters/control_flow.py,49,method,
631,visit_FunctionDef,tensorflow/tensorflow/python/autograph/converters/control_flow.py,54,method,
632,visit_If,tensorflow/tensorflow/python/autograph/converters/control_flow.py,206,method,
633,visit_While,tensorflow/tensorflow/python/autograph/converters/control_flow.py,261,method,
634,visit_For,tensorflow/tensorflow/python/autograph/converters/control_flow.py,309,method,
635,AnnotatedDef,tensorflow/tensorflow/python/autograph/converters/control_flow.py,395,class,
636,transform,tensorflow/tensorflow/python/autograph/converters/control_flow.py,402,function,
637,ControlFlowTransformer,tensorflow/tensorflow/python/autograph/converters/control_flow_deprecated_py2.py,44,class,Transforms control flow structures like loops an conditionals.
638,visit_If,tensorflow/tensorflow/python/autograph/converters/control_flow_deprecated_py2.py,200,method,
639,visit_While,tensorflow/tensorflow/python/autograph/converters/control_flow_deprecated_py2.py,400,method,
640,visit_For,tensorflow/tensorflow/python/autograph/converters/control_flow_deprecated_py2.py,490,method,
641,AnnotatedDef,tensorflow/tensorflow/python/autograph/converters/control_flow_deprecated_py2.py,623,class,
642,transform,tensorflow/tensorflow/python/autograph/converters/control_flow_deprecated_py2.py,630,function,
643,DirectivesTransformer,tensorflow/tensorflow/python/autograph/converters/directives.py,90,class,Parses compiler directives and converts them into AST annotations.
644,visit_Name,tensorflow/tensorflow/python/autograph/converters/directives.py,117,method,
645,visit_Attribute,tensorflow/tensorflow/python/autograph/converters/directives.py,126,method,
646,visit_Assign,tensorflow/tensorflow/python/autograph/converters/directives.py,134,method,
647,visit_AugAssign,tensorflow/tensorflow/python/autograph/converters/directives.py,138,method,
648,visit_Expr,tensorflow/tensorflow/python/autograph/converters/directives.py,142,method,
649,visit_While,tensorflow/tensorflow/python/autograph/converters/directives.py,173,method,
650,visit_For,tensorflow/tensorflow/python/autograph/converters/directives.py,176,method,
651,transform,tensorflow/tensorflow/python/autograph/converters/directives.py,180,function,
652,FunctionTransformer,tensorflow/tensorflow/python/autograph/converters/functions.py,38,class,Wraps function bodies around autograph-specific boilerplate.
653,visit_Lambda,tensorflow/tensorflow/python/autograph/converters/functions.py,53,method,
654,visit_FunctionDef,tensorflow/tensorflow/python/autograph/converters/functions.py,81,method,
655,transform,tensorflow/tensorflow/python/autograph/converters/functions.py,134,function,
656,FunctionTransformer,tensorflow/tensorflow/python/autograph/converters/functions_test.py,31,class,
657,f,tensorflow/tensorflow/python/autograph/converters/functions_test.py,35,method,Docstring.
658,f,tensorflow/tensorflow/python/autograph/converters/functions_test.py,49,method,"First sentence.
Second sentence.
Returns:
Something."
659,f,tensorflow/tensorflow/python/autograph/converters/functions_test.py,68,method,
660,inner_fn_callee,tensorflow/tensorflow/python/autograph/converters/functions_test.py,85,method,
661,f,tensorflow/tensorflow/python/autograph/converters/functions_test.py,89,method,
662,f,tensorflow/tensorflow/python/autograph/converters/functions_test.py,121,method,
663,inner_fn,tensorflow/tensorflow/python/autograph/converters/functions_test.py,70,method,
664,inner_fn,tensorflow/tensorflow/python/autograph/converters/functions_test.py,90,method,
665,f,tensorflow/tensorflow/python/autograph/converters/functions_test.py,104,method,
666,inner_fn,tensorflow/tensorflow/python/autograph/converters/functions_test.py,106,method,
667,ListCompTransformer,tensorflow/tensorflow/python/autograph/converters/list_comprehensions.py,42,class,Lowers list comprehensions into standard control flow.
668,visit_Assign,tensorflow/tensorflow/python/autograph/converters/list_comprehensions.py,45,method,
669,transform,tensorflow/tensorflow/python/autograph/converters/list_comprehensions.py,81,function,
670,ListTransformer,tensorflow/tensorflow/python/autograph/converters/lists.py,51,class,Converts lists and related operations to their TF counterpart.
671,visit_List,tensorflow/tensorflow/python/autograph/converters/lists.py,54,method,
672,visit_Call,tensorflow/tensorflow/python/autograph/converters/lists.py,131,method,
673,visit_FunctionDef,tensorflow/tensorflow/python/autograph/converters/lists.py,209,method,
674,visit_For,tensorflow/tensorflow/python/autograph/converters/lists.py,215,method,
675,visit_While,tensorflow/tensorflow/python/autograph/converters/lists.py,221,method,
676,visit_If,tensorflow/tensorflow/python/autograph/converters/lists.py,227,method,
677,visit_With,tensorflow/tensorflow/python/autograph/converters/lists.py,233,method,
678,transform,tensorflow/tensorflow/python/autograph/converters/lists.py,239,function,
679,LogicalExpressionTransformer,tensorflow/tensorflow/python/autograph/converters/logical_expressions.py,49,class,Converts logical expressions to corresponding TF calls.
680,visit_Compare,tensorflow/tensorflow/python/autograph/converters/logical_expressions.py,83,method,
681,visit_UnaryOp,tensorflow/tensorflow/python/autograph/converters/logical_expressions.py,114,method,
682,visit_BoolOp,tensorflow/tensorflow/python/autograph/converters/logical_expressions.py,123,method,
683,transform,tensorflow/tensorflow/python/autograph/converters/logical_expressions.py,135,function,
684,ConditionalReturnRewriter,tensorflow/tensorflow/python/autograph/converters/return_statements.py,43,class,"Rewrites a a pattern where it's unobvious that all paths return a value.
This rewrite allows avoiding intermediate None return values.
The following pattern:
if cond:
<block 1>
return
else:
<block 2>
<block 3>
is converted to:
if cond:
<block 1>
return
else:
<block 2>
<block 3>
and vice-versa (if the else returns, subsequent statements are moved under the
if branch)."
685,visit_Return,tensorflow/tensorflow/python/autograph/converters/return_statements.py,70,method,
686,visit_While,tensorflow/tensorflow/python/autograph/converters/return_statements.py,100,method,
687,visit_For,tensorflow/tensorflow/python/autograph/converters/return_statements.py,106,method,
688,visit_With,tensorflow/tensorflow/python/autograph/converters/return_statements.py,113,method,
689,visit_Try,tensorflow/tensorflow/python/autograph/converters/return_statements.py,120,method,
690,visit_ExceptHandler,tensorflow/tensorflow/python/autograph/converters/return_statements.py,130,method,
691,visit_If,tensorflow/tensorflow/python/autograph/converters/return_statements.py,135,method,
692,visit_FunctionDef,tensorflow/tensorflow/python/autograph/converters/return_statements.py,153,method,
693,ReturnStatementsTransformer,tensorflow/tensorflow/python/autograph/converters/return_statements.py,183,class,"Lowers return statements into variables and conditionals.
Specifically, the following pattern:
<block 1>
return val
<block 2>
is converted to:
do_return = False
retval = None
<block 1>
do_return = True
retval = val
if not do_return:
<block 2>
return retval
The conversion adjusts loops as well:
<block 1>
while cond:
<block 2>
return retval
is converted to:
<block 1>
while not do_return and cond:
<block 2>
do_return = True
retval = val"
694,visit_Return,tensorflow/tensorflow/python/autograph/converters/return_statements.py,227,method,
695,visit_While,tensorflow/tensorflow/python/autograph/converters/return_statements.py,283,method,
696,visit_For,tensorflow/tensorflow/python/autograph/converters/return_statements.py,297,method,
697,visit_With,tensorflow/tensorflow/python/autograph/converters/return_statements.py,319,method,
698,visit_Try,tensorflow/tensorflow/python/autograph/converters/return_statements.py,324,method,
699,visit_ExceptHandler,tensorflow/tensorflow/python/autograph/converters/return_statements.py,331,method,
700,visit_If,tensorflow/tensorflow/python/autograph/converters/return_statements.py,335,method,
701,visit_FunctionDef,tensorflow/tensorflow/python/autograph/converters/return_statements.py,341,method,
702,transform,tensorflow/tensorflow/python/autograph/converters/return_statements.py,392,function,"Ensure a function has only a single return, at the end."
703,SliceTransformer,tensorflow/tensorflow/python/autograph/converters/slices.py,28,class,"Converts slicing operations to their TF counterpart.
Currently, relying on the default slice operator that Tensor uses is
insufficient, because TensorArray and tensor lists use dedicated index read
and write functions."
704,visit_Assign,tensorflow/tensorflow/python/autograph/converters/slices.py,48,method,
705,visit_Subscript,tensorflow/tensorflow/python/autograph/converters/slices.py,58,method,
706,transform,tensorflow/tensorflow/python/autograph/converters/slices.py,84,function,
707,VariableAccessTransformer,tensorflow/tensorflow/python/autograph/converters/variables.py,28,class,"Rewrites basic symbol reads.
This transformer rewrites variable reads with a ""read"" operator which allows
tracking activity.
Example:
For a basic statement:
a = b + c
This is translated to:
a = ld(b) + ld(c)
Augmented assignment operations also introduce a `ld` operator:
a += b
The assignment target also receives an operator to properly represent the
read:
a = ld(a)
a += ld(b)"
708,visit_Name,tensorflow/tensorflow/python/autograph/converters/variables.py,55,method,
709,visit_Delete,tensorflow/tensorflow/python/autograph/converters/variables.py,63,method,
710,visit_AugAssign,tensorflow/tensorflow/python/autograph/converters/variables.py,88,method,
711,transform,tensorflow/tensorflow/python/autograph/converters/variables.py,100,function,
712,control_status_ctx,tensorflow/tensorflow/python/autograph/core/ag_ctx.py,35,function,
713,Status,tensorflow/tensorflow/python/autograph/core/ag_ctx.py,40,class,
714,ControlStatusCtx,tensorflow/tensorflow/python/autograph/core/ag_ctx.py,46,class,A context that tracks whether autograph is enabled by the user.
715,NullCtx,tensorflow/tensorflow/python/autograph/core/ag_ctx.py,66,class,Helper substitute for contextlib.nullcontext.
716,Rule,tensorflow/tensorflow/python/autograph/core/config_lib.py,27,class,Base class for conversion rules.
717,matches,tensorflow/tensorflow/python/autograph/core/config_lib.py,33,method,
718,Action,tensorflow/tensorflow/python/autograph/core/config_lib.py,38,class,
719,DoNotConvert,tensorflow/tensorflow/python/autograph/core/config_lib.py,44,class,Indicates that this module should be not converted.
720,get_action,tensorflow/tensorflow/python/autograph/core/config_lib.py,50,method,
721,Convert,tensorflow/tensorflow/python/autograph/core/config_lib.py,56,class,Indicates that this module should be converted.
722,get_action,tensorflow/tensorflow/python/autograph/core/config_lib.py,62,method,
723,Feature,tensorflow/tensorflow/python/autograph/core/converter.py,83,class,"This enumeration represents optional conversion options.
These conversion options are experimental. They are subject to change without
notice and offer no guarantees.
_Example Usage_
```python
optionals= tf.autograph.experimental.Feature.EQUALITY_OPERATORS
@tf.function(experimental_autograph_options=optionals)
def f(i):
if i == 0: # EQUALITY_OPERATORS allows the use of == here.
tf.print('i is zero')
```
Attributes:
ALL: Enable all features.
AUTO_CONTROL_DEPS: Insert of control dependencies in the generated code.
ASSERT_STATEMENTS: Convert Tensor-dependent assert statements to tf.Assert.
BUILTIN_FUNCTIONS: Convert builtin functions applied to Tensors to
their TF counterparts.
EQUALITY_OPERATORS: Whether to convert the comparison operators, like
equality. This is soon to be deprecated as support is being added to the
Tensor class.
LISTS: Convert list idioms, like initializers, slices, append, etc.
NAME_SCOPES: Insert name scopes that name ops according to context, like the
function they were defined in."
724,all,tensorflow/tensorflow/python/autograph/core/converter.py,123,method,Returns a tuple that enables all options.
725,all_but,tensorflow/tensorflow/python/autograph/core/converter.py,128,method,Returns a tuple that enables all but the excluded options.
726,ConversionOptions,tensorflow/tensorflow/python/autograph/core/converter.py,138,class,"Immutable container for global conversion flags.
Attributes:
recursive: bool, whether to recursively convert any user functions or
classes that the converted function may use.
user_requested: bool, whether the conversion was explicitly requested by
the user, as opposed to being performed as a result of other logic. This
value always auto-resets resets to False in child conversions.
optional_features: Union[Feature, Set[Feature]], controls the use of
optional features in the conversion process. See Feature for available
options."
727,as_tuple,tensorflow/tensorflow/python/autograph/core/converter.py,169,method,
728,uses,tensorflow/tensorflow/python/autograph/core/converter.py,183,method,
729,call_options,tensorflow/tensorflow/python/autograph/core/converter.py,187,method,Returns the corresponding options to be used for recursive conversion.
730,to_ast,tensorflow/tensorflow/python/autograph/core/converter.py,195,method,"Returns a representation of this object as an AST node.
The AST node encodes a constructor that would create an object with the
same contents.
Returns:
ast.Node"
731,list_of_features,tensorflow/tensorflow/python/autograph/core/converter.py,215,method,
732,ProgramContext,tensorflow/tensorflow/python/autograph/core/converter.py,236,class,"ProgramContext keeps track of converting function hierarchies.
Attributes:
options: ConversionOptions
autograph_module: Deprecated. Do not use."
733,Base,tensorflow/tensorflow/python/autograph/core/converter.py,249,class,"All converters should inherit from this class.
Attributes:
ctx: EntityContext"
734,get_definition_directive,tensorflow/tensorflow/python/autograph/core/converter.py,262,method,"Returns the unique directive argument for a symbol.
See lang/directives.py for details on directives.
Example:
# Given a directive in the code:
ag.foo_directive(bar, baz=1)
# One can write for an AST node Name(id='bar'):
get_definition_directive(node, ag.foo_directive, 'baz')
Args:
node: ast.AST, the node representing the symbol for which the directive
argument is needed.
directive: Callable[..., Any], the directive to search.
arg: str, the directive argument to return.
default: Any
Raises:
ValueError: if conflicting annotations have been found"
735,visit,tensorflow/tensorflow/python/autograph/core/converter.py,311,method,
736,allowlist,tensorflow/tensorflow/python/autograph/core/converter_testing.py,35,function,Helper that marks a callable as whtelitisted.
737,is_inside_generated_code,tensorflow/tensorflow/python/autograph/core/converter_testing.py,47,function,Tests whether the caller is generated code. Implementation-specific.
738,FunctionScope,tensorflow/tensorflow/python/autograph/core/function_wrappers.py,33,class,"Context manager that wraps the body of a converted function.
This context manager handles various operations related to the scope of a
function:
* optional TF name scopes - these name scopes match the name of the
function, for easy visualization in tensorBoard;
* optional automatic control dependencies - this adds the same mechanism
for control dependencies that is used by `@tf.function`; it can be
optionally enabled when using `tf.autograph.to_graph`;
* tracking of autograph conversion state (whether it's enabled by the user,
conversion options;"
739,ret,tensorflow/tensorflow/python/autograph/core/function_wrappers.py,91,method,Marks a value as returned from the function guarded by the scope.
740,with_function_scope,tensorflow/tensorflow/python/autograph/core/function_wrappers.py,114,function,Inline version of the FunctionScope context manager.
741,UnsupportedFeaturesChecker,tensorflow/tensorflow/python/autograph/core/unsupported_features_checker.py,26,class,"Quick check for Python features we know we don't support.
Any features detected will cause AutoGraph to not compile a function."
742,visit_Attribute,tensorflow/tensorflow/python/autograph/core/unsupported_features_checker.py,32,method,
743,visit_For,tensorflow/tensorflow/python/autograph/core/unsupported_features_checker.py,39,method,
744,visit_While,tensorflow/tensorflow/python/autograph/core/unsupported_features_checker.py,45,method,
745,visit_Yield,tensorflow/tensorflow/python/autograph/core/unsupported_features_checker.py,53,method,
746,visit_YieldFrom,tensorflow/tensorflow/python/autograph/core/unsupported_features_checker.py,56,method,
747,verify,tensorflow/tensorflow/python/autograph/core/unsupported_features_checker.py,60,function,
748,is_autograph_strict_conversion_mode,tensorflow/tensorflow/python/autograph/impl/api.py,72,function,
749,AutoGraphError,tensorflow/tensorflow/python/autograph/impl/api.py,82,class,Base class for all AutoGraph exceptions.
750,ConversionError,tensorflow/tensorflow/python/autograph/impl/api.py,87,class,Raised during the conversion process.
751,StagingError,tensorflow/tensorflow/python/autograph/impl/api.py,92,class,Raised during the staging (i.e. Python execution) of converted code.
752,StackTraceMapper,tensorflow/tensorflow/python/autograph/impl/api.py,166,class,Remaps generated code to code it originated from.
753,get_effective_source_map,tensorflow/tensorflow/python/autograph/impl/api.py,172,method,
754,PyToTF,tensorflow/tensorflow/python/autograph/impl/api.py,203,class,The TensorFlow AutoGraph transformer.
755,get_transformed_name,tensorflow/tensorflow/python/autograph/impl/api.py,228,method,
756,get_extra_locals,tensorflow/tensorflow/python/autograph/impl/api.py,231,method,
757,get_caching_key,tensorflow/tensorflow/python/autograph/impl/api.py,234,method,
758,initial_analysis,tensorflow/tensorflow/python/autograph/impl/api.py,237,method,
759,transform_ast,tensorflow/tensorflow/python/autograph/impl/api.py,250,method,
760,autograph_artifact,tensorflow/tensorflow/python/autograph/impl/api.py,298,function,
761,is_autograph_artifact,tensorflow/tensorflow/python/autograph/impl/api.py,303,function,
762,converted_call,tensorflow/tensorflow/python/autograph/impl/api.py,307,function,"Converts a function call inline.
For internal use only.
Note: The argument list is optimized for readability of generated code, which
may look like this:
ag__.converted_call(f, (arg1, arg2), None, fscope)
ag__.converted_call(f, (), dict(arg1=val1, **kwargs), fscope)
ag__.converted_call(f, (arg1, arg2) + varargs, dict(**kwargs), lscope)
Args:
f: The function to convert.
args: Tuple, the original positional arguments of f
kwargs: Optional[Dict], the original keyword arguments of f
caller_fn_scope: Optional[function_wrappers.FunctionScope], the function
scope of the converted function in which this call was originally made.
options: Optional[converter.ConversionOptions], conversion options. If not
specified, the value of caller_fn_scope.callopts is used. Either options
or caller_fn_scope must be present.
Returns:
Any, the result of executing a possibly-converted `f` with the given
arguments."
763,tf_convert,tensorflow/tensorflow/python/autograph/impl/api.py,506,function,"Decorator that applies AutoGraph to a function.
Use in internal APIs.
This API is suitable for high order functions internal to the TensorFlow API,
and more generally any function to which Autograph is not applied.
Guidance: convert was a decorator meant for use directly by developers, and
will be soon deprecated in favor of tf.function. tf_convert is to be called
from high order functions internal to TF.
Args:
f: Callable.
ctx: ag_ctx.ControlStatusCtx, the Autograph context in which `f` is used.
convert_by_default: bool, whether to use AutoGraph when the context doesn't
specify.
user_requested: bool, whether to ignore the conversion allowlist. See
ConversionOptions.user_requested.
Returns:
Either `f or the converted version of `f`."
764,call_with_unspecified_conversion_status,tensorflow/tensorflow/python/autograph/impl/api.py,565,function,Decorator that resets the conversion context to the unspecified status.
765,do_not_convert,tensorflow/tensorflow/python/autograph/impl/api.py,599,function,"Decorator that suppresses the conversion of a function.
Args:
func: function to decorate.
Returns:
If `func` is not None, returns a `Callable` which is equivalent to
`func`, but is not converted by AutoGraph.
If `func` is None, returns a decorator that, when invoked with a
single `func` argument, returns a `Callable` equivalent to the
above case."
766,convert,tensorflow/tensorflow/python/autograph/impl/api.py,626,function,"Decorator that compiles a function to use TensorFlow ops.
The decorator is dynamic - it recompiles the target whenever the decorated
function is called. This means the parameter values are known at conversion.
It also means that repeated calls with different types of parameters will be
correctly processed.
Args:
recursive: bool, whether to recursively convert any functions or classes
that the converted function may use.
optional_features: converted.Feature, allows toggling optional or
experimental features. When set to None, only the core features are
enabled.
user_requested: bool, whether this is a function that the user explicitly
asked to be converted. See ConversionOptions.user_requested.
conversion_ctx: Optional ag_ctx.ControlStatusCtx, the Autograph context in
which `f` is used.
Returns:
Callable, a decorator that converts the given function into an equivalent
function that uses TensorFlow ops."
767,to_graph,tensorflow/tensorflow/python/autograph/impl/api.py,682,function,"Converts a Python entity into a TensorFlow graph.
Also see: `tf.autograph.to_code`, `tf.function`.
Unlike `tf.function`, `to_graph` is a low-level transpiler that converts
Python code to TensorFlow graph code. It does not implement any caching,
variable management or create any actual ops, and is best used where greater
control over the generated TensorFlow graph is desired. Another difference
from `tf.function` is that `to_graph` will not wrap the graph into a
TensorFlow function or a Python callable. Internally, `tf.function` uses
`to_graph`.
Example usage:
>>> def f(x):
... if x > 0:
... y = x * x
... else:
... y = -x
... return y
...
>>> converted_f = to_graph(f)
>>> x = tf.constant(2)
>>> converted_f(x) # converted_foo is like a TensorFlow Op.
<tf.Tensor: shape=(), dtype=int32, numpy=4>
Supported Python entities include:
* functions
* classes
* object methods
Functions are converted into new functions with converted code.
Classes are converted by generating a new class whose methods use converted
code.
Methods are converted into unbound function that have an additional first
argument called `self`.
For a tutorial, see the
[tf.function and AutoGraph guide](https://www.tensorflow.org/guide/function).
For more detailed information, see the
[AutoGraph reference documentation](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/autograph/g3doc/reference/index.md).
Args:
entity: Python callable or class to convert.
recursive: Whether to recursively convert any functions that the converted
function may call.
experimental_optional_features: `None`, a tuple of, or a single
`tf.autograph.experimental.Feature` value.
Returns:
Same as `entity`, the converted Python function or class.
Raises:
ValueError: If the entity could not be converted."
768,to_graph_v1,tensorflow/tensorflow/python/autograph/impl/api.py,754,function,"Converts a Python entity into a TensorFlow graph.
Also see: `tf.autograph.to_code`, `tf.function`.
Unlike `tf.function`, `to_graph` is a low-level transpiler that converts
Python code to TensorFlow graph code. It does not implement any caching,
variable management or create any actual ops, and is best used where greater
control over the generated TensorFlow graph is desired. Another difference
from `tf.function` is that `to_graph` will not wrap the graph into a
TensorFlow function or a Python callable. Internally, `tf.function` uses
`to_graph`.
_Example Usage_
```python
def foo(x):
if x > 0:
y = x * x
else:
y = -x
return y
converted_foo = to_graph(foo)
x = tf.constant(1)
y = converted_foo(x) # converted_foo is a TensorFlow Op-like.
assert is_tensor(y)
```
Supported Python entities include:
* functions
* classes
* object methods
Functions are converted into new functions with converted code.
Classes are converted by generating a new class whose methods use converted
code.
Methods are converted into unbound function that have an additional first
argument called `self`.
Args:
entity: Python callable or class to convert.
recursive: Whether to recursively convert any functions that the converted
function may call.
arg_values: Deprecated.
arg_types: Deprecated.
experimental_optional_features: `None`, a tuple of, or a single
`tf.autograph.experimental.Feature` value.
Returns:
Same as `entity`, the converted Python function or class.
Raises:
ValueError: If the entity could not be converted."
769,to_code_v1,tensorflow/tensorflow/python/autograph/impl/api.py,825,function,"Returns the source code generated by AutoGraph, as a string.
Example usage:
>>> def f(x):
... if x < 0:
... x = -x
... return x
>>> tf.autograph.to_code(f)
""...def tf__f(x):...""
Also see: `tf.autograph.to_graph`.
Note: If a function has been decorated with `tf.function`, pass its
underlying Python function, rather than the callable that `tf.function
creates:
>>> @tf.function
... def f(x):
... if x < 0:
... x = -x
... return x
>>> tf.autograph.to_code(f.python_function)
""...def tf__f(x):...""
Args:
entity: Python callable or class.
recursive: Whether to recursively convert any functions that the converted
function may call.
arg_values: Deprecated.
arg_types: Deprecated.
indentation: Deprecated.
experimental_optional_features: `None`, a tuple of, or a single
`tf.autograph.experimental.Feature` value.
Returns:
The converted code as string."
770,to_code,tensorflow/tensorflow/python/autograph/impl/api.py,879,function,"Returns the source code generated by AutoGraph, as a string.
Example usage:
>>> def f(x):
... if x < 0:
... x = -x
... return x
>>> tf.autograph.to_code(f)
""...def tf__f(x):...""
Also see: `tf.autograph.to_graph`.
Note: If a function has been decorated with `tf.function`, pass its
underlying Python function, rather than the callable that `tf.function
creates:
>>> @tf.function
... def f(x):
... if x < 0:
... x = -x
... return x
>>> tf.autograph.to_code(f.python_function)
""...def tf__f(x):...""
Args:
entity: Python callable or class to convert.
recursive: Whether to recursively convert any functions that the converted
function may call.
experimental_optional_features: `None`, a tuple of, or a single
`tf.autograph.experimental.Feature` value.
Returns:
The converted code as string."
771,is_unsupported,tensorflow/tensorflow/python/autograph/impl/conversion.py,73,function,Checks whether an entity is supported by AutoGraph at all.
772,is_allowlisted,tensorflow/tensorflow/python/autograph/impl/conversion.py,116,function,"Checks whether an entity is allowed for use in graph mode.
Examples of allowed entities include all members of the tensorflow
package.
Args:
o: A Python entity.
check_call_override: Reserved for internal use. When set to `False`, it
disables the rule according to which classes are allowed if their
__call__ method is allowed.
allow_namedtuple_subclass: Reserved for internal use. When `True`,
namedtuple subclasses are not allowed.
Returns:
Boolean"
773,is_in_allowlist_cache,tensorflow/tensorflow/python/autograph/impl/conversion.py,221,function,
774,cache_allowlisted,tensorflow/tensorflow/python/autograph/impl/conversion.py,229,function,
775,set_element_type,tensorflow/tensorflow/python/autograph/lang/directives.py,33,function,"Indicates that the entity is expected hold items of specified type/shape.
The staged TensorFlow ops will reflect and assert this data type. Ignored
otherwise.
Args:
entity: The entity to annotate.
dtype: TensorFlow dtype value to assert for entity.
shape: Optional shape to assert for entity."
776,set_loop_options,tensorflow/tensorflow/python/autograph/lang/directives.py,50,function,"Specifies additional arguments to be passed to the enclosing while_loop.
The parameters apply to and only to the immediately enclosing loop. It only
has effect if the loop is staged as a TF while_loop; otherwise the parameters
have no effect.
Usage:
>>> @tf.function(autograph=True)
... def f():
... n = 0
... for i in tf.range(10):
... tf.autograph.experimental.set_loop_options(maximum_iterations=3)
... n += 1
... return n
>>> @tf.function(autograph=True)
... def f():
... v = tf.constant((0,))
... for i in tf.range(3):
... tf.autograph.experimental.set_loop_options(
... shape_invariants=[(v, tf.TensorShape([None]))]
... )
... v = tf.concat((v, [i]), 0)
... return v
Also see tf.while_loop.
Args:
parallel_iterations: The maximum number of iterations allowed to run in
parallel at any given time. Note that this does not guarantee parallel
execution.
swap_memory: Whether to store intermediate values needed for
gradients on the CPU instead of GPU.
maximum_iterations: Allows limiting the total number of iterations executed
by the loop.
shape_invariants: Allows controlling the argument with the same name passed
to tf.while_loop. Unlike tf.while_loop, this is a list of
`(tensor, shape)` pairs."
777,match_staging_level,tensorflow/tensorflow/python/autograph/lang/special_functions.py,50,function,Casts a value to be staged at the same level as another.
778,tensor_list,tensorflow/tensorflow/python/autograph/lang/special_functions.py,57,function,"Creates an tensor list and populates it with the given elements.
This function provides a more uniform access to tensor lists and tensor
arrays, and allows optional initialization.
Note: this function is a simplified wrapper. If you need greater control,
it is recommended to use the underlying implementation directly.
Args:
elements: Iterable[tf.Tensor, ...], the elements to initially fill the list
with
element_dtype: Optional[tf.DType], data type for the elements in the list;
required if the list is empty
element_shape: Optional[tf.TensorShape], shape for the elements in the list;
required if the list is empty
use_tensor_array: bool, whether to use the more compatible but restrictive
tf.TensorArray implementation
Returns:
Union[tf.Tensor, tf.TensorArray], the new list.
Raises:
ValueError: for invalid arguments"
779,stack,tensorflow/tensorflow/python/autograph/lang/special_functions.py,92,function,"Stacks the input, if it admits the notion of stacking.
For example, a list of tensors can be stacked into a larger tensor. This
function is similar to tf.stack, but it accepts non-lists and lists of
non-tensors as arguments. In the latter case, the function does nothing.
Args:
list_or_tensor: Any
element_dtype: tf.DType, optional dtypedtype for the elements in the list.
Required if the input is stackable, and the list is untyped.
strict: bool, if True an error is raised if the input is not stackable.
Otherwise the function is a no-op.
Returns:
Any, if the input is stackable, the result will be a tf.Tensor. Otherwise,
if strict=False, the result will be list_or_tensor.
Raises:
ValueError: if strict=True and the input is not stackable."
780,if_exp,tensorflow/tensorflow/python/autograph/operators/conditional_expressions.py,27,function,
781,verify_single_cond_var,tensorflow/tensorflow/python/autograph/operators/control_flow.py,233,function,Verifies whether body_var and orelse_var are consistent.
782,for_stmt,tensorflow/tensorflow/python/autograph/operators/control_flow.py,291,function,"Functional form of a for statement.
The loop operates on a state, which includes all symbols that are
variant across loop iterations, excluding the variables local to the loop.
For example, given the loop below that calculates the geometric and
arithmetic means or some numbers:
```
geo_mean = 1
arith_mean = 0
for i in range(n):
a = numbers[i]
geo_mean *= a
arith_mean += a
```
The state is represented by the variables geo_mean and arith_mean. The
`extra_test`, `body`, `get_state` and `set_state` functions must bind to the
original `geo_mean` and `arith_mean` symbols, using `nonlocal`.
The inputs and outputs of the callables representing the loop blocks are not
explicit - instead, these functions must use nonlocal/global for side effects.
The inputs and outputs are instead controlled by the set_state/get_state
functions.
Args:
iter_: The entity being iterated over.
extra_test: Callable with boolean return type.
An additional loop condition.
body: Callable representing the actual loop body.
get_state: Additional callable which can capture additional state (such as
the values of composite symbols). This is only useful when staging the
loop.
set_state: Additional callable which save values captured by get_state back
into the Python environment. This is only useful when staging the loop.
symbol_names: Tuple containing names of the loop variables returned by
get_state.
opts: Optional dict of extra loop parameters.
Returns:
Tuple containing the final state."
783,while_stmt,tensorflow/tensorflow/python/autograph/operators/control_flow.py,727,function,"Functional form of a while statement.
The loop operates on a so-called state, which includes all symbols that are
variant across loop iterations. In what follows we refer to state as either
a tuple of entities that represent an actual state, or a list of arguments
of the corresponding types.
The inputs and outputs of the callables representing the loop blocks are not
explicit - instead, these functions must use nonlocal/global for side effects.
The inputs and outputs are instead controlled by the set_state/get_state
functions.
Args:
test: Callable with boolean return type. The loop condition.
body: Callable representing the actual loop body.
get_state: Additional callable which can capture additional state (such as
the values of composite symbols). This is only useful when staging the
loop.
set_state: Additional callable which save values captured by get_state back
into the Python environment. This is only useful when staging the loop.
symbol_names: Tuple containing the names of all loop variables.
opts: Optional dict of extra loop parameters.
Returns:
Tuple containing the final state."
784,if_stmt,tensorflow/tensorflow/python/autograph/operators/control_flow.py,915,function,"Functional form of an if statement.
The conditional operates on a state, which includes all symbols whose values
are a function of the branch taken.
For example, given the code below that calculates the abs function:
```
x = 1
if x > 0:
x = -x
```
The state is represented by the variable `x`. The `body, `orelse` and
`set_state` functions must bind to the original `x` symbol, using `nonlocal`.
The inputs and outputs of the callables representing the loop blocks are not
explicit - instead, these functions must use nonlocal/global for side effects.
The inputs and outputs are instead controlled by the set_state/get_state
functions.
Args:
cond: Boolean.
body: Callable representing the main block of the conditional.
orelse: Callable representing the else block of the conditional.
get_state: Function that returns a tuple containing the values of all
composite symbols modified within the conditional. This allows access to
state that branches may mutate through side effects. This function is not
needed and should not be called when dispatching to code matching Python's
default semantics. This is useful for checkpointing to avoid unintended
side-effects when staging requires evaluating all code-paths.
set_state: Function to set the values of all composite symbols modified
within the conditional. This is the complement to get_state, used to
restore checkpointed values. The single argument a tuple containing values
for each composite symbol that may be modified in a branch of the
conditional. The is usually the result of a call to get_state.
symbol_names: Tuple containing basic loop var names.
nouts: Number of variables output by the statement. Vars which are
not outputs will not be passed through staged control flow such as
tf.cond. This includes variables that are defined before the conditional,
but are not used after it."
785,for_stmt,tensorflow/tensorflow/python/autograph/operators/control_flow_deprecated_py2.py,279,function,"Functional form of a for statement.
The loop operates on a state, which includes all symbols that are
variant across loop iterations, excluding the iterate as well as the
variables local to the loop.
For example, given the loop below that calculates the geometric and
arithmetic means or some numbers:
geo_mean = 1
arith_mean = 0
for i in range(n):
a = numbers[i]
geo_mean *= a
arith_mean += a
The state is represented by the variables geo_mean and arith_mean. The
argument for initial_state may contain the tuple (1, 0), the body will
include the arguments geo_mean and arith_mean and will return a tuple
representing the new values for geo_mean and respectively arith_mean.
Args:
iter_: The entity being iterated over.
extra_test: Callable with the state as arguments, and boolean return type.
An additional loop condition.
body: Callable with the iterate and the state as arguments, and state as
return type. The actual loop body.
get_state: Additional callable which can capture additional state (such as
the values of composite symbols). This is only useful when staging the
loop.
set_state: Additional callable which save values captured by get_state back
into the Python environment. This is only useful when staging the loop.
init_vars: Tuple containing the initial state.
basic_symbol_names: Tuple containing basic loop var names.
composite_symbol_names: Tuple containing composite loop var names.
opts: Optional dict of extra loop parameters.
Returns:
Tuple containing the final state."
786,while_stmt,tensorflow/tensorflow/python/autograph/operators/control_flow_deprecated_py2.py,817,function,"Functional form of a while statement.
The loop operates on a so-called state, which includes all symbols that are
variant across loop iterations. In what follows we refer to state as either
a tuple of entities that represent an actual state, or a list of arguments
of the corresponding types.
Args:
test: Callable with the state as arguments, and boolean return type. The
loop condition.
body: Callable with the state as arguments, and state as return type. The
actual loop body.
get_state: Additional callable which can capture additional state (such as
the values of composite symbols). This is only useful when staging the
loop.
set_state: Additional callable which save values captured by get_state back
into the Python environment. This is only useful when staging the loop.
init_vars: Tuple containing the initial state.
basic_symbol_names: Tuple containing basic loop var names.
composite_symbol_names: Tuple containing composite loop var names.
opts: Optional dict of extra loop parameters.
Returns:
Tuple containing the final state."
787,if_stmt,tensorflow/tensorflow/python/autograph/operators/control_flow_deprecated_py2.py,1008,function,"Functional form of an if statement.
Args:
cond: Boolean.
body: Callable with no arguments, and outputs of the positive (if) branch as
return type.
orelse: Callable with no arguments, and outputs of the negative (else)
branch as return type.
get_state: Function that returns a tuple containing the values of all
composite symbols modified within the conditional. This allows access to
state that branches may mutate through side effects. This function is not
needed and should not be called when dispatching to code matching Python's
default semantics. This is useful for checkpointing to avoid unintended
side-effects when staging requires evaluating all code-paths.
set_state: Function to set the values of all composite symbols modified
within the conditional. This is the complement to get_state, used to
restore checkpointed values. The single argument a tuple containing values
for each composite symbol that may be modified in a branch of the
conditional. The is usually the result of a call to get_state.
basic_symbol_names: Tuple containing basic loop var names.
composite_symbol_names: Tuple containing composite loop var names.
Returns:
Tuple containing the statement outputs."
788,tf_if_stmt,tensorflow/tensorflow/python/autograph/operators/control_flow_deprecated_py2.py,1048,function,Overload of if_stmt that stages a TF cond.
789,new_list,tensorflow/tensorflow/python/autograph/operators/data_structures.py,36,function,"The list constructor.
Args:
iterable: Optional elements to fill the list with.
Returns:
A list-like object. The exact return value depends on the initial elements."
790,tf_tensor_array_new,tensorflow/tensorflow/python/autograph/operators/data_structures.py,57,function,Overload of new_list that stages a Tensor list creation.
791,tf_tensor_list_new,tensorflow/tensorflow/python/autograph/operators/data_structures.py,107,function,Overload of new_list that stages a Tensor list creation.
792,list_append,tensorflow/tensorflow/python/autograph/operators/data_structures.py,171,function,"The list append function.
Note: it is unspecified where list_ will be mutated or not. If list_ is
a TensorFlow entity, it will not be typically mutated. If list_ is a plain
list, it will be. In general, if the list is mutated then the return value
should point to the original entity.
Args:
list_: An entity that supports append semantics.
x: The element to append.
Returns:
Same as list_, after the append was performed.
Raises:
ValueError: if list_ is not of a known list-like type."
793,ListPopOpts,tensorflow/tensorflow/python/autograph/operators/data_structures.py,230,class,
794,list_pop,tensorflow/tensorflow/python/autograph/operators/data_structures.py,235,function,"The list pop function.
Note: it is unspecified where list_ will be mutated or not. If list_ is
a TensorFlow entity, it will not be typically mutated. If list_ is a plain
list, it will be. In general, if the list is mutated then the return value
should point to the original entity.
Args:
list_: An entity that supports pop semantics.
i: Optional index to pop from. May be None.
opts: A ListPopOpts.
Returns:
Tuple (x, out_list_):
out_list_: same as list_, after the removal was performed.
x: the removed element value.
Raises:
ValueError: if list_ is not of a known list-like type or the operation is
not supported for that type."
795,ListStackOpts,tensorflow/tensorflow/python/autograph/operators/data_structures.py,299,class,
796,list_stack,tensorflow/tensorflow/python/autograph/operators/data_structures.py,305,function,"The list stack function.
This does not have a direct correspondent in Python. The closest idiom to
this is tf.append or np.stack. It's different from those in the sense that it
accepts a Tensor list, rather than a list of tensors. It can also accept
TensorArray. When the target is anything else, the dispatcher will rely on
ctx.original_call for fallback.
Args:
list_: An entity that supports append semantics.
opts: A ListStackOpts object.
Returns:
The output of the stack operation, typically a Tensor."
797,DispatchContext,tensorflow/tensorflow/python/autograph/operators/dispatch_context.py,27,class,"Allows passing additional parameters to the specific implementations.
Attributes:
options: Optional dict of extra arguments that may be required by specific
implementations."
798,option,tensorflow/tensorflow/python/autograph/operators/dispatch_context.py,37,method,
799,assert_stmt,tensorflow/tensorflow/python/autograph/operators/exceptions.py,26,function,"Functional form of an assert statement.
This follows the semantics of the Python assert statement, however the
concrete implementations may deviate from it. See the respective
implementation for details.
In general, the assert statement should not be used for control flow.
Furthermore, it is encouraged that the assertion expressions should not have
side effects.
Args:
expression1: Any
expression2: Callable[[], Any], returns the expression to include in the
error message when expression1 evaluates to False. When expression1 is
True, the result of expression2 will not be evaluated, however,
expression2 itself may be evaluated in some implementations.
Returns:
Any, implementation-dependent.
Raises:
ValueError: if any arguments are illegal."
800,not_,tensorflow/tensorflow/python/autograph/operators/logical.py,26,function,"Functional form of ""not""."
801,and_,tensorflow/tensorflow/python/autograph/operators/logical.py,43,function,"Functional form of ""and"". Uses lazy evaluation semantics."
802,or_,tensorflow/tensorflow/python/autograph/operators/logical.py,62,function,"Functional form of ""or"". Uses lazy evaluation semantics."
803,eq,tensorflow/tensorflow/python/autograph/operators/logical.py,81,function,"Functional form of ""equal""."
804,not_eq,tensorflow/tensorflow/python/autograph/operators/logical.py,98,function,"Functional form of ""not-equal""."
805,overload_of,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,65,function,
806,locals_in_original_context,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,92,function,Executes the locals function in the context of a specified function.
807,globals_in_original_context,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,97,function,Executes the locals function in the context of a specified function.
808,eval_in_original_context,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,102,function,Executes the eval function in the context of a specified function.
809,super_in_original_context,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,117,function,"Executes the super function in the context of a specified function.
See https://docs.python.org/3/library/functions.html#super for the exact
details
Args:
f: Callable, typically the super builtin
args: List[Any], the original call arguments
caller_fn_scope: Optional[function_wrappers.FunctionScope], the function
scope of the converted function in which this call was originally made
Returns:
The result of calling `f` as if it was called in the frame indicated by
`caller_fn_scope`."
810,abs_,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,179,function,
811,float_,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,202,function,
812,int_,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,219,function,
813,len_,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,241,function,
814,print_,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,318,function,Overload of the print builtin.
815,range_,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,360,function,
816,enumerate_,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,391,function,
817,zip_,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,409,function,
818,map_,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,423,function,
819,next_,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,437,function,
820,next_tf_iterator,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,509,function,
821,next_py,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,521,function,
822,filter_,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,527,function,
823,any_,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,541,function,
824,all_,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,570,function,
825,sorted_,tensorflow/tensorflow/python/autograph/operators/py_builtins.py,596,function,
826,GetItemOpts,tensorflow/tensorflow/python/autograph/operators/slices.py,34,class,
827,get_item,tensorflow/tensorflow/python/autograph/operators/slices.py,38,function,"The slice read operator (i.e. __getitem__).
Note: it is unspecified whether target will be mutated or not. In general,
if target is mutable (like Python lists), it will be mutated.
Args:
target: An entity that supports getitem semantics.
i: Index to read from.
opts: A GetItemOpts object.
Returns:
The read element.
Raises:
ValueError: if target is not of a supported type."
828,set_item,tensorflow/tensorflow/python/autograph/operators/slices.py,100,function,"The slice write operator (i.e. __setitem__).
Note: it is unspecified whether target will be mutated or not. In general,
if target is mutable (like Python lists), it will be mutated.
Args:
target: An entity that supports setitem semantics.
i: Index to modify.
x: The new element value.
Returns:
Same as target, after the update was performed.
Raises:
ValueError: if target is not of a supported type."
829,ld,tensorflow/tensorflow/python/autograph/operators/variables.py,22,function,Load variable operator.
830,ldu,tensorflow/tensorflow/python/autograph/operators/variables.py,29,function,"Load variable operator that returns Undefined when failing to evaluate.
Note: the name (""load or return undefined"") is abbreviated to minimize
the amount of clutter in generated code.
This variant of `ld` is useful when loading symbols that may be undefined at
runtime, such as composite symbols, and whether they are defined or not cannot
be determined statically. For example `d['a']` is undefined when `d` is an
empty dict.
Args:
load_v: Lambda that executes the actual read.
name: Human-readable name of the symbol being read.
Returns:
Either the value of the symbol, or Undefined, if the symbol is not fully
defined."
831,Undefined,tensorflow/tensorflow/python/autograph/operators/variables.py,54,class,"Represents an undefined symbol in Python.
This is used to reify undefined symbols, which is required to use the
functional form of loops.
Example:
while n > 0:
n = n - 1
s = n
return s # Runtime error if n == 0
This is valid Python code and will not result in an error as long as n
is positive. The use of this class is to stay as close to Python semantics
as possible for staged code of this nature.
Converted version of the above showing the possible usage of this class:
s = Undefined('s')
init_state = (s,)
s = while_loop(cond, body, init_state)
return s # s is an instance of Undefined if the loop never runs
Attributes:
symbol_name: Text, identifier for the undefined symbol"
832,read,tensorflow/tensorflow/python/autograph/operators/variables.py,86,method,
833,UndefinedReturnValue,tensorflow/tensorflow/python/autograph/operators/variables.py,106,class,Represents a return value that is undefined.
834,NoValue,tensorflow/tensorflow/python/autograph/pyct/anno.py,37,class,
835,Basic,tensorflow/tensorflow/python/autograph/pyct/anno.py,43,class,"Container for basic annotation keys.
The enum values are used strictly for documentation purposes."
836,Static,tensorflow/tensorflow/python/autograph/pyct/anno.py,67,class,"Container for static analysis annotation keys.
The enum values are used strictly for documentation purposes."
837,keys,tensorflow/tensorflow/python/autograph/pyct/anno.py,110,function,
838,getanno,tensorflow/tensorflow/python/autograph/pyct/anno.py,116,function,
839,hasanno,tensorflow/tensorflow/python/autograph/pyct/anno.py,123,function,
840,setanno,tensorflow/tensorflow/python/autograph/pyct/anno.py,127,function,
841,delanno,tensorflow/tensorflow/python/autograph/pyct/anno.py,137,function,
842,copyanno,tensorflow/tensorflow/python/autograph/pyct/anno.py,145,function,
843,dup,tensorflow/tensorflow/python/autograph/pyct/anno.py,154,function,"Recursively copies annotations in an AST tree.
Args:
node: ast.AST
copy_map: Dict[Hashable, Hashable], maps a source anno key to a destination
key. All annotations with the source key will be copied to identical
annotations with the destination key.
field_name: str"
844,CleanCopier,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,30,class,NodeTransformer-like visitor that copies an AST.
845,copy,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,37,method,"Returns a deep copy of node (excluding some fields, see copy_clean)."
846,copy_clean,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,63,function,"Creates a deep copy of an AST.
The copy will not include fields that are prefixed by '__', with the
exception of user-specified annotations.
Args:
node: ast.AST
preserve_annos: Optional[Set[Hashable]], annotation keys to include in the
copy
Returns:
ast.AST"
847,SymbolRenamer,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,79,class,Transformer that can rename symbols to a simple names.
848,visit_Nonlocal,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,106,method,
849,visit_Global,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,110,method,
850,visit_Name,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,114,method,
851,visit_Attribute,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,117,method,
852,visit_FunctionDef,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,123,method,
853,rename_symbols,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,130,function,Renames symbols in an AST. Requires qual_names annotations.
854,keywords_to_dict,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,140,function,Converts a list of ast.keyword objects to a dict.
855,PatternMatcher,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,150,class,Matches a node against a pattern represented by a node.
856,compare_and_visit,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,158,method,
857,no_match,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,164,method,
858,is_wildcard,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,168,method,
859,generic_visit,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,177,method,
860,matches,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,214,function,"Basic pattern matcher for AST.
The pattern may contain wildcards represented by the symbol '_'. A node
matches a pattern if for every node in the tree, either there is a node of
the same type in pattern, or a Name node with id='_'.
Args:
node: ast.AST
pattern: ast.AST
Returns:
bool"
861,apply_to_single_assignments,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,236,function,"Applies a function to each individual assignment.
This function can process a possibly-unpacked (e.g. a, b = c, d) assignment.
It tries to break down the unpacking if possible. In effect, it has the same
effect as passing the assigned values in SSA form to apply_fn.
Examples:
The following will result in apply_fn(a, c), apply_fn(b, d):
a, b = c, d
The following will result in apply_fn(a, c[0]), apply_fn(b, c[1]):
a, b = c
The following will result in apply_fn(a, (b, c)):
a = b, c
It uses the visitor pattern to allow subclasses to process single
assignments individually.
Args:
targets: Union[List[ast.AST, ...], Tuple[ast.AST, ...], ast.AST, should be
used with the targets field of an ast.Assign node
values: ast.AST
apply_fn: Callable[[ast.AST, ast.AST], None], called with the
respective nodes of each single assignment"
862,parallel_walk,tensorflow/tensorflow/python/autograph/pyct/ast_util.py,283,function,"Walks two ASTs in parallel.
The two trees must have identical structure.
Args:
node: Union[ast.AST, Iterable[ast.AST]]
other: Union[ast.AST, Iterable[ast.AST]]
Yields:
Tuple[ast.AST, ast.AST]
Raises:
ValueError: if the two trees don't have identical structure."
863,CodeObjectCache,tensorflow/tensorflow/python/autograph/pyct/cache.py,63,class,"A function cache based on code objects.
Code objects are good proxies for the source code of a function.
This cache efficiently handles functions that share code objects, such as
functions defined in a loop, bound methods, etc.
The cache falls back to the function object, if it doesn't have a code object."
864,UnboundInstanceCache,tensorflow/tensorflow/python/autograph/pyct/cache.py,81,class,"A function cache based on unbound function objects.
Using the function for the cache key allows efficient handling of object
methods.
Unlike the _CodeObjectCache, this discriminates between different functions
even if they have the same code. This is needed for decorators that may
masquerade as another function."
865,Node,tensorflow/tensorflow/python/autograph/pyct/cfg.py,54,class,"A node in the CFG.
Although new instances of this class are mutable, the objects that a user
finds in the CFG are typically not.
The nodes represent edges in the CFG graph, and maintain pointers to allow
efficient walking in both forward and reverse order. The following property
holds for all nodes: ""child in node.next"" iff ""node in child.prev"".
Attributes:
next: FrozenSet[Node, ...], the nodes that follow this node, in control
flow order
prev: FrozenSet[Node, ...], the nodes that precede this node, in reverse
control flow order
ast_node: ast.AST, the AST node corresponding to this CFG node"
866,freeze,tensorflow/tensorflow/python/autograph/pyct/cfg.py,77,method,
867,Graph,tensorflow/tensorflow/python/autograph/pyct/cfg.py,95,class,"A Control Flow Graph.
The CFG maintains an index to allow looking up a CFG node by the AST node to
which it is associated. The index can also be enumerated in top-down, depth
first order.
Walking the graph in forward or reverse order is supported by double
parent-child links.
Note: the error nodes are not wired to their corresponding finally guards,
because these are shared, and wiring them would create a reverse path from
normal control flow into the error nodes, which we want to avoid.
The graph also maintains edges corresponding to higher level statements
like for-else loops. A node is considered successor of a statement if there
is an edge from a node that is lexically a child of that statement to a node
that is not. Statement predecessors are analogously defined.
Attributes:
entry: Node, the entry node
exit: FrozenSet[Node, ...], the exit nodes
error: FrozenSet[Node, ...], nodes that exit due to an explicitly raised
error (errors propagated from function calls are not accounted)
index: Dict[ast.Node, Node], mapping AST nodes to the respective CFG
node
stmt_prev: Dict[ast.Node, FrozenSet[Node, ...]], mapping statement AST
nodes to their predecessor CFG nodes
stmt_next: Dict[ast.Node, FrozenSet[Node, ...]], mapping statement AST
nodes to their successor CFG nodes"
868,as_dot,tensorflow/tensorflow/python/autograph/pyct/cfg.py,133,method,Print CFG in DOT format.
869,GraphVisitor,tensorflow/tensorflow/python/autograph/pyct/cfg.py,152,class,"Base class for a CFG visitors.
This implementation is not thread safe.
The visitor has some facilities to simplify dataflow analyses. In particular,
it allows revisiting the nodes at the decision of the subclass. This can be
used to visit the graph until the state reaches a fixed point.
For more details on dataflow analysis, see
https://www.seas.harvard.edu/courses/cs252/2011sp/slides/Lec02-Dataflow.pdf
Note: the literature generally suggests visiting successor nodes only when the
state of the current node changed, regardless of whether that successor has
ever been visited. This implementation visits every successor at least once.
Attributes:
graph: Graph
in_: Dict[Node, Any], stores node-keyed state during a visit
out: Dict[Node, Any], stores node-keyed state during a visit"
870,init_state,tensorflow/tensorflow/python/autograph/pyct/cfg.py,178,method,"State initialization function. Optional to overload.
An in/out state slot will be created for each node in the graph. Subclasses
must overload this to control what that is initialized to.
Args:
node: Node"
871,visit_node,tensorflow/tensorflow/python/autograph/pyct/cfg.py,190,method,"Visitor function.
Args:
node: Node
Returns:
bool, whether the node should be revisited; subclasses can visit every
reachable node exactly once by always returning False"
872,reset,tensorflow/tensorflow/python/autograph/pyct/cfg.py,201,method,
873,can_ignore,tensorflow/tensorflow/python/autograph/pyct/cfg.py,209,method,Returns True if the node can safely be assumed not to touch variables.
874,visit_forward,tensorflow/tensorflow/python/autograph/pyct/cfg.py,245,method,
875,visit_reverse,tensorflow/tensorflow/python/autograph/pyct/cfg.py,248,method,
876,GraphBuilder,tensorflow/tensorflow/python/autograph/pyct/cfg.py,252,class,"Builder that constructs a CFG from a given AST.
This GraphBuilder facilitates constructing the DAG that forms the CFG when
nodes
are supplied in lexical order (i.e., top-down, depth first). Under these
conditions, it supports building patterns found in typical structured
programs.
This builder ignores the flow generated by exceptions, which are assumed to
always be catastrophic and present purely for diagnostic purposes (e.g. to
print debug information). Statements like raise and try/catch sections are
allowed and will generate control flow edges, but ordinary statements are
assumed not to raise exceptions.
Finally sections are also correctly interleaved between break/continue/return
nodes and their subsequent statements.
Important concepts:
* nodes - nodes refer refer to CFG nodes; AST nodes are qualified explicitly
* leaf set - since the graph is constructed gradually, a leaf set maintains
the CFG nodes that will precede the node that the builder expects to
receive next; when an ordinary node is added, it is connected to the
existing leaves and it in turn becomes the new leaf
* jump nodes - nodes that should generate edges other than what
ordinary nodes would; these correspond to break, continue and return
statements
* sections - logical delimiters for subgraphs that require special
edges; there are various types of nodes, each admitting various
types of jump nodes; sections are identified by their corresponding AST
node"
877,reset,tensorflow/tensorflow/python/autograph/pyct/cfg.py,292,method,Resets the state of this factory.
878,begin_statement,tensorflow/tensorflow/python/autograph/pyct/cfg.py,372,method,"Marks the beginning of a statement.
Args:
stmt: Hashable, a key by which the statement can be identified in
the CFG's stmt_prev and stmt_next attributes"
879,end_statement,tensorflow/tensorflow/python/autograph/pyct/cfg.py,381,method,"Marks the end of a statement.
Args:
stmt: Hashable, a key by which the statement can be identified in
the CFG's stmt_prev and stmt_next attributes; must match a key
previously passed to begin_statement."
880,add_ordinary_node,tensorflow/tensorflow/python/autograph/pyct/cfg.py,391,method,"Grows the graph by adding an ordinary CFG node.
Ordinary nodes are followed by the next node, in lexical order, that is,
they become the new leaf set.
Args:
ast_node: ast.AST
Returns:
Node"
881,add_exit_node,tensorflow/tensorflow/python/autograph/pyct/cfg.py,438,method,"Grows the graph by adding an exit node.
This node becomes an exit for the current section.
Args:
ast_node: ast.AST
section_id: Hashable, the node for which ast_node should be considered
to be an exit node
guards: Tuple[ast.AST, ...], the finally sections that guard ast_node
Returns:
Node"
882,add_continue_node,tensorflow/tensorflow/python/autograph/pyct/cfg.py,455,method,"Grows the graph by adding a reentry node.
This node causes control flow to go back to the loop section's entry.
Args:
ast_node: ast.AST
section_id: Hashable, the node for which ast_node should be considered
to be an exit node
guards: Tuple[ast.AST, ...], the finally sections that guard ast_node"
883,connect_raise_node,tensorflow/tensorflow/python/autograph/pyct/cfg.py,469,method,"Adds extra connection between a raise node and containing except guards.
The node is a graph node, not an ast node.
Args:
node: Node
except_guards: Tuple[ast.AST, ...], the except sections that guard node"
884,enter_section,tensorflow/tensorflow/python/autograph/pyct/cfg.py,484,method,"Enters a regular section.
Regular sections admit exit jumps, which end the section.
Args:
section_id: Hashable, the same node that will be used in calls to the
ast_node arg passed to add_exit_node"
885,exit_section,tensorflow/tensorflow/python/autograph/pyct/cfg.py,496,method,Exits a regular section.
886,enter_loop_section,tensorflow/tensorflow/python/autograph/pyct/cfg.py,505,method,"Enters a loop section.
Loop sections define an entry node. The end of the section always flows back
to the entry node. These admit continue jump nodes which also flow to the
entry node.
Args:
section_id: Hashable, the same node that will be used in calls to the
ast_node arg passed to add_continue_node
entry_node: ast.AST, the entry node into the loop (e.g. the test node
for while loops)"
887,exit_loop_section,tensorflow/tensorflow/python/autograph/pyct/cfg.py,524,method,Exits a loop section.
888,enter_cond_section,tensorflow/tensorflow/python/autograph/pyct/cfg.py,539,method,"Enters a conditional section.
Conditional sections define an entry node, and one or more branches.
Args:
section_id: Hashable, the same node that will be used in calls to the
section_id arg passed to new_cond_branch"
889,new_cond_branch,tensorflow/tensorflow/python/autograph/pyct/cfg.py,553,method,Begins a new branch in a cond section.
890,exit_cond_section,tensorflow/tensorflow/python/autograph/pyct/cfg.py,567,method,Exits a conditional section.
891,enter_except_section,tensorflow/tensorflow/python/autograph/pyct/cfg.py,574,method,Enters an except section.
892,enter_finally_section,tensorflow/tensorflow/python/autograph/pyct/cfg.py,579,method,Enters a finally section.
893,exit_finally_section,tensorflow/tensorflow/python/autograph/pyct/cfg.py,589,method,Exits a finally section.
894,build,tensorflow/tensorflow/python/autograph/pyct/cfg.py,599,method,"Returns the CFG accumulated so far and resets the builder.
Returns:
Graph"
895,AstToCfg,tensorflow/tensorflow/python/autograph/pyct/cfg.py,647,class,"Converts an AST to CFGs.
A separate CFG will be constructed for each function."
896,visit_ClassDef,tensorflow/tensorflow/python/autograph/pyct/cfg.py,714,method,
897,visit_FunctionDef,tensorflow/tensorflow/python/autograph/pyct/cfg.py,767,method,
898,visit_Lambda,tensorflow/tensorflow/python/autograph/pyct/cfg.py,770,method,
899,visit_Return,tensorflow/tensorflow/python/autograph/pyct/cfg.py,773,method,
900,visit_Import,tensorflow/tensorflow/python/autograph/pyct/cfg.py,776,method,
901,visit_ImportFrom,tensorflow/tensorflow/python/autograph/pyct/cfg.py,779,method,
902,visit_Expr,tensorflow/tensorflow/python/autograph/pyct/cfg.py,782,method,
903,visit_Assign,tensorflow/tensorflow/python/autograph/pyct/cfg.py,785,method,
904,visit_AnnAssign,tensorflow/tensorflow/python/autograph/pyct/cfg.py,788,method,
905,visit_AugAssign,tensorflow/tensorflow/python/autograph/pyct/cfg.py,791,method,
906,visit_Pass,tensorflow/tensorflow/python/autograph/pyct/cfg.py,794,method,
907,visit_Global,tensorflow/tensorflow/python/autograph/pyct/cfg.py,797,method,
908,visit_Nonlocal,tensorflow/tensorflow/python/autograph/pyct/cfg.py,800,method,
909,visit_Print,tensorflow/tensorflow/python/autograph/pyct/cfg.py,803,method,
910,visit_Raise,tensorflow/tensorflow/python/autograph/pyct/cfg.py,806,method,
911,visit_Assert,tensorflow/tensorflow/python/autograph/pyct/cfg.py,811,method,
912,visit_Delete,tensorflow/tensorflow/python/autograph/pyct/cfg.py,815,method,
913,visit_If,tensorflow/tensorflow/python/autograph/pyct/cfg.py,818,method,
914,visit_While,tensorflow/tensorflow/python/autograph/pyct/cfg.py,840,method,
915,visit_For,tensorflow/tensorflow/python/autograph/pyct/cfg.py,863,method,
916,visit_Break,tensorflow/tensorflow/python/autograph/pyct/cfg.py,894,method,
917,visit_Continue,tensorflow/tensorflow/python/autograph/pyct/cfg.py,897,method,
918,visit_ExceptHandler,tensorflow/tensorflow/python/autograph/pyct/cfg.py,900,method,
919,visit_Try,tensorflow/tensorflow/python/autograph/pyct/cfg.py,914,method,
920,visit_With,tensorflow/tensorflow/python/autograph/pyct/cfg.py,956,method,
921,build,tensorflow/tensorflow/python/autograph/pyct/cfg.py,964,function,
922,CountingVisitor,tensorflow/tensorflow/python/autograph/pyct/cfg_test.py,28,class,
923,init_state,tensorflow/tensorflow/python/autograph/pyct/cfg_test.py,34,method,
924,visit_node,tensorflow/tensorflow/python/autograph/pyct/cfg_test.py,37,method,
925,FrameInfo,tensorflow/tensorflow/python/autograph/pyct/error_utils.py,26,class,
926,MultilineMessageKeyError,tensorflow/tensorflow/python/autograph/pyct/error_utils.py,141,class,
927,ErrorMetadataBase,tensorflow/tensorflow/python/autograph/pyct/error_utils.py,153,class,"Container objects attached to exceptions raised in user code.
This metadata allows re-raising exceptions that occur in generated code, with
a custom error message that includes a stack trace relative to user-readable
code from which the generated code originated."
928,get_message,tensorflow/tensorflow/python/autograph/pyct/error_utils.py,177,method,Returns the message for the underlying exception.
929,create_exception,tensorflow/tensorflow/python/autograph/pyct/error_utils.py,211,method,
930,to_exception,tensorflow/tensorflow/python/autograph/pyct/error_utils.py,221,method,
931,PyCTError,tensorflow/tensorflow/python/autograph/pyct/errors.py,22,class,Base class for all exceptions.
932,UnsupportedLanguageElementError,tensorflow/tensorflow/python/autograph/pyct/errors.py,27,class,Raised for code patterns that AutoGraph does not support.
933,is_literal,tensorflow/tensorflow/python/autograph/pyct/gast_util.py,40,function,Tests whether node represents a Python literal.
934,islambda,tensorflow/tensorflow/python/autograph/pyct/inspect_utils.py,60,function,
935,isnamedtuple,tensorflow/tensorflow/python/autograph/pyct/inspect_utils.py,68,function,Returns True if the argument is a namedtuple-like.
936,isbuiltin,tensorflow/tensorflow/python/autograph/pyct/inspect_utils.py,82,function,Returns True if the argument is a built-in function.
937,isconstructor,tensorflow/tensorflow/python/autograph/pyct/inspect_utils.py,96,function,"Returns True if the argument is an object constructor.
In general, any object of type class is a constructor, with the exception
of classes created using a callable metaclass.
See below for why a callable metaclass is not a trivial combination:
https://docs.python.org/2.7/reference/datamodel.html#customizing-class-creation
Args:
cls: Any
Returns:
Bool"
938,getimmediatesource,tensorflow/tensorflow/python/autograph/pyct/inspect_utils.py,143,function,A variant of inspect.getsource that ignores the __wrapped__ property.
939,getnamespace,tensorflow/tensorflow/python/autograph/pyct/inspect_utils.py,151,function,"Returns the complete namespace of a function.
Namespace is defined here as the mapping of all non-local variables to values.
This includes the globals and the closure variables. Note that this captures
the entire globals collection of the function, and may contain extra symbols
that it does not actually use.
Args:
f: User defined function.
Returns:
A dict mapping symbol names to values."
940,getqualifiedname,tensorflow/tensorflow/python/autograph/pyct/inspect_utils.py,177,function,"Returns the name by which a value can be referred to in a given namespace.
If the object defines a parent module, the function attempts to use it to
locate the object.
This function will recurse inside modules, but it will not search objects for
attributes. The recursion depth is controlled by max_depth.
Args:
namespace: Dict[str, Any], the namespace to search into.
object_: Any, the value to search.
max_depth: Optional[int], a limit to the recursion depth when searching
inside modules.
visited: Optional[Set[int]], ID of modules to avoid visiting.
Returns: Union[str, None], the fully-qualified name that resolves to the value
o, or None if it couldn't be found."
941,getdefiningclass,tensorflow/tensorflow/python/autograph/pyct/inspect_utils.py,250,function,Resolves the class (e.g. one of the superclasses) that defined a method.
942,getmethodclass,tensorflow/tensorflow/python/autograph/pyct/inspect_utils.py,265,function,"Resolves a function's owner, e.g. a method's class.
Note that this returns the object that the function was retrieved from, not
necessarily the class where it was defined.
This function relies on Python stack frame support in the interpreter, and
has the same limitations that inspect.currentframe.
Limitations. This function will only work correctly if the owned class is
visible in the caller's global or local variables.
Args:
m: A user defined function
Returns:
The class that this function was retrieved from, or None if the function
is not an object or class method, or the class that owns the object or
method is not visible to m.
Raises:
ValueError: if the class could not be resolved for any unexpected reason."
943,getfutureimports,tensorflow/tensorflow/python/autograph/pyct/inspect_utils.py,339,function,"Detects what future imports are necessary to safely execute entity source.
Args:
entity: Any object
Returns:
A tuple of future strings"
944,decorator,tensorflow/tensorflow/python/autograph/pyct/inspect_utils_test.py,37,function,
945,function_decorator,tensorflow/tensorflow/python/autograph/pyct/inspect_utils_test.py,41,function,
946,wrapping_decorator,tensorflow/tensorflow/python/autograph/pyct/inspect_utils_test.py,47,function,
947,free_function,tensorflow/tensorflow/python/autograph/pyct/inspect_utils_test.py,85,function,
948,factory,tensorflow/tensorflow/python/autograph/pyct/inspect_utils_test.py,89,function,
949,free_factory,tensorflow/tensorflow/python/autograph/pyct/inspect_utils_test.py,93,function,
950,load_source,tensorflow/tensorflow/python/autograph/pyct/loader.py,50,function,Loads the given source code as a Python module.
951,load_ast,tensorflow/tensorflow/python/autograph/pyct/loader.py,70,function,"Loads the given AST as a Python module.
Compiling the AST code this way ensures that the source code is readable by
e.g. `pdb` or `inspect`.
Args:
nodes: Union[ast.AST, Iterable[ast.AST]], the code to compile, as an AST
object.
indentation: Text, the string to use for indentation.
include_source_map: bool, whether return a source map.
delete_on_exit: bool, whether to delete the temporary file used for
compilation on exit.
Returns:
Tuple[module, Text, Dict[LineLocation, OriginInfo]], containing:
the module containing the unparsed nodes, the source code corresponding to
nodes, and the source map. Is include_source_map is False, the source map
will be None."
952,load_source,tensorflow/tensorflow/python/autograph/pyct/loader_deprecated_py2.py,40,function,Loads the given source code as a Python module.
953,load_ast,tensorflow/tensorflow/python/autograph/pyct/loader_deprecated_py2.py,58,function,"Loads the given AST as a Python module.
Compiling the AST code this way ensures that the source code is readable by
e.g. `pdb` or `inspect`.
Args:
nodes: Union[ast.AST, Iterable[ast.AST]], the code to compile, as an AST
object.
indentation: Text, the string to use for indentation.
include_source_map: bool, whether return a source map.
delete_on_exit: bool, whether to delete the temporary file used for
compilation on exit.
Returns:
Tuple[module, Text, Dict[LineLocation, OriginInfo]], containing:
the module containing the unparsed nodes, the source code corresponding to
nodes, and the source map. Is include_source_map is False, the source map
will be None."
954,Namer,tensorflow/tensorflow/python/autograph/pyct/naming.py,24,class,Symbol name generator.
955,new_symbol,tensorflow/tensorflow/python/autograph/pyct/naming.py,31,method,See control_flow.SymbolNamer.new_symbol.
956,LineLocation,tensorflow/tensorflow/python/autograph/pyct/origin_info.py,35,class,"Similar to Location, but without column information.
Attributes:
filename: Text
lineno: int, 1-based"
957,Location,tensorflow/tensorflow/python/autograph/pyct/origin_info.py,46,class,"Encodes code location information.
Attributes:
filename: Text
lineno: int, 1-based
col_offset: int
line_loc: LineLocation"
958,line_loc,tensorflow/tensorflow/python/autograph/pyct/origin_info.py,58,method,
959,OriginInfo,tensorflow/tensorflow/python/autograph/pyct/origin_info.py,62,class,"Container for information about the source code before conversion.
Attributes:
loc: Location
function_name: Optional[Text]
source_code_line: Text
comment: Optional[Text]"
960,as_frame,tensorflow/tensorflow/python/autograph/pyct/origin_info.py,75,method,Returns a 4-tuple consistent with the return of traceback.extract_tb.
961,create_source_map,tensorflow/tensorflow/python/autograph/pyct/origin_info.py,89,function,"Creates a source map between an annotated AST and the code it compiles to.
Note: this function assumes nodes nodes, code and filepath correspond to the
same code.
Args:
nodes: Iterable[ast.AST, ...], one or more AST modes.
code: Text, the source code in which nodes are found.
filepath: Text
Returns:
Dict[LineLocation, OriginInfo], mapping locations in code to locations
indicated by origin annotations in node."
962,OriginResolver,tensorflow/tensorflow/python/autograph/pyct/origin_info.py,166,class,Annotates an AST with additional source information like file name.
963,visit,tensorflow/tensorflow/python/autograph/pyct/origin_info.py,212,method,
964,resolve,tensorflow/tensorflow/python/autograph/pyct/origin_info.py,226,function,"Adds origin information to an AST, based on the source it was loaded from.
This allows us to map the original source code line numbers to generated
source code.
Note: the AST may be a part of a larger context (e.g. a function is part of
a module that may contain other things). However, this function does not
assume the source argument contains the entire context, nor that it contains
only code corresponding to node itself. However, it assumes that node was
parsed from the given source code.
For this reason, two extra arguments are required, and they indicate the
location of the node in the original context.
Args:
node: gast.AST, the AST to annotate.
source: Text, the source code representing node.
context_filepath: Text
context_lineno: int
context_col_offset: int"
965,resolve_entity,tensorflow/tensorflow/python/autograph/pyct/origin_info.py,271,function,"Like resolve, but extracts the context information from an entity."
966,dedent_block,tensorflow/tensorflow/python/autograph/pyct/parser.py,65,function,Dedents a code so that its first line starts at row zero.
967,parse_entity,tensorflow/tensorflow/python/autograph/pyct/parser.py,133,function,"Returns the AST and source code of given entity.
Args:
entity: Any, Python function/method/class
future_features: Iterable[Text], future features to use (e.g.
'print_statement'). See
https://docs.python.org/2/reference/simple_stmts.html#future
Returns:
gast.AST, Text: the parsed AST node; the source code that was parsed to
generate the AST (including any prefixes that this function may have added)."
968,parse,tensorflow/tensorflow/python/autograph/pyct/parser.py,323,function,"Returns the AST of given piece of code.
Args:
src: Text
preamble_len: Int, indicates leading nodes in the parsed AST which should be
dropped.
single_node: Bool, whether `src` is assumed to be represented by exactly one
AST node.
Returns:
ast.AST"
969,parse_expression,tensorflow/tensorflow/python/autograph/pyct/parser.py,347,function,"Returns the AST of given identifier.
Args:
src: A piece of code that represents a single Python expression
Returns:
A gast.AST object.
Raises:
ValueError: if src does not consist of a single Expression."
970,unparse,tensorflow/tensorflow/python/autograph/pyct/parser.py,366,function,"Returns the source code of given AST.
Args:
node: The code to compile, as an AST object.
indentation: Unused, deprecated. The returning code will always be indented
at 4 spaces.
include_encoding_marker: Bool, thether to include a comment on the first
line to explicitly specify UTF-8 encoding.
Returns:
code: The source code generated from the AST object
source_mapping: A mapping between the user and AutoGraph generated code."
971,PrettyPrinter,tensorflow/tensorflow/python/autograph/pyct/pretty_printer.py,26,class,Print AST nodes.
972,generic_visit,tensorflow/tensorflow/python/autograph/pyct/pretty_printer.py,59,method,
973,fmt,tensorflow/tensorflow/python/autograph/pyct/pretty_printer.py,128,function,
974,CallerMustSetThis,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,36,class,
975,Symbol,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,40,class,Represents a Python symbol.
976,Literal,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,44,class,Represents a Python numeric literal.
977,QN,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,57,class,Represents a qualified name.
978,is_symbol,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,95,method,
979,is_simple,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,98,method,
980,is_composite,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,101,method,
981,has_subscript,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,104,method,
982,has_attr,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,107,method,
983,parent,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,111,method,
984,owner_set,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,117,method,"Returns all the symbols (simple or composite) that own this QN.
In other words, if this symbol was modified, the symbols in the owner set
may also be affected.
Examples:
'a.b[c.d]' has two owners, 'a' and 'a.b'"
985,support_set,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,133,method,"Returns the set of simple symbols that this QN relies on.
This would be the smallest set of symbols necessary for the QN to
statically resolve (assuming properties and index ranges are verified
at runtime).
Examples:
'a.b' has only one support symbol, 'a'
'a[i]' has two support symbols, 'a' and 'i'"
986,ssf,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,175,method,Simple symbol form.
987,ast,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,187,method,AST representation.
988,QnResolver,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,210,class,"Annotates nodes with QN information.
Note: Not using NodeAnnos to avoid circular dependencies."
989,visit_Name,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,216,method,
990,visit_Attribute,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,221,method,
991,visit_Subscript,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,228,method,
992,resolve,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,251,function,
993,from_str,tensorflow/tensorflow/python/autograph/pyct/qual_names.py,255,function,
994,ContextAdjuster,tensorflow/tensorflow/python/autograph/pyct/templates.py,35,class,"Adjusts the ctx field of nodes to ensure consistency.
This transformer can change the ctx fields of a variable, tuple and other
AST elements that allow one, based on whether the element is being read or
written."
995,visit,tensorflow/tensorflow/python/autograph/pyct/templates.py,46,method,
996,visit_Attribute,tensorflow/tensorflow/python/autograph/pyct/templates.py,58,method,
997,visit_Tuple,tensorflow/tensorflow/python/autograph/pyct/templates.py,64,method,
998,visit_List,tensorflow/tensorflow/python/autograph/pyct/templates.py,68,method,
999,visit_Name,tensorflow/tensorflow/python/autograph/pyct/templates.py,72,method,
1000,visit_Call,tensorflow/tensorflow/python/autograph/pyct/templates.py,76,method,
1001,visit_Dict,tensorflow/tensorflow/python/autograph/pyct/templates.py,83,method,
1002,visit_Subscript,tensorflow/tensorflow/python/autograph/pyct/templates.py,89,method,
1003,visit_comprehension,tensorflow/tensorflow/python/autograph/pyct/templates.py,95,method,
1004,visit_Lambda,tensorflow/tensorflow/python/autograph/pyct/templates.py,101,method,
1005,ReplaceTransformer,tensorflow/tensorflow/python/autograph/pyct/templates.py,108,class,Replace AST nodes.
1006,visit_Expr,tensorflow/tensorflow/python/autograph/pyct/templates.py,146,method,
1007,visit_keyword,tensorflow/tensorflow/python/autograph/pyct/templates.py,154,method,
1008,visit_FunctionDef,tensorflow/tensorflow/python/autograph/pyct/templates.py,172,method,
1009,visit_Attribute,tensorflow/tensorflow/python/autograph/pyct/templates.py,185,method,
1010,visit_Name,tensorflow/tensorflow/python/autograph/pyct/templates.py,197,method,
1011,replace,tensorflow/tensorflow/python/autograph/pyct/templates.py,234,function,"Replaces placeholders in a Python template.
AST Name and Tuple nodes always receive the context that inferred from
the template. However, when replacing more complex nodes (that can potentially
contain Name children), then the caller is responsible for setting the
appropriate context.
Args:
template: A string representing Python code. Any symbol name can be used
that appears in the template code can be used as placeholder.
**replacements: A mapping from placeholder names to (lists of) AST nodes
that these placeholders will be replaced by. String values are also
supported as a shorthand for AST Name nodes with the respective ID.
Returns:
An AST node or list of AST nodes with the replacements made. If the
template was a function, a list will be returned. If the template was a
node, the same node will be returned. If the template was a string, an
AST node will be returned (a `Module` node in the case of a multi-line
string, an `Expr` node otherwise).
Raises:
ValueError: if the arguments are incorrect."
1012,replace_as_expression,tensorflow/tensorflow/python/autograph/pyct/templates.py,279,function,"Variant of replace that generates expressions, instead of code blocks."
1013,AnalysisLevel,tensorflow/tensorflow/python/autograph/pyct/transformer.py,32,class,
1014,Context,tensorflow/tensorflow/python/autograph/pyct/transformer.py,41,class,"Contains information about a source code transformation.
This object is mutable, and is updated during conversion. Not thread safe.
Attributes:
info: EntityInfo, immutable.
namer: naming.Namer.
current_origin: origin_info.OriginInfo, holds the OriginInfo of the last
AST node to be processed successfully. Useful for error handling.
user: An user-supplied context object. The object is opaque to the
infrastructure, but will pe passed through to all custom transformations."
1015,EntityInfo,tensorflow/tensorflow/python/autograph/pyct/transformer.py,63,class,"Contains information about a Python entity.
Immutable.
Examples of entities include functions and classes.
Attributes:
name: The name that identifies this entity.
source_code: The entity's source code.
source_file: The entity's source file.
future_features: Tuple[Text], the future features that this entity was
compiled with. See
https://docs.python.org/2/reference/simple_stmts.html#future.
namespace: Dict[str, ], containing symbols visible to the entity (excluding
parameters)."
1016,NodeStateTracker,tensorflow/tensorflow/python/autograph/pyct/transformer.py,200,class,"Base class for general-purpose Python code transformation.
This abstract class provides helpful functions, like state tracking within
the scope of arbitrary node, helpers for processing code blocks, debugging,
mapping of transformed code to original code, and others.
Scope-local state tracking: to keep state across nodes, at the level of
(possibly nested) scopes, use enter/exit_local_scope and set/get_local.
You must call enter/exit_local_scope manually, but the transformer detects
when they are not properly paired.
The transformer allows keeping state across calls that is local
to arbitrary nodes and their descendants, using the self.state attribute.
Multiple independent scopes are allowed and automatically constructed.
For example, to keep track of the `If` node that encloses any `Name` node,
one can write:
```
class FooType(object):
def __init__(self):
self.foo_property = None
class DummyTransformer(NodeStateTracker, ast.NodeTransformer):
def visit_If(self, node):
self.state[FooType].enter()
self.state[FooType].foo_property = node
node = self.veneric_visit(node)
self.state[FooType].exit()
return node
def visit_Name(self, node):
self.state[FooType].foo_property # will hold the innermost enclosing if
```
Alternatively, the `enter()`/`exit()` calls can be managed by a `with`
statement:
```
def visit_If(self, node):
with self.state[FooType] as foo:
foo.foo_property = node
return self.generic_visit(node)
```"
1017,debug_print,tensorflow/tensorflow/python/autograph/pyct/transformer.py,270,method,Helper method useful for debugging. Prints the AST.
1018,debug_print_src,tensorflow/tensorflow/python/autograph/pyct/transformer.py,276,method,Helper method useful for debugging. Prints the AST as code.
1019,visit_block,tensorflow/tensorflow/python/autograph/pyct/transformer.py,282,method,"A more powerful version of generic_visit for statement blocks.
An example of a block is the body of an if statement.
This function allows specifying a postprocessing callback (the
after_visit argument) argument which can be used to move nodes to a new
destination. This is done by after_visit by returning a non-null
second return value, e.g. return new_node, new_destination.
For example, a transformer could perform the following move:
foo()
bar()
baz()
foo()
if cond:
bar()
baz()
The above could be done with a postprocessor of this kind:
def after_visit(node):
if node_is_function_call(bar):
new_container_node = build_cond()
new_container_node.body.append(node)
return new_container_node, new_container_node.body
else:
# Once we set a new destination, all subsequent items will be
# moved to it, so we don't need to explicitly handle baz.
return node, None
Args:
nodes: enumerable of AST node objects. If None, the function returns None.
before_visit: optional callable that is called before visiting each item
in nodes
after_visit: optional callable that takes in an AST node and returns a
tuple (new_node, new_destination). It is called after visiting each item
in nodes. Is used in the same was as the
visit_* methods: new_node will replace the node; if not None,
new_destination must be a list, and subsequent nodes will be placed
in this list instead of the list returned by visit_block.
Returns:
A list of AST node objects containing the transformed items fron nodes,
except those nodes that have been relocated using after_visit."
1020,Base,tensorflow/tensorflow/python/autograph/pyct/transformer.py,360,class,"Base class for general-purpose Python-to-Python code transformation.
This is an extension of ast.NodeTransformer that provides the additional
functions offered by NodeStateTracker."
1021,create_assignment,tensorflow/tensorflow/python/autograph/pyct/transformer.py,367,method,
1022,apply_to_single_assignments,tensorflow/tensorflow/python/autograph/pyct/transformer.py,374,method,"Applies a function to each individual assignment.
This function can process a possibly-unpacked (e.g. a, b = c, d) assignment.
It tries to break down the unpacking if possible. In effect, it has the same
effect as passing the assigned values in SSA form to apply_fn.
Examples:
The following will result in apply_fn(a, c), apply_fn(b, d):
a, b = c, d
The following will result in apply_fn(a, c[0]), apply_fn(b, c[1]):
a, b = c
The following will result in apply_fn(a, (b, c)):
a = b, c
It uses the visitor pattern to allow subclasses to process single
assignments individually.
Args:
targets: list, tuple of or individual AST node. Should be used with the
targets field of an ast.Assign node.
values: an AST node.
apply_fn: a function of a single argument, which will be called with the
respective nodes of each single assignment. The signature is
apply_fn(target, value), no return value."
1023,visit,tensorflow/tensorflow/python/autograph/pyct/transformer.py,421,method,
1024,CodeGenerator,tensorflow/tensorflow/python/autograph/pyct/transformer.py,478,class,"Base class for general-purpose Python-to-string code transformation.
Similar to Base, but outputs arbitrary strings instead of a Python AST.
This uses the same visitor mechanism that the standard NodeVisitor uses,
meaning that subclasses write handlers for the different kinds of nodes.
New code is generated using the emit method, which appends to a code buffer
that can be afterwards obtained from code_buffer.
Example:
class SimpleCodeGen(CodeGenerator):
def visitIf(self, node):
self.emit('if ')
self.visit(node.test)
self.emit(' { ')
self.visit(node.body)
self.emit(' } else { ')
self.visit(node.orelse)
self.emit(' } ')
node = ast.parse(...)
gen = SimpleCodeGen()
gen.visit(node)
# gen.code_buffer contains the resulting code"
1025,emit,tensorflow/tensorflow/python/autograph/pyct/transformer.py,513,method,
1026,code_buffer,tensorflow/tensorflow/python/autograph/pyct/transformer.py,517,method,
1027,visit,tensorflow/tensorflow/python/autograph/pyct/transformer.py,520,method,
1028,GenericTranspiler,tensorflow/tensorflow/python/autograph/pyct/transpiler.py,227,class,"A generic transpiler for Python functions.
Its interface is the `transform` API, which can process Python function
objects. Internally, it handles parsing.
Users typically subclass this, customizing the `transform_ast` method. The
output of transformed_ast is returned directly by `transform`. Existing
methods like `transform_function` may also be overloaded.
Example:
class MyTransformer(GenericTranspiler):
def transform_ast(self, node, ctx):
result = <<transform node>>
return result
transformer = MyTransfomer()
result = transformer.transform(f, ...)
# result is the output"
1029,get_transformed_name,tensorflow/tensorflow/python/autograph/pyct/transpiler.py,251,method,Returns a name for the output function. Subclasses may override this.
1030,transform_ast,tensorflow/tensorflow/python/autograph/pyct/transpiler.py,259,method,"Performs an actual transformation of a function's AST.
Subclasses must implement this method, and do not usually call it.
Args:
node: One or more ast.AST nodes representing the AST to be transformed.
ctx: transformer.Context."
1031,transform,tensorflow/tensorflow/python/autograph/pyct/transpiler.py,270,method,"Transforms a Python object.
Users typically call this method.
Args:
obj: A Python object, function, type, etc.
user_context: An opaque object (may be None) that is forwarded to
transform_ast, through the ctx.user_context argument.
Returns:
Tre result of calling transform_function.
Raises:
NotImplementedError: if the type of obj is not handled."
1032,transform_module,tensorflow/tensorflow/python/autograph/pyct/transpiler.py,300,method,"Transforms a module.
Subclasses may override this method. The return value is opaque.
The method receives the original AST. The result is passed as-is to the
output of `transform`.
Args:
mod: A Python module.
user_context: An opaque object (may be None) that is forwarded to
transform_ast, through the ctx.user_context argument.
Returns:
List[Tuple[Any, Any]]. By default it returns the output of transform_ast,
evaluated on each supported member, other than modules, together with a
`transformer.Context` containing information about the transformation
process."
1033,transform_function,tensorflow/tensorflow/python/autograph/pyct/transpiler.py,328,method,"Transforms a function.
Subclasses may override this method. The return value is opaque.
The method receives the original AST. The result is passed as-is to the
output of `transform`.
Args:
fn: A function or lambda.
user_context: An opaque object (may be None) that is forwarded to
transform_ast, through the ctx.user_context argument.
Returns:
Tuple[Any, Any]. By default it returns the output of transform_ast,
together with a `transformer.Context` containing information about the
transformation process."
1034,PyToPy,tensorflow/tensorflow/python/autograph/pyct/transpiler.py,368,class,"A generic Python-to-Python transpiler.
Its `transform` method offers a function-in, function-out interface.
Internally, it takes care of parsing, caching and loading of the translated
code.
Users typically subclass this, overriding `transform_ast`.
Usually, instances of this class are singletons, since each instance manages
its own cache. The caching can be controlled by overriding `get_caching_key`.
Example:
class MyTransformer(PyToPy):
def transform_ast(self, node, ctx):
node = <<transform node, usually using ast.NodeTransformer classes>>
return node
transformer = MyTransfomer()
new_f, module, source_map = transformer.transform_function(f, ...)
# new_f is a function with signature identical to f
The transformed function has access to the same namespace as the original
function. To allow access to internal APIs, users may inject additional
symbols by overriding `get_extra_locals`."
1035,get_extra_locals,tensorflow/tensorflow/python/autograph/pyct/transpiler.py,402,method,"Returns extra static local variables to be made to transformed code.
Subclasses must override this.
Returns:
extra_locals: A Dict[Text, Any] containing additional variables to make
available to the transformed code."
1036,get_caching_key,tensorflow/tensorflow/python/autograph/pyct/transpiler.py,413,method,"Returns a unique key to use for caching.
Subclasses must override this.
Calls made to `transform_function` with functions that have the same code
object and caching key will return a cached instance on subsequent
invocations.
Args:
user_context: The context object which was passed to `transform`.
Returns:
extra_locals: A hashable."
1037,transform_function,tensorflow/tensorflow/python/autograph/pyct/transpiler.py,436,method,"Transforms a function. See GenericTranspiler.trasnform_function.
This overload wraps the parent's `transform_function`, adding caching and
facilities to instantiate the output as a Python object. It also
adds facilities to make new symbols available to the generated Python code,
visible as local variables - see `get_extra_locals`.
Args:
fn: A function or lambda.
user_context: An opaque object (may be None) that is forwarded to
transform_ast, through the ctx.user_context argument.
Returns:
A tuple:
* A function or lambda with the same signature and closure as `fn`
* The temporary module into which the transformed function was loaded
* The source map as a
Dict[origin_info.LineLocation, origin_info.OriginInfo]"
1038,FlipSignTransformer,tensorflow/tensorflow/python/autograph/pyct/transpiler_test.py,30,class,
1039,visit_BinOp,tensorflow/tensorflow/python/autograph/pyct/transpiler_test.py,32,method,
1040,DummyGensym,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,40,class,A dumb gensym that suffixes a stem by sequential numbers from 1000.
1041,new_name,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,50,method,
1042,ASTEdgePattern,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,60,class,"A pattern defining a type of AST edge.
This consists of three components:
- The type of the parent node, checked with isinstance,
- The name of the field, checked with string equality, and
- The type of the child node, also checked with isinstance.
If all three match, the whole pattern is considered to match.
In all three slots, the special value `anf.ANY` is treated as ""match
anything"". The internal nodes are produced from the `gast` library rather
than the standard `ast` module, which may affect `isinstance` checks."
1043,matches,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,76,method,Computes whether this pattern matches the given edge.
1044,AnfTransformer,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,89,class,Performs the conversion to A-normal form (ANF).
1045,visit_Return,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,256,method,
1046,visit_Delete,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,259,method,
1047,visit_Assign,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,262,method,
1048,visit_AugAssign,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,265,method,
1049,visit_Print,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,268,method,
1050,visit_For,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,271,method,
1051,visit_AsyncFor,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,290,method,
1052,visit_While,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,295,method,
1053,visit_If,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,307,method,
1054,visit_With,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,324,method,
1055,visit_AsyncWith,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,343,method,
1056,visit_Raise,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,348,method,
1057,visit_Assert,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,353,method,
1058,visit_Exec,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,361,method,
1059,visit_Expr,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,366,method,
1060,visit_BoolOp,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,371,method,
1061,visit_BinOp,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,376,method,
1062,visit_UnaryOp,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,379,method,
1063,visit_Lambda,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,382,method,
1064,visit_IfExp,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,387,method,
1065,visit_Dict,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,393,method,
1066,visit_Set,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,396,method,
1067,visit_ListComp,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,399,method,
1068,visit_SetComp,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,405,method,
1069,visit_DictComp,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,411,method,
1070,visit_GeneratorExp,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,417,method,
1071,visit_Await,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,423,method,
1072,visit_Yield,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,428,method,
1073,visit_YieldFrom,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,431,method,
1074,visit_Compare,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,436,method,
1075,visit_Call,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,443,method,
1076,visit_Repr,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,446,method,
1077,visit_FormattedValue,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,451,method,
1078,visit_JoinedStr,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,456,method,
1079,visit_Attribute,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,461,method,
1080,visit_Subscript,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,464,method,
1081,visit_List,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,470,method,
1082,visit_Tuple,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,476,method,
1083,transform,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf.py,527,function,"Converts the given node to A-normal form (ANF).
The general idea of A-normal form: https://en.wikipedia.org/wiki/A-normal_form
The specific converters used here are based on Python AST semantics as
documented at https://greentreesnakes.readthedocs.io/en/latest/.
What exactly should be considered A-normal form for any given programming
language is not completely obvious. The transformation defined here is
therefore configurable as to which syntax to replace with a fresh variable and
which to leave be. The configuration is intentionally flexible enough to
define very precise variable insertion transformations, should that be
desired.
The configuration is a list of syntax rules, each of which is a 2-tuple:
- An `ASTEdgePattern` (which see) defining a type of AST edge, and
- Whether to transform children of such edges.
The special object `anf.ANY` may be used as a pattern that matches all edges.
Each replacement directive is one of three possible things:
- The object `anf.REPLACE`, meaning ""Replace this child node with a variable"",
- The object `anf.LEAVE`, meaning ""Do not replace this child node with a
variable"", or
- A Python callable. If a callable, it is called with the parent node, the
field name, and the child node, and must compute a boolean indicating
whether to transform the child node or not. The callable is free to use
whatever context information it chooses. The callable may be invoked more
than once on the same link, and must produce the same answer each time.
The syntax rules are tested in order, and the first match governs. If no rule
matches, the node is not transformed.
The above rules notwithstanding,
- Variable references are never replaced with (fresh) variables, as that would
accomplish nothing.
- The left-hand children of Assign and AugAssign nodes, and the children of
Del nodes, are never replaced with variables, as that would break their
semantics.
- The right-hand children of Assign nodes are never replaced with variables,
as the original assignment would still have to be present in the result
to define the new variable. (That is, there's no point in transforming
`x = sin(y)` into `tmp = sin(y); x = tmp`.)
- The right-hand children of AugAssign nodes are never replaced with variables
either, but only because the difference from Assign was considered a
potential source of confusion (and it would have been slightly awkward in
the code to treat the RHS differently than the LHS).
- Various special-purpose AST nodes are not exposed to the configuration, lest
the transform produce invalid syntax like, e.g., `tmp = +; x = 1 tmp 2`.
For example, the configuration
```python
[(anf.ASTEdgePattern(anf.ANY, anf.ANY, gast.expr), anf.REPLACE)]
```
gives explicit fresh names to all expressions regardless of context (except as
outlined above), whereas
```python
[(anf.ASTEdgePattern(gast.If, ""test"", anf.ANY), anf.REPLACE)]
```
only transforms the conditionals of `if` statements (but not, e.g., `while`).
If no configuration is supplied, the default behavior is to transform all
expressions except literal constants, which is defined as a configuration as
```python
# For Python 3, and gast library versions before 0.3
literals = (gast.Num, gast.Str, gast.Bytes, gast.NameConstant)
[(anf.ASTEdgePattern(anf.ANY, anf.ANY, literals), anf.LEAVE),
(anf.ASTEdgePattern(anf.ANY, anf.ANY, gast.expr), anf.REPLACE)]
```
Args:
node: The node to transform.
ctx: transformer.EntityInfo. TODO(mdan): What information does this
argument provide?
config: Optional ANF configuration. If omitted, ANF replaces all expression
expect literal constants."
1084,exec_expected_result,tensorflow/tensorflow/python/autograph/pyct/common_transformers/anf_test.py,40,function,
1085,Scope,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,36,class,"Encloses local symbol definition and usage information.
This can track for instance whether a symbol is modified in the current scope.
Note that scopes do not necessarily align with Python's scopes. For example,
the body of an if statement may be considered a separate scope.
Caution - the AST references held by this object are weak.
Scope objects are mutable during construction only, and must be frozen using
`Scope.finalize()` before use. Furthermore, a scope is consistent only after
all its children have been frozen. While analysing code blocks, scopes are
being gradually built, from the innermost scope outward. Freezing indicates
that the analysis of a code block is complete. Once frozen, mutation is no
longer allowed. `is_final` tracks whether the scope is frozen or not. Certain
properties, like `referenced`, are only accurate when called on frozen scopes.
Attributes:
parent: Optional[Scope], the parent scope, if any.
isolated: bool, whether the scope is a true Python scope (e.g. the scope of
a function), or just a surrogate tracking an ordinary code block. Using
the terminology of the Python 3 reference documentation, True roughly
represents an actual scope, whereas False represents an ordinary code
block.
function_name: Optional[str], name of the function owning this scope.
isolated_names: Set[qual_names.QN], identifiers that are isolated to this
scope (even if the scope is not isolated).
annotations: Set[qual_names.QN], identifiers used as type annotations
in this scope.
read: Set[qual_names.QN], identifiers read in this scope.
modified: Set[qual_names.QN], identifiers modified in this scope.
deleted: Set[qual_names.QN], identifiers deleted in this scope.
bound: Set[qual_names.QN], names that are bound to this scope. See
https://docs.python.org/3/reference/executionmodel.html#binding-of-names
for a precise definition.
globals: Set[qual_names.QN], names that are explicitly marked as global in
this scope. Note that this doesn't include free read-only vars bound to
global symbols.
nonlocals: Set[qual_names.QN], names that are explicitly marked as nonlocal
in this scope. Note that this doesn't include free read-only vars bound to
global symbols.
free_vars: Set[qual_names.QN], the free variables in this scope. See
https://docs.python.org/3/reference/executionmodel.html for a precise
definition.
params: WeakValueDictionary[qual_names.QN, ast.Node], function arguments
visible in this scope, mapped to the function node that defines them.
enclosing_scope: Scope, the innermost isolated scope that is a transitive
parent of this scope. May be the scope itself.
referenced: Set[qual_names.QN], the totality of the symbols used by this
scope and its parents.
is_final: bool, whether the scope is frozen or not.
Note - simple statements may never delete and modify a symbol at the same
time. However, compound ones like if statements can. In that latter case, it's
undefined whether the symbol is actually modified or deleted upon statement
exit. Certain analyses like reaching definitions need to be careful about
this."
1086,enclosing_scope,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,130,method,
1087,referenced,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,137,method,
1088,free_vars,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,143,method,
1089,copy_from,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,147,method,Recursively copies the contents of this scope from another scope.
1090,copy_of,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,162,method,
1091,merge_from,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,172,method,Adds all activity from another scope to this scope.
1092,finalize,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,185,method,Freezes this scope.
1093,mark_param,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,207,method,
1094,ActivityAnalyzer,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,230,class,"Annotates nodes with local scope information.
See Scope.
The use of this class requires that qual_names.resolve() has been called on
the node. This class will ignore nodes have not been
annotated with their qualified names."
1095,visit_Import,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,353,method,
1096,visit_ImportFrom,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,356,method,
1097,visit_Global,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,359,method,
1098,visit_Nonlocal,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,368,method,
1099,visit_Expr,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,378,method,
1100,visit_Raise,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,381,method,
1101,visit_Return,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,384,method,
1102,visit_Assign,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,387,method,
1103,visit_AnnAssign,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,390,method,
1104,visit_AugAssign,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,399,method,
1105,visit_Delete,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,413,method,
1106,visit_Name,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,416,method,
1107,visit_alias,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,422,method,
1108,visit_Attribute,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,435,method,
1109,visit_Subscript,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,443,method,
1110,visit_Print,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,450,method,
1111,visit_Assert,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,457,method,
1112,visit_Call,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,460,method,
1113,visit_comprehension,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,509,method,
1114,visit_DictComp,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,516,method,
1115,visit_ListComp,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,519,method,
1116,visit_SetComp,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,522,method,
1117,visit_GeneratorExp,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,525,method,
1118,visit_ClassDef,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,528,method,
1119,visit_FunctionDef,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,568,method,
1120,visit_Lambda,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,606,method,
1121,visit_With,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,648,method,
1122,visit_withitem,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,654,method,
1123,visit_If,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,657,method,
1124,visit_For,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,668,method,
1125,visit_While,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,685,method,
1126,visit_ExceptHandler,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,696,method,
1127,resolve,tensorflow/tensorflow/python/autograph/pyct/static_analysis/activity.py,707,function,
1128,NoValue,tensorflow/tensorflow/python/autograph/pyct/static_analysis/annos.py,27,class,
1129,NodeAnno,tensorflow/tensorflow/python/autograph/pyct/static_analysis/annos.py,33,class,"Additional annotations used by the static analyzer.
These are in addition to the basic annotations declared in anno.py."
1130,Analyzer,tensorflow/tensorflow/python/autograph/pyct/static_analysis/liveness.py,40,class,CFG visitor that performs liveness analysis at statement level.
1131,init_state,tensorflow/tensorflow/python/autograph/pyct/static_analysis/liveness.py,47,method,
1132,visit_node,tensorflow/tensorflow/python/autograph/pyct/static_analysis/liveness.py,50,method,
1133,TreeAnnotator,tensorflow/tensorflow/python/autograph/pyct/static_analysis/liveness.py,96,class,"Runs liveness analysis on each of the functions defined in the AST.
If a function defined other local functions, those will have separate CFGs.
However, dataflow analysis needs to tie up these CFGs to properly emulate the
effect of closures. In the case of liveness, the parent function's live
variables must account for the variables that are live at the entry of each
subfunction. For example:
def foo():
# baz is live from here on
def bar():
print(baz)
This analyzer runs liveness analysis on each individual function, accounting
for the effect above."
1134,visit,tensorflow/tensorflow/python/autograph/pyct/static_analysis/liveness.py,121,method,
1135,visit_Lambda,tensorflow/tensorflow/python/autograph/pyct/static_analysis/liveness.py,142,method,
1136,visit_FunctionDef,tensorflow/tensorflow/python/autograph/pyct/static_analysis/liveness.py,145,method,
1137,visit_If,tensorflow/tensorflow/python/autograph/pyct/static_analysis/liveness.py,168,method,
1138,visit_For,tensorflow/tensorflow/python/autograph/pyct/static_analysis/liveness.py,173,method,
1139,visit_While,tensorflow/tensorflow/python/autograph/pyct/static_analysis/liveness.py,178,method,
1140,visit_Try,tensorflow/tensorflow/python/autograph/pyct/static_analysis/liveness.py,183,method,
1141,visit_ExceptHandler,tensorflow/tensorflow/python/autograph/pyct/static_analysis/liveness.py,188,method,
1142,visit_With,tensorflow/tensorflow/python/autograph/pyct/static_analysis/liveness.py,193,method,
1143,visit_Expr,tensorflow/tensorflow/python/autograph/pyct/static_analysis/liveness.py,197,method,
1144,resolve,tensorflow/tensorflow/python/autograph/pyct/static_analysis/liveness.py,206,function,"Resolves the live symbols at the exit of control flow statements.
Args:
node: ast.AST
source_info: transformer.SourceInfo
graphs: Dict[ast.FunctionDef, cfg.Graph]
include_annotations: Bool, whether type annotations should be included in
the analysis.
Returns:
ast.AST"
1145,Definition,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_definitions.py,40,class,"Definition objects describe a unique definition of a variable.
Subclasses of this may be used by passing an appropriate factory function to
resolve.
Attributes:
param_of: Optional[ast.AST]
directives: Dict, optional definition annotations"
1146,Analyzer,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_definitions.py,112,class,CFG visitor that determines reaching definitions at statement level.
1147,init_state,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_definitions.py,120,method,
1148,visit_node,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_definitions.py,123,method,
1149,TreeAnnotator,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_definitions.py,169,class,"AST visitor that annotates each symbol name with its reaching definitions.
Simultaneously, the visitor runs the dataflow analysis on each function node,
accounting for the effect of closures. For example:
def foo():
bar = 1
def baz():
# bar = 1 reaches here"
1150,visit_FunctionDef,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_definitions.py,189,method,
1151,visit_Name,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_definitions.py,204,method,
1152,visit_If,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_definitions.py,233,method,
1153,visit_For,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_definitions.py,237,method,
1154,visit_While,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_definitions.py,253,method,
1155,visit_Try,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_definitions.py,257,method,
1156,visit_ExceptHandler,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_definitions.py,261,method,
1157,visit,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_definitions.py,267,method,
1158,resolve,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_definitions.py,279,function,"Resolves reaching definitions for each symbol.
Args:
node: ast.AST
source_info: transformer.SourceInfo
graphs: Dict[ast.FunctionDef, cfg.Graph]
definition_factory: Callable[[], Definition]
Returns:
ast.AST"
1159,Definition,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_fndefs.py,32,class,Definition objects describe a unique definition of a function.
1160,Analyzer,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_fndefs.py,76,class,CFG visitor that determines reaching definitions at statement level.
1161,init_state,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_fndefs.py,85,method,
1162,visit_node,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_fndefs.py,88,method,
1163,TreeAnnotator,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_fndefs.py,109,class,"AST visitor that annotates each symbol name with its reaching definitions.
Simultaneously, the visitor runs the dataflow analysis on each function node,
accounting for the effect of closures. For example:
def foo():
def f():
pass
def g():
# `def f` reaches here"
1164,visit_FunctionDef,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_fndefs.py,147,method,
1165,visit_Lambda,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_fndefs.py,150,method,
1166,visit,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_fndefs.py,153,method,
1167,resolve,tensorflow/tensorflow/python/autograph/pyct/static_analysis/reaching_fndefs.py,170,function,"Resolves reaching definitions for each symbol.
Args:
node: ast.AST
source_info: transformer.SourceInfo
graphs: Dict[ast.FunctionDef, cfg.Graph]
Returns:
ast.AST"
1168,Resolver,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,41,class,"Resolver objects handle the process of looking up actual names and types.
All resolve_* methods:
* have a first namespace argument, mapping string to actual values
* specify names as QN objects
* specify types as a Set of inferred types
All resolve_* methods must return either:
* a set of `type` objects
* None"
1169,res_name,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,54,method,"Resolves the type an external (e.g. closure, global) variable."
1170,res_value,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,58,method,Resolves the type a literal value.
1171,res_call,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,63,method,"Resolves the return type an external function or method call.
Args:
ns: namespace
name: str, the function name
target: if this is a method call, the types of the method target, None
otherwise
args: list or argument types
keywords: dict of name to argument types
starargs: list of types of the *args arguments (should be at most one)
kwargs: list of types of the **kwargs arguments (in order of appearance)"
1172,res_arg,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,78,method,Resolves the type of a (possibly annotated) function argument.
1173,StmtInferrer,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,162,class,"Runs type inference on a single AST statement.
This visitor annotates most nodes with type information. It also sets types
for the symbols modified by this statement in its types_out property."
1174,visit,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,178,method,
1175,visit_FunctionDef,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,185,method,
1176,visit_Constant,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,189,method,
1177,visit_Tuple,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,192,method,
1178,visit_List,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,203,method,
1179,visit_Set,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,212,method,
1180,visit_Name,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,215,method,
1181,visit_Call,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,244,method,
1182,visit_Index,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,273,method,
1183,visit_Assign,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,276,method,
1184,visit_Subscript,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,284,method,
1185,visit_Compare,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,294,method,
1186,visit_BinOp,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,316,method,
1187,Analyzer,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,329,class,CFG visitor that propagates type information across statements.
1188,init_state,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,348,method,
1189,visit_node,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,364,method,
1190,FunctionVisitor,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,394,class,AST visitor that applies type inference to each function separately.
1191,visit_FunctionDef,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,402,method,
1192,resolve,tensorflow/tensorflow/python/autograph/pyct/static_analysis/type_inference.py,417,function,"Performs type inference.
Args:
node: ast.AST
source_info: transformer.SourceInfo
graphs: Dict[ast.FunctionDef, cfg.Graph]
resolver: Resolver
Returns:
ast.AST"
1193,simple_function,tensorflow/tensorflow/python/autograph/pyct/testing/basic_definitions.py,23,function,Docstring.
1194,nested_functions,tensorflow/tensorflow/python/autograph/pyct/testing/basic_definitions.py,28,function,Docstring.
1195,function_with_print,tensorflow/tensorflow/python/autograph/pyct/testing/basic_definitions.py,37,function,
1196,SimpleClass,tensorflow/tensorflow/python/autograph/pyct/testing/basic_definitions.py,44,class,
1197,simple_method,tensorflow/tensorflow/python/autograph/pyct/testing/basic_definitions.py,46,method,
1198,method_with_print,tensorflow/tensorflow/python/autograph/pyct/testing/basic_definitions.py,49,method,
1199,function_with_multiline_call,tensorflow/tensorflow/python/autograph/pyct/testing/basic_definitions.py,53,function,Docstring.
1200,basic_decorator,tensorflow/tensorflow/python/autograph/pyct/testing/basic_definitions.py,61,function,
1201,decorated_function,tensorflow/tensorflow/python/autograph/pyct/testing/basic_definitions.py,67,function,
1202,NodeSampler,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,30,class,
1203,sample,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,33,method,
1204,StatementSampler,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,39,class,
1205,ExpressionSampler,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,49,class,
1206,CompareSampler,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,58,class,
1207,BinaryOpSampler,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,71,class,
1208,UnaryOpSampler,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,83,class,
1209,NameSampler,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,87,class,
1210,CodeGenerator,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,98,class,Generate random syntactically-valid Python ASTs.
1211,generate_statement,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,105,method,"Generate a statement node, dispatching to the correct class method."
1212,sample_node_list,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,124,method,"Generate a list of statements of random length.
Args:
low: Fewest number of statements to generate.
high: Highest number of statements to generate.
generator: Function to call to generate nodes.
Returns:
A list of statements."
1213,generate_Name,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,140,method,
1214,generate_BinOp,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,145,method,
1215,generate_Compare,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,151,method,
1216,generate_UnaryOp,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,155,method,
1217,generate_expression,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,160,method,
1218,generate_Assign,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,167,method,Generate an Assign node.
1219,generate_If,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,177,method,Generate an If node.
1220,generate_While,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,196,method,Generate a While node.
1221,generate_Call,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,207,method,
1222,generate_Return,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,210,method,
1223,generate_Print,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,213,method,
1224,generate_FunctionDef,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,216,method,Generate a FunctionDef node.
1225,generate_random_functiondef,tensorflow/tensorflow/python/autograph/pyct/testing/codegen.py,233,function,
1226,wrapping_decorator,tensorflow/tensorflow/python/autograph/pyct/testing/decorators.py,24,function,
1227,standalone_decorator,tensorflow/tensorflow/python/autograph/pyct/testing/decorators.py,33,function,
1228,functional_decorator,tensorflow/tensorflow/python/autograph/pyct/testing/decorators.py,41,function,
1229,set_verbosity,tensorflow/tensorflow/python/autograph/utils/ag_logging.py,41,function,"Sets the AutoGraph verbosity level.
_Debug logging in AutoGraph_
More verbose logging is useful to enable when filing bug reports or doing
more in-depth debugging.
There are two means to control the logging verbosity:
* The `set_verbosity` function
* The `AUTOGRAPH_VERBOSITY` environment variable
`set_verbosity` takes precedence over the environment variable.
For example:
```python
import os
import tensorflow as tf
os.environ['AUTOGRAPH_VERBOSITY'] = 5
# Verbosity is now 5
tf.autograph.set_verbosity(0)
# Verbosity is now 0
os.environ['AUTOGRAPH_VERBOSITY'] = 1
# No effect, because set_verbosity was already called.
```
Logs entries are output to [absl](https://abseil.io)'s
[default output](https://abseil.io/docs/python/guides/logging),
with `INFO` level.
Logs can be mirrored to stdout by using the `alsologtostdout` argument.
Mirroring is enabled by default when Python runs in interactive mode.
Args:
level: int, the verbosity level; larger values specify increased verbosity;
0 means no logging. When reporting bugs, it is recommended to set this
value to a larger number, like 10.
alsologtostdout: bool, whether to also output log messages to `sys.stdout`."
1230,trace,tensorflow/tensorflow/python/autograph/utils/ag_logging.py,92,function,"Traces argument information at compilation time.
`trace` is useful when debugging, and it always executes during the tracing
phase, that is, when the TF graph is constructed.
_Example usage_
```python
import tensorflow as tf
for i in tf.range(10):
tf.autograph.trace(i)
# Output: <Tensor ...>
```
Args:
*args: Arguments to print to `sys.stdout`."
1231,get_verbosity,tensorflow/tensorflow/python/autograph/utils/ag_logging.py,114,function,
1232,has_verbosity,tensorflow/tensorflow/python/autograph/utils/ag_logging.py,121,function,
1233,error,tensorflow/tensorflow/python/autograph/utils/ag_logging.py,131,function,
1234,log,tensorflow/tensorflow/python/autograph/utils/ag_logging.py,138,function,
1235,warn,tensorflow/tensorflow/python/autograph/utils/ag_logging.py,145,function,
1236,BasicRef,tensorflow/tensorflow/python/autograph/utils/compat_util.py,27,class,This shim emulates the nonlocal keyword in Py2-compatible source.
1237,deprecated_py2_support,tensorflow/tensorflow/python/autograph/utils/compat_util.py,34,function,Swaps calling module with a Py2-specific implementation. Noop in Py3.
1238,control_dependency_on_returns,tensorflow/tensorflow/python/autograph/utils/context_managers.py,27,function,"Create a TF control dependency on the return values of a function.
If the function had no return value, a no-op context is returned.
Args:
return_value: The return value to set as control dependency.
Returns:
A context manager."
1239,alias_tensors,tensorflow/tensorflow/python/autograph/utils/misc.py,27,function,"Wraps any Tensor arguments with an identity op.
Any other argument, including Variables, is returned unchanged.
Args:
*args: Any arguments. Must contain at least one element.
Returns:
Same as *args, with Tensor instances replaced as described.
Raises:
ValueError: If args doesn't meet the requirements."
1240,get_range_len,tensorflow/tensorflow/python/autograph/utils/misc.py,55,function,
1241,MatchDType,tensorflow/tensorflow/python/autograph/utils/py_func.py,28,class,"Allows matching the dtype of an argument.
Used in conjunction with function calls. For example, MatchDType(0) will
match the DType of the first argument."
1242,wrap_py_func,tensorflow/tensorflow/python/autograph/utils/py_func.py,38,function,"Helper that wraps a callable to py_func.
The helper passes tensor arguments through the py_func interface. Non-tensor
arguments are allowed, and will be passed to f directly. Note that non-tensor
arguments are captured by f will not update every time the wrapper is
called (this is consistent with its argument list, which only includes
the tensor arguments). In general, it's safest not to reuse this wrapper.
Args:
f: Callable
return_dtypes: None, individual of tuple/list of DType or MatchDType, the
data type for each of f's return value(s). Set to None if f has no
return values or use_dummy_return is True. Use MatchDType to define a
dtype identical to that of `i`th argument (argument 0 is the first);
an argument must of Tensor type if it is to be used with MatchDType.
args: Positional arguments for f, as list or tuple.
kwargs: Keyword arguments for f, as dict with string keys. May be None.
use_dummy_return: If True, the function will return a dummy value of 1
and discard its actual return value.
Returns:
The return values of f converted to tensor.
Raises:
ValueError: if any of the arguments are incorrect."
1243,dynamic_list_append,tensorflow/tensorflow/python/autograph/utils/tensor_list.py,26,function,Converts a list append call inline.
1244,TensorList,tensorflow/tensorflow/python/autograph/utils/tensor_list.py,43,class,Tensor list wrapper API-compatible with Python built-in list.
1245,append,tensorflow/tensorflow/python/autograph/utils/tensor_list.py,51,method,
1246,pop,tensorflow/tensorflow/python/autograph/utils/tensor_list.py,54,method,
1247,clear,tensorflow/tensorflow/python/autograph/utils/tensor_list.py,58,method,
1248,count,tensorflow/tensorflow/python/autograph/utils/tensor_list.py,61,method,
1249,is_dense_tensor,tensorflow/tensorflow/python/autograph/utils/tensors.py,32,function,
1250,is_tensor_array,tensorflow/tensorflow/python/autograph/utils/tensors.py,38,function,
1251,is_tensor_list,tensorflow/tensorflow/python/autograph/utils/tensors.py,42,function,
1252,is_range_tensor,tensorflow/tensorflow/python/autograph/utils/tensors.py,51,function,Returns True if a tensor is the result of a tf.range op. Best effort.
1253,list_local_devices,tensorflow/tensorflow/python/client/device_lib.py,25,function,"List the available devices available in the local process.
Args:
session_config: a session config proto or None to use the default config.
Returns:
A list of `DeviceAttribute` protocol buffers."
1254,TF_NewSessionOptions,tensorflow/tensorflow/python/client/pywrap_tf_session.py,51,function,
1255,TF_Reset,tensorflow/tensorflow/python/client/pywrap_tf_session.py,65,function,
1256,SessionInterface,tensorflow/tensorflow/python/client/session.py,51,class,Base class for implementations of TensorFlow client sessions.
1257,graph,tensorflow/tensorflow/python/client/session.py,55,method,"The underlying TensorFlow graph, to be used in building Operations."
1258,sess_str,tensorflow/tensorflow/python/client/session.py,60,method,The TensorFlow process to which this session will connect.
1259,run,tensorflow/tensorflow/python/client/session.py,64,method,Runs operations in the session. See `BaseSession.run()` for details.
1260,partial_run_setup,tensorflow/tensorflow/python/client/session.py,68,method,Sets up the feeds and fetches for partial runs in the session.
1261,partial_run,tensorflow/tensorflow/python/client/session.py,72,method,Continues the execution with additional feeds and fetches.
1262,register_session_run_conversion_functions,tensorflow/tensorflow/python/client/session.py,144,function,"Register fetch and feed conversion functions for `tf.Session.run()`.
This function registers a triple of conversion functions for fetching and/or
feeding values of user-defined types in a call to tf.Session.run().
An example
```python
class SquaredTensor(object):
def __init__(self, tensor):
self.sq = tf.square(tensor)
#you can define conversion functions as follows:
fetch_function = lambda squared_tensor:([squared_tensor.sq],
lambda val: val[0])
feed_function = lambda feed, feed_val: [(feed.sq, feed_val)]
feed_function_for_partial_run = lambda feed: [feed.sq]
#then after invoking this register function, you can use as follows:
session.run(squared_tensor1,
feed_dict = {squared_tensor2 : some_numpy_array})
```
Args:
tensor_type: The type for which you want to register a conversion function.
fetch_function: A callable that takes an object of type `tensor_type` and
returns a tuple, where the first element is a list of `tf.Tensor` objects,
and the second element is a callable that takes a list of ndarrays and
returns an object of some value type that corresponds to `tensor_type`.
fetch_function describes how to expand fetch into its component Tensors
and how to contract the fetched results back into a single return value.
feed_function: A callable that takes feed_key and feed_value as input, and
returns a list of tuples (feed_tensor, feed_val), feed_key must have type
`tensor_type`, and feed_tensor must have type `tf.Tensor`. Each feed
function describes how to unpack a single fed value and map it to feeds of
one or more tensors and their corresponding values.
feed_function_for_partial_run: A callable for specifying tensor values to
feed when setting up a partial run, which takes a `tensor_type` type
object as input, and returns a list of Tensors.
Raises:
ValueError: If `tensor_type` has already been registered."
1263,BaseSession,tensorflow/tensorflow/python/client/session.py,627,class,"A class for interacting with a TensorFlow computation.
The BaseSession enables incremental graph building with inline
execution of Operations and evaluation of Tensors."
1264,list_devices,tensorflow/tensorflow/python/client/session.py,706,method,"Lists available devices in this session.
```python
devices = sess.list_devices()
for d in devices:
print(d.name)
```
Where:
Each element in the list has the following properties
name: A string with the full name of the device. ex:
`/job:worker/replica:0/task:3/device:CPU:0`
device_type: The type of the device (e.g. `CPU`, `GPU`, `TPU`.)
memory_limit: The maximum amount of memory available on the device.
Note: depending on the device, it is possible the usable memory could
be substantially less.
Raises:
tf.errors.OpError: If it encounters an error (e.g. session is in an
invalid state, or network errors occur).
Returns:
A list of devices in the session."
1265,close,tensorflow/tensorflow/python/client/session.py,744,method,"Closes this session.
Calling this method frees all resources associated with the session.
Raises:
tf.errors.OpError: Or one of its subclasses if an error occurs while
closing the TensorFlow session."
1266,graph,tensorflow/tensorflow/python/client/session.py,775,method,The graph that was launched in this session.
1267,graph_def,tensorflow/tensorflow/python/client/session.py,780,method,"A serializable version of the underlying TensorFlow graph.
Returns:
A graph_pb2.GraphDef proto containing nodes for all of the Operations in
the underlying TensorFlow graph."
1268,sess_str,tensorflow/tensorflow/python/client/session.py,790,method,
1269,as_default,tensorflow/tensorflow/python/client/session.py,793,method,"Returns a context manager that makes this object the default session.
Use with the `with` keyword to specify that calls to
`tf.Operation.run` or `tf.Tensor.eval` should be executed in
this session.
```python
c = tf.constant(..)
sess = tf.compat.v1.Session()
with sess.as_default():
assert tf.compat.v1.get_default_session() is sess
print(c.eval())
```
To get the current default session, use `tf.compat.v1.get_default_session`.
*N.B.* The `as_default` context manager *does not* close the
session when you exit the context, and you must close the session
explicitly.
```python
c = tf.constant(...)
sess = tf.compat.v1.Session()
with sess.as_default():
print(c.eval())
# ...
with sess.as_default():
print(c.eval())
sess.close()
```
Alternatively, you can use `with tf.compat.v1.Session():` to create a
session that is automatically closed on exiting the context,
including when an uncaught exception is raised.
*N.B.* The default session is a property of the current thread. If you
create a new thread, and wish to use the default session in that
thread, you must explicitly add a `with sess.as_default():` in that
thread's function.
*N.B.* Entering a `with sess.as_default():` block does not affect
the current default graph. If you are using multiple graphs, and
`sess.graph` is different from the value of
`tf.compat.v1.get_default_graph`, you must explicitly enter a
`with sess.graph.as_default():` block to make `sess.graph` the default
graph.
Returns:
A context manager using this session as the default session."
1270,run,tensorflow/tensorflow/python/client/session.py,848,method,"Runs operations and evaluates tensors in `fetches`.
This method runs one ""step"" of TensorFlow computation, by
running the necessary graph fragment to execute every `Operation`
and evaluate every `Tensor` in `fetches`, substituting the values in
`feed_dict` for the corresponding input values.
The `fetches` argument may be a single graph element, or an arbitrarily
nested list, tuple, namedtuple, dict, or OrderedDict containing graph
elements at its leaves. A graph element can be one of the following types:
* A `tf.Operation`.
The corresponding fetched value will be `None`.
* A `tf.Tensor`.
The corresponding fetched value will be a numpy ndarray containing the
value of that tensor.
* A `tf.sparse.SparseTensor`.
The corresponding fetched value will be a
`tf.compat.v1.SparseTensorValue`
containing the value of that sparse tensor.
* A `get_tensor_handle` op. The corresponding fetched value will be a
numpy ndarray containing the handle of that tensor.
* A `string` which is the name of a tensor or operation in the graph.
The value returned by `run()` has the same shape as the `fetches` argument,
where the leaves are replaced by the corresponding values returned by
TensorFlow.
Example:
```python
a = tf.constant([10, 20])
b = tf.constant([1.0, 2.0])
# 'fetches' can be a singleton
v = session.run(a)
# v is the numpy array [10, 20]
# 'fetches' can be a list.
v = session.run([a, b])
# v is a Python list with 2 numpy arrays: the 1-D array [10, 20] and the
# 1-D array [1.0, 2.0]
# 'fetches' can be arbitrary lists, tuples, namedtuple, dicts:
MyData = collections.namedtuple('MyData', ['a', 'b'])
v = session.run({'k1': MyData(a, b), 'k2': [b, a]})
# v is a dict with
# v['k1'] is a MyData namedtuple with 'a' (the numpy array [10, 20]) and
# 'b' (the numpy array [1.0, 2.0])
# v['k2'] is a list with the numpy array [1.0, 2.0] and the numpy array
# [10, 20].
```
The optional `feed_dict` argument allows the caller to override
the value of tensors in the graph. Each key in `feed_dict` can be
one of the following types:
* If the key is a `tf.Tensor`, the
value may be a Python scalar, string, list, or numpy ndarray
that can be converted to the same `dtype` as that
tensor. Additionally, if the key is a
`tf.compat.v1.placeholder`, the shape of
the value will be checked for compatibility with the placeholder.
* If the key is a
`tf.sparse.SparseTensor`,
the value should be a
`tf.compat.v1.SparseTensorValue`.
* If the key is a nested tuple of `Tensor`s or `SparseTensor`s, the value
should be a nested tuple with the same structure that maps to their
corresponding values as above.
Each value in `feed_dict` must be convertible to a numpy array of the dtype
of the corresponding key.
The optional `options` argument expects a [`RunOptions`] proto. The options
allow controlling the behavior of this particular step (e.g. turning tracing
on).
The optional `run_metadata` argument expects a [`RunMetadata`] proto. When
appropriate, the non-Tensor output of this step will be collected there. For
example, when users turn on tracing in `options`, the profiled info will be
collected into this argument and passed back.
Args:
fetches: A single graph element, a list of graph elements, or a dictionary
whose values are graph elements or lists of graph elements (described
above).
feed_dict: A dictionary that maps graph elements to values (described
above).
options: A [`RunOptions`] protocol buffer
run_metadata: A [`RunMetadata`] protocol buffer
Returns:
Either a single value if `fetches` is a single graph element, or
a list of values if `fetches` is a list, or a dictionary with the
same keys as `fetches` if that is a dictionary (described above).
Order in which `fetches` operations are evaluated inside the call
is undefined.
Raises:
RuntimeError: If this `Session` is in an invalid state (e.g. has been
closed).
TypeError: If `fetches` or `feed_dict` keys are of an inappropriate type.
ValueError: If `fetches` or `feed_dict` keys are invalid or refer to a
`Tensor` that doesn't exist."
1271,partial_run,tensorflow/tensorflow/python/client/session.py,969,method,"Continues the execution with more feeds and fetches.
This is EXPERIMENTAL and subject to change.
To use partial execution, a user first calls `partial_run_setup()` and
then a sequence of `partial_run()`. `partial_run_setup` specifies the
list of feeds and fetches that will be used in the subsequent
`partial_run` calls.
The optional `feed_dict` argument allows the caller to override
the value of tensors in the graph. See run() for more information.
Below is a simple example:
```python
a = array_ops.placeholder(dtypes.float32, shape=[])
b = array_ops.placeholder(dtypes.float32, shape=[])
c = array_ops.placeholder(dtypes.float32, shape=[])
r1 = math_ops.add(a, b)
r2 = math_ops.multiply(r1, c)
h = sess.partial_run_setup([r1, r2], [a, b, c])
res = sess.partial_run(h, r1, feed_dict={a: 1, b: 2})
res = sess.partial_run(h, r2, feed_dict={c: res})
```
Args:
handle: A handle for a sequence of partial runs.
fetches: A single graph element, a list of graph elements, or a dictionary
whose values are graph elements or lists of graph elements (see
documentation for `run`).
feed_dict: A dictionary that maps graph elements to values (described
above).
Returns:
Either a single value if `fetches` is a single graph element, or
a list of values if `fetches` is a list, or a dictionary with the
same keys as `fetches` if that is a dictionary
(see documentation for `run`).
Raises:
tf.errors.OpError: Or one of its subclasses on error."
1272,partial_run_setup,tensorflow/tensorflow/python/client/session.py,1016,method,"Sets up a graph with feeds and fetches for partial run.
This is EXPERIMENTAL and subject to change.
Note that contrary to `run`, `feeds` only specifies the graph elements.
The tensors will be supplied by the subsequent `partial_run` calls.
Args:
fetches: A single graph element, or a list of graph elements.
feeds: A single graph element, or a list of graph elements.
Returns:
A handle for partial run.
Raises:
RuntimeError: If this `Session` is in an invalid state (e.g. has been
closed).
TypeError: If `fetches` or `feed_dict` keys are of an inappropriate type.
tf.errors.OpError: Or one of its subclasses if a TensorFlow error happens."
1273,make_callable,tensorflow/tensorflow/python/client/session.py,1186,method,"Returns a Python callable that runs a particular step.
The returned callable will take `len(feed_list)` arguments whose types
must be compatible feed values for the respective elements of `feed_list`.
For example, if element `i` of `feed_list` is a `tf.Tensor`, the `i`th
argument to the returned callable must be a numpy ndarray (or something
convertible to an ndarray) with matching element type and shape. See
`tf.Session.run` for details of the allowable feed key and value types.
The returned callable will have the same return type as
`tf.Session.run(fetches, ...)`. For example, if `fetches` is a `tf.Tensor`,
the callable will return a numpy ndarray; if `fetches` is a `tf.Operation`,
it will return `None`.
Args:
fetches: A value or list of values to fetch. See `tf.Session.run` for
details of the allowable fetch types.
feed_list: (Optional.) A list of `feed_dict` keys. See `tf.Session.run`
for details of the allowable feed key types.
accept_options: (Optional.) If `True`, the returned `Callable` will be
able to accept `tf.compat.v1.RunOptions` and `tf.compat.v1.RunMetadata`
as optional keyword arguments `options` and `run_metadata`,
respectively, with the same syntax and semantics as `tf.Session.run`,
which is useful for certain use cases (profiling and debugging) but will
result in measurable slowdown of the `Callable`'s
performance. Default: `False`.
Returns:
A function that when called will execute the step defined by
`feed_list` and `fetches` in this session.
Raises:
TypeError: If `fetches` or `feed_list` cannot be interpreted
as arguments to `tf.Session.run`."
1274,Session,tensorflow/tensorflow/python/client/session.py,1509,class,"A class for running TensorFlow operations.
A `Session` object encapsulates the environment in which `Operation`
objects are executed, and `Tensor` objects are evaluated. For
example:
```python
tf.compat.v1.disable_eager_execution() # need to disable eager in TF2.x
# Build a graph.
a = tf.constant(5.0)
b = tf.constant(6.0)
c = a * b
# Launch the graph in a session.
sess = tf.compat.v1.Session()
# Evaluate the tensor `c`.
print(sess.run(c)) # prints 30.0
```
A session may own resources, such as
`tf.Variable`, `tf.queue.QueueBase`,
and `tf.compat.v1.ReaderBase`. It is important to release
these resources when they are no longer required. To do this, either
invoke the `tf.Session.close` method on the session, or use
the session as a context manager. The following two examples are
equivalent:
```python
# Using the `close()` method.
sess = tf.compat.v1.Session()
sess.run(...)
sess.close()
# Using the context manager.
with tf.compat.v1.Session() as sess:
sess.run(...)
```
The
[`ConfigProto`](https://www.tensorflow.org/code/tensorflow/core/protobuf/config.proto)
protocol buffer exposes various configuration options for a
session. For example, to create a session that uses soft constraints
for device placement, and log the resulting placement decisions,
create a session as follows:
```python
# Launch the graph in a session that allows soft device placement and
# logs the placement decisions.
sess = tf.compat.v1.Session(config=tf.compat.v1.ConfigProto(
allow_soft_placement=True,
log_device_placement=True))
```"
1275,reset,tensorflow/tensorflow/python/client/session.py,1644,method,"Resets resource containers on `target`, and close all connected sessions.
A resource container is distributed across all workers in the
same cluster as `target`. When a resource container on `target`
is reset, resources associated with that container will be cleared.
In particular, all Variables in the container will become undefined:
they lose their values and shapes.
NOTE:
(i) reset() is currently only implemented for distributed sessions.
(ii) Any sessions on the master named by `target` will be closed.
If no resource containers are provided, all containers are reset.
Args:
target: The execution engine to connect to.
containers: A list of resource container name strings, or `None` if all of
all the containers are to be reset.
config: (Optional.) Protocol buffer with configuration options.
Raises:
tf.errors.OpError: Or one of its subclasses if an error occurs while
resetting containers."
1276,InteractiveSession,tensorflow/tensorflow/python/client/session.py,1679,class,"A TensorFlow `Session` for use in interactive contexts, such as a shell.
The only difference with a regular `Session` is that an `InteractiveSession`
installs itself as the default session on construction.
The methods `tf.Tensor.eval`
and `tf.Operation.run`
will use that session to run ops.
This is convenient in interactive shells and [IPython
notebooks](http://ipython.org), as it avoids having to pass an explicit
`Session` object to run ops.
For example:
```python
sess = tf.compat.v1.InteractiveSession()
a = tf.constant(5.0)
b = tf.constant(6.0)
c = a * b
# We can just use 'c.eval()' without passing 'sess'
print(c.eval())
sess.close()
```
Note that a regular session installs itself as the default session when it
is created in a `with` statement. The common usage in non-interactive
programs is to follow that pattern:
```python
a = tf.constant(5.0)
b = tf.constant(6.0)
c = a * b
with tf.compat.v1.Session():
# We can also use 'c.eval()' here.
print(c.eval())
```"
1277,close,tensorflow/tensorflow/python/client/session.py,1771,method,Closes an `InteractiveSession`.
1278,SessionBenchmark,tensorflow/tensorflow/python/client/session_benchmark.py,36,class,Tests and benchmarks for interacting with the `tf.compat.v1.Session`.
1279,benchmarkGrpcSession,tensorflow/tensorflow/python/client/session_benchmark.py,178,method,
1280,benchmarkDirectSession,tensorflow/tensorflow/python/client/session_benchmark.py,204,method,
1281,AllocationMaximum,tensorflow/tensorflow/python/client/timeline.py,32,class,"Stores the maximum allocation for a given allocator within the timelne.
Parameters:
timestamp: `tensorflow::Env::NowMicros()` when this maximum was reached.
num_bytes: the total memory used at this time.
tensors: the set of tensors allocated at this time."
1282,StepStatsAnalysis,tensorflow/tensorflow/python/client/timeline.py,44,class,"Stores the step stats analysis output.
Parameters:
chrome_trace: A dict containing the chrome trace analysis.
allocator_maximums: A dict mapping allocator names to AllocationMaximum."
1283,Timeline,tensorflow/tensorflow/python/client/timeline.py,346,class,A class for visualizing execution timelines of TensorFlow steps.
1284,analyze_step_stats,tensorflow/tensorflow/python/client/timeline.py,674,method,"Analyze the step stats and format it into Chrome Trace Format.
Args:
show_dataflow: (Optional.) If True, add flow events to the trace
connecting producers and consumers of tensors.
show_memory: (Optional.) If True, add object snapshot events to the trace
showing the sizes and lifetimes of tensors.
op_time: (Optional.) How the execution time of op is shown in timeline.
Possible values are ""schedule"", ""gpu"" and ""all"". ""schedule"" will show op
from the time it is scheduled to the end of the scheduling. Notice by
the end of its scheduling its async kernels may not start yet. It is
shown using the default value from step_stats. ""gpu"" will show op with
the execution time of its kernels on GPU. ""all"" will show op from the
start of its scheduling to the end of its last kernel.
Returns:
A 'StepStatsAnalysis' object."
1285,generate_chrome_trace_format,tensorflow/tensorflow/python/client/timeline.py,707,method,"Produces a trace in Chrome Trace Format.
Args:
show_dataflow: (Optional.) If True, add flow events to the trace
connecting producers and consumers of tensors.
show_memory: (Optional.) If True, add object snapshot events to the trace
showing the sizes and lifetimes of tensors.
op_time: (Optional.) How the execution time of op is shown in timeline.
Possible values are ""schedule"", ""gpu"" and ""all"".
""schedule"" will show op from the time it is scheduled to the end of
the scheduling.
Notice by the end of its scheduling its async kernels may not start
yet. It is shown using the default value from step_stats.
""gpu"" will show op with the execution time of its kernels on GPU.
""all"" will show op from the start of its scheduling to the end of
its last kernel.
Returns:
A JSON formatted string in Chrome Trace format."
1286,forward_compatible,tensorflow/tensorflow/python/compat/compat.py,70,function,"Return true if the forward compatibility window has expired.
See [Version
compatibility](https://tensorflow.org/guide/version_compat#backward_forward).
Forward-compatibility refers to scenarios where the producer of a TensorFlow
model (a GraphDef or SavedModel) is compiled against a version of the
TensorFlow library newer than what the consumer was compiled against. The
""producer"" is typically a Python program that constructs and trains a model
while the ""consumer"" is typically another program that loads and serves the
model.
TensorFlow has been supporting a 3 week forward-compatibility window for
programs compiled from source at HEAD.
For example, consider the case where a new operation `MyNewAwesomeAdd` is
created with the intent of replacing the implementation of an existing Python
wrapper - `tf.add`. The Python wrapper implementation should change from
something like:
```python
def add(inputs, name=None):
return gen_math_ops.add(inputs, name)
```
to:
```python
from tensorflow.python.compat import compat
def add(inputs, name=None):
if compat.forward_compatible(year, month, day):
# Can use the awesome new implementation.
return gen_math_ops.my_new_awesome_add(inputs, name)
# To maintain forward compatibility, use the old implementation.
return gen_math_ops.add(inputs, name)
```
Where `year`, `month`, and `day` specify the date beyond which binaries
that consume a model are expected to have been updated to include the
new operations. This date is typically at least 3 weeks beyond the date
the code that adds the new operation is committed.
Args:
year: A year (e.g., 2018). Must be an `int`.
month: A month (1 <= month <= 12) in year. Must be an `int`.
day: A day (1 <= day <= 31, or 30, or 29, or 28) in month. Must be an
`int`.
Returns:
True if the caller can expect that serialized TensorFlow graphs produced
can be consumed by programs that are compiled with the TensorFlow library
source code after (year, month, day)."
1287,forward_compatibility_horizon,tensorflow/tensorflow/python/compat/compat.py,131,function,"Context manager for testing forward compatibility of generated graphs.
See [Version
compatibility](https://tensorflow.org/guide/version_compat#backward_forward).
To ensure forward compatibility of generated graphs (see `forward_compatible`)
with older binaries, new features can be gated with:
```python
if compat.forward_compatible(year=2018, month=08, date=01):
generate_graph_with_new_features()
else:
generate_graph_so_older_binaries_can_consume_it()
```
However, when adding new features, one may want to unittest it before
the forward compatibility window expires. This context manager enables
such tests. For example:
```python
from tensorflow.python.compat import compat
def testMyNewFeature(self):
with compat.forward_compatibility_horizon(2018, 08, 02):
# Test that generate_graph_with_new_features() has an effect
```
Args:
year: A year (e.g., 2018). Must be an `int`.
month: A month (1 <= month <= 12) in year. Must be an `int`.
day: A day (1 <= day <= 31, or 30, or 29, or 28) in month. Must be an
`int`.
Yields:
Nothing."
1288,enable_v2_behavior,tensorflow/tensorflow/python/compat/v2_compat.py,43,function,"Enables TensorFlow 2.x behaviors.
This function can be called at the beginning of the program (before `Tensors`,
`Graphs` or other structures have been created, and before devices have been
initialized. It switches all global behaviors that are different between
TensorFlow 1.x and 2.x to behave as intended for 2.x.
This function is called in the main TensorFlow `__init__.py` file, user should
not need to call it, except during complex migrations."
1289,disable_v2_behavior,tensorflow/tensorflow/python/compat/v2_compat.py,82,function,"Disables TensorFlow 2.x behaviors.
This function can be called at the beginning of the program (before `Tensors`,
`Graphs` or other structures have been created, and before devices have been
initialized. It switches all global behaviors that are different between
TensorFlow 1.x and 2.x to behave as intended for 1.x.
User can call this function to disable 2.x behavior during complex migrations."
1290,convert_graph_def,tensorflow/tensorflow/python/compiler/mlir/mlir.py,26,function,"Import a GraphDef and convert it to a textual MLIR module.
Args:
graph_def: An object of type graph_pb2.GraphDef or a textual proto
representation of a valid GraphDef.
pass_pipeline: A textual description of an MLIR Pass Pipeline to run on the
module, see MLIR documentation for the
[textual pass pipeline syntax](https://github.com/tensorflow/mlir/blob/master/g3doc/WritingAPass.md#textual-pass-pipeline-specification).
Returns:
A textual representation of the MLIR module corresponding to the graphdef.
Raises a RuntimeError on error."
1291,TrtPrecisionMode,tensorflow/tensorflow/python/compiler/tensorrt/trt_convert.py,98,class,
1292,supported_precision_modes,tensorflow/tensorflow/python/compiler/tensorrt/trt_convert.py,104,method,
1293,TrtConversionParams,tensorflow/tensorflow/python/compiler/tensorrt/trt_convert.py,117,class,"Parameters that are used for TF-TRT conversion.
Fields:
rewriter_config_template: a template RewriterConfig proto used to create a
TRT-enabled RewriterConfig. If None, it will use a default one.
max_workspace_size_bytes: the maximum GPU temporary memory which the TRT
engine can use at execution time. This corresponds to the
'workspaceSize' parameter of nvinfer1::IBuilder::setMaxWorkspaceSize().
precision_mode: one the strings in
TrtPrecisionMode.supported_precision_modes().
minimum_segment_size: the minimum number of nodes required for a subgraph
to be replaced by TRTEngineOp.
is_dynamic_op: whether to generate dynamic TRT ops which will build the
TRT network and engine at run time. i.e. Since TensorRT version < 6.0
does not support dynamic dimensions other than the batch dimension, when
the TensorFlow graph has a non-batch dimension of dynamic size, we would
need to enable this option. This option should be set to True in TF 2.0.
maximum_cached_engines: max number of cached TRT engines for dynamic TRT
ops. Created TRT engines for a dynamic dimension are cached. This is the
maximum number of engines that can be cached. If the number of cached
engines is already at max but none of them supports the input shapes,
the TRTEngineOp will fall back to run the original TF subgraph that
corresponds to the TRTEngineOp.
use_calibration: this argument is ignored if precision_mode is not INT8.
If set to True, a calibration graph will be created to calibrate the
missing ranges. The calibration graph must be converted to an inference
graph by running calibration with calibrate(). If set to False,
quantization nodes will be expected for every tensor in the graph
(excluding those which will be fused). If a range is missing, an error
will occur. Please note that accuracy may be negatively affected if
there is a mismatch between which tensors TRT quantizes and which
tensors were trained with fake quantization.
max_batch_size: max size for the input batch. This parameter is only
effective when is_dynamic_op=False which is not supported in TF 2.0.
allow_build_at_runtime: whether to build TensorRT engines during runtime.
If no TensorRT engine can be found in cache that can handle the given
inputs during runtime, then a new TensorRT engine is built at runtime if
allow_build_at_runtime=True, and otherwise native TF is used. This
argument is only effective if is_dynamic_op=True."
1294,get_tensorrt_rewriter_config,tensorflow/tensorflow/python/compiler/tensorrt/trt_convert.py,292,function,"Returns a RewriterConfig proto for TRT transformation.
Args:
conversion_params: a TrtConversionParams instance.
is_v2: whether we're getting a RewriterConfig for TF 2.0.
disable_non_trt_optimizers: Turn off all default Grappler optimizers.
Returns:
A RewriterConfig proto which sets a TensorRTOptimizer to run Grappler.
Raises:
TypeError: if any of the parameters are of unexpected type.
ValueError: if any of the parameters are of unexpected value."
1295,is_explicit_batch_mode_enabled,tensorflow/tensorflow/python/compiler/tensorrt/trt_convert.py,387,function,Checks whether explicit batch is enabled by the rewriter config.
1296,TrtGraphConverter,tensorflow/tensorflow/python/compiler/tensorrt/trt_convert.py,398,class,"A converter for TF-TRT transformation for TF 1.x GraphDef/SavedModels.
To run the conversion without quantization calibration (e.g. for FP32/FP16
precision modes):
```python
converter = TrtGraphConverter(
input_saved_model_dir=""my_dir"",
precision_mode=TrtPrecisionMode.FP16)
converted_graph_def = converter.convert()
converter.save(output_saved_model_dir)
```
To run the conversion with quantization calibration:
```python
converter = TrtGraphConverter(
input_saved_model_dir=""my_dir"",
precision_mode=TrtPrecisionMode.INT8)
converter.convert()
# Run calibration 10 times.
converted_graph_def = converter.calibrate(
fetch_names=['output:0'],
num_runs=10,
feed_dict_fn=lambda: {'input:0': my_next_data()})
converter.save(output_saved_model_dir)
```"
1297,convert,tensorflow/tensorflow/python/compiler/tensorrt/trt_convert.py,643,method,"Run the TF-TRT conversion.
Returns:
The converted GraphDef for TF 1.x."
1298,calibrate,tensorflow/tensorflow/python/compiler/tensorrt/trt_convert.py,656,method,"Run the calibration and return the calibrated GraphDef.
Args:
fetch_names: a list of output tensor name to fetch during calibration.
num_runs: number of runs of the graph during calibration.
feed_dict_fn: a function that returns a dictionary mapping input names (as
strings) in the GraphDef to be calibrated to values (e.g. Python list,
numpy arrays, etc). One and only one of `feed_dict_fn` and
`input_map_fn` should be specified.
input_map_fn: a function that returns a dictionary mapping input names (as
strings) in the GraphDef to be calibrated to Tensor objects. The values
of the named input tensors in the GraphDef to be calibrated will be
re-mapped to the respective `Tensor` values during calibration. One and
only one of `feed_dict_fn` and `input_map_fn` should be specified.
Raises:
ValueError: if the input combination is invalid.
RuntimeError: if this method is called in eager mode.
Returns:
The GraphDef after the calibration."
1299,save,tensorflow/tensorflow/python/compiler/tensorrt/trt_convert.py,749,method,"Save the converted graph as a SavedModel.
Args:
output_saved_model_dir: construct a SavedModel using the converted
GraphDef and save it to the specified directory. This option only works
when the input graph is loaded from a SavedModel, i.e. when
input_saved_model_dir is specified and input_graph_def is None in
__init__().
Raises:
ValueError: if the input to the converter is a GraphDef instead of a
SavedModel."
1300,TrtGraphConverterV2,tensorflow/tensorflow/python/compiler/tensorrt/trt_convert.py,880,class,"An offline converter for TF-TRT transformation for TF 2.0 SavedModels.
Currently this is not available on Windows platform.
Note that in V2, is_dynamic_op=False is not supported, meaning TRT engines
will be built only when the corresponding TRTEngineOp is executed. But we
still provide a way to avoid the cost of building TRT engines during inference
(see more below).
There are several ways to run the conversion:
1. FP32/FP16 precision
```python
params = tf.experimental.tensorrt.ConversionParams(
precision_mode='FP16')
converter = tf.experimental.tensorrt.Converter(
input_saved_model_dir=""my_dir"", conversion_params=params)
converter.convert()
converter.save(output_saved_model_dir)
```
In this case, no TRT engines will be built or saved in the converted
SavedModel. But if input data is available during conversion, we can still
build and save the TRT engines to reduce the cost during inference (see
option 2 below).
2. FP32/FP16 precision with pre-built engines
```python
params = tf.experimental.tensorrt.ConversionParams(
precision_mode='FP16',
# Set this to a large enough number so it can cache all the engines.
maximum_cached_engines=16)
converter = tf.experimental.tensorrt.Converter(
input_saved_model_dir=""my_dir"", conversion_params=params)
converter.convert()
# Define a generator function that yields input data, and use it to execute
# the graph to build TRT engines.
# With TensorRT 5.1, different engines will be built (and saved later) for
# different input shapes to the TRTEngineOp.
def my_input_fn():
for _ in range(num_runs):
inp1, inp2 = ...
yield inp1, inp2
converter.build(input_fn=my_input_fn) # Generate corresponding TRT engines
converter.save(output_saved_model_dir) # Generated engines will be saved.
```
In this way, one engine will be built/saved for each unique input shapes of
the TRTEngineOp. This is good for applications that cannot afford building
engines during inference but have access to input data that is similar to
the one used in production (for example, that has the same input shapes).
Also, the generated TRT engines is platform dependent, so we need to run
`build()` in an environment that is similar to production (e.g. with
same type of GPU).
3. INT8 precision and calibration with pre-built engines
```python
params = tf.experimental.tensorrt.ConversionParams(
precision_mode='INT8',
# Currently only one INT8 engine is supported in this mode.
maximum_cached_engines=1,
use_calibration=True)
converter = tf.experimental.tensorrt.Converter(
input_saved_model_dir=""my_dir"", conversion_params=params)
# Define a generator function that yields input data, and run INT8
# calibration with the data. All input data should have the same shape.
# At the end of convert(), the calibration stats (e.g. range information)
# will be saved and can be used to generate more TRT engines with different
# shapes. Also, one TRT engine will be generated (with the same shape as
# the calibration data) for save later.
def my_calibration_input_fn():
for _ in range(num_runs):
inp1, inp2 = ...
yield inp1, inp2
converter.convert(calibration_input_fn=my_calibration_input_fn)
# (Optional) Generate more TRT engines offline (same as the previous
# option), to avoid the cost of generating them during inference.
def my_input_fn():
for _ in range(num_runs):
inp1, inp2 = ...
yield inp1, inp2
converter.build(input_fn=my_input_fn)
# Save the TRT engine and the engines.
converter.save(output_saved_model_dir)
```"
1301,convert,tensorflow/tensorflow/python/compiler/tensorrt/trt_convert.py,1060,method,"Convert the input SavedModel in 2.0 format.
Args:
calibration_input_fn: a generator function that yields input data as a
list or tuple, which will be used to execute the converted signature for
calibration. All the returned input data should have the same shape.
Example: `def input_fn(): yield input1, input2, input3`
Raises:
ValueError: if the input combination is invalid.
Returns:
The TF-TRT converted Function."
1302,build,tensorflow/tensorflow/python/compiler/tensorrt/trt_convert.py,1126,method,"Run inference with converted graph in order to build TensorRT engines.
Args:
input_fn: a generator function that yields input data as a list or tuple,
which will be used to execute the converted signature to generate TRT
engines. Example:
`def input_fn():
# Let's assume a network with 2 input tensors. We generate 3 sets
# of dummy input data:
input_shapes = [[(1, 16), (2, 16)], # 1st input list
[(2, 32), (4, 32)], # 2nd list of two tensors
[(4, 32), (8, 32)]] # 3rd input list
for shapes in input_shapes:
# return a list of input tensors
yield [np.zeros(x).astype(np.float32) for x in shapes]`
Raises:
NotImplementedError: build() is already called.
RuntimeError: the input_fx is None."
1303,save,tensorflow/tensorflow/python/compiler/tensorrt/trt_convert.py,1189,method,"Save the converted SavedModel.
Args:
output_saved_model_dir: directory to saved the converted SavedModel."
1304,create_inference_graph,tensorflow/tensorflow/python/compiler/tensorrt/trt_convert.py,1270,function,"Python wrapper for the TRT transformation.
Args:
input_graph_def: a GraphDef object containing a model to be transformed. If
set to None, the graph will be read from the SavedModel loaded from
input_saved_model_dir.
outputs: list of tensors or node names for the model outputs. Only used when
input_graph_def is not None.
max_batch_size: max size for the input batch.
max_workspace_size_bytes: the maximum GPU temporary memory which the TRT
engine can use at execution time. This corresponds to the 'workspaceSize'
parameter of nvinfer1::IBuilder::setMaxWorkspaceSize().
precision_mode: one of TrtPrecisionMode.supported_precision_modes().
minimum_segment_size: the minimum number of nodes required for a subgraph to
be replaced by TRTEngineOp.
is_dynamic_op: whether to generate dynamic TRT ops which will build the TRT
network and engine at run time.
maximum_cached_engines: max number of cached TRT engines in dynamic TRT ops.
If the number of cached engines is already at max but none of them can
serve the input, the TRTEngineOp will fall back to run the TF function
based on which the TRTEngineOp is created.
input_saved_model_dir: the directory to load the SavedModel which contains
the input graph to transforms. Used only when input_graph_def is None.
input_saved_model_tags: list of tags to load the SavedModel.
input_saved_model_signature_key: the key of the signature to optimize the
graph for.
output_saved_model_dir: if not None, construct a SavedModel using the
returned GraphDef and save it to the specified directory. This option only
works when the input graph is loaded from a SavedModel, i.e. when
input_saved_model_dir is specified and input_graph_def is None.
session_config: the ConfigProto used to create a Session. It's also used as
a template to create a TRT-enabled ConfigProto for conversion. If not
specified, a default ConfigProto will be used.
Returns:
A GraphDef transformed from input_graph_def (or the SavedModel graph def
loaded from input_saved_model_dir, if input_graph_def is not present), where
all TRT compatible subgraphs are replaced with TRTEngineOps, and a TF
function is added for each of the subgraphs.
If is_dynamic_op is True, each TRTEngineOp will contain a serialized
subgraph GraphDef, which will be converted to a TRT engine at execution time
and the TRT engine will be cached for future usage. A new TRT engine will be
created each time when none of the cached engines match the input shapes. If
it fails to execute the TRT engine or the number of cached engines reaches
maximum_cached_engines, the op will fall back to call the corresponding TF
function.
If is_dynamic_op is False, each TRTEngineOp will contain a serialized TRT
engine created from the corresponding subgraph. No more engines will be
created on the fly, and the op will fall back to call the corresponding TF
function when it fails to execute the engine.
Raises:
ValueError: if the combination of the parameters is invalid."
1305,TrtPrecisionMode,tensorflow/tensorflow/python/compiler/tensorrt/trt_convert_windows.py,31,class,
1306,TrtConversionParams,tensorflow/tensorflow/python/compiler/tensorrt/trt_convert_windows.py,43,class,"Parameters that are used for TF-TRT conversion.
Fields:
rewriter_config_template: a template RewriterConfig proto used to create a
TRT-enabled RewriterConfig. If None, it will use a default one.
max_workspace_size_bytes: the maximum GPU temporary memory which the TRT
engine can use at execution time. This corresponds to the
'workspaceSize' parameter of nvinfer1::IBuilder::setMaxWorkspaceSize().
precision_mode: one the strings in
TrtPrecisionMode.supported_precision_modes().
minimum_segment_size: the minimum number of nodes required for a subgraph
to be replaced by TRTEngineOp.
is_dynamic_op: whether to generate dynamic TRT ops which will build the
TRT network and engine at run time. i.e. Since TensorRT version < 6.0
does not support dynamic dimensions other than the batch dimension, when
the TensorFlow graph has a non-batch dimension of dynamic size, we would
need to enable this option. This option should be set to True in TF 2.0.
maximum_cached_engines: max number of cached TRT engines for dynamic TRT
ops. Created TRT engines for a dynamic dimension are cached. This is the
maximum number of engines that can be cached. If the number of cached
engines is already at max but none of them supports the input shapes,
the TRTEngineOp will fall back to run the original TF subgraph that
corresponds to the TRTEngineOp.
use_calibration: this argument is ignored if precision_mode is not INT8.
If set to True, a calibration graph will be created to calibrate the
missing ranges. The calibration graph must be converted to an inference
graph by running calibration with calibrate(). If set to False,
quantization nodes will be expected for every tensor in the graph
(exlcuding those which will be fused). If a range is missing, an error
will occur. Please note that accuracy may be negatively affected if
there is a mismatch between which tensors TRT quantizes and which
tensors were trained with fake quantization.
max_batch_size: max size for the input batch. This parameter is only
effective when is_dynamic_op=False which is not supported in TF 2.0."
1307,TrtConverterWindows,tensorflow/tensorflow/python/compiler/tensorrt/trt_convert_windows.py,97,class,"An offline converter for TF-TRT transformation for TF 2.0 SavedModels.
Currently this is not available on Windows platform."
1308,conv2d_layer,tensorflow/tensorflow/python/compiler/tensorrt/test/conv2d_test.py,32,function,
1309,div_round_up,tensorflow/tensorflow/python/compiler/tensorrt/test/conv2d_test.py,62,function,
1310,build_graph,tensorflow/tensorflow/python/compiler/tensorrt/test/conv2d_test.py,66,function,
1311,CalibrationInt32Support,tensorflow/tensorflow/python/compiler/tensorrt/test/int32_test.py,68,class,Test execution of calibration with int32 input
1312,GraphFn,tensorflow/tensorflow/python/compiler/tensorrt/test/int32_test.py,71,method,
1313,GetParams,tensorflow/tensorflow/python/compiler/tensorrt/test/int32_test.py,76,method,
1314,ExpectedEnginesToBuild,tensorflow/tensorflow/python/compiler/tensorrt/test/int32_test.py,87,method,
1315,IsQuantizationMode,tensorflow/tensorflow/python/compiler/tensorrt/test/tf_trt_integration_test_base.py,95,function,
1316,IsQuantizationWithCalibration,tensorflow/tensorflow/python/compiler/tensorrt/test/tf_trt_integration_test_base.py,99,function,
1317,GraphState,tensorflow/tensorflow/python/compiler/tensorrt/test/tf_trt_integration_test_base.py,103,class,
1318,GetGraph,tensorflow/tensorflow/python/compiler/tensorrt/test/testdata/gen_tftrt_model.py,49,function,Define graph.
1319,GenerateModelV2,tensorflow/tensorflow/python/compiler/tensorrt/test/testdata/gen_tftrt_model.py,59,function,Generate and convert a model using TFv2 API.
1320,GenerateModelV1,tensorflow/tensorflow/python/compiler/tensorrt/test/testdata/gen_tftrt_model.py,90,function,Generate and convert a model using TFv1 API.
1321,experimental_jit_scope,tensorflow/tensorflow/python/compiler/xla/jit.py,42,function,"Enable or disable JIT compilation of operators within the scope.
NOTE: This is an experimental feature.
The compilation is a hint and only supported on a best-effort basis.
Example usage:
```python
with tf.xla.experimental.jit_scope():
c = tf.matmul(a, b) # compiled
with tf.xla.experimental.jit_scope(compile_ops=False):
d = tf.matmul(a, c) # not compiled
with tf.xla.experimental.jit_scope(
compile_ops=lambda node_def: 'matmul' in node_def.op.lower()):
e = tf.matmul(a, b) + d # matmul is compiled, the addition is not.
```
Example of `separate_compiled_gradients`:
```python
# In the example below, the computations for f, g and h will all be compiled
# in separate scopes.
with tf.xla.experimental.jit_scope(
separate_compiled_gradients=True):
f = tf.matmul(a, b)
g = tf.gradients([f], [a, b], name='mygrads1')
h = tf.gradients([f], [a, b], name='mygrads2')
```
Args:
compile_ops: Whether to enable or disable compilation in the scope.
Either a Python bool, or a callable that accepts the parameter
`node_def` and returns a python bool.
separate_compiled_gradients: If true put each gradient subgraph into a
separate compilation scope. This gives fine-grained control over which
portions of the graph will be compiled as a single unit. Compiling
gradients separately may yield better performance for some graphs.
The scope is named based on the scope of the forward computation as well
as the name of the gradients. As a result, the gradients will be compiled
in a scope that is separate from both the forward computation, and from
other gradients.
Raises:
RuntimeError: if called when eager execution is enabled.
Yields:
The current scope, enabling or disabling compilation."
1322,enable_jit_nonstateful,tensorflow/tensorflow/python/compiler/xla/jit_test.py,39,function,
1323,compile,tensorflow/tensorflow/python/compiler/xla/xla.py,67,function,"Builds an operator that compiles and runs `computation` with XLA.
NOTE: In eager mode, `computation` will have `@tf.function` semantics.
Args:
computation: A Python function that builds a computation to apply to the
input. If the function takes n inputs, 'inputs' should be a list of n
tensors.
`computation` may return a list of operations and tensors. Tensors must
come before operations in the returned list. The return value of
`compile` is a list of tensors corresponding to the tensors from the
output of `computation`.
All `Operation`s returned from `computation` will be executed when
evaluating any of the returned output tensors.
inputs: A list of inputs or `None` (equivalent to an empty list). Each input
can be a nested structure containing values that are convertible to
tensors. Note that passing an N-dimension list of compatible values will
result in a N-dimension list of scalar tensors rather than a single Rank-N
tensors. If you need different behavior, convert part of inputs to tensors
with `tf.convert_to_tensor`.
Returns:
Same data structure as if computation(*inputs) is called directly with some
exceptions for correctness. Exceptions include:
1) None output: a NoOp would be returned which control-depends on
computation.
2) Single value output: A tuple containing the value would be returned.
3) Operation-only outputs: a NoOp would be returned which
control-depends on computation.
TODO(b/121383831): Investigate into removing these special cases.
Raises:
RuntimeError: if called when eager execution is enabled.
Known issues:
When a tf.random operation is built with XLA, the implementation doesn't
pass the user provided seed to the XLA compiler. As such, the XLA compiler
generates a random number and uses it as a seed when compiling the
operation. This implementation causes a violation of the Tensorflow
defined semantics in two aspects. First, changing the value of the user
defined seed doesn't change the numbers generated by the operation.
Second, when a seed is not specified, running the program multiple times
will generate the same numbers."
1324,XLACompileContext,tensorflow/tensorflow/python/compiler/xla/xla.py,125,class,"A `ControlFlowContext` for nodes inside an XLA computation cluster.
THIS IS ONLY FOR TENSORFLOW INTERNAL IMPLEMENTATION, DO NO USE DIRECTLY.
The primary role of `XLACompileContext` is to mark operators inside a
xla.compile() computation with attribute ""_xla_compile_id=XYZ"", where XYZ is
a unique name.
`ControlFlowContext` is used to perform the annotation since it integrates
with Tensorflow constructs like ResourceVariables. For example, if a
`ResourceVariable` is constructed inside a xla.compile() block, the
`ResourceVariable` implementation can use
`with ops.control_dependencies(None)` to build the variable's definition
outside the compiled computation."
1325,report_unsupported_operations,tensorflow/tensorflow/python/compiler/xla/xla.py,159,method,
1326,AddOp,tensorflow/tensorflow/python/compiler/xla/xla.py,195,method,Create op in XLACompileContext and notifies outer context recursively.
1327,AddValue,tensorflow/tensorflow/python/compiler/xla/xla.py,268,method,Add `val` to the current context and its outer context recursively.
1328,AddInnerOp,tensorflow/tensorflow/python/compiler/xla/xla.py,285,method,
1329,grad_state,tensorflow/tensorflow/python/compiler/xla/xla.py,291,method,
1330,back_prop,tensorflow/tensorflow/python/compiler/xla/xla.py,299,method,"Forwards to the enclosing while context, if any."
1331,is_flat,tensorflow/tensorflow/python/compiler/xla/xla.py,409,function,"Checks if outputs is a flat structure.
Following structures and values are considered flat:
1) None
2) A single object
3) A list or tuple of Tensors/Operations
The only structures that this function understands are sequences,
dictionaries and types defined using the attrs library. E.g. this means
that if outputs contains a single user-defined Object, it is considered to
be flat. Errors are raised later on if that Object cannot be converted to a
Tensor.
Args:
outputs: Output from `computation` inside `xla.compile`.
Returns:
A boolean indicates whether outputs is flat."
1332,check_function_argument_count,tensorflow/tensorflow/python/compiler/xla/xla.py,591,function,"Validate the number of input arguments to an XLA function.
Args:
func: the Python function that will be called to generate the body of an XLA
computation graph.
input_arity: the number of explicit arguments supplied by the caller.
infeed_queue: if not None, the infeed queue that will supply
additional arguments to the function.
Returns:
None if function can be called with the supplied number of
arguments, or an error string if it cannot."
1333,BatchBenchmark,tensorflow/tensorflow/python/data/benchmarks/batch_benchmark.py,27,class,Benchmarks for `tf.data.Dataset.batch()`.
1334,benchmark_batch_sparse,tensorflow/tensorflow/python/data/benchmarks/batch_benchmark.py,30,method,
1335,benchmark_batch_dense,tensorflow/tensorflow/python/data/benchmarks/batch_benchmark.py,51,method,
1336,DatasetBenchmarkBase,tensorflow/tensorflow/python/data/benchmarks/benchmark_base.py,31,class,Base class for dataset benchmarks.
1337,run_benchmark,tensorflow/tensorflow/python/data/benchmarks/benchmark_base.py,34,method,"Benchmarks the dataset.
Runs the dataset `iters` times. In each iteration, the benchmark measures
the time it takes to go through `num_elements` elements of the dataset.
Args:
dataset: Dataset to benchmark.
num_elements: Number of dataset elements to iterate through each benchmark
iteration.
iters: Number of times to repeat the timing.
warmup: If true, warms up the session caches by running an untimed run.
apply_default_optimizations: Determines whether default optimizations
should be applied.
Returns:
A float, representing the per-element wall time of the dataset in seconds.
This is the median time (with respect to `iters`) it takes for the dataset
to go through `num_elements` elements, divided by `num_elements.`"
1338,run_and_report_benchmark,tensorflow/tensorflow/python/data/benchmarks/benchmark_base.py,87,method,
1339,FilterBenchmark,tensorflow/tensorflow/python/data/benchmarks/filter_benchmark.py,26,class,Benchmarks for `tf.data.Dataset.filter()`.
1340,benchmark_simple_function,tensorflow/tensorflow/python/data/benchmarks/filter_benchmark.py,34,method,
1341,benchmark_return_component_optimization,tensorflow/tensorflow/python/data/benchmarks/filter_benchmark.py,37,method,
1342,SingleThreadedFlatMapDataset,tensorflow/tensorflow/python/data/benchmarks/from_tensor_slices_benchmark.py,30,class,A `Dataset` that maps a function over its input and flattens the result.
1343,element_spec,tensorflow/tensorflow/python/data/benchmarks/from_tensor_slices_benchmark.py,54,method,
1344,FromTensorSlicesBenchmark,tensorflow/tensorflow/python/data/benchmarks/from_tensor_slices_benchmark.py,62,class,Benchmarks for `tf.data.Dataset.from_tensor_slices()`.
1345,benchmark_slice_repeat_batch,tensorflow/tensorflow/python/data/benchmarks/from_tensor_slices_benchmark.py,65,method,
1346,benchmark_reshape_slice_repeat,tensorflow/tensorflow/python/data/benchmarks/from_tensor_slices_benchmark.py,82,method,
1347,benchmark_slice_repeat_sparse,tensorflow/tensorflow/python/data/benchmarks/from_tensor_slices_benchmark.py,101,method,
1348,benchmark_slice_batch_cache_repeat,tensorflow/tensorflow/python/data/benchmarks/from_tensor_slices_benchmark.py,132,method,
1349,make_dataset,tensorflow/tensorflow/python/data/benchmarks/from_tensor_slices_benchmark.py,116,method,
1350,ListFilesBenchmark,tensorflow/tensorflow/python/data/benchmarks/list_files_benchmark.py,35,class,Benchmarks for `tf.data.Dataset.list_files()`.
1351,benchmark_nested_directories,tensorflow/tensorflow/python/data/benchmarks/list_files_benchmark.py,38,method,
1352,MapBenchmark,tensorflow/tensorflow/python/data/benchmarks/map_benchmark.py,32,class,Benchmarks for `tf.data.Dataset.map()`.
1353,benchmark_chain_of_maps,tensorflow/tensorflow/python/data/benchmarks/map_benchmark.py,35,method,
1354,benchmark_map_fan_out,tensorflow/tensorflow/python/data/benchmarks/map_benchmark.py,53,method,
1355,benchmark_stats,tensorflow/tensorflow/python/data/benchmarks/map_benchmark.py,72,method,
1356,benchmark_sequential_control_flow,tensorflow/tensorflow/python/data/benchmarks/map_benchmark.py,85,method,
1357,benchmark_parallel_control_flow,tensorflow/tensorflow/python/data/benchmarks/map_benchmark.py,103,method,
1358,benchmark_helper,tensorflow/tensorflow/python/data/benchmarks/map_benchmark.py,37,method,
1359,benchmark_helper,tensorflow/tensorflow/python/data/benchmarks/map_benchmark.py,56,method,
1360,fn,tensorflow/tensorflow/python/data/benchmarks/map_benchmark.py,88,method,
1361,fn,tensorflow/tensorflow/python/data/benchmarks/map_benchmark.py,107,method,
1362,body,tensorflow/tensorflow/python/data/benchmarks/map_benchmark.py,91,method,
1363,MetaBenchmark,tensorflow/tensorflow/python/data/benchmarks/meta_benchmark.py,31,class,Benchmark that compares various ways of running tf.data benchmarks.
1364,setup_fast_dataset,tensorflow/tensorflow/python/data/benchmarks/meta_benchmark.py,37,method,
1365,benchmark_fast_dataset_with_only_cpp_iterations,tensorflow/tensorflow/python/data/benchmarks/meta_benchmark.py,44,method,
1366,benchmark_fast_dataset_with_session_run,tensorflow/tensorflow/python/data/benchmarks/meta_benchmark.py,48,method,
1367,benchmark_fast_dataset_with_session_callable,tensorflow/tensorflow/python/data/benchmarks/meta_benchmark.py,52,method,
1368,benchmark_fast_dataset_in_eager,tensorflow/tensorflow/python/data/benchmarks/meta_benchmark.py,56,method,
1369,setup_slow_dataset,tensorflow/tensorflow/python/data/benchmarks/meta_benchmark.py,61,method,
1370,benchmark_slow_dataset_with_only_cpp_iterations,tensorflow/tensorflow/python/data/benchmarks/meta_benchmark.py,67,method,
1371,benchmark_slow_dataset_with_session_run,tensorflow/tensorflow/python/data/benchmarks/meta_benchmark.py,71,method,
1372,benchmark_slow_dataset_with_session_callable,tensorflow/tensorflow/python/data/benchmarks/meta_benchmark.py,75,method,
1373,benchmark_slow_dataset_in_eager,tensorflow/tensorflow/python/data/benchmarks/meta_benchmark.py,79,method,
1374,report,tensorflow/tensorflow/python/data/benchmarks/meta_benchmark.py,84,method,
1375,run_benchmark_in_eager,tensorflow/tensorflow/python/data/benchmarks/meta_benchmark.py,105,method,
1376,run_benchmark_with_session_run,tensorflow/tensorflow/python/data/benchmarks/meta_benchmark.py,113,method,
1377,run_benchmark_with_only_cpp_iterations,tensorflow/tensorflow/python/data/benchmarks/meta_benchmark.py,132,method,Benchmarks the dataset with the iterations performed in C++.
1378,PrefetchBenchmark,tensorflow/tensorflow/python/data/benchmarks/prefetch_benchmark.py,24,class,Benchmarks for `tf.data.Dataset.prefetch()`.
1379,benchmark_prefetch,tensorflow/tensorflow/python/data/benchmarks/prefetch_benchmark.py,27,method,
1380,RangeBenchmark,tensorflow/tensorflow/python/data/benchmarks/range_benchmark.py,24,class,Benchmarks for `tf.data.Dataset.range()`.
1381,benchmark_range,tensorflow/tensorflow/python/data/benchmarks/range_benchmark.py,27,method,
1382,AutotuneBenchmark,tensorflow/tensorflow/python/data/experimental/benchmarks/autotune_benchmark.py,31,class,Benchmarks for autotuning performance knobs.
1383,benchmark_map,tensorflow/tensorflow/python/data/experimental/benchmarks/autotune_benchmark.py,66,method,
1384,benchmark_map_and_batch,tensorflow/tensorflow/python/data/experimental/benchmarks/autotune_benchmark.py,87,method,
1385,benchmark_interleave,tensorflow/tensorflow/python/data/experimental/benchmarks/autotune_benchmark.py,110,method,
1386,benchmark_map_and_interleave,tensorflow/tensorflow/python/data/experimental/benchmarks/autotune_benchmark.py,134,method,
1387,benchmark_map_batch_and_interleave,tensorflow/tensorflow/python/data/experimental/benchmarks/autotune_benchmark.py,182,method,
1388,f1,tensorflow/tensorflow/python/data/experimental/benchmarks/autotune_benchmark.py,152,method,
1389,f2,tensorflow/tensorflow/python/data/experimental/benchmarks/autotune_benchmark.py,155,method,
1390,CsvDatasetBenchmark,tensorflow/tensorflow/python/data/experimental/benchmarks/csv_dataset_benchmark.py,38,class,Benchmarks for `tf.data.experimental.CsvDataset`.
1391,benchmark_map_with_floats,tensorflow/tensorflow/python/data/experimental/benchmarks/csv_dataset_benchmark.py,90,method,
1392,benchmark_map_with_strings,tensorflow/tensorflow/python/data/experimental/benchmarks/csv_dataset_benchmark.py,100,method,
1393,benchmark_csv_dataset_with_floats,tensorflow/tensorflow/python/data/experimental/benchmarks/csv_dataset_benchmark.py,110,method,
1394,benchmark_csv_dataset_with_strings,tensorflow/tensorflow/python/data/experimental/benchmarks/csv_dataset_benchmark.py,120,method,
1395,MapAndBatchBenchmark,tensorflow/tensorflow/python/data/experimental/benchmarks/map_and_batch_benchmark.py,40,class,Benchmarks for `tf.data.experimental.map_and_batch()`.
1396,benchmark_map_and_batch,tensorflow/tensorflow/python/data/experimental/benchmarks/map_and_batch_benchmark.py,43,method,Measures the performance of parallelized batching.
1397,benchmark_map_and_batch_chaining_versus_fusing,tensorflow/tensorflow/python/data/experimental/benchmarks/map_and_batch_benchmark.py,97,method,"Compares the performance of chaining and fusing map and batch.
NOTE: It is recommended to build the benchmark with
`-c opt --copt=-mavx --copt=-mavx2 --copt=-mfma --copt=-gmlt`
and execute it on a machine with at least 32 CPU cores."
1398,name,tensorflow/tensorflow/python/data/experimental/benchmarks/map_and_batch_benchmark.py,116,method,
1399,benchmark,tensorflow/tensorflow/python/data/experimental/benchmarks/map_and_batch_benchmark.py,126,method,Runs benchmark the given series.
1400,make_dataset,tensorflow/tensorflow/python/data/experimental/benchmarks/map_and_batch_benchmark.py,129,method,
1401,MapDefunBenchmark,tensorflow/tensorflow/python/data/experimental/benchmarks/map_defun_benchmark.py,34,class,Benchmarks for MapDefunOp.
1402,benchmark_defun_vs_map_fn,tensorflow/tensorflow/python/data/experimental/benchmarks/map_defun_benchmark.py,52,method,Benchmarks to compare the performance of MapDefun vs tf.map_fn.
1403,defun,tensorflow/tensorflow/python/data/experimental/benchmarks/map_defun_benchmark.py,56,method,
1404,fn,tensorflow/tensorflow/python/data/experimental/benchmarks/map_defun_benchmark.py,59,method,
1405,MapVectorizationBenchmark,tensorflow/tensorflow/python/data/experimental/benchmarks/map_vectorization_benchmark.py,97,class,Benchmarks for the `MapVectorization` optimization.
1406,benchmark_identity,tensorflow/tensorflow/python/data/experimental/benchmarks/map_vectorization_benchmark.py,147,method,
1407,benchmark_add_const,tensorflow/tensorflow/python/data/experimental/benchmarks/map_vectorization_benchmark.py,151,method,
1408,benchmark_return_const,tensorflow/tensorflow/python/data/experimental/benchmarks/map_vectorization_benchmark.py,154,method,
1409,benchmark_select,tensorflow/tensorflow/python/data/experimental/benchmarks/map_vectorization_benchmark.py,157,method,
1410,benchmark_cast,tensorflow/tensorflow/python/data/experimental/benchmarks/map_vectorization_benchmark.py,160,method,
1411,benchmark_reshape,tensorflow/tensorflow/python/data/experimental/benchmarks/map_vectorization_benchmark.py,164,method,
1412,benchmark_decode_csv,tensorflow/tensorflow/python/data/experimental/benchmarks/map_vectorization_benchmark.py,168,method,
1413,benchmark_parse_single_example,tensorflow/tensorflow/python/data/experimental/benchmarks/map_vectorization_benchmark.py,172,method,
1414,MatchingFilesBenchmark,tensorflow/tensorflow/python/data/experimental/benchmarks/matching_files_benchmark.py,35,class,Benchmark for the experimental `MatchingFilesDataset`.
1415,benchmark_nested_directories,tensorflow/tensorflow/python/data/experimental/benchmarks/matching_files_benchmark.py,38,method,
1416,OptimizationBenchmark,tensorflow/tensorflow/python/data/experimental/benchmarks/optimize_benchmark.py,32,class,Benchmarks for static optimizations.
1417,benchmark_map_fusion,tensorflow/tensorflow/python/data/experimental/benchmarks/optimize_benchmark.py,35,method,Evaluates performance map of fusion.
1418,benchmark_map_and_filter_fusion,tensorflow/tensorflow/python/data/experimental/benchmarks/optimize_benchmark.py,76,method,Evaluates performance map of fusion.
1419,benchmark_filter_fusion,tensorflow/tensorflow/python/data/experimental/benchmarks/optimize_benchmark.py,119,method,
1420,ParallelInterleaveBenchmark,tensorflow/tensorflow/python/data/experimental/benchmarks/parallel_interleave_benchmark.py,68,class,Benchmarks for `tf.data.experimental.parallel_interleave()`.
1421,apply_interleave,tensorflow/tensorflow/python/data/experimental/benchmarks/parallel_interleave_benchmark.py,71,method,
1422,make_dataset,tensorflow/tensorflow/python/data/experimental/benchmarks/parallel_interleave_benchmark.py,89,method,
1423,benchmark_remote_file_simulation,tensorflow/tensorflow/python/data/experimental/benchmarks/parallel_interleave_benchmark.py,126,method,
1424,benchmark_fast_input,tensorflow/tensorflow/python/data/experimental/benchmarks/parallel_interleave_benchmark.py,135,method,
1425,benchmark_single_cycle,tensorflow/tensorflow/python/data/experimental/benchmarks/parallel_interleave_benchmark.py,142,method,
1426,benchmark_single_parallel_call,tensorflow/tensorflow/python/data/experimental/benchmarks/parallel_interleave_benchmark.py,152,method,
1427,benchmark_long_cycle,tensorflow/tensorflow/python/data/experimental/benchmarks/parallel_interleave_benchmark.py,159,method,
1428,benchmark_stats,tensorflow/tensorflow/python/data/experimental/benchmarks/parallel_interleave_benchmark.py,167,method,
1429,RejectionResampleBenchmark,tensorflow/tensorflow/python/data/experimental/benchmarks/rejection_resample_benchmark.py,56,class,Benchmarks for `tf.data.experimental.rejection_resample()`.
1430,benchmark_resample_performance,tensorflow/tensorflow/python/data/experimental/benchmarks/rejection_resample_benchmark.py,59,method,
1431,SnapshotDatasetBenchmark,tensorflow/tensorflow/python/data/experimental/benchmarks/snapshot_dataset_benchmark.py,34,class,Benchmarks for `tf.data.experimental.snapshot()`.
1432,benchmarkWriteSnapshotGzipCompression,tensorflow/tensorflow/python/data/experimental/benchmarks/snapshot_dataset_benchmark.py,68,method,
1433,benchmarkWriteSnapshotSnappyCompression,tensorflow/tensorflow/python/data/experimental/benchmarks/snapshot_dataset_benchmark.py,76,method,
1434,benchmarkWriteSnapshotSimple,tensorflow/tensorflow/python/data/experimental/benchmarks/snapshot_dataset_benchmark.py,84,method,
1435,benchmarkPassthroughSnapshotSimple,tensorflow/tensorflow/python/data/experimental/benchmarks/snapshot_dataset_benchmark.py,94,method,
1436,benchmarkReadSnapshotSimple,tensorflow/tensorflow/python/data/experimental/benchmarks/snapshot_dataset_benchmark.py,104,method,
1437,benchmarkReadSnapshotGzipCompression,tensorflow/tensorflow/python/data/experimental/benchmarks/snapshot_dataset_benchmark.py,114,method,
1438,benchmarkReadSnapshotSnappyCompression,tensorflow/tensorflow/python/data/experimental/benchmarks/snapshot_dataset_benchmark.py,123,method,
1439,UnbatchBenchmark,tensorflow/tensorflow/python/data/experimental/benchmarks/unbatch_benchmark.py,32,class,Benchmarks for `tf.data.Dataset.unbatch()`.
1440,benchmark_native_unbatch,tensorflow/tensorflow/python/data/experimental/benchmarks/unbatch_benchmark.py,35,method,
1441,benchmark_old_unbatch_implementation,tensorflow/tensorflow/python/data/experimental/benchmarks/unbatch_benchmark.py,72,method,
1442,chunk,tensorflow/tensorflow/python/data/experimental/kernel_tests/auto_shard_dataset_test.py,46,function,
1443,remove_variants,tensorflow/tensorflow/python/data/experimental/kernel_tests/serialization/dataset_serialization_test_base.py,41,function,"Remove variants from a nest structure, so sess.run will execute."
1444,dense_to_ragged_batch,tensorflow/tensorflow/python/data/experimental/ops/batching.py,36,function,"A transformation that batches ragged elements into `tf.RaggedTensor`s.
This transformation combines multiple consecutive elements of the input
dataset into a single element.
Like `tf.data.Dataset.batch`, the components of the resulting element will
have an additional outer dimension, which will be `batch_size` (or
`N % batch_size` for the last element if `batch_size` does not divide the
number of input elements `N` evenly and `drop_remainder` is `False`). If
your program depends on the batches having the same outer dimension, you
should set the `drop_remainder` argument to `True` to prevent the smaller
batch from being produced.
Unlike `tf.data.Dataset.batch`, the input elements to be batched may have
different shapes:
* If an input element is a `tf.Tensor` whose static `tf.TensorShape` is
fully defined, then it is batched as normal.
* If an input element is a `tf.Tensor` whose static `tf.TensorShape` contains
one or more axes with unknown size (i.e., `shape[i]=None`), then the output
will contain a `tf.RaggedTensor` that is ragged up to any of such
dimensions.
* If an input element is a `tf.RaggedTensor` or any other type, then it is
batched as normal.
Example:
>>> dataset = tf.data.Dataset.from_tensor_slices(np.arange(6))
>>> dataset = dataset.map(lambda x: tf.range(x))
>>> dataset.element_spec.shape
TensorShape([None])
>>> dataset = dataset.apply(
... tf.data.experimental.dense_to_ragged_batch(batch_size=2))
>>> for batch in dataset:
... print(batch)
<tf.RaggedTensor [[], [0]]>
<tf.RaggedTensor [[0, 1], [0, 1, 2]]>
<tf.RaggedTensor [[0, 1, 2, 3], [0, 1, 2, 3, 4]]>
Args:
batch_size: A `tf.int64` scalar `tf.Tensor`, representing the number of
consecutive elements of this dataset to combine in a single batch.
drop_remainder: (Optional.) A `tf.bool` scalar `tf.Tensor`, representing
whether the last batch should be dropped in the case it has fewer than
`batch_size` elements; the default behavior is not to drop the smaller
batch.
row_splits_dtype: The dtype that should be used for the `row_splits` of any
new ragged tensors. Existing `tf.RaggedTensor` elements do not have their
row_splits dtype changed.
Returns:
Dataset: A `Dataset`."
1445,dense_to_sparse_batch,tensorflow/tensorflow/python/data/experimental/ops/batching.py,102,function,"A transformation that batches ragged elements into `tf.sparse.SparseTensor`s.
Like `Dataset.padded_batch()`, this transformation combines multiple
consecutive elements of the dataset, which might have different
shapes, into a single element. The resulting element has three
components (`indices`, `values`, and `dense_shape`), which
comprise a `tf.sparse.SparseTensor` that represents the same data. The
`row_shape` represents the dense shape of each row in the
resulting `tf.sparse.SparseTensor`, to which the effective batch size is
prepended. For example:
```python
# NOTE: The following examples use `{ ... }` to represent the
# contents of a dataset.
a = { ['a', 'b', 'c'], ['a', 'b'], ['a', 'b', 'c', 'd'] }
a.apply(tf.data.experimental.dense_to_sparse_batch(
batch_size=2, row_shape=[6])) ==
{
([[0, 0], [0, 1], [0, 2], [1, 0], [1, 1]], # indices
['a', 'b', 'c', 'a', 'b'], # values
[2, 6]), # dense_shape
([[0, 0], [0, 1], [0, 2], [0, 3]],
['a', 'b', 'c', 'd'],
[1, 6])
}
```
Args:
batch_size: A `tf.int64` scalar `tf.Tensor`, representing the number of
consecutive elements of this dataset to combine in a single batch.
row_shape: A `tf.TensorShape` or `tf.int64` vector tensor-like object
representing the equivalent dense shape of a row in the resulting
`tf.sparse.SparseTensor`. Each element of this dataset must have the same
rank as `row_shape`, and must have size less than or equal to `row_shape`
in each dimension.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1446,map_and_batch_with_legacy_function,tensorflow/tensorflow/python/data/experimental/ops/batching.py,153,function,"Fused implementation of `map` and `batch`.
NOTE: This is an escape hatch for existing uses of `map_and_batch` that do not
work with V2 functions. New uses are strongly discouraged and existing uses
should migrate to `map_and_batch` as this method will not be removed in V2.
Args:
map_func: A function mapping a nested structure of tensors to another
nested structure of tensors.
batch_size: A `tf.int64` scalar `tf.Tensor`, representing the number of
consecutive elements of this dataset to combine in a single batch.
num_parallel_batches: (Optional.) A `tf.int64` scalar `tf.Tensor`,
representing the number of batches to create in parallel. On one hand,
higher values can help mitigate the effect of stragglers. On the other
hand, higher values can increase contention if CPU is scarce.
drop_remainder: (Optional.) A `tf.bool` scalar `tf.Tensor`, representing
whether the last batch should be dropped in case its size is smaller than
desired; the default behavior is not to drop the smaller batch.
num_parallel_calls: (Optional.) A `tf.int32` scalar `tf.Tensor`,
representing the number of elements to process in parallel. If not
specified, `batch_size * num_parallel_batches` elements will be processed
in parallel. If the value `tf.data.experimental.AUTOTUNE` is used, then
the number of parallel calls is set dynamically based on available CPU.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`.
Raises:
ValueError: If both `num_parallel_batches` and `num_parallel_calls` are
specified."
1447,map_and_batch,tensorflow/tensorflow/python/data/experimental/ops/batching.py,213,function,"Fused implementation of `map` and `batch`.
Maps `map_func` across `batch_size` consecutive elements of this dataset
and then combines them into a batch. Functionally, it is equivalent to `map`
followed by `batch`. This API is temporary and deprecated since input pipeline
optimization now fuses consecutive `map` and `batch` operations automatically.
Args:
map_func: A function mapping a nested structure of tensors to another
nested structure of tensors.
batch_size: A `tf.int64` scalar `tf.Tensor`, representing the number of
consecutive elements of this dataset to combine in a single batch.
num_parallel_batches: (Optional.) A `tf.int64` scalar `tf.Tensor`,
representing the number of batches to create in parallel. On one hand,
higher values can help mitigate the effect of stragglers. On the other
hand, higher values can increase contention if CPU is scarce.
drop_remainder: (Optional.) A `tf.bool` scalar `tf.Tensor`, representing
whether the last batch should be dropped in case its size is smaller than
desired; the default behavior is not to drop the smaller batch.
num_parallel_calls: (Optional.) A `tf.int32` scalar `tf.Tensor`,
representing the number of elements to process in parallel. If not
specified, `batch_size * num_parallel_batches` elements will be processed
in parallel. If the value `tf.data.experimental.AUTOTUNE` is used, then
the number of parallel calls is set dynamically based on available CPU.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`.
Raises:
ValueError: If both `num_parallel_batches` and `num_parallel_calls` are
specified."
1448,unbatch,tensorflow/tensorflow/python/data/experimental/ops/batching.py,269,function,"Splits elements of a dataset into multiple elements on the batch dimension.
For example, if elements of the dataset are shaped `[B, a0, a1, ...]`,
where `B` may vary for each input element, then for each element in the
dataset, the unbatched dataset will contain `B` consecutive elements
of shape `[a0, a1, ...]`.
```python
# NOTE: The following example uses `{ ... }` to represent the contents
# of a dataset.
a = { ['a', 'b', 'c'], ['a', 'b'], ['a', 'b', 'c', 'd'] }
a.unbatch() == {
'a', 'b', 'c', 'a', 'b', 'a', 'b', 'c', 'd'}
```
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1449,cardinality,tensorflow/tensorflow/python/data/experimental/ops/cardinality.py,38,function,"Returns the cardinality of `dataset`, if known.
The operation returns the cardinality of `dataset`. The operation may return
`tf.data.experimental.INFINITE_CARDINALITY` if `dataset` contains an infinite
number of elements or `tf.data.experimental.UNKNOWN_CARDINALITY` if the
analysis fails to determine the number of elements in `dataset` (e.g. when the
dataset source is a file).
>>> dataset = tf.data.Dataset.range(42)
>>> print(tf.data.experimental.cardinality(dataset).numpy())
42
>>> dataset = dataset.repeat()
>>> cardinality = tf.data.experimental.cardinality(dataset)
>>> print((cardinality == tf.data.experimental.INFINITE_CARDINALITY).numpy())
True
>>> dataset = dataset.filter(lambda x: True)
>>> cardinality = tf.data.experimental.cardinality(dataset)
>>> print((cardinality == tf.data.experimental.UNKNOWN_CARDINALITY).numpy())
True
Args:
dataset: A `tf.data.Dataset` for which to determine cardinality.
Returns:
A scalar `tf.int64` `Tensor` representing the cardinality of `dataset`. If
the cardinality is infinite or unknown, the operation returns the named
constant `INFINITE_CARDINALITY` and `UNKNOWN_CARDINALITY` respectively."
1450,assert_cardinality,tensorflow/tensorflow/python/data/experimental/ops/cardinality.py,72,function,"Asserts the cardinality of the input dataset.
NOTE: The following assumes that ""examples.tfrecord"" contains 42 records.
>>> dataset = tf.data.TFRecordDataset(""examples.tfrecord"")
>>> cardinality = tf.data.experimental.cardinality(dataset)
>>> print((cardinality == tf.data.experimental.UNKNOWN_CARDINALITY).numpy())
True
>>> dataset = dataset.apply(tf.data.experimental.assert_cardinality(42))
>>> print(tf.data.experimental.cardinality(dataset).numpy())
42
Args:
expected_cardinality: The expected cardinality of the input dataset.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`.
Raises:
FailedPreconditionError: The assertion is checked at runtime (when iterating
the dataset) and an error is raised if the actual and expected cardinality
differ."
1451,compress,tensorflow/tensorflow/python/data/experimental/ops/compression_ops.py,24,function,"Compress a dataset element.
Args:
element: A nested structure of types supported by Tensorflow.
Returns:
A variant tensor representing the compressed element. This variant can be
passed to `uncompress` to get back the original element."
1452,uncompress,tensorflow/tensorflow/python/data/experimental/ops/compression_ops.py,39,function,"Uncompress a compressed dataset element.
Args:
element: A scalar variant tensor to uncompress. The element should have been
created by calling `compress`.
output_spec: A nested structure of `tf.TypeSpec` representing the type(s) of
the uncompressed element.
Returns:
The uncompressed element."
1453,CounterV2,tensorflow/tensorflow/python/data/experimental/ops/counter.py,29,function,"Creates a `Dataset` that counts from `start` in steps of size `step`.
For example:
```python
Dataset.count() == [0, 1, 2, ...)
Dataset.count(2) == [2, 3, ...)
Dataset.count(2, 5) == [2, 7, 12, ...)
Dataset.count(0, -1) == [0, -1, -2, ...)
Dataset.count(10, -1) == [10, 9, ...)
```
Args:
start: (Optional.) The starting value for the counter. Defaults to 0.
step: (Optional.) The step size for the counter. Defaults to 1.
dtype: (Optional.) The data type for counter elements. Defaults to
`tf.int64`.
Returns:
A `Dataset` of scalar `dtype` elements."
1454,CounterV1,tensorflow/tensorflow/python/data/experimental/ops/counter.py,59,function,
1455,ProcessingMode,tensorflow/tensorflow/python/data/experimental/ops/data_service_ops.py,36,class,
1456,validate,tensorflow/tensorflow/python/data/experimental/ops/data_service_ops.py,40,method,Raises a ValueError if the given object is not a valid processing mode.
1457,distribute,tensorflow/tensorflow/python/data/experimental/ops/data_service_ops.py,295,function,"A transformation that moves dataset processing to the tf.data service.
When you iterate over a dataset containing the `distribute` transformation,
the tf.data service creates a ""job"" which produces data for the dataset
iteration.
The `processing_mode` argument controls what data is produced by a tf.data
service job. Currently, the only supported mode is ""parallel_epochs"".
processing_mode=""parallel_epochs"" means that multiple tf.data workers will
iterate through the dataset in parallel, each producing all elements of the
dataset. For example, if the dataset contains {0, 1, 2}, every tf.data worker
used for execution will produce {0, 1, 2}. If there are 3 workers, the job
will produce the elements {0, 0, 0, 1, 1, 1, 2, 2, 2} (though not necessarily
in that order). To account for this, it is recommended to randomly shuffle
your dataset, so that different tf.data workers will iterate through the
dataset in different orders.
In the future, there will be additional processing modes. For example,
a ""one_epoch"" mode which partitions the dataset across the tf.data
workers, so that the consumers see each element of the dataset only once.
```
dataset = tf.data.Dataset.range(5)
dataset = dataset.map(lambda x: x*x)
dataset = dataset.apply(
tf.data.experimental.service.distribute(""parallel_epochs"",
""grpc://dataservice:5000""))
dataset = dataset.map(lambda x: x+1)
for element in dataset:
print(element) # prints { 1, 2, 5, 10, 17 }
```
In the above example, the first two lines (before the call to `distribute`)
will be executed on tf.data workers, and the elements provided over
RPC. The remaining transformations (after the call to `distribute`) will be
executed locally.
The `job_name` argument allows jobs to be shared across multiple
datasets. Instead of each dataset creating its own job, all
datasets with the same `job_name` will consume from the same job. A new job
will be created for each iteration of the dataset (with each repetition of
`Dataset.repeat` counting as a new iteration). Suppose two training workers
(in either a single client or multi-client setup) iterate over the below
dataset, and there is a single tf.data worker:
```
range5_dataset = tf.data.Dataset.range(5)
dataset = range5_dataset.apply(tf.data.experimental.service.distribute(
""parallel_epochs"", ""grpc://dataservice:5000"", job_name=""my_job_name""))
for iteration in range(3):
print(list(dataset))
```
The elements of each job will be split between the two processes, with
elements being consumed by the processes on a first-come first-served basis.
One possible result is that process 1 prints
```
[0, 2, 4]
[0, 1, 3]
[1]
```
and process 2 prints
```
[1, 3]
[2, 4]
[0, 2, 3, 4]
```
Job names must not be re-used across different training jobs within the
lifetime of the tf.data service. In general, the tf.data service is expected
to live for the duration of a single training job.
To use the tf.data service with multiple training jobs, make sure to use
different job names to avoid conflicts. For example, suppose a training job
calls `distribute` with `job_name=""job""` and reads until end of input. If
another independent job connects to the same tf.data service and tries to read
from `job_name=""job""`, it will immediately receive end of input, without
getting any data.
**Keras and Distribution Strategies**
The dataset produced by the `distribute` transformation can be passed to
Keras' `Model.fit` or Distribution Strategy's
`tf.distribute.Strategy.experimental_distribute_dataset` like any other
`tf.data.Dataset`. We recommend setting a `job_name` on the call to
`distribute` so that if there are multiple workers, they read data from the
same job. Note that the autosharding normally performed by
`experimental_distribute_dataset` will be disabled when setting a `job_name`,
since sharing the job already results in splitting data across the workers.
When using a shared job, data will be dynamically balanced across workers, so
that they reach end of input about the same time. This results in better
worker utilization than with autosharding, where each worker processes an
independent set of files, and some workers may run out of data earlier than
others.
Args:
processing_mode: A string specifying the policy for how data should be
processed by tf.data workers. Currently, the only supported value is
""parallel_epochs"".
service: A string indicating how to connect to the tf.data service. The
string should be in the format ""protocol://address"", e.g.
""grpc://localhost:5000"".
job_name: (Optional.) The name of the job. This argument makes it possible
for multiple datasets to share the same job. The default behavior is that
the dataset creates anonymous, exclusively owned jobs.
max_outstanding_requests: (Optional.) A limit on how many elements may be
requested at the same time. You can use this option to control the amount
of memory used, since `distribute` won't use more than `element_size` *
`max_outstanding_requests` of memory.
Returns:
Dataset: A `Dataset` of the elements produced by the data service."
1458,register_dataset,tensorflow/tensorflow/python/data/experimental/ops/data_service_ops.py,424,function,"Registers a dataset with the tf.data service.
`register_dataset` registers a dataset with the tf.data service so that
datasets can be created later with
`tf.data.experimental.service.from_dataset_id`. This is useful when the
dataset
is registered by one process, then used in another process. When the same
process is both registering and reading from the dataset, it is simpler to use
`tf.data.experimental.service.distribute` instead.
If the dataset is already registered with the tf.data service,
`register_dataset` returns the already-registered dataset's id.
>>> dispatcher = tf.data.experimental.service.DispatchServer(port=0)
>>> dispatcher_address = dispatcher.target.split(""://"")[1]
>>> worker = tf.data.experimental.service.WorkerServer(
... port=0, dispatcher_address=dispatcher_address)
>>> dataset = tf.data.Dataset.range(10)
>>> dataset_id = tf.data.experimental.service.register_dataset(
... dispatcher.target, dataset)
>>> dataset = tf.data.experimental.service.from_dataset_id(
... processing_mode=""parallel_epochs"",
... service=dispatcher.target,
... dataset_id=dataset_id,
... element_spec=dataset.element_spec)
>>> print(list(dataset.as_numpy_iterator()))
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
Args:
service: A string indicating how to connect to the tf.data service. The
string should be in the format ""protocol://address"", e.g.
""grpc://localhost:5000"".
dataset: A `tf.data.Dataset` to register with the tf.data service.
Returns:
A scalar int64 tensor of the registered dataset's id."
1459,from_dataset_id,tensorflow/tensorflow/python/data/experimental/ops/data_service_ops.py,491,function,"Creates a dataset which reads data from the tf.data service.
This is useful when the dataset is registered by one process, then used in
another process. When the same process is both registering and reading from
the dataset, it is simpler to use `tf.data.experimental.service.distribute`
instead.
Before using `from_dataset_id`, the dataset must have been registered with the
tf.data service using `tf.data.experimental.service.register_dataset`.
`register_dataset` returns a dataset id for the registered dataset. That is
the `dataset_id` which should be passed to `from_dataset_id`.
The `element_spec` argument indicates the `tf.TypeSpec`s for the elements
produced by the dataset. Currently `element_spec` must be explicitly
specified, and match the dataset registered under `dataset_id`. `element_spec`
defaults to `None` so that in the future we can support automatically
discovering the `element_spec` by querying the tf.data service.
`tf.data.experimental.service.distribute` is a convenience method which
combines `register_dataset` and `from_dataset_id` into a dataset
transformation.
See the documentation for `tf.data.experimental.service.distribute` for more
detail about how `from_dataset_id` works.
>>> dispatcher = tf.data.experimental.service.DispatchServer(port=0)
>>> dispatcher_address = dispatcher.target.split(""://"")[1]
>>> worker = tf.data.experimental.service.WorkerServer(
... port=0, dispatcher_address=dispatcher_address)
>>> dataset = tf.data.Dataset.range(10)
>>> dataset_id = tf.data.experimental.service.register_dataset(
... dispatcher.target, dataset)
>>> dataset = tf.data.experimental.service.from_dataset_id(
... processing_mode=""parallel_epochs"",
... service=dispatcher.target,
... dataset_id=dataset_id,
... element_spec=dataset.element_spec)
>>> print(list(dataset.as_numpy_iterator()))
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
Args:
processing_mode: A string specifying the policy for how data should be
processed by tf.data workers. Currently, the only supported value is
""parallel_epochs"".
service: A string indicating how to connect to the tf.data service. The
string should be in the format ""protocol://address"", e.g.
""grpc://localhost:5000"".
dataset_id: The id of the dataset to read from. This id is returned by
`register_dataset` when the dataset is registered with the tf.data
service.
element_spec: A nested structure of `tf.TypeSpec`s representing the type of
elements produced by the dataset. Use `tf.data.Dataset.element_spec` to
see the element spec for a given dataset.
job_name: (Optional.) The name of the job. This argument makes it possible
for multiple datasets to share the same job. The default behavior is that
the dataset creates anonymous, exclusively owned jobs.
max_outstanding_requests: (Optional.) A limit on how many elements may be
requested at the same time. You can use this option to control the amount
of memory used, since `distribute` won't use more than `element_size` *
`max_outstanding_requests` of memory.
Returns:
A `tf.data.Dataset` which reads from the tf.data service."
1460,replicate,tensorflow/tensorflow/python/data/experimental/ops/distribute.py,294,function,"A transformation that replicates `dataset` onto a list of devices.
Args:
dataset: A `tf.data.Dataset` object.
devices: A list of devices to replicate the dataset on.
Returns:
A dictionary mapping device name to a dataset on that device."
1461,batch_sizes_for_worker,tensorflow/tensorflow/python/data/experimental/ops/distribute.py,328,function,"Determines how to rebatch a dataset for the given worker.
Given the global batch size, number of workers, number of replicas per worker,
and worker index, returns the correct batch sizes for rebatching a dataset
on worker `worker_index` of `num_workers`, such that each global step (across
all workers and replicas) will consume global_batch_size elements. The
returned value should be passed as the `batch_sizes` input parameter to
`tf.data.experimental.rebatch()`. The returned batch sizes meet the following
constraints:
Let G = global_batch_size, W = num_workers, R = num_replicas_per_worker
(A) for any worker, len(batch_sizes) = W * R
(B) for any worker, sum(batch_sizes) == G
(C) for any global step (i.e. R iterations on each worker), the sum of batches
consumed by replicas across all workers is G.
(D) any two batch sizes of any two replicas differs by at most one.
For example, suppose we have G = 7, W = 2, R = 2, and suppose we have two
files which each contain 7 elements:
```python
# WORKER 0
batch_sizes_0 = batch_sizes_for_worker(global_batch_size=global_batch_size,
num_workers=2,
num_replicas_per_worker=2,
worker_index=0)
print(batch_sizes_0)
>> [2, 2, 2, 1]
dataset_0 = tf.data.Dataset.from_tensor_slices([""file_a"", ""file_b""])
dataset_0 = dataset_0.shard(num_shards, index=0)
dataset_0 = dataset_0.batch(7)
dataset_0 = dataset_0.apply(tf.data.experimental.rebatch(batch_sizes_0))
for elem in dataset_0:
print(elem)
>> [[A0, A1], [A2, A3], [A4, A5], [A6]]
# WORKER 1
batch_sizes_1 = batch_sizes_for_worker(global_batch_size=global_batch_size,
num_workers=2,
num_replicas_per_worker=2,
worker_index=1)
print(batch_sizes_1)
>> [2, 1, 2, 2]
dataset_1 = tf.data.Dataset.from_tensor_slices([""file_a"", ""file_b""])
dataset_1 = dataset_1.shard(num_shards, index=1)
dataset_1 = dataset_1.batch(7)
dataset_1 = dataset_1.apply(tf.data.experimental.rebatch(batch_sizes_1))
for elem in dataset_1:
print(elem)
>> [[B0, B1], [B2], [B3, B4], [B5, B6]]
```
The above example will produce the following elements:
Step 1:
Worker 0 Replica 0: [A0, A1]
Worker 0 Replica 1: [A2, A3]
Worker 1 Replica 0: [B0, B1]
Worker 1 Replica 1: [B2]
Total batch size = 7
Step 2:
Worker 0 Replica 0: [A4, A5]
Worker 0 Replica 1: [A6]
Worker 1 Replica 0: [B3, B4]
Worker 1 Replica 1: [B5, B6]
Total batch size = 7
Args:
global_batch_size: A `tf.int64` scalar, representing the global batch size.
num_workers: An integer representing the number of workers the dataset will
be distributed across.
num_replicas_per_worker: An integer representing the number of replicas per
worker. All workers are assumed to have the same number of replicas.
worker_index: An integer index of the worker to be rebatched.
Returns:
A `tf.int64` vector, representing the batch sizes to rebatch the dataset
into."
1462,compute_batch_size,tensorflow/tensorflow/python/data/experimental/ops/distribute.py,436,function,"An operation that returns the batch size of the dataset.
This op tries to infer the batch size statically by walking up the dataset
tree from the final dataset node and returning the batch size of the first
batching dataset (such as from .batch() and .padded_batch()) that it
encounters. This differs from using the `element_spec` of a dataset in that it
does not account for partial batches.
This operation may fail if it encounters contradictory batch sizes (for
example, if the dataset is created by zipping together two datasets with
different batch sizes), if there are no explicit batching transformations, or
if there are operations downstream from the batching transformation that may
modify its batch size. In these cases, it returns a -1.
Args:
dataset: A `tf.data.Dataset` object.
Returns:
A `tf.int64` Tensor representing the batch size of the dataset sans partial
batches. If this cannot be inferred statically, the value of this tensor
will be -1."
1463,AutoShardPolicy,tensorflow/tensorflow/python/data/experimental/ops/distribute_options.py,27,class,"Represents the type of auto-sharding we enable.
Please see the DistributeOptions.auto_shard_policy documentation for more
information on each type of autosharding."
1464,ExternalStatePolicy,tensorflow/tensorflow/python/data/experimental/ops/distribute_options.py,39,class,
1465,DistributeOptions,tensorflow/tensorflow/python/data/experimental/ops/distribute_options.py,46,class,"Represents options for distributed data processing.
You can set the distribution options of a dataset through the
`experimental_distribute` property of `tf.data.Options`; the property is
an instance of `tf.data.experimental.DistributeOptions`.
```python
options = tf.data.Options()
options.experimental_distribute.auto_shard_policy = AutoShardPolicy.OFF
dataset = dataset.with_options(options)
```"
1466,enumerate_dataset,tensorflow/tensorflow/python/data/experimental/ops/enumerate_ops.py,26,function,"A transformation that enumerates the elements of a dataset.
It is similar to python's `enumerate`.
For example:
```python
# NOTE: The following examples use `{ ... }` to represent the
# contents of a dataset.
a = { 1, 2, 3 }
b = { (7, 8), (9, 10) }
# The nested structure of the `datasets` argument determines the
# structure of elements in the resulting dataset.
a.apply(tf.data.experimental.enumerate_dataset(start=5))
=> { (5, 1), (6, 2), (7, 3) }
b.apply(tf.data.experimental.enumerate_dataset())
=> { (0, (7, 8)), (1, (9, 10)) }
```
Args:
start: A `tf.int64` scalar `tf.Tensor`, representing the start value for
enumeration.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1467,ignore_errors,tensorflow/tensorflow/python/data/experimental/ops/error_ops.py,26,function,"Creates a `Dataset` from another `Dataset` and silently ignores any errors.
Use this transformation to produce a dataset that contains the same elements
as the input, but silently drops any elements that caused an error. For
example:
```python
dataset = tf.data.Dataset.from_tensor_slices([1., 2., 0., 4.])
# Computing `tf.debugging.check_numerics(1. / 0.)` will raise an
InvalidArgumentError.
dataset = dataset.map(lambda x: tf.debugging.check_numerics(1. / x, ""error""))
# Using `ignore_errors()` will drop the element that causes an error.
dataset =
dataset.apply(tf.data.experimental.ignore_errors()) # ==> {1., 0.5, 0.2}
```
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1468,get_single_element,tensorflow/tensorflow/python/data/experimental/ops/get_single_element.py,27,function,"Returns the single element in `dataset` as a nested structure of tensors.
This function enables you to use a `tf.data.Dataset` in a stateless
""tensor-in tensor-out"" expression, without creating an iterator.
This can be useful when your preprocessing transformations are expressed
as a `Dataset`, and you want to use the transformation at serving time.
For example:
```python
def preprocessing_fn(input_str):
# ...
return image, label
input_batch = ... # input batch of BATCH_SIZE elements
dataset = (tf.data.Dataset.from_tensor_slices(input_batch)
.map(preprocessing_fn, num_parallel_calls=BATCH_SIZE)
.batch(BATCH_SIZE))
image_batch, label_batch = tf.data.experimental.get_single_element(dataset)
```
Args:
dataset: A `tf.data.Dataset` object containing a single element.
Returns:
A nested structure of `tf.Tensor` objects, corresponding to the single
element of `dataset`.
Raises:
TypeError: if `dataset` is not a `tf.data.Dataset` object.
InvalidArgumentError (at runtime): if `dataset` does not contain exactly
one element."
1469,group_by_reducer,tensorflow/tensorflow/python/data/experimental/ops/grouping.py,38,function,"A transformation that groups elements and performs a reduction.
This transformation maps element of a dataset to a key using `key_func` and
groups the elements by key. The `reducer` is used to process each group; its
`init_func` is used to initialize state for each group when it is created, the
`reduce_func` is used to update the state every time an element is mapped to
the matching group, and the `finalize_func` is used to map the final state to
an output value.
Args:
key_func: A function mapping a nested structure of tensors
(having shapes and types defined by `self.output_shapes` and
`self.output_types`) to a scalar `tf.int64` tensor.
reducer: An instance of `Reducer`, which captures the reduction logic using
the `init_func`, `reduce_func`, and `finalize_func` functions.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1470,group_by_window,tensorflow/tensorflow/python/data/experimental/ops/grouping.py,68,function,"A transformation that groups windows of elements by key and reduces them.
This transformation maps each consecutive element in a dataset to a key
using `key_func` and groups the elements by key. It then applies
`reduce_func` to at most `window_size_func(key)` elements matching the same
key. All except the final window for each key will contain
`window_size_func(key)` elements; the final window may be smaller.
You may provide either a constant `window_size` or a window size determined by
the key through `window_size_func`.
Args:
key_func: A function mapping a nested structure of tensors
(having shapes and types defined by `self.output_shapes` and
`self.output_types`) to a scalar `tf.int64` tensor.
reduce_func: A function mapping a key and a dataset of up to `window_size`
consecutive elements matching that key to another dataset.
window_size: A `tf.int64` scalar `tf.Tensor`, representing the number of
consecutive elements matching the same key to combine in a single
batch, which will be passed to `reduce_func`. Mutually exclusive with
`window_size_func`.
window_size_func: A function mapping a key to a `tf.int64` scalar
`tf.Tensor`, representing the number of consecutive elements matching
the same key to combine in a single batch, which will be passed to
`reduce_func`. Mutually exclusive with `window_size`.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`.
Raises:
ValueError: if neither or both of {`window_size`, `window_size_func`} are
passed."
1471,bucket_by_sequence_length,tensorflow/tensorflow/python/data/experimental/ops/grouping.py,128,function,"A transformation that buckets elements in a `Dataset` by length.
Elements of the `Dataset` are grouped together by length and then are padded
and batched.
This is useful for sequence tasks in which the elements have variable length.
Grouping together elements that have similar lengths reduces the total
fraction of padding in a batch which increases training step efficiency.
Args:
element_length_func: function from element in `Dataset` to `tf.int32`,
determines the length of the element, which will determine the bucket it
goes into.
bucket_boundaries: `list<int>`, upper length boundaries of the buckets.
bucket_batch_sizes: `list<int>`, batch size per bucket. Length should be
`len(bucket_boundaries) + 1`.
padded_shapes: Nested structure of `tf.TensorShape` to pass to
`tf.data.Dataset.padded_batch`. If not provided, will use
`dataset.output_shapes`, which will result in variable length dimensions
being padded out to the maximum length in each batch.
padding_values: Values to pad with, passed to
`tf.data.Dataset.padded_batch`. Defaults to padding with 0.
pad_to_bucket_boundary: bool, if `False`, will pad dimensions with unknown
size to maximum length in batch. If `True`, will pad dimensions with
unknown size to bucket boundary minus 1 (i.e., the maximum length in each
bucket), and caller must ensure that the source `Dataset` does not contain
any elements with length longer than `max(bucket_boundaries)`.
no_padding: `bool`, indicates whether to pad the batch features (features
need to be either of type `tf.sparse.SparseTensor` or of same shape).
drop_remainder: (Optional.) A `tf.bool` scalar `tf.Tensor`, representing
whether the last batch should be dropped in the case it has fewer than
`batch_size` elements; the default behavior is not to drop the smaller
batch.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`.
Raises:
ValueError: if `len(bucket_batch_sizes) != len(bucket_boundaries) + 1`."
1472,Reducer,tensorflow/tensorflow/python/data/experimental/ops/grouping.py,443,class,"A reducer is used for reducing a set of elements.
A reducer is represented as a tuple of the three functions:
1) initialization function: key => initial state
2) reduce function: (old state, input) => new state
3) finalization function: state => result"
1473,init_func,tensorflow/tensorflow/python/data/experimental/ops/grouping.py,458,method,
1474,reduce_func,tensorflow/tensorflow/python/data/experimental/ops/grouping.py,462,method,
1475,finalize_func,tensorflow/tensorflow/python/data/experimental/ops/grouping.py,466,method,
1476,parallel_interleave,tensorflow/tensorflow/python/data/experimental/ops/interleave_ops.py,43,function,"A parallel version of the `Dataset.interleave()` transformation.
`parallel_interleave()` maps `map_func` across its input to produce nested
datasets, and outputs their elements interleaved. Unlike
`tf.data.Dataset.interleave`, it gets elements from `cycle_length` nested
datasets in parallel, which increases the throughput, especially in the
presence of stragglers. Furthermore, the `sloppy` argument can be used to
improve performance, by relaxing the requirement that the outputs are produced
in a deterministic order, and allowing the implementation to skip over nested
datasets whose elements are not readily available when requested.
Example usage:
```python
# Preprocess 4 files concurrently.
filenames = tf.data.Dataset.list_files(""/path/to/data/train*.tfrecords"")
dataset = filenames.apply(
tf.data.experimental.parallel_interleave(
lambda filename: tf.data.TFRecordDataset(filename),
cycle_length=4))
```
WARNING: If `sloppy` is `True`, the order of produced elements is not
deterministic.
Args:
map_func: A function mapping a nested structure of tensors to a `Dataset`.
cycle_length: The number of input `Dataset`s to interleave from in parallel.
block_length: The number of consecutive elements to pull from an input
`Dataset` before advancing to the next input `Dataset`.
sloppy: A boolean controlling whether determinism should be traded for
performance by allowing elements to be produced out of order. If
`sloppy` is `None`, the `tf.data.Options.experimental_deterministic`
dataset option (`True` by default) is used to decide whether to enforce a
deterministic order.
buffer_output_elements: The number of elements each iterator being
interleaved should buffer (similar to the `.prefetch()` transformation for
each interleaved iterator).
prefetch_input_elements: The number of input elements to transform to
iterators before they are needed for interleaving.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1477,sample_from_datasets_v2,tensorflow/tensorflow/python/data/experimental/ops/interleave_ops.py,146,function,"Samples elements at random from the datasets in `datasets`.
Args:
datasets: A list of `tf.data.Dataset` objects with compatible structure.
weights: (Optional.) A list of `len(datasets)` floating-point values where
`weights[i]` represents the probability with which an element should be
sampled from `datasets[i]`, or a `tf.data.Dataset` object where each
element is such a list. Defaults to a uniform distribution across
`datasets`.
seed: (Optional.) A `tf.int64` scalar `tf.Tensor`, representing the
random seed that will be used to create the distribution. See
`tf.random.set_seed` for behavior.
Returns:
A dataset that interleaves elements from `datasets` at random, according to
`weights` if provided, otherwise with uniform probability.
Raises:
TypeError: If the `datasets` or `weights` arguments have the wrong type.
ValueError: If the `weights` argument is specified and does not match the
length of the `datasets` element."
1478,sample_from_datasets_v1,tensorflow/tensorflow/python/data/experimental/ops/interleave_ops.py,230,function,
1479,choose_from_datasets_v2,tensorflow/tensorflow/python/data/experimental/ops/interleave_ops.py,237,function,"Creates a dataset that deterministically chooses elements from `datasets`.
For example, given the following datasets:
```python
datasets = [tf.data.Dataset.from_tensors(""foo"").repeat(),
tf.data.Dataset.from_tensors(""bar"").repeat(),
tf.data.Dataset.from_tensors(""baz"").repeat()]
# Define a dataset containing `[0, 1, 2, 0, 1, 2, 0, 1, 2]`.
choice_dataset = tf.data.Dataset.range(3).repeat(3)
result = tf.data.experimental.choose_from_datasets(datasets, choice_dataset)
```
The elements of `result` will be:
```
""foo"", ""bar"", ""baz"", ""foo"", ""bar"", ""baz"", ""foo"", ""bar"", ""baz""
```
Args:
datasets: A list of `tf.data.Dataset` objects with compatible structure.
choice_dataset: A `tf.data.Dataset` of scalar `tf.int64` tensors between
`0` and `len(datasets) - 1`.
Returns:
A dataset that interleaves elements from `datasets` according to the values
of `choice_dataset`.
Raises:
TypeError: If the `datasets` or `choice_dataset` arguments have the wrong
type."
1480,choose_from_datasets_v1,tensorflow/tensorflow/python/data/experimental/ops/interleave_ops.py,280,function,
1481,save,tensorflow/tensorflow/python/data/experimental/ops/io.py,34,function,"Saves the content of the given dataset.
Example usage:
>>> import tempfile
>>> path = os.path.join(tempfile.gettempdir(), ""saved_data"")
>>> # Save a dataset
>>> dataset = tf.data.Dataset.range(2)
>>> tf.data.experimental.save(dataset, path)
>>> new_dataset = tf.data.experimental.load(path,
... tf.TensorSpec(shape=(), dtype=tf.int64))
>>> for elem in new_dataset:
... print(elem)
tf.Tensor(0, shape=(), dtype=int64)
tf.Tensor(1, shape=(), dtype=int64)
The saved dataset is saved in multiple file ""shards"". By default, the dataset
output is divided to shards in a round-robin fashion but custom sharding can
be specified via the `shard_func` function. For example, you can save the
dataset to using a single shard as follows:
```python
dataset = make_dataset()
def custom_shard_func(element):
return 0
dataset = tf.data.experimental.save(
path=""/path/to/data"", ..., shard_func=custom_shard_func)
```
NOTE: The directory layout and file format used for saving the dataset is
considered an implementation detail and may change. For this reason, datasets
saved through `tf.data.experimental.save` should only be consumed through
`tf.data.experimental.load`, which is guaranteed to be backwards compatible.
Args:
dataset: The dataset to save.
path: Required. A directory to use for saving the dataset.
compression: Optional. The algorithm to use to compress data when writing
it. Supported options are `GZIP` and `NONE`. Defaults to `NONE`.
shard_func: Optional. A function to control the mapping of dataset elements
to file shards. The function is expected to map elements of the input
dataset to int64 shard IDs. If present, the function will be traced and
executed as graph computation."
1482,load,tensorflow/tensorflow/python/data/experimental/ops/io.py,146,function,"Loads a previously saved dataset.
Example usage:
>>> import tempfile
>>> path = os.path.join(tempfile.gettempdir(), ""saved_data"")
>>> # Save a dataset
>>> dataset = tf.data.Dataset.range(2)
>>> tf.data.experimental.save(dataset, path)
>>> new_dataset = tf.data.experimental.load(path,
... tf.TensorSpec(shape=(), dtype=tf.int64))
>>> for elem in new_dataset:
... print(elem)
tf.Tensor(0, shape=(), dtype=int64)
tf.Tensor(1, shape=(), dtype=int64)
Note that to load a previously saved dataset, you need to specify
`element_spec` -- a type signature of the elements of the saved dataset, which
can be obtained via `tf.data.Dataset.element_spec`. This requirement exists so
that shape inference of the loaded dataset does not need to perform I/O.
If the default option of sharding the saved dataset was used, the element
order of the saved dataset will be preserved when loading it.
The `reader_func` argument can be used to specify a custom order in which
elements should be loaded from the individual shards. The `reader_func` is
expected to take a single argument -- a dataset of datasets, each containing
elements of one of the shards -- and return a dataset of elements. For
example, the order of shards can be shuffled when loading them as follows:
```python
def custom_reader_func(datasets):
datasets = datasets.shuffle(NUM_SHARDS)
return datasets.interleave(lambda x: x, num_parallel_calls=AUTOTUNE)
dataset = tf.data.experimental.load(
path=""/path/to/data"", ..., reader_func=custom_reader_func)
```
Args:
path: Required. A path pointing to a previously saved dataset.
element_spec: Required. A nested structure of `tf.TypeSpec` objects matching
the structure of an element of the saved dataset and specifying the type
of individual element components.
compression: Optional. The algorithm to use to decompress the data when
reading it. Supported options are `GZIP` and `NONE`. Defaults to `NONE`.
reader_func: Optional. A function to control how to read data from shards.
If present, the function will be traced and executed as graph computation.
Returns:
A `tf.data.Dataset` instance."
1483,make_saveable_from_iterator,tensorflow/tensorflow/python/data/experimental/ops/iterator_ops.py,49,function,"Returns a SaveableObject for saving/restoring iterator state using Saver.
Args:
iterator: Iterator.
external_state_policy: A string that identifies how to handle input
pipelines that depend on external state. Possible values are
'ignore': The external state is silently ignored.
'warn': The external state is ignored, logging a warning.
'fail': The operation fails upon encountering external state.
By default we set it to 'fail'.
Returns:
A SaveableObject for saving/restoring iterator state using Saver.
Raises:
ValueError: If iterator does not support checkpointing.
ValueError: If `external_state_policy` is not one of 'warn', 'ignore' or
'fail'.
For example:
```python
with tf.Graph().as_default():
ds = tf.data.Dataset.range(10)
iterator = ds.make_initializable_iterator()
# Build the iterator SaveableObject.
saveable_obj = tf.data.experimental.make_saveable_from_iterator(iterator)
# Add the SaveableObject to the SAVEABLE_OBJECTS collection so
# it can be automatically saved using Saver.
tf.compat.v1.add_to_collection(tf.GraphKeys.SAVEABLE_OBJECTS, saveable_obj)
saver = tf.compat.v1.train.Saver()
while continue_training:
... Perform training ...
if should_save_checkpoint:
saver.save()
```
Note: When restoring the iterator, the existing iterator state is completely
discarded. This means that any changes you may have made to the Dataset
graph will be discarded as well! This includes the new Dataset graph
that you may have built during validation. So, while running validation,
make sure to run the initializer for the validation input pipeline after
restoring the checkpoint.
Note: Not all iterators support checkpointing yet. Attempting to save the
state of an unsupported iterator will throw an error."
1484,CheckpointInputPipelineHook,tensorflow/tensorflow/python/data/experimental/ops/iterator_ops.py,106,class,"Checkpoints input pipeline state every N steps or seconds.
This hook saves the state of the iterators in the `Graph` so that when
training is resumed the input pipeline continues from where it left off.
This could potentially avoid overfitting in certain pipelines where the
number of training steps per eval are small compared to the dataset
size or if the training pipeline is pre-empted.
Differences from `CheckpointSaverHook`:
1. Saves only the input pipelines in the ""iterators"" collection and not the
global variables or other saveable objects.
2. Does not write the `GraphDef` and `MetaGraphDef` to the summary.
Example of checkpointing the training pipeline:
```python
est = tf.estimator.Estimator(model_fn)
while True:
est.train(
train_input_fn,
hooks=[tf.data.experimental.CheckpointInputPipelineHook(est)],
steps=train_steps_per_eval)
# Note: We do not pass the hook here.
metrics = est.evaluate(eval_input_fn)
if should_stop_the_training(metrics):
break
```
This hook should be used if the input pipeline state needs to be saved
separate from the model checkpoint. Doing so may be useful for a few reasons:
1. The input pipeline checkpoint may be large, if there are large shuffle
or prefetch buffers for instance, and may bloat the checkpoint size.
2. If the input pipeline is shared between training and validation, restoring
the checkpoint during validation may override the validation input
pipeline.
For saving the input pipeline checkpoint alongside the model weights use
`tf.data.experimental.make_saveable_from_iterator` directly to create a
`SaveableObject` and add to the `SAVEABLE_OBJECTS` collection. Note, however,
that you will need to be careful not to restore the training iterator during
eval. You can do that by not adding the iterator to the SAVEABLE_OBJECTS
collector when building the eval graph."
1485,begin,tensorflow/tensorflow/python/data/experimental/ops/iterator_ops.py,227,method,
1486,after_create_session,tensorflow/tensorflow/python/data/experimental/ops/iterator_ops.py,244,method,
1487,before_run,tensorflow/tensorflow/python/data/experimental/ops/iterator_ops.py,284,method,
1488,after_run,tensorflow/tensorflow/python/data/experimental/ops/iterator_ops.py,290,method,
1489,end,tensorflow/tensorflow/python/data/experimental/ops/iterator_ops.py,293,method,
1490,map_defun,tensorflow/tensorflow/python/data/experimental/ops/map_defun.py,26,function,"Map a function on the list of tensors unpacked from `elems` on dimension 0.
Args:
fn: A function (`function.defun`) that takes a list of tensors and returns
another list of tensors. The output list has the same types as
output_dtypes. The elements of the output list have the same dimension 0
as `elems`, and the remaining dimensions correspond to those of
`fn_output_shapes`.
elems: A list of tensors.
output_dtypes: A list of dtypes corresponding to the output types of the
function.
output_shapes: A list of `TensorShape`s corresponding to the output shapes
from each invocation of the function on slices of inputs.
max_intra_op_parallelism: An integer. If positive, sets the max parallelism
limit of each function call to this.
Raises:
ValueError: if any of the inputs are malformed.
Returns:
A list of `Tensor` objects with the same types as `output_dtypes`."
1491,MatchingFilesDataset,tensorflow/tensorflow/python/data/experimental/ops/matching_files.py,28,class,A `Dataset` that list the files according to the input patterns.
1492,element_spec,tensorflow/tensorflow/python/data/experimental/ops/matching_files.py,38,method,
1493,model,tensorflow/tensorflow/python/data/experimental/ops/optimization.py,24,function,"A transformation that models performance.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1494,optimize,tensorflow/tensorflow/python/data/experimental/ops/optimization.py,39,function,"A transformation that applies optimizations.
Args:
optimizations: (Optional.) A `tf.string` vector `tf.Tensor` identifying
optimizations to use. If not specified, the default set of optimizations
is applied.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1495,MapVectorizationOptions,tensorflow/tensorflow/python/data/experimental/ops/optimization_options.py,36,class,Represents options for the MapVectorization optimization.
1496,OptimizationOptions,tensorflow/tensorflow/python/data/experimental/ops/optimization_options.py,70,class,"Represents options for dataset optimizations.
You can set the optimization options of a dataset through the
`experimental_optimization` property of `tf.data.Options`; the property is
an instance of `tf.data.experimental.OptimizationOptions`.
```python
options = tf.data.Options()
options.experimental_optimization.noop_elimination = True
options.experimental_optimization.map_vectorization.enabled = True
options.experimental_optimization.apply_default_optimizations = False
dataset = dataset.with_options(options)
```"
1497,parse_example_dataset,tensorflow/tensorflow/python/data/experimental/ops/parsing_ops.py,110,function,"A transformation that parses `Example` protos into a `dict` of tensors.
Parses a number of serialized `Example` protos given in `serialized`. We refer
to `serialized` as a batch with `batch_size` many entries of individual
`Example` protos.
This op parses serialized examples into a dictionary mapping keys to `Tensor`,
`SparseTensor`, and `RaggedTensor` objects. `features` is a dict from keys to
`VarLenFeature`, `RaggedFeature`, `SparseFeature`, and `FixedLenFeature`
objects. Each `VarLenFeature` and `SparseFeature` is mapped to a
`SparseTensor`; each `RaggedFeature` is mapped to a `RaggedTensor`; and each
`FixedLenFeature` is mapped to a `Tensor`. See `tf.io.parse_example` for more
details about feature dictionaries.
Args:
features: A `dict` mapping feature keys to `FixedLenFeature`,
`VarLenFeature`, `RaggedFeature`, and `SparseFeature` values.
num_parallel_calls: (Optional.) A `tf.int32` scalar `tf.Tensor`,
representing the number of parsing processes to call in parallel.
deterministic: (Optional.) A boolean controlling whether determinism
should be traded for performance by allowing elements to be produced out
of order if some parsing calls complete faster than others. If
`deterministic` is `None`, the
`tf.data.Options.experimental_deterministic` dataset option (`True` by
default) is used to decide whether to produce elements
deterministically.
Returns:
A dataset transformation function, which can be passed to
`tf.data.Dataset.apply`.
Raises:
ValueError: if features argument is None."
1498,prefetch_to_device,tensorflow/tensorflow/python/data/experimental/ops/prefetching_ops.py,37,function,"A transformation that prefetches dataset values to the given `device`.
NOTE: Although the transformation creates a `tf.data.Dataset`, the
transformation must be the final `Dataset` in the input pipeline.
Args:
device: A string. The name of a device to which elements will be prefetched.
buffer_size: (Optional.) The number of elements to buffer on `device`.
Defaults to an automatically chosen value.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1499,copy_to_device,tensorflow/tensorflow/python/data/experimental/ops/prefetching_ops.py,60,function,"A transformation that copies dataset elements to the given `target_device`.
Args:
target_device: The name of a device to which elements will be copied.
source_device: The original device on which `input_dataset` will be placed.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1500,map_on_gpu,tensorflow/tensorflow/python/data/experimental/ops/prefetching_ops.py,260,function,"Maps `map_func` across the elements of this dataset.
NOTE: This is a highly experimental version of `tf.data.Dataset.map` that runs
`map_func` on GPU. It must be used after applying the
`tf.data.experimental.copy_to_device` transformation with a GPU device
argument.
Args:
map_func: A function mapping a nested structure of tensors (having shapes
and types defined by `self.output_shapes` and `self.output_types`) to
another nested structure of tensors.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1501,RandomDatasetV2,tensorflow/tensorflow/python/data/experimental/ops/random_ops.py,32,class,A `Dataset` of pseudorandom values.
1502,element_spec,tensorflow/tensorflow/python/data/experimental/ops/random_ops.py,43,method,
1503,RandomDatasetV1,tensorflow/tensorflow/python/data/experimental/ops/random_ops.py,48,class,A `Dataset` of pseudorandom values.
1504,make_tf_record_dataset,tensorflow/tensorflow/python/data/experimental/ops/readers.py,223,function,"Reads and optionally parses TFRecord files into a dataset.
Provides common functionality such as batching, optional parsing, shuffling,
and performant defaults.
Args:
file_pattern: List of files or patterns of TFRecord file paths.
See `tf.io.gfile.glob` for pattern rules.
batch_size: An int representing the number of records to combine
in a single batch.
parser_fn: (Optional.) A function accepting string input to parse
and process the record contents. This function must map records
to components of a fixed shape, so they may be batched. By
default, uses the record contents unmodified.
num_epochs: (Optional.) An int specifying the number of times this
dataset is repeated. If None (the default), cycles through the
dataset forever.
shuffle: (Optional.) A bool that indicates whether the input
should be shuffled. Defaults to `True`.
shuffle_buffer_size: (Optional.) Buffer size to use for
shuffling. A large buffer size ensures better shuffling, but
increases memory usage and startup time.
shuffle_seed: (Optional.) Randomization seed to use for shuffling.
prefetch_buffer_size: (Optional.) An int specifying the number of
feature batches to prefetch for performance improvement.
Defaults to auto-tune. Set to 0 to disable prefetching.
num_parallel_reads: (Optional.) Number of threads used to read
records from files. By default or if set to a value >1, the
results will be interleaved. Defaults to `24`.
num_parallel_parser_calls: (Optional.) Number of parallel
records to parse in parallel. Defaults to `batch_size`.
drop_final_batch: (Optional.) Whether the last batch should be
dropped in case its size is smaller than `batch_size`; the
default behavior is not to drop the smaller batch.
Returns:
A dataset, where each element matches the output of `parser_fn`
except it will have an additional leading `batch-size` dimension,
or a `batch_size`-length 1-D tensor of strings if `parser_fn` is
unspecified."
1505,make_csv_dataset_v2,tensorflow/tensorflow/python/data/experimental/ops/readers.py,322,function,"Reads CSV files into a dataset.
Reads CSV files into a dataset, where each element is a (features, labels)
tuple that corresponds to a batch of CSV rows. The features dictionary
maps feature column names to `Tensor`s containing the corresponding
feature data, and labels is a `Tensor` containing the batch's label data.
Args:
file_pattern: List of files or patterns of file paths containing CSV
records. See `tf.io.gfile.glob` for pattern rules.
batch_size: An int representing the number of records to combine
in a single batch.
column_names: An optional list of strings that corresponds to the CSV
columns, in order. One per column of the input record. If this is not
provided, infers the column names from the first row of the records.
These names will be the keys of the features dict of each dataset element.
column_defaults: A optional list of default values for the CSV fields. One
item per selected column of the input record. Each item in the list is
either a valid CSV dtype (float32, float64, int32, int64, or string), or a
`Tensor` with one of the aforementioned types. The tensor can either be
a scalar default value (if the column is optional), or an empty tensor (if
the column is required). If a dtype is provided instead of a tensor, the
column is also treated as required. If this list is not provided, tries
to infer types based on reading the first num_rows_for_inference rows of
files specified, and assumes all columns are optional, defaulting to `0`
for numeric values and `""""` for string values. If both this and
`select_columns` are specified, these must have the same lengths, and
`column_defaults` is assumed to be sorted in order of increasing column
index.
label_name: A optional string corresponding to the label column. If
provided, the data for this column is returned as a separate `Tensor` from
the features dictionary, so that the dataset complies with the format
expected by a `tf.Estimator.train` or `tf.Estimator.evaluate` input
function.
select_columns: An optional list of integer indices or string column
names, that specifies a subset of columns of CSV data to select. If
column names are provided, these must correspond to names provided in
`column_names` or inferred from the file header lines. When this argument
is specified, only a subset of CSV columns will be parsed and returned,
corresponding to the columns specified. Using this results in faster
parsing and lower memory usage. If both this and `column_defaults` are
specified, these must have the same lengths, and `column_defaults` is
assumed to be sorted in order of increasing column index.
field_delim: An optional `string`. Defaults to `"",""`. Char delimiter to
separate fields in a record.
use_quote_delim: An optional bool. Defaults to `True`. If false, treats
double quotation marks as regular characters inside of the string fields.
na_value: Additional string to recognize as NA/NaN.
header: A bool that indicates whether the first rows of provided CSV files
correspond to header lines with column names, and should not be included
in the data.
num_epochs: An int specifying the number of times this dataset is repeated.
If None, cycles through the dataset forever.
shuffle: A bool that indicates whether the input should be shuffled.
shuffle_buffer_size: Buffer size to use for shuffling. A large buffer size
ensures better shuffling, but increases memory usage and startup time.
shuffle_seed: Randomization seed to use for shuffling.
prefetch_buffer_size: An int specifying the number of feature
batches to prefetch for performance improvement. Recommended value is the
number of batches consumed per training step. Defaults to auto-tune.
num_parallel_reads: Number of threads used to read CSV records from files.
If >1, the results will be interleaved. Defaults to `1`.
sloppy: If `True`, reading performance will be improved at
the cost of non-deterministic ordering. If `False`, the order of elements
produced is deterministic prior to shuffling (elements are still
randomized if `shuffle=True`. Note that if the seed is set, then order
of elements after shuffling is deterministic). Defaults to `False`.
num_rows_for_inference: Number of rows of a file to use for type inference
if record_defaults is not provided. If None, reads all the rows of all
the files. Defaults to 100.
compression_type: (Optional.) A `tf.string` scalar evaluating to one of
`""""` (no compression), `""ZLIB""`, or `""GZIP""`. Defaults to no compression.
ignore_errors: (Optional.) If `True`, ignores errors with CSV file parsing,
such as malformed data or empty lines, and moves on to the next valid
CSV record. Otherwise, the dataset raises an error and stops processing
when encountering any invalid records. Defaults to `False`.
Returns:
A dataset, where each element is a (features, labels) tuple that corresponds
to a batch of `batch_size` CSV rows. The features dictionary maps feature
column names to `Tensor`s containing the corresponding column data, and
labels is a `Tensor` containing the column data for the label column
specified by `label_name`.
Raises:
ValueError: If any of the arguments is malformed."
1506,make_csv_dataset_v1,tensorflow/tensorflow/python/data/experimental/ops/readers.py,569,function,
1507,CsvDatasetV2,tensorflow/tensorflow/python/data/experimental/ops/readers.py,604,class,A Dataset comprising lines from one or more CSV files.
1508,element_spec,tensorflow/tensorflow/python/data/experimental/ops/readers.py,778,method,
1509,CsvDatasetV1,tensorflow/tensorflow/python/data/experimental/ops/readers.py,783,class,A Dataset comprising lines from one or more CSV files.
1510,make_batched_features_dataset_v2,tensorflow/tensorflow/python/data/experimental/ops/readers.py,874,function,"Returns a `Dataset` of feature dictionaries from `Example` protos.
If label_key argument is provided, returns a `Dataset` of tuple
comprising of feature dictionaries and label.
Example:
```
serialized_examples = [
features {
feature { key: ""age"" value { int64_list { value: [ 0 ] } } }
feature { key: ""gender"" value { bytes_list { value: [ ""f"" ] } } }
feature { key: ""kws"" value { bytes_list { value: [ ""code"", ""art"" ] } } }
},
features {
feature { key: ""age"" value { int64_list { value: [] } } }
feature { key: ""gender"" value { bytes_list { value: [ ""f"" ] } } }
feature { key: ""kws"" value { bytes_list { value: [ ""sports"" ] } } }
}
]
```
We can use arguments:
```
features: {
""age"": FixedLenFeature([], dtype=tf.int64, default_value=-1),
""gender"": FixedLenFeature([], dtype=tf.string),
""kws"": VarLenFeature(dtype=tf.string),
}
```
And the expected output is:
```python
{
""age"": [[0], [-1]],
""gender"": [[""f""], [""f""]],
""kws"": SparseTensor(
indices=[[0, 0], [0, 1], [1, 0]],
values=[""code"", ""art"", ""sports""]
dense_shape=[2, 2]),
}
```
Args:
file_pattern: List of files or patterns of file paths containing
`Example` records. See `tf.io.gfile.glob` for pattern rules.
batch_size: An int representing the number of records to combine
in a single batch.
features: A `dict` mapping feature keys to `FixedLenFeature` or
`VarLenFeature` values. See `tf.io.parse_example`.
reader: A function or class that can be
called with a `filenames` tensor and (optional) `reader_args` and returns
a `Dataset` of `Example` tensors. Defaults to `tf.data.TFRecordDataset`.
label_key: (Optional) A string corresponding to the key labels are stored in
`tf.Examples`. If provided, it must be one of the `features` key,
otherwise results in `ValueError`.
reader_args: Additional arguments to pass to the reader class.
num_epochs: Integer specifying the number of times to read through the
dataset. If None, cycles through the dataset forever. Defaults to `None`.
shuffle: A boolean, indicates whether the input should be shuffled. Defaults
to `True`.
shuffle_buffer_size: Buffer size of the ShuffleDataset. A large capacity
ensures better shuffling but would increase memory usage and startup time.
shuffle_seed: Randomization seed to use for shuffling.
prefetch_buffer_size: Number of feature batches to prefetch in order to
improve performance. Recommended value is the number of batches consumed
per training step. Defaults to auto-tune.
reader_num_threads: Number of threads used to read `Example` records. If >1,
the results will be interleaved. Defaults to `1`.
parser_num_threads: Number of threads to use for parsing `Example` tensors
into a dictionary of `Feature` tensors. Defaults to `2`.
sloppy_ordering: If `True`, reading performance will be improved at
the cost of non-deterministic ordering. If `False`, the order of elements
produced is deterministic prior to shuffling (elements are still
randomized if `shuffle=True`. Note that if the seed is set, then order
of elements after shuffling is deterministic). Defaults to `False`.
drop_final_batch: If `True`, and the batch size does not evenly divide the
input dataset size, the final smaller batch will be dropped. Defaults to
`False`.
Returns:
A dataset of `dict` elements, (or a tuple of `dict` elements and label).
Each `dict` maps feature keys to `Tensor` or `SparseTensor` objects.
Raises:
TypeError: If `reader` is of the wrong type.
ValueError: If `label_key` is not one of the `features` keys."
1511,make_batched_features_dataset_v1,tensorflow/tensorflow/python/data/experimental/ops/readers.py,1058,function,
1512,SqlDatasetV2,tensorflow/tensorflow/python/data/experimental/ops/readers.py,1114,class,A `Dataset` consisting of the results from a SQL query.
1513,element_spec,tensorflow/tensorflow/python/data/experimental/ops/readers.py,1155,method,
1514,SqlDatasetV1,tensorflow/tensorflow/python/data/experimental/ops/readers.py,1160,class,A `Dataset` consisting of the results from a SQL query.
1515,rejection_resample,tensorflow/tensorflow/python/data/experimental/ops/resampling.py,37,function,"A transformation that resamples a dataset to achieve a target distribution.
**NOTE** Resampling is performed via rejection sampling; some fraction
of the input values will be dropped.
Args:
class_func: A function mapping an element of the input dataset to a scalar
`tf.int32` tensor. Values should be in `[0, num_classes)`.
target_dist: A floating point type tensor, shaped `[num_classes]`.
initial_dist: (Optional.) A floating point type tensor, shaped
`[num_classes]`. If not provided, the true class distribution is
estimated live in a streaming fashion.
seed: (Optional.) Python integer seed for the resampler.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1516,scan,tensorflow/tensorflow/python/data/experimental/ops/scan_ops.py,158,function,"A transformation that scans a function across an input dataset.
This transformation is a stateful relative of `tf.data.Dataset.map`.
In addition to mapping `scan_func` across the elements of the input dataset,
`scan()` accumulates one or more state tensors, whose initial values are
`initial_state`.
Args:
initial_state: A nested structure of tensors, representing the initial state
of the accumulator.
scan_func: A function that maps `(old_state, input_element)` to
`(new_state, output_element)`. It must take two arguments and return a
pair of nested structures of tensors. The `new_state` must match the
structure of `initial_state`.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1517,shuffle_and_repeat,tensorflow/tensorflow/python/data/experimental/ops/shuffle_ops.py,60,function,"Shuffles and repeats a Dataset, reshuffling with each repetition.
>>> d = tf.data.Dataset.from_tensor_slices([1, 2, 3])
>>> d = d.apply(tf.data.experimental.shuffle_and_repeat(2, count=2))
>>> [elem.numpy() for elem in d] # doctest: +SKIP
[2, 3, 1, 1, 3, 2]
```python
dataset.apply(
tf.data.experimental.shuffle_and_repeat(buffer_size, count, seed))
```
produces the same output as
```python
dataset.shuffle(
buffer_size, seed=seed, reshuffle_each_iteration=True).repeat(count)
```
In each repetition, this dataset fills a buffer with `buffer_size` elements,
then randomly samples elements from this buffer, replacing the selected
elements with new elements. For perfect shuffling, set the buffer size equal
to the full size of the dataset.
For instance, if your dataset contains 10,000 elements but `buffer_size` is
set to 1,000, then `shuffle` will initially select a random element from
only the first 1,000 elements in the buffer. Once an element is selected,
its space in the buffer is replaced by the next (i.e. 1,001-st) element,
maintaining the 1,000 element buffer.
Args:
buffer_size: A `tf.int64` scalar `tf.Tensor`, representing the maximum
number elements that will be buffered when prefetching.
count: (Optional.) A `tf.int64` scalar `tf.Tensor`, representing the number
of times the dataset should be repeated. The default behavior (if `count`
is `None` or `-1`) is for the dataset be repeated indefinitely.
seed: (Optional.) A `tf.int64` scalar `tf.Tensor`, representing the random
seed that will be used to create the distribution. See
`tf.random.set_seed` for behavior.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1518,sleep,tensorflow/tensorflow/python/data/experimental/ops/sleep.py,37,function,"Sleeps for `sleep_microseconds` before producing each input element.
Args:
sleep_microseconds: The number of microseconds to sleep before producing an
input element.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1519,legacy_snapshot,tensorflow/tensorflow/python/data/experimental/ops/snapshot.py,108,function,"Writes to/reads from a snapshot of a dataset.
This function attempts to determine whether a valid snapshot exists at the
`path`, and reads from the snapshot if so. If not, it will run the
preprocessing pipeline as usual, and write out a snapshot of the data
processed for future use.
Args:
path: A directory where we want to save our snapshots and/or read from a
previously saved snapshot.
compression: The type of compression to apply to the Dataset. Currently
supports ""GZIP"" or None. Defaults to None (no compression).
reader_path_prefix: A prefix to add to the path when reading from snapshots.
Defaults to None.
writer_path_prefix: A prefix to add to the path when writing to snapshots.
Defaults to None.
shard_size_bytes: The size of each shard to be written by the snapshot
dataset op. Defaults to 10 GiB.
pending_snapshot_expiry_seconds: How long to wait (in seconds) before the
snapshot op considers a previously unfinished snapshot to be stale.
num_reader_threads: Number of threads to parallelize reading from snapshot.
Especially useful if compression is turned on since the decompression
operation tends to be intensive. Defaults to 1. If > 1, then this might
introduce non-determinism i.e. the order in which the elements are read
from the snapshot are different from the order they're written.
reader_buffer_size: Maximum number of elements we can prefetch reading from
the snapshot. Defaults to 1. Increasing this might improve performance but
will increase memory consumption.
num_writer_threads: Number of threads to parallelize writing from snapshot.
We'll open up `num_writer_threads` files and write to them in parallel.
Especially useful if compression is turned on since the compression
operation tends to be intensive. Defaults to 1. If > 1, then this might
introduce non-determinism i.e. the order in which the elements are read
from the upstream iterator are different from the order they're written.
writer_buffer_size: Maximum number of pipeline elements to fill up the
buffer before writing them out using `num_writer_threads`.
shuffle_on_read: If this is True, then the order in which examples are
produced when reading from a snapshot will be random. Defaults to False.
shuffle_seed: Optional. If shuffle_seed is set, the random number generator
used for shuffling (when shuffle_on_read is turned on) is seeded by the
given seed. Otherwise, it is seeded by a random seed that differs for
every run.
mode: The mode at which snapshot should operate. Valid options are ""auto"",
""read"", ""write"", and ""passthrough"". The default mode is ""auto"", where the
snapshot op will automatically determine what mode to operate in.
snapshot_name: If set, use the supplied string as a named snapshot name
instead of introspecting the data pipeline and automatically generating a
unique identifier for the snapshot.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1520,snapshot,tensorflow/tensorflow/python/data/experimental/ops/snapshot.py,258,function,"API to persist the output of the input dataset.
The snapshot API allows users to transparently persist the output of their
preprocessing pipeline to disk, and materialize the pre-processed data on a
different training run.
This API enables repeated preprocessing steps to be consolidated, and allows
re-use of already processed data, trading off disk storage and network
bandwidth for freeing up more valuable CPU resources and accelerator compute
time.
https://github.com/tensorflow/community/blob/master/rfcs/20200107-tf-data-snapshot.md
has detailed design documentation of this feature.
Users can specify various options to control the behavior of snapshot,
including how snapshots are read from and written to by passing in
user-defined functions to the `reader_func` and `shard_func` parameters.
`shard_func` is a user specified function that maps input elements to snapshot
shards.
Users may want to specify this function to control how snapshot files should
be written to disk. Below is an example of how a potential shard_func could
be written.
```python
dataset = ...
dataset = dataset.enumerate()
dataset = dataset.apply(tf.data.experimental.snapshot(""/path/to/snapshot/dir"",
shard_func=lambda x, y: x % NUM_SHARDS, ...))
dataset = dataset.map(lambda x, y: y)
```
`reader_func` is a user specified function that accepts a single argument:
(1) a Dataset of Datasets, each representing a ""split"" of elements of the
original dataset. The cardinality of the input dataset matches the
number of the shards specified in the `shard_func` (see above). The function
should return a Dataset of elements of the original dataset.
Users may want specify this function to control how snapshot files should be
read from disk, including the amount of shuffling and parallelism.
Here is an example of a standard reader function a user can define. This
function enables both dataset shuffling and parallel reading of datasets:
```python
def user_reader_func(datasets):
# shuffle the datasets splits
datasets = datasets.shuffle(NUM_CORES)
# read datasets in parallel and interleave their elements
return datasets.interleave(lambda x: x, num_parallel_calls=AUTOTUNE)
dataset = dataset.apply(tf.data.experimental.snapshot(""/path/to/snapshot/dir"",
reader_func=user_reader_func))
```
By default, snapshot parallelizes reads by the number of cores available on
the system, but will not attempt to shuffle the data.
Args:
path: Required. A directory to use for storing / loading the snapshot to /
from.
compression: Optional. The type of compression to apply to the snapshot
written to disk. Supported options are `GZIP`, `SNAPPY`, `AUTO` or None.
Defaults to AUTO, which attempts to pick an appropriate compression
algorithm for the dataset.
reader_func: Optional. A function to control how to read data from snapshot
shards.
shard_func: Optional. A function to control how to shard data when writing a
snapshot.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1521,StatsAggregatorV2,tensorflow/tensorflow/python/data/experimental/ops/stats_aggregator.py,31,class,"A stateful resource that aggregates statistics from one or more iterators.
To record statistics, use one of the custom transformation functions defined
in this module when defining your `tf.data.Dataset`. All statistics will be
aggregated by the `StatsAggregator` that is associated with a particular
iterator (see below). For example, to record the latency of producing each
element by iterating over a dataset:
```python
dataset = ...
dataset = dataset.apply(tf.data.experimental.latency_stats(""total_bytes""))
```
To associate a `StatsAggregator` with a `tf.data.Dataset` object, use
the following pattern:
```python
aggregator = tf.data.experimental.StatsAggregator()
dataset = ...
# Apply `StatsOptions` to associate `dataset` with `aggregator`.
options = tf.data.Options()
options.experimental_stats.aggregator = aggregator
dataset = dataset.with_options(options)
```
Note: This interface is experimental and expected to change. In particular,
we expect to add other implementations of `StatsAggregator` that provide
different ways of exporting statistics, and add more types of statistics."
1522,StatsAggregatorV1,tensorflow/tensorflow/python/data/experimental/ops/stats_aggregator.py,82,class,"A stateful resource that aggregates statistics from one or more iterators.
To record statistics, use one of the custom transformation functions defined
in this module when defining your `tf.data.Dataset`. All statistics will be
aggregated by the `StatsAggregator` that is associated with a particular
iterator (see below). For example, to record the latency of producing each
element by iterating over a dataset:
```python
dataset = ...
dataset = dataset.apply(tf.data.experimental.latency_stats(""total_bytes""))
```
To associate a `StatsAggregator` with a `tf.data.Dataset` object, use
the following pattern:
```python
aggregator = tf.data.experimental.StatsAggregator()
dataset = ...
# Apply `StatsOptions` to associate `dataset` with `aggregator`.
options = tf.data.Options()
options.experimental_stats.aggregator = aggregator
dataset = dataset.with_options(options)
```
To get a protocol buffer summary of the currently aggregated statistics,
use the `StatsAggregator.get_summary()` tensor. The easiest way to do this
is to add the returned tensor to the `tf.GraphKeys.SUMMARIES` collection,
so that the summaries will be included with any existing summaries.
```python
aggregator = tf.data.experimental.StatsAggregator()
# ...
stats_summary = aggregator.get_summary()
tf.compat.v1.add_to_collection(tf.GraphKeys.SUMMARIES, stats_summary)
```
Note: This interface is experimental and expected to change. In particular,
we expect to add other implementations of `StatsAggregator` that provide
different ways of exporting statistics, and add more types of statistics."
1523,get_summary,tensorflow/tensorflow/python/data/experimental/ops/stats_aggregator.py,130,method,"Returns a string `tf.Tensor` that summarizes the aggregated statistics.
The returned tensor will contain a serialized `tf.compat.v1.summary.Summary`
protocol
buffer, which can be used with the standard TensorBoard logging facilities.
Returns:
A scalar string `tf.Tensor` that summarizes the aggregated statistics."
1524,set_stats_aggregator,tensorflow/tensorflow/python/data/experimental/ops/stats_ops.py,29,function,"Set the given `stats_aggregator` for aggregating the input dataset stats.
Args:
stats_aggregator: A `tf.data.experimental.StatsAggregator` object.
prefix: (Optional) String, all statistics recorded for the input `dataset`
will have given `prefix` prepend with the name.
counter_prefix: (Optional) String, all statistics recorded as `counters`
will have the given `prefix` for the counter. Defaults to ""/tensorflow"".
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1525,bytes_produced_stats,tensorflow/tensorflow/python/data/experimental/ops/stats_ops.py,52,function,"Records the number of bytes produced by each element of the input dataset.
To consume the statistics, associate a `StatsAggregator` with the output
dataset.
Args:
tag: String. All statistics recorded by the returned transformation will
be associated with the given `tag`.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1526,latency_stats,tensorflow/tensorflow/python/data/experimental/ops/stats_ops.py,75,function,"Records the latency of producing each element of the input dataset.
To consume the statistics, associate a `StatsAggregator` with the output
dataset.
Args:
tag: String. All statistics recorded by the returned transformation will
be associated with the given `tag`.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1527,StatsOptions,tensorflow/tensorflow/python/data/experimental/ops/stats_options.py,28,class,"Represents options for collecting dataset stats using `StatsAggregator`.
You can set the stats options of a dataset through the `experimental_stats`
property of `tf.data.Options`; the property is an instance of
`tf.data.experimental.StatsOptions`. For example, to collect latency stats
on all dataset edges, use the following pattern:
```python
aggregator = tf.data.experimental.StatsAggregator()
options = tf.data.Options()
options.experimental_stats.aggregator = aggregator
options.experimental_stats.latency_all_edges = True
dataset = dataset.with_options(options)
```"
1528,take_while,tensorflow/tensorflow/python/data/experimental/ops/take_while_ops.py,56,function,"A transformation that stops dataset iteration based on a `predicate`.
Args:
predicate: A function that maps a nested structure of tensors (having shapes
and types defined by `self.output_shapes` and `self.output_types`) to a
scalar `tf.bool` tensor.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1529,assert_next,tensorflow/tensorflow/python/data/experimental/ops/testing.py,26,function,"A transformation that asserts which transformations happen next.
Transformations should be referred to by their base name, not including
version suffix. For example, use ""Batch"" instead of ""BatchV2"". ""Batch"" will
match any of ""Batch"", ""BatchV1"", ""BatchV2"", etc.
Args:
transformations: A `tf.string` vector `tf.Tensor` identifying the
transformations that are expected to happen next.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1530,non_serializable,tensorflow/tensorflow/python/data/experimental/ops/testing.py,49,function,"A non-serializable identity transformation.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1531,sleep,tensorflow/tensorflow/python/data/experimental/ops/testing.py,64,function,"Sleeps for `sleep_microseconds` before producing each input element.
Args:
sleep_microseconds: The number of microseconds to sleep before producing an
input element.
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1532,ThreadingOptions,tensorflow/tensorflow/python/data/experimental/ops/threading_options.py,26,class,"Represents options for dataset threading.
You can set the threading options of a dataset through the
`experimental_threading` property of `tf.data.Options`; the property is
an instance of `tf.data.experimental.ThreadingOptions`.
```python
options = tf.data.Options()
options.experimental_threading.private_threadpool_size = 10
dataset = dataset.with_options(options)
```"
1533,PrivateThreadPool,tensorflow/tensorflow/python/data/experimental/ops/threadpool.py,41,class,A stateful resource that represents a private thread pool.
1534,override_threadpool,tensorflow/tensorflow/python/data/experimental/ops/threadpool.py,78,function,"Returns a new dataset that uses the given thread pool for its operations.
Args:
dataset: A `tf.data.Dataset` object.
thread_pool: A `PrivateThreadPool` object.
Returns:
A dataset containing the same values as `dataset`, but which uses
`thread_pool` to compute any of its parallel operations (such as
`tf.data.Dataset.map`)."
1535,unique,tensorflow/tensorflow/python/data/experimental/ops/unique.py,27,function,"Creates a `Dataset` from another `Dataset`, discarding duplicates.
Use this transformation to produce a dataset that contains one instance of
each unique element in the input. For example:
```python
dataset = tf.data.Dataset.from_tensor_slices([1, 37, 2, 37, 2, 1])
# Using `unique()` will drop the duplicate elements.
dataset = dataset.apply(tf.data.experimental.unique()) # ==> { 1, 37, 2 }
```
Returns:
A `Dataset` transformation function, which can be passed to
`tf.data.Dataset.apply`."
1536,TFRecordWriter,tensorflow/tensorflow/python/data/experimental/ops/writers.py,30,class,"Writes a dataset to a TFRecord file.
The elements of the dataset must be scalar strings. To serialize dataset
elements as strings, you can use the `tf.io.serialize_tensor` function.
```python
dataset = tf.data.Dataset.range(3)
dataset = dataset.map(tf.io.serialize_tensor)
writer = tf.data.experimental.TFRecordWriter(""/path/to/file.tfrecord"")
writer.write(dataset)
```
To read back the elements, use `TFRecordDataset`.
```python
dataset = tf.data.TFRecordDataset(""/path/to/file.tfrecord"")
dataset = dataset.map(lambda x: tf.io.parse_tensor(x, tf.int64))
```
To shard a `dataset` across multiple TFRecord files:
```python
dataset = ... # dataset to be written
def reduce_func(key, dataset):
filename = tf.strings.join([PATH_PREFIX, tf.strings.as_string(key)])
writer = tf.data.experimental.TFRecordWriter(filename)
writer.write(dataset.map(lambda _, x: x))
return tf.data.Dataset.from_tensors(filename)
dataset = dataset.enumerate()
dataset = dataset.apply(tf.data.experimental.group_by_window(
lambda i, _: i % NUM_SHARDS, reduce_func, tf.int64.max
))
```"
1537,write,tensorflow/tensorflow/python/data/experimental/ops/writers.py,85,method,"Writes a dataset to a TFRecord file.
An operation that writes the content of the specified dataset to the file
specified in the constructor.
If the file exists, it will be overwritten.
Args:
dataset: a `tf.data.Dataset` whose elements are to be written to a file
Returns:
In graph mode, this returns an operation which when executed performs the
write. In eager mode, the write is performed by the method itself and
there is no return value.
Raises
TypeError: if `dataset` is not a `tf.data.Dataset`.
TypeError: if the elements produced by the dataset are not scalar strings."
1538,DispatchServer,tensorflow/tensorflow/python/data/experimental/service/server_lib.py,29,class,"An in-process tf.data service dispatch server.
A `tf.data.experimental.service.DispatchServer` coordinates a cluster of
`tf.data.experimental.service.WorkerServer`s. When the workers start, they
register themselves with the dispatcher.
>>> dispatcher = tf.data.experimental.service.DispatchServer(port=0)
>>> dispatcher_address = dispatcher.target.split(""://"")[1]
>>> worker = tf.data.experimental.service.WorkerServer(
... port=0, dispatcher_address=dispatcher_address)
>>> dataset = tf.data.Dataset.range(10)
>>> dataset = dataset.apply(tf.data.experimental.service.distribute(
... processing_mode=""parallel_epochs"", service=dispatcher.target))
>>> print(list(dataset.as_numpy_iterator()))
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
When starting a dedicated tf.data dispatch process, use join() to block
indefinitely after starting up the server.
```
dispatcher = tf.data.experimental.service.DispatchServer(port=5050)
dispatcher.join()
```"
1539,start,tensorflow/tensorflow/python/data/experimental/service/server_lib.py,78,method,"Starts this server.
>>> dispatcher = tf.data.experimental.service.DispatchServer(port=0,
... start=False)
>>> dispatcher.start()
Raises:
tf.errors.OpError: Or one of its subclasses if an error occurs while
starting the server."
1540,join,tensorflow/tensorflow/python/data/experimental/service/server_lib.py,91,method,"Blocks until the server has shut down.
This is useful when starting a dedicated dispatch process.
```
dispatcher = tf.data.experimental.service.DispatchServer(port=5050)
dispatcher.join()
```
Raises:
tf.errors.OpError: Or one of its subclasses if an error occurs while
joining the server."
1541,target,tensorflow/tensorflow/python/data/experimental/service/server_lib.py,108,method,"Returns a target that can be used to connect to the server.
>>> dispatcher = tf.data.experimental.service.DispatchServer(port=0)
>>> dataset = tf.data.Dataset.range(10)
>>> dataset = dataset.apply(tf.data.experimental.service.distribute(
... processing_mode=""parallel_epochs"", service=dispatcher.target))
The returned string will be in the form protocol://address, e.g.
""grpc://localhost:5050""."
1542,WorkerServer,tensorflow/tensorflow/python/data/experimental/service/server_lib.py,148,class,"An in-process tf.data service worker server.
A `tf.data.experimental.service.WorkerServer` performs `tf.data.Dataset`
processing for user-defined datasets, and provides the resulting elements over
RPC. A worker is associated with a single
`tf.data.experimental.service.DispatchServer`.
>>> dispatcher = tf.data.experimental.service.DispatchServer(port=0)
>>> dispatcher_address = dispatcher.target.split(""://"")[1]
>>> worker = tf.data.experimental.service.WorkerServer(
... port=0, dispatcher_address=dispatcher_address)
>>> dataset = tf.data.Dataset.range(10)
>>> dataset = dataset.apply(tf.data.experimental.service.distribute(
... processing_mode=""parallel_epochs"", service=dispatcher.target))
>>> print(list(dataset.as_numpy_iterator()))
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
When starting a dedicated tf.data worker process, use join() to block
indefinitely after starting up the server.
```
worker = tf.data.experimental.service.WorkerServer(
port=5051, dispatcher_address=""grpc://localhost:5050"")
worker.join()
```"
1543,start,tensorflow/tensorflow/python/data/experimental/service/server_lib.py,213,method,"Starts this server.
Raises:
tf.errors.OpError: Or one of its subclasses if an error occurs while
starting the server."
1544,join,tensorflow/tensorflow/python/data/experimental/service/server_lib.py,222,method,"Blocks until the server has shut down.
This is useful when starting a dedicated worker process.
```
worker_server = tf.data.experimental.service.WorkerServer(
port=5051, dispatcher_address=""grpc://localhost:5050"")
worker_server.join()
```
This method currently blocks forever.
Raises:
tf.errors.OpError: Or one of its subclasses if an error occurs while
joining the server."
1545,Foo,tensorflow/tensorflow/python/data/kernel_tests/map_test.py,153,class,Dummy class used for invalid return value tests.
1546,eager_only_combinations,tensorflow/tensorflow/python/data/kernel_tests/test_base.py,44,function,Returns the default test combinations for eager mode only tf.data tests.
1547,graph_only_combinations,tensorflow/tensorflow/python/data/kernel_tests/test_base.py,49,function,Returns the default test combinations for graph mode only tf.data tests.
1548,v2_only_combinations,tensorflow/tensorflow/python/data/kernel_tests/test_base.py,54,function,Returns the default test combinations for v1 only tf.data tests.
1549,DatasetV2,tensorflow/tensorflow/python/data/ops/dataset_ops.py,106,class,"Represents a potentially large set of elements.
The `tf.data.Dataset` API supports writing descriptive and efficient input
pipelines. `Dataset` usage follows a common pattern:
1. Create a source dataset from your input data.
2. Apply dataset transformations to preprocess the data.
3. Iterate over the dataset and process the elements.
Iteration happens in a streaming fashion, so the full dataset does not need to
fit into memory.
Source Datasets:
The simplest way to create a dataset is to create it from a python `list`:
>>> dataset = tf.data.Dataset.from_tensor_slices([1, 2, 3])
>>> for element in dataset:
... print(element)
tf.Tensor(1, shape=(), dtype=int32)
tf.Tensor(2, shape=(), dtype=int32)
tf.Tensor(3, shape=(), dtype=int32)
To process lines from files, use `tf.data.TextLineDataset`:
>>> dataset = tf.data.TextLineDataset([""file1.txt"", ""file2.txt""])
To process records written in the `TFRecord` format, use `TFRecordDataset`:
>>> dataset = tf.data.TFRecordDataset([""file1.tfrecords"", ""file2.tfrecords""])
To create a dataset of all files matching a pattern, use
`tf.data.Dataset.list_files`:
>>> dataset = tf.data.Dataset.list_files(""/path/*.txt"") # doctest: +SKIP
See `tf.data.FixedLengthRecordDataset` and `tf.data.Dataset.from_generator`
for more ways to create datasets.
Transformations:
Once you have a dataset, you can apply transformations to prepare the data for
your model:
>>> dataset = tf.data.Dataset.from_tensor_slices([1, 2, 3])
>>> dataset = dataset.map(lambda x: x*2)
>>> list(dataset.as_numpy_iterator())
[2, 4, 6]
Common Terms:
**Element**: A single output from calling `next()` on a dataset iterator.
Elements may be nested structures containing multiple components. For
example, the element `(1, (3, ""apple""))` has one tuple nested in another
tuple. The components are `1`, `3`, and `""apple""`.
**Component**: The leaf in the nested structure of an element.
Supported types:
Elements can be nested structures of tuples, named tuples, and dictionaries.
Note that Python lists are *not* treated as nested structures of components.
Instead, lists are converted to tensors and treated as components. For
example, the element `(1, [1, 2, 3])` has only two components; the tensor `1`
and the tensor `[1, 2, 3]`. Element components can be of any type
representable by `tf.TypeSpec`, including `tf.Tensor`, `tf.data.Dataset`,
`tf.sparse.SparseTensor`, `tf.RaggedTensor`, and `tf.TensorArray`.
>>> a = 1 # Integer element
>>> b = 2.0 # Float element
>>> c = (1, 2) # Tuple element with 2 components
>>> d = {""a"": (2, 2), ""b"": 3} # Dict element with 3 components
>>> Point = collections.namedtuple(""Point"", [""x"", ""y""]) # doctest: +SKIP
>>> e = Point(1, 2) # Named tuple # doctest: +SKIP
>>> f = tf.data.Dataset.range(10) # Dataset element"
1550,options,tensorflow/tensorflow/python/data/ops/dataset_ops.py,348,method,"Returns the options for this dataset and its inputs.
Returns:
A `tf.data.Options` object representing the dataset options."
1551,element_spec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,451,method,"The type specification of an element of this dataset.
>>> dataset = tf.data.Dataset.from_tensor_slices([1, 2, 3])
>>> dataset.element_spec
TensorSpec(shape=(), dtype=tf.int32, name=None)
Returns:
A nested structure of `tf.TypeSpec` objects matching the structure of an
element of this dataset and specifying the type of individual components."
1552,as_numpy_iterator,tensorflow/tensorflow/python/data/ops/dataset_ops.py,472,method,"Returns an iterator which converts all elements of the dataset to numpy.
Use `as_numpy_iterator` to inspect the content of your dataset. To see
element shapes and types, print dataset elements directly instead of using
`as_numpy_iterator`.
>>> dataset = tf.data.Dataset.from_tensor_slices([1, 2, 3])
>>> for element in dataset:
... print(element)
tf.Tensor(1, shape=(), dtype=int32)
tf.Tensor(2, shape=(), dtype=int32)
tf.Tensor(3, shape=(), dtype=int32)
This method requires that you are running in eager mode and the dataset's
element_spec contains only `TensorSpec` components.
>>> dataset = tf.data.Dataset.from_tensor_slices([1, 2, 3])
>>> for element in dataset.as_numpy_iterator():
... print(element)
1
2
3
>>> dataset = tf.data.Dataset.from_tensor_slices([1, 2, 3])
>>> print(list(dataset.as_numpy_iterator()))
[1, 2, 3]
`as_numpy_iterator()` will preserve the nested structure of dataset
elements.
>>> dataset = tf.data.Dataset.from_tensor_slices({'a': ([1, 2], [3, 4]),
... 'b': [5, 6]})
>>> list(dataset.as_numpy_iterator()) == [{'a': (1, 3), 'b': 5},
... {'a': (2, 4), 'b': 6}]
True
Returns:
An iterable over the elements of the dataset, with their tensors converted
to numpy arrays.
Raises:
TypeError: if an element contains a non-`Tensor` value.
RuntimeError: if eager execution is not enabled."
1553,from_tensors,tensorflow/tensorflow/python/data/ops/dataset_ops.py,570,method,"Creates a `Dataset` with a single element, comprising the given tensors.
`from_tensors` produces a dataset containing only a single element. To slice
the input tensor into multiple elements, use `from_tensor_slices` instead.
>>> dataset = tf.data.Dataset.from_tensors([1, 2, 3])
>>> list(dataset.as_numpy_iterator())
[array([1, 2, 3], dtype=int32)]
>>> dataset = tf.data.Dataset.from_tensors(([1, 2, 3], 'A'))
>>> list(dataset.as_numpy_iterator())
[(array([1, 2, 3], dtype=int32), b'A')]
>>> # You can use `from_tensors` to produce a dataset which repeats
>>> # the same example many times.
>>> example = tf.constant([1,2,3])
>>> dataset = tf.data.Dataset.from_tensors(example).repeat(2)
>>> list(dataset.as_numpy_iterator())
[array([1, 2, 3], dtype=int32), array([1, 2, 3], dtype=int32)]
Note that if `tensors` contains a NumPy array, and eager execution is not
enabled, the values will be embedded in the graph as one or more
`tf.constant` operations. For large datasets (> 1 GB), this can waste
memory and run into byte limits of graph serialization. If `tensors`
contains one or more large NumPy arrays, consider the alternative described
in [this
guide](https://tensorflow.org/guide/data#consuming_numpy_arrays).
Args:
tensors: A dataset element.
Returns:
Dataset: A `Dataset`."
1554,from_tensor_slices,tensorflow/tensorflow/python/data/ops/dataset_ops.py,607,method,"Creates a `Dataset` whose elements are slices of the given tensors.
The given tensors are sliced along their first dimension. This operation
preserves the structure of the input tensors, removing the first dimension
of each tensor and using it as the dataset dimension. All input tensors
must have the same size in their first dimensions.
>>> # Slicing a 1D tensor produces scalar tensor elements.
>>> dataset = tf.data.Dataset.from_tensor_slices([1, 2, 3])
>>> list(dataset.as_numpy_iterator())
[1, 2, 3]
>>> # Slicing a 2D tensor produces 1D tensor elements.
>>> dataset = tf.data.Dataset.from_tensor_slices([[1, 2], [3, 4]])
>>> list(dataset.as_numpy_iterator())
[array([1, 2], dtype=int32), array([3, 4], dtype=int32)]
>>> # Slicing a tuple of 1D tensors produces tuple elements containing
>>> # scalar tensors.
>>> dataset = tf.data.Dataset.from_tensor_slices(([1, 2], [3, 4], [5, 6]))
>>> list(dataset.as_numpy_iterator())
[(1, 3, 5), (2, 4, 6)]
>>> # Dictionary structure is also preserved.
>>> dataset = tf.data.Dataset.from_tensor_slices({""a"": [1, 2], ""b"": [3, 4]})
>>> list(dataset.as_numpy_iterator()) == [{'a': 1, 'b': 3},
... {'a': 2, 'b': 4}]
True
>>> # Two tensors can be combined into one Dataset object.
>>> features = tf.constant([[1, 3], [2, 1], [3, 3]]) # ==> 3x2 tensor
>>> labels = tf.constant(['A', 'B', 'A']) # ==> 3x1 tensor
>>> dataset = Dataset.from_tensor_slices((features, labels))
>>> # Both the features and the labels tensors can be converted
>>> # to a Dataset object separately and combined after.
>>> features_dataset = Dataset.from_tensor_slices(features)
>>> labels_dataset = Dataset.from_tensor_slices(labels)
>>> dataset = Dataset.zip((features_dataset, labels_dataset))
>>> # A batched feature and label set can be converted to a Dataset
>>> # in similar fashion.
>>> batched_features = tf.constant([[[1, 3], [2, 3]],
... [[2, 1], [1, 2]],
... [[3, 3], [3, 2]]], shape=(3, 2, 2))
>>> batched_labels = tf.constant([['A', 'A'],
... ['B', 'B'],
... ['A', 'B']], shape=(3, 2, 1))
>>> dataset = Dataset.from_tensor_slices((batched_features, batched_labels))
>>> for element in dataset.as_numpy_iterator():
... print(element)
(array([[1, 3],
[2, 3]], dtype=int32), array([[b'A'],
[b'A']], dtype=object))
(array([[2, 1],
[1, 2]], dtype=int32), array([[b'B'],
[b'B']], dtype=object))
(array([[3, 3],
[3, 2]], dtype=int32), array([[b'A'],
[b'B']], dtype=object))
Note that if `tensors` contains a NumPy array, and eager execution is not
enabled, the values will be embedded in the graph as one or more
`tf.constant` operations. For large datasets (> 1 GB), this can waste
memory and run into byte limits of graph serialization. If `tensors`
contains one or more large NumPy arrays, consider the alternative described
in [this guide](
https://tensorflow.org/guide/data#consuming_numpy_arrays).
Args:
tensors: A dataset element, with each component having the same size in
the first dimension.
Returns:
Dataset: A `Dataset`."
1555,from_generator,tensorflow/tensorflow/python/data/ops/dataset_ops.py,721,method,"Creates a `Dataset` whose elements are generated by `generator`.
The `generator` argument must be a callable object that returns
an object that supports the `iter()` protocol (e.g. a generator function).
The elements generated by `generator` must be compatible with the given
`output_types` and (optional) `output_shapes` arguments.
>>> import itertools
>>>
>>> def gen():
... for i in itertools.count(1):
... yield (i, [1] * i)
>>>
>>> dataset = tf.data.Dataset.from_generator(
... gen,
... (tf.int64, tf.int64),
... (tf.TensorShape([]), tf.TensorShape([None])))
>>>
>>> list(dataset.take(3).as_numpy_iterator())
[(1, array([1])), (2, array([1, 1])), (3, array([1, 1, 1]))]
Note: The current implementation of `Dataset.from_generator()` uses
`tf.numpy_function` and inherits the same constraints. In particular, it
requires the dataset and iterator related operations to be placed
on a device in the same process as the Python program that called
`Dataset.from_generator()`. The body of `generator` will not be
serialized in a `GraphDef`, and you should not use this method if you
need to serialize your model and restore it in a different environment.
Note: If `generator` depends on mutable global variables or other external
state, be aware that the runtime may invoke `generator` multiple times
(in order to support repeating the `Dataset`) and at any time
between the call to `Dataset.from_generator()` and the production of the
first element from the generator. Mutating global variables or external
state can cause undefined behavior, and we recommend that you explicitly
cache any external state in `generator` before calling
`Dataset.from_generator()`.
Args:
generator: A callable object that returns an object that supports the
`iter()` protocol. If `args` is not specified, `generator` must take no
arguments; otherwise it must take as many arguments as there are values
in `args`.
output_types: A nested structure of `tf.DType` objects corresponding to
each component of an element yielded by `generator`.
output_shapes: (Optional.) A nested structure of `tf.TensorShape` objects
corresponding to each component of an element yielded by `generator`.
args: (Optional.) A tuple of `tf.Tensor` objects that will be evaluated
and passed to `generator` as NumPy-array arguments.
Returns:
Dataset: A `Dataset`."
1556,range,tensorflow/tensorflow/python/data/ops/dataset_ops.py,921,method,"Creates a `Dataset` of a step-separated range of values.
>>> list(Dataset.range(5).as_numpy_iterator())
[0, 1, 2, 3, 4]
>>> list(Dataset.range(2, 5).as_numpy_iterator())
[2, 3, 4]
>>> list(Dataset.range(1, 5, 2).as_numpy_iterator())
[1, 3]
>>> list(Dataset.range(1, 5, -2).as_numpy_iterator())
[]
>>> list(Dataset.range(5, 1).as_numpy_iterator())
[]
>>> list(Dataset.range(5, 1, -2).as_numpy_iterator())
[5, 3]
>>> list(Dataset.range(2, 5, output_type=tf.int32).as_numpy_iterator())
[2, 3, 4]
>>> list(Dataset.range(1, 5, 2, output_type=tf.float32).as_numpy_iterator())
[1.0, 3.0]
Args:
*args: follows the same semantics as python's xrange.
len(args) == 1 -> start = 0, stop = args[0], step = 1.
len(args) == 2 -> start = args[0], stop = args[1], step = 1.
len(args) == 3 -> start = args[0], stop = args[1], step = args[2].
**kwargs:
- output_type: Its expected dtype. (Optional, default: `tf.int64`).
Returns:
Dataset: A `RangeDataset`.
Raises:
ValueError: if len(args) == 0."
1557,zip,tensorflow/tensorflow/python/data/ops/dataset_ops.py,958,method,"Creates a `Dataset` by zipping together the given datasets.
This method has similar semantics to the built-in `zip()` function
in Python, with the main difference being that the `datasets`
argument can be an arbitrary nested structure of `Dataset` objects.
>>> # The nested structure of the `datasets` argument determines the
>>> # structure of elements in the resulting dataset.
>>> a = tf.data.Dataset.range(1, 4) # ==> [ 1, 2, 3 ]
>>> b = tf.data.Dataset.range(4, 7) # ==> [ 4, 5, 6 ]
>>> ds = tf.data.Dataset.zip((a, b))
>>> list(ds.as_numpy_iterator())
[(1, 4), (2, 5), (3, 6)]
>>> ds = tf.data.Dataset.zip((b, a))
>>> list(ds.as_numpy_iterator())
[(4, 1), (5, 2), (6, 3)]
>>>
>>> # The `datasets` argument may contain an arbitrary number of datasets.
>>> c = tf.data.Dataset.range(7, 13).batch(2) # ==> [ [7, 8],
... # [9, 10],
... # [11, 12] ]
>>> ds = tf.data.Dataset.zip((a, b, c))
>>> for element in ds.as_numpy_iterator():
... print(element)
(1, 4, array([7, 8]))
(2, 5, array([ 9, 10]))
(3, 6, array([11, 12]))
>>>
>>> # The number of elements in the resulting dataset is the same as
>>> # the size of the smallest dataset in `datasets`.
>>> d = tf.data.Dataset.range(13, 15) # ==> [ 13, 14 ]
>>> ds = tf.data.Dataset.zip((a, d))
>>> list(ds.as_numpy_iterator())
[(1, 13), (2, 14)]
Args:
datasets: A nested structure of datasets.
Returns:
Dataset: A `Dataset`."
1558,concatenate,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1002,method,"Creates a `Dataset` by concatenating the given dataset with this dataset.
>>> a = tf.data.Dataset.range(1, 4) # ==> [ 1, 2, 3 ]
>>> b = tf.data.Dataset.range(4, 8) # ==> [ 4, 5, 6, 7 ]
>>> ds = a.concatenate(b)
>>> list(ds.as_numpy_iterator())
[1, 2, 3, 4, 5, 6, 7]
>>> # The input dataset and dataset to be concatenated should have the same
>>> # nested structures and output types.
>>> c = tf.data.Dataset.zip((a, b))
>>> a.concatenate(c)
Traceback (most recent call last):
TypeError: Two datasets to concatenate have different types
<dtype: 'int64'> and (tf.int64, tf.int64)
>>> d = tf.data.Dataset.from_tensor_slices([""a"", ""b"", ""c""])
>>> a.concatenate(d)
Traceback (most recent call last):
TypeError: Two datasets to concatenate have different types
<dtype: 'int64'> and <dtype: 'string'>
Args:
dataset: `Dataset` to be concatenated.
Returns:
Dataset: A `Dataset`."
1559,prefetch,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1031,method,"Creates a `Dataset` that prefetches elements from this dataset.
Most dataset input pipelines should end with a call to `prefetch`. This
allows later elements to be prepared while the current element is being
processed. This often improves latency and throughput, at the cost of
using additional memory to store prefetched elements.
Note: Like other `Dataset` methods, prefetch operates on the
elements of the input dataset. It has no concept of examples vs. batches.
`examples.prefetch(2)` will prefetch two elements (2 examples),
while `examples.batch(20).prefetch(2)` will prefetch 2 elements
(2 batches, of 20 examples each).
>>> dataset = tf.data.Dataset.range(3)
>>> dataset = dataset.prefetch(2)
>>> list(dataset.as_numpy_iterator())
[0, 1, 2]
Args:
buffer_size: A `tf.int64` scalar `tf.Tensor`, representing the maximum
number of elements that will be buffered when prefetching.
Returns:
Dataset: A `Dataset`."
1560,list_files,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1060,method,"A dataset of all files matching one or more glob patterns.
The `file_pattern` argument should be a small number of glob patterns.
If your filenames have already been globbed, use
`Dataset.from_tensor_slices(filenames)` instead, as re-globbing every
filename with `list_files` may result in poor performance with remote
storage systems.
Note: The default behavior of this method is to return filenames in
a non-deterministic random shuffled order. Pass a `seed` or `shuffle=False`
to get results in a deterministic order.
Example:
If we had the following files on our filesystem:
- /path/to/dir/a.txt
- /path/to/dir/b.py
- /path/to/dir/c.py
If we pass ""/path/to/dir/*.py"" as the directory, the dataset
would produce:
- /path/to/dir/b.py
- /path/to/dir/c.py
Args:
file_pattern: A string, a list of strings, or a `tf.Tensor` of string type
(scalar or vector), representing the filename glob (i.e. shell wildcard)
pattern(s) that will be matched.
shuffle: (Optional.) If `True`, the file names will be shuffled randomly.
Defaults to `True`.
seed: (Optional.) A `tf.int64` scalar `tf.Tensor`, representing the random
seed that will be used to create the distribution. See
`tf.random.set_seed` for behavior.
Returns:
Dataset: A `Dataset` of strings corresponding to file names."
1561,repeat,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1128,method,"Repeats this dataset so each original value is seen `count` times.
>>> dataset = tf.data.Dataset.from_tensor_slices([1, 2, 3])
>>> dataset = dataset.repeat(3)
>>> list(dataset.as_numpy_iterator())
[1, 2, 3, 1, 2, 3, 1, 2, 3]
Note: If this dataset is a function of global state (e.g. a random number
generator), then different repetitions may produce different elements.
Args:
count: (Optional.) A `tf.int64` scalar `tf.Tensor`, representing the
number of times the dataset should be repeated. The default behavior (if
`count` is `None` or `-1`) is for the dataset be repeated indefinitely.
Returns:
Dataset: A `Dataset`."
1562,enumerate,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1149,method,"Enumerates the elements of this dataset.
It is similar to python's `enumerate`.
>>> dataset = tf.data.Dataset.from_tensor_slices([1, 2, 3])
>>> dataset = dataset.enumerate(start=5)
>>> for element in dataset.as_numpy_iterator():
... print(element)
(5, 1)
(6, 2)
(7, 3)
>>> # The nested structure of the input dataset determines the structure of
>>> # elements in the resulting dataset.
>>> dataset = tf.data.Dataset.from_tensor_slices([(7, 8), (9, 10)])
>>> dataset = dataset.enumerate()
>>> for element in dataset.as_numpy_iterator():
... print(element)
(0, array([7, 8], dtype=int32))
(1, array([ 9, 10], dtype=int32))
Args:
start: A `tf.int64` scalar `tf.Tensor`, representing the start value for
enumeration.
Returns:
Dataset: A `Dataset`."
1563,shuffle,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1182,method,"Randomly shuffles the elements of this dataset.
This dataset fills a buffer with `buffer_size` elements, then randomly
samples elements from this buffer, replacing the selected elements with new
elements. For perfect shuffling, a buffer size greater than or equal to the
full size of the dataset is required.
For instance, if your dataset contains 10,000 elements but `buffer_size` is
set to 1,000, then `shuffle` will initially select a random element from
only the first 1,000 elements in the buffer. Once an element is selected,
its space in the buffer is replaced by the next (i.e. 1,001-st) element,
maintaining the 1,000 element buffer.
`reshuffle_each_iteration` controls whether the shuffle order should be
different for each epoch. In TF 1.X, the idiomatic way to create epochs
was through the `repeat` transformation:
>>> dataset = tf.data.Dataset.range(3)
>>> dataset = dataset.shuffle(3, reshuffle_each_iteration=True)
>>> dataset = dataset.repeat(2) # doctest: +SKIP
[1, 0, 2, 1, 2, 0]
>>> dataset = tf.data.Dataset.range(3)
>>> dataset = dataset.shuffle(3, reshuffle_each_iteration=False)
>>> dataset = dataset.repeat(2) # doctest: +SKIP
[1, 0, 2, 1, 0, 2]
In TF 2.0, `tf.data.Dataset` objects are Python iterables which makes it
possible to also create epochs through Python iteration:
>>> dataset = tf.data.Dataset.range(3)
>>> dataset = dataset.shuffle(3, reshuffle_each_iteration=True)
>>> list(dataset.as_numpy_iterator()) # doctest: +SKIP
[1, 0, 2]
>>> list(dataset.as_numpy_iterator()) # doctest: +SKIP
[1, 2, 0]
>>> dataset = tf.data.Dataset.range(3)
>>> dataset = dataset.shuffle(3, reshuffle_each_iteration=False)
>>> list(dataset.as_numpy_iterator()) # doctest: +SKIP
[1, 0, 2]
>>> list(dataset.as_numpy_iterator()) # doctest: +SKIP
[1, 0, 2]
Args:
buffer_size: A `tf.int64` scalar `tf.Tensor`, representing the number of
elements from this dataset from which the new dataset will sample.
seed: (Optional.) A `tf.int64` scalar `tf.Tensor`, representing the random
seed that will be used to create the distribution. See
`tf.random.set_seed` for behavior.
reshuffle_each_iteration: (Optional.) A boolean, which if true indicates
that the dataset should be pseudorandomly reshuffled each time it is
iterated over. (Defaults to `True`.)
Returns:
Dataset: A `Dataset`."
1564,cache,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1242,method,"Caches the elements in this dataset.
The first time the dataset is iterated over, its elements will be cached
either in the specified file or in memory. Subsequent iterations will
use the cached data.
Note: For the cache to be finalized, the input dataset must be iterated
through in its entirety. Otherwise, subsequent iterations will not use
cached data.
>>> dataset = tf.data.Dataset.range(5)
>>> dataset = dataset.map(lambda x: x**2)
>>> dataset = dataset.cache()
>>> # The first time reading through the data will generate the data using
>>> # `range` and `map`.
>>> list(dataset.as_numpy_iterator())
[0, 1, 4, 9, 16]
>>> # Subsequent iterations read from the cache.
>>> list(dataset.as_numpy_iterator())
[0, 1, 4, 9, 16]
When caching to a file, the cached data will persist across runs. Even the
first iteration through the data will read from the cache file. Changing
the input pipeline before the call to `.cache()` will have no effect until
the cache file is removed or the filename is changed.
>>> dataset = tf.data.Dataset.range(5)
>>> dataset = dataset.cache(""/path/to/file"") # doctest: +SKIP
>>> list(dataset.as_numpy_iterator()) # doctest: +SKIP
[0, 1, 2, 3, 4]
>>> dataset = tf.data.Dataset.range(10)
>>> dataset = dataset.cache(""/path/to/file"") # Same file! # doctest: +SKIP
>>> list(dataset.as_numpy_iterator()) # doctest: +SKIP
[0, 1, 2, 3, 4]
Note: `cache` will produce exactly the same elements during each iteration
through the dataset. If you wish to randomize the iteration order, make sure
to call `shuffle` *after* calling `cache`.
Args:
filename: A `tf.string` scalar `tf.Tensor`, representing the name of a
directory on the filesystem to use for caching elements in this Dataset.
If a filename is not provided, the dataset will be cached in memory.
Returns:
Dataset: A `Dataset`."
1565,take,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1292,method,"Creates a `Dataset` with at most `count` elements from this dataset.
>>> dataset = tf.data.Dataset.range(10)
>>> dataset = dataset.take(3)
>>> list(dataset.as_numpy_iterator())
[0, 1, 2]
Args:
count: A `tf.int64` scalar `tf.Tensor`, representing the number of
elements of this dataset that should be taken to form the new dataset.
If `count` is -1, or if `count` is greater than the size of this
dataset, the new dataset will contain all elements of this dataset.
Returns:
Dataset: A `Dataset`."
1566,skip,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1311,method,"Creates a `Dataset` that skips `count` elements from this dataset.
>>> dataset = tf.data.Dataset.range(10)
>>> dataset = dataset.skip(7)
>>> list(dataset.as_numpy_iterator())
[7, 8, 9]
Args:
count: A `tf.int64` scalar `tf.Tensor`, representing the number of
elements of this dataset that should be skipped to form the new dataset.
If `count` is greater than the size of this dataset, the new dataset
will contain no elements. If `count` is -1, skips the entire dataset.
Returns:
Dataset: A `Dataset`."
1567,shard,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1330,method,"Creates a `Dataset` that includes only 1/`num_shards` of this dataset.
`shard` is deterministic. The Dataset produced by `A.shard(n, i)` will
contain all elements of A whose index mod n = i.
>>> A = tf.data.Dataset.range(10)
>>> B = A.shard(num_shards=3, index=0)
>>> list(B.as_numpy_iterator())
[0, 3, 6, 9]
>>> C = A.shard(num_shards=3, index=1)
>>> list(C.as_numpy_iterator())
[1, 4, 7]
>>> D = A.shard(num_shards=3, index=2)
>>> list(D.as_numpy_iterator())
[2, 5, 8]
This dataset operator is very useful when running distributed training, as
it allows each worker to read a unique subset.
When reading a single input file, you can shard elements as follows:
```python
d = tf.data.TFRecordDataset(input_file)
d = d.shard(num_workers, worker_index)
d = d.repeat(num_epochs)
d = d.shuffle(shuffle_buffer_size)
d = d.map(parser_fn, num_parallel_calls=num_map_threads)
```
Important caveats:
- Be sure to shard before you use any randomizing operator (such as
shuffle).
- Generally it is best if the shard operator is used early in the dataset
pipeline. For example, when reading from a set of TFRecord files, shard
before converting the dataset to input samples. This avoids reading every
file on every worker. The following is an example of an efficient
sharding strategy within a complete pipeline:
```python
d = Dataset.list_files(pattern)
d = d.shard(num_workers, worker_index)
d = d.repeat(num_epochs)
d = d.shuffle(shuffle_buffer_size)
d = d.interleave(tf.data.TFRecordDataset,
cycle_length=num_readers, block_length=1)
d = d.map(parser_fn, num_parallel_calls=num_map_threads)
```
Args:
num_shards: A `tf.int64` scalar `tf.Tensor`, representing the number of
shards operating in parallel.
index: A `tf.int64` scalar `tf.Tensor`, representing the worker index.
Returns:
Dataset: A `Dataset`.
Raises:
InvalidArgumentError: if `num_shards` or `index` are illegal values.
Note: error checking is done on a best-effort basis, and errors aren't
guaranteed to be caught upon dataset creation. (e.g. providing in a
placeholder tensor bypasses the early checking, and will instead result
in an error during a session.run call.)"
1568,batch,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1398,method,"Combines consecutive elements of this dataset into batches.
>>> dataset = tf.data.Dataset.range(8)
>>> dataset = dataset.batch(3)
>>> list(dataset.as_numpy_iterator())
[array([0, 1, 2]), array([3, 4, 5]), array([6, 7])]
>>> dataset = tf.data.Dataset.range(8)
>>> dataset = dataset.batch(3, drop_remainder=True)
>>> list(dataset.as_numpy_iterator())
[array([0, 1, 2]), array([3, 4, 5])]
The components of the resulting element will have an additional outer
dimension, which will be `batch_size` (or `N % batch_size` for the last
element if `batch_size` does not divide the number of input elements `N`
evenly and `drop_remainder` is `False`). If your program depends on the
batches having the same outer dimension, you should set the `drop_remainder`
argument to `True` to prevent the smaller batch from being produced.
Args:
batch_size: A `tf.int64` scalar `tf.Tensor`, representing the number of
consecutive elements of this dataset to combine in a single batch.
drop_remainder: (Optional.) A `tf.bool` scalar `tf.Tensor`, representing
whether the last batch should be dropped in the case it has fewer than
`batch_size` elements; the default behavior is not to drop the smaller
batch.
Returns:
Dataset: A `Dataset`."
1569,padded_batch,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1431,method,"Combines consecutive elements of this dataset into padded batches.
This transformation combines multiple consecutive elements of the input
dataset into a single element.
Like `tf.data.Dataset.batch`, the components of the resulting element will
have an additional outer dimension, which will be `batch_size` (or
`N % batch_size` for the last element if `batch_size` does not divide the
number of input elements `N` evenly and `drop_remainder` is `False`). If
your program depends on the batches having the same outer dimension, you
should set the `drop_remainder` argument to `True` to prevent the smaller
batch from being produced.
Unlike `tf.data.Dataset.batch`, the input elements to be batched may have
different shapes, and this transformation will pad each component to the
respective shape in `padded_shapes`. The `padded_shapes` argument
determines the resulting shape for each dimension of each component in an
output element:
* If the dimension is a constant, the component will be padded out to that
length in that dimension.
* If the dimension is unknown, the component will be padded out to the
maximum length of all elements in that dimension.
>>> A = (tf.data.Dataset
... .range(1, 5, output_type=tf.int32)
... .map(lambda x: tf.fill([x], x)))
>>> # Pad to the smallest per-batch size that fits all elements.
>>> B = A.padded_batch(2)
>>> for element in B.as_numpy_iterator():
... print(element)
[[1 0]
[2 2]]
[[3 3 3 0]
[4 4 4 4]]
>>> # Pad to a fixed size.
>>> C = A.padded_batch(2, padded_shapes=5)
>>> for element in C.as_numpy_iterator():
... print(element)
[[1 0 0 0 0]
[2 2 0 0 0]]
[[3 3 3 0 0]
[4 4 4 4 0]]
>>> # Pad with a custom value.
>>> D = A.padded_batch(2, padded_shapes=5, padding_values=-1)
>>> for element in D.as_numpy_iterator():
... print(element)
[[ 1 -1 -1 -1 -1]
[ 2 2 -1 -1 -1]]
[[ 3 3 3 -1 -1]
[ 4 4 4 4 -1]]
>>> # Components of nested elements can be padded independently.
>>> elements = [([1, 2, 3], [10]),
... ([4, 5], [11, 12])]
>>> dataset = tf.data.Dataset.from_generator(
... lambda: iter(elements), (tf.int32, tf.int32))
>>> # Pad the first component of the tuple to length 4, and the second
>>> # component to the smallest size that fits.
>>> dataset = dataset.padded_batch(2,
... padded_shapes=([4], [None]),
... padding_values=(-1, 100))
>>> list(dataset.as_numpy_iterator())
[(array([[ 1, 2, 3, -1], [ 4, 5, -1, -1]], dtype=int32),
array([[ 10, 100], [ 11, 12]], dtype=int32))]
>>> # Pad with a single value and multiple components.
>>> E = tf.data.Dataset.zip((A, A)).padded_batch(2, padding_values=-1)
>>> for element in E.as_numpy_iterator():
... print(element)
(array([[ 1, -1],
[ 2, 2]], dtype=int32), array([[ 1, -1],
[ 2, 2]], dtype=int32))
(array([[ 3, 3, 3, -1],
[ 4, 4, 4, 4]], dtype=int32), array([[ 3, 3, 3, -1],
[ 4, 4, 4, 4]], dtype=int32))
See also `tf.data.experimental.dense_to_sparse_batch`, which combines
elements that may have different shapes into a `tf.sparse.SparseTensor`.
Args:
batch_size: A `tf.int64` scalar `tf.Tensor`, representing the number of
consecutive elements of this dataset to combine in a single batch.
padded_shapes: (Optional.) A nested structure of `tf.TensorShape` or
`tf.int64` vector tensor-like objects representing the shape to which
the respective component of each input element should be padded prior
to batching. Any unknown dimensions will be padded to the maximum size
of that dimension in each batch. If unset, all dimensions of all
components are padded to the maximum size in the batch. `padded_shapes`
must be set if any component has an unknown rank.
padding_values: (Optional.) A nested structure of scalar-shaped
`tf.Tensor`, representing the padding values to use for the respective
components. None represents that the nested structure should be padded
with default values. Defaults are `0` for numeric types and the empty
string for string types. The `padding_values` should have the
same structure as the input dataset. If `padding_values` is a single
element and the input dataset has multiple components, then the same
`padding_values` will be used to pad every component of the dataset.
If `padding_values` is a scalar, then its value will be broadcasted
to match the shape of each component.
drop_remainder: (Optional.) A `tf.bool` scalar `tf.Tensor`, representing
whether the last batch should be dropped in the case it has fewer than
`batch_size` elements; the default behavior is not to drop the smaller
batch.
Returns:
Dataset: A `Dataset`.
Raises:
ValueError: If a component has an unknown rank, and the `padded_shapes`
argument is not set."
1570,map,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1557,method,"Maps `map_func` across the elements of this dataset.
This transformation applies `map_func` to each element of this dataset, and
returns a new dataset containing the transformed elements, in the same
order as they appeared in the input. `map_func` can be used to change both
the values and the structure of a dataset's elements. For example, adding 1
to each element, or projecting a subset of element components.
>>> dataset = Dataset.range(1, 6) # ==> [ 1, 2, 3, 4, 5 ]
>>> dataset = dataset.map(lambda x: x + 1)
>>> list(dataset.as_numpy_iterator())
[2, 3, 4, 5, 6]
The input signature of `map_func` is determined by the structure of each
element in this dataset.
>>> dataset = Dataset.range(5)
>>> # `map_func` takes a single argument of type `tf.Tensor` with the same
>>> # shape and dtype.
>>> result = dataset.map(lambda x: x + 1)
>>> # Each element is a tuple containing two `tf.Tensor` objects.
>>> elements = [(1, ""foo""), (2, ""bar""), (3, ""baz"")]
>>> dataset = tf.data.Dataset.from_generator(
... lambda: elements, (tf.int32, tf.string))
>>> # `map_func` takes two arguments of type `tf.Tensor`. This function
>>> # projects out just the first component.
>>> result = dataset.map(lambda x_int, y_str: x_int)
>>> list(result.as_numpy_iterator())
[1, 2, 3]
>>> # Each element is a dictionary mapping strings to `tf.Tensor` objects.
>>> elements = ([{""a"": 1, ""b"": ""foo""},
... {""a"": 2, ""b"": ""bar""},
... {""a"": 3, ""b"": ""baz""}])
>>> dataset = tf.data.Dataset.from_generator(
... lambda: elements, {""a"": tf.int32, ""b"": tf.string})
>>> # `map_func` takes a single argument of type `dict` with the same keys
>>> # as the elements.
>>> result = dataset.map(lambda d: str(d[""a""]) + d[""b""])
The value or values returned by `map_func` determine the structure of each
element in the returned dataset.
>>> dataset = tf.data.Dataset.range(3)
>>> # `map_func` returns two `tf.Tensor` objects.
>>> def g(x):
... return tf.constant(37.0), tf.constant([""Foo"", ""Bar"", ""Baz""])
>>> result = dataset.map(g)
>>> result.element_spec
(TensorSpec(shape=(), dtype=tf.float32, name=None), TensorSpec(shape=(3,), dtype=tf.string, name=None))
>>> # Python primitives, lists, and NumPy arrays are implicitly converted to
>>> # `tf.Tensor`.
>>> def h(x):
... return 37.0, [""Foo"", ""Bar""], np.array([1.0, 2.0], dtype=np.float64)
>>> result = dataset.map(h)
>>> result.element_spec
(TensorSpec(shape=(), dtype=tf.float32, name=None), TensorSpec(shape=(2,), dtype=tf.string, name=None), TensorSpec(shape=(2,), dtype=tf.float64, name=None))
>>> # `map_func` can return nested structures.
>>> def i(x):
... return (37.0, [42, 16]), ""foo""
>>> result = dataset.map(i)
>>> result.element_spec
((TensorSpec(shape=(), dtype=tf.float32, name=None),
TensorSpec(shape=(2,), dtype=tf.int32, name=None)),
TensorSpec(shape=(), dtype=tf.string, name=None))
`map_func` can accept as arguments and return any type of dataset element.
Note that irrespective of the context in which `map_func` is defined (eager
vs. graph), tf.data traces the function and executes it as a graph. To use
Python code inside of the function you have a few options:
1) Rely on AutoGraph to convert Python code into an equivalent graph
computation. The downside of this approach is that AutoGraph can convert
some but not all Python code.
2) Use `tf.py_function`, which allows you to write arbitrary Python code but
will generally result in worse performance than 1). For example:
>>> d = tf.data.Dataset.from_tensor_slices(['hello', 'world'])
>>> # transform a string tensor to upper case string using a Python function
>>> def upper_case_fn(t: tf.Tensor):
... return t.numpy().decode('utf-8').upper()
>>> d = d.map(lambda x: tf.py_function(func=upper_case_fn,
... inp=[x], Tout=tf.string))
>>> list(d.as_numpy_iterator())
[b'HELLO', b'WORLD']
3) Use `tf.numpy_function`, which also allows you to write arbitrary
Python code. Note that `tf.py_function` accepts `tf.Tensor` whereas
`tf.numpy_function` accepts numpy arrays and returns only numpy arrays.
For example:
>>> d = tf.data.Dataset.from_tensor_slices(['hello', 'world'])
>>> def upper_case_fn(t: np.ndarray):
... return t.decode('utf-8').upper()
>>> d = d.map(lambda x: tf.numpy_function(func=upper_case_fn,
... inp=[x], Tout=tf.string))
>>> list(d.as_numpy_iterator())
[b'HELLO', b'WORLD']
Note that the use of `tf.numpy_function` and `tf.py_function`
in general precludes the possibility of executing user-defined
transformations in parallel (because of Python GIL).
Performance can often be improved by setting `num_parallel_calls` so that
`map` will use multiple threads to process elements. If deterministic order
isn't required, it can also improve performance to set
`deterministic=False`.
>>> dataset = Dataset.range(1, 6) # ==> [ 1, 2, 3, 4, 5 ]
>>> dataset = dataset.map(lambda x: x + 1,
... num_parallel_calls=tf.data.experimental.AUTOTUNE,
... deterministic=False)
Args:
map_func: A function mapping a dataset element to another dataset element.
num_parallel_calls: (Optional.) A `tf.int32` scalar `tf.Tensor`,
representing the number elements to process asynchronously in parallel.
If not specified, elements will be processed sequentially. If the value
`tf.data.experimental.AUTOTUNE` is used, then the number of parallel
calls is set dynamically based on available CPU.
deterministic: (Optional.) A boolean controlling whether determinism
should be traded for performance by allowing elements to be produced out
of order. If `deterministic` is `None`, the
`tf.data.Options.experimental_deterministic` dataset option (`True` by
default) is used to decide whether to produce elements
deterministically.
Returns:
Dataset: A `Dataset`."
1571,flat_map,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1704,method,"Maps `map_func` across this dataset and flattens the result.
Use `flat_map` if you want to make sure that the order of your dataset
stays the same. For example, to flatten a dataset of batches into a
dataset of their elements:
>>> dataset = tf.data.Dataset.from_tensor_slices(
... [[1, 2, 3], [4, 5, 6], [7, 8, 9]])
>>> dataset = dataset.flat_map(lambda x: Dataset.from_tensor_slices(x))
>>> list(dataset.as_numpy_iterator())
[1, 2, 3, 4, 5, 6, 7, 8, 9]
`tf.data.Dataset.interleave()` is a generalization of `flat_map`, since
`flat_map` produces the same output as
`tf.data.Dataset.interleave(cycle_length=1)`
Args:
map_func: A function mapping a dataset element to a dataset.
Returns:
Dataset: A `Dataset`."
1572,interleave,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1729,method,"Maps `map_func` across this dataset, and interleaves the results.
For example, you can use `Dataset.interleave()` to process many input files
concurrently:
>>> # Preprocess 4 files concurrently, and interleave blocks of 16 records
>>> # from each file.
>>> filenames = [""/var/data/file1.txt"", ""/var/data/file2.txt"",
... ""/var/data/file3.txt"", ""/var/data/file4.txt""]
>>> dataset = tf.data.Dataset.from_tensor_slices(filenames)
>>> def parse_fn(filename):
... return tf.data.Dataset.range(10)
>>> dataset = dataset.interleave(lambda x:
... tf.data.TextLineDataset(x).map(parse_fn, num_parallel_calls=1),
... cycle_length=4, block_length=16)
The `cycle_length` and `block_length` arguments control the order in which
elements are produced. `cycle_length` controls the number of input elements
that are processed concurrently. If you set `cycle_length` to 1, this
transformation will handle one input element at a time, and will produce
identical results to `tf.data.Dataset.flat_map`. In general,
this transformation will apply `map_func` to `cycle_length` input elements,
open iterators on the returned `Dataset` objects, and cycle through them
producing `block_length` consecutive elements from each iterator, and
consuming the next input element each time it reaches the end of an
iterator.
For example:
>>> dataset = Dataset.range(1, 6) # ==> [ 1, 2, 3, 4, 5 ]
>>> # NOTE: New lines indicate ""block"" boundaries.
>>> dataset = dataset.interleave(
... lambda x: Dataset.from_tensors(x).repeat(6),
... cycle_length=2, block_length=4)
>>> list(dataset.as_numpy_iterator())
[1, 1, 1, 1,
2, 2, 2, 2,
1, 1,
2, 2,
3, 3, 3, 3,
4, 4, 4, 4,
3, 3,
4, 4,
5, 5, 5, 5,
5, 5]
Note: The order of elements yielded by this transformation is
deterministic, as long as `map_func` is a pure function and
`deterministic=True`. If `map_func` contains any stateful operations, the
order in which that state is accessed is undefined.
Performance can often be improved by setting `num_parallel_calls` so that
`interleave` will use multiple threads to fetch elements. If determinism
isn't required, it can also improve performance to set
`deterministic=False`.
>>> filenames = [""/var/data/file1.txt"", ""/var/data/file2.txt"",
... ""/var/data/file3.txt"", ""/var/data/file4.txt""]
>>> dataset = tf.data.Dataset.from_tensor_slices(filenames)
>>> dataset = dataset.interleave(lambda x: tf.data.TFRecordDataset(x),
... cycle_length=4, num_parallel_calls=tf.data.experimental.AUTOTUNE,
... deterministic=False)
Args:
map_func: A function mapping a dataset element to a dataset.
cycle_length: (Optional.) The number of input elements that will be
processed concurrently. If not set, the tf.data runtime decides what it
should be based on available CPU. If `num_parallel_calls` is set to
`tf.data.experimental.AUTOTUNE`, the `cycle_length` argument identifies
the maximum degree of parallelism.
block_length: (Optional.) The number of consecutive elements to produce
from each input element before cycling to another input element. If not
set, defaults to 1.
num_parallel_calls: (Optional.) If specified, the implementation creates a
threadpool, which is used to fetch inputs from cycle elements
asynchronously and in parallel. The default behavior is to fetch inputs
from cycle elements synchronously with no parallelism. If the value
`tf.data.experimental.AUTOTUNE` is used, then the number of parallel
calls is set dynamically based on available CPU.
deterministic: (Optional.) A boolean controlling whether determinism
should be traded for performance by allowing elements to be produced out
of order. If `deterministic` is `None`, the
`tf.data.Options.experimental_deterministic` dataset option (`True` by
default) is used to decide whether to produce elements
deterministically.
Returns:
Dataset: A `Dataset`."
1573,filter,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1841,method,"Filters this dataset according to `predicate`.
>>> dataset = tf.data.Dataset.from_tensor_slices([1, 2, 3])
>>> dataset = dataset.filter(lambda x: x < 3)
>>> list(dataset.as_numpy_iterator())
[1, 2]
>>> # `tf.math.equal(x, y)` is required for equality comparison
>>> def filter_fn(x):
... return tf.math.equal(x, 1)
>>> dataset = dataset.filter(filter_fn)
>>> list(dataset.as_numpy_iterator())
[1]
Args:
predicate: A function mapping a dataset element to a boolean.
Returns:
Dataset: The `Dataset` containing the elements of this dataset for which
`predicate` is `True`."
1574,apply,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1864,method,"Applies a transformation function to this dataset.
`apply` enables chaining of custom `Dataset` transformations, which are
represented as functions that take one `Dataset` argument and return a
transformed `Dataset`.
>>> dataset = tf.data.Dataset.range(100)
>>> def dataset_fn(ds):
... return ds.filter(lambda x: x < 5)
>>> dataset = dataset.apply(dataset_fn)
>>> list(dataset.as_numpy_iterator())
[0, 1, 2, 3, 4]
Args:
transformation_func: A function that takes one `Dataset` argument and
returns a `Dataset`.
Returns:
Dataset: The `Dataset` returned by applying `transformation_func` to this
dataset."
1575,window,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1894,method,"Combines (nests of) input elements into a dataset of (nests of) windows.
A ""window"" is a finite dataset of flat elements of size `size` (or possibly
fewer if there are not enough input elements to fill the window and
`drop_remainder` evaluates to `False`).
The `shift` argument determines the number of input elements by which the
window moves on each iteration. If windows and elements are both numbered
starting at 0, the first element in window `k` will be element `k * shift`
of the input dataset. In particular, the first element of the first window
will always be the first element of the input dataset.
The `stride` argument determines the stride of the input elements, and the
`shift` argument determines the shift of the window.
For example:
>>> dataset = tf.data.Dataset.range(7).window(2)
>>> for window in dataset:
... print(list(window.as_numpy_iterator()))
[0, 1]
[2, 3]
[4, 5]
[6]
>>> dataset = tf.data.Dataset.range(7).window(3, 2, 1, True)
>>> for window in dataset:
... print(list(window.as_numpy_iterator()))
[0, 1, 2]
[2, 3, 4]
[4, 5, 6]
>>> dataset = tf.data.Dataset.range(7).window(3, 1, 2, True)
>>> for window in dataset:
... print(list(window.as_numpy_iterator()))
[0, 2, 4]
[1, 3, 5]
[2, 4, 6]
Note that when the `window` transformation is applied to a dataset of
nested elements, it produces a dataset of nested windows.
>>> nested = ([1, 2, 3, 4], [5, 6, 7, 8])
>>> dataset = tf.data.Dataset.from_tensor_slices(nested).window(2)
>>> for window in dataset:
... def to_numpy(ds):
... return list(ds.as_numpy_iterator())
... print(tuple(to_numpy(component) for component in window))
([1, 2], [5, 6])
([3, 4], [7, 8])
>>> dataset = tf.data.Dataset.from_tensor_slices({'a': [1, 2, 3, 4]})
>>> dataset = dataset.window(2)
>>> for window in dataset:
... def to_numpy(ds):
... return list(ds.as_numpy_iterator())
... print({'a': to_numpy(window['a'])})
{'a': [1, 2]}
{'a': [3, 4]}
Args:
size: A `tf.int64` scalar `tf.Tensor`, representing the number of elements
of the input dataset to combine into a window. Must be positive.
shift: (Optional.) A `tf.int64` scalar `tf.Tensor`, representing the
number of input elements by which the window moves in each iteration.
Defaults to `size`. Must be positive.
stride: (Optional.) A `tf.int64` scalar `tf.Tensor`, representing the
stride of the input elements in the sliding window. Must be positive.
The default value of 1 means ""retain every input element"".
drop_remainder: (Optional.) A `tf.bool` scalar `tf.Tensor`, representing
whether the last window should be dropped if its size is smaller than
`size`.
Returns:
Dataset: A `Dataset` of (nests of) windows -- a finite datasets of flat
elements created from the (nests of) input elements."
1576,reduce,tensorflow/tensorflow/python/data/ops/dataset_ops.py,1975,method,"Reduces the input dataset to a single element.
The transformation calls `reduce_func` successively on every element of
the input dataset until the dataset is exhausted, aggregating information in
its internal state. The `initial_state` argument is used for the initial
state and the final state is returned as the result.
>>> tf.data.Dataset.range(5).reduce(np.int64(0), lambda x, _: x + 1).numpy()
5
>>> tf.data.Dataset.range(5).reduce(np.int64(0), lambda x, y: x + y).numpy()
10
Args:
initial_state: An element representing the initial state of the
transformation.
reduce_func: A function that maps `(old_state, input_element)` to
`new_state`. It must take two arguments and return a new element
The structure of `new_state` must match the structure of
`initial_state`.
Returns:
A dataset element corresponding to the final state of the transformation."
1577,unbatch,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2088,method,"Splits elements of a dataset into multiple elements.
For example, if elements of the dataset are shaped `[B, a0, a1, ...]`,
where `B` may vary for each input element, then for each element in the
dataset, the unbatched dataset will contain `B` consecutive elements
of shape `[a0, a1, ...]`.
>>> elements = [ [1, 2, 3], [1, 2], [1, 2, 3, 4] ]
>>> dataset = tf.data.Dataset.from_generator(lambda: elements, tf.int64)
>>> dataset = dataset.unbatch()
>>> list(dataset.as_numpy_iterator())
[1, 2, 3, 1, 2, 1, 2, 3, 4]
Note: `unbatch` requires a data copy to slice up the batched tensor into
smaller, unbatched tensors. When optimizing performance, try to avoid
unnecessary usage of `unbatch`.
Returns:
A `Dataset`."
1578,with_options,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2112,method,"Returns a new `tf.data.Dataset` with the given options set.
The options are ""global"" in the sense they apply to the entire dataset.
If options are set multiple times, they are merged as long as different
options do not use different non-default values.
>>> ds = tf.data.Dataset.range(5)
>>> ds = ds.interleave(lambda x: tf.data.Dataset.range(5),
... cycle_length=3,
... num_parallel_calls=3)
>>> options = tf.data.Options()
>>> # This will make the interleave order non-deterministic.
>>> options.experimental_deterministic = False
>>> ds = ds.with_options(options)
Args:
options: A `tf.data.Options` that identifies the options the use.
Returns:
Dataset: A `Dataset` with the given options.
Raises:
ValueError: when an option is set more than once to a non-default value"
1579,cardinality,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2139,method,"Returns the cardinality of the dataset, if known.
`cardinality` may return `tf.data.INFINITE_CARDINALITY` if the dataset
contains an infinite number of elements or `tf.data.UNKNOWN_CARDINALITY` if
the analysis fails to determine the number of elements in the dataset
(e.g. when the dataset source is a file).
>>> dataset = tf.data.Dataset.range(42)
>>> print(dataset.cardinality().numpy())
42
>>> dataset = dataset.repeat()
>>> cardinality = dataset.cardinality()
>>> print((cardinality == tf.data.INFINITE_CARDINALITY).numpy())
True
>>> dataset = dataset.filter(lambda x: True)
>>> cardinality = dataset.cardinality()
>>> print((cardinality == tf.data.UNKNOWN_CARDINALITY).numpy())
True
Returns:
A scalar `tf.int64` `Tensor` representing the cardinality of the dataset.
If the cardinality is infinite or unknown, `cardinality` returns the
named constants `tf.data.INFINITE_CARDINALITY` and
`tf.data.UNKNOWN_CARDINALITY` respectively."
1580,is_tensor_or_parent_ref,tensorflow/tensorflow/python/data/ops/dataset_ops.py,323,method,
1581,get_next_id,tensorflow/tensorflow/python/data/ops/dataset_ops.py,699,method,
1582,get_iterator,tensorflow/tensorflow/python/data/ops/dataset_ops.py,709,method,
1583,iterator_completed,tensorflow/tensorflow/python/data/ops/dataset_ops.py,717,method,
1584,get_iterator_id_fn,tensorflow/tensorflow/python/data/ops/dataset_ops.py,793,method,"Creates a unique `iterator_id` for each pass over the dataset.
The returned `iterator_id` disambiguates between multiple concurrently
existing iterators.
Args:
unused_dummy: Ignored value.
Returns:
A `tf.int64` tensor whose value uniquely identifies an iterator in
`generator_state`."
1585,generator_next_fn,tensorflow/tensorflow/python/data/ops/dataset_ops.py,809,method,"Generates the next element from iterator with ID `iterator_id_t`.
We map this function across an infinite repetition of the
`iterator_id_t`, and raise `StopIteration` to terminate the iteration.
Args:
iterator_id_t: A `tf.int64` tensor whose value uniquely identifies the
iterator in `generator_state` from which to generate an element.
Returns:
The next element to generate from the iterator."
1586,finalize_fn,tensorflow/tensorflow/python/data/ops/dataset_ops.py,878,method,Releases host-side state for the iterator with ID `iterator_id_t`.
1587,flat_map_fn,tensorflow/tensorflow/python/data/ops/dataset_ops.py,895,method,
1588,generator_py_func,tensorflow/tensorflow/python/data/ops/dataset_ops.py,823,method,A `py_func` that will be called to invoke the iterator.
1589,finalize_py_func,tensorflow/tensorflow/python/data/ops/dataset_ops.py,881,method,
1590,DatasetV1,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2169,class,"Represents a potentially large set of elements.
A `Dataset` can be used to represent an input pipeline as a
collection of elements and a ""logical plan"" of transformations that act on
those elements."
1591,make_one_shot_iterator,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2216,method,"Creates an iterator for elements of this dataset.
Note: The returned iterator will be initialized automatically.
A ""one-shot"" iterator does not currently support re-initialization. For
that see `make_initializable_iterator`.
Example:
```python
# Building graph ...
dataset = ...
next_value = dataset.make_one_shot_iterator().get_next()
# ... from within a session ...
try:
while True:
value = sess.run(next_value)
...
except tf.errors.OutOfRangeError:
pass
```
Returns:
An `tf.data.Iterator` for elements of this dataset."
1592,make_initializable_iterator,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2309,method,"Creates an iterator for elements of this dataset.
Note: The returned iterator will be in an uninitialized state,
and you must run the `iterator.initializer` operation before using it:
```python
# Building graph ...
dataset = ...
iterator = dataset.make_initializable_iterator()
next_value = iterator.get_next() # This is a Tensor.
# ... from within a session ...
sess.run(iterator.initializer)
try:
while True:
value = sess.run(next_value)
...
except tf.errors.OutOfRangeError:
pass
```
Args:
shared_name: (Optional.) If non-empty, the returned iterator will be
shared under the given name across multiple sessions that share the same
devices (e.g. when using a remote server).
Returns:
A `tf.data.Iterator` for elements of this dataset.
Raises:
RuntimeError: If eager execution is enabled."
1593,output_classes,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2367,method,"Returns the class of each component of an element of this dataset.
Returns:
A nested structure of Python `type` objects corresponding to each
component of an element of this dataset."
1594,output_shapes,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2381,method,"Returns the shape of each component of an element of this dataset.
Returns:
A nested structure of `tf.TensorShape` objects corresponding to each
component of an element of this dataset."
1595,output_types,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2395,method,"Returns the type of each component of an element of this dataset.
Returns:
A nested structure of `tf.DType` objects corresponding to each component
of an element of this dataset."
1596,element_spec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2407,method,
1597,from_tensors,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2415,method,
1598,from_tensor_slices,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2420,method,
1599,from_sparse_tensor_slices,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2425,method,"Splits each rank-N `tf.sparse.SparseTensor` in this dataset row-wise.
Args:
sparse_tensor: A `tf.sparse.SparseTensor`.
Returns:
Dataset: A `Dataset` of rank-(N-1) sparse tensors."
1600,from_generator,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2438,method,
1601,range,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2444,method,
1602,zip,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2449,method,
1603,concatenate,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2453,method,
1604,prefetch,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2457,method,
1605,list_files,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2462,method,
1606,repeat,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2466,method,
1607,shuffle,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2470,method,
1608,cache,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2475,method,
1609,take,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2479,method,
1610,skip,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2483,method,
1611,shard,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2487,method,
1612,batch,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2491,method,
1613,padded_batch,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2496,method,
1614,map,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2506,method,
1615,map_with_legacy_function,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2520,method,"Maps `map_func` across the elements of this dataset.
Note: This is an escape hatch for existing uses of `map` that do not work
with V2 functions. New uses are strongly discouraged and existing uses
should migrate to `map` as this method will be removed in V2.
Args:
map_func: A function mapping a nested structure of tensors (having shapes
and types defined by `self.output_shapes` and `self.output_types`) to
another nested structure of tensors.
num_parallel_calls: (Optional.) A `tf.int32` scalar `tf.Tensor`,
representing the number elements to process asynchronously in parallel.
If not specified, elements will be processed sequentially. If the value
`tf.data.experimental.AUTOTUNE` is used, then the number of parallel
calls is set dynamically based on available CPU.
deterministic: (Optional.) A boolean controlling whether determinism
should be traded for performance by allowing elements to be produced out
of order. If `deterministic` is `None`, the
`tf.data.Options.experimental_deterministic` dataset option (`True` by
default) is used to decide whether to produce elements
deterministically.
Returns:
Dataset: A `Dataset`."
1616,flat_map,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2567,method,
1617,interleave,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2571,method,
1618,filter,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2582,method,
1619,filter_with_legacy_function,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2586,method,"Filters this dataset according to `predicate`.
Note: This is an escape hatch for existing uses of `filter` that do not work
with V2 functions. New uses are strongly discouraged and existing uses
should migrate to `filter` as this method will be removed in V2.
Args:
predicate: A function mapping a nested structure of tensors (having shapes
and types defined by `self.output_shapes` and `self.output_types`) to a
scalar `tf.bool` tensor.
Returns:
Dataset: The `Dataset` containing the elements of this dataset for which
`predicate` is `True`."
1620,apply,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2605,method,
1621,window,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2609,method,
1622,unbatch,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2614,method,
1623,with_options,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2618,method,
1624,DatasetV1Adapter,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2628,class,Wraps a V2 `Dataset` object in the `tf.compat.v1.data.Dataset` API.
1625,options,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2647,method,
1626,element_spec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2651,method,
1627,make_one_shot_iterator,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2684,function,"Creates an iterator for elements of `dataset`.
Note: The returned iterator will be initialized automatically.
A ""one-shot"" iterator does not support re-initialization.
Args:
dataset: A `tf.data.Dataset`.
Returns:
A `tf.data.Iterator` for elements of `dataset`."
1628,make_initializable_iterator,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2705,function,"Creates an iterator for elements of `dataset`.
Note: The returned iterator will be in an uninitialized state,
and you must run the `iterator.initializer` operation before using it:
```python
dataset = ...
iterator = tf.compat.v1.data.make_initializable_iterator(dataset)
# ...
sess.run(iterator.initializer)
```
Args:
dataset: A `tf.data.Dataset`.
shared_name: (Optional.) If non-empty, the returned iterator will be shared
under the given name across multiple sessions that share the same devices
(e.g. when using a remote server).
Returns:
A `tf.data.Iterator` for elements of `dataset`.
Raises:
RuntimeError: If eager execution is enabled."
1629,get_structure,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2739,function,"Returns the type signature for elements of the input dataset / iterator.
Args:
dataset_or_iterator: A `tf.data.Dataset` or an `tf.data.Iterator`.
Returns:
A nested structure of `tf.TypeSpec` objects matching the structure of an
element of `dataset_or_iterator` and specifying the type of individual
components.
Raises:
TypeError: If input is not a `tf.data.Dataset` or an `tf.data.Iterator`
object."
1630,get_legacy_output_classes,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2763,function,"Returns the output classes for elements of the input dataset / iterator.
Args:
dataset_or_iterator: A `tf.data.Dataset` or `tf.data.Iterator`.
Returns:
A nested structure of Python `type` objects matching the structure of the
dataset / iterator elements and specifying the class of the individual
components."
1631,get_legacy_output_shapes,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2780,function,"Returns the output shapes for elements of the input dataset / iterator.
Args:
dataset_or_iterator: A `tf.data.Dataset` or `tf.data.Iterator`.
Returns:
A nested structure of `tf.TensorShape` objects matching the structure of
the dataset / iterator elements and specifying the shape of the individual
components."
1632,get_legacy_output_types,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2797,function,"Returns the output shapes for elements of the input dataset / iterator.
Args:
dataset_or_iterator: A `tf.data.Dataset` or `tf.data.Iterator`.
Returns:
A nested structure of `tf.DType` objects objects matching the structure of
dataset / iterator elements and specifying the shape of the individual
components."
1633,Options,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2814,class,"Represents options for tf.data.Dataset.
An `Options` object can be, for instance, used to control which graph
optimizations to apply or whether to use performance modeling to dynamically
tune the parallelism of operations such as `tf.data.Dataset.map` or
`tf.data.Dataset.interleave`.
After constructing an `Options` object, use `dataset.with_options(options)` to
apply the options to a dataset.
>>> dataset = tf.data.Dataset.range(3)
>>> options = tf.data.Options()
>>> # Set options here.
>>> dataset = dataset.with_options(options)"
1634,merge,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2927,method,"Merges itself with the given `tf.data.Options`.
The given `tf.data.Options` can be merged as long as there does not exist an
attribute that is set to different values in `self` and `options`.
Args:
options: a `tf.data.Options` to merge with
Raises:
ValueError: if the given `tf.data.Options` cannot be merged
Returns:
New `tf.data.Options()` object which is the result of merging self with
the input `tf.data.Options`."
1635,DatasetSource,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2946,class,Abstract class representing a dataset with no inputs.
1636,UnaryDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2953,class,Abstract class representing a dataset with one input.
1637,UnaryUnchangedStructureDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2964,class,Represents a unary dataset with the same input and output structure.
1638,element_spec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2973,method,
1639,TensorDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2977,class,A `Dataset` with a single element.
1640,element_spec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2992,method,
1641,TensorSliceDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,2996,class,A `Dataset` of slices from a dataset element.
1642,element_spec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3019,method,
1643,SparseTensorSliceDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3023,class,A `Dataset` that splits a rank-N `tf.sparse.SparseTensor` into its rows.
1644,element_spec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3048,method,
1645,from_variant,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3080,function,"Constructs a dataset from the given variant and structure.
Args:
variant: A scalar `tf.variant` tensor representing a dataset.
structure: A `tf.data.experimental.Structure` object representing the
structure of each element in the dataset.
Returns:
A `tf.data.Dataset` instance."
1646,to_variant,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3095,function,"Returns a variant representing the given dataset.
Args:
dataset: A `tf.data.Dataset`.
Returns:
A scalar `tf.variant` tensor representing the given dataset."
1647,DatasetSpec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3110,class,"Type specification for `tf.data.Dataset`.
See `tf.TypeSpec` for more information about TensorFlow type specifications.
>>> dataset = tf.data.Dataset.range(3)
>>> tf.data.DatasetSpec.from_value(dataset)
DatasetSpec(TensorSpec(shape=(), dtype=tf.int64, name=None), TensorShape([]))"
1648,value_type,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3127,method,
1649,from_value,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3154,method,Creates a `DatasetSpec` for the given `tf.data.Dataset` value.
1650,StructuredFunctionWrapper,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3183,class,A function wrapper that supports structured arguments and return values.
1651,output_structure,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3388,method,
1652,output_classes,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3392,method,
1653,output_shapes,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3398,method,
1654,output_types,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3404,method,
1655,function,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3410,method,
1656,wrapper_fn,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3330,method,
1657,wrapper_fn,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3363,method,
1658,ZipDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3469,class,A `Dataset` that zips its inputs together.
1659,element_spec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3497,method,
1660,ConcatenateDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3501,class,A `Dataset` that concatenates its input with given dataset.
1661,element_spec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3545,method,
1662,RepeatDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3549,class,A `Dataset` that repeats its input several times.
1663,RangeDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3567,class,A `Dataset` of a step separated range of values.
1664,element_spec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3606,method,
1665,CacheDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3610,class,A `Dataset` that caches elements of its input.
1666,ShuffleDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3632,class,A `Dataset` that randomly shuffles the elements of its input.
1667,TakeDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3688,class,A `Dataset` containing the first `count` elements from its input.
1668,SkipDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3702,class,A `Dataset` skipping the first `count` elements from its input.
1669,ShardDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3716,class,A `Dataset` for sharding its input.
1670,BatchDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3733,class,A `Dataset` that batches contiguous elements from its input.
1671,element_spec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3766,method,
1672,PaddedBatchDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3939,class,A `Dataset` that batches and pads contiguous elements from its input.
1673,element_spec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,4021,method,
1674,check_types,tensorflow/tensorflow/python/data/ops/dataset_ops.py,3947,method,
1675,MapDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,4030,class,A `Dataset` that maps a function over elements in its input.
1676,element_spec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,4061,method,
1677,ParallelMapDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,4068,class,A `Dataset` that maps a function over elements in its input in parallel.
1678,element_spec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,4111,method,
1679,FlatMapDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,4118,class,A `Dataset` that maps a function over its input and flattens the result.
1680,element_spec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,4142,method,
1681,InterleaveDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,4149,class,A `Dataset` that interleaves the result of transformed inputs.
1682,element_spec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,4181,method,
1683,ParallelInterleaveDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,4188,class,A `Dataset` that maps a function over its input and interleaves the result.
1684,element_spec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,4249,method,
1685,FilterDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,4256,class,A `Dataset` that filters its input according to a predicate function.
1686,PrefetchDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,4288,class,A `Dataset` that asynchronously prefetches its input.
1687,WindowDataset,tensorflow/tensorflow/python/data/ops/dataset_ops.py,4321,class,A dataset that creates window datasets from the input elements.
1688,element_spec,tensorflow/tensorflow/python/data/ops/dataset_ops.py,4353,method,
1689,normalize_to_dense,tensorflow/tensorflow/python/data/ops/dataset_ops.py,4456,function,"Normalizes non-tensor components in a dataset to dense representations.
This is necessary for dataset transformations that slice along the batch
dimension and are oblivious to non-tensors, e.g. `unbatch`, `rebatch`.
Args:
dataset: Dataset to normalize.
Returns:
A dataset whose sparse and ragged tensors have been normalized to their
dense representations."
1690,Iterator,tensorflow/tensorflow/python/data/ops/iterator_ops.py,81,class,Represents the state of iterating through a `Dataset`.
1691,from_structure,tensorflow/tensorflow/python/data/ops/iterator_ops.py,125,method,"Creates a new, uninitialized `Iterator` with the given structure.
This iterator-constructing method can be used to create an iterator that
is reusable with many different datasets.
The returned iterator is not bound to a particular dataset, and it has
no `initializer`. To initialize the iterator, run the operation returned by
`Iterator.make_initializer(dataset)`.
The following is an example
```python
iterator = Iterator.from_structure(tf.int64, tf.TensorShape([]))
dataset_range = Dataset.range(10)
range_initializer = iterator.make_initializer(dataset_range)
dataset_evens = dataset_range.filter(lambda x: x % 2 == 0)
evens_initializer = iterator.make_initializer(dataset_evens)
# Define a model based on the iterator; in this example, the model_fn
# is expected to take scalar tf.int64 Tensors as input (see
# the definition of 'iterator' above).
prediction, loss = model_fn(iterator.get_next())
# Train for `num_epochs`, where for each epoch, we first iterate over
# dataset_range, and then iterate over dataset_evens.
for _ in range(num_epochs):
# Initialize the iterator to `dataset_range`
sess.run(range_initializer)
while True:
try:
pred, loss_val = sess.run([prediction, loss])
except tf.errors.OutOfRangeError:
break
# Initialize the iterator to `dataset_evens`
sess.run(evens_initializer)
while True:
try:
pred, loss_val = sess.run([prediction, loss])
except tf.errors.OutOfRangeError:
break
```
Args:
output_types: A nested structure of `tf.DType` objects corresponding to
each component of an element of this dataset.
output_shapes: (Optional.) A nested structure of `tf.TensorShape` objects
corresponding to each component of an element of this dataset. If
omitted, each component will have an unconstrainted shape.
shared_name: (Optional.) If non-empty, this iterator will be shared under
the given name across multiple sessions that share the same devices
(e.g. when using a remote server).
output_classes: (Optional.) A nested structure of Python `type` objects
corresponding to each component of an element of this iterator. If
omitted, each component is assumed to be of type `tf.Tensor`.
Returns:
An `Iterator`.
Raises:
TypeError: If the structures of `output_shapes` and `output_types` are
not the same."
1692,from_string_handle,tensorflow/tensorflow/python/data/ops/iterator_ops.py,229,method,"Creates a new, uninitialized `Iterator` based on the given handle.
This method allows you to define a ""feedable"" iterator where you can choose
between concrete iterators by feeding a value in a `tf.Session.run` call.
In that case, `string_handle` would be a `tf.compat.v1.placeholder`, and you
would
feed it with the value of `tf.data.Iterator.string_handle` in each step.
For example, if you had two iterators that marked the current position in
a training dataset and a test dataset, you could choose which to use in
each step as follows:
```python
train_iterator = tf.data.Dataset(...).make_one_shot_iterator()
train_iterator_handle = sess.run(train_iterator.string_handle())
test_iterator = tf.data.Dataset(...).make_one_shot_iterator()
test_iterator_handle = sess.run(test_iterator.string_handle())
handle = tf.compat.v1.placeholder(tf.string, shape=[])
iterator = tf.data.Iterator.from_string_handle(
handle, train_iterator.output_types)
next_element = iterator.get_next()
loss = f(next_element)
train_loss = sess.run(loss, feed_dict={handle: train_iterator_handle})
test_loss = sess.run(loss, feed_dict={handle: test_iterator_handle})
```
Args:
string_handle: A scalar `tf.Tensor` of type `tf.string` that evaluates to
a handle produced by the `Iterator.string_handle()` method.
output_types: A nested structure of `tf.DType` objects corresponding to
each component of an element of this dataset.
output_shapes: (Optional.) A nested structure of `tf.TensorShape` objects
corresponding to each component of an element of this dataset. If
omitted, each component will have an unconstrainted shape.
output_classes: (Optional.) A nested structure of Python `type` objects
corresponding to each component of an element of this iterator. If
omitted, each component is assumed to be of type `tf.Tensor`.
Returns:
An `Iterator`."
1693,initializer,tensorflow/tensorflow/python/data/ops/iterator_ops.py,307,method,"A `tf.Operation` that should be run to initialize this iterator.
Returns:
A `tf.Operation` that should be run to initialize this iterator
Raises:
ValueError: If this iterator initializes itself automatically."
1694,make_initializer,tensorflow/tensorflow/python/data/ops/iterator_ops.py,323,method,"Returns a `tf.Operation` that initializes this iterator on `dataset`.
Args:
dataset: A `Dataset` with compatible structure to this iterator.
name: (Optional.) A name for the created operation.
Returns:
A `tf.Operation` that can be run to initialize this iterator on the given
`dataset`.
Raises:
TypeError: If `dataset` and this iterator do not have a compatible
element structure."
1695,get_next,tensorflow/tensorflow/python/data/ops/iterator_ops.py,379,method,"Returns a nested structure of `tf.Tensor`s representing the next element.
In graph mode, you should typically call this method *once* and use its
result as the input to another computation. A typical loop will then call
`tf.Session.run` on the result of that computation. The loop will terminate
when the `Iterator.get_next()` operation raises
`tf.errors.OutOfRangeError`. The following skeleton shows how to use
this method when building a training loop:
```python
dataset = ... # A `tf.data.Dataset` object.
iterator = dataset.make_initializable_iterator()
next_element = iterator.get_next()
# Build a TensorFlow graph that does something with each element.
loss = model_function(next_element)
optimizer = ... # A `tf.compat.v1.train.Optimizer` object.
train_op = optimizer.minimize(loss)
with tf.compat.v1.Session() as sess:
try:
while True:
sess.run(train_op)
except tf.errors.OutOfRangeError:
pass
```
NOTE: It is legitimate to call `Iterator.get_next()` multiple times, e.g.
when you are distributing different elements to multiple devices in a single
step. However, a common pitfall arises when users call `Iterator.get_next()`
in each iteration of their training loop. `Iterator.get_next()` adds ops to
the graph, and executing each op allocates resources (including threads); as
a consequence, invoking it in every iteration of a training loop causes
slowdown and eventual resource exhaustion. To guard against this outcome, we
log a warning when the number of uses crosses a fixed threshold of
suspiciousness.
Args:
name: (Optional.) A name for the created operation.
Returns:
A nested structure of `tf.Tensor` objects."
1696,string_handle,tensorflow/tensorflow/python/data/ops/iterator_ops.py,435,method,"Returns a string-valued `tf.Tensor` that represents this iterator.
Args:
name: (Optional.) A name for the created operation.
Returns:
A scalar `tf.Tensor` of type `tf.string`."
1697,output_classes,tensorflow/tensorflow/python/data/ops/iterator_ops.py,453,method,"Returns the class of each component of an element of this iterator.
The expected values are `tf.Tensor` and `tf.sparse.SparseTensor`.
Returns:
A nested structure of Python `type` objects corresponding to each
component of an element of this dataset."
1698,output_shapes,tensorflow/tensorflow/python/data/ops/iterator_ops.py,469,method,"Returns the shape of each component of an element of this iterator.
Returns:
A nested structure of `tf.TensorShape` objects corresponding to each
component of an element of this dataset."
1699,output_types,tensorflow/tensorflow/python/data/ops/iterator_ops.py,483,method,"Returns the type of each component of an element of this iterator.
Returns:
A nested structure of `tf.DType` objects corresponding to each component
of an element of this dataset."
1700,element_spec,tensorflow/tensorflow/python/data/ops/iterator_ops.py,495,method,
1701,IteratorResourceDeleter,tensorflow/tensorflow/python/data/ops/iterator_ops.py,518,class,"An object which cleans up an iterator resource handle.
An alternative to defining a __del__ method on an object. Even if the parent
object is part of a reference cycle, the cycle will be collectable."
1702,IteratorBase,tensorflow/tensorflow/python/data/ops/iterator_ops.py,548,class,"Represents an iterator of a `tf.data.Dataset`.
`tf.data.Iterator` is the primary mechanism for enumerating elements of a
`tf.data.Dataset`. It supports the Python Iterator protocol, which means
it can be iterated over using a for-loop:
>>> dataset = tf.data.Dataset.range(2)
>>> for element in dataset:
... print(element)
tf.Tensor(0, shape=(), dtype=int64)
tf.Tensor(1, shape=(), dtype=int64)
or by fetching individual elements explicitly via `get_next()`:
>>> dataset = tf.data.Dataset.range(2)
>>> iterator = iter(dataset)
>>> print(iterator.get_next())
tf.Tensor(0, shape=(), dtype=int64)
>>> print(iterator.get_next())
tf.Tensor(1, shape=(), dtype=int64)
In addition, non-raising iteration is supported via `get_next_as_optional()`,
which returns the next element (if available) wrapped in a
`tf.experimental.Optional`.
>>> dataset = tf.data.Dataset.from_tensors(42)
>>> iterator = iter(dataset)
>>> optional = iterator.get_next_as_optional()
>>> print(optional.has_value())
tf.Tensor(True, shape=(), dtype=bool)
>>> optional = iterator.get_next_as_optional()
>>> print(optional.has_value())
tf.Tensor(False, shape=(), dtype=bool)"
1703,element_spec,tensorflow/tensorflow/python/data/ops/iterator_ops.py,586,method,"The type specification of an element of this iterator.
>>> dataset = tf.data.Dataset.from_tensors(42)
>>> iterator = iter(dataset)
>>> iterator.element_spec
tf.TensorSpec(shape=(), dtype=tf.int32, name=None)
Returns:
A nested structure of `tf.TypeSpec` objects matching the structure of an
element of this iterator, specifying the type of individual components."
1704,get_next,tensorflow/tensorflow/python/data/ops/iterator_ops.py,601,method,"Returns a nested structure of `tf.Tensor`s containing the next element.
>>> dataset = tf.data.Dataset.from_tensors(42)
>>> iterator = iter(dataset)
>>> print(iterator.get_next())
tf.Tensor(42, shape=(), dtype=int32)
Returns:
A nested structure of `tf.Tensor` objects.
Raises:
`tf.errors.OutOfRangeError`: If the end of the iterator has been reached."
1705,get_next_as_optional,tensorflow/tensorflow/python/data/ops/iterator_ops.py,618,method,"Returns a `tf.experimental.Optional` which contains the next element.
If the iterator has reached the end of the sequence, the returned
`tf.experimental.Optional` will have no value.
>>> dataset = tf.data.Dataset.from_tensors(42)
>>> iterator = iter(dataset)
>>> optional = iterator.get_next_as_optional()
>>> print(optional.has_value())
tf.Tensor(True, shape=(), dtype=bool)
>>> print(optional.get_value())
tf.Tensor(42, shape=(), dtype=int32)
>>> optional = iterator.get_next_as_optional()
>>> print(optional.has_value())
tf.Tensor(False, shape=(), dtype=bool)
Returns:
A `tf.experimental.Optional` object representing the next element."
1706,OwnedIterator,tensorflow/tensorflow/python/data/ops/iterator_ops.py,641,class,"An iterator producing tf.Tensor objects from a tf.data.Dataset.
The iterator resource created through `OwnedIterator` is owned by the Python
object and the life time of the underlying resource is tied to the life time
of the `OwnedIterator` object. This makes `OwnedIterator` appropriate for use
in eager mode and inside of tf.functions."
1707,next,tensorflow/tensorflow/python/data/ops/iterator_ops.py,737,method,
1708,output_classes,tensorflow/tensorflow/python/data/ops/iterator_ops.py,781,method,"Returns the class of each component of an element of this iterator.
The expected values are `tf.Tensor` and `tf.sparse.SparseTensor`.
Returns:
A nested structure of Python `type` objects corresponding to each
component of an element of this dataset."
1709,output_shapes,tensorflow/tensorflow/python/data/ops/iterator_ops.py,797,method,"Returns the shape of each component of an element of this iterator.
Returns:
A nested structure of `tf.TensorShape` objects corresponding to each
component of an element of this dataset."
1710,output_types,tensorflow/tensorflow/python/data/ops/iterator_ops.py,811,method,"Returns the type of each component of an element of this iterator.
Returns:
A nested structure of `tf.DType` objects corresponding to each component
of an element of this dataset."
1711,element_spec,tensorflow/tensorflow/python/data/ops/iterator_ops.py,823,method,
1712,get_next,tensorflow/tensorflow/python/data/ops/iterator_ops.py,826,method,
1713,get_next_as_optional,tensorflow/tensorflow/python/data/ops/iterator_ops.py,829,method,
1714,IteratorSpec,tensorflow/tensorflow/python/data/ops/iterator_ops.py,857,class,"Type specification for `tf.data.Iterator`.
For instance, `tf.data.IteratorSpec` can be used to define a tf.function that
takes `tf.data.Iterator` as an input argument:
>>> @tf.function(input_signature=[tf.data.IteratorSpec(
... tf.TensorSpec(shape=(), dtype=tf.int32, name=None))])
... def square(iterator):
... x = iterator.get_next()
... return x * x
>>> dataset = tf.data.Dataset.from_tensors(5)
>>> iterator = iter(dataset)
>>> print(square(iterator))
tf.Tensor(25, shape=(), dtype=int32)
Attributes:
element_spec: A nested structure of `TypeSpec` objects that represents the
type specification of the iterator elements."
1715,value_type,tensorflow/tensorflow/python/data/ops/iterator_ops.py,884,method,
1716,from_value,tensorflow/tensorflow/python/data/ops/iterator_ops.py,907,method,
1717,get_next_as_optional,tensorflow/tensorflow/python/data/ops/iterator_ops.py,939,function,"Returns a `tf.experimental.Optional` with the next element of the iterator.
If the iterator has reached the end of the sequence, the returned
`tf.experimental.Optional` will have no value.
Args:
iterator: A `tf.data.Iterator`.
Returns:
A `tf.experimental.Optional` object which either contains the next element
of the iterator (if it exists) or no value."
1718,MultiDeviceIterator,tensorflow/tensorflow/python/data/ops/multi_device_iterator_ops.py,211,class,An iterator over multiple devices.
1719,get_next,tensorflow/tensorflow/python/data/ops/multi_device_iterator_ops.py,322,method,"Returns the next element given a `device`, else returns all in a list."
1720,get_next_as_optional,tensorflow/tensorflow/python/data/ops/multi_device_iterator_ops.py,334,method,
1721,initializer,tensorflow/tensorflow/python/data/ops/multi_device_iterator_ops.py,343,method,
1722,element_spec,tensorflow/tensorflow/python/data/ops/multi_device_iterator_ops.py,369,method,
1723,MultiDeviceIteratorResourceDeleter,tensorflow/tensorflow/python/data/ops/multi_device_iterator_ops.py,373,class,"An object which cleans up a Multi Device Iterator resource.
An alternative to defining a __del__ method on an object. Even if the parent
object is part of a reference cycle, the cycle will be collectible."
1724,MultiDeviceIteratorSpec,tensorflow/tensorflow/python/data/ops/multi_device_iterator_ops.py,411,class,Type specification for `OwnedMultiDeviceIterator`.
1725,value_type,tensorflow/tensorflow/python/data/ops/multi_device_iterator_ops.py,422,method,
1726,from_value,tensorflow/tensorflow/python/data/ops/multi_device_iterator_ops.py,453,method,
1727,OwnedMultiDeviceIterator,tensorflow/tensorflow/python/data/ops/multi_device_iterator_ops.py,461,class,"An iterator over multiple devices.
The multi-device iterator resource created through `OwnedMultiDeviceIterator`
is owned by the Python object and the life time of the underlying resource is
tied to the life time of the `OwnedMultiDeviceIterator` object. This makes
`OwnedMultiDeviceIterator` appropriate for use in eager mode and inside of
tf.functions."
1728,get_next,tensorflow/tensorflow/python/data/ops/multi_device_iterator_ops.py,577,method,"Returns the next element given a `device`, else returns all in a list."
1729,next,tensorflow/tensorflow/python/data/ops/multi_device_iterator_ops.py,592,method,
1730,get_next_as_optional,tensorflow/tensorflow/python/data/ops/multi_device_iterator_ops.py,601,method,
1731,element_spec,tensorflow/tensorflow/python/data/ops/multi_device_iterator_ops.py,610,method,
1732,Optional,tensorflow/tensorflow/python/data/ops/optional_ops.py,38,class,"Represents a value that may or may not be present.
A `tf.experimental.Optional` can represent the result of an operation that may
fail as a value, rather than raising an exception and halting execution. For
example, `tf.data.Iterator.get_next_as_optional()` returns a
`tf.experimental.Optional` that either contains the next element of an
iterator if one exists, or an ""empty"" value that indicates the end of the
sequence has been reached.
`tf.experimental.Optional` can only be used with values that are convertible
to `tf.Tensor` or `tf.CompositeTensor`.
One can create a `tf.experimental.Optional` from a value using the
`from_value()` method:
>>> optional = tf.experimental.Optional.from_value(42)
>>> print(optional.has_value())
tf.Tensor(True, shape=(), dtype=bool)
>>> print(optional.get_value())
tf.Tensor(42, shape=(), dtype=int32)
or without a value using the `empty()` method:
>>> optional = tf.experimental.Optional.empty(
... tf.TensorSpec(shape=(), dtype=tf.int32, name=None))
>>> print(optional.has_value())
tf.Tensor(False, shape=(), dtype=bool)"
1733,has_value,tensorflow/tensorflow/python/data/ops/optional_ops.py,69,method,"Returns a tensor that evaluates to `True` if this optional has a value.
>>> optional = tf.experimental.Optional.from_value(42)
>>> print(optional.has_value())
tf.Tensor(True, shape=(), dtype=bool)
Args:
name: (Optional.) A name for the created operation.
Returns:
A scalar `tf.Tensor` of type `tf.bool`."
1734,get_value,tensorflow/tensorflow/python/data/ops/optional_ops.py,85,method,"Returns the value wrapped by this optional.
If this optional does not have a value (i.e. `self.has_value()` evaluates to
`False`), this operation will raise `tf.errors.InvalidArgumentError` at
runtime.
>>> optional = tf.experimental.Optional.from_value(42)
>>> print(optional.get_value())
tf.Tensor(42, shape=(), dtype=int32)
Args:
name: (Optional.) A name for the created operation.
Returns:
The wrapped value."
1735,element_spec,tensorflow/tensorflow/python/data/ops/optional_ops.py,105,method,"The type specification of an element of this optional.
>>> optional = tf.experimental.Optional.from_value(42)
>>> print(optional.element_spec)
tf.TensorSpec(shape=(), dtype=tf.int32, name=None)
Returns:
A nested structure of `tf.TypeSpec` objects matching the structure of an
element of this optional, specifying the type of individual components."
1736,empty,tensorflow/tensorflow/python/data/ops/optional_ops.py,119,method,"Returns an `Optional` that has no value.
NOTE: This method takes an argument that defines the structure of the value
that would be contained in the returned `Optional` if it had a value.
>>> optional = tf.experimental.Optional.empty(
... tf.TensorSpec(shape=(), dtype=tf.int32, name=None))
>>> print(optional.has_value())
tf.Tensor(False, shape=(), dtype=bool)
Args:
element_spec: A nested structure of `tf.TypeSpec` objects matching the
structure of an element of this optional.
Returns:
A `tf.experimental.Optional` with no value."
1737,from_value,tensorflow/tensorflow/python/data/ops/optional_ops.py,140,method,"Returns a `tf.experimental.Optional` that wraps the given value.
>>> optional = tf.experimental.Optional.from_value(42)
>>> print(optional.has_value())
tf.Tensor(True, shape=(), dtype=bool)
>>> print(optional.get_value())
tf.Tensor(42, shape=(), dtype=int32)
Args:
value: A value to wrap. The value must be convertible to `tf.Tensor` or
`tf.CompositeTensor`.
Returns:
A `tf.experimental.Optional` that wraps `value`."
1738,OptionalSpec,tensorflow/tensorflow/python/data/ops/optional_ops.py,206,class,"Type specification for `tf.experimental.Optional`.
For instance, `tf.OptionalSpec` can be used to define a tf.function that takes
`tf.experimental.Optional` as an input argument:
>>> @tf.function(input_signature=[tf.OptionalSpec(
... tf.TensorSpec(shape=(), dtype=tf.int32, name=None))])
... def maybe_square(optional):
... if optional.has_value():
... x = optional.get_value()
... return x * x
... return -1
>>> optional = tf.experimental.Optional.from_value(5)
>>> print(maybe_square(optional))
tf.Tensor(25, shape=(), dtype=int32)
Attributes:
element_spec: A nested structure of `TypeSpec` objects that represents the
type specification of the optional element."
1739,value_type,tensorflow/tensorflow/python/data/ops/optional_ops.py,234,method,
1740,from_value,tensorflow/tensorflow/python/data/ops/optional_ops.py,252,method,
1741,TextLineDatasetV2,tensorflow/tensorflow/python/data/ops/readers.py,134,class,A `Dataset` comprising lines from one or more text files.
1742,element_spec,tensorflow/tensorflow/python/data/ops/readers.py,174,method,
1743,creator_fn,tensorflow/tensorflow/python/data/ops/readers.py,164,method,
1744,TextLineDatasetV1,tensorflow/tensorflow/python/data/ops/readers.py,179,class,A `Dataset` comprising lines from one or more text files.
1745,ParallelInterleaveDataset,tensorflow/tensorflow/python/data/ops/readers.py,235,class,A `Dataset` that maps a function over its input and flattens the result.
1746,element_spec,tensorflow/tensorflow/python/data/ops/readers.py,282,method,
1747,TFRecordDatasetV2,tensorflow/tensorflow/python/data/ops/readers.py,290,class,A `Dataset` comprising records from one or more TFRecord files.
1748,element_spec,tensorflow/tensorflow/python/data/ops/readers.py,349,method,
1749,creator_fn,tensorflow/tensorflow/python/data/ops/readers.py,327,method,
1750,TFRecordDatasetV1,tensorflow/tensorflow/python/data/ops/readers.py,354,class,A `Dataset` comprising records from one or more TFRecord files.
1751,FixedLengthRecordDatasetV2,tensorflow/tensorflow/python/data/ops/readers.py,439,class,A `Dataset` of fixed-length records from one or more binary files.
1752,element_spec,tensorflow/tensorflow/python/data/ops/readers.py,492,method,
1753,creator_fn,tensorflow/tensorflow/python/data/ops/readers.py,481,method,
1754,FixedLengthRecordDatasetV1,tensorflow/tensorflow/python/data/ops/readers.py,497,class,A `Dataset` of fixed-length records from one or more binary files.
1755,optional_param_to_tensor,tensorflow/tensorflow/python/data/util/convert.py,26,function,
1756,partial_shape_to_tensor,tensorflow/tensorflow/python/data/util/convert.py,38,function,"Returns a `tf.Tensor` that represents the given shape.
Args:
shape_like: A value that can be converted to a `tf.TensorShape` or a
`tf.Tensor`.
Returns:
A 1-D `tf.Tensor` of `tf.int64` elements representing the given shape, where
`-1` is substituted for any unknown dimensions."
1757,assert_same_structure,tensorflow/tensorflow/python/data/util/nest.py,104,function,"Asserts that two structures are nested in the same way.
Args:
nest1: an arbitrarily nested structure.
nest2: an arbitrarily nested structure.
check_types: if `True` (default) types of sequences should be same as
well. For dictionary, ""type"" of dictionary is considered to include its
keys. In other words, two dictionaries with different keys are considered
to have a different ""type"". If set to `False`, two iterables are
considered same as long as they yield the elements that have same
structures.
Raises:
ValueError: If the two structures do not have the same number of elements or
if the two structures are not nested in the same way.
TypeError: If the two structures differ in the type of sequence in any of
their substructures. Only possible if `check_types` is `True`."
1758,pack_sequence_as,tensorflow/tensorflow/python/data/util/nest.py,157,function,"Returns a given flattened sequence packed into a nest.
If `structure` is a scalar, `flat_sequence` must be a single-element list;
in this case the return value is `flat_sequence[0]`.
Args:
structure: tuple or list constructed of scalars and/or other tuples/lists,
or a scalar. Note: numpy arrays are considered scalars.
flat_sequence: flat sequence to pack.
Returns:
packed: `flat_sequence` converted to have the same recursive structure as
`structure`.
Raises:
ValueError: If nest and structure have different element counts."
1759,map_structure,tensorflow/tensorflow/python/data/util/nest.py,195,function,"Applies `func` to each entry in `structure` and returns a new structure.
Applies `func(x[0], x[1], ...)` where x[i] is an entry in
`structure[i]`. All structures in `structure` must have the same arity,
and the return value will contain the results in the same structure.
Args:
func: A callable that accepts as many arguments are there are structures.
*structure: scalar, or tuple or list of constructed scalars and/or other
tuples/lists, or scalars. Note: numpy arrays are considered scalars.
**check_types_dict: only valid keyword argument is `check_types`. If set to
`True` (default) the types of iterables within the structures have to be
same (e.g. `map_structure(func, [1], (1,))` raises a `TypeError`
exception). To allow this set this argument to `False`.
Returns:
A new structure with the same arity as `structure`, whose values correspond
to `func(x[0], x[1], ...)` where `x[i]` is a value in the corresponding
location in `structure[i]`. If there are different sequence types and
`check_types` is `False` the sequence types of the first structure will be
used.
Raises:
TypeError: If `func` is not callable or if the structures do not match
each other by depth tree.
ValueError: If no structure is provided or if the structures do not match
each other by type.
ValueError: If wrong keyword arguments are provided."
1760,assert_shallow_structure,tensorflow/tensorflow/python/data/util/nest.py,259,function,"Asserts that `shallow_tree` is a shallow structure of `input_tree`.
That is, this function tests if the `input_tree` structure can be created from
the `shallow_tree` structure by replacing its leaf nodes with deeper
tree structures.
Examples:
The following code will raise an exception:
```python
shallow_tree = [""a"", ""b""]
input_tree = [""c"", [""d"", ""e""], ""f""]
assert_shallow_structure(shallow_tree, input_tree)
```
The following code will not raise an exception:
```python
shallow_tree = [""a"", ""b""]
input_tree = [""c"", [""d"", ""e""]]
assert_shallow_structure(shallow_tree, input_tree)
```
Args:
shallow_tree: an arbitrarily nested structure.
input_tree: an arbitrarily nested structure.
check_types: if `True` (default) the sequence types of `shallow_tree` and
`input_tree` have to be the same.
Raises:
TypeError: If `shallow_tree` is a sequence but `input_tree` is not.
TypeError: If the sequence types of `shallow_tree` are different from
`input_tree`. Only raised if `check_types` is `True`.
ValueError: If the sequence lengths of `shallow_tree` are different from
`input_tree`."
1761,flatten_up_to,tensorflow/tensorflow/python/data/util/nest.py,327,function,"Flattens `input_tree` up to `shallow_tree`.
Any further depth in structure in `input_tree` is retained as elements in the
partially flatten output.
If `shallow_tree` and `input_tree` are not sequences, this returns a
single-element list: `[input_tree]`.
Use Case:
Sometimes we may wish to partially flatten a nested sequence, retaining some
of the nested structure. We achieve this by specifying a shallow structure,
`shallow_tree`, we wish to flatten up to.
The input, `input_tree`, can be thought of as having the same structure as
`shallow_tree`, but with leaf nodes that are themselves tree structures.
Examples:
```python
input_tree = [[[2, 2], [3, 3]], [[4, 9], [5, 5]]]
shallow_tree = [[True, True], [False, True]]
flattened_input_tree = flatten_up_to(shallow_tree, input_tree)
flattened_shallow_tree = flatten_up_to(shallow_tree, shallow_tree)
# Output is:
# [[2, 2], [3, 3], [4, 9], [5, 5]]
# [True, True, False, True]
```
```python
input_tree = [[('a', 1), [('b', 2), [('c', 3), [('d', 4)]]]]]
shallow_tree = [['level_1', ['level_2', ['level_3', ['level_4']]]]]
input_tree_flattened_as_shallow_tree = flatten_up_to(shallow_tree, input_tree)
input_tree_flattened = flatten(input_tree)
# Output is:
# [('a', 1), ('b', 2), ('c', 3), ('d', 4)]
# ['a', 1, 'b', 2, 'c', 3, 'd', 4]
```
Non-Sequence Edge Cases:
```python
flatten_up_to(0, 0) # Output: [0]
flatten_up_to(0, [0, 1, 2]) # Output: [[0, 1, 2]]
flatten_up_to([0, 1, 2], 0) # Output: TypeError
flatten_up_to([0, 1, 2], [0, 1, 2]) # Output: [0, 1, 2]
```
Args:
shallow_tree: a possibly pruned structure of input_tree.
input_tree: an arbitrarily nested structure or a scalar object.
Note, numpy arrays are considered scalars.
Returns:
A Python list, the partially flattened version of `input_tree` according to
the structure of `shallow_tree`.
Raises:
TypeError: If `shallow_tree` is a sequence but `input_tree` is not.
TypeError: If the sequence types of `shallow_tree` are different from
`input_tree`.
ValueError: If the sequence lengths of `shallow_tree` are different from
`input_tree`."
1762,map_structure_up_to,tensorflow/tensorflow/python/data/util/nest.py,400,function,"Applies a function or op to a number of partially flattened inputs.
The `inputs` are flattened up to `shallow_tree` before being mapped.
Use Case:
Sometimes we wish to apply a function to a partially flattened
sequence (for example when the function itself takes sequence inputs). We
achieve this by specifying a shallow structure, `shallow_tree` we wish to
flatten up to.
The `inputs`, can be thought of as having the same structure as
`shallow_tree`, but with leaf nodes that are themselves tree structures.
This function, therefore, will return something with the same base structure
as `shallow_tree`.
Examples:
```python
ab_tuple = collections.namedtuple(""ab_tuple"", ""a, b"")
op_tuple = collections.namedtuple(""op_tuple"", ""add, mul"")
inp_val = ab_tuple(a=2, b=3)
inp_ops = ab_tuple(a=op_tuple(add=1, mul=2), b=op_tuple(add=2, mul=3))
out = map_structure_up_to(inp_val, lambda val, ops: (val + ops.add) * ops.mul,
inp_val, inp_ops)
# Output is: ab_tuple(a=6, b=15)
```
```python
data_list = [[2, 4, 6, 8], [[1, 3, 5, 7, 9], [3, 5, 7]]]
name_list = ['evens', ['odds', 'primes']]
out = map_structure_up_to(
name_list,
lambda name, sec: ""first_{}_{}"".format(len(sec), name),
name_list, data_list)
# Output is: ['first_4_evens', ['first_5_odds', 'first_3_primes']]
```
Args:
shallow_tree: a shallow tree, common to all the inputs.
func: callable which will be applied to each input individually.
*inputs: arbitrarily nested combination of objects that are compatible with
shallow_tree. The function `func` is applied to corresponding
partially flattened elements of each input, so the function must support
arity of `len(inputs)`.
Raises:
TypeError: If `shallow_tree` is a sequence but `input_tree` is not.
TypeError: If the sequence types of `shallow_tree` are different from
`input_tree`.
ValueError: If the sequence lengths of `shallow_tree` are different from
`input_tree`.
Returns:
result of repeatedly applying `func`, with same structure as
`shallow_tree`."
1763,OptionsBase,tensorflow/tensorflow/python/data/util/options.py,26,class,"Base class for representing a set of tf.data options.
Attributes:
_options: Stores the option values."
1764,create_option,tensorflow/tensorflow/python/data/util/options.py,59,function,"Creates a type-checked property.
Args:
name: The name to use.
ty: The type to use. The type of the property will be validated when it
is set.
docstring: The docstring to use.
default_factory: A callable that takes no arguments and returns a default
value to use if not set.
Returns:
A type-checked property."
1765,merge_options,tensorflow/tensorflow/python/data/util/options.py,89,function,"Merges the given options, returning the result as a new options object.
The input arguments are expected to have a matching type that derives from
`OptionsBase` (and thus each represent a set of options). The method outputs
an object of the same type created by merging the sets of options represented
by the input arguments.
The sets of options can be merged as long as there does not exist an option
with different non-default values.
If an option is an instance of `OptionsBase` itself, then this method is
applied recursively to the set of options represented by this option.
Args:
*options_list: options to merge
Raises:
TypeError: if the input arguments are incompatible or not derived from
`OptionsBase`
ValueError: if the given options cannot be merged
Returns:
A new options object which is the result of merging the given options."
1766,get_seed,tensorflow/tensorflow/python/data/util/random_seed.py,29,function,"Returns the local seeds an operation should use given an op-specific seed.
See `random_seed.get_seed` for more details. This wrapper adds support for
the case where `seed` may be a tensor.
Args:
seed: An integer or a `tf.int64` scalar tensor.
Returns:
A tuple of two `tf.int64` scalar tensors that should be used for the local
seed of the calling dataset."
1767,any_sparse,tensorflow/tensorflow/python/data/util/sparse.py,28,function,"Checks for sparse tensor.
Args:
classes: a structure of objects that identify the dataset item classes
Returns:
`True` if `classes` contains a sparse tensor type and `False` otherwise."
1768,as_dense_shapes,tensorflow/tensorflow/python/data/util/sparse.py,40,function,"Converts sparse tensor shapes to their physical shapes.
Args:
shapes: a structure of shapes to convert.
classes: a structure of objects that identify the dataset item classes
Returns:
a structure matching the nested structure of `shapes`, containing
`tensor_shape.unknown_shape()` at positions where `classes` contains
`tf.sparse.SparseTensor` and matching contents of `shapes` otherwise"
1769,as_dense_types,tensorflow/tensorflow/python/data/util/sparse.py,59,function,"Converts sparse tensor types to `dtypes.variant`.
Args:
types: a structure of types to convert.
classes: a structure of objects that identify the dataset item classes
Returns:
a structure matching the nested structure of `types`, containing
`dtypes.variant` at positions where `classes` contains
`tf.sparse.SparseTensor` and matching contents of `types` otherwise"
1770,deserialize_sparse_tensors,tensorflow/tensorflow/python/data/util/sparse.py,78,function,"Deserializes sparse tensors.
Args:
tensors: a structure of tensors to deserialize.
types: a structure that holds information about types of `tensors`
shapes: a structure that holds information about shapes of `tensors`
classes: a structure of objects that identify the dataset item classes
Returns:
`tensors` with any serialized sparse tensors replaced by their deserialized
version."
1771,get_classes,tensorflow/tensorflow/python/data/util/sparse.py,101,function,"Gets classes for a structure of tensors.
Args:
tensors: the tensor structure to get classes for.
Returns:
a structure matching the nested structure of `tensors`, containing
`tf.sparse.SparseTensor` at positions where `tensors` contains a sparse
tensor and `tf.Tensor` otherwise."
1772,serialize_many_sparse_tensors,tensorflow/tensorflow/python/data/util/sparse.py,119,function,"Serializes many sparse tensors into a batch.
Args:
tensors: a tensor structure to serialize.
Returns:
`tensors` with any sparse tensors replaced by the serialized batch."
1773,serialize_sparse_tensors,tensorflow/tensorflow/python/data/util/sparse.py,137,function,"Serializes sparse tensors.
Args:
tensors: a tensor structure to serialize.
Returns:
`tensors` with any sparse tensors replaced by their serialized version."
1774,normalize_element,tensorflow/tensorflow/python/data/util/structure.py,70,function,"Normalizes a nested structure of element components.
* Components matching `SparseTensorSpec` are converted to `SparseTensor`.
* Components matching `RaggedTensorSpec` are converted to `RaggedTensor`.
* Components matching `DatasetSpec` or `TensorArraySpec` are passed through.
* `CompositeTensor` components are passed through.
* All other components are converted to `Tensor`.
Args:
element: A nested structure of individual components.
Returns:
A nested structure of `Tensor`, `Dataset`, `SparseTensor`, `RaggedTensor`,
or `TensorArray` objects."
1775,convert_legacy_structure,tensorflow/tensorflow/python/data/util/structure.py,119,function,"Returns a `Structure` that represents the given legacy structure.
This method provides a way to convert from the existing `Dataset` and
`Iterator` structure-related properties to a `Structure` object. A ""legacy""
structure is represented by the `tf.data.Dataset.output_types`,
`tf.data.Dataset.output_shapes`, and `tf.data.Dataset.output_classes`
properties.
TODO(b/110122868): Remove this function once `Structure` is used throughout
`tf.data`.
Args:
output_types: A nested structure of `tf.DType` objects corresponding to
each component of a structured value.
output_shapes: A nested structure of `tf.TensorShape` objects
corresponding to each component a structured value.
output_classes: A nested structure of Python `type` objects corresponding
to each component of a structured value.
Returns:
A `Structure`.
Raises:
TypeError: If a structure cannot be built from the arguments, because one of
the component classes in `output_classes` is not supported."
1776,from_compatible_tensor_list,tensorflow/tensorflow/python/data/util/structure.py,210,function,"Returns an element constructed from the given spec and tensor list.
Args:
element_spec: A nested structure of `tf.TypeSpec` objects representing to
element type specification.
tensor_list: A list of tensors to use for constructing the value.
Returns:
An element constructed from the given spec and tensor list.
Raises:
ValueError: If the number of tensors needed to construct an element for
the given spec does not match the given number of tensors."
1777,from_tensor_list,tensorflow/tensorflow/python/data/util/structure.py,233,function,"Returns an element constructed from the given spec and tensor list.
Args:
element_spec: A nested structure of `tf.TypeSpec` objects representing to
element type specification.
tensor_list: A list of tensors to use for constructing the value.
Returns:
An element constructed from the given spec and tensor list.
Raises:
ValueError: If the number of tensors needed to construct an element for
the given spec does not match the given number of tensors or the given
spec is not compatible with the tensor list."
1778,get_flat_tensor_specs,tensorflow/tensorflow/python/data/util/structure.py,257,function,"Returns a list `tf.TypeSpec`s for the element tensor representation.
Args:
element_spec: A nested structure of `tf.TypeSpec` objects representing to
element type specification.
Returns:
A list `tf.TypeSpec`s for the element tensor representation."
1779,get_flat_tensor_shapes,tensorflow/tensorflow/python/data/util/structure.py,273,function,"Returns a list `tf.TensorShapes`s for the element tensor representation.
Args:
element_spec: A nested structure of `tf.TypeSpec` objects representing to
element type specification.
Returns:
A list `tf.TensorShapes`s for the element tensor representation."
1780,get_flat_tensor_types,tensorflow/tensorflow/python/data/util/structure.py,286,function,"Returns a list `tf.DType`s for the element tensor representation.
Args:
element_spec: A nested structure of `tf.TypeSpec` objects representing to
element type specification.
Returns:
A list `tf.DType`s for the element tensor representation."
1781,to_batched_tensor_list,tensorflow/tensorflow/python/data/util/structure.py,329,function,"Returns a tensor list representation of the element.
Args:
element_spec: A nested structure of `tf.TypeSpec` objects representing to
element type specification.
element: The element to convert to tensor list representation.
Returns:
A tensor list representation of `element`.
Raises:
ValueError: If `element_spec` and `element` do not have the same number of
elements or if the two structures are not nested in the same way or the
rank of any of the tensors in the tensor list representation is 0.
TypeError: If `element_spec` and `element` differ in the type of sequence
in any of their substructures."
1782,to_tensor_list,tensorflow/tensorflow/python/data/util/structure.py,355,function,"Returns a tensor list representation of the element.
Args:
element_spec: A nested structure of `tf.TypeSpec` objects representing to
element type specification.
element: The element to convert to tensor list representation.
Returns:
A tensor list representation of `element`.
Raises:
ValueError: If `element_spec` and `element` do not have the same number of
elements or if the two structures are not nested in the same way.
TypeError: If `element_spec` and `element` differ in the type of sequence
in any of their substructures."
1783,are_compatible,tensorflow/tensorflow/python/data/util/structure.py,380,function,"Indicates whether two type specifications are compatible.
Two type specifications are compatible if they have the same nested structure
and the their individual components are pair-wise compatible.
Args:
spec1: A `tf.TypeSpec` object to compare.
spec2: A `tf.TypeSpec` object to compare.
Returns:
`True` if the two type specifications are compatible and `False` otherwise."
1784,type_spec_from_value,tensorflow/tensorflow/python/data/util/structure.py,407,function,"Creates a type specification for the given value.
Args:
element: The element to create the type specification for.
use_fallback: Whether to fall back to converting the element to a tensor
in order to compute its `TypeSpec`.
Returns:
A nested structure of `TypeSpec`s that represents the type specification
of `element`.
Raises:
TypeError: If a `TypeSpec` cannot be built for `element`, because its type
is not supported."
1785,NoneTensor,tensorflow/tensorflow/python/data/util/structure.py,471,class,Composite tensor representation for `None` value.
1786,NoneTensorSpec,tensorflow/tensorflow/python/data/util/structure.py,481,class,Type specification for `None` value.
1787,value_type,tensorflow/tensorflow/python/data/util/structure.py,485,method,
1788,from_value,tensorflow/tensorflow/python/data/util/structure.py,505,method,
1789,CustomMap,tensorflow/tensorflow/python/data/util/structure_test.py,759,class,"Custom, immutable map."
1790,obtain_all_variant_tensor_ops,tensorflow/tensorflow/python/data/util/traverse.py,25,function,"Given an input dataset, finds all dataset ops used for construction.
A series of transformations would have created this dataset with each
transformation including zero or more Dataset ops, each producing a dataset
variant tensor. This method outputs all of them.
Args:
dataset: Dataset to find variant tensors for.
Returns:
A list of variant_tensor producing dataset ops used to construct this
dataset."
1791,DebugAnalyzer,tensorflow/tensorflow/python/debug/cli/analyzer_cli.py,130,class,Analyzer for debug data from dump directories.
1792,add_tensor_filter,tensorflow/tensorflow/python/debug/cli/analyzer_cli.py,420,method,"Add a tensor filter.
A tensor filter is a named callable of the signature:
filter_callable(dump_datum, tensor),
wherein dump_datum is an instance of debug_data.DebugTensorDatum carrying
metadata about the dumped tensor, including tensor name, timestamps, etc.
tensor is the value of the dumped tensor as an numpy.ndarray object.
The return value of the function is a bool.
This is the same signature as the input argument to
debug_data.DebugDumpDir.find().
Args:
filter_name: (str) name of the filter. Cannot be empty.
filter_callable: (callable) a filter function of the signature described
as above.
Raises:
ValueError: If filter_name is an empty str.
TypeError: If filter_name is not a str.
Or if filter_callable is not callable."
1793,get_tensor_filter,tensorflow/tensorflow/python/debug/cli/analyzer_cli.py,460,method,"Retrieve filter function by name.
Args:
filter_name: Name of the filter set during add_tensor_filter() call.
Returns:
The callable associated with the filter name.
Raises:
ValueError: If there is no tensor filter of the specified filter name."
1794,get_help,tensorflow/tensorflow/python/debug/cli/analyzer_cli.py,478,method,
1795,list_tensors,tensorflow/tensorflow/python/debug/cli/analyzer_cli.py,481,method,"Command handler for list_tensors.
List tensors dumped during debugged Session.run() call.
Args:
args: Command-line arguments, excluding the command prefix, as a list of
str.
screen_info: Optional dict input containing screen information such as
cols.
Returns:
Output text lines as a RichTextLines object.
Raises:
ValueError: If `--filter_exclude_node_names` is used without `-f` or
`--tensor_filter` being used."
1796,node_info,tensorflow/tensorflow/python/debug/cli/analyzer_cli.py,740,method,"Command handler for node_info.
Query information about a given node.
Args:
args: Command-line arguments, excluding the command prefix, as a list of
str.
screen_info: Optional dict input containing screen information such as
cols.
Returns:
Output text lines as a RichTextLines object."
1797,list_inputs,tensorflow/tensorflow/python/debug/cli/analyzer_cli.py,875,method,"Command handler for inputs.
Show inputs to a given node.
Args:
args: Command-line arguments, excluding the command prefix, as a list of
str.
screen_info: Optional dict input containing screen information such as
cols.
Returns:
Output text lines as a RichTextLines object."
1798,print_tensor,tensorflow/tensorflow/python/debug/cli/analyzer_cli.py,911,method,"Command handler for print_tensor.
Print value of a given dumped tensor.
Args:
args: Command-line arguments, excluding the command prefix, as a list of
str.
screen_info: Optional dict input containing screen information such as
cols.
Returns:
Output text lines as a RichTextLines object."
1799,list_outputs,tensorflow/tensorflow/python/debug/cli/analyzer_cli.py,1053,method,"Command handler for inputs.
Show inputs to a given node.
Args:
args: Command-line arguments, excluding the command prefix, as a list of
str.
screen_info: Optional dict input containing screen information such as
cols.
Returns:
Output text lines as a RichTextLines object."
1800,evaluate_expression,tensorflow/tensorflow/python/debug/cli/analyzer_cli.py,1089,method,
1801,print_source,tensorflow/tensorflow/python/debug/cli/analyzer_cli.py,1112,method,Print the content of a source file.
1802,list_source,tensorflow/tensorflow/python/debug/cli/analyzer_cli.py,1243,method,List Python source files that constructed nodes and tensors.
1803,create_analyzer_ui,tensorflow/tensorflow/python/debug/cli/analyzer_cli.py,1583,function,"Create an instance of CursesUI based on a DebugDumpDir object.
Args:
debug_dump: (debug_data.DebugDumpDir) The debug dump to use.
tensor_filters: (dict) A dict mapping tensor filter name (str) to tensor
filter (Callable).
ui_type: (str) requested UI type, e.g., ""curses"", ""readline"".
on_ui_exit: (`Callable`) the callback to be called when the UI exits.
config: A `cli_config.CLIConfig` object.
Returns:
(base_ui.BaseUI) A BaseUI subtype object with a set of standard analyzer
commands and tab-completions registered."
1804,no_rewrite_session_config,tensorflow/tensorflow/python/debug/cli/analyzer_cli_test.py,62,function,
1805,line_number_above,tensorflow/tensorflow/python/debug/cli/analyzer_cli_test.py,74,function,
1806,parse_op_and_node,tensorflow/tensorflow/python/debug/cli/analyzer_cli_test.py,78,function,"Parse a line containing an op node followed by a node name.
For example, if the line is
"" [Variable] hidden/weights"",
this function will return (""Variable"", ""hidden/weights"")
Args:
line: The line to be parsed, as a str.
Returns:
Name of the parsed op type.
Name of the parsed node."
1807,assert_column_header_command_shortcut,tensorflow/tensorflow/python/debug/cli/analyzer_cli_test.py,102,function,
1808,assert_listed_tensors,tensorflow/tensorflow/python/debug/cli/analyzer_cli_test.py,116,function,"Check RichTextLines output for list_tensors commands.
Args:
tst: A test_util.TensorFlowTestCase instance.
out: The RichTextLines object to be checked.
expected_tensor_names: (list of str) Expected tensor names in the list.
expected_op_types: (list of str) Expected op types of the tensors, in the
same order as the expected_tensor_names.
node_name_regex: Optional: node name regex filter.
op_type_regex: Optional: op type regex filter.
tensor_filter_name: Optional: name of the tensor filter.
sort_by: (str) (timestamp | op_type | tensor_name) the field by which the
tensors in the list are sorted.
reverse: (bool) whether the sorting is in reverse (i.e., descending) order."
1809,assert_node_attribute_lines,tensorflow/tensorflow/python/debug/cli/analyzer_cli_test.py,261,function,"Check RichTextLines output for node_info commands.
Args:
tst: A test_util.TensorFlowTestCase instance.
out: The RichTextLines object to be checked.
node_name: Name of the node.
op_type: Op type of the node, as a str.
device: Name of the device on which the node resides.
input_op_type_node_name_pairs: A list of 2-tuples of op type and node name,
for the (non-control) inputs to the node.
ctrl_input_op_type_node_name_pairs: A list of 2-tuples of op type and node
name, for the control inputs to the node.
recipient_op_type_node_name_pairs: A list of 2-tuples of op type and node
name, for the (non-control) output recipients to the node.
ctrl_recipient_op_type_node_name_pairs: A list of 2-tuples of op type and
node name, for the control output recipients to the node.
attr_key_val_pairs: Optional: attribute key-value pairs of the node, as a
list of 2-tuples.
num_dumped_tensors: Optional: number of tensor dumps from the node.
show_stack_trace: (bool) whether the stack trace of the node's
construction is asserted to be present.
stack_trace_available: (bool) whether Python stack trace is available."
1810,check_syntax_error_output,tensorflow/tensorflow/python/debug/cli/analyzer_cli_test.py,425,function,Check RichTextLines output for valid command prefix but invalid syntax.
1811,check_error_output,tensorflow/tensorflow/python/debug/cli/analyzer_cli_test.py,434,function,"Check RichTextLines output from invalid/erroneous commands.
Args:
tst: A test_util.TensorFlowTestCase instance.
out: The RichTextLines object to be checked.
command_prefix: The command prefix of the command that caused the error.
args: The arguments (excluding prefix) of the command that caused the error."
1812,check_menu_item,tensorflow/tensorflow/python/debug/cli/analyzer_cli_test.py,497,function,
1813,create_analyzer_cli,tensorflow/tensorflow/python/debug/cli/analyzer_cli_test.py,514,function,"Create an analyzer CLI.
Args:
dump: A `DebugDumpDir` object to base the analyzer CLI on.
Returns:
1) A `DebugAnalyzer` object created based on `dump`.
2) A `CommandHandlerRegistry` that is based on the `DebugAnalyzer` object
and has the common tfdbg commands, e.g., lt, ni, li, lo, registered."
1814,BaseUI,tensorflow/tensorflow/python/debug/cli/base_ui.py,27,class,Base class of tfdbg user interface.
1815,set_help_intro,tensorflow/tensorflow/python/debug/cli/base_ui.py,72,method,"Set an introductory message to the help output of the command registry.
Args:
help_intro: (RichTextLines) Rich text lines appended to the beginning of
the output of the command ""help"", as introductory information."
1816,register_command_handler,tensorflow/tensorflow/python/debug/cli/base_ui.py,82,method,"A wrapper around CommandHandlerRegistry.register_command_handler().
In addition to calling the wrapped register_command_handler() method, this
method also registers the top-level tab-completion context based on the
command prefixes and their aliases.
See the doc string of the wrapped method for more details on the args.
Args:
prefix: (str) command prefix.
handler: (callable) command handler.
help_info: (str) help information.
prefix_aliases: (list of str) aliases of the command prefix."
1817,register_tab_comp_context,tensorflow/tensorflow/python/debug/cli/base_ui.py,109,method,Wrapper around TabCompletionRegistry.register_tab_comp_context().
1818,run_ui,tensorflow/tensorflow/python/debug/cli/base_ui.py,114,method,"Run the UI until user- or command- triggered exit.
Args:
init_command: (str) Optional command to run on CLI start up.
title: (str) Optional title to display in the CLI.
title_color: (str) Optional color of the title, e.g., ""yellow"".
enable_mouse_on_start: (bool) Whether the mouse mode is to be enabled on
start-up.
Returns:
An exit token of arbitrary type. Can be None."
1819,config,tensorflow/tensorflow/python/debug/cli/base_ui.py,200,method,Obtain the CLIConfig of this `BaseUI` instance.
1820,CLIConfig,tensorflow/tensorflow/python/debug/cli/cli_config.py,30,class,Client-facing configurations for TFDBG command-line interfaces.
1821,get,tensorflow/tensorflow/python/debug/cli/cli_config.py,52,method,
1822,set,tensorflow/tensorflow/python/debug/cli/cli_config.py,57,method,"Set the value of a property.
Supports limitd property value types: `bool`, `int` and `str`.
Args:
property_name: Name of the property.
property_val: Value of the property. If the property has `bool` type and
this argument has `str` type, the `str` value will be parsed as a `bool`
Raises:
ValueError: if a `str` property_value fails to be parsed as a `bool`.
KeyError: if `property_name` is an invalid property name."
1823,set_callback,tensorflow/tensorflow/python/debug/cli/cli_config.py,99,method,"Set a set-callback for given property.
Args:
property_name: Name of the property.
callback: The callback as a `callable` of signature:
def cbk(config):
where config is the config after it is set to the new value.
The callback is invoked each time the set() method is called with the
matching property_name.
Raises:
KeyError: If property_name does not exist.
TypeError: If `callback` is not callable."
1824,summarize,tensorflow/tensorflow/python/debug/cli/cli_config.py,130,method,"Get a text summary of the config.
Args:
highlight: A property name to highlight in the output.
Returns:
A `RichTextLines` output."
1825,bytes_to_readable_str,tensorflow/tensorflow/python/debug/cli/cli_shared.py,55,function,"Generate a human-readable string representing number of bytes.
The units B, kB, MB and GB are used.
Args:
num_bytes: (`int` or None) Number of bytes.
include_b: (`bool`) Include the letter B at the end of the unit.
Returns:
(`str`) A string representing the number of bytes in a human-readable way,
including a unit at the end."
1826,time_to_readable_str,tensorflow/tensorflow/python/debug/cli/cli_shared.py,85,function,"Convert time value to human-readable string.
Args:
value_us: time value in microseconds.
force_time_unit: force the output to use the specified time unit. Must be
in TIME_UNITS.
Returns:
Human-readable string representation of the time value.
Raises:
ValueError: if force_time_unit value is not in TIME_UNITS."
1827,parse_ranges_highlight,tensorflow/tensorflow/python/debug/cli/cli_shared.py,113,function,"Process ranges highlight string.
Args:
ranges_string: (str) A string representing a numerical range of a list of
numerical ranges. See the help info of the -r flag of the print_tensor
command for more details.
Returns:
An instance of tensor_format.HighlightOptions, if range_string is a valid
representation of a range or a list of ranges."
1828,numpy_printoptions_from_screen_info,tensorflow/tensorflow/python/debug/cli/cli_shared.py,143,function,
1829,format_tensor,tensorflow/tensorflow/python/debug/cli/cli_shared.py,150,function,"Generate formatted str to represent a tensor or its slices.
Args:
tensor: (numpy ndarray) The tensor value.
tensor_name: (str) Name of the tensor, e.g., the tensor's debug watch key.
np_printoptions: (dict) Numpy tensor formatting options.
print_all: (bool) Whether the tensor is to be displayed in its entirety,
instead of printing ellipses, even if its number of elements exceeds
the default numpy display threshold.
(Note: Even if this is set to true, the screen output can still be cut
off by the UI frontend if it consist of more lines than the frontend
can handle.)
tensor_slicing: (str or None) Slicing of the tensor, e.g., ""[:, 1]"". If
None, no slicing will be performed on the tensor.
highlight_options: (tensor_format.HighlightOptions) options to highlight
elements of the tensor. See the doc of tensor_format.format_tensor()
for more details.
include_numeric_summary: Whether a text summary of the numeric values (if
applicable) will be included.
write_path: A path to save the tensor value (after any slicing) to
(optional). `numpy.save()` is used to save the value.
Returns:
An instance of `debugger_cli_common.RichTextLines` representing the
(potentially sliced) tensor."
1830,error,tensorflow/tensorflow/python/debug/cli/cli_shared.py,218,function,"Generate a RichTextLines output for error.
Args:
msg: (str) The error message.
Returns:
(debugger_cli_common.RichTextLines) A representation of the error message
for screen output."
1831,get_tfdbg_logo,tensorflow/tensorflow/python/debug/cli/cli_shared.py,261,function,Make an ASCII representation of the tfdbg logo.
1832,get_run_start_intro,tensorflow/tensorflow/python/debug/cli/cli_shared.py,279,function,"Generate formatted intro for run-start UI.
Args:
run_call_count: (int) Run call counter.
fetches: Fetches of the `Session.run()` call. See doc of `Session.run()`
for more details.
feed_dict: Feeds to the `Session.run()` call. See doc of `Session.run()`
for more details.
tensor_filters: (dict) A dict from tensor-filter name to tensor-filter
callable.
is_callable_runner: (bool) whether a runner returned by
Session.make_callable is being run.
Returns:
(RichTextLines) Formatted intro message about the `Session.run()` call."
1833,get_run_short_description,tensorflow/tensorflow/python/debug/cli/cli_shared.py,386,function,"Get a short description of the run() call.
Args:
run_call_count: (int) Run call counter.
fetches: Fetches of the `Session.run()` call. See doc of `Session.run()`
for more details.
feed_dict: Feeds to the `Session.run()` call. See doc of `Session.run()`
for more details.
is_callable_runner: (bool) whether a runner returned by
Session.make_callable is being run.
Returns:
(str) A short description of the run() call, including information about
the fetche(s) and feed(s)."
1834,get_error_intro,tensorflow/tensorflow/python/debug/cli/cli_shared.py,434,function,"Generate formatted intro for TensorFlow run-time error.
Args:
tf_error: (errors.OpError) TensorFlow run-time error object.
Returns:
(RichTextLines) Formatted intro message about the run-time OpError, with
sample commands for debugging."
1835,assert_lines_equal_ignoring_whitespace,tensorflow/tensorflow/python/debug/cli/cli_test_utils.py,25,function,"Assert equality in lines, ignoring all whitespace.
Args:
test: An instance of unittest.TestCase or its subtypes (e.g.,
TensorFlowTestCase).
expected_lines: Expected lines as an iterable of strings.
actual_lines: Actual lines as an iterable of strings."
1836,assert_array_lines_close,tensorflow/tensorflow/python/debug/cli/cli_test_utils.py,48,function,"Assert that the array value represented by lines is close to expected.
Note that the shape of the array represented by the `array_lines` is ignored.
Args:
test: An instance of TensorFlowTestCase.
expected_array: Expected value of the array.
array_lines: A list of strings representing the array.
E.g., ""array([[ 1.0, 2.0 ], [ 3.0, 4.0 ]])""
Assumes that values are separated by commas, parentheses, brackets, ""|""
characters and whitespace."
1837,Interval,tensorflow/tensorflow/python/debug/cli/command_parser.py,33,class,Represents an interval between a start and end value.
1838,contains,tensorflow/tensorflow/python/debug/cli/command_parser.py,42,method,
1839,parse_command,tensorflow/tensorflow/python/debug/cli/command_parser.py,56,function,"Parse command string into a list of arguments.
- Disregards whitespace inside double quotes and brackets.
- Strips paired leading and trailing double quotes in arguments.
- Splits the command at whitespace.
Nested double quotes and brackets are not handled.
Args:
command: (str) Input command.
Returns:
(list of str) List of arguments."
1840,extract_output_file_path,tensorflow/tensorflow/python/debug/cli/command_parser.py,104,function,"Extract output file path from command arguments.
Args:
args: (list of str) command arguments.
Returns:
(list of str) Command arguments with the output file path part stripped.
(str or None) Output file path (if any).
Raises:
SyntaxError: If there is no file path after the last "">"" character."
1841,parse_tensor_name_with_slicing,tensorflow/tensorflow/python/debug/cli/command_parser.py,151,function,"Parse tensor name, potentially suffixed by slicing string.
Args:
in_str: (str) Input name of the tensor, potentially followed by a slicing
string. E.g.: Without slicing string: ""hidden/weights/Variable:0"", with
slicing string: ""hidden/weights/Variable:0[1, :]""
Returns:
(str) name of the tensor
(str) slicing string, if any. If no slicing string is present, return """"."
1842,validate_slicing_string,tensorflow/tensorflow/python/debug/cli/command_parser.py,174,function,"Validate a slicing string.
Check if the input string contains only brackets, digits, commas and
colons that are valid characters in numpy-style array slicing.
Args:
slicing_string: (str) Input slicing string to be validated.
Returns:
(bool) True if and only if the slicing string is valid."
1843,parse_indices,tensorflow/tensorflow/python/debug/cli/command_parser.py,219,function,"Parse a string representing indices.
For example, if the input is ""[1, 2, 3]"", the return value will be a list of
indices: [1, 2, 3]
Args:
indices_string: (str) a string representing indices. Can optionally be
surrounded by a pair of brackets.
Returns:
(list of int): Parsed indices."
1844,parse_ranges,tensorflow/tensorflow/python/debug/cli/command_parser.py,243,function,"Parse a string representing numerical range(s).
Args:
range_string: (str) A string representing a numerical range or a list of
them. For example:
""[-1.0,1.0]"", ""[-inf, 0]"", ""[[-inf, -1.0], [1.0, inf]]""
Returns:
(list of list of float) A list of numerical ranges parsed from the input
string.
Raises:
ValueError: If the input doesn't represent a range or a list of ranges."
1845,parse_memory_interval,tensorflow/tensorflow/python/debug/cli/command_parser.py,284,function,"Convert a human-readable memory interval to a tuple of start and end value.
Args:
interval_str: (`str`) A human-readable str representing an interval
(e.g., ""[10kB, 20kB]"", ""<100M"", "">100G""). Only the units ""kB"", ""MB"", ""GB""
are supported. The ""B character at the end of the input `str` may be
omitted.
Returns:
`Interval` object where start and end are in bytes.
Raises:
ValueError: if the input is not valid."
1846,parse_time_interval,tensorflow/tensorflow/python/debug/cli/command_parser.py,314,function,"Convert a human-readable time interval to a tuple of start and end value.
Args:
interval_str: (`str`) A human-readable str representing an interval
(e.g., ""[10us, 20us]"", ""<100s"", "">100ms""). Supported time suffixes are
us, ms, s.
Returns:
`Interval` object where start and end are in microseconds.
Raises:
ValueError: if the input is not valid."
1847,parse_readable_size_str,tensorflow/tensorflow/python/debug/cli/command_parser.py,409,function,"Convert a human-readable str representation to number of bytes.
Only the units ""kB"", ""MB"", ""GB"" are supported. The ""B character at the end
of the input `str` may be omitted.
Args:
size_str: (`str`) A human-readable str representing a number of bytes
(e.g., ""0"", ""1023"", ""1.1kB"", ""24 MB"", ""23GB"", ""100 G"".
Returns:
(`int`) The parsed number of bytes.
Raises:
ValueError: on failure to parse the input `size_str`."
1848,parse_readable_time_str,tensorflow/tensorflow/python/debug/cli/command_parser.py,443,function,"Parses a time string in the format N, Nus, Nms, Ns.
Args:
time_str: (`str`) string consisting of an integer time value optionally
followed by 'us', 'ms', or 's' suffix. If suffix is not specified,
value is assumed to be in microseconds. (e.g. 100us, 8ms, 5s, 100).
Returns:
Microseconds value."
1849,evaluate_tensor_slice,tensorflow/tensorflow/python/debug/cli/command_parser.py,471,function,"Call eval on the slicing of a tensor, with validation.
Args:
tensor: (numpy ndarray) The tensor value.
tensor_slicing: (str or None) Slicing of the tensor, e.g., ""[:, 1]"". If
None, no slicing will be performed on the tensor.
Returns:
(numpy ndarray) The sliced tensor.
Raises:
ValueError: If tensor_slicing is not a valid numpy ndarray slicing str."
1850,get_print_tensor_argparser,tensorflow/tensorflow/python/debug/cli/command_parser.py,494,function,"Get an ArgumentParser for a command that prints tensor values.
Examples of such commands include print_tensor and print_feed.
Args:
description: Description of the ArgumentParser.
Returns:
An instance of argparse.ArgumentParser."
1851,ParseInterval,tensorflow/tensorflow/python/debug/cli/command_parser_test.py,391,class,
1852,ScrollBar,tensorflow/tensorflow/python/debug/cli/curses_ui.py,71,class,"Vertical ScrollBar for Curses-based CLI.
An object of this class has knowledge of the location of the scroll bar
in the screen coordinates, the current scrolling position, and the total
number of text lines in the screen text. By using this information, it
can generate text rendering of the scroll bar, which consists of and UP
button on the top and a DOWN button on the bottom, in addition to a scroll
block in between, whose exact location is determined by the scrolling
position. The object can also calculate the scrolling command (e.g.,
_SCROLL_UP_A_LINE, _SCROLL_DOWN) from the coordinate of a mouse click
event in the screen region it occupies."
1853,layout,tensorflow/tensorflow/python/debug/cli/curses_ui.py,154,method,"Get the RichTextLines layout of the scroll bar.
Returns:
(debugger_cli_common.RichTextLines) The text layout of the scroll bar."
1854,get_click_command,tensorflow/tensorflow/python/debug/cli/curses_ui.py,192,method,
1855,CursesUI,tensorflow/tensorflow/python/debug/cli/curses_ui.py,209,class,"Curses-based Command-line UI.
In this class, the methods with the prefix ""_screen_"" are the methods that
interact with the actual terminal using the curses library."
1856,run_ui,tensorflow/tensorflow/python/debug/cli/curses_ui.py,487,method,Run the CLI: See the doc of base_ui.BaseUI.run_ui for more details.
1857,get_help,tensorflow/tensorflow/python/debug/cli/curses_ui.py,520,method,
1858,string_to_codes,tensorflow/tensorflow/python/debug/cli/curses_ui_test.py,39,function,
1859,codes_to_string,tensorflow/tensorflow/python/debug/cli/curses_ui_test.py,43,function,
1860,MockCursesUI,tensorflow/tensorflow/python/debug/cli/curses_ui_test.py,48,class,Mock subclass of CursesUI that bypasses actual terminal manipulations.
1861,NavigationHistoryItem,tensorflow/tensorflow/python/debug/cli/curses_widgets.py,26,class,Individual item in navigation history.
1862,CursesNavigationHistory,tensorflow/tensorflow/python/debug/cli/curses_widgets.py,42,class,"Navigation history containing commands, outputs and scroll info."
1863,add_item,tensorflow/tensorflow/python/debug/cli/curses_widgets.py,66,method,"Add an item to the navigation histoyr.
Args:
command: command line text.
screen_output: screen output produced for the command.
scroll_position: (`int`) scroll position in the screen output."
1864,update_scroll_position,tensorflow/tensorflow/python/debug/cli/curses_widgets.py,82,method,"Update the scroll position of the currently-pointed-to history item.
Args:
new_scroll_position: (`int`) new scroll-position value.
Raises:
ValueError: If the history is empty."
1865,size,tensorflow/tensorflow/python/debug/cli/curses_widgets.py,95,method,
1866,pointer,tensorflow/tensorflow/python/debug/cli/curses_widgets.py,98,method,
1867,go_back,tensorflow/tensorflow/python/debug/cli/curses_widgets.py,101,method,"Go back one place in the history, if possible.
Decrease the pointer value by 1, if possible. Otherwise, the pointer value
will be unchanged.
Returns:
The updated pointer value.
Raises:
ValueError: If history is empty."
1868,go_forward,tensorflow/tensorflow/python/debug/cli/curses_widgets.py,120,method,"Go forward one place in the history, if possible.
Increase the pointer value by 1, if possible. Otherwise, the pointer value
will be unchanged.
Returns:
The updated pointer value.
Raises:
ValueError: If history is empty."
1869,can_go_back,tensorflow/tensorflow/python/debug/cli/curses_widgets.py,139,method,"Test whether client can go back one place.
Returns:
(`bool`) Whether going back one place is possible."
1870,can_go_forward,tensorflow/tensorflow/python/debug/cli/curses_widgets.py,147,method,"Test whether client can go forward one place.
Returns:
(`bool`) Whether going back one place is possible."
1871,render,tensorflow/tensorflow/python/debug/cli/curses_widgets.py,155,method,"Render the rich text content of the single-line navigation bar.
Args:
max_length: (`int`) Maximum length of the navigation bar, in characters.
backward_command: (`str`) command for going backward. Used to construct
the shortcut menu item.
forward_command: (`str`) command for going forward. Used to construct the
shortcut menu item.
latest_command_attribute: font attribute for lastest command.
old_command_attribute: font attribute for old (non-latest) command.
Returns:
(`debugger_cli_common.RichTextLines`) the navigation bar text with
attributes."
1872,CommandLineExit,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,41,class,
1873,exit_token,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,48,method,
1874,RichLine,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,52,class,"Rich single-line text.
Attributes:
text: A plain string, the raw text represented by this object. Should not
contain newlines.
font_attr_segs: A list of (start, end, font attribute) triples, representing
richness information applied to substrings of text."
1875,rich_text_lines_from_rich_line_list,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,113,function,"Convert a list of RichLine objects or strings to a RichTextLines object.
Args:
rich_text_list: a list of RichLine objects or strings
annotations: annotations for the resultant RichTextLines object.
Returns:
A corresponding RichTextLines object."
1876,get_tensorflow_version_lines,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,135,function,"Generate RichTextLines with TensorFlow version info.
Args:
include_dependency_versions: Include the version of TensorFlow's key
dependencies, such as numpy.
Returns:
A formatted, multi-line `RichTextLines` object."
1877,RichTextLines,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,154,class,"Rich multi-line text.
Line-by-line text output, with font attributes (e.g., color) and annotations
(e.g., indices in a multi-dimensional tensor). Used as the text output of CLI
commands. Can be rendered on terminal environments such as curses.
This is not to be confused with Rich Text Format (RTF). This class is for text
lines only."
1878,lines,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,216,method,
1879,font_attr_segs,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,220,method,
1880,annotations,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,224,method,
1881,num_lines,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,227,method,
1882,slice,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,230,method,"Slice a RichTextLines object.
The object itself is not changed. A sliced instance is returned.
Args:
begin: (int) Beginning line index (inclusive). Must be >= 0.
end: (int) Ending line index (exclusive). Must be >= 0.
Returns:
(RichTextLines) Sliced output instance of RichTextLines.
Raises:
ValueError: If begin or end is negative."
1883,extend,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,270,method,"Extend this instance of RichTextLines with another instance.
The extension takes effect on the text lines, the font attribute segments,
as well as the annotations. The line indices in the font attribute
segments and the annotations are adjusted to account for the existing
lines. If there are duplicate, non-line-index fields in the annotations,
the value from the input argument ""other"" will override that in this
instance.
Args:
other: (RichTextLines) The other RichTextLines instance to be appended at
the end of this instance."
1884,append,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,334,method,"Append a single line of text.
Args:
line: (str) The text to be added to the end.
font_attr_segs: (list of tuples) Font attribute segments of the appended
line."
1885,append_rich_line,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,347,method,
1886,prepend,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,350,method,"Prepend (i.e., add to the front) a single line of text.
Args:
line: (str) The text to be added to the front.
font_attr_segs: (list of tuples) Font attribute segments of the appended
line."
1887,write_to_file,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,364,method,"Write the object itself to file, in a plain format.
The font_attr_segs and annotations are ignored.
Args:
file_path: (str) path of the file to write to."
1888,regex_find,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,381,function,"Perform regex match in rich text lines.
Produces a new RichTextLines object with font_attr_segs containing highlighted
regex matches.
Example use cases include:
1) search for specific items in a large list of items, and
2) search for specific numerical values in a large tensor.
Args:
orig_screen_output: The original RichTextLines, in which the regex find
is to be performed.
regex: The regex used for matching.
font_attr: Font attribute used for highlighting the found result.
Returns:
A modified copy of orig_screen_output.
Raises:
ValueError: If input str regex is not a valid regular expression."
1889,wrap_rich_text_lines,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,434,function,"Wrap RichTextLines according to maximum number of columns.
Produces a new RichTextLines object with the text lines, font_attr_segs and
annotations properly wrapped. This ought to be used sparingly, as in most
cases, command handlers producing RichTextLines outputs should know the
screen/panel width via the screen_info kwarg and should produce properly
length-limited lines in the output accordingly.
Args:
inp: Input RichTextLines object.
cols: Number of columns, as an int.
Returns:
1) A new instance of RichTextLines, with line lengths limited to cols.
2) A list of new (wrapped) line index. For example, if the original input
consists of three lines and only the second line is wrapped, and it's
wrapped into two lines, this return value will be: [0, 1, 3].
Raises:
ValueError: If inputs have invalid types."
1890,CommandHandlerRegistry,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,530,class,"Registry of command handlers for CLI.
Handler methods (callables) for user commands can be registered with this
class, which then is able to dispatch commands to the correct handlers and
retrieve the RichTextLines output.
For example, suppose you have the following handler defined:
def echo(argv, screen_info=None):
return RichTextLines([""arguments = %s"" % "" "".join(argv),
""screen_info = "" + repr(screen_info)])
you can register the handler with the command prefix ""echo"" and alias ""e"":
registry = CommandHandlerRegistry()
registry.register_command_handler(""echo"", echo,
""Echo arguments, along with screen info"", prefix_aliases=[""e""])
then to invoke this command handler with some arguments and screen_info, do:
registry.dispatch_command(""echo"", [""foo"", ""bar""], screen_info={""cols"": 80})
or with the prefix alias:
registry.dispatch_command(""e"", [""foo"", ""bar""], screen_info={""cols"": 80})
The call will return a RichTextLines object which can be rendered by a CLI."
1891,register_command_handler,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,591,method,"Register a callable as a command handler.
Args:
prefix: Command prefix, i.e., the first word in a command, e.g.,
""print"" as in ""print tensor_1"".
handler: A callable of the following signature:
foo_handler(argv, screen_info=None),
where argv is the argument vector (excluding the command prefix) and
screen_info is a dictionary containing information about the screen,
such as number of columns, e.g., {""cols"": 100}.
The callable should return:
1) a RichTextLines object representing the screen output.
The callable can also raise an exception of the type CommandLineExit,
which if caught by the command-line interface, will lead to its exit.
The exception can optionally carry an exit token of arbitrary type.
help_info: A help string.
prefix_aliases: Aliases for the command prefix, as a list of str. E.g.,
shorthands for the command prefix: [""p"", ""pr""]
Raises:
ValueError: If
1) the prefix is empty, or
2) handler is not callable, or
3) a handler is already registered for the prefix, or
4) elements in prefix_aliases clash with existing aliases.
5) help_info is not a str."
1892,dispatch_command,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,657,method,"Handles a command by dispatching it to a registered command handler.
Args:
prefix: Command prefix, as a str, e.g., ""print"".
argv: Command argument vector, excluding the command prefix, represented
as a list of str, e.g.,
[""tensor_1""]
screen_info: A dictionary containing screen info, e.g., {""cols"": 100}.
Returns:
An instance of RichTextLines or None. If any exception is caught during
the invocation of the command handler, the RichTextLines will wrap the
error type and message.
Raises:
ValueError: If
1) prefix is empty, or
2) no command handler is registered for the command prefix, or
3) the handler is found for the prefix, but it fails to return a
RichTextLines or raise any exception.
CommandLineExit:
If the command handler raises this type of exception, this method will
simply pass it along."
1893,is_registered,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,718,method,"Test if a command prefix or its alias is has a registered handler.
Args:
prefix: A prefix or its alias, as a str.
Returns:
True iff a handler is registered for prefix."
1894,get_help,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,729,method,"Compile help information into a RichTextLines object.
Args:
cmd_prefix: Optional command prefix. As the prefix itself or one of its
aliases.
Returns:
A RichTextLines object containing the help information. If cmd_prefix
is None, the return value will be the full command-line help. Otherwise,
it will be the help information for the specified command."
1895,set_help_intro,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,759,method,"Set an introductory message to help output.
Args:
help_intro: (RichTextLines) Rich text lines appended to the
beginning of the output of the command ""help"", as introductory
information."
1896,TabCompletionRegistry,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,846,class,Registry for tab completion responses.
1897,register_tab_comp_context,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,855,method,"Register a tab-completion context.
Register that, for each word in context_words, the potential tab-completions
are the words in comp_items.
A context word is a pre-existing, completed word in the command line that
determines how tab-completion works for another, incomplete word in the same
command line.
Completion items consist of potential candidates for the incomplete word.
To give a general example, a context word can be ""drink"", and the completion
items can be [""coffee"", ""tea"", ""water""]
Note: A context word can be empty, in which case the context is for the
top-level commands.
Args:
context_words: A list of context words belonging to the context being
registered. It is a list of str, instead of a single string, to support
synonym words triggering the same tab-completion context, e.g.,
both ""drink"" and the short-hand ""dr"" can trigger the same context.
comp_items: A list of completion items, as a list of str.
Raises:
TypeError: if the input arguments are not all of the correct types."
1898,deregister_context,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,898,method,"Deregister a list of context words.
Args:
context_words: A list of context words to deregister, as a list of str.
Raises:
KeyError: if there are word(s) in context_words that do not correspond
to any registered contexts."
1899,extend_comp_items,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,917,method,"Add a list of completion items to a completion context.
Args:
context_word: A single completion word as a string. The extension will
also apply to all other context words of the same context.
new_comp_items: (list of str) New completion items to add.
Raises:
KeyError: if the context word has not been registered."
1900,remove_comp_items,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,936,method,"Remove a list of completion items from a completion context.
Args:
context_word: A single completion word as a string. The removal will
also apply to all other context words of the same context.
comp_items: Completion items to remove.
Raises:
KeyError: if the context word has not been registered."
1901,get_completions,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,955,method,"Get the tab completions given a context word and a prefix.
Args:
context_word: The context word.
prefix: The prefix of the incomplete word.
Returns:
(1) None if no registered context matches the context_word.
A list of str for the matching completion items. Can be an empty list
of a matching context exists, but no completion item matches the
prefix.
(2) Common prefix of all the words in the first return value. If the
first return value is None, this return value will be None, too. If
the first return value is not None, i.e., a list, this return value
will be a str, which can be an empty str if there is no common
prefix among the items of the list."
1902,CommandHistory,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,1004,class,Keeps command history and supports lookup.
1903,add_command,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,1053,method,"Add a command to the command history.
Args:
command: The history command, as a str.
Raises:
TypeError: if command is not a str."
1904,most_recent_n,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,1077,method,"Look up the n most recent commands.
Args:
n: Number of most recent commands to look up.
Returns:
A list of n most recent commands, or all available most recent commands,
if n exceeds size of the command history, in chronological order."
1905,lookup_prefix,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,1090,method,"Look up the n most recent commands that starts with prefix.
Args:
prefix: The prefix to lookup.
n: Number of most recent commands to look up.
Returns:
A list of n most recent commands that have the specified prefix, or all
available most recent commands that have the prefix, if n exceeds the
number of history commands with the prefix."
1906,MenuItem,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,1110,class,A class for an item in a text-based menu.
1907,caption,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,1130,method,
1908,type,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,1134,method,
1909,content,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,1138,method,
1910,is_enabled,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,1141,method,
1911,disable,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,1144,method,
1912,enable,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,1147,method,
1913,Menu,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,1151,class,A class for text-based menu.
1914,append,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,1164,method,"Append an item to the Menu.
Args:
item: (MenuItem) the item to be appended."
1915,insert,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,1172,method,
1916,num_items,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,1175,method,
1917,captions,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,1178,method,
1918,caption_to_item,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,1181,method,"Get a MenuItem from the caption.
Args:
caption: (str) The caption to look up.
Returns:
(MenuItem) The first-match menu item with the caption, if any.
Raises:
LookupError: If a menu item with the caption does not exist."
1919,format_as_single_line,tensorflow/tensorflow/python/debug/cli/debugger_cli_common.py,1201,method,"Format the menu as a single-line RichTextLines object.
Args:
prefix: (str) String added to the beginning of the line.
divider: (str) The dividing string between the menu items.
enabled_item_attrs: (list or str) Attributes applied to each enabled
menu item, e.g., [""bold"", ""underline""].
disabled_item_attrs: (list or str) Attributes applied to each
disabled menu item, e.g., [""red""].
Returns:
(RichTextLines) A single-line output representing the menu, with
font_attr_segs marking the individual menu items."
1920,ExpressionEvaluator,tensorflow/tensorflow/python/debug/cli/evaluator.py,106,class,Evaluates Python expressions using debug tensor values from a dump.
1921,evaluate,tensorflow/tensorflow/python/debug/cli/evaluator.py,118,method,"Parse an expression.
Args:
expression: the expression to be parsed.
Returns:
The result of the evaluation.
Raises:
ValueError: If the value of one or more of the debug tensors in the
expression are not available."
1922,ProfileDataTableView,tensorflow/tensorflow/python/debug/cli/profile_analyzer_cli.py,48,class,Table View of profiling data.
1923,value,tensorflow/tensorflow/python/debug/cli/profile_analyzer_cli.py,81,method,"Get the content of a cell of the table.
Args:
row: (int) row index.
col: (int) column index.
device_name_filter: Regular expression to filter by device name.
node_name_filter: Regular expression to filter by node name.
op_type_filter: Regular expression to filter by op type.
Returns:
A debuggre_cli_common.RichLine object representing the content of the
cell, potentially with a clickable MenuItem.
Raises:
IndexError: if row index is out of range."
1924,row_count,tensorflow/tensorflow/python/debug/cli/profile_analyzer_cli.py,133,method,
1925,column_count,tensorflow/tensorflow/python/debug/cli/profile_analyzer_cli.py,136,method,
1926,column_names,tensorflow/tensorflow/python/debug/cli/profile_analyzer_cli.py,139,method,
1927,column_sort_id,tensorflow/tensorflow/python/debug/cli/profile_analyzer_cli.py,142,method,
1928,ProfileAnalyzer,tensorflow/tensorflow/python/debug/cli/profile_analyzer_cli.py,223,class,Analyzer for profiling data.
1929,list_profile,tensorflow/tensorflow/python/debug/cli/profile_analyzer_cli.py,382,method,"Command handler for list_profile.
List per-operation profile information.
Args:
args: Command-line arguments, excluding the command prefix, as a list of
str.
screen_info: Optional dict input containing screen information such as
cols.
Returns:
Output text lines as a RichTextLines object."
1930,print_source,tensorflow/tensorflow/python/debug/cli/profile_analyzer_cli.py,601,method,"Print a Python source file with line-level profile information.
Args:
args: Command-line arguments, excluding the command prefix, as a list of
str.
screen_info: Optional dict input containing screen information such as
cols.
Returns:
Output text lines as a RichTextLines object."
1931,get_help,tensorflow/tensorflow/python/debug/cli/profile_analyzer_cli.py,764,method,
1932,profile_data_generator,tensorflow/tensorflow/python/debug/cli/profile_analyzer_cli.py,463,method,
1933,create_profiler_ui,tensorflow/tensorflow/python/debug/cli/profile_analyzer_cli.py,768,function,"Create an instance of CursesUI based on a `tf.Graph` and `RunMetadata`.
Args:
graph: Python `Graph` object.
run_metadata: A `RunMetadata` protobuf object.
ui_type: (str) requested UI type, e.g., ""curses"", ""readline"".
on_ui_exit: (`Callable`) the callback to be called when the UI exits.
config: An instance of `cli_config.CLIConfig`.
Returns:
(base_ui.BaseUI) A BaseUI subtype object with a set of standard analyzer
commands and tab-completions registered."
1934,no_rewrite_session_config,tensorflow/tensorflow/python/debug/cli/profile_analyzer_cli_test.py,39,function,
1935,ReadlineUI,tensorflow/tensorflow/python/debug/cli/readline_ui.py,28,class,Readline-based Command-line UI.
1936,run_ui,tensorflow/tensorflow/python/debug/cli/readline_ui.py,53,method,Run the CLI: See the doc of base_ui.BaseUI.run_ui for more details.
1937,MockReadlineUI,tensorflow/tensorflow/python/debug/cli/readline_ui_test.py,34,class,Test subclass of ReadlineUI that bypasses terminal manipulations.
1938,HighlightOptions,tensorflow/tensorflow/python/debug/cli/tensor_format.py,40,class,Options for highlighting elements of a tensor.
1939,format_tensor,tensorflow/tensorflow/python/debug/cli/tensor_format.py,72,function,"Generate a RichTextLines object showing a tensor in formatted style.
Args:
tensor: The tensor to be displayed, as a numpy ndarray or other
appropriate format (e.g., None representing uninitialized tensors).
tensor_label: A label for the tensor, as a string. If set to None, will
suppress the tensor name line in the return value.
include_metadata: Whether metadata such as dtype and shape are to be
included in the formatted text.
auxiliary_message: An auxiliary message to display under the tensor label,
dtype and shape information lines.
include_numeric_summary: Whether a text summary of the numeric values (if
applicable) will be included.
np_printoptions: A dictionary of keyword arguments that are passed to a
call of np.set_printoptions() to set the text format for display numpy
ndarrays.
highlight_options: (HighlightOptions) options for highlighting elements
of the tensor.
Returns:
A RichTextLines object. Its annotation field has line-by-line markups to
indicate which indices in the array the first element of each line
corresponds to."
1940,locate_tensor_element,tensorflow/tensorflow/python/debug/cli/tensor_format.py,282,function,"Locate a tensor element in formatted text lines, given element indices.
Given a RichTextLines object representing a tensor and indices of the sought
element, return the row number at which the element is located (if exists).
Args:
formatted: A RichTextLines object containing formatted text lines
representing the tensor.
indices: Indices of the sought element, as a list of int or a list of list
of int. The former case is for a single set of indices to look up,
whereas the latter case is for looking up a batch of indices sets at once.
In the latter case, the indices must be in ascending order, or a
ValueError will be raised.
Returns:
1) A boolean indicating whether the element falls into an omitted line.
2) Row index.
3) Column start index, i.e., the first column in which the representation
of the specified tensor starts, if it can be determined. If it cannot
be determined (e.g., due to ellipsis), None.
4) Column end index, i.e., the column right after the last column that
represents the specified tensor. Iff it cannot be determined, None.
For return values described above are based on a single set of indices to
look up. In the case of batch mode (multiple sets of indices), the return
values will be lists of the types described above.
Raises:
AttributeError: If:
Input argument ""formatted"" does not have the required annotations.
ValueError: If:
1) Indices do not match the dimensions of the tensor, or
2) Indices exceed sizes of the tensor, or
3) Indices contain negative value(s).
4) If in batch mode, and if not all sets of indices are in ascending
order."
1941,numeric_summary,tensorflow/tensorflow/python/debug/cli/tensor_format.py,488,function,"Get a text summary of a numeric tensor.
This summary is only available for numeric (int*, float*, complex*) and
Boolean tensors.
Args:
tensor: (`numpy.ndarray`) the tensor value object to be summarized.
Returns:
The summary text as a `RichTextLines` object. If the type of `tensor` is not
numeric or Boolean, a single-line `RichTextLines` object containing a
warning message will reflect that."
1942,get_ui,tensorflow/tensorflow/python/debug/cli/ui_factory.py,26,function,"Create a `base_ui.BaseUI` subtype.
This factory method attempts to fallback to other available ui_types on
ImportError. For example, if `ui_type` is `curses`, but `curses` cannot be
imported properly, e.g., on Windows, will fallback to `readline`.
Args:
ui_type: (`str`) requested UI type. Currently supported:
(curses | readline)
on_ui_exit: (`Callable`) the callback to be called when the UI exits.
available_ui_types: (`None` or `list` of `str`) Manually-set available
ui_types.
config: An instance of `cli_config.CLIConfig()` carrying user-facing
configurations.
Returns:
A `base_ui.BaseUI` subtype object.
Raises:
ValueError: on invalid ui_type or on exhausting or fallback ui_types."
1943,parse_args,tensorflow/tensorflow/python/debug/examples/v1/debug_mnist_v1.py,45,function,"Parses commandline arguments.
Returns:
A tuple (parsed, unparsed) of the parsed object and a group of unparsed
arguments that did not match the parser."
1944,parse_args,tensorflow/tensorflow/python/debug/examples/v2/debug_mnist_v2.py,45,function,"Parses commandline arguments.
Returns:
A tuple (parsed, unparsed) of the parsed object and a group of unparsed
arguments that did not match the parser."
1945,limit_string_length,tensorflow/tensorflow/python/debug/lib/check_numerics_callback.py,98,function,"Limit the length of input string.
Args:
string: Input string.
max_len: (int or None) If int, the length limit. If None, no limit.
Returns:
Possibly length-limited string."
1946,get_check_numerics_error_message,tensorflow/tensorflow/python/debug/lib/check_numerics_callback.py,127,function,"Create a meaningful and user-friendly error message about offending tensor.
The error message reveals the following info about the op that outputs
NaN/Infinity: dtype, shape (to the extent known at graph-construction time),
input tensors, stack trace for op creation (if is graph mode).
Args:
slot: (int) slot index of the tensor output.
num_outputs: (int) total number of outputs of the op.
op_type: (str) Type of the that generates `tensor`.
tensor: (Tensor) the offending tensor, i.e., the tensor that contains
Infinities or NaNs.
inputs: (array of Tensor) inputs to the op that generates `tensor`.
graph: (tf.Graph) the graph object that `tensor` belongs to. Available only
under graph mode.
traceback: (list of trace frames) the stack trace of the op's creation.
Available only under graph model.
stack_height_limit: (int or None) If int, limit to the height of the stack
trace printed in the error message. If None, no limit to the height.
path_length_limit: (int or None) Length limit for file paths included in the
formatted stack trace.
Returns:
(str) A formatted error message."
1947,CheckNumericsCallback,tensorflow/tensorflow/python/debug/lib/check_numerics_callback.py,227,class,Wrapper for the numerics-checking callback for thread locality.
1948,callback,tensorflow/tensorflow/python/debug/lib/check_numerics_callback.py,239,method,Eager-function unified callback for checking numerics.
1949,enable_check_numerics,tensorflow/tensorflow/python/debug/lib/check_numerics_callback.py,339,function,"Enable tensor numerics checking in an eager/graph unified fashion.
The numerics checking mechanism will cause any TensorFlow eager execution or
graph execution to error out as soon as an op's output tensor contains
infinity or NaN.
This method is idempotent. Calling it multiple times has the same effect
as calling it once.
This method takes effect only on the thread in which it is called.
When a op's float-type output tensor contains any Infinity or NaN, an
`tf.errors.InvalidArgumentError` will be thrown, with an error message that
reveals the following information:
- The type of the op that generated the tensor with bad numerics.
- Data type (dtype) of the tensor.
- Shape of the tensor (to the extent known at the time of eager execution
or graph construction).
- Name of the containing graph (if available).
- (Graph mode only): The stack trace of the intra-graph op's creation,
with a stack-height limit and a path-length limit for visual clarity.
The stack frames that belong to the user's code (as opposed to
tensorflow's internal code) are highlighted with a text arrow (""->"").
- (Eager mode only): How many of the offending tensor's elements are
`Infinity` and `NaN`, respectively.
Once enabled, the check-numerics mechanism can be disabled by using
`tf.debugging.disable_check_numerics()`.
Example usage:
1. Catching infinity during the execution of a `tf.function` graph:
```py
import tensorflow as tf
tf.debugging.enable_check_numerics()
@tf.function
def square_log_x_plus_1(x):
v = tf.math.log(x + 1)
return tf.math.square(v)
x = -1.0
# When the following line runs, a function graph will be compiled
# from the Python function `square_log_x_plus_1()`. Due to the
# `enable_check_numerics()` call above, the graph will contain
# numerics checking ops that will run during the function graph's
# execution. The function call generates an -infinity when the Log
# (logarithm) op operates on the output tensor of the Add op.
# The program errors out at this line, printing an error message.
y = square_log_x_plus_1(x)
z = -y
```
2. Catching NaN during eager execution:
```py
import numpy as np
import tensorflow as tf
tf.debugging.enable_check_numerics()
x = np.array([[0.0, -1.0], [4.0, 3.0]])
# The following line executes the Sqrt op eagerly. Due to the negative
# element in the input array, a NaN is generated. Due to the
# `enable_check_numerics()` call above, the program errors immediately
# at this line, printing an error message.
y = tf.math.sqrt(x)
z = tf.matmul(y, y)
```
NOTE: If your code is running on TPUs, be sure to call
`tf.config.set_soft_device_placement(True)` before calling
`tf.debugging.enable_check_numerics()` as this API uses automatic outside
compilation on TPUs. For example:
```py
tf.config.set_soft_device_placement(True)
tf.debugging.enable_check_numerics()
resolver = tf.distribute.cluster_resolver.TPUClusterResolver(tpu='')
strategy = tf.distribute.TPUStrategy(resolver)
with strategy.scope():
# ...
```
Args:
stack_height_limit: Limit to the height of the printed stack trace.
Applicable only to ops in `tf.function`s (graphs).
path_length_limit: Limit to the file path included in the printed stack
trace. Applicable only to ops in `tf.function`s (graphs)."
1950,disable_check_numerics,tensorflow/tensorflow/python/debug/lib/check_numerics_callback.py,448,function,"Disable the eager/graph unified numerics checking mechanism.
This method can be used after a call to `tf.debugging.enable_check_numerics()`
to disable the numerics-checking mechanism that catches infinity and NaN
values output by ops executed eagerly or in tf.function-compiled graphs.
This method is idempotent. Calling it multiple times has the same effect
as calling it once.
This method takes effect only on the thread in which it is called."
1951,get_graph_element_name,tensorflow/tensorflow/python/debug/lib/common.py,29,function,"Obtain the name or string representation of a graph element.
If the graph element has the attribute ""name"", return name. Otherwise, return
a __str__ representation of the graph element. Certain graph elements, such as
`SparseTensor`s, do not have the attribute ""name"".
Args:
elem: The graph element in question.
Returns:
If the attribute 'name' is available, return the name. Otherwise, return
str(fetch)."
1952,get_flattened_names,tensorflow/tensorflow/python/debug/lib/common.py,47,function,"Get a flattened list of the names in run() call feeds or fetches.
Args:
feeds_or_fetches: Feeds or fetches of the `Session.run()` call. It maybe
a Tensor, an Operation or a Variable. It may also be nested lists, tuples
or dicts. See doc of `Session.run()` for more details.
Returns:
(list of str) A flattened list of fetch names from `feeds_or_fetches`."
1953,get_run_key,tensorflow/tensorflow/python/debug/lib/common.py,74,function,"Summarize the names of feeds and fetches as a RunKey JSON string.
Args:
feed_dict: The feed_dict given to the `Session.run()` call.
fetches: The fetches from the `Session.run()` call.
Returns:
A JSON Array consisting of two items. They first items is a flattened
Array of the names of the feeds. The second item is a flattened Array of
the names of the fetches."
1954,InconvertibleTensorProto,tensorflow/tensorflow/python/debug/lib/debug_data.py,59,class,Represents a TensorProto that cannot be converted to np.ndarray.
1955,initialized,tensorflow/tensorflow/python/debug/lib/debug_data.py,79,method,
1956,load_tensor_from_event_file,tensorflow/tensorflow/python/debug/lib/debug_data.py,83,function,"Load a tensor from an event file.
Assumes that the event file contains a `Event` protobuf and the `Event`
protobuf contains a `Tensor` value.
Args:
event_file_path: (`str`) path to the event file.
Returns:
The tensor value loaded from the event file, as a `numpy.ndarray`. For
uninitialized Tensors, returns `None`. For Tensors of data types that
cannot be converted to `numpy.ndarray` (e.g., `tf.resource`), return
`None`."
1957,load_tensor_from_event,tensorflow/tensorflow/python/debug/lib/debug_data.py,105,function,"Load a tensor from an Event proto.
Args:
event: The Event proto, assumed to hold a tensor value in its
summary.value[0] field.
Returns:
The tensor value loaded from the event file, as a `numpy.ndarray`, if
representation of the tensor value by a `numpy.ndarray` is possible.
For uninitialized Tensors, returns `None`. For Tensors of data types that
cannot be represented as `numpy.ndarray` (e.g., `tf.resource`), return
the `TensorProto` protobuf object without converting it to a
`numpy.ndarray`."
1958,has_inf_or_nan,tensorflow/tensorflow/python/debug/lib/debug_data.py,202,function,"A predicate for whether a tensor consists of any bad numerical values.
This predicate is common enough to merit definition in this module.
Bad numerical values include `nan`s and `inf`s.
The signature of this function follows the requirement of the method
`DebugDumpDir.find()`.
Args:
datum: (`DebugTensorDatum`) Datum metadata.
tensor: (`numpy.ndarray` or None) Value of the tensor. None represents
an uninitialized tensor.
Returns:
(`bool`) True if and only if tensor consists of any nan or inf values."
1959,extract_core_metadata_from_event_proto,tensorflow/tensorflow/python/debug/lib/debug_data.py,240,function,
1960,device_name_to_device_path,tensorflow/tensorflow/python/debug/lib/debug_data.py,250,function,Convert device name to device path.
1961,device_path_to_device_name,tensorflow/tensorflow/python/debug/lib/debug_data.py,257,function,"Parse device name from device path.
Args:
device_dir: (str) a directory name for the device.
Returns:
(str) parsed device name."
1962,DebugTensorDatum,tensorflow/tensorflow/python/debug/lib/debug_data.py,273,class,"A single tensor dumped by TensorFlow Debugger (tfdbg).
Contains metadata about the dumped tensor, including `timestamp`,
`node_name`, `output_slot`, `debug_op`, and path to the dump file
(`file_path`).
This type does not hold the generally space-expensive tensor value (numpy
array). Instead, it points to the file from which the tensor value can be
loaded (with the `get_tensor` method) if needed."
1963,get_tensor,tensorflow/tensorflow/python/debug/lib/debug_data.py,343,method,"Get tensor from the dump (`Event`) file.
Returns:
The tensor loaded from the dump (`Event`) file."
1964,timestamp,tensorflow/tensorflow/python/debug/lib/debug_data.py,354,method,"Timestamp of when this tensor value was dumped.
Returns:
(`int`) The timestamp in microseconds."
1965,extended_timestamp,tensorflow/tensorflow/python/debug/lib/debug_data.py,364,method,"Extended timestamp, possibly with an index suffix.
The index suffix, e.g., ""-1"", is for disambiguating multiple dumps of the
same tensor with the same timestamp, which can occur if the dumping events
are spaced by shorter than the temporal resolution of the timestamps.
Returns:
(`str`) The extended timestamp."
1966,debug_op,tensorflow/tensorflow/python/debug/lib/debug_data.py,378,method,"Name of the debug op.
Returns:
(`str`) debug op name (e.g., `DebugIdentity`)."
1967,device_name,tensorflow/tensorflow/python/debug/lib/debug_data.py,388,method,"Name of the device that the tensor belongs to.
Returns:
(`str`) device name."
1968,node_name,tensorflow/tensorflow/python/debug/lib/debug_data.py,398,method,"Name of the node from which the tensor value was dumped.
Returns:
(`str`) name of the node watched by the debug op."
1969,output_slot,tensorflow/tensorflow/python/debug/lib/debug_data.py,408,method,"Output slot index from which the tensor value was dumped.
Returns:
(`int`) output slot index watched by the debug op."
1970,tensor_name,tensorflow/tensorflow/python/debug/lib/debug_data.py,418,method,"Name of the tensor watched by the debug op.
Returns:
(`str`) `Tensor` name, in the form of `node_name`:`output_slot`"
1971,watch_key,tensorflow/tensorflow/python/debug/lib/debug_data.py,428,method,"Watch key identities a debug watch on a tensor.
Returns:
(`str`) A watch key, in the form of `tensor_name`:`debug_op`."
1972,file_path,tensorflow/tensorflow/python/debug/lib/debug_data.py,439,method,Path to the file which stores the value of the dumped tensor.
1973,dump_size_bytes,tensorflow/tensorflow/python/debug/lib/debug_data.py,445,method,"Size of the dump file.
Unit: byte.
Returns:
If the dump file exists, size of the dump file, in bytes.
If the dump file does not exist, None."
1974,WatchKeyDoesNotExistInDebugDumpDirError,tensorflow/tensorflow/python/debug/lib/debug_data.py,458,class,
1975,DebugDumpDir,tensorflow/tensorflow/python/debug/lib/debug_data.py,462,class,"Data set from a debug-dump directory on filesystem.
An instance of `DebugDumpDir` contains all `DebugTensorDatum` instances
in a tfdbg dump root directory."
1976,set_python_graph,tensorflow/tensorflow/python/debug/lib/debug_data.py,658,method,"Provide Python `Graph` object to the wrapper.
Unlike the partition graphs, which are protobuf `GraphDef` objects, `Graph`
is a Python object and carries additional information such as the traceback
of the construction of the nodes in the graph.
Args:
python_graph: (ops.Graph) The Python Graph object."
1977,python_graph,tensorflow/tensorflow/python/debug/lib/debug_data.py,676,method,"Get the Python graph.
Returns:
If the Python graph has been set, returns a `tf.Graph` object. Otherwise,
returns None."
1978,core_metadata,tensorflow/tensorflow/python/debug/lib/debug_data.py,687,method,"Metadata about the `Session.run()` call from the core runtime.
Of the three counters available in the return value, `global_step` is
supplied by the caller of the debugged `Session.run()`, while
`session_run_index` and `executor_step_index` are determined by the state
of the core runtime, automatically. For the same fetch list, feed keys and
debug tensor watch options, the same executor will be used and
`executor_step_index` should increase by one at a time. However, runs with
different fetch lists, feed keys and debug_tensor watch options that all
share the same `Session` object can lead to gaps in `session_run_index`.
Returns:
If core metadata are loaded, a `namedtuple` with the fields:
`global_step`: A global step count supplied by the caller of
`Session.run()`. It is optional to the caller. If the caller did not
supply this parameter, its value will be -1.
`session_run_index`: A sorted index for Run() calls to the underlying
TensorFlow `Session` object.
`executor_step_index`: A counter for invocations of a given runtime
executor. The same executor is re-used for the same fetched tensors,
target nodes, input feed keys and debug tensor watch options.
`input_names`: Names of the input (feed) Tensors.
`output_names`: Names of the output (fetched) Tensors.
`target_nodes`: Names of the target nodes.
If the core metadata have not been loaded, `None`.
If more than one core metadata files exist, return a list of the
`nametuple` described above."
1979,dumped_tensor_data,tensorflow/tensorflow/python/debug/lib/debug_data.py,721,method,Retrieve dumped tensor data.
1980,t0,tensorflow/tensorflow/python/debug/lib/debug_data.py,733,method,"Absolute timestamp of the first dumped tensor across all devices.
Returns:
(`int`) absolute timestamp of the first dumped tensor, in microseconds."
1981,size,tensorflow/tensorflow/python/debug/lib/debug_data.py,742,method,"Total number of dumped tensors in the dump root directory.
Returns:
(`int`) The total number of dumped tensors in the dump root directory."
1982,loaded_partition_graphs,tensorflow/tensorflow/python/debug/lib/debug_data.py,920,method,Test whether partition graphs have been loaded.
1983,partition_graphs,tensorflow/tensorflow/python/debug/lib/debug_data.py,924,method,"Get the partition graphs.
Returns:
Partition graphs as a list of GraphDef.
Raises:
LookupError: If no partition graphs have been loaded."
1984,reconstructed_non_debug_partition_graphs,tensorflow/tensorflow/python/debug/lib/debug_data.py,938,method,"Reconstruct partition graphs with the debugger-inserted ops stripped.
The reconstructed partition graphs are identical to the original (i.e.,
non-debugger-decorated) partition graphs except in the following respects:
1) The exact names of the runtime-inserted internal nodes may differ.
These include _Send, _Recv, _HostSend, _HostRecv, _Retval ops.
2) As a consequence of 1, the nodes that receive input directly from such
send- and recv-type ops will have different input names.
3) The parallel_iteration attribute of while-loop Enter ops are set to 1.
Returns:
A dict mapping device names (`str`s) to reconstructed
`tf.compat.v1.GraphDef`s."
1985,run_fetches_info,tensorflow/tensorflow/python/debug/lib/debug_data.py,959,method,"Get a str representation of the fetches used in the Session.run() call.
Returns:
If the information is available from one `Session.run` call, a `str`
obtained from `repr(fetches)`.
If the information is available from multiple `Session.run` calls, a
`list` of `str` from `repr(fetches)`.
If the information is not available, `None`."
1986,run_feed_keys_info,tensorflow/tensorflow/python/debug/lib/debug_data.py,974,method,"Get a str representation of the feed_dict used in the Session.run() call.
Returns:
If the information is available from one `Session.run` call, a `str`
obtained from `repr(feed_dict)`.
If the information is available from multiple `Session.run` calls, a
`list` of `str` obtained from `repr(feed_dict)`.
If the information is not available, `None`."
1987,nodes,tensorflow/tensorflow/python/debug/lib/debug_data.py,1022,method,"Get a list of all nodes from the partition graphs.
Args:
device_name: (`str`) name of device. If None, all nodes from all available
devices will be included.
Returns:
All nodes' names, as a list of str.
Raises:
LookupError: If no partition graphs have been loaded.
ValueError: If specified node name does not exist."
1988,node_attributes,tensorflow/tensorflow/python/debug/lib/debug_data.py,1048,method,"Get the attributes of a node.
Args:
node_name: Name of the node in question.
device_name: (`str`) name of the device. If there is only one device or if
node_name exists on only one device, this argument is optional.
Returns:
Attributes of the node.
Raises:
LookupError: If no partition graphs have been loaded."
1989,node_inputs,tensorflow/tensorflow/python/debug/lib/debug_data.py,1068,method,"Get the inputs of given node according to partition graphs.
Args:
node_name: Name of the node.
is_control: (`bool`) Whether control inputs, rather than non-control
inputs, are to be returned.
device_name: (`str`) name of the device. If there is only one device or if
node_name exists on only one device, this argument is optional.
Returns:
(`list` of `str`) inputs to the node, as a list of node names.
Raises:
LookupError: If node inputs and control inputs have not been loaded
from partition graphs yet."
1990,transitive_inputs,tensorflow/tensorflow/python/debug/lib/debug_data.py,1095,method,"Get the transitive inputs of given node according to partition graphs.
Args:
node_name: Name of the node.
include_control: Include control inputs (True by default).
include_reversed_ref: Whether a ref input, say from A to B, is to be also
considered as an input from B to A. The rationale is that ref inputs
generally let the recipient (e.g., B in this case) mutate the value of
the source (e.g., A in this case). So the reverse direction of the ref
edge reflects the direction of information flow.
device_name: (`str`) name of the device. If there is only one device or if
node_name exists on only one device, this argument is optional.
Returns:
(`list` of `str`) all transitive inputs to the node, as a list of node
names.
Raises:
LookupError: If node inputs and control inputs have not been loaded
from partition graphs yet."
1991,find_some_path,tensorflow/tensorflow/python/debug/lib/debug_data.py,1153,method,"Find a path between a source node and a destination node.
Limitation: the source and destination are required to be on the same
device, i.e., this method does not yet take into account Send/Recv nodes
across devices.
TODO(cais): Make this method work across device edges by tracing Send/Recv
nodes.
Args:
src_node_name: (`str`) name of the source node or name of an output tensor
of the node.
dst_node_name: (`str`) name of the destination node or name of an output
tensor of the node.
include_control: (`bool`) whrther control edges are considered in the
graph tracing.
include_reversed_ref: Whether a ref input, say from A to B, is to be also
considered as an input from B to A. The rationale is that ref inputs
generally let the recipient (e.g., B in this case) mutate the value of
the source (e.g., A in this case). So the reverse direction of the ref
edge reflects the direction of information flow.
device_name: (`str`) name of the device. If there is only one device or if
node_name exists on only one device, this argument is optional.
Returns:
A path from the src_node_name to dst_node_name, as a `list` of `str`, if
it exists. The list includes src_node_name as the first item and
dst_node_name as the last.
If such a path does not exist, `None`.
Raises:
ValueError: If the source and destination nodes are not on the same
device."
1992,node_recipients,tensorflow/tensorflow/python/debug/lib/debug_data.py,1231,method,"Get recipient of the given node's output according to partition graphs.
Args:
node_name: (`str`) name of the node.
is_control: (`bool`) whether control outputs, rather than non-control
outputs, are to be returned.
device_name: (`str`) name of the device. If there is only one device or if
node_name exists on only one device, this argument is optional.
Returns:
(`list` of `str`) all inputs to the node, as a list of node names.
Raises:
LookupError: If node inputs and control inputs have not been loaded
from partition graphs yet."
1993,devices,tensorflow/tensorflow/python/debug/lib/debug_data.py,1260,method,"Get the list of device names.
Returns:
(`list` of `str`) names of the devices."
1994,node_exists,tensorflow/tensorflow/python/debug/lib/debug_data.py,1268,method,"Test if a node exists in the partition graphs.
Args:
node_name: (`str`) name of the node to be checked.
device_name: optional device name. If None, will search for the node
on all available devices. Otherwise, search for the node only on
the given device.
Returns:
A boolean indicating whether the node exists.
Raises:
LookupError: If no partition graphs have been loaded yet.
ValueError: If device_name is specified but cannot be found."
1995,node_device,tensorflow/tensorflow/python/debug/lib/debug_data.py,1297,method,"Get the names of the devices that has nodes of the specified name.
Args:
node_name: (`str`) name of the node.
Returns:
(`str` or `list` of `str`) name of the device(s) on which the node of the
given name is found. Returns a `str` if there is only one such device,
otherwise return a `list` of `str`.
Raises:
LookupError: If node inputs and control inputs have not been loaded
from partition graphs yet.
ValueError: If the node does not exist in partition graphs."
1996,node_op_type,tensorflow/tensorflow/python/debug/lib/debug_data.py,1324,method,"Get the op type of given node.
Args:
node_name: (`str`) name of the node.
device_name: (`str`) name of the device. If there is only one device or if
node_name exists on only one device, this argument is optional.
Returns:
(`str`) op type of the node.
Raises:
LookupError: If node op types have not been loaded
from partition graphs yet."
1997,debug_watch_keys,tensorflow/tensorflow/python/debug/lib/debug_data.py,1346,method,"Get all tensor watch keys of given node according to partition graphs.
Args:
node_name: (`str`) name of the node.
device_name: (`str`) name of the device. If there is only one device or if
node_name exists on only one device, this argument is optional.
Returns:
(`list` of `str`) all debug tensor watch keys. Returns an empty list if
the node name does not correspond to any debug watch keys.
Raises:
`LookupError`: If debug watch information has not been loaded from
partition graphs yet."
1998,watch_key_to_data,tensorflow/tensorflow/python/debug/lib/debug_data.py,1380,method,"Get all `DebugTensorDatum` instances corresponding to a debug watch key.
Args:
debug_watch_key: (`str`) debug watch key.
device_name: (`str`) name of the device. If there is only one device or if
the specified debug_watch_key exists on only one device, this argument
is optional.
Returns:
A list of `DebugTensorDatum` instances that correspond to the debug watch
key. If the watch key does not exist, returns an empty list.
Raises:
ValueError: If there are multiple devices that have the debug_watch_key,
but device_name is not specified."
1999,find,tensorflow/tensorflow/python/debug/lib/debug_data.py,1417,method,"Find dumped tensor data by a certain predicate.
Args:
predicate: A callable that takes two input arguments:
```python
def predicate(debug_tensor_datum, tensor):
# returns a bool
```
where `debug_tensor_datum` is an instance of `DebugTensorDatum`, which
carries the metadata, such as the `Tensor`'s node name, output slot
timestamp, debug op name, etc.; and `tensor` is the dumped tensor value
as a `numpy.ndarray`.
first_n: (`int`) return only the first n `DebugTensotDatum` instances (in
time order) for which the predicate returns True. To return all the
`DebugTensotDatum` instances, let first_n be <= 0.
device_name: optional device name.
exclude_node_names: Optional regular expression to exclude nodes with
names matching the regular expression.
Returns:
A list of all `DebugTensorDatum` objects in this `DebugDumpDir` object
for which predicate returns True, sorted in ascending order of the
timestamp."
2000,get_tensor_file_paths,tensorflow/tensorflow/python/debug/lib/debug_data.py,1466,method,"Get the file paths from a debug-dumped tensor.
Args:
node_name: (`str`) name of the node that the tensor is produced by.
output_slot: (`int`) output slot index of tensor.
debug_op: (`str`) name of the debug op.
device_name: (`str`) name of the device. If there is only one device or if
the specified debug_watch_key exists on only one device, this argument
is optional.
Returns:
List of file path(s) loaded. This is a list because each debugged tensor
may be dumped multiple times.
Raises:
WatchKeyDoesNotExistInDebugDumpDirError: If the tensor does not exist in
the debug-dump data."
2001,get_tensors,tensorflow/tensorflow/python/debug/lib/debug_data.py,1500,method,"Get the tensor value from for a debug-dumped tensor.
The tensor may be dumped multiple times in the dump root directory, so a
list of tensors (`numpy.ndarray`) is returned.
Args:
node_name: (`str`) name of the node that the tensor is produced by.
output_slot: (`int`) output slot index of tensor.
debug_op: (`str`) name of the debug op.
device_name: (`str`) name of the device. If there is only one device or if
the specified debug_watch_key exists on only one device, this argument
is optional.
Returns:
List of tensors (`numpy.ndarray`) loaded from the debug-dump file(s).
Raises:
WatchKeyDoesNotExistInDebugDumpDirError: If the tensor does not exist in
the debug-dump data."
2002,get_rel_timestamps,tensorflow/tensorflow/python/debug/lib/debug_data.py,1532,method,"Get the relative timestamp from for a debug-dumped tensor.
Relative timestamp means (absolute timestamp - `t0`), where `t0` is the
absolute timestamp of the first dumped tensor in the dump root. The tensor
may be dumped multiple times in the dump root directory, so a list of
relative timestamps (`numpy.ndarray`) is returned.
Args:
node_name: (`str`) name of the node that the tensor is produced by.
output_slot: (`int`) output slot index of tensor.
debug_op: (`str`) name of the debug op.
device_name: (`str`) name of the device. If there is only one device or if
the specified debug_watch_key exists on only one device, this argument
is optional.
Returns:
(`list` of `int`) list of relative timestamps.
Raises:
WatchKeyDoesNotExistInDebugDumpDirError: If the tensor watch key does not
exist in the debug dump data."
2003,get_dump_sizes_bytes,tensorflow/tensorflow/python/debug/lib/debug_data.py,1569,method,"Get the sizes of the dump files for a debug-dumped tensor.
Unit of the file size: byte.
Args:
node_name: (`str`) name of the node that the tensor is produced by.
output_slot: (`int`) output slot index of tensor.
debug_op: (`str`) name of the debug op.
device_name: (`str`) name of the device. If there is only one device or if
the specified debug_watch_key exists on only one device, this argument
is optional.
Returns:
(`list` of `int`): list of dump file sizes in bytes.
Raises:
WatchKeyDoesNotExistInDebugDumpDirError: If the tensor watch key does not
exist in the debug dump data."
2004,node_traceback,tensorflow/tensorflow/python/debug/lib/debug_data.py,1603,method,"Try to retrieve the Python traceback of node's construction.
Args:
element_name: (`str`) Name of a graph element (node or tensor).
Returns:
(list) The traceback list object as returned by the `extract_trace`
method of Python's traceback module.
Raises:
LookupError: If Python graph is not available for traceback lookup.
KeyError: If the node cannot be found in the Python graph loaded."
2005,BaseMonitor,tensorflow/tensorflow/python/debug/lib/debug_events_monitors.py,49,class,Base class for debug event data monitors.
2006,on_execution,tensorflow/tensorflow/python/debug/lib/debug_events_monitors.py,56,method,"Monitor method for top-level execution events.
Return values (if any) are ignored by the associated DebugDataReader.
Args:
execution_index: The index of the top-level execution event, as an int.
execution: An Execution data object, for a top-level op or function
execution event."
2007,on_graph_execution_trace,tensorflow/tensorflow/python/debug/lib/debug_events_monitors.py,67,method,"Monitor method for intra-graph execution events.
Return values (if any) are ignored by the associated DebugDataReader.
Args:
graph_execution_trace_index: The index of the intra-graph execution
event, as an int.
graph_execution_trace: A GraphExecutionTrace data object, for an
intra-graph tensor event."
2008,InfNanAlert,tensorflow/tensorflow/python/debug/lib/debug_events_monitors.py,84,class,Alert for Infinity and NaN values.
2009,wall_time,tensorflow/tensorflow/python/debug/lib/debug_events_monitors.py,108,method,
2010,op_type,tensorflow/tensorflow/python/debug/lib/debug_events_monitors.py,112,method,
2011,output_slot,tensorflow/tensorflow/python/debug/lib/debug_events_monitors.py,116,method,
2012,size,tensorflow/tensorflow/python/debug/lib/debug_events_monitors.py,120,method,
2013,num_neg_inf,tensorflow/tensorflow/python/debug/lib/debug_events_monitors.py,124,method,
2014,num_pos_inf,tensorflow/tensorflow/python/debug/lib/debug_events_monitors.py,128,method,
2015,num_nan,tensorflow/tensorflow/python/debug/lib/debug_events_monitors.py,132,method,
2016,execution_index,tensorflow/tensorflow/python/debug/lib/debug_events_monitors.py,136,method,
2017,graph_execution_trace_index,tensorflow/tensorflow/python/debug/lib/debug_events_monitors.py,140,method,
2018,InfNanMonitor,tensorflow/tensorflow/python/debug/lib/debug_events_monitors.py,144,class,Monitor for Infinity and NaN in tensor values.
2019,on_execution,tensorflow/tensorflow/python/debug/lib/debug_events_monitors.py,262,method,
2020,on_graph_execution_trace,tensorflow/tensorflow/python/debug/lib/debug_events_monitors.py,286,method,Monitor method for GraphExecutionTrace data object.
2021,alerts,tensorflow/tensorflow/python/debug/lib/debug_events_monitors.py,310,method,
2022,DebugEventsReader,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,39,class,Reader class for a tfdbg v2 DebugEvents directory.
2023,starting_wall_time,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,151,method,"Get the starting timestamp of the instrumented TensorFlow program.
When there are multiple hosts (i.e., multiple tfdbg file sets), the earliest
timestamp among the file sets is returned. It is assumed to be the job that
starts first (e.g., the coordinator).
Returns:
Starting timestamp in seconds since the epoch, as a float."
2024,tfdbg_run_id,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,163,method,Get the run ID of the instrumented TensorFlow program.
2025,tensorflow_version,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,167,method,Get the version string of TensorFlow that the debugged program ran on.
2026,tfdbg_file_version,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,171,method,Get the tfdbg file format version.
2027,source_files_iterator,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,236,method,
2028,stack_frames_iterator,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,239,method,
2029,graphs_iterator,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,242,method,
2030,read_source_files_event,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,245,method,Read a DebugEvent proto at given offset from the .source_files file.
2031,read_graphs_event,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,251,method,"Read a DebugEvent proto at a given offset from the .graphs file.
Args:
offset: Offset to read the DebugEvent proto from.
Returns:
A DebugEventProto.
Raises:
`errors.DataLossError` if offset is at a wrong location.
`IndexError` if offset is out of range of the file."
2032,execution_iterator,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,267,method,
2033,read_execution_event,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,270,method,"Read a DebugEvent proto at a given offset from the .execution file.
Args:
offset: Offset to read the DebugEvent proto from.
Returns:
A DebugEventProto.
Raises:
`errors.DataLossError` if offset is at a wrong location.
`IndexError` if offset is out of range of the file."
2034,graph_execution_traces_iterators,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,287,method,
2035,read_graph_execution_traces_event,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,293,method,"Read DebugEvent at given offset from given .graph_execution_traces file.
Args:
locator: A (file_index, offset) tuple that locates the DebugEvent
containing the graph execution trace.
Returns:
A DebugEventProto.
Raises:
`errors.DataLossError` if offset is at a wrong location.
`IndexError` if offset is out of range of the file."
2036,close,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,314,method,
2037,BaseDigest,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,322,class,"Base class for digest.
Properties:
wall_time: A timestamp for the digest as a `float` (unit: s).
locator: A datum that allows tracng the digest to its original
location. It can be either of the two:
1. Bytes offset from the beginning of the file as a single integer,
for the case of all digests of the same kind coming from the same
file.
2. A tuple of a file index and a byte offset. This applies to case
in which the same type of debugger data may come from multple files,
e.g., graph execution traces."
2038,wall_time,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,342,method,
2039,locator,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,346,method,
2040,to_json,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,349,method,
2041,ExecutionDigest,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,353,class,"Light-weight digest summarizing top-level execution event.
Use `DebugDataReader.read_execution(execution_digest)` to load the more
detailed data object concerning the execution event (`Execution`).
Properties:
op_type: Type name of the executed op. In the case of the eager execution of
an individual op, it is the name of the op (e.g., ""MatMul"").
In the case of the execution of a tf.function (FuncGraph), this is the
internally-generated name of the function (e.g.,
""__inference_my_func_123"").
output_tensor_device_ids: IDs of the devices on which the output tensors of
the execution reside. For no-output execution, this is `None`."
2042,op_type,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,379,method,
2043,output_tensor_device_ids,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,383,method,
2044,to_json,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,386,method,
2045,Execution,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,399,class,"Detailed data relating to a top-level execution event.
The execution is of an individual op or a tf.function, which may have any
number of output tensors.
Properties (beyond the base class `ExecutionDigest`):
host_name: Name of the host on which the execution happened.
stack_frame_ids: Reference IDs for stack frames, ordered from bottommost to
topmost. Use `DebugDataReader.read_execution_stack_trace()` to load the
detailed stack frames (filepath, lineno and function name).
tensor_debug_mode: TensorDebugMode enum value, as an `int`.
graph_id: ID of the executed FuncGraph (applicable only the execution of a
tf.function). `None` for the eager execution of an individual op.
input_tensor_ids: IDs of the input (eager) tensor(s) for this execution, if
any. If the eager execution has no input tensor, this is `None`. Else,
this is a `tuple` of `int`s.
output_tensor_ids: IDs of the output (eager) tensor(s) from this execution,
if any. If the eager execution produces no output tensor, this is `None`.
Else, this is a `tuple` of `int`s.
debug_tensor_values: Values of the debug tensor(s), applicable only to
non-FULL_TENSOR tensor debug mode. A tuple of list of numbers. Each
element of the tuple corresponds to an output tensor of the execution.
See documentation of the various TensorDebugModes for the semantics of the
numbers. If the eager execution produces no output tensor, this is
`None`. Else, this is a `tuple` of `list`s."
2046,host_name,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,450,method,
2047,stack_frame_ids,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,454,method,
2048,tensor_debug_mode,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,458,method,
2049,graph_id,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,462,method,
2050,input_tensor_ids,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,466,method,
2051,num_outputs,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,470,method,
2052,output_tensor_ids,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,474,method,
2053,debug_tensor_values,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,478,method,
2054,to_json,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,481,method,
2055,DebuggedGraph,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,495,class,"Data object representing debugging information about a tf.Graph.
Includes `FuncGraph`s.
Properties:
name: Name of the graph (if any). May be `None` for non-function graphs.
graph_id: Debugger-generated ID for the graph.
inner_graph_ids: A list of the debugger-generated IDs for the graphs
enclosed by this graph.
outer_graph_id: If this graph is nested within an outer graph, ID of the
outer graph. If this is an outermost graph, `None`."
2056,add_inner_graph_id,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,522,method,"Add the debugger-generated ID of a graph nested within this graph.
Args:
inner_graph_id: The debugger-generated ID of the nested inner graph."
2057,add_op,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,531,method,"Add an op creation data object.
Args:
graph_op_creation_digest: A GraphOpCreationDigest data object describing
the creation of an op inside this graph."
2058,add_op_consumer,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,545,method,"Add a consuming op for this op.
Args:
src_op_name: Name of the op of which the output tensor is being consumed.
src_slot: 0-based output slot of the op being consumed.
dst_op_name: Name of the consuming op (e.g., ""Conv2D_3/BiasAdd"")
dst_slot: 0-based input slot of the consuming op that receives the tensor
from this op."
2059,name,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,558,method,
2060,graph_id,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,562,method,
2061,outer_graph_id,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,566,method,
2062,inner_graph_ids,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,570,method,
2063,get_tensor_id,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,573,method,Get the ID of a symbolic tensor in this graph.
2064,get_op_creation_digest,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,577,method,Get the GraphOpCreationDigest for a op in the graph.
2065,get_op_consumers,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,581,method,"Get all the downstream consumers of this op.
Only data (non-control) edges are tracked.
Args:
src_op_name: Name of the op providing the tensor being consumed.
Returns:
A list of (src_slot, dst_op_name, dst_slot) tuples. In each item of
the list:
src_slot: 0-based output slot of the op of which the output tensor
is being consumed.
dst_op_name: Name of the consuming op (e.g., ""Conv2D_3/BiasAdd"")
dst_slot: 0-based input slot of the consuming op that receives
the tensor from this op."
2066,to_json,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,600,method,
2067,DebuggedDevice,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,609,class,"Debugger data regarding a device involved in the debugged program.
Properties:
device_name: Name of the device, as a str.
device_id: An integer ID for the device, unique for each device within
the scope of the debugged TensorFlow program."
2068,device_name,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,625,method,
2069,device_id,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,629,method,
2070,to_json,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,632,method,
2071,GraphOpCreationDigest,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,639,class,"Data object describing the creation of an op inside a graph.
For size efficiency, this digest object does not contain any stack frames or
any references to them. To obtain the stack frames, use
`DataReader.read_graph_op_creation_stack_trace()`.
Properties (beyond the base class):
graph_id: Debugger-generated ID of the immediately-enclosing graph.
op_type: Type name of the op (e.g., ""MatMul"").
op_name: Name of the op (e.g., ""dense_1/MatMul"").
output_tensor_ids: Debugger-generated IDs for the output(s) of the op.
If the op produces no output tensor, this is `None`. Else, this is a
`tuple` of `int`s.
input_names: Names of the input tensors to the op.
device_name: The name of the device that the op is placed on (if available).
host_name: Name of the host on which the op is created.
stack_frame_ids: IDs of the frames of the stack trace at which the op
is created."
2072,graph_id,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,682,method,
2073,op_type,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,686,method,
2074,op_name,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,690,method,
2075,output_tensor_ids,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,694,method,
2076,num_outputs,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,698,method,
2077,input_names,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,702,method,
2078,device_name,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,706,method,
2079,host_name,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,710,method,
2080,stack_frame_ids,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,714,method,
2081,to_json,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,717,method,
2082,GraphExecutionTraceDigest,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,732,class,"Light-weight summary of a intra-graph tensor execution event.
Use `DebugDataReader.read_graph_execution_trace()` on this object to read more
detailed data (`GraphExecutionTrace`).
Properties (beyond the base class):
op_type: Type name of the executed op (e.g., ""Conv2D"").
op_name: Name of the op (e.g., ""conv_2d_3/Conv2D"").
output_slot: Output slot index of the tensor.
graph_id: The debugger-generated ID of the innermost (immediately-enclosing)
graph."
2083,op_type,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,755,method,
2084,op_name,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,759,method,
2085,output_slot,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,763,method,
2086,graph_id,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,767,method,
2087,to_json,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,770,method,
2088,GraphExecutionTrace,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,781,class,"Detailed data object describing an intra-graph tensor execution.
Attributes (in addition to GraphExecutionTraceDigest):
graph_ids: The debugger-generated IDs of the graphs that enclose the
executed op (tensor), ordered from the outermost to the innermost.
graph_id: The debugger-generated ID of the innermost (immediately-enclosing)
graph.
tensor_debug_mode: TensorDebugMode enum value.
debug_tensor_value: Debug tensor values (only for non-FULL_TENSOR
tensor_debug_mode). A list of numbers. See the documentation of the
TensorDebugModes for the semantics of the numbers.
device_name: Device on which the tensor resides (if available)"
2089,graph_ids,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,815,method,
2090,graph_id,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,819,method,
2091,tensor_debug_mode,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,823,method,
2092,debug_tensor_value,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,827,method,
2093,device_name,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,831,method,
2094,to_json,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,834,method,
2095,DebugDataReader,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,917,class,"A reader that reads structured debugging data in the tfdbg v2 format.
The set of data read by an object of this class concerns the execution history
of a tfdbg2-instrumented TensorFlow program.
Note:
- An object of this class incrementally reads data from files that belong to
the tfdbg v2 DebugEvent file set. Calling `update()` triggers the reading
from the last-successful reading positions in the files.
- This object can be used as a context manager. Its `__exit__()` call
closes the file readers cleanly."
2096,update,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1112,method,Perform incremental read of the file set.
2097,source_file_list,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1120,method,"Get a list of source files known to the debugger data reader.
Returns:
A tuple of `(host_name, file_path)` tuples."
2098,source_lines,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1128,method,"Read the line-by-line content of a source file.
Args:
host_name: Host name on which the source file is located.
file_path: File path at which the source file is located.
Returns:
Lines of the source file as a `list` of `str`s."
2099,starting_wall_time,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1141,method,"Wall timestamp for when the debugged TensorFlow program started.
Returns:
Stating wall time as seconds since the epoch, as a `float`."
2100,tensorflow_version,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1149,method,"TensorFlow version used in the debugged TensorFlow program.
Note: this is not necessarily the same as the version of TensorFlow used to
load the DebugEvent file set.
Returns:
TensorFlow version used by the debugged program, as a `str`."
2101,tfdbg_run_id,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1160,method,Get the debugger run ID of the debugged TensorFlow program.
2102,outermost_graphs,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1164,method,Get the number of outer most graphs read so far.
2103,graph_by_id,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1169,method,Get a DebuggedGraph object by its ID.
2104,device_name_by_id,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1173,method,Get the name of a device by the debugger-generated ID of the device.
2105,device_name_map,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1177,method,Get a map mapping device IDs to device names.
2106,graph_op_digests,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1182,method,"Get the list of the digests for graph-op creation so far.
Args:
op_type: Optional op type to filter the creation events with.
Returns:
A list of `GraphOpCreationDigest` objects."
2107,graph_execution_traces,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1197,method,"Get all the intra-graph execution tensor traces read so far.
Args:
digest: Whether the results will be returned in the more light-weight
digest form.
begin: Optional beginning index for the requested traces or their digests.
Python-style negative indices are supported.
end: Optional ending index for the requested traces or their digests.
Python-style negative indices are supported.
Returns:
If `digest`: a `list` of `GraphExecutionTraceDigest` objects.
Else: a `list` of `GraphExecutionTrace` objects."
2108,num_graph_execution_traces,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1222,method,Get the number of graph execution traces read so far.
2109,executions,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1226,method,"Get `Execution`s or `ExecutionDigest`s this reader has read so far.
Args:
digest: Whether the results are returned in a digest form, i.e.,
`ExecutionDigest` format, instead of the more detailed `Execution`
format.
begin: Optional beginning index for the requested execution data objects
or their digests. Python-style negative indices are supported.
end: Optional ending index for the requested execution data objects or
their digests. Python-style negative indices are supported.
Returns:
If `digest`: a `list` of `ExecutionDigest` objects.
Else: a `list` of `Execution` objects."
2110,num_executions,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1253,method,Get the number of execution events read so far.
2111,read_execution,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1257,method,Read a detailed Execution object.
2112,read_graph_execution_trace,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1263,method,"Read the detailed graph execution trace.
Args:
graph_execution_trace_digest: A `GraphExecutionTraceDigest` object.
Returns:
The corresponding `GraphExecutionTrace` object."
2113,read_execution_stack_trace,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1277,method,"Read the stack trace of a given Execution object.
Args:
execution: The Execution object of interest.
Returns:
1. The host name.
2. The stack trace, as a list of (file_path, lineno, func) tuples."
2114,read_graph_op_creation_stack_trace,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1292,method,"Read the stack trace of a given graph op creation object.
Args:
graph_op_creation_digest: The GraphOpCreationDigest object of interest.
Returns:
A tuple consisting of:
1. The host name.
2. The stack trace, as a list of (file_path, lineno, func) tuples."
2115,execution_to_tensor_values,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1312,method,"Read the full tensor values from an Execution or ExecutionDigest.
Args:
execution: An `ExecutionDigest` or `ExeuctionDigest` object.
Returns:
A list of numpy arrays representing the output tensor values of the
execution event."
2116,graph_execution_trace_to_tensor_value,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1326,method,"Read full tensor values from an Execution or ExecutionDigest.
Args:
trace: An `GraphExecutionTraceDigest` or `GraphExecutionTrace` object.
Returns:
A numpy array representing the output tensor value of the intra-graph
tensor execution event."
2117,symbolic_tensor_id,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1339,method,"Get the ID of a symbolic tensor.
Args:
graph_id: The ID of the immediately-enclosing graph.
op_name: Name of the op.
output_slot: Output slot as an int.
Returns:
The ID of the symbolic tensor as an int."
2118,graph_execution_trace_to_tensor_id,tensorflow/tensorflow/python/debug/lib/debug_events_reader.py,1352,method,Get symbolic tensor ID from a GraphExecutoinTraceDigest object.
2119,DebugEventsWriter,tensorflow/tensorflow/python/debug/lib/debug_events_writer.py,30,class,A writer for TF debugging events. Used by tfdbg v2.
2120,WriteSourceFile,tensorflow/tensorflow/python/debug/lib/debug_events_writer.py,60,method,"Write a SourceFile proto with the writer.
Args:
source_file: A SourceFile proto, describing the content of a source file
involved in the execution of the debugged TensorFlow program."
2121,WriteGraphOpCreation,tensorflow/tensorflow/python/debug/lib/debug_events_writer.py,86,method,"Write a GraphOpCreation proto with the writer.
Args:
graph_op_creation: A GraphOpCreation proto, describing the details of the
creation of an op inside a TensorFlow Graph."
2122,WriteDebuggedGraph,tensorflow/tensorflow/python/debug/lib/debug_events_writer.py,99,method,"Write a DebuggedGraph proto with the writer.
Args:
debugged_graph: A DebuggedGraph proto, describing the details of a
TensorFlow Graph that has completed its construction."
2123,WriteExecution,tensorflow/tensorflow/python/debug/lib/debug_events_writer.py,110,method,"Write a Execution proto with the writer.
Args:
execution: An Execution proto, describing a TensorFlow op or graph
execution event."
2124,WriteGraphExecutionTrace,tensorflow/tensorflow/python/debug/lib/debug_events_writer.py,121,method,"Write a GraphExecutionTrace proto with the writer.
Args:
graph_execution_trace: A GraphExecutionTrace proto, concerning the value
of an intermediate tensor or a list of intermediate tensors that are
computed during the graph's execution."
2125,RegisterDeviceAndGetId,tensorflow/tensorflow/python/debug/lib/debug_events_writer.py,135,method,
2126,FlushNonExecutionFiles,tensorflow/tensorflow/python/debug/lib/debug_events_writer.py,139,method,Flush the non-execution debug event files.
2127,FlushExecutionFiles,tensorflow/tensorflow/python/debug/lib/debug_events_writer.py,143,method,"Flush the execution debug event files.
Causes the current content of the cyclic buffers to be written to
the .execution and .graph_execution_traces debug events files.
Also clears those cyclic buffers."
2128,Close,tensorflow/tensorflow/python/debug/lib/debug_events_writer.py,152,method,Close the writer.
2129,dump_root,tensorflow/tensorflow/python/debug/lib/debug_events_writer.py,157,method,
2130,GradientsDebugger,tensorflow/tensorflow/python/debug/lib/debug_gradients.py,68,class,"Gradients Debugger.
Allows retrieval of gradient tensors created by TensorFlow's automatic
differentiation algorithm, i.e., `tf.gradients` and optimizer classes that
use it."
2131,y_tensor,tensorflow/tensorflow/python/debug/lib/debug_gradients.py,101,method,
2132,graph,tensorflow/tensorflow/python/debug/lib/debug_gradients.py,105,method,
2133,identify_gradient,tensorflow/tensorflow/python/debug/lib/debug_gradients.py,114,method,"Create a debug identity tensor that registers and forwards gradients.
The side effect of this method is that when gradient tensor(s) are created
with respect to the any paths that include the `input_tensor`, the gradient
tensor(s) with respect to `input_tensor` will be registered with this
this `GradientsDebugger` instance and can later be retrieved, with the
methods `gradient_tensor` and `gradient_tensors`.
Example:
```python
x = tf.Variable(1.0)
y = tf.add(x, x)
grad_debugger = tf_debug.GradientsDebugger()
debug_y = grad_debugger.identify_gradient(y)
z = tf.square(debug_y)
# Create a train op under the grad_debugger context.
with grad_debugger:
train_op = tf.compat.v1.train.GradientDescentOptimizer(z)
# Now we can reflect through grad_debugger to get the gradient tensor
# with respect to y.
y_grad = grad_debugger.gradient_tensor(y)
```
Args:
input_tensor: the input `tf.Tensor` object whose related gradient tensors
are to be registered with this `GradientsDebugger` instance when they
are created, e.g., during `tf.gradients` calls or the construction
of optimization (training) op that uses `tf.gradients`.
Returns:
A forwarded identity of `input_tensor`, as a `tf.Tensor`.
Raises:
ValueError: If an op with name that duplicates the gradient-debugging op
already exists in the graph (highly unlikely)."
2134,watch_gradients_by_tensors,tensorflow/tensorflow/python/debug/lib/debug_gradients.py,171,method,"Watch gradient tensors by x-tensor(s).
The side effect of this method is that when gradient tensor(s) are created
with respect to the any paths that include the `x_tensor`s, the gradient
tensor(s) with respect to the tensor will be registered with this
this `GradientsDebugger` instance and can later be retrieved, with the
methods `gradient_tensor` and `gradient_tensors`.
Unlike the method `identify_gradient`, this method is used to retrieve
gradient tensors after the construction of the forward subgraph has
completed (but before the construction of the backward subgraph).
This method is the same as `watch_gradients_by_x_tensor_names` except that
the tensors are specified by the Python `tf.Tensor` or `tf.Variable`
objects, instead by name patterns.
Example:
```python
x = tf.Variable(1.0)
y = tf.add(x, x, name=""y"")
z = tf.square(debug_y)
# Create a train op under the grad_debugger context.
grad_debugger = tf_debug.GradientsDebugger()
with grad_debugger.watch_gradients_by_tensors(y):
train_op = tf.compat.v1.train.GradientDescentOptimizer(z)
# Now we can reflect through grad_debugger to get the gradient tensor
# with respect to y.
y_grad = grad_debugger.gradient_tensor(y)
# or
y_grad = grad_debugger.gradient_tensor(""y:0"")
```
Args:
graph: the `tf.Graph` to watch the gradients on.
tensors: a `tf.Tensor` or `tf.Variable` object, or a list of such objects.
Returns:
The GradientsDebugger instance itself."
2135,watch_gradients_by_tensor_names,tensorflow/tensorflow/python/debug/lib/debug_gradients.py,224,method,"Watch gradient tensors by name(s) of the x-tensor(s).
The side effect of this method is that when gradient tensor(s) are created
with respect to the x-tensors, the gradient tensor(s) will be registered
with this `GradientsDebugger` instance and can later be retrieved.
Unlike the `identify_gradient` method, this method is used after the
construction of the forward graph has completed. Unlike the
`watch_gradients_by_tensor` method, this method does not use handles to the
tensors of interest; it uses their names.
This method is the same as `watch_gradients_by_tensors` except that the
x-tensors are specified by name patterns, instead of `tf.Tensor` or
`tf.Variable` objects.
Example:
```python
x = tf.Variable(1.0, name=""x"")
y = tf.add(x, x, name=""y"")
z = tf.square(debug_y)
# Create a train op under the grad_debugger context.
grad_debugger = tf_debug.GradientsDebugger()
with grad_debugger.watch_gradients_by_tensor_names(r""(x|y):0$""):
train_op = tf.compat.v1.train.GradientDescentOptimizer(z)
# Now we can reflect through grad_debugger to get the gradient tensor
# with respect to x and y.
x_grad = grad_debugger.gradient_tensor(""x:0"")
y_grad = grad_debugger.gradient_tensor(""y:0"")
```
Args:
graph: the `tf.Graph` to watch the gradients on.
tensor_name_regex: the regular-expression pattern of the name(s) of the
x-tensor(s) to watch. x-tensor refers to the tensors on the denominator
of the differentiation.
Returns:
The GradientsDebugger instance itself."
2136,register_gradient_tensor,tensorflow/tensorflow/python/debug/lib/debug_gradients.py,294,method,"Register the gradient tensor for an x-tensor.
Args:
x_tensor_name: (`str`) the name of the independent `tf.Tensor`, i.e.,
the tensor on the denominator of the differentiation.
gradient_tensor: the gradient `tf.Tensor`."
2137,gradient_tensor,tensorflow/tensorflow/python/debug/lib/debug_gradients.py,308,method,"Get the gradient tensor of an x-tensor.
Args:
x_tensor: (`tf.Tensor`, `tf.Variable` or `str`) The x-tensor object or its
name. x-tensor refers to the independent `tf.Tensor`, i.e., the tensor
on the denominator of the differentiation.
Returns:
If found, the gradient tensor.
Raises:
TypeError: If `x_tensor` is not a `tf.Tensor`, `tf.Variable` or `str`.
LookupError: If the `x_tensor` has not been registered with a gradient
tensor."
2138,gradient_tensors,tensorflow/tensorflow/python/debug/lib/debug_gradients.py,331,method,"Get the gradient tensors that this object is aware of.
Returns:
A dict mapping x-tensor names to gradient tensor objects. x-tensor refers
to the tensors on the denominator of the differentation."
2139,clear_gradient_debuggers,tensorflow/tensorflow/python/debug/lib/debug_gradients.py,351,function,Clear all globally registered gradient debuggers.
2140,gradient_values_from_dump,tensorflow/tensorflow/python/debug/lib/debug_gradients.py,372,function,"Find gradient values from a `DebugDumpDir` object.
Args:
grad_debugger: the `tf_debug.GradientsDebugger` instance to be used.
x_tensor: (`tf.Tensor`, `tf.Variable` or `str`) The x-tensor object or its
name. x-tensor refers to the independent `tf.Tensor`, i.e., the tensor
on the denominator of the differentiation.
dump: A `tfdbg.DebugDumpDir` object.
Returns:
If this `GradientsDebugger` instance has the gradient tensor of `x_tensor`
registered: a list of `numpy.ndarray` representing the value of the
gradient tensor from `dump`. The list could be empty, if the gradient
tensor is not executed in the `tf.Session.run()` call that generated
the `dump`. The list could also contain multiple values of the gradient
tensor, e.g., if gradient tensor is computed repeatedly in a
`tf.while_loop` during the run that generated the `dump`.
Raises:
LookupError: If this `GradientsDebugger` instance does not have the
gradient tensor of `x_tensor` registered.
ValueError: If this `GradientsDebugger` has a `tf.Graph` object that
does not match the `tf.Graph` object of the `dump`.
TypeError: If `x_tensor` is not a `tf.Tensor`, `tf.Variable` or `str`."
2141,parse_node_or_tensor_name,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,25,function,"Get the node name from a string that can be node or tensor name.
Args:
name: An input node name (e.g., ""node_a"") or tensor name (e.g.,
""node_a:0""), as a str.
Returns:
1) The node name, as a str. If the input name is a tensor name, i.e.,
consists of a colon, the final colon and the following output slot
will be stripped.
2) If the input name is a tensor name, the output slot, as an int. If
the input name is not a tensor name, None."
2142,get_node_name,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,49,function,
2143,get_output_slot,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,54,function,"Get the output slot number from the name of a graph element.
If element_name is a node name without output slot at the end, 0 will be
assumed.
Args:
element_name: (`str`) name of the graph element in question.
Returns:
(`int`) output slot number."
2144,is_copy_node,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,70,function,"Determine whether a node name is that of a debug Copy node.
Such nodes are inserted by TensorFlow core upon request in
RunOptions.debug_options.debug_tensor_watch_opts.
Args:
node_name: Name of the node.
Returns:
A bool indicating whether the input argument is the name of a debug Copy
node."
2145,is_debug_node,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,86,function,"Determine whether a node name is that of a debug node.
Such nodes are inserted by TensorFlow core upon request in
RunOptions.debug_options.debug_tensor_watch_opts.
Args:
node_name: Name of the node.
Returns:
A bool indicating whether the input argument is the name of a debug node."
2146,parse_debug_node_name,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,101,function,"Parse the name of a debug node.
Args:
node_name: Name of the debug node.
Returns:
1. Name of the watched node, as a str.
2. Output slot index of the watched tensor, as an int.
3. Index of the debug node, as an int.
4. Name of the debug op, as a str, e.g, ""DebugIdentity"".
Raises:
ValueError: If the input node name is not a valid debug node name."
2147,GraphTracingReachedDestination,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,142,class,
2148,DFSGraphTracer,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,146,class,Graph input tracer using depth-first search.
2149,trace,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,180,method,"Trace inputs.
Args:
graph_element_name: Name of the node or an output tensor of the node, as a
str.
Raises:
GraphTracingReachedDestination: if destination_node_name of this tracer
object is not None and the specified node is reached."
2150,inputs,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,216,method,
2151,depth_list,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,219,method,
2152,DebugGraph,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,237,class,Represents a debugger-decorated graph.
2153,device_name,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,434,method,
2154,debug_graph_def,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,438,method,The debugger-decorated GraphDef.
2155,non_debug_graph_def,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,443,method,The GraphDef without the Copy* and Debug* nodes added by the debugger.
2156,node_devices,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,449,method,
2157,node_op_types,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,453,method,
2158,node_attributes,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,457,method,
2159,node_inputs,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,461,method,
2160,node_ctrl_inputs,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,465,method,
2161,node_reversed_ref_inputs,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,469,method,
2162,node_recipients,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,473,method,
2163,node_ctrl_recipients,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,477,method,
2164,reconstruct_non_debug_graph_def,tensorflow/tensorflow/python/debug/lib/debug_graphs.py,481,function,"Reconstruct original (non-debugger-decorated) partition GraphDef.
This method strips the input `tf.compat.v1.GraphDef` of the Copy* and
Debug*-type nodes inserted by the debugger.
The reconstructed partition graph is identical to the original (i.e.,
non-debugger-decorated) partition graph except in the following respects:
1) The exact names of the runtime-inserted internal nodes may differ.
These include _Send, _Recv, _HostSend, _HostRecv, _Retval ops.
2) As a consequence of 1, the nodes that receive input directly from such
send- and recv-type ops will have different input names.
3) The parallel_iteration attribute of while-loop Enter ops are set to 1.
Args:
debug_graph_def: The debugger-decorated `tf.compat.v1.GraphDef`, with the
debugger-inserted Copy* and Debug* nodes.
Returns:
The reconstructed `tf.compat.v1.GraphDef` stripped of the debugger-inserted
nodes."
2165,EventListenerStub,tensorflow/tensorflow/python/debug/lib/debug_service_pb2_grpc.py,30,class,"EventListener: Receives Event protos, e.g., from debugged TensorFlow
runtime(s)."
2166,EventListenerServicer,tensorflow/tensorflow/python/debug/lib/debug_service_pb2_grpc.py,58,class,"EventListener: Receives Event protos, e.g., from debugged TensorFlow
runtime(s)."
2167,SendEvents,tensorflow/tensorflow/python/debug/lib/debug_service_pb2_grpc.py,63,method,"Client(s) can use this RPC method to send the EventListener Event protos.
The Event protos can hold information such as:
1) intermediate tensors from a debugged graph being executed, which can
be sent from DebugIdentity ops configured with grpc URLs.
2) GraphDefs of partition graphs, which can be sent from special debug
ops that get executed immediately after the beginning of the graph
execution."
2168,SendTracebacks,tensorflow/tensorflow/python/debug/lib/debug_service_pb2_grpc.py,76,method,"Send the tracebacks of ops in a Python graph definition.
"
2169,SendSourceFiles,tensorflow/tensorflow/python/debug/lib/debug_service_pb2_grpc.py,83,method,"Send a collection of source code files being debugged.
"
2170,add_EventListenerServicer_to_server,tensorflow/tensorflow/python/debug/lib/debug_service_pb2_grpc.py,91,function,
2171,add_debug_tensor_watch,tensorflow/tensorflow/python/debug/lib/debug_utils.py,26,function,"Add watch on a `Tensor` to `RunOptions`.
N.B.:
1. Under certain circumstances, the `Tensor` may not get actually watched
(e.g., if the node of the `Tensor` is constant-folded during runtime).
2. For debugging purposes, the `parallel_iteration` attribute of all
`tf.while_loop`s in the graph are set to 1 to prevent any node from
being executed multiple times concurrently. This change does not affect
subsequent non-debugged runs of the same `tf.while_loop`s.
Args:
run_options: An instance of `config_pb2.RunOptions` to be modified.
node_name: (`str`) name of the node to watch.
output_slot: (`int`) output slot index of the tensor from the watched node.
debug_ops: (`str` or `list` of `str`) name(s) of the debug op(s). Can be a
`list` of `str` or a single `str`. The latter case is equivalent to a
`list` of `str` with only one element.
For debug op types with customizable attributes, each debug op string can
optionally contain a list of attribute names, in the syntax of:
debug_op_name(attr_name_1=attr_value_1;attr_name_2=attr_value_2;...)
debug_urls: (`str` or `list` of `str`) URL(s) to send debug values to,
e.g., `file:///tmp/tfdbg_dump_1`, `grpc://localhost:12345`.
tolerate_debug_op_creation_failures: (`bool`) Whether to tolerate debug op
creation failures by not throwing exceptions.
global_step: (`int`) Optional global_step count for this debug tensor
watch."
2172,watch_graph,tensorflow/tensorflow/python/debug/lib/debug_utils.py,82,function,"Add debug watches to `RunOptions` for a TensorFlow graph.
To watch all `Tensor`s on the graph, let both `node_name_regex_allowlist`
and `op_type_regex_allowlist` be the default (`None`).
N.B.:
1. Under certain circumstances, the `Tensor` may not get actually watched
(e.g., if the node of the `Tensor` is constant-folded during runtime).
2. For debugging purposes, the `parallel_iteration` attribute of all
`tf.while_loop`s in the graph are set to 1 to prevent any node from
being executed multiple times concurrently. This change does not affect
subsequent non-debugged runs of the same `tf.while_loop`s.
Args:
run_options: An instance of `config_pb2.RunOptions` to be modified.
graph: An instance of `ops.Graph`.
debug_ops: (`str` or `list` of `str`) name(s) of the debug op(s) to use.
debug_urls: URLs to send debug values to. Can be a list of strings,
a single string, or None. The case of a single string is equivalent to
a list consisting of a single string, e.g., `file:///tmp/tfdbg_dump_1`,
`grpc://localhost:12345`.
For debug op types with customizable attributes, each debug op name string
can optionally contain a list of attribute names, in the syntax of:
debug_op_name(attr_name_1=attr_value_1;attr_name_2=attr_value_2;...)
node_name_regex_allowlist: Regular-expression allowlist for node_name,
e.g., `""(weight_[0-9]+|bias_.*)""`
op_type_regex_allowlist: Regular-expression allowlist for the op type of
nodes, e.g., `""(Variable|Add)""`.
If both `node_name_regex_allowlist` and `op_type_regex_allowlist`
are set, the two filtering operations will occur in a logical `AND`
relation. In other words, a node will be included if and only if it
hits both allowlists.
tensor_dtype_regex_allowlist: Regular-expression allowlist for Tensor
data type, e.g., `""^int.*""`.
This allowlist operates in logical `AND` relations to the two allowlists
above.
tolerate_debug_op_creation_failures: (`bool`) whether debug op creation
failures (e.g., due to dtype incompatibility) are to be tolerated by not
throwing exceptions.
global_step: (`int`) Optional global_step count for this debug tensor
watch.
reset_disk_byte_usage: (`bool`) whether to reset the tracked disk byte
usage to zero (default: `False`)."
2173,watch_graph_with_denylists,tensorflow/tensorflow/python/debug/lib/debug_utils.py,202,function,"Add debug tensor watches, denylisting nodes and op types.
This is similar to `watch_graph()`, but the node names and op types are
denylisted, instead of allowlisted.
N.B.:
1. Under certain circumstances, the `Tensor` may not get actually watched
(e.g., if the node of the `Tensor` is constant-folded during runtime).
2. For debugging purposes, the `parallel_iteration` attribute of all
`tf.while_loop`s in the graph are set to 1 to prevent any node from
being executed multiple times concurrently. This change does not affect
subsequent non-debugged runs of the same `tf.while_loop`s.
Args:
run_options: An instance of `config_pb2.RunOptions` to be modified.
graph: An instance of `ops.Graph`.
debug_ops: (`str` or `list` of `str`) name(s) of the debug op(s) to use. See
the documentation of `watch_graph` for more details.
debug_urls: URL(s) to send debug values to, e.g.,
`file:///tmp/tfdbg_dump_1`, `grpc://localhost:12345`.
node_name_regex_denylist: Regular-expression denylist for node_name. This
should be a string, e.g., `""(weight_[0-9]+|bias_.*)""`.
op_type_regex_denylist: Regular-expression denylist for the op type of
nodes, e.g., `""(Variable|Add)""`. If both node_name_regex_denylist and
op_type_regex_denylist are set, the two filtering operations will occur in
a logical `OR` relation. In other words, a node will be excluded if it
hits either of the two denylists; a node will be included if and only if
it hits neither of the denylists.
tensor_dtype_regex_denylist: Regular-expression denylist for Tensor data
type, e.g., `""^int.*""`. This denylist operates in logical `OR` relations
to the two allowlists above.
tolerate_debug_op_creation_failures: (`bool`) whether debug op creation
failures (e.g., due to dtype incompatibility) are to be tolerated by not
throwing exceptions.
global_step: (`int`) Optional global_step count for this debug tensor watch.
reset_disk_byte_usage: (`bool`) whether to reset the tracked disk byte
usage to zero (default: `False`)."
2174,is_op_type_function,tensorflow/tensorflow/python/debug/lib/dumping_callback.py,61,function,
2175,enable_dump_debug_info,tensorflow/tensorflow/python/debug/lib/dumping_callback.py,687,function,"Enable dumping debugging information from a TensorFlow program.
The debugging information is dumped to a directory on the file system
specified as `dump_root`.
The dumped debugging information can be ingested by debugger UIs.
The files in the dump directory contain the following information:
- TensorFlow Function construction (e.g., compilation of Python functions
decorated with @tf.function), the op types, names (if available), context,
the input and output tensors, and the associated stack traces.
- Execution of TensorFlow operations (ops) and Functions and their stack
traces, op types, names (if available) and contexts. In addition,
depending on the value of the `tensor_debug_mode` argument (see Args
section below), the value(s) of the output tensors or more concise
summaries of the tensor values will be dumped.
- A snapshot of Python source files involved in the execution of the
TensorFlow program.
Once enabled, the dumping can be disabled with the corresponding
`disable_dump_debug_info()` method under the same Python namespace.
Calling this method more than once with the same `dump_root` is idempotent.
Calling this method more than once with different `tensor_debug_mode`s
leads to a `ValueError`.
Calling this method more than once with different `circular_buffer_size`s
leads to a `ValueError`.
Calling this method with a different `dump_root` abolishes the
previously-enabled `dump_root`.
Usage example:
```py
tf.debugging.experimental.enable_dump_debug_info('/tmp/my-tfdbg-dumps')
# Code to build, train and run your TensorFlow model...
```
NOTE: If your code is running on TPUs, be sure to call
`tf.config.set_soft_device_placement(True)` before calling
`tf.debugging.experimental.enable_dump_debug_info()` as this API uses
automatic outside compilation on TPUs. For example:
```py
tf.config.set_soft_device_placement(True)
tf.debugging.experimental.enable_dump_debug_info(
logdir, tensor_debug_mode=""FULL_HEALTH"")
resolver = tf.distribute.cluster_resolver.TPUClusterResolver(tpu='')
strategy = tf.distribute.TPUStrategy(resolver)
with strategy.scope():
# ...
```
Args:
dump_root: The directory path where the dumping information will be written.
tensor_debug_mode: Debug mode for tensor values, as a string.
The currently supported options are:
- ""NO_TENSOR"": (Default) Only traces the output tensors of all executed
ops (including those executed eagerly at the Python level or as a part
of a TensorFlow graph) and functions, while not extracting any
information from the values of the tensors.
- ""CURT_HEALTH"": For each floating-dtype tensor (e.g., tensors of dtypes
such as `float32`, `float64` and `bfloat16`), extracts a binary bit
indicating whether it contains any -infinity, +infinity or NaN.
- ""CONCISE_HEALTH"": For each floating-dtype tensor, extract total
element count, and counts of -infinity, +infinity and NaN elements.
- ""FULL_HEALTH"": For each floating-dtype tensor, extracts the dtype,
rank (number of dimensions), total element count, and counts of
-infinity, +infinity and NaN elements.
- ""SHAPE"": For each tensor (regardless of dtype), extracts its dtype,
rank, total element count and shape.
circular_buffer_size: Size of the circular buffers for execution events.
These circular buffers are designed to reduce the overhead of debugging
dumping. They hold the most recent debug events concerning eager execution
of ops and `tf.function`s and traces of tensor values computed inside
`tf.function`s. They are written to the file system only when the proper
flushing method is called (see description of return values below).
Expected to be an integer. If <= 0, the circular-buffer behavior will be
disabled, i.e., the execution debug events will be written to the file
writers in the same way as non-execution events such as op creations and
source-file snapshots.
op_regex: Dump data from only the tensors from op types that matches to the
regular expression (through Python's `re.match()`).
""Op type"" refers to the names of the TensorFlow operations (e.g.,
""MatMul"", ""LogSoftmax""), which may repeat in a TensorFlow
function. It does *not* refer to the names of nodes (e.g.,
""dense/MatMul"", ""dense_1/MatMul_1"") which are unique within a function.
- Example 1: Dump tensor data from only MatMul and Relu ops
`op_regex=""^(MatMul|Relu)$""`.
- Example 2: Dump tensors from all ops *except* Relu:
`op_regex=""(?!^Relu$)""`.
This filter operates in a logical AND relation with `tensor_dtypes`.
tensor_dtypes: Dump data from only the tensors of which the specified
dtypes. This optional argument can be in any of the following format:
- a list or tuple of `DType` objects or strings that can be converted
to `DType` objects via `tf.as_dtype()`. Examples:
- `tensor_dtype=[tf.float32, tf.float64]`,
- `tensor_dtype=[""float32"", ""float64""]`,
- `tensor_dtypes=(tf.int32, tf.bool)`,
- `tensor_dtypes=(""int32"", ""bool"")`
- a callable that takes a single `DType` argument and returns a Python
`boolean` indicating whether the dtype is to be included in the data
dumping. Examples:
- `tensor_dtype=lambda dtype: dtype.is_integer`.
This filter operates in a logical AND relation with `op_regex`.
Returns:
A DebugEventsWriter instance used by the dumping callback. The caller
may use its flushing methods, including `FlushNonExecutionFiles()` and
`FlushExecutionFiles()`."
2176,disable_dump_debug_info,tensorflow/tensorflow/python/debug/lib/dumping_callback.py,878,function,"Disable the currently-enabled debugging dumping.
If the `enable_dump_debug_info()` method under the same Python namespace
has been invoked before, calling this method disables it. If no call to
`enable_dump_debug_info()` has been made, calling this method is a no-op.
Calling this method more than once is idempotent."
2177,EventListenerBaseStreamHandler,tensorflow/tensorflow/python/debug/lib/grpc_debug_server.py,50,class,Per-stream handler of EventListener gRPC streams.
2178,on_core_metadata_event,tensorflow/tensorflow/python/debug/lib/grpc_debug_server.py,56,method,"Callback for core metadata.
Args:
event: The Event proto that carries a JSON string in its
`log_message.message` field.
Returns:
`None` or an `EventReply` proto to be sent back to the client. If `None`,
an `EventReply` proto construct with the default no-arg constructor will
be sent back to the client."
2179,on_graph_def,tensorflow/tensorflow/python/debug/lib/grpc_debug_server.py,72,method,"Callback for Event proto received through the gRPC stream.
This Event proto carries a GraphDef, encoded as bytes, in its graph_def
field.
Args:
graph_def: A GraphDef object.
device_name: Name of the device on which the graph was created.
wall_time: An epoch timestamp (in microseconds) for the graph.
Returns:
`None` or an `EventReply` proto to be sent back to the client. If `None`,
an `EventReply` proto construct with the default no-arg constructor will
be sent back to the client."
2180,on_value_event,tensorflow/tensorflow/python/debug/lib/grpc_debug_server.py,91,method,"Callback for Event proto received through the gRPC stream.
This Event proto carries a Tensor in its summary.value[0] field.
Args:
event: The Event proto from the stream to be processed."
2181,EventListenerBaseServicer,tensorflow/tensorflow/python/debug/lib/grpc_debug_server.py,103,class,Base Python class for gRPC debug server.
2182,SendEvents,tensorflow/tensorflow/python/debug/lib/grpc_debug_server.py,127,method,"Implementation of the SendEvents service method.
This method receives streams of Event protos from the client, and processes
them in ways specified in the on_event() callback. The stream is
bi-directional, but currently only the client-to-server stream (i.e., the
stream from the debug ops to the server) is used.
Args:
request_iterator: The incoming stream of Event protos.
context: Server context.
Raises:
ValueError: If there are more than one core metadata events.
Yields:
An empty stream of responses."
2183,run_server,tensorflow/tensorflow/python/debug/lib/grpc_debug_server.py,332,method,"Start running the server.
Args:
blocking: If `True`, block until `stop_server()` is invoked.
Raises:
ValueError: If server stop has already been requested, or if the server
has already started running."
2184,stop_server,tensorflow/tensorflow/python/debug/lib/grpc_debug_server.py,365,method,"Request server stopping.
Once stopped, server cannot be stopped or started again. This method is
non-blocking. Call `wait()` on the returned event to block until the server
has completely stopped.
Args:
grace: Grace period in seconds to be used when calling `server.stop()`.
Raises:
ValueError: If server stop has already been requested, or if the server
has not started running yet.
Returns:
A threading.Event that will be set when the server has completely stopped."
2185,request_watch,tensorflow/tensorflow/python/debug/lib/grpc_debug_server.py,394,method,"Request enabling a debug tensor watchpoint or breakpoint.
This will let the server send a EventReply to the client side
(i.e., the debugged TensorFlow runtime process) to request adding a watch
key (i.e., <node_name>:<output_slot>:<debug_op>) to the list of enabled
watch keys. The list applies only to debug ops with the attribute
gated_grpc=True.
To disable the watch, use `request_unwatch()`.
Args:
node_name: (`str`) name of the node that the to-be-watched tensor belongs
to, e.g., ""hidden/Weights"".
output_slot: (`int`) output slot index of the tensor to watch.
debug_op: (`str`) name of the debug op to enable. This should not include
any attribute substrings.
breakpoint: (`bool`) Iff `True`, the debug op will block and wait until it
receives an `EventReply` response from the server. The `EventReply`
proto may carry a TensorProto that modifies the value of the debug op's
output tensor."
2186,request_unwatch,tensorflow/tensorflow/python/debug/lib/grpc_debug_server.py,423,method,"Request disabling a debug tensor watchpoint or breakpoint.
This is the opposite of `request_watch()`.
Args:
node_name: (`str`) name of the node that the to-be-watched tensor belongs
to, e.g., ""hidden/Weights"".
output_slot: (`int`) output slot index of the tensor to watch.
debug_op: (`str`) name of the debug op to enable. This should not include
any attribute substrings."
2187,breakpoints,tensorflow/tensorflow/python/debug/lib/grpc_debug_server.py,441,method,"Get a set of the currently-activated breakpoints.
Returns:
A `set` of 3-tuples: (node_name, output_slot, debug_op), e.g.,
{(""MatMul"", 0, ""DebugIdentity"")}."
2188,gated_grpc_debug_watches,tensorflow/tensorflow/python/debug/lib/grpc_debug_server.py,450,method,"Get the list of debug watches with attribute gated_grpc=True.
Since the server receives `GraphDef` from the debugged runtime, it can only
return such debug watches that it has received so far.
Returns:
A `list` of `DebugWatch` `namedtuples` representing the debug watches with
gated_grpc=True. Each `namedtuple` element has the attributes:
`node_name` as a `str`,
`output_slot` as an `int`,
`debug_op` as a `str`."
2189,SendTracebacks,tensorflow/tensorflow/python/debug/lib/grpc_debug_server.py,465,method,"Base implementation of the handling of SendTracebacks calls.
The base implementation does nothing with the incoming request.
Override in an implementation of the server if necessary.
Args:
request: A `CallTraceback` proto, containing information about the
type (e.g., graph vs. eager execution) and source-code traceback of the
call and (any) associated `tf.Graph`s.
context: Server context.
Returns:
A `EventReply` proto."
2190,SendSourceFiles,tensorflow/tensorflow/python/debug/lib/grpc_debug_server.py,482,method,"Base implementation of the handling of SendSourceFiles calls.
The base implementation does nothing with the incoming request.
Override in an implementation of the server if necessary.
Args:
request: A `DebuggedSourceFiles` proto, containing the path, content, size
and last-modified timestamp of source files.
context: Server context.
Returns:
A `EventReply` proto."
2191,start_server_on_separate_thread,tensorflow/tensorflow/python/debug/lib/grpc_debug_test_server.py,368,function,"Create a test gRPC debug server and run on a separate thread.
Args:
dump_to_filesystem: (bool) whether the debug server will dump debug data
to the filesystem.
server_start_delay_sec: (float) amount of time (in sec) to delay the server
start up for.
poll_server: (bool) whether the server will be polled till success on
startup.
blocking: (bool) whether the server should be started in a blocking mode.
toggle_watch_on_core_metadata: A list of
(node_name, output_slot, debug_op) tuples to toggle the
watchpoint status during the on_core_metadata calls (optional).
Returns:
server_port: (int) Port on which the server runs.
debug_server_url: (str) grpc:// URL to the server.
server_dump_dir: (str) The debug server's dump directory.
server_thread: The server Thread object.
server: The `EventListenerTestServicer` object.
Raises:
ValueError: If polling the server process for ready state is not successful
within maximum polling count."
2192,parse_cluster_spec,tensorflow/tensorflow/python/debug/lib/grpc_tensorflow_server.py,45,function,"Parse content of cluster_spec string and inject info into cluster protobuf.
Args:
cluster_spec: cluster specification string, e.g.,
""local|localhost:2222;localhost:2223""
cluster: cluster protobuf.
verbose: If verbose logging is requested.
Raises:
ValueError: if the cluster_spec string is invalid."
2193,ProfileDatum,tensorflow/tensorflow/python/debug/lib/profiling.py,24,class,Profile data point.
2194,exec_time,tensorflow/tensorflow/python/debug/lib/profiling.py,60,method,Op execution time plus pre- and post-processing.
2195,AggregateProfile,tensorflow/tensorflow/python/debug/lib/profiling.py,65,class,Profile summary data for aggregating a number of ProfileDatum.
2196,add,tensorflow/tensorflow/python/debug/lib/profiling.py,82,method,"Accumulate a new instance of ProfileDatum.
Args:
profile_datum: (`ProfileDatum`) an instance of `ProfileDatum` to
accumulate to this object."
2197,node_count,tensorflow/tensorflow/python/debug/lib/profiling.py,103,method,
2198,node_exec_count,tensorflow/tensorflow/python/debug/lib/profiling.py,107,method,
2199,AggregateProfile,tensorflow/tensorflow/python/debug/lib/profiling_test.py,27,class,
2200,setUp,tensorflow/tensorflow/python/debug/lib/profiling_test.py,29,method,
2201,no_rewrite_session_config,tensorflow/tensorflow/python/debug/lib/session_debug_testlib.py,58,function,
2202,send_graph_tracebacks,tensorflow/tensorflow/python/debug/lib/source_remote.py,176,function,"Send the tracebacks of a graph execution call to debug server(s).
Args:
destinations: gRPC destination addresses, a `str` or a `list` of `str`s,
e.g., ""localhost:4242"". If a `list`, gRPC requests containing the same
`CallTraceback` proto payload will be sent to all the destinations.
run_key: A string describing the feeds, fetches (and targets) names of the
`tf.Session.run` call.
origin_stack: The traceback of the `tf.Session.run()` invocation.
graph: A Python `tf.Graph` object (i.e., *not* a `tf.compat.v1.GraphDef`),
which contains op tracebacks.
send_source: Whether the source files involved in the op tracebacks but
outside the TensorFlow library are to be sent."
2203,send_eager_tracebacks,tensorflow/tensorflow/python/debug/lib/source_remote.py,200,function,"Send the tracebacks of an eager execution call to debug server(s).
Args:
destinations: gRPC destination addresses, a `str` or a `list` of `str`s,
e.g., ""localhost:4242"". If a `list`, gRPC requests containing the same
origin_stack: The traceback of the eager operation invocation.
send_source: Whether the source files involved in the op tracebacks but
outside the TensorFlow library are to be sent."
2204,line_number_above,tensorflow/tensorflow/python/debug/lib/source_remote_test.py,42,function,
2205,is_extension_uncompiled_python_source,tensorflow/tensorflow/python/debug/lib/source_utils.py,47,function,
2206,is_extension_compiled_python_source,tensorflow/tensorflow/python/debug/lib/source_utils.py,52,function,
2207,guess_is_tensorflow_py_library,tensorflow/tensorflow/python/debug/lib/source_utils.py,61,function,"Guess whether a Python source file is a part of the tensorflow library.
Special cases:
1) Returns False for unit-test files in the library (*_test.py),
2) Returns False for files under python/debug/examples.
Args:
py_file_path: full path of the Python source file in question.
Returns:
(`bool`) Whether the file is inferred to be a part of the tensorflow
library."
2208,load_source,tensorflow/tensorflow/python/debug/lib/source_utils.py,86,function,"Load the content of a Python source code file.
This function covers the following case:
1. source_file_path points to an existing Python (.py) file on the
file system.
2. source_file_path is a path within a .par file (i.e., a zip-compressed,
self-contained Python executable).
Args:
source_file_path: Path to the Python source file to read.
Returns:
A length-2 tuple:
- Lines of the source file, as a `list` of `str`s.
- The width of the string needed to show the line number in the file.
This is calculated based on the number of lines in the source file.
Raises:
IOError: if loading is unsuccessful."
2209,annotate_source,tensorflow/tensorflow/python/debug/lib/source_utils.py,156,function,"Annotate a Python source file with a list of ops created at each line.
(The annotation doesn't change the source file itself.)
Args:
dump: (`DebugDumpDir`) A `DebugDumpDir` object of which the Python graph
has been loaded.
source_file_path: (`str`) Path to the source file being annotated.
do_dumped_tensors: (`str`) Whether dumped Tensors, instead of ops are to be
used to annotate the source file.
file_stack_top: (`bool`) Whether only the top stack trace in the
specified source file is to be annotated.
min_line: (`None` or `int`) The 1-based line to start annotate the source
file from (inclusive).
max_line: (`None` or `int`) The 1-based line number to end the annotation
at (exclusive).
Returns:
A `dict` mapping 1-based line number to a list of op name(s) created at
that line, or tensor names if `do_dumped_tensors` is True.
Raises:
ValueError: If the dump object does not have a Python graph set."
2210,list_source_files_against_dump,tensorflow/tensorflow/python/debug/lib/source_utils.py,223,function,"Generate a list of source files with information regarding ops and tensors.
Args:
dump: (`DebugDumpDir`) A `DebugDumpDir` object of which the Python graph
has been loaded.
path_regex_allowlist: A regular-expression filter for source file path.
node_name_regex_allowlist: A regular-expression filter for node names.
Returns:
A list of tuples regarding the Python source files involved in constructing
the ops and tensors contained in `dump`. Each tuple is:
(source_file_path, is_tf_library, num_nodes, num_tensors, num_dumps,
first_line)
is_tf_library: (`bool`) A guess of whether the file belongs to the
TensorFlow Python library.
num_nodes: How many nodes were created by lines of this source file.
These include nodes with dumps and those without.
num_tensors: How many Tensors were created by lines of this source file.
These include Tensors with dumps and those without.
num_dumps: How many debug Tensor dumps were from nodes (and Tensors)
that were created by this source file.
first_line: The first line number (1-based) that created any nodes or
Tensors in this source file.
The list is sorted by ascending order of source_file_path.
Raises:
ValueError: If the dump object does not have a Python graph set."
2211,annotate_source_against_profile,tensorflow/tensorflow/python/debug/lib/source_utils.py,324,function,"Annotate a Python source file with profiling information at each line.
(The annotation doesn't change the source file itself.)
Args:
profile_data: (`list` of `ProfileDatum`) A list of `ProfileDatum`.
source_file_path: (`str`) Path to the source file being annotated.
node_name_filter: Regular expression to filter by node name.
op_type_filter: Regular expression to filter by op type.
min_line: (`None` or `int`) The 1-based line to start annotate the source
file from (inclusive).
max_line: (`None` or `int`) The 1-based line number to end the annotation
at (exclusive).
Returns:
A `dict` mapping 1-based line number to a the namedtuple
`profiling.LineOrFuncProfileSummary`."
2212,line_number_above,tensorflow/tensorflow/python/debug/lib/source_utils_test.py,47,function,"Get lineno of the AST node immediately above this function's call site.
It is assumed that there is no empty line(s) between the call site and the
preceding AST node.
Returns:
The lineno of the preceding AST node, at the same level of the AST.
If the preceding AST spans multiple lines:
- In Python 3.8+, the lineno of the first line is returned.
- In older Python versions, the lineno of the last line is returned."
2213,DumpingDebugWrapperSession,tensorflow/tensorflow/python/debug/wrappers/dumping_wrapper.py,31,class,Debug Session wrapper that dumps debug data to filesystem.
2214,prepare_run_debug_urls,tensorflow/tensorflow/python/debug/wrappers/dumping_wrapper.py,92,method,"Implementation of abstract method in superclass.
See doc of `NonInteractiveDebugWrapperSession.prepare_run_debug_urls()`
for details. This implementation creates a run-specific subdirectory under
self._session_root and stores information regarding run `fetches` and
`feed_dict.keys()` in the subdirectory.
Args:
fetches: Same as the `fetches` argument to `Session.run()`
feed_dict: Same as the `feed_dict` argument to `Session.run()`
Returns:
debug_urls: (`str` or `list` of `str`) file:// debug URLs to be used in
this `Session.run()` call."
2215,OnSessionInitRequest,tensorflow/tensorflow/python/debug/wrappers/framework.py,135,class,"Request to an on-session-init callback.
This callback is invoked during the __init__ call to a debug-wrapper session."
2216,OnSessionInitAction,tensorflow/tensorflow/python/debug/wrappers/framework.py,152,class,Enum-like values for possible action to take on session init.
2217,OnSessionInitResponse,tensorflow/tensorflow/python/debug/wrappers/framework.py,168,class,Response from an on-session-init callback.
2218,OnRunStartRequest,tensorflow/tensorflow/python/debug/wrappers/framework.py,181,class,"Request to an on-run-start callback.
This callback is invoked during a run() call of the debug-wrapper
session, immediately after the run() call counter is incremented."
2219,OnRunStartAction,tensorflow/tensorflow/python/debug/wrappers/framework.py,212,class,Enum-like values for possible action to take on start of a run() call.
2220,OnRunStartResponse,tensorflow/tensorflow/python/debug/wrappers/framework.py,226,class,"Request from an on-run-start callback.
The caller of the callback can use this response object to specify what
action the debug-wrapper session actually takes on the run() call."
2221,OnRunEndRequest,tensorflow/tensorflow/python/debug/wrappers/framework.py,274,class,"Request to an on-run-end callback.
The callback is invoked immediately before the wrapped run() call ends."
2222,OnRunEndResponse,tensorflow/tensorflow/python/debug/wrappers/framework.py,308,class,Response from an on-run-end callback.
2223,BaseDebugWrapperSession,tensorflow/tensorflow/python/debug/wrappers/framework.py,318,class,"Base class of debug-wrapper session classes.
Concrete classes that inherit from this class need to implement the abstract
methods such as on_session_init, on_run_start and on_run_end."
2224,graph,tensorflow/tensorflow/python/debug/wrappers/framework.py,382,method,
2225,graph_def,tensorflow/tensorflow/python/debug/wrappers/framework.py,386,method,
2226,sess_str,tensorflow/tensorflow/python/debug/wrappers/framework.py,390,method,
2227,session,tensorflow/tensorflow/python/debug/wrappers/framework.py,394,method,
2228,run,tensorflow/tensorflow/python/debug/wrappers/framework.py,397,method,"Wrapper around Session.run() that inserts tensor watch options.
Args:
fetches: Same as the `fetches` arg to regular `Session.run()`.
feed_dict: Same as the `feed_dict` arg to regular `Session.run()`.
options: Same as the `options` arg to regular `Session.run()`.
run_metadata: Same as the `run_metadata` arg to regular `Session.run()`.
callable_runner: A `callable` returned by `Session.make_callable()`.
If not `None`, `fetches` and `feed_dict` must both be `None`.
Mutually exclusive with `callable_options`.
callable_runner_args: An optional list of arguments to `callable_runner`
or for `callable_options`.
callable_options: An instance of `config_pb2.CallableOptions`, to be
used with `Session._make_callable_from_options()`. Mutually exclusive
with `callable_runner`.
Returns:
Simply forwards the output of the wrapped `Session.run()` call.
Raises:
ValueError: On invalid `OnRunStartAction` value. Or if `callable_runner`
is not `None` and either or both of `fetches` and `feed_dict` is `None`."
2229,run_step_fn,tensorflow/tensorflow/python/debug/wrappers/framework.py,643,method,
2230,partial_run_setup,tensorflow/tensorflow/python/debug/wrappers/framework.py,647,method,Sets up the feeds and fetches for partial runs in the session.
2231,partial_run,tensorflow/tensorflow/python/debug/wrappers/framework.py,652,method,
2232,list_devices,tensorflow/tensorflow/python/debug/wrappers/framework.py,656,method,
2233,reset,tensorflow/tensorflow/python/debug/wrappers/framework.py,659,method,
2234,make_callable,tensorflow/tensorflow/python/debug/wrappers/framework.py,662,method,
2235,run_call_count,tensorflow/tensorflow/python/debug/wrappers/framework.py,686,method,
2236,increment_run_call_count,tensorflow/tensorflow/python/debug/wrappers/framework.py,689,method,
2237,on_session_init,tensorflow/tensorflow/python/debug/wrappers/framework.py,755,method,"Callback invoked during construction of the debug-wrapper session.
This is a blocking callback.
The invocation happens right before the constructor ends.
Args:
request: (`OnSessionInitRequest`) callback request carrying information
such as the session being wrapped.
Returns:
An instance of `OnSessionInitResponse`."
2238,on_run_start,tensorflow/tensorflow/python/debug/wrappers/framework.py,770,method,"Callback invoked on run() calls to the debug-wrapper session.
This is a blocking callback.
The invocation happens after the wrapper's run() call is entered,
after an increment of run call counter.
Args:
request: (`OnRunStartRequest`) callback request object carrying
information about the run call such as the fetches, feed dict, run
options, run metadata, and how many `run()` calls to this wrapper
session have occurred.
Returns:
An instance of `OnRunStartResponse`, carrying information to
debug URLs used to watch the tensors."
2239,on_run_end,tensorflow/tensorflow/python/debug/wrappers/framework.py,789,method,"Callback invoked on run() calls to the debug-wrapper session.
This is a blocking callback.
The invocation happens right before the wrapper exits its run() call.
Args:
request: (`OnRunEndRequest`) callback request object carrying information
such as the actual action performed by the session wrapper for the
run() call.
Returns:
An instance of `OnRunStartResponse`."
2240,as_default,tensorflow/tensorflow/python/debug/wrappers/framework.py,804,method,
2241,close,tensorflow/tensorflow/python/debug/wrappers/framework.py,820,method,
2242,should_stop,tensorflow/tensorflow/python/debug/wrappers/framework.py,826,method,
2243,is_empty,tensorflow/tensorflow/python/debug/wrappers/framework.py,444,method,Check whether a possibly nested structure is empty.
2244,wrapped_runner,tensorflow/tensorflow/python/debug/wrappers/framework.py,668,method,
2245,wrapped_runner,tensorflow/tensorflow/python/debug/wrappers/framework.py,678,method,
2246,WatchOptions,tensorflow/tensorflow/python/debug/wrappers/framework.py,835,class,Type for return values of watch_fn.
2247,NonInteractiveDebugWrapperSession,tensorflow/tensorflow/python/debug/wrappers/framework.py,885,class,"Base class for non-interactive (i.e., non-CLI) debug wrapper sessions."
2248,on_session_init,tensorflow/tensorflow/python/debug/wrappers/framework.py,924,method,See doc of BaseDebugWrapperSession.on_run_start.
2249,prepare_run_debug_urls,tensorflow/tensorflow/python/debug/wrappers/framework.py,930,method,"Abstract method to be implemented by concrete subclasses.
This method prepares the run-specific debug URL(s).
Args:
fetches: Same as the `fetches` argument to `Session.run()`
feed_dict: Same as the `feed_dict` argument to `Session.run()`
Returns:
debug_urls: (`str` or `list` of `str`) Debug URLs to be used in
this `Session.run()` call."
2250,on_run_start,tensorflow/tensorflow/python/debug/wrappers/framework.py,944,method,See doc of BaseDebugWrapperSession.on_run_start.
2251,on_run_end,tensorflow/tensorflow/python/debug/wrappers/framework.py,985,method,See doc of BaseDebugWrapperSession.on_run_end.
2252,publish_traceback,tensorflow/tensorflow/python/debug/wrappers/grpc_wrapper.py,31,function,"Publish traceback and source code if graph version is new.
`graph.version` is compared with `old_graph_version`. If the former is higher
(i.e., newer), the graph traceback and the associated source code is sent to
the debug server at the specified gRPC URLs.
Args:
debug_server_urls: A single gRPC debug server URL as a `str` or a `list` of
debug server URLs.
graph: A Python `tf.Graph` object.
feed_dict: Feed dictionary given to the `Session.run()` call.
fetches: Fetches from the `Session.run()` call.
old_graph_version: Old graph version to compare to.
Returns:
If `graph.version > old_graph_version`, the new graph version as an `int`.
Else, the `old_graph_version` is returned."
2253,GrpcDebugWrapperSession,tensorflow/tensorflow/python/debug/wrappers/grpc_wrapper.py,69,class,Debug Session wrapper that send debug data to gRPC stream(s).
2254,prepare_run_debug_urls,tensorflow/tensorflow/python/debug/wrappers/grpc_wrapper.py,122,method,"Implementation of abstract method in superclass.
See doc of `NonInteractiveDebugWrapperSession.prepare_run_debug_urls()`
for details.
Args:
fetches: Same as the `fetches` argument to `Session.run()`
feed_dict: Same as the `feed_dict` argument to `Session.run()`
Returns:
debug_urls: (`str` or `list` of `str`) file:// debug URLs to be used in
this `Session.run()` call."
2255,register_signal_handler,tensorflow/tensorflow/python/debug/wrappers/grpc_wrapper.py,154,function,
2256,TensorBoardDebugWrapperSession,tensorflow/tensorflow/python/debug/wrappers/grpc_wrapper.py,162,class,"A tfdbg Session wrapper that can be used with TensorBoard Debugger Plugin.
This wrapper is the same as `GrpcDebugWrapperSession`, except that it uses a
predefined `watch_fn` that
1) uses `DebugIdentity` debug ops with the `gated_grpc` attribute set to
`True` to allow the interactive enabling and disabling of tensor
breakpoints.
2) watches all tensors in the graph.
This saves the need for the user to define a `watch_fn`."
2257,run,tensorflow/tensorflow/python/debug/wrappers/grpc_wrapper.py,212,method,
2258,LocalCLIDebugHook,tensorflow/tensorflow/python/debug/wrappers/hooks.py,30,class,"Command-line-interface debugger hook.
Can be used as a hook for `tf.compat.v1.train.MonitoredSession`s and
`tf.estimator.Estimator`s. Provides a substitute for
`tfdbg.LocalCLIDebugWrapperSession` in cases where the session is not directly
available."
2259,add_tensor_filter,tensorflow/tensorflow/python/debug/wrappers/hooks.py,67,method,"Add a tensor filter.
See doc of `LocalCLIDebugWrapperSession.add_tensor_filter()` for details.
Override default behavior to accommodate the possibility of this method
being
called prior to the initialization of the underlying
`LocalCLIDebugWrapperSession` object.
Args:
filter_name: See doc of `LocalCLIDebugWrapperSession.add_tensor_filter()`
for details.
tensor_filter: See doc of
`LocalCLIDebugWrapperSession.add_tensor_filter()` for details."
2260,begin,tensorflow/tensorflow/python/debug/wrappers/hooks.py,88,method,
2261,before_run,tensorflow/tensorflow/python/debug/wrappers/hooks.py,91,method,
2262,after_run,tensorflow/tensorflow/python/debug/wrappers/hooks.py,143,method,
2263,DumpingDebugHook,tensorflow/tensorflow/python/debug/wrappers/hooks.py,151,class,"A debugger hook that dumps debug data to filesystem.
Can be used as a hook for `tf.compat.v1.train.MonitoredSession`s and
`tf.estimator.Estimator`s."
2264,begin,tensorflow/tensorflow/python/debug/wrappers/hooks.py,182,method,
2265,before_run,tensorflow/tensorflow/python/debug/wrappers/hooks.py,185,method,
2266,after_run,tensorflow/tensorflow/python/debug/wrappers/hooks.py,219,method,
2267,GrpcDebugHook,tensorflow/tensorflow/python/debug/wrappers/hooks.py,223,class,"A hook that streams debugger-related events to any grpc_debug_server.
For example, the debugger data server is a grpc_debug_server. The debugger
data server writes debugger-related events it receives via GRPC to logdir.
This enables debugging features in Tensorboard such as health pills.
When the arguments of debug_utils.watch_graph changes, strongly consider
changing arguments here too so that features are available to tflearn users.
Can be used as a hook for `tf.compat.v1.train.MonitoredSession`s and
`tf.estimator.Estimator`s."
2268,before_run,tensorflow/tensorflow/python/debug/wrappers/hooks.py,266,method,"Called right before a session is run.
Args:
run_context: A session_run_hook.SessionRunContext. Encapsulates
information on the run.
Returns:
A session_run_hook.SessionRunArgs object."
2269,TensorBoardDebugHook,tensorflow/tensorflow/python/debug/wrappers/hooks.py,305,class,"A tfdbg hook that can be used with TensorBoard Debugger Plugin.
This hook is the same as `GrpcDebugHook`, except that it uses a predefined
`watch_fn` that
1) uses `DebugIdentity` debug ops with the `gated_grpc` attribute set to
`True`, to allow the interactive enabling and disabling of tensor
breakpoints.
2) watches all tensors in the graph.
This saves the need for the user to define a `watch_fn`."
2270,before_run,tensorflow/tensorflow/python/debug/wrappers/hooks.py,351,method,
2271,LocalCLIDebugWrapperSession,tensorflow/tensorflow/python/debug/wrappers/local_cli_wrapper.py,43,class,"Concrete subclass of BaseDebugWrapperSession implementing a local CLI.
This class has all the methods that a `session.Session` object has, in order
to support debugging with minimal code changes. Invoking its `run()` method
will launch the command-line interface (CLI) of tfdbg."
2272,add_tensor_filter,tensorflow/tensorflow/python/debug/wrappers/local_cli_wrapper.py,209,method,"Add a tensor filter.
Args:
filter_name: (`str`) name of the filter.
tensor_filter: (`callable`) the filter callable. See the doc string of
`DebugDumpDir.find()` for more details about its signature."
2273,on_session_init,tensorflow/tensorflow/python/debug/wrappers/local_cli_wrapper.py,220,method,"Overrides on-session-init callback.
Args:
request: An instance of `OnSessionInitRequest`.
Returns:
An instance of `OnSessionInitResponse`."
2274,on_run_start,tensorflow/tensorflow/python/debug/wrappers/local_cli_wrapper.py,233,method,"Overrides on-run-start callback.
Args:
request: An instance of `OnRunStartRequest`.
Returns:
An instance of `OnRunStartResponse`."
2275,on_run_end,tensorflow/tensorflow/python/debug/wrappers/local_cli_wrapper.py,306,method,"Overrides on-run-end callback.
Actions taken:
1) Load the debug dump.
2) Bring up the Analyzer CLI.
Args:
request: An instance of OnSessionInitRequest.
Returns:
An instance of OnSessionInitResponse."
2276,build_ring_all_reduce,tensorflow/tensorflow/python/distribute/all_reduce.py,254,function,"Construct a subgraph performing a ring-style all-reduce of input_tensors.
Args:
input_tensors: a list of `tf.Tensor` objects, which must all
have the same shape and type.
num_workers: number of worker tasks spanned by input_tensors.
num_subchunks: number of subchunks each device should process in one tick.
gpu_perm: a list of ints giving a ring-wise rank ordering of GPUs at
each worker. All workers must have the same number of
GPUs with the same rank ordering. If NVLINK is available, this should
be a ring order supported by NVLINK edges.
red_op: a binary operator for elementwise reduction.
un_op: an optional unary operator to apply to fully reduced values.
Raises:
ValueError: empty input_tensors or they don't all have same
size.
Returns:
a list of `tf.Tensor` identical sum-reductions of input_tensors."
2277,build_recursive_hd_all_reduce,tensorflow/tensorflow/python/distribute/all_reduce.py,426,function,"Construct a subgraph for recursive halving-doubling all-reduce.
The recursive halving-doubling algorithm is described in
(Thakur et al., 2015).
The concept is to arrange the participating n devices in
a linear sequence where devices exchange data pairwise
with one other device in each round. During the gather
phase there are lg(n) rounds where devices exchange
increasingly smaller sub-tensors with another device
at increasingly greater distances, until at the top
each device has 1/n of the fully reduced values. During the
scatter phase each device exchanges its fully reduced
sub-tensor (which doubles in length at each round)
with one other device at increasingly smaller distances
until each device has all of the fully reduced values.
Note: this preliminary version requires that len(input_tensors) be a
power of 2. TODO(tucker): relax this restriction. Also, the
number of elements in each tensor must be divisible by 2^h where h
is the number of hops in each phase. This will also be relaxed in
the future with edge-case specific logic.
Args:
input_tensors: list of `tf.Tensor` to be elementwise reduced.
red_op: a binary elementwise reduction Op.
un_op: an optional unary elementwise Op to apply to reduced values.
Returns:
list of `tf.Tensor` which are the fully reduced tensors, one
at each device of input_tensors.
Raises:
ValueError: num_devices not a power of 2, or tensor len not divisible
by 2 the proper number of times.
References:
Optimization of Collective Communication Operations in MPICH:
[Thakur et al., 2005]
(https://journals.sagepub.com/doi/abs/10.1177/1094342005051521)
([pdf](http://wwwi10.lrr.in.tum.de/~gerndt/home/Teaching/HPCSeminar/mpich_multi_coll.pdf))"
2278,build_shuffle_all_reduce,tensorflow/tensorflow/python/distribute/all_reduce.py,558,function,"Construct a subgraph for shuffle all-reduce.
Shuffle reduce is essentially the algorithm implemented when using
parameter servers. Suppose tensor length is n, there are d devices
and g gather shards. Each device sends a n/g length sub-tensor to
each gather shard. The gather shards perform a reduction across d
fragments, then broadcast the result back to each device. The
devices then join the g fully reduced fragments they receive from
the shards. The gather shards could perform d-1 pairwise
reductions, or one d-way reduction. The first is better where
reduction Op time is low compared to transmission time, the second
better in the other case.
Args:
input_tensors: list of `tf.Tensor` values to be reduced.
gather_devices: list of names of devices on which reduction shards
should be placed.
red_op: an n-array elementwise reduction Op
un_op: optional elementwise unary Op to be applied to fully-reduced values.
Returns:
list of `tf.Tensor` which are the fully reduced tensors."
2279,build_nccl_all_reduce,tensorflow/tensorflow/python/distribute/all_reduce.py,685,function,"Build a subgraph that does one full all-reduce, using NCCL.
Args:
input_tensors: list of `tf.Tensor` of same-shape and type values to
be reduced.
red_op: binary elementwise reduction operator. Must be one of
{tf.add}
un_op: optional unary elementwise Op to apply to fully-reduce values.
Returns:
list of `tf.Tensor` of reduced values.
Raises:
ValueError: red_op not supported."
2280,build_nccl_then_ring,tensorflow/tensorflow/python/distribute/all_reduce.py,779,function,"Construct hybrid of NCCL within workers, Ring across workers."
2281,build_nccl_then_recursive_hd,tensorflow/tensorflow/python/distribute/all_reduce.py,788,function,"Construct hybrid of NCCL within workers, Recursive-HD across workers."
2282,build_nccl_then_shuffle,tensorflow/tensorflow/python/distribute/all_reduce.py,794,function,"Construct hybrid of NCCL within workers, Shuffle across workers."
2283,build_shuffle_then_ring,tensorflow/tensorflow/python/distribute/all_reduce.py,845,function,"Construct hybrid of Shuffle within workers, Ring across workers."
2284,build_shuffle_then_shuffle,tensorflow/tensorflow/python/distribute/all_reduce.py,857,function,"Construct hybrid of Shuffle within workers, Shuffle across workers."
2285,CentralStorageStrategy,tensorflow/tensorflow/python/distribute/central_storage_strategy.py,28,class,"A one-machine strategy that puts all variables on a single device.
Variables are assigned to local CPU or the only GPU. If there is more
than one GPU, compute operations (other than variable update operations)
will be replicated across all GPUs.
For Example:
```
strategy = tf.distribute.experimental.CentralStorageStrategy()
# Create a dataset
ds = tf.data.Dataset.range(5).batch(2)
# Distribute that dataset
dist_dataset = strategy.experimental_distribute_dataset(ds)
with strategy.scope():
@tf.function
def train_step(val):
return val + 1
# Iterate over the distributed dataset
for x in dist_dataset:
# process dataset elements
strategy.run(train_step, args=(x,))
```"
2286,experimental_distribute_dataset,tensorflow/tensorflow/python/distribute/central_storage_strategy.py,77,method,"Distributes a tf.data.Dataset instance provided via dataset.
The returned dataset is a wrapped strategy dataset which creates a
multidevice iterator under the hood. It prefetches the input data to the
specified devices on the worker. The returned distributed dataset can be
iterated over similar to how regular datasets can.
NOTE: Currently, the user cannot add any more transformations to a
distributed dataset.
For Example:
```
strategy = tf.distribute.CentralStorageStrategy() # with 1 CPU and 1 GPU
dataset = tf.data.Dataset.range(10).batch(2)
dist_dataset = strategy.experimental_distribute_dataset(dataset)
for x in dist_dataset:
print(x) # Prints PerReplica values [0, 1], [2, 3],...
```
Args:
dataset: `tf.data.Dataset` to be prefetched to device.
options: `tf.distribute.InputOptions` used to control options on how this
dataset is distributed.
Returns:
A ""distributed `Dataset`"" that the caller can iterate over."
2287,experimental_distribute_datasets_from_function,tensorflow/tensorflow/python/distribute/central_storage_strategy.py,108,method,"Distributes `tf.data.Dataset` instances created by calls to `dataset_fn`.
`dataset_fn` will be called once for each worker in the strategy. In this
case, we only have one worker so `dataset_fn` is called once. Each replica
on this worker will then dequeue a batch of elements from this local
dataset.
The `dataset_fn` should take an `tf.distribute.InputContext` instance where
information about batching and input replication can be accessed.
For Example:
```
def dataset_fn(input_context):
batch_size = input_context.get_per_replica_batch_size(global_batch_size)
d = tf.data.Dataset.from_tensors([[1.]]).repeat().batch(batch_size)
return d.shard(
input_context.num_input_pipelines, input_context.input_pipeline_id)
inputs = strategy.experimental_distribute_datasets_from_function(dataset_fn)
for batch in inputs:
replica_results = strategy.run(replica_fn, args=(batch,))
```
IMPORTANT: The `tf.data.Dataset` returned by `dataset_fn` should have a
per-replica batch size, unlike `experimental_distribute_dataset`, which uses
the global batch size. This may be computed using
`input_context.get_per_replica_batch_size`.
Args:
dataset_fn: A function taking a `tf.distribute.InputContext` instance and
returning a `tf.data.Dataset`.
options: `tf.distribute.InputOptions` used to control options on how this
dataset is distributed.
Returns:
A ""distributed `Dataset`"", which the caller can iterate over like regular
datasets."
2288,experimental_local_results,tensorflow/tensorflow/python/distribute/central_storage_strategy.py,154,method,"Returns the list of all local per-replica values contained in `value`.
In `CentralStorageStrategy` there is a single worker so the value returned
will be all the values on that worker.
Args:
value: A value returned by `run()`, `extended.call_for_each_replica()`,
or a variable created in `scope`.
Returns:
A tuple of values contained in `value`. If `value` represents a single
value, this returns `(value,).`"
2289,run,tensorflow/tensorflow/python/distribute/central_storage_strategy.py,170,method,"Run `fn` on each replica, with the given arguments.
In `CentralStorageStrategy`, `fn` is called on each of the compute
replicas, with the provided ""per replica"" arguments specific to that device.
Args:
fn: The function to run. The output must be a `tf.nest` of `Tensor`s.
args: (Optional) Positional arguments to `fn`.
kwargs: (Optional) Keyword arguments to `fn`.
options: (Optional) An instance of `tf.distribute.RunOptions` specifying
the options to run `fn`.
Returns:
Return value from running `fn`."
2290,reduce,tensorflow/tensorflow/python/distribute/central_storage_strategy.py,188,method,"Reduce `value` across replicas.
Given a per-replica value returned by `run`, say a
per-example loss, the batch will be divided across all the replicas. This
function allows you to aggregate across replicas and optionally also across
batch elements. For example, if you have a global batch size of 8 and 2
replicas, values for examples `[0, 1, 2, 3]` will be on replica 0 and
`[4, 5, 6, 7]` will be on replica 1. By default, `reduce` will just
aggregate across replicas, returning `[0+4, 1+5, 2+6, 3+7]`. This is useful
when each replica is computing a scalar or some other value that doesn't
have a ""batch"" dimension (like a gradient). More often you will want to
aggregate across the global batch, which you can get by specifying the batch
dimension as the `axis`, typically `axis=0`. In this case it would return a
scalar `0+1+2+3+4+5+6+7`.
If there is a last partial batch, you will need to specify an axis so
that the resulting shape is consistent across replicas. So if the last
batch has size 6 and it is divided into [0, 1, 2, 3] and [4, 5], you
would get a shape mismatch unless you specify `axis=0`. If you specify
`tf.distribute.ReduceOp.MEAN`, using `axis=0` will use the correct
denominator of 6. Contrast this with computing `reduce_mean` to get a
scalar value on each replica and this function to average those means,
which will weigh some values `1/8` and others `1/4`.
For Example:
```
strategy = tf.distribute.experimental.CentralStorageStrategy(
compute_devices=['CPU:0', 'GPU:0'], parameter_device='CPU:0')
ds = tf.data.Dataset.range(10)
# Distribute that dataset
dist_dataset = strategy.experimental_distribute_dataset(ds)
with strategy.scope():
@tf.function
def train_step(val):
# pass through
return val
# Iterate over the distributed dataset
for x in dist_dataset:
result = strategy.run(train_step, args=(x,))
result = strategy.reduce(tf.distribute.ReduceOp.SUM, result,
axis=None).numpy()
# result: array([ 4, 6, 8, 10])
result = strategy.reduce(tf.distribute.ReduceOp.SUM, result, axis=0).numpy()
# result: 28
```
Args:
reduce_op: A `tf.distribute.ReduceOp` value specifying how values should
be combined.
value: A ""per replica"" value, e.g. returned by `run` to
be combined into a single tensor.
axis: Specifies the dimension to reduce along within each
replica's tensor. Should typically be set to the batch dimension, or
`None` to only reduce across replicas (e.g. if the tensor has no batch
dimension).
Returns:
A `Tensor`."
2291,CentralStorageStrategyV1,tensorflow/tensorflow/python/distribute/central_storage_strategy.py,256,class,
2292,CollectiveAllReduceStrategy,tensorflow/tensorflow/python/distribute/collective_all_reduce_strategy.py,48,class,"A distribution strategy for synchronous training on multiple workers.
This strategy implements synchronous distributed training across multiple
workers, each with potentially multiple GPUs. Similar to
`tf.distribute.MirroredStrategy`, it replicates all variables and computations
to each local device. The difference is that it uses a distributed collective
implementation (e.g. all-reduce), so that multiple workers can work together.
You need to launch your program on each worker and configure
`cluster_resolver` correctly. For example, if you are using
`tf.distribute.cluster_resolver.TFConfigClusterResolver`, each worker needs to
have its corresponding `task_type` and `task_id` set in the `TF_CONFIG`
environment variable.
Your program runs on each worker as-is. Note that collectives require each
worker to participate. All `tf.distribute` and non `tf.distribute` API may use
collectives internally, e.g. checkpointing and saving since reading a
`tf.Variable` with `tf.VariableSynchronization.ON_READ` all-reduces the value.
Therefore it's recommended to run exactly the same program on each worker.
Dispatching based on `task_type` or `task_id` of the worker is error-prone.
`cluster_resolver.num_accelerators()` determines the number of GPUs the
strategy uses. If it's zero, the strategy uses the CPU. All workers need to
use the same number of devices, otherwise the behavior is undefined.
This strategy is not intended for TPU. Use
`tf.distribute.experimental.TPUStrategy` instead.
__Saving__
You need to save and checkpoint on all workers instead of just one. This is
because variables whose synchronization=ON_READ triggers aggregation during
saving. It's recommended to save to a different path on each worker to avoid
race conditions. Each worker saves the same thing. See
[Multi-worker training with Keras](https://www.tensorflow.org/tutorials/distribute/multi_worker_with_keras#model_saving_and_loading)
tutorial for examples.
__Known Issues__
* `tf.distribute.cluster_resolver.TFConfigClusterResolver` does not return the
correct number of accelerators. The strategy uses all available GPUs if
`cluster_resolver` is `tf.distribute.cluster_resolver.TFConfigClusterResolver`
or `None`.
* In eager mode, the strategy needs to be created before calling any other
Tensorflow API."
2293,cluster_resolver,tensorflow/tensorflow/python/distribute/collective_all_reduce_strategy.py,140,method,"Returns the cluster resolver associated with this strategy.
As a multi-worker strategy,
`tf.distribute.experimental.MultiWorkerMirroredStrategy` provides the
associated `tf.distribute.cluster_resolver.ClusterResolver`. If the user
provides one in `__init__`, that instance is returned; if the user does
not, a default `TFConfigClusterResolver` is provided."
2294,CollectiveAllReduceStrategyV1,tensorflow/tensorflow/python/distribute/collective_all_reduce_strategy.py,153,class,
2295,CollectiveAllReduceExtended,tensorflow/tensorflow/python/distribute/collective_all_reduce_strategy.py,176,class,Implementation of CollectiveAllReduceStrategy.
2296,experimental_between_graph,tensorflow/tensorflow/python/distribute/collective_all_reduce_strategy.py,622,method,
2297,experimental_should_init,tensorflow/tensorflow/python/distribute/collective_all_reduce_strategy.py,626,method,
2298,should_checkpoint,tensorflow/tensorflow/python/distribute/collective_all_reduce_strategy.py,630,method,
2299,should_save_summary,tensorflow/tensorflow/python/distribute/collective_all_reduce_strategy.py,634,method,
2300,initial_value_fn,tensorflow/tensorflow/python/distribute/collective_all_reduce_strategy.py,403,method,
2301,LocalCollectiveAllReduceStrategy,tensorflow/tensorflow/python/distribute/collective_all_reduce_strategy_test.py,468,class,
2302,fn,tensorflow/tensorflow/python/distribute/collective_all_reduce_strategy_test.py,491,method,
2303,Hints,tensorflow/tensorflow/python/distribute/collective_util.py,25,class,"Hints for collective operations like AllReduce.
This can be passed to methods like
`tf.distribute.get_replica_context().all_reduce()` to optimize collective
operation performance. Note that these are only hints, which may or may not
change the actual behavior. Some options only apply to certain strategy and
are ignored by others.
One common optimization is to break gradients all-reduce into multiple packs
so that weight updates can overlap with gradient all-reduce.
Example:
```python
hints = tf.distribute.experimental.CollectiveHints(
bytes_per_pack=50 * 1024 * 1024)
grads = tf.distribute.get_replica_context().all_reduce(
'sum', grads, experimental_hints=hints)
optimizer.apply_gradients(zip(grads, vars),
experimental_aggregate_gradients=False)
```"
2304,DistributionParameter,tensorflow/tensorflow/python/distribute/combinations.py,53,class,"Transforms arguments of type `NamedDistribution`.
Convert all arguments of type `NamedDistribution` to the value of their
`strategy` property."
2305,modified_arguments,tensorflow/tensorflow/python/distribute/combinations.py,60,method,
2306,ClusterParameters,tensorflow/tensorflow/python/distribute/combinations.py,69,class,"Adds cluster parameters if a `NamedDistribution` has it.
It needs to be before DistributionParameter."
2307,modified_arguments,tensorflow/tensorflow/python/distribute/combinations.py,75,method,
2308,NamedGPUCombination,tensorflow/tensorflow/python/distribute/combinations.py,96,class,"Enable tests to request GPU hardware and skip non-GPU combinations.
This class expects test_combinations to be generated with `NamedDistribution`
wrapping instances of `tf.distribute.Strategy`.
Optionally, the `required_gpus` argument is supported. GPU hardware is
required, if its value is `True` or > 0.
Attributes:
GPU_TEST: The environment is considered to have GPU hardware available if
the name of the program contains ""test_gpu"" or ""test_xla_gpu""."
2309,should_execute_combination,tensorflow/tensorflow/python/distribute/combinations.py,112,method,
2310,parameter_modifiers,tensorflow/tensorflow/python/distribute/combinations.py,134,method,
2311,GPUCombination,tensorflow/tensorflow/python/distribute/combinations.py,138,class,NamedGPUCombination that passes `tf.distribute.Strategy` to the tests.
2312,parameter_modifiers,tensorflow/tensorflow/python/distribute/combinations.py,141,method,
2313,NamedTPUCombination,tensorflow/tensorflow/python/distribute/combinations.py,148,class,"Allow to request TPU hardware and skip non-TPU combinations.
This class expects test_combinations to be generated with `NamedDistribution`
wrapping instances of `tf.distribute.Strategy`.
Optionally, the `required_tpus` parameter is supported. TPU hardware is
required, if its argument is `True` or > 0.
Optionally, the `use_cloud_tpu` parameter is supported. If TPU hardware is
required by `required_tpus`, it specifically must be a Cloud TPU (specified
with `--tpu`) if `use_cloud_tpu` is `True`.
Attributes:
TPU_TEST: The environment is considered to have TPU hardware available if
the name of the program contains ""test_tpu""."
2314,should_execute_combination,tensorflow/tensorflow/python/distribute/combinations.py,168,method,
2315,parameter_modifiers,tensorflow/tensorflow/python/distribute/combinations.py,202,method,
2316,TPUCombination,tensorflow/tensorflow/python/distribute/combinations.py,210,class,NamedTPUCombination that passes `tf.distribute.Strategy` to the tests.
2317,parameter_modifiers,tensorflow/tensorflow/python/distribute/combinations.py,213,method,
2318,NamedDistribution,tensorflow/tensorflow/python/distribute/combinations.py,220,class,Wraps a `tf.distribute.Strategy` and adds a name for test titles.
2319,runner,tensorflow/tensorflow/python/distribute/combinations.py,268,method,
2320,strategy,tensorflow/tensorflow/python/distribute/combinations.py,272,method,
2321,concat,tensorflow/tensorflow/python/distribute/combinations.py,279,function,Concats combinations.
2322,generate,tensorflow/tensorflow/python/distribute/combinations.py,287,function,"Distributed adapter of `framework.combinations_lib.generate`.
All tests with distributed strategy should use this one instead of
`framework.test_combinations.generate`. This function has support of strategy
combinations, GPU/TPU and multi worker support.
See `framework.test_combinations_lib.generate` for usage."
2323,check_destinations,tensorflow/tensorflow/python/distribute/cross_device_ops.py,49,function,"Checks whether `destinations` is not empty.
Args:
destinations: a `DistributedValues`, variable, or string object.
Returns:
Boolean which is True if `destinations` is not empty."
2324,validate_destinations,tensorflow/tensorflow/python/distribute/cross_device_ops.py,65,function,Validates the `destination` is one of expected types.
2325,reduce_non_distributed_value,tensorflow/tensorflow/python/distribute/cross_device_ops.py,79,function,Reduce a non-DistributedValue `value` to `destinations`.
2326,get_devices_from,tensorflow/tensorflow/python/distribute/cross_device_ops.py,159,function,
2327,simple_broadcast,tensorflow/tensorflow/python/distribute/cross_device_ops.py,181,function,Broadcast `value` to `destinations` using simple copies.
2328,CrossDeviceOps,tensorflow/tensorflow/python/distribute/cross_device_ops.py,217,class,Base class for cross-device reduction and broadcasting algorithms.
2329,reduce,tensorflow/tensorflow/python/distribute/cross_device_ops.py,228,method,"Reduce `per_replica_value` to `destinations`.
It runs the reduction operation defined by `reduce_op` and put the
result on `destinations`.
Args:
reduce_op: An instance of `tf.distribute.ReduceOp` that indicates how
per_replica_value will be reduced.
per_replica_value: A `tf.distribute.DistributedValues` object or a tensor
with device set.
destinations: the reduction destinations.
experimental_hints: A `tf.distrbute.experimental.CollectiveHints`. Hints
to perform collective operations.
Returns:
a Mirrored object.
Raises:
ValueError: if per_replica_value can't be converted to a PerReplica
object or if destinations aren't strings, Variables or DistributedValues"
2330,batch_reduce,tensorflow/tensorflow/python/distribute/cross_device_ops.py,272,method,"Reduce PerReplica objects in a batch.
Reduce each first element in `value_destination_pairs` to each second
element which indicates the destinations.
This can be faster than multiple individual `reduce`s because we can
fuse several tensors into one or multiple packs before reduction.
Args:
reduce_op: An instance of `tf.distribute.ReduceOp` that indicates how the
`per_replica_value` will be reduced.
value_destination_pairs: A list or a tuple of PerReplica objects (or
tensors with device set if there is one device) and destinations.
experimental_hints: A `tf.distrbute.experimental.CollectiveHints`. Hints
to perform collective operations.
Returns:
a list of Mirrored objects.
Raises:
ValueError: if `value_destination_pairs` is not an iterable of
tuples of PerReplica objects and destinations."
2331,broadcast,tensorflow/tensorflow/python/distribute/cross_device_ops.py,324,method,"Broadcast the `tensor` to destinations.
Args:
tensor: the tensor to broadcast.
destinations: the broadcast destinations.
Returns:
a Mirrored object."
2332,reduce_implementation,tensorflow/tensorflow/python/distribute/cross_device_ops.py,338,method,"The implementation of reduce of `per_replica_value` to `destinations`.
Overriding this method is useful for subclass implementers.
It runs the reduction operation defined by `reduce_op` and put the
result on `destinations`.
Args:
reduce_op: An instance `tf.distribute.ReduceOp` that indicates of how
per_replica_value will be reduced.
per_replica_value: A PerReplica object or a tensor with device set.
destinations: the reduction destinations.
experimental_hints: A `tf.distrbute.experimental.CollectiveHints`. Hints
to perform collective operations.
Returns:
a Mirrored object.
Raises:
ValueError: if per_replica_value can't be converted to a PerReplica
object."
2333,batch_reduce_implementation,tensorflow/tensorflow/python/distribute/cross_device_ops.py,366,method,"Implementation of reduce PerReplica objects in a batch.
Overriding this method is useful for subclass implementers.
Reduce each first element in `value_destination_pairs` to each second
element which indicates the destinations.
Args:
reduce_op: An instance of `tf.distribute.ReduceOp` that indicates how
per_replica_value will be reduced.
value_destination_pairs: An iterable of tuples of PerReplica objects
(or tensors with device set if there is one device) and destinations.
experimental_hints: A `tf.distrbute.experimental.CollectiveHints`. Hints
to perform collective operations.
Returns:
a list of Mirrored objects.
Raises:
ValueError: if `value_destination_pairs` is not an iterable of
tuples of PerReplica objects and destinations"
2334,broadcast_implementation,tensorflow/tensorflow/python/distribute/cross_device_ops.py,395,method,"Implementation of broadcast the `tensor` to destinations.
Args:
tensor: the tensor to broadcast.
destinations: the broadcast destinations.
Returns:
a Mirrored object."
2335,ReductionToOneDevice,tensorflow/tensorflow/python/distribute/cross_device_ops.py,409,class,"Always do reduction to one device first and then do broadcasting.
Batch reduction is done by reduction on each element one by one.
```
mirrored_strategy = tf.distribute.MirroredStrategy(
cross_device_ops=tf.distribute.ReductionToOneDevice())
```"
2336,reduce_implementation,tensorflow/tensorflow/python/distribute/cross_device_ops.py,433,method,
2337,batch_reduce_implementation,tensorflow/tensorflow/python/distribute/cross_device_ops.py,448,method,
2338,AllReduceCrossDeviceOps,tensorflow/tensorflow/python/distribute/cross_device_ops.py,642,class,Reduction using all-reduce.
2339,reduce_implementation,tensorflow/tensorflow/python/distribute/cross_device_ops.py,661,method,
2340,batch_reduce_implementation,tensorflow/tensorflow/python/distribute/cross_device_ops.py,670,method,
2341,NcclAllReduce,tensorflow/tensorflow/python/distribute/cross_device_ops.py,747,class,Reduction using NCCL all-reduce.
2342,HierarchicalCopyAllReduce,tensorflow/tensorflow/python/distribute/cross_device_ops.py,773,class,"Reduction using hierarchical copy all-reduce.
It reduces to one GPU along edges in some hierarchy and broadcasts back to
each GPU along the same path. Before performing all-reduce, tensors will be
repacked or aggregated for more efficient cross-device transportation.
This is a reduction created for Nvidia DGX-1 which assumes GPUs connects like
that on DGX-1 machine. If you have different GPU inter-connections, it is
likely that it would be slower than `tf.distribute.ReductionToOneDevice`."
2343,MultiWorkerAllReduce,tensorflow/tensorflow/python/distribute/cross_device_ops.py,804,class,All-reduce algorithms for distributed TensorFlow.
2344,validate_and_complete_spec,tensorflow/tensorflow/python/distribute/cross_device_ops.py,843,method,Validate and complete the all-reduce spec.
2345,CollectiveCommunication,tensorflow/tensorflow/python/distribute/cross_device_ops.py,916,class,"Communication choices for CollectiveOps.
* `AUTO`: Default to runtime's automatic choices.
* `RING`: TensorFlow's ring algorithms for all-reduce and
all-gather.
* `NCCL`: Use ncclAllReduce for all-reduce, and ring algorithms for
all-gather."
2346,CollectiveAllReduce,tensorflow/tensorflow/python/distribute/cross_device_ops.py,932,class,"All-reduce cross device ops using collective ops.
In the between-graph replicated training, it will still do all-reduces across
all workers and then put results on the right destinations."
2347,reduce_implementation,tensorflow/tensorflow/python/distribute/cross_device_ops.py,981,method,
2348,batch_reduce_implementation,tensorflow/tensorflow/python/distribute/cross_device_ops.py,1012,method,
2349,choose_the_best,tensorflow/tensorflow/python/distribute/cross_device_ops.py,1167,function,"Find the best CrossDeviceOps locally given a `tf.compat.v1.ConfigProto`.
Args:
devices: a list of devices passed to `tf.distribute.Strategy`.
session_config: a `tf.compat.v1.ConfigProto` or `None`. If `None`, it will
make decision based on all logical devices.
Returns:
A subclass of `CrossDeviceOps`."
2350,aggregate_gradients_using_nccl,tensorflow/tensorflow/python/distribute/cross_device_utils.py,42,function,Aggregate gradients using nccl allreduce.
2351,aggregate_gradients_using_hierarchical_copy,tensorflow/tensorflow/python/distribute/cross_device_utils.py,56,function,"Aggregate gradients using hierarchical copies.
Args:
avail_devices: available GPU devices.
replica_grads: List of lists of (gradient, variable) tuples. The outer list
is over replicas. The inner list is over individual gradients.
Returns:
The list of (aggregated_gradient, variable), where the gradient has been
summed across all replicas and the variable is chosen from the first
replica."
2352,aggregate_single_gradient_using_copy,tensorflow/tensorflow/python/distribute/cross_device_utils.py,138,function,"Calculate the average gradient for a shared variable across all replicas.
Note that this function provides a synchronization point across all replicas.
Args:
grad_and_vars: A list or tuple of (gradient, variable) tuples. Each
(gradient, variable) pair within the outer list represents the gradient
of the variable calculated for a single replica, and the number of pairs
equals the number of replicas.
use_mean: if True, mean is taken, else sum of gradients is taken.
check_inf_nan: check grads for nans and infs.
Returns:
The tuple ([(average_gradient, variable),], has_nan_or_inf) where the
gradient has been averaged across all replicas. The variable is chosen
from the first replica. The has_nan_or_inf indicates the grads has nan or
inf."
2353,group_device_names,tensorflow/tensorflow/python/distribute/cross_device_utils.py,173,function,"Group device names into groups of group_size.
Args:
devices: a list of canonical device strings.
group_size: integer which is equal to or greater than 1.
Returns:
list of lists of devices, where each inner list is group_size long,
and each device appears at least once in an inner list. If
len(devices) % group_size == 0 then each device will appear exactly once.
Raises:
ValueError: if group_size > len(devices)"
2354,split_grads_by_size,tensorflow/tensorflow/python/distribute/cross_device_utils.py,200,function,"Break gradients into two sets according to tensor size.
Args:
threshold_size: int size cutoff for small vs large tensor.
device_grads: List of lists of (gradient, variable) tuples. The outer
list is over devices. The inner list is over individual gradients.
Returns:
small_grads: Subset of device_grads where shape is <= threshold_size
elements.
large_grads: Subset of device_grads where shape is > threshold_size
elements."
2355,CollectiveKeys,tensorflow/tensorflow/python/distribute/cross_device_utils.py,241,class,"Class that manages collective keys.
We need to manage three different keys for collective:
*Group key*: an integer key to identify the set of cooperative devices.
Collective ops work under the same set of devices must using the same group
key.
*Instance key*: an integer key to identify the set of same counterpart of
tensors on different devices in a device group that need to be all-reduced.
""Graph key"": an integer key that is unique key graph. This is used to support
multiple graphs per client session. It must be non-zero and set in the
`config` argument of each call to `session.run`."
2356,get_group_key,tensorflow/tensorflow/python/distribute/cross_device_utils.py,284,method,"Returns a group key for the set of devices.
Args:
devices: list of strings naming devices in a collective group.
Returns:
int key uniquely identifying the set of device names."
2357,get_op_instance_key,tensorflow/tensorflow/python/distribute/cross_device_utils.py,308,method,Returns a new instance key for use in defining a collective op.
2358,get_variable_instance_key,tensorflow/tensorflow/python/distribute/cross_device_utils.py,314,method,Returns a new instance key for use in creating a Variable.
2359,build_collective_reduce,tensorflow/tensorflow/python/distribute/cross_device_utils.py,321,function,"Build a subgraph that does one full all-reduce, using the collective Op.
If called in eager mode, it's required to supply a list of async executors for
each input Tensor.
Args:
input_tensors: tensors within a single worker graph that are to be reduced
together; must be one per device.
devices: a list of device strings to run the collective on.
group_size: total number of devices globally that will be doing this same
reduction. The reduction will actually include the corresponding tensors
at all these workers.
collective_keys: a CollectiveKeys object.
reduction_op: string naming the reduction op.
unary_op: string naming the unary final op.
communication_hint: string providing hint to runtime for choosing collective
implementation.
control_inputs: if not None, add control edges between control_inputs and
(index-wise) corresponding collective_reduce tensors
executors: a list of async executor. Required for eager execution.
Returns:
An array of final tensors, one per device, computed by the full reduction.
Raises:
ValueError: There must be at least two tensors over all the workers."
2360,build_collective_gather,tensorflow/tensorflow/python/distribute/cross_device_utils.py,391,function,"Build a subgraph that does one full all-gather, using the collective Op.
This method must be called in graph mode or inside a tf.function.
Args:
input_tensors: tensors within a single worker graph that are to be gathered
together; must be one per device.
devices: a list of device strings to run the collective on.
group_size: total number of devices globally that will be doing this same
gathering. The gathering will actually include the corresponding tensors
at all these workers.
collective_keys: a CollectiveKeys object.
communication_hint: string providing hint to runtime for choosing collective
implementation.
control_inputs: if not None, add control edges between control_inputs and
(index-wise) corresponding collective_gather tensors
Returns:
An array of final tensors, one per device, computed by the full gather."
2361,build_collective_gather_indexed_slices,tensorflow/tensorflow/python/distribute/cross_device_utils.py,442,function,"Build a subgraph that all-gathers IndexedSlices using the collective Op.
This method must be called in graph mode or inside a tf.function.
Args:
input_slices_list: a list of IndexedSlices within a single worker graph that
are to be gathered together; must be one per device.
devices: a list of device strings to run the collective on.
group_size: total number of devices globally that will be doing this same
gathering. The gathering will actually include the corresponding tensors
at all these workers.
collective_keys: a CollectiveKeys object.
communication_hint: string providing hint to runtime for choosing collective
implementation.
control_inputs: if not None, add control edges between control_inputs and
(index-wise) corresponding collective_reduce tensors
Returns:
An array of final IndexedSlices, one per device, computed by the full
gather.
Raises:
ValueError: if control_inputs is not None and doesn't match the length and
devices of inputs."
2362,sum_grad_and_var_all_reduce,tensorflow/tensorflow/python/distribute/cross_device_utils.py,549,function,Apply all-reduce algorithm over specified gradient tensors.
2363,sum_gradients_all_reduce,tensorflow/tensorflow/python/distribute/cross_device_utils.py,590,function,"Apply all-reduce algorithm over specified gradient tensors.
Args:
dev_prefixes: list of prefix strings to use to generate PS device names.
replica_grads: the gradients to reduce.
num_workers: number of worker processes across entire job.
alg: the all-reduce algorithm to apply.
num_shards: alg-specific sharding factor.
gpu_indices: indices of local GPUs in order usable for ring-reduce.
Returns:
list of reduced tensors"
2364,extract_ranges,tensorflow/tensorflow/python/distribute/cross_device_utils.py,632,function,"Extract consecutive ranges and singles from index_list.
Args:
index_list: List of monotone increasing non-negative integers.
range_size_limit: Largest size range to return. If a larger
consecutive range exists, it will be returned as multiple
ranges.
Returns:
(ranges, singles) where ranges is a list of [first, last] pairs of
consecutive elements in index_list, and singles is all of the
other elements, in original order."
2365,pack_range,tensorflow/tensorflow/python/distribute/cross_device_utils.py,672,function,"Form the concatenation of a specified range of gradient tensors.
Args:
key: Value under which to store meta-data in packing that will be used
later to restore the grad_var list structure.
packing: Dict holding data describing packed ranges of small tensors.
grad_vars: List of (grad, var) pairs for one replica.
rng: A pair of integers giving the first, last indices of a consecutive
range of tensors to be packed.
Returns:
A tensor that is the concatenation of all the specified small tensors."
2366,unpack_grad_tuple,tensorflow/tensorflow/python/distribute/cross_device_utils.py,704,function,"Unpack a previously packed collection of gradient tensors.
Args:
gv: A (grad, var) pair to be unpacked.
gpt: A GradPackTuple describing the packing operation that produced gv.
Returns:
A list of (grad, var) pairs corresponding to the values that were
originally packed into gv, maybe following subsequent operations like
reduction."
2367,pack_small_tensors,tensorflow/tensorflow/python/distribute/cross_device_utils.py,727,function,"Concatenate small gradient tensors together for reduction.
Args:
replica_grads: List of lists of (gradient, variable) tuples.
max_bytes: Int giving max number of bytes in a tensor that
may be considered small.
max_group: Int giving max number of small tensors that may be
concatenated into one new tensor.
Returns:
new_replica_grads, packing where new_replica_grads is identical to
replica_grads except that all feasible small_tensors have been removed
from their places and concatenated into larger tensors that are
now in the front of the list for each replica, and packing contains
the data necessary to restore the replica_grads structure.
Look through the first replica for gradients of the same type (float),
and small size, that are all sequential. For each such group,
replace by a new tensor that is a flattened concatenation. Note
that the corresponding variable will be absent, which doesn't matter
because it isn't used during all-reduce.
Requires:
Every gv_list in replicas must have isomorphic structure including identical
tensor sizes and types."
2368,unpack_small_tensors,tensorflow/tensorflow/python/distribute/cross_device_utils.py,783,function,"Undo the structure alterations to replica_grads done by pack_small_tensors.
Args:
replica_grads: List of List of (grad, var) tuples.
packing: A dict generated by pack_small_tensors describing the changes
it made to replica_grads.
Returns:
new_replica_grads: identical to replica_grads except that concatenations
of small tensors have been split apart and returned to their original
positions, paired with their original variables."
2369,aggregate_tensors_or_indexed_slices,tensorflow/tensorflow/python/distribute/cross_device_utils.py,815,function,Aggregate tensors using `accumulation_fn` and IndexedSlices via concat.
2370,divide_by_n_tensors_or_indexed_slices,tensorflow/tensorflow/python/distribute/cross_device_utils.py,823,function,
2371,copy_tensor_or_indexed_slices_to_device,tensorflow/tensorflow/python/distribute/cross_device_utils.py,832,function,
2372,contains_indexed_slices,tensorflow/tensorflow/python/distribute/cross_device_utils.py,844,function,Check whether the value is `IndexedSlices` or contains `IndexedSlices`.
2373,is_indexed_slices,tensorflow/tensorflow/python/distribute/cross_device_utils.py,856,function,
2374,split_by_sparsity,tensorflow/tensorflow/python/distribute/cross_device_utils.py,863,function,"Split values into dense and sparse values.
Args:
values: a list of tensors or `PerReplica`s.
Returns:
Four lists:
a list of dense values, a list of their indices in `values` and
a list of sparse values, a list of their indices in `values`."
2375,stitch_values,tensorflow/tensorflow/python/distribute/cross_device_utils.py,888,function,"Stitch values together according to their indices.
Args:
values_and_indices_list: a list of tuples of values and indices indicating
the values and positions in the returned list.
Returns:
a stitched list of values."
2376,per_replica_num_elements,tensorflow/tensorflow/python/distribute/cross_device_utils.py,911,function,"Returns the static number of elements of one replica.
Args:
per_replica: A PerReplica of Tensor or IndexedSlices.
Returns:
Number of elements. None if some replica has a different or unknown shape."
2377,pack_by_size,tensorflow/tensorflow/python/distribute/cross_device_utils.py,930,function,"Packs `per_replica_list` into chunks of `bytes_per_pack`.
The method preserves the original order of `per_replica_list`. The packing is
best effort, each pack could have more or less bytes than `bytes_per_pack`.
It only packs values with known shape. Note that, the usage is different from
`cross_device_ops._pack_tensors`, this function is intended to work with the
ScopeAllocator style batching used in `CollectiveAllReduce`.
Args:
per_replica_list: A list of PerReplica.
bytes_per_pack: Bytes per pack.
Returns:
A list of packs of PerReplica. All values are packed into one pack if
`bytes_per_pack` is zero or any of the value has unknown shape."
2378,get_dataset_from_tensor_slices,tensorflow/tensorflow/python/distribute/custom_training_loop_gradient_test.py,34,function,
2379,AssertFlattenedMixin,tensorflow/tensorflow/python/distribute/custom_training_loop_gradient_test.py,42,class,Mixin for specialized asserts.
2380,assert_equal_flattened,tensorflow/tensorflow/python/distribute/custom_training_loop_gradient_test.py,45,method,"Asserts that flattened results are equal.
Due to the number of replicas in the strategy, the output may have a
different structure and needs to be flattened for comparison.
Args:
expected_results: The results expected as a result of a computation.
actual_results: The actual results of a computation."
2381,get_dataset_from_tensor_slices,tensorflow/tensorflow/python/distribute/custom_training_loop_input_test.py,43,function,
2382,AssertFlattenedMixin,tensorflow/tensorflow/python/distribute/custom_training_loop_input_test.py,51,class,Mixin for specialized asserts.
2383,assert_equal_flattened,tensorflow/tensorflow/python/distribute/custom_training_loop_input_test.py,54,method,"Asserts that flattened results are equal.
Due to the number of replicas in the strategy, the output may have a
different structure and needs to be flattened for comparison.
Args:
expected_results: The results expected as a result of a computation.
actual_results: The actual results of a computation."
2384,canonicalize,tensorflow/tensorflow/python/distribute/device_util.py,27,function,"Canonicalize device string.
If d has missing components, the rest would be deduced from the `default`
argument or from '/replica:0/task:0/device:CPU:0'. For example:
If d = '/cpu:0', default='/job:worker/task:1', it returns
'/job:worker/replica:0/task:1/device:CPU:0'.
If d = '/cpu:0', default='/job:worker', it returns
'/job:worker/replica:0/task:0/device:CPU:0'.
If d = '/gpu:0', default=None, it returns
'/replica:0/task:0/device:GPU:0'.
Note: This uses ""job:localhost"" as the default if executing eagerly.
Args:
d: a device string or tf.config.LogicalDevice
default: a string for default device if d doesn't have all components.
Returns:
a canonicalized device string."
2385,resolve,tensorflow/tensorflow/python/distribute/device_util.py,79,function,Canonicalize `d` with current device as default.
2386,current,tensorflow/tensorflow/python/distribute/device_util.py,110,function,Return a string (not canonicalized) for the current device.
2387,get_host_for_device,tensorflow/tensorflow/python/distribute/device_util.py,122,function,Returns the corresponding host device for the given device.
2388,local_devices_from_num_gpus,tensorflow/tensorflow/python/distribute/device_util.py,130,function,Returns device strings for local GPUs or CPU.
2389,DistributeConfig,tensorflow/tensorflow/python/distribute/distribute_config.py,24,class,"A config tuple for distribution strategies.
Attributes:
train_distribute: a `DistributionStrategy` object for training.
eval_distribute: an optional `DistributionStrategy` object for
evaluation.
remote_cluster: a dict, `ClusterDef` or `ClusterSpec` object specifying
the cluster configurations. If this is given, the `train_and_evaluate`
method will be running as a standalone client which connects to the
cluster for training."
2390,CoordinatorMode,tensorflow/tensorflow/python/distribute/distribute_coordinator.py,50,class,Specify how distribute coordinator runs.
2391,run_standard_tensorflow_server,tensorflow/tensorflow/python/distribute/distribute_coordinator.py,555,function,"Starts a standard TensorFlow server.
This method parses configurations from ""TF_CONFIG"" environment variable and
starts a TensorFlow server. The ""TF_CONFIG"" is typically a json string and
must have information of the cluster and the role of the server in the
cluster. One example is:
TF_CONFIG='{
""cluster"": {
""worker"": [""host1:2222"", ""host2:2222"", ""host3:2222""],
""ps"": [""host4:2222"", ""host5:2222""]
},
""task"": {""type"": ""worker"", ""index"": 1}
}'
This ""TF_CONFIG"" specifies there are 3 workers and 2 ps tasks in the cluster
and the current role is worker 1.
Valid task types are ""chief"", ""worker"", ""ps"" and ""evaluator"" and you can have
at most one ""chief"" and at most one ""evaluator"".
An optional key-value can be specified is ""rpc_layer"". The default value is
""grpc"".
Args:
session_config: an optional `tf.compat.v1.ConfigProto` object. Users can
pass in the session config object to configure server-local devices.
Returns:
a `tf.distribute.Server` object which has already been started.
Raises:
ValueError: if the ""TF_CONFIG"" environment is not complete."
2392,run_distribute_coordinator,tensorflow/tensorflow/python/distribute/distribute_coordinator.py,631,function,"Runs the coordinator for distributed TensorFlow.
This function runs a split coordinator for distributed TensorFlow in its
default mode, i.e the STANDALONE_CLIENT mode. Given a `cluster_spec`
specifying server addresses and their roles in a cluster, this coordinator
will figure out how to set them up, give the underlying function the right
targets for master sessions via a scope object and coordinate their training.
The cluster consisting of standard servers needs to be brought up either with
the standard server binary or with a binary running distribute coordinator
with `task_type` set to non-client type which will then turn into standard
servers.
In addition to be the distribute coordinator, this is also the source of
configurations for each job in the distributed training. As there are multiple
ways to configure a distributed TensorFlow cluster, its context object
provides these configurations so that users or higher-level APIs don't have to
figure out the configuration for each job by themselves.
In the between-graph replicated training, this coordinator will create
multiple threads and each calls the `worker_fn` which is supposed to create
its own graph and connect to one worker master given by its context object. In
the in-graph replicated training, it has only one thread calling this
`worker_fn`.
Another mode is the INDEPENDENT_WORKER mode where each server runs a
distribute coordinator which will start a standard server and optionally runs
`worker_fn` depending whether it is between-graph training or in-graph
replicated training.
The `strategy` object is expected to be a DistributionStrategy object which
has implemented methods needed by distributed coordinator such as
`configure(session_config, cluster_spec, task_type, task_id)` which configures
the strategy object for a specific task and `experimental_should_init`
property which instructs the distribute coordinator whether to run init ops
for a task. The distribute coordinator will make a copy of the `strategy`
object, call its `configure` method and pass it to `worker_fn` as an argument.
The `worker_fn` defines the training logic and is called under its own
worker context which can be accessed to via `get_current_worker_context`. A
worker context provides access to configurations for each task, e.g. the
task_type, task_id, master target and so on. Since `worker_fn` will be called
in a thread and possibly multiple times, caller should be careful when it
accesses global data. For example, it is unsafe to define flags in a
`worker_fn` or to define different environment variables for different
`worker_fn`s.
The `worker_fn` for the between-graph replication is defined as if there is
only one worker corresponding to the `worker_fn` and possibly ps jobs. For
example, when training with parameter servers, it assigns variables to
parameter servers and all other operations to that worker. In the in-graph
replication case, the `worker_fn` has to define operations for all worker
jobs. Using a distribution strategy can simplify the `worker_fn` by not having
to worry about the replication and device assignment of variables and
operations.
This method is intended to be invoked by high-level APIs so that users don't
have to explicitly call it to run this coordinator. For those who don't use
high-level APIs, to change a program to use this coordinator, wrap everything
in a the program after global data definitions such as commandline flag
definition into the `worker_fn` and get task-specific configurations from
the worker context.
The `cluster_spec` can be either passed by the argument or parsed from the
""TF_CONFIG"" environment variable. Example of a TF_CONFIG:
```
cluster = {'chief': ['host0:2222'],
'ps': ['host1:2222', 'host2:2222'],
'worker': ['host3:2222', 'host4:2222', 'host5:2222']}
os.environ['TF_CONFIG'] = json.dumps({'cluster': cluster})
```
If `cluster_spec` is not given in any format, it becomes local training and
this coordinator will connect to a local session.
For evaluation, if ""evaluator"" exists in the cluster_spec, a separate thread
will be created to call `eval_fn` with its `task_type` set to ""evaluator"". If
`eval_fn` is not defined, fall back to `worker_fn`. This implies that
evaluation will be done on a single machine if there is an ""evaluator"" task.
If ""evaluator"" doesn't exist in the cluster_spec, it entirely depends on the
`worker_fn` for how to do evaluation.
Args:
worker_fn: the function to be called. The function should accept a
`strategy` object and will be given access to a context object via a
context manager scope.
strategy: a DistributionStrategy object specifying whether it should
run between-graph replicated training or not, whether to run init ops,
etc. This object will also be configured given `session_config`,
`cluster_spec`, `task_type` and `task_id`.
eval_fn: optional function for ""evaluator"" task. If `eval_fn` is not passed
in but a ""evaluator"" task is found in the `cluster_spec`, the `worker_fn`
will be used for this task.
eval_strategy: optional DistributionStrategy object for ""evaluator"" task.
mode: in which mode this distribute coordinator runs.
cluster_spec: a dict, ClusterDef or ClusterSpec specifying servers and roles
in a cluster. If not set or empty, fall back to local training.
task_type: the current task type, optional if this is a client.
task_id: the current task id, optional if this is a client.
session_config: an optional `tf.compat.v1.ConfigProto` object which will be
passed to `strategy`'s `configure` method and used to create a session.
rpc_layer: optional string, the protocol for RPC, e.g. ""grpc"".
Raises:
ValueError: if `cluster_spec` is supplied but not a dict or a ClusterDef or
a ClusterSpec.
Returns:
In the client job, return the value returned by `worker_fn` if
it is in-graph replication or INDEPENDENT_WORKER mode; return None
otherwise."
2393,get_current_worker_context,tensorflow/tensorflow/python/distribute/distribute_coordinator_context.py,26,function,Returns the current task context.
2394,MockExtended,tensorflow/tensorflow/python/distribute/distribute_coordinator_test.py,85,class,
2395,MockStrategy,tensorflow/tensorflow/python/distribute/distribute_coordinator_test.py,98,class,
2396,configure,tensorflow/tensorflow/python/distribute/distribute_coordinator_test.py,108,method,
2397,MockServer,tensorflow/tensorflow/python/distribute/distribute_coordinator_test.py,141,class,
2398,start,tensorflow/tensorflow/python/distribute/distribute_coordinator_test.py,147,method,
2399,join,tensorflow/tensorflow/python/distribute/distribute_coordinator_test.py,150,method,
2400,joined,tensorflow/tensorflow/python/distribute/distribute_coordinator_test.py,155,method,
2401,started,tensorflow/tensorflow/python/distribute/distribute_coordinator_test.py,159,method,
2402,new_init,tensorflow/tensorflow/python/distribute/distribute_coordinator_test.py,954,function,
2403,get_update_replica_id,tensorflow/tensorflow/python/distribute/distribute_lib.py,242,function,Get the current device if in a `tf.distribute.Strategy.update()` call.
2404,UpdateContext,tensorflow/tensorflow/python/distribute/distribute_lib.py,250,class,Context manager when you are in `update()` or `update_non_slot()`.
2405,get_loss_reduction,tensorflow/tensorflow/python/distribute/distribute_lib.py,273,function,"`tf.distribute.ReduceOp` corresponding to the last loss reduction.
This is used to decide whether loss should be scaled in optimizer (used only
for estimator + v1 optimizer use case).
Returns:
`tf.distribute.ReduceOp` corresponding to the last loss reduction for
estimator and v1 optimizer use case. `tf.distribute.ReduceOp.SUM` otherwise."
2406,require_replica_context,tensorflow/tensorflow/python/distribute/distribute_lib.py,327,function,Verify in `replica_ctx` replica context.
2407,InputReplicationMode,tensorflow/tensorflow/python/distribute/distribute_lib.py,434,class,"Replication mode for input function.
* `PER_WORKER`: The input function will be called on each worker
independently, creating as many input pipelines as number of workers.
Replicas will dequeue from the local Dataset on their worker.
`tf.distribute.Strategy` doesn't manage any state sharing between such
separate input pipelines."
2408,InputContext,tensorflow/tensorflow/python/distribute/distribute_lib.py,447,class,"A class wrapping information needed by an input function.
This is a context class that is passed to the user's input function and
contains information about the compute replicas and input pipelines. The
number of compute replicas (in sync training) helps compute the local batch
size from the desired global batch size for each replica. The input pipeline
information can be used to return a different subset of the input in each
replica (for e.g. shard the input pipeline, use a different input
source etc)."
2409,num_replicas_in_sync,tensorflow/tensorflow/python/distribute/distribute_lib.py,480,method,Returns the number of compute replicas in sync.
2410,input_pipeline_id,tensorflow/tensorflow/python/distribute/distribute_lib.py,485,method,Returns the input pipeline ID.
2411,num_input_pipelines,tensorflow/tensorflow/python/distribute/distribute_lib.py,490,method,Returns the number of input pipelines.
2412,get_per_replica_batch_size,tensorflow/tensorflow/python/distribute/distribute_lib.py,494,method,"Returns the per-replica batch size.
Args:
global_batch_size: the global batch size which should be divisible by
`num_replicas_in_sync`.
Returns:
the per-replica batch size.
Raises:
ValueError: if `global_batch_size` not divisible by
`num_replicas_in_sync`."
2413,ValueContext,tensorflow/tensorflow/python/distribute/distribute_lib.py,520,class,"A class wrapping information needed by a distribute function.
This is a context class that is passed to the `value_fn` in
`strategy.experimental_distribute_values_from_function` and contains
information about the compute replicas. The `num_replicas_in_sync` and
`replica_id` can be used to customize the value on each replica.
Example usage:
1. Directly constructed.
>>> def value_fn(context):
... return context.replica_id_in_sync_group/context.num_replicas_in_sync
>>> context = tf.distribute.experimental.ValueContext(
... replica_id_in_sync_group=2, num_replicas_in_sync=4)
>>> per_replica_value = value_fn(context)
>>> per_replica_value
0.5
2. Passed in by `experimental_distribute_values_from_function`.
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> def value_fn(value_context):
... return value_context.num_replicas_in_sync
>>> distributed_values = (
... strategy.experimental_distribute_values_from_function(
... value_fn))
>>> local_result = strategy.experimental_local_results(distributed_values)
>>> local_result
(2, 2)"
2414,num_replicas_in_sync,tensorflow/tensorflow/python/distribute/distribute_lib.py,570,method,Returns the number of compute replicas in sync.
2415,replica_id_in_sync_group,tensorflow/tensorflow/python/distribute/distribute_lib.py,575,method,Returns the replica ID.
2416,RunOptions,tensorflow/tensorflow/python/distribute/distribute_lib.py,586,class,"Run options for `strategy.run`.
This can be used to hold some strategy specific configs.
Attributes:
experimental_enable_dynamic_batch_size: Boolean. Only applies to
TPUStrategy. Default to True. If True, TPUStrategy will enable dynamic
padder to support dynamic batch size for the inputs. Otherwise only static
shape inputs are allowed.
experimental_bucketizing_dynamic_shape: Boolean. Only applies to
TPUStrategy. Default to False. If True, TPUStrategy will automatic
bucketize inputs passed into `run` if the input shape is
dynamic. This is a performance optimization to reduce XLA recompilation,
which should not have impact on correctness."
2417,InputOptions,tensorflow/tensorflow/python/distribute/distribute_lib.py,616,class,"Run options for `experimental_distribute_dataset(s_from_function)`.
This can be used to hold some strategy specific configs.
```python
# Setup TPUStrategy
resolver = tf.distribute.cluster_resolver.TPUClusterResolver(tpu='')
tf.config.experimental_connect_to_cluster(resolver)
tf.tpu.experimental.initialize_tpu_system(resolver)
strategy = tf.distribute.TPUStrategy(resolver)
dataset = tf.data.Dataset.range(16)
distributed_dataset_on_host = (
strategy.experimental_distribute_dataset(
dataset,
tf.distribute.InputOptions(
experimental_prefetch_to_device=False)))
```
Attributes:
experimental_prefetch_to_device: Boolean. Defaults to True. If True, dataset
elements will be prefetched to accelerator device memory. When False,
dataset elements are prefetched to host device memory. Must be False when
using TPUEmbedding API."
2418,StrategyBase,tensorflow/tensorflow/python/distribute/distribute_lib.py,657,class,"A state & compute distribution policy on a list of devices.
See [the guide](https://www.tensorflow.org/guide/distributed_training)
for overview and examples. See `tf.distribute.StrategyExtended` and
[`tf.distribute`](https://www.tensorflow.org/api_docs/python/tf/distribute)
for a glossory of concepts mentioned on this page such as ""per-replica"",
_replica_, and _reduce_.
In short:
* To use it with Keras `compile`/`fit`,
[please
read](https://www.tensorflow.org/guide/distributed_training#using_tfdistributestrategy_with_keras).
* You may pass descendant of `tf.distribute.Strategy` to
`tf.estimator.RunConfig` to specify how a `tf.estimator.Estimator`
should distribute its computation. See
[guide](https://www.tensorflow.org/guide/distributed_training#using_tfdistributestrategy_with_estimator_limited_support).
* Otherwise, use `tf.distribute.Strategy.scope` to specify that a
strategy should be used when building an executing your model.
(This puts you in the ""cross-replica context"" for this strategy, which
means the strategy is put in control of things like variable placement.)
* If you are writing a custom training loop, you will need to call a few more
methods,
[see the
guide](https://www.tensorflow.org/guide/distributed_training#using_tfdistributestrategy_with_custom_training_loops):
* Start by either creating a `tf.data.Dataset` normally or using
`tf.distribute.experimental_make_numpy_dataset` to make a dataset out of
a `numpy` array.
* Use `tf.distribute.Strategy.experimental_distribute_dataset` to convert
a `tf.data.Dataset` to something that produces ""per-replica"" values.
If you want to manually specify how the dataset should be partitioned
across replicas, use
`tf.distribute.Strategy.experimental_distribute_datasets_from_function`
instead.
* Use `tf.distribute.Strategy.run` to run a function
once per replica, taking values that may be ""per-replica"" (e.g.
from a `tf.distribute.DistributedDataset` object) and returning
""per-replica"" values.
This function is executed in ""replica context"", which means each
operation is performed separately on each replica.
* Finally use a method (such as `tf.distribute.Strategy.reduce`) to
convert the resulting ""per-replica"" values into ordinary `Tensor`s.
A custom training loop can be as simple as:
```
with my_strategy.scope():
@tf.function
def distribute_train_epoch(dataset):
def replica_fn(input):
# process input and return result
return result
total_result = 0
for x in dataset:
per_replica_result = my_strategy.run(replica_fn, args=(x,))
total_result += my_strategy.reduce(tf.distribute.ReduceOp.SUM,
per_replica_result, axis=None)
return total_result
dist_dataset = my_strategy.experimental_distribute_dataset(dataset)
for _ in range(EPOCHS):
train_result = distribute_train_epoch(dist_dataset)
```
This takes an ordinary `dataset` and `replica_fn` and runs it
distributed using a particular `tf.distribute.Strategy` named
`my_strategy` above. Any variables created in `replica_fn` are created
using `my_strategy`'s policy, and library functions called by
`replica_fn` can use the `get_replica_context()` API to implement
distributed-specific behavior.
You can use the `reduce` API to aggregate results across replicas and use
this as a return value from one iteration over a
`tf.distribute.DistributedDataset`. Or
you can use `tf.keras.metrics` (such as loss, accuracy, etc.) to
accumulate metrics across steps in a given epoch.
See the
[custom training loop
tutorial](https://www.tensorflow.org/tutorials/distribute/custom_training)
for a more detailed example.
Note: `tf.distribute.Strategy` currently does not support TensorFlow's
partitioned variables (where a single variable is split across multiple
devices) at this time."
2419,extended,tensorflow/tensorflow/python/distribute/distribute_lib.py,779,method,`tf.distribute.StrategyExtended` with additional methods.
2420,scope,tensorflow/tensorflow/python/distribute/distribute_lib.py,798,method,"Context manager to make the strategy current and distribute variables.
This method returns a context manager, and is used as follows:
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> # Variable created inside scope:
>>> with strategy.scope():
... mirrored_variable = tf.Variable(1.)
>>> mirrored_variable
MirroredVariable:{
0: <tf.Variable 'Variable:0' shape=() dtype=float32, numpy=1.0>,
1: <tf.Variable 'Variable/replica_1:0' shape=() dtype=float32, numpy=1.0>
}
>>> # Variable created outside scope:
>>> regular_variable = tf.Variable(1.)
>>> regular_variable
<tf.Variable 'Variable:0' shape=() dtype=float32, numpy=1.0>
_What happens when Strategy.scope is entered?_
* `strategy` is installed in the global context as the ""current"" strategy.
Inside this scope, `tf.distribute.get_strategy()` will now return this
strategy. Outside this scope, it returns the default no-op strategy.
* Entering the scope also enters the ""cross-replica context"". See
`tf.distribute.StrategyExtended` for an explanation on cross-replica and
replica contexts.
* Variable creation inside `scope` is intercepted by the strategy. Each
strategy defines how it wants to affect the variable creation. Sync
strategies like `MirroredStrategy`, `TPUStrategy` and
`MultiWorkerMiroredStrategy` create variables replicated on each replica,
whereas `ParameterServerStrategy` creates variables on the parameter
servers. This is done using a custom `tf.variable_creator_scope`.
* In some strategies, a default device scope may also be entered: in
`MultiWorkerMiroredStrategy`, a default device scope of ""/CPU:0"" is
entered on each worker.
Note: Entering a scope does not automatically distribute a computation, except
in the case of high level training framework like keras `model.fit`. If
you're not using `model.fit`, you
need to use `strategy.run` API to explicitly distribute that computation.
See an example in the [custom training loop tutorial](https://www.tensorflow.org/tutorials/distribute/custom_training).
_What should be in scope and what should be outside?_
There are a number of requirements on what needs to happen inside the scope.
However, in places where we have information about which strategy is in use,
we often enter the scope for the user, so they don't have to do it
explicitly (i.e. calling those either inside or outside the scope is OK).
* Anything that creates variables that should be distributed variables
must be in `strategy.scope`. This can be either by directly putting it in
scope, or relying on another API like `strategy.run` or `model.fit` to
enter it for you. Any variable that is created outside scope will not be
distributed and may have performance implications. Common things that
create variables in TF: models, optimizers, metrics. These should always
be created inside the scope. Another source of variable creation can be
a checkpoint restore - when variables are created lazily. Note that any
variable created inside a strategy captures the strategy information. So
reading and writing to these variables outside the `strategy.scope` can
also work seamlessly, without the user having to enter the scope.
* Some strategy APIs (such as `strategy.run` and `strategy.reduce`) which
require to be in a strategy's scope, enter the scope for you
automatically, which means when using those APIs you don't need to
enter the scope yourself.
* When a `tf.keras.Model` is created inside a `strategy.scope`, we capture
this information. When high level training frameworks methods such as
`model.compile`, `model.fit` etc are then called
on this model, we automatically enter the scope, as well as use this
strategy to distribute the training etc. See
detailed example in [distributed keras tutorial](https://www.tensorflow.org/tutorials/distribute/keras).
Note that simply calling the `model(..)` is not impacted - only high
level training framework APIs are. `model.compile`, `model.fit`,
`model.evaluate`, `model.predict` and `model.save` can all be called
inside or outside the scope.
* The following can be either inside or outside the scope:
** Creating the input datasets
** Defining `tf.function`s that represent your training step
** Saving APIs such as `tf.saved_model.save`. Loading creates variables,
so that should go inside the scope if you want to train the model in a
distributed way.
** Checkpoint saving. As mentioned above - `checkpoint.restore` may
sometimes need to be inside scope if it creates variables.
Returns:
A context manager."
2421,colocate_vars_with,tensorflow/tensorflow/python/distribute/distribute_lib.py,890,method,DEPRECATED: use extended.colocate_vars_with() instead.
2422,make_dataset_iterator,tensorflow/tensorflow/python/distribute/distribute_lib.py,895,method,DEPRECATED TF 1.x ONLY.
2423,make_input_fn_iterator,tensorflow/tensorflow/python/distribute/distribute_lib.py,900,method,DEPRECATED TF 1.x ONLY.
2424,experimental_make_numpy_dataset,tensorflow/tensorflow/python/distribute/distribute_lib.py,913,method,"Makes a `tf.data.Dataset` from a numpy array.
This avoids adding `numpy_input` as a large constant in the graph,
and copies the data to the machine or machines that will be processing
the input.
Note that you will likely need to use `experimental_distribute_dataset`
with the returned dataset to further distribute it with the strategy.
Example:
>>> strategy = tf.distribute.MirroredStrategy()
>>> numpy_input = np.ones([10], dtype=np.float32)
>>> dataset = strategy.experimental_make_numpy_dataset(numpy_input)
>>> dataset
<TensorSliceDataset shapes: (), types: tf.float32>
>>> dataset = dataset.batch(2)
>>> dist_dataset = strategy.experimental_distribute_dataset(dataset)
Args:
numpy_input: a nest of NumPy input arrays that will be converted into a
dataset. Note that the NumPy arrays are stacked, as that is normal
`tf.data.Dataset` behavior.
Returns:
A `tf.data.Dataset` representing `numpy_input`."
2425,experimental_run,tensorflow/tensorflow/python/distribute/distribute_lib.py,945,method,DEPRECATED TF 1.x ONLY.
2426,experimental_distribute_dataset,tensorflow/tensorflow/python/distribute/distribute_lib.py,951,method,"Creates `tf.distribute.DistributedDataset` from `tf.data.Dataset`.
The returned `tf.distribute.DistributedDataset` can be iterated over
similar to regular datasets.
NOTE: The user cannot add any more transformations to a
`tf.distribute.DistributedDataset`. You can only create an iterator or
examine the `tf.TypeSpec` of the data generated by it. See API docs of
`tf.distribute.DistributedDataset` to learn more.
The following is an example:
>>> global_batch_size = 2
>>> # Passing the devices is optional.
... strategy = tf.distribute.MirroredStrategy(devices=[""GPU:0"", ""GPU:1""])
>>> # Create a dataset
... dataset = tf.data.Dataset.range(4).batch(global_batch_size)
>>> # Distribute that dataset
... dist_dataset = strategy.experimental_distribute_dataset(dataset)
>>> @tf.function
... def replica_fn(input):
... return input*2
>>> result = []
>>> # Iterate over the `tf.distribute.DistributedDataset`
... for x in dist_dataset:
... # process dataset elements
... result.append(strategy.run(replica_fn, args=(x,)))
>>> print(result)
[PerReplica:{
0: <tf.Tensor: shape=(1,), dtype=int64, numpy=array([0])>,
1: <tf.Tensor: shape=(1,), dtype=int64, numpy=array([2])>
}, PerReplica:{
0: <tf.Tensor: shape=(1,), dtype=int64, numpy=array([4])>,
1: <tf.Tensor: shape=(1,), dtype=int64, numpy=array([6])>
}]
Three key actions happending under the hood of this method are batching,
sharding, and prefetching.
In the code snippet above, `dataset` is batched by `global_batch_size`, and
calling `experimental_distribute_dataset` on it rebatches `dataset` to a
new batch size that is equal to the global batch size divided by the number
of replicas in sync. We iterate through it using a Pythonic for loop.
`x` is a `tf.distribute.DistributedValues` containing data for all replicas,
and each replica gets data of the new batch size.
`tf.distribute.Strategy.run` will take care of feeding the right per-replica
data in `x` to the right `replica_fn` executed on each replica.
Sharding contains autosharding across multiple workers and within every
worker. First, in multi-worker distributed training (i.e. when you use
`tf.distribute.experimental.MultiWorkerMirroredStrategy`
or `tf.distribute.TPUStrategy`), autosharding a dataset over a set of
workers means that each worker is assigned a subset of the entire dataset
(if the right `tf.data.experimental.AutoShardPolicy` is set). This is to
ensure that at each step, a global batch size of non-overlapping dataset
elements will be processed by each worker. Autosharding has a couple of
different options that can be specified using
`tf.data.experimental.DistributeOptions`. Then, sharding within each worker
means the method will split the data among all the worker devices (if more
than one a present). This will happen regardless of multi-worker
autosharding.
Note: for autosharding across multiple workers, the default mode is
`tf.data.experimental.AutoShardPolicy.AUTO`. This mode
will attempt to shard the input dataset by files if the dataset is
being created out of reader datasets (e.g. `tf.data.TFRecordDataset`,
`tf.data.TextLineDataset`, etc.) or otherwise shard the dataset by data,
where each of the workers will read the entire dataset and only process the
shard assigned to it. However, if you have less than one input file per
worker, we suggest that you disable dataset autosharding across workers by
setting the `tf.data.experimental.DistributeOptions.auto_shard_policy` to be
`tf.data.experimental.AutoShardPolicy.OFF`.
By default, this method adds a prefetch transformation at the end of the
user provided `tf.data.Dataset` instance. The argument to the prefetch
transformation which is `buffer_size` is equal to the number of replicas in
sync.
If the above batch splitting and dataset sharding logic is undesirable,
please use
`tf.distribute.Strategy.experimental_distribute_datasets_from_function`
instead, which does not do any automatic batching or sharding for you.
Note: If you are using TPUStrategy, the order in which the data is processed
by the workers when using
`tf.distribute.Strategy.experimental_distribute_dataset` or
`tf.distribute.Strategy.experimental_distribute_datasets_from_function` is
not guaranteed. This is typically required if you are using
`tf.distribute` to scale prediction. You can however insert an index for
each element in the batch and order outputs accordingly. Refer to [this
snippet](https://www.tensorflow.org/tutorials/distribute/input#caveats)
for an example of how to order outputs.
Note: Stateful dataset transformations are currently not supported with
`tf.distribute.experimental_distribute_dataset` or
`tf.distribute.experimental_distribute_datasets_from_function`. Any stateful
ops that the dataset may have are currently ignored. For example, if your
dataset has a `map_fn` that uses `tf.random.uniform` to rotate an image,
then you have a dataset graph that depends on state (i.e the random seed) on
the local machine where the python process is being executed.
For a tutorial on more usage and properties of this method, refer to the
[tutorial on distributed input](https://www.tensorflow.org/tutorials/distribute/input#tfdistributestrategyexperimental_distribute_dataset).
If you are interested in last partial batch handling, read [this section](https://www.tensorflow.org/tutorials/distribute/input#partial_batches).
Args:
dataset: `tf.data.Dataset` that will be sharded across all replicas using
the rules stated above.
options: `tf.distribute.InputOptions` used to control options on how this
dataset is distributed.
Returns:
A `tf.distribute.DistributedDataset`."
2427,experimental_distribute_datasets_from_function,tensorflow/tensorflow/python/distribute/distribute_lib.py,1070,method,"Distributes `tf.data.Dataset` instances created by calls to `dataset_fn`.
The argument `dataset_fn` that users pass in is an input function that has a
`tf.distribute.InputContext` argument and returns a `tf.data.Dataset`
instance. It is expected that the returned dataset from `dataset_fn` is
already batched by per-replica batch size (i.e. global batch size divided by
the number of replicas in sync) and sharded.
`tf.distribute.Strategy.experimental_distribute_datasets_from_function` does
not batch or shard the `tf.data.Dataset` instance
returned from the input function. `dataset_fn` will be called on the CPU
device of each of the workers and each generates a dataset where every
replica on that worker will dequeue one batch of inputs (i.e. if a worker
has two replicas, two batches will be dequeued from the `Dataset` every
step).
This method can be used for several purposes. First, it allows you to
specify your own batching and sharding logic. (In contrast,
`tf.distribute.experimental_distribute_dataset` does batching and sharding
for you.)For example, where
`experimental_distribute_dataset` is unable to shard the input files, this
method might be used to manually shard the dataset (avoiding the slow
fallback behavior in `experimental_distribute_dataset`). In cases where the
dataset is infinite, this sharding can be done by creating dataset replicas
that differ only in their random seed.
The `dataset_fn` should take an `tf.distribute.InputContext` instance where
information about batching and input replication can be accessed.
You can use `element_spec` property of the
`tf.distribute.DistributedDataset` returned by this API to query the
`tf.TypeSpec` of the elements returned by the iterator. This can be used to
set the `input_signature` property of a `tf.function`. Follow
`tf.distribute.DistributedDataset.element_spec` to see an example.
IMPORTANT: The `tf.data.Dataset` returned by `dataset_fn` should have a
per-replica batch size, unlike `experimental_distribute_dataset`, which uses
the global batch size. This may be computed using
`input_context.get_per_replica_batch_size`.
Note: If you are using TPUStrategy, the order in which the data is processed
by the workers when using
`tf.distribute.Strategy.experimental_distribute_dataset` or
`tf.distribute.Strategy.experimental_distribute_datasets_from_function` is
not guaranteed. This is typically required if you are using
`tf.distribute` to scale prediction. You can however insert an index for
each element in the batch and order outputs accordingly. Refer to [this
snippet](https://www.tensorflow.org/tutorials/distribute/input#caveats)
for an example of how to order outputs.
Note: Stateful dataset transformations are currently not supported with
`tf.distribute.experimental_distribute_dataset` or
`tf.distribute.experimental_distribute_datasets_from_function`. Any stateful
ops that the dataset may have are currently ignored. For example, if your
dataset has a `map_fn` that uses `tf.random.uniform` to rotate an image,
then you have a dataset graph that depends on state (i.e the random seed) on
the local machine where the python process is being executed.
For a tutorial on more usage and properties of this method, refer to the
[tutorial on distributed input](https://www.tensorflow.org/tutorials/distribute/input#tfdistributestrategyexperimental_distribute_datasets_from_function).
If you are interested in last partial batch handling, read [this section](https://www.tensorflow.org/tutorials/distribute/input#partial_batches).
Args:
dataset_fn: A function taking a `tf.distribute.InputContext` instance and
returning a `tf.data.Dataset`.
options: `tf.distribute.InputOptions` used to control options on how this
dataset is distributed.
Returns:
A `tf.distribute.DistributedDataset`."
2428,run,tensorflow/tensorflow/python/distribute/distribute_lib.py,1147,method,"Run `fn` on each replica, with the given arguments.
Executes ops specified by `fn` on each replica. If `args` or `kwargs` have
`tf.distribute.DistributedValues`, such as those produced by a
`tf.distribute.DistributedDataset` from
`tf.distribute.Strategy.experimental_distribute_dataset` or
`tf.distribute.Strategy.experimental_distribute_datasets_from_function`,
when `fn` is executed on a particular replica, it will be executed with the
component of `tf.distribute.DistributedValues` that correspond to that
replica.
`fn` may call `tf.distribute.get_replica_context()` to access members such
as `all_reduce`.
All arguments in `args` or `kwargs` should either be nest of tensors or
`tf.distribute.DistributedValues` containing tensors or composite tensors.
IMPORTANT: Depending on the implementation of `tf.distribute.Strategy` and
whether eager execution is enabled, `fn` may be called one or more times. If
`fn` is annotated with `tf.function` or `tf.distribute.Strategy.run` is
called inside a `tf.function`, eager execution is disabled and `fn` is
called once (or once per replica, if you are using MirroredStrategy) to
generate a Tensorflow graph, which will then be reused for execution with
new inputs. Otherwise, if eager execution is enabled, `fn` will be called
every step just like regular python code.
Example usage:
1. Constant tensor input.
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> tensor_input = tf.constant(3.0)
>>> @tf.function
... def replica_fn(input):
... return input*2.0
>>> result = strategy.run(replica_fn, args=(tensor_input,))
>>> result
PerReplica:{
0: <tf.Tensor: shape=(), dtype=float32, numpy=6.0>,
1: <tf.Tensor: shape=(), dtype=float32, numpy=6.0>
}
2. DistributedValues input.
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> @tf.function
... def run():
... def value_fn(value_context):
... return value_context.num_replicas_in_sync
... distributed_values = (
... strategy.experimental_distribute_values_from_function(
... value_fn))
... def replica_fn2(input):
... return input*2
... return strategy.run(replica_fn2, args=(distributed_values,))
>>> result = run()
>>> result
<tf.Tensor: shape=(), dtype=int32, numpy=4>
Args:
fn: The function to run. The output must be a `tf.nest` of `Tensor`s.
args: (Optional) Positional arguments to `fn`.
kwargs: (Optional) Keyword arguments to `fn`.
options: (Optional) An instance of `tf.distribute.RunOptions` specifying
the options to run `fn`.
Returns:
Merged return value of `fn` across replicas. The structure of the return
value is the same as the return value from `fn`. Each element in the
structure can either be `tf.distribute.DistributedValues`, `Tensor`
objects, or `Tensor`s (for example, if running on a single replica)."
2429,reduce,tensorflow/tensorflow/python/distribute/distribute_lib.py,1233,method,"Reduce `value` across replicas and return result on current device.
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> def step_fn():
... i = tf.distribute.get_replica_context().replica_id_in_sync_group
... return tf.identity(i)
>>>
>>> per_replica_result = strategy.run(step_fn)
>>> total = strategy.reduce(""SUM"", per_replica_result, axis=None)
>>> total
<tf.Tensor: shape=(), dtype=int32, numpy=1>
To see how this would look with multiple replicas, consider the same
example with MirroredStrategy with 2 GPUs:
```python
strategy = tf.distribute.MirroredStrategy(devices=[""GPU:0"", ""GPU:1""])
def step_fn():
i = tf.distribute.get_replica_context().replica_id_in_sync_group
return tf.identity(i)
per_replica_result = strategy.run(step_fn)
# Check devices on which per replica result is:
strategy.experimental_local_results(per_replica_result)[0].device
# /job:localhost/replica:0/task:0/device:GPU:0
strategy.experimental_local_results(per_replica_result)[1].device
# /job:localhost/replica:0/task:0/device:GPU:1
total = strategy.reduce(""SUM"", per_replica_result, axis=None)
# Check device on which reduced result is:
total.device
# /job:localhost/replica:0/task:0/device:CPU:0
```
This API is typically used for aggregating the results returned from
different replicas, for reporting etc. For example, loss computed from
different replicas can be averaged using this API before printing.
Note: The result is copied to the ""current"" device - which would typically
be the CPU of the worker on which the program is running. For `TPUStrategy`,
it is the first TPU host. For multi client `MultiWorkerMirroredStrategy`,
this is CPU of each worker.
There are a number of different tf.distribute APIs for reducing values
across replicas:
* `tf.distribute.ReplicaContext.all_reduce`: This differs from
`Strategy.reduce` in that it is for replica context and does
not copy the results to the host device. `all_reduce` should be typically
used for reductions inside the training step such as gradients.
* `tf.distribute.StrategyExtended.reduce_to` and
`tf.distribute.StrategyExtended.batch_reduce_to`: These APIs are more
advanced versions of `Strategy.reduce` as they allow customizing the
destination of the result. They are also called in cross replica context.
_What should axis be?_
Given a per-replica value returned by `run`, say a
per-example loss, the batch will be divided across all the replicas. This
function allows you to aggregate across replicas and optionally also across
batch elements by specifying the axis parameter accordingly.
For example, if you have a global batch size of 8 and 2
replicas, values for examples `[0, 1, 2, 3]` will be on replica 0 and
`[4, 5, 6, 7]` will be on replica 1. With `axis=None`, `reduce` will
aggregate only across replicas, returning `[0+4, 1+5, 2+6, 3+7]`.
This is useful when each replica is computing a scalar or some other value
that doesn't have a ""batch"" dimension (like a gradient or loss).
```
strategy.reduce(""sum"", per_replica_result, axis=None)
```
Sometimes, you will want to aggregate across both the global batch _and_
all replicas. You can get this behavior by specifying the batch
dimension as the `axis`, typically `axis=0`. In this case it would return a
scalar `0+1+2+3+4+5+6+7`.
```
strategy.reduce(""sum"", per_replica_result, axis=0)
```
If there is a last partial batch, you will need to specify an axis so
that the resulting shape is consistent across replicas. So if the last
batch has size 6 and it is divided into [0, 1, 2, 3] and [4, 5], you
would get a shape mismatch unless you specify `axis=0`. If you specify
`tf.distribute.ReduceOp.MEAN`, using `axis=0` will use the correct
denominator of 6. Contrast this with computing `reduce_mean` to get a
scalar value on each replica and this function to average those means,
which will weigh some values `1/8` and others `1/4`.
Args:
reduce_op: a `tf.distribute.ReduceOp` value specifying how values should
be combined. Allows using string representation of the enum such as
""SUM"", ""MEAN"".
value: a `tf.distribute.DistributedValues` instance, e.g. returned by
`Strategy.run`, to be combined into a single tensor. It can also be a
regular tensor when used with `OneDeviceStrategy` or default strategy.
axis: specifies the dimension to reduce along within each
replica's tensor. Should typically be set to the batch dimension, or
`None` to only reduce across replicas (e.g. if the tensor has no batch
dimension).
Returns:
A `Tensor`."
2430,unwrap,tensorflow/tensorflow/python/distribute/distribute_lib.py,1433,method,"Returns the list of all local per-replica values contained in `value`.
DEPRECATED: Please use `experimental_local_results` instead.
Note: This only returns values on the workers initiated by this client.
When using a `tf.distribute.Strategy` like
`tf.distribute.experimental.MultiWorkerMirroredStrategy`, each worker
will be its own client, and this function will only return values
computed on that worker.
Args:
value: A value returned by `experimental_run()`,
`extended.call_for_each_replica()`, or a variable created in `scope`.
Returns:
A tuple of values contained in `value`. If `value` represents a single
value, this returns `(value,).`"
2431,experimental_local_results,tensorflow/tensorflow/python/distribute/distribute_lib.py,1454,method,"Returns the list of all local per-replica values contained in `value`.
Note: This only returns values on the worker initiated by this client.
When using a `tf.distribute.Strategy` like
`tf.distribute.experimental.MultiWorkerMirroredStrategy`, each worker
will be its own client, and this function will only return values
computed on that worker.
Args:
value: A value returned by `experimental_run()`, `run()`,
`extended.call_for_each_replica()`, or a variable created in `scope`.
Returns:
A tuple of values contained in `value`. If `value` represents a single
value, this returns `(value,).`"
2432,group,tensorflow/tensorflow/python/distribute/distribute_lib.py,1474,method,Shortcut for `tf.group(self.experimental_local_results(value))`.
2433,num_replicas_in_sync,tensorflow/tensorflow/python/distribute/distribute_lib.py,1479,method,Returns number of replicas over which gradients are aggregated.
2434,configure,tensorflow/tensorflow/python/distribute/distribute_lib.py,1484,method,"DEPRECATED: use `update_config_proto` instead.
Configures the strategy class.
DEPRECATED: This method's functionality has been split into the strategy
constructor and `update_config_proto`. In the future, we will allow passing
cluster and config_proto to the constructor to configure the strategy. And
`update_config_proto` can be used to update the config_proto based on the
specific strategy."
2435,update_config_proto,tensorflow/tensorflow/python/distribute/distribute_lib.py,1504,method,DEPRECATED TF 1.x ONLY.
2436,cluster_resolver,tensorflow/tensorflow/python/distribute/distribute_lib.py,1524,method,"Returns the cluster resolver associated with this strategy.
In general, when using a multi-worker `tf.distribute` strategy such as
`tf.distribute.experimental.MultiWorkerMirroredStrategy` or
`tf.distribute.TPUStrategy()`, there is a
`tf.distribute.cluster_resolver.ClusterResolver` associated with the
strategy used, and such an instance is returned by this property.
Strategies that intend to have an associated
`tf.distribute.cluster_resolver.ClusterResolver` must set the
relevant attribute, or override this property; otherwise, `None` is returned
by default. Those strategies should also provide information regarding what
is returned by this property.
Single-worker strategies usually do not have a
`tf.distribute.cluster_resolver.ClusterResolver`, and in those cases this
property will return `None`.
The `tf.distribute.cluster_resolver.ClusterResolver` may be useful when the
user needs to access information such as the cluster spec, task type or task
id. For example,
```python
os.environ['TF_CONFIG'] = json.dumps({
'cluster': {
'worker': [""localhost:12345"", ""localhost:23456""],
'ps': [""localhost:34567""]
},
'task': {'type': 'worker', 'index': 0}
})
# This implicitly uses TF_CONFIG for the cluster and current task info.
strategy = tf.distribute.experimental.MultiWorkerMirroredStrategy()
...
if strategy.cluster_resolver.task_type == 'worker':
# Perform something that's only applicable on workers. Since we set this
# as a worker above, this block will run on this particular instance.
elif strategy.cluster_resolver.task_type == 'ps':
# Perform something that's only applicable on parameter servers. Since we
# set this as a worker above, this block will not run on this particular
# instance.
```
For more information, please see
`tf.distribute.cluster_resolver.ClusterResolver`'s API docstring.
Returns:
The cluster resolver associated with this strategy. Returns `None` if a
cluster resolver is not applicable or available in this strategy."
2437,mean_reduce_helper,tensorflow/tensorflow/python/distribute/distribute_lib.py,1372,method,Computes the numerator and denominator on each replica.
2438,reduce_sum,tensorflow/tensorflow/python/distribute/distribute_lib.py,1346,method,
2439,mean_reduce_fn,tensorflow/tensorflow/python/distribute/distribute_lib.py,1417,method,
2440,reduce_sum_fn,tensorflow/tensorflow/python/distribute/distribute_lib.py,1356,method,
2441,Strategy,tensorflow/tensorflow/python/distribute/distribute_lib.py,1584,class,
2442,experimental_assign_to_logical_device,tensorflow/tensorflow/python/distribute/distribute_lib.py,1588,method,"Adds annotation that `tensor` will be assigned to a logical device.
NOTE: This API is only supported in TPUStrategy for now.
This adds an annotation to `tensor` specifying that operations on
`tensor` will be invoked on logical core device id `logical_device_id`.
When model parallelism is used, the default behavior is that all ops
are placed on zero-th logical device.
```python
# Initializing TPU system with 2 logical devices and 4 replicas.
resolver = tf.distribute.cluster_resolver.TPUClusterResolver(tpu='')
tf.config.experimental_connect_to_cluster(resolver)
topology = tf.tpu.experimental.initialize_tpu_system(resolver)
device_assignment = tf.tpu.experimental.DeviceAssignment.build(
topology,
computation_shape=[1, 1, 1, 2],
num_replicas=4)
strategy = tf.distribute.TPUStrategy(
resolver, experimental_device_assignment=device_assignment)
iterator = iter(inputs)
@tf.function()
def step_fn(inputs):
output = tf.add(inputs, inputs)
# Add operation will be executed on logical device 0.
output = strategy.experimental_assign_to_logical_device(output, 0)
return output
strategy.run(step_fn, args=(next(iterator),))
```
Args:
tensor: Input tensor to annotate.
logical_device_id: Id of the logical core to which the tensor will be
assigned.
Raises:
ValueError: The logical device id presented is not consistent with total
number of partitions specified by the device assignment.
Returns:
Annotated tensor with idential value as `tensor`."
2443,experimental_split_to_logical_devices,tensorflow/tensorflow/python/distribute/distribute_lib.py,1637,method,"Adds annotation that `tensor` will be split across logical devices.
NOTE: This API is only supported in TPUStrategy for now.
This adds an annotation to tensor `tensor` specifying that operations on
`tensor` will be be split among multiple logical devices. Tensor `tensor`
will be split across dimensions specified by `partition_dimensions`.
The dimensions of `tensor` must be divisible by corresponding value in
`partition_dimensions`.
For example, for system with 8 logical devices, if `tensor` is an image
tensor with shape (batch_size, width, height, channel) and
`partition_dimensions` is [1, 2, 4, 1], then `tensor` will be split
2 in width dimension and 4 way in height dimension and the split
tensor values will be fed into 8 logical devices.
```python
# Initializing TPU system with 8 logical devices and 1 replica.
resolver = tf.distribute.cluster_resolver.TPUClusterResolver(tpu='')
tf.config.experimental_connect_to_cluster(resolver)
topology = tf.tpu.experimental.initialize_tpu_system(resolver)
device_assignment = tf.tpu.experimental.DeviceAssignment.build(
topology,
computation_shape=[1, 2, 2, 2],
num_replicas=1)
strategy = tf.distribute.TPUStrategy(
resolver, experimental_device_assignment=device_assignment)
iterator = iter(inputs)
@tf.function()
def step_fn(inputs):
inputs = strategy.experimental_split_to_logical_devices(
inputs, [1, 2, 4, 1])
# model() function will be executed on 8 logical devices with `inputs`
# split 2 * 4 ways.
output = model(inputs)
return output
strategy.run(step_fn, args=(next(iterator),))
```
Args:
tensor: Input tensor to annotate.
partition_dimensions: An unnested list of integers with the size equal to
rank of `tensor` specifying how `tensor` will be partitioned. The
product of all elements in `partition_dimensions` must be equal to the
total number of logical devices per replica.
Raises:
ValueError: 1) If the size of partition_dimensions does not equal to rank
of `tensor` or 2) if product of elements of `partition_dimensions` does
not match the number of logical devices per replica defined by the
implementing DistributionStrategy's device specification or
3) if a known size of `tensor` is not divisible by corresponding
value in `partition_dimensions`.
Returns:
Annotated tensor with idential value as `tensor`."
2444,experimental_replicate_to_logical_devices,tensorflow/tensorflow/python/distribute/distribute_lib.py,1700,method,"Adds annotation that `tensor` will be replicated to all logical devices.
NOTE: This API is only supported in TPUStrategy for now.
This adds an annotation to tensor `tensor` specifying that operations on
`tensor` will be invoked on all logical devices.
```python
# Initializing TPU system with 2 logical devices and 4 replicas.
resolver = tf.distribute.cluster_resolver.TPUClusterResolver(tpu='')
tf.config.experimental_connect_to_cluster(resolver)
topology = tf.tpu.experimental.initialize_tpu_system(resolver)
device_assignment = tf.tpu.experimental.DeviceAssignment.build(
topology,
computation_shape=[1, 1, 1, 2],
num_replicas=4)
strategy = tf.distribute.TPUStrategy(
resolver, experimental_device_assignment=device_assignment)
iterator = iter(inputs)
@tf.function()
def step_fn(inputs):
images, labels = inputs
images = strategy.experimental_split_to_logical_devices(
inputs, [1, 2, 4, 1])
# model() function will be executed on 8 logical devices with `inputs`
# split 2 * 4 ways.
output = model(inputs)
# For loss calculation, all logical devices share the same logits
# and labels.
labels = strategy.experimental_replicate_to_logical_devices(labels)
output = strategy.experimental_replicate_to_logical_devices(output)
loss = loss_fn(labels, output)
return loss
strategy.run(step_fn, args=(next(iterator),))
```
Args:
tensor: Input tensor to annotate.
Returns:
Annotated tensor with idential value as `tensor`."
2445,experimental_distribute_values_from_function,tensorflow/tensorflow/python/distribute/distribute_lib.py,1749,method,"Generates `tf.distribute.DistributedValues` from `value_fn`.
This function is to generate `tf.distribute.DistributedValues` to pass
into `run`, `reduce`, or other methods that take
distributed values when not using datasets.
Args:
value_fn: The function to run to generate values. It is called for
each replica with `tf.distribute.ValueContext` as the sole argument. It
must return a Tensor or a type that can be converted to a Tensor.
Returns:
A `tf.distribute.DistributedValues` containing a value for each replica.
Example usage:
1. Return constant value per replica:
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> def value_fn(ctx):
... return tf.constant(1.)
>>> distributed_values = (
... strategy.experimental_distribute_values_from_function(
... value_fn))
>>> local_result = strategy.experimental_local_results(distributed_values)
>>> local_result
(<tf.Tensor: shape=(), dtype=float32, numpy=1.0>,
<tf.Tensor: shape=(), dtype=float32, numpy=1.0>)
2. Distribute values in array based on replica_id:
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> array_value = np.array([3., 2., 1.])
>>> def value_fn(ctx):
... return array_value[ctx.replica_id_in_sync_group]
>>> distributed_values = (
... strategy.experimental_distribute_values_from_function(
... value_fn))
>>> local_result = strategy.experimental_local_results(distributed_values)
>>> local_result
(3.0, 2.0)
3. Specify values using num_replicas_in_sync:
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> def value_fn(ctx):
... return ctx.num_replicas_in_sync
>>> distributed_values = (
... strategy.experimental_distribute_values_from_function(
... value_fn))
>>> local_result = strategy.experimental_local_results(distributed_values)
>>> local_result
(2, 2)
4. Place values on devices and distribute:
```
strategy = tf.distribute.TPUStrategy()
worker_devices = strategy.extended.worker_devices
multiple_values = []
for i in range(strategy.num_replicas_in_sync):
with tf.device(worker_devices[i]):
multiple_values.append(tf.constant(1.0))
def value_fn(ctx):
return multiple_values[ctx.replica_id_in_sync_group]
distributed_values = strategy.
experimental_distribute_values_from_function(
value_fn)
```"
2446,StrategyV1,tensorflow/tensorflow/python/distribute/distribute_lib.py,1828,class,"A list of devices with a state & compute distribution policy.
See [the guide](https://www.tensorflow.org/guide/distribute_strategy)
for overview and examples.
Note: Not all `tf.distribute.Strategy` implementations currently support
TensorFlow's partitioned variables (where a single variable is split across
multiple devices) at this time."
2447,make_dataset_iterator,tensorflow/tensorflow/python/distribute/distribute_lib.py,1839,method,"Makes an iterator for input provided via `dataset`.
DEPRECATED: This method is not available in TF 2.x.
Data from the given dataset will be distributed evenly across all the
compute replicas. We will assume that the input dataset is batched by the
global batch size. With this assumption, we will make a best effort to
divide each batch across all the replicas (one or more workers).
If this effort fails, an error will be thrown, and the user should instead
use `make_input_fn_iterator` which provides more control to the user, and
does not try to divide a batch across replicas.
The user could also use `make_input_fn_iterator` if they want to
customize which input is fed to which replica/worker etc.
Args:
dataset: `tf.data.Dataset` that will be distributed evenly across all
replicas.
Returns:
An `tf.distribute.InputIterator` which returns inputs for each step of the
computation. User should call `initialize` on the returned iterator."
2448,make_input_fn_iterator,tensorflow/tensorflow/python/distribute/distribute_lib.py,1865,method,"Returns an iterator split across replicas created from an input function.
DEPRECATED: This method is not available in TF 2.x.
The `input_fn` should take an `tf.distribute.InputContext` object where
information about batching and input sharding can be accessed:
```
def input_fn(input_context):
batch_size = input_context.get_per_replica_batch_size(global_batch_size)
d = tf.data.Dataset.from_tensors([[1.]]).repeat().batch(batch_size)
return d.shard(input_context.num_input_pipelines,
input_context.input_pipeline_id)
with strategy.scope():
iterator = strategy.make_input_fn_iterator(input_fn)
replica_results = strategy.experimental_run(replica_fn, iterator)
```
The `tf.data.Dataset` returned by `input_fn` should have a per-replica
batch size, which may be computed using
`input_context.get_per_replica_batch_size`.
Args:
input_fn: A function taking a `tf.distribute.InputContext` object and
returning a `tf.data.Dataset`.
replication_mode: an enum value of `tf.distribute.InputReplicationMode`.
Only `PER_WORKER` is supported currently, which means there will be
a single call to `input_fn` per worker. Replicas will dequeue from the
local `tf.data.Dataset` on their worker.
Returns:
An iterator object that should first be `.initialize()`-ed. It may then
either be passed to `strategy.experimental_run()` or you can
`iterator.get_next()` to get the next value to pass to
`strategy.extended.call_for_each_replica()`."
2449,experimental_make_numpy_dataset,tensorflow/tensorflow/python/distribute/distribute_lib.py,1907,method,"Makes a tf.data.Dataset for input provided via a numpy array.
This avoids adding `numpy_input` as a large constant in the graph,
and copies the data to the machine or machines that will be processing
the input.
Note that you will likely need to use
tf.distribute.Strategy.experimental_distribute_dataset
with the returned dataset to further distribute it with the strategy.
Example:
```
numpy_input = np.ones([10], dtype=np.float32)
dataset = strategy.experimental_make_numpy_dataset(numpy_input)
dist_dataset = strategy.experimental_distribute_dataset(dataset)
```
Args:
numpy_input: A nest of NumPy input arrays that will be converted into a
dataset. Note that lists of Numpy arrays are stacked, as that is normal
`tf.data.Dataset` behavior.
session: (TensorFlow v1.x graph execution only) A session used for
initialization.
Returns:
A `tf.data.Dataset` representing `numpy_input`."
2450,experimental_run,tensorflow/tensorflow/python/distribute/distribute_lib.py,1938,method,"Runs ops in `fn` on each replica, with inputs from `input_iterator`.
DEPRECATED: This method is not available in TF 2.x. Please switch
to using `run` instead.
When eager execution is enabled, executes ops specified by `fn` on each
replica. Otherwise, builds a graph to execute the ops on each replica.
Each replica will take a single, different input from the inputs provided by
one `get_next` call on the input iterator.
`fn` may call `tf.distribute.get_replica_context()` to access members such
as `replica_id_in_sync_group`.
IMPORTANT: Depending on the `tf.distribute.Strategy` implementation being
used, and whether eager execution is enabled, `fn` may be called one or more
times (once for each replica).
Args:
fn: The function to run. The inputs to the function must match the outputs
of `input_iterator.get_next()`. The output must be a `tf.nest` of
`Tensor`s.
input_iterator: (Optional) input iterator from which the inputs are taken.
Returns:
Merged return value of `fn` across replicas. The structure of the return
value is the same as the return value from `fn`. Each element in the
structure can either be `PerReplica` (if the values are unsynchronized),
`Mirrored` (if the values are kept in sync), or `Tensor` (if running on a
single replica)."
2451,reduce,tensorflow/tensorflow/python/distribute/distribute_lib.py,1973,method,
2452,update_config_proto,tensorflow/tensorflow/python/distribute/distribute_lib.py,1978,method,"Returns a copy of `config_proto` modified for use with this strategy.
DEPRECATED: This method is not available in TF 2.x.
The updated config has something needed to run a strategy, e.g.
configuration to run collective ops, or device filters to improve
distributed training performance.
Args:
config_proto: a `tf.ConfigProto` object.
Returns:
The updated copy of the `config_proto`."
2453,StrategyExtendedV2,tensorflow/tensorflow/python/distribute/distribute_lib.py,1999,class,"Additional APIs for algorithms that need to be distribution-aware.
Note: For most usage of `tf.distribute.Strategy`, there should be no need to
call these methods, since TensorFlow libraries (such as optimizers) already
call these methods when needed on your behalf.
Some common use cases of functions on this page:
* _Locality_
`tf.distribute.DistributedValues` can have the same _locality_ as a
_distributed variable_, which leads to a mirrored value residing on the same
devices as the variable (as opposed to the compute devices). Such values may
be passed to a call to `tf.distribute.StrategyExtended.update` to update the
value of a variable. You may use
`tf.distribute.StrategyExtended.colocate_vars_with` to give a variable the
same locality as another variable. You may convert a ""PerReplica"" value to a
variable's locality by using `tf.distribute.StrategyExtended.reduce_to` or
`tf.distribute.StrategyExtended.batch_reduce_to`.
* _How to update a distributed variable_
A distributed variable is variables created on multiple devices. As discussed
in the [glossary](https://www.tensorflow.org/api_docs/python/tf/distribute),
mirrored variable and SyncOnRead variable are two examples. The standard
pattern for updating distributed variables is to:
1. In your function passed to `tf.distribute.Strategy.run`,
compute a list of (update, variable) pairs. For example, the update might
be a gradient of the loss with respect to the variable.
2. Switch to cross-replica mode by calling
`tf.distribute.get_replica_context().merge_call()` with the updates and
variables as arguments.
3. Call
`tf.distribute.StrategyExtended.reduce_to(VariableAggregation.SUM, t, v)`
(for one variable) or `tf.distribute.StrategyExtended.batch_reduce_to`
(for a list of variables) to sum the updates.
4. Call `tf.distribute.StrategyExtended.update(v)` for each variable to update
its value.
Steps 2 through 4 are done automatically by class
`tf.keras.optimizers.Optimizer` if you call its
`tf.keras.optimizers.Optimizer.apply_gradients` method in a replica context.
In fact, a higher-level solution to update a distributed variable is by
calling `assign` on the variable as you would do to a regular `tf.Variable`.
You can call the method in both _replica context_ and _cross-replica context_.
For a _mirrored variable_, calling `assign` in _replica context_ requires you
to specify the `aggregation` type in the variable constructor. In that case,
the context switching and sync described in steps 2 through 4 are handled for
you. If you call `assign` on _mirrored variable_ in _cross-replica context_,
you can only assign a single value or assign values from another mirrored
variable or a mirrored `tf.distribute.DistributedValues`. For a _SyncOnRead
variable_, in _replica context_, you can simply call `assign` on it and no
aggregation happens under the hood. In _cross-replica context_, you can only
assign a single value to a SyncOnRead variable. One example case is restoring
from a checkpoint: if the `aggregation` type of the variable is
`tf.VariableAggregation.SUM`, it is assumed that replica values were added
before checkpointing, so at the time of restoring, the value is divided by
the number of replicas and then assigned to each replica; if the `aggregation`
type is `tf.VariableAggregation.MEAN`, the value is assigned to each replica
directly."
2454,variable_created_in_scope,tensorflow/tensorflow/python/distribute/distribute_lib.py,2139,method,"Tests whether `v` was created while this strategy scope was active.
Variables created inside the strategy scope are ""owned"" by it:
>>> strategy = tf.distribute.MirroredStrategy()
>>> with strategy.scope():
... v = tf.Variable(1.)
>>> strategy.extended.variable_created_in_scope(v)
True
Variables created outside the strategy are not owned by it:
>>> strategy = tf.distribute.MirroredStrategy()
>>> v = tf.Variable(1.)
>>> strategy.extended.variable_created_in_scope(v)
False
Args:
v: A `tf.Variable` instance.
Returns:
True if `v` was created inside the scope, False if not."
2455,colocate_vars_with,tensorflow/tensorflow/python/distribute/distribute_lib.py,2165,method,"Scope that controls which devices variables will be created on.
No operations should be added to the graph inside this scope, it
should only be used when creating variables (some implementations
work by changing variable creation, others work by using a
tf.compat.v1.colocate_with() scope).
This may only be used inside `self.scope()`.
Example usage:
```
with strategy.scope():
var1 = tf.Variable(...)
with strategy.extended.colocate_vars_with(var1):
# var2 and var3 will be created on the same device(s) as var1
var2 = tf.Variable(...)
var3 = tf.Variable(...)
def fn(v1, v2, v3):
# operates on v1 from var1, v2 from var2, and v3 from var3
# `fn` runs on every device `var1` is on, `var2` and `var3` will be there
# too.
strategy.extended.update(var1, fn, args=(var2, var3))
```
Args:
colocate_with_variable: A variable created in this strategy's `scope()`.
Variables created while in the returned context manager will be on the
same set of devices as `colocate_with_variable`.
Returns:
A context manager."
2456,reduce_to,tensorflow/tensorflow/python/distribute/distribute_lib.py,2250,method,"Combine (via e.g. sum or mean) values across replicas.
`reduce_to` aggregates `tf.distribute.DistributedValues` and distributed
variables. It supports both dense values and `tf.IndexedSlices`.
This API currently can only be called in cross-replica context. Other
variants to reduce values across replicas are:
* `tf.distribute.StrategyExtended.batch_reduce_to`: the batch version of
this API.
* `tf.distribute.ReplicaContext.all_reduce`: the counterpart of this API
in replica context. It supports both batched and non-batched all-reduce.
* `tf.distribute.Strategy.reduce`: a more convenient method to reduce
to the host in cross-replica context.
`destinations` specifies where to reduce the value to, e.g. ""GPU:0"". You can
also pass in a `Tensor`, and the destinations will be the device of that
tensor. For all-reduce, pass the same to `value` and `destinations`.
It can be used in `tf.distribute.ReplicaContext.merge_call` to write code
that works for all `tf.distribute.Strategy`.
>>> @tf.function
... def step_fn(var):
...
... def merge_fn(strategy, value, var):
... # All-reduce the value. Note that `value` here is a
... # `tf.distribute.DistributedValues`.
... reduced = strategy.extended.reduce_to(tf.distribute.ReduceOp.SUM,
... value, destinations=var)
... strategy.extended.update(var, lambda var, value: var.assign(value),
... args=(reduced,))
...
... value = tf.identity(1.)
... tf.distribute.get_replica_context().merge_call(merge_fn,
... args=(value, var))
>>>
>>> def run(strategy):
... with strategy.scope():
... v = tf.Variable(0.)
... strategy.run(step_fn, args=(v,))
... return v
>>>
>>> run(tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""]))
MirroredVariable:{
0: <tf.Variable 'Variable:0' shape=() dtype=float32, numpy=2.0>,
1: <tf.Variable 'Variable/replica_1:0' shape=() dtype=float32, numpy=2.0>
}
>>> run(tf.distribute.experimental.CentralStorageStrategy(
... compute_devices=[""GPU:0"", ""GPU:1""], parameter_device=""CPU:0""))
<tf.Variable 'Variable:0' shape=() dtype=float32, numpy=2.0>
>>> run(tf.distribute.OneDeviceStrategy(""GPU:0""))
<tf.Variable 'Variable:0' shape=() dtype=float32, numpy=1.0>
Args:
reduce_op: a `tf.distribute.ReduceOp` or string. How to reduce the value.
value: a `tf.distribute.DistributedValue`, or a `tf.Tensor` like object.
destinations: a `tf.distribute.DistributedValue`, a `tf.Variable`, a
`tf.Tensor` alike object, or a device string. It specifies the devices
to reduce to. To perform an all-reduce, pass the same to `value` and
`destinations`. Note that if it's a `tf.Variable`, the value is reduced
to the devices of that variable, this method doesn't update the variable.
experimental_hints: a `tf.distrbute.experimental.CollectiveHints`. Hints
to perform collective operations. See
`tf.distrbute.experimental.CollectiveHints` for details.
Returns:
A tensor or value reduced to `destinations`."
2457,batch_reduce_to,tensorflow/tensorflow/python/distribute/distribute_lib.py,2333,method,"Combine multiple `reduce_to` calls into one for faster execution.
Similar to `reduce_to`, but accepts a list of (value, destinations) pairs.
It's more efficient than reduce each value separately.
This API currently can only be called in cross-replica context. Other
variants to reduce values across replicas are:
* `tf.distribute.StrategyExtended.reduce_to`: the non-batch version of
this API.
* `tf.distribute.ReplicaContext.all_reduce`: the counterpart of this API
in replica context. It supports both batched and non-batched all-reduce.
* `tf.distribute.Strategy.reduce`: a more convenient method to reduce
to the host in cross-replica context.
See `reduce_to` for more information.
>>> @tf.function
... def step_fn(var):
...
... def merge_fn(strategy, value, var):
... # All-reduce the value. Note that `value` here is a
... # `tf.distribute.DistributedValues`.
... reduced = strategy.extended.batch_reduce_to(
... tf.distribute.ReduceOp.SUM, [(value, var)])[0]
... strategy.extended.update(var, lambda var, value: var.assign(value),
... args=(reduced,))
...
... value = tf.identity(1.)
... tf.distribute.get_replica_context().merge_call(merge_fn,
... args=(value, var))
>>>
>>> def run(strategy):
... with strategy.scope():
... v = tf.Variable(0.)
... strategy.run(step_fn, args=(v,))
... return v
>>>
>>> run(tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""]))
MirroredVariable:{
0: <tf.Variable 'Variable:0' shape=() dtype=float32, numpy=2.0>,
1: <tf.Variable 'Variable/replica_1:0' shape=() dtype=float32, numpy=2.0>
}
>>> run(tf.distribute.experimental.CentralStorageStrategy(
... compute_devices=[""GPU:0"", ""GPU:1""], parameter_device=""CPU:0""))
<tf.Variable 'Variable:0' shape=() dtype=float32, numpy=2.0>
>>> run(tf.distribute.OneDeviceStrategy(""GPU:0""))
<tf.Variable 'Variable:0' shape=() dtype=float32, numpy=1.0>
Args:
reduce_op: a `tf.distribute.ReduceOp`. How to reduce the value.
value_destination_pairs: a sequence of (value, destinations) pairs. See
`reduce_to()` for descriptions.
experimental_hints: a `tf.distrbute.experimental.CollectiveHints`. Hints
to perform collective operations.
Returns:
A list of reduced values, one per pair in `value_destination_pairs`."
2458,update,tensorflow/tensorflow/python/distribute/distribute_lib.py,2412,method,"Run `fn` to update `var` using inputs mirrored to the same devices.
`tf.distribute.StrategyExtended.update` takes a distributed variable `var`
to be updated, an update function `fn`, and `args` and `kwargs` for `fn`. It
applies `fn` to each component variable of `var` and passes corresponding
values from `args` and `kwargs`. Neither `args` nor `kwargs` may contain
per-replica values. If they contain mirrored values, they will be unwrapped
before calling `fn`. For example, `fn` can be `assign_add` and `args` can be
a mirrored DistributedValues where each component contains the value to be
added to this mirrored variable `var`. Calling `update` will call
`assign_add` on each component variable of `var` with the corresponding
tensor value on that device.
Example usage:
```python
strategy = tf.distribute.MirroredStrategy(['GPU:0', 'GPU:1']) # With 2 devices
with strategy.scope():
v = tf.Variable(5.0, aggregation=tf.VariableAggregation.SUM)
def update_fn(v):
return v.assign(1.0)
result = strategy.extended.update(v, update_fn)
# result is
# Mirrored:{
# 0: tf.Tensor(1.0, shape=(), dtype=float32),
# 1: tf.Tensor(1.0, shape=(), dtype=float32)
# }
```
If `var` is mirrored across multiple devices, then this method implements
logic as following:
```python
results = {}
for device, v in var:
with tf.device(device):
# args and kwargs will be unwrapped if they are mirrored.
results[device] = fn(v, *args, **kwargs)
return merged(results)
```
Otherwise, this method returns `fn(var, *args, **kwargs)` colocated with
`var`.
Args:
var: Variable, possibly mirrored to multiple devices, to operate on.
fn: Function to call. Should take the variable as the first argument.
args: Tuple or list. Additional positional arguments to pass to `fn()`.
kwargs: Dict with keyword arguments to pass to `fn()`.
group: Boolean. Defaults to True. If False, the return value will be
unwrapped.
Returns:
By default, the merged return value of `fn` across all replicas. The
merged result has dependencies to make sure that if it is evaluated at
all, the side effects (updates) will happen on every replica. If instead
""group=False"" is specified, this function will return a nest of lists
where each list has an element per replica, and the caller is responsible
for ensuring all elements are executed."
2459,update_non_slot,tensorflow/tensorflow/python/distribute/distribute_lib.py,2485,method,"Runs `fn(*args, **kwargs)` on `colocate_with` devices.
Used to update non-slot variables.
Args:
colocate_with: Devices returned by `non_slot_devices()`.
fn: Function to execute.
args: Tuple or list. Positional arguments to pass to `fn()`.
kwargs: Dict with keyword arguments to pass to `fn()`.
group: Boolean. Defaults to True. If False, the return value will be
unwrapped.
Returns:
Return value of `fn`, possibly merged across devices."
2460,value_container,tensorflow/tensorflow/python/distribute/distribute_lib.py,2516,method,"Returns the container that this per-replica `value` belongs to.
Args:
value: A value returned by `run()` or a variable created in `scope()`.
Returns:
A container that `value` belongs to.
If value does not belong to any container (including the case of
container having been destroyed), returns the value itself.
`value in experimental_local_results(value_container(value))` will
always be true."
2461,experimental_require_static_shapes,tensorflow/tensorflow/python/distribute/distribute_lib.py,2544,method,Returns `True` if static shape is required; `False` otherwise.
2462,worker_devices,tensorflow/tensorflow/python/distribute/distribute_lib.py,2554,method,"Returns the tuple of all devices used to for compute replica execution.
"
2463,parameter_devices,tensorflow/tensorflow/python/distribute/distribute_lib.py,2561,method,Returns the tuple of all devices used to place variables.
2464,non_slot_devices,tensorflow/tensorflow/python/distribute/distribute_lib.py,2567,method,"Device(s) for non-slot variables.
This method returns non-slot devices where non-slot variables are placed.
Users can create non-slot variables on these devices by using a block:
```python
with tf.distribute.StrategyExtended.colocate_vars_with(tf.distribute.StrategyExtended.non_slot_devices(...)):
...
```
Args:
var_list: The list of variables being optimized, needed with the
default `tf.distribute.Strategy`.
Returns:
A sequence of devices for non-slot variables."
2465,creator_with_resource_vars,tensorflow/tensorflow/python/distribute/distribute_lib.py,2090,method,Variable creator to use in `_CurrentDistributionContext`.
2466,distributed_getter,tensorflow/tensorflow/python/distribute/distribute_lib.py,2117,method,
2467,create_colocated_variable,tensorflow/tensorflow/python/distribute/distribute_lib.py,2202,method,
2468,StrategyExtendedV1,tensorflow/tensorflow/python/distribute/distribute_lib.py,2618,class,
2469,experimental_make_numpy_dataset,tensorflow/tensorflow/python/distribute/distribute_lib.py,2622,method,"Makes a dataset for input provided via a numpy array.
This avoids adding `numpy_input` as a large constant in the graph,
and copies the data to the machine or machines that will be processing
the input.
Args:
numpy_input: A nest of NumPy input arrays that will be distributed evenly
across all replicas. Note that lists of Numpy arrays are stacked, as
that is normal `tf.data.Dataset` behavior.
session: (TensorFlow v1.x graph execution only) A session used for
initialization.
Returns:
A `tf.data.Dataset` representing `numpy_input`."
2470,broadcast_to,tensorflow/tensorflow/python/distribute/distribute_lib.py,2645,method,"Mirror a tensor on one device to all worker devices.
Args:
tensor: A Tensor value to broadcast.
destinations: A mirrored variable or device string specifying the
destination devices to copy `tensor` to.
Returns:
A value mirrored to `destinations` devices."
2471,experimental_run_steps_on_iterator,tensorflow/tensorflow/python/distribute/distribute_lib.py,2665,method,"DEPRECATED: please use `run` instead.
Run `fn` with input from `iterator` for `iterations` times.
This method can be used to run a step function for training a number of
times using input from a dataset.
Args:
fn: function to run using this distribution strategy. The function must
have the following signature: `def fn(context, inputs)`. `context` is an
instance of `MultiStepContext` that will be passed when `fn` is run.
`context` can be used to specify the outputs to be returned from `fn`
by calling `context.set_last_step_output`. It can also be used to
capture non tensor outputs by `context.set_non_tensor_output`. See
`MultiStepContext` documentation for more information. `inputs` will
have same type/structure as `iterator.get_next()`. Typically, `fn`
will use `call_for_each_replica` method of the strategy to distribute
the computation over multiple replicas.
iterator: Iterator of a dataset that represents the input for `fn`. The
caller is responsible for initializing the iterator as needed.
iterations: (Optional) Number of iterations that `fn` should be run.
Defaults to 1.
initial_loop_values: (Optional) Initial values to be passed into the
loop that runs `fn`. Defaults to `None`. # TODO(priyag): Remove
initial_loop_values argument when we have a mechanism to infer the
outputs of `fn`.
Returns:
Returns the `MultiStepContext` object which has the following properties,
among other things:
- run_op: An op that runs `fn` `iterations` times.
- last_step_outputs: A dictionary containing tensors set using
`context.set_last_step_output`. Evaluating this returns the value of
the tensors after the last iteration.
- non_tensor_outputs: A dictionary containing anything that was set by
`fn` by calling `context.set_non_tensor_output`."
2472,call_for_each_replica,tensorflow/tensorflow/python/distribute/distribute_lib.py,2716,method,"Run `fn` once per replica.
`fn` may call `tf.get_replica_context()` to access methods such as
`replica_id_in_sync_group` and `merge_call()`.
`merge_call()` is used to communicate between the replicas and
re-enter the cross-replica context. All replicas pause their execution
having encountered a `merge_call()` call. After that the
`merge_fn`-function is executed. Its results are then unwrapped and
given back to each replica call. After that execution resumes until
`fn` is complete or encounters another `merge_call()`. Example:
```python
# Called once in ""cross-replica"" context.
def merge_fn(distribution, three_plus_replica_id):
# sum the values across replicas
return sum(distribution.experimental_local_results(three_plus_replica_id))
# Called once per replica in `distribution`, in a ""replica"" context.
def fn(three):
replica_ctx = tf.get_replica_context()
v = three + replica_ctx.replica_id_in_sync_group
# Computes the sum of the `v` values across all replicas.
s = replica_ctx.merge_call(merge_fn, args=(v,))
return s + v
with distribution.scope():
# in ""cross-replica"" context
...
merged_results = distribution.run(fn, args=[3])
# merged_results has the values from every replica execution of `fn`.
# This statement prints a list:
print(distribution.experimental_local_results(merged_results))
```
Args:
fn: function to run (will be run once per replica).
args: Tuple or list with positional arguments for `fn`.
kwargs: Dict with keyword arguments for `fn`.
Returns:
Merged return value of `fn` across all replicas."
2473,read_var,tensorflow/tensorflow/python/distribute/distribute_lib.py,2769,method,"Reads the value of a variable.
Returns the aggregate value of a replica-local variable, or the
(read-only) value of any other variable.
Args:
v: A variable allocated within the scope of this `tf.distribute.Strategy`.
Returns:
A tensor representing the value of `v`, aggregated across replicas if
necessary."
2474,experimental_between_graph,tensorflow/tensorflow/python/distribute/distribute_lib.py,2785,method,"Whether the strategy uses between-graph replication or not.
This is expected to return a constant value that will not be changed
throughout its life cycle."
2475,experimental_should_init,tensorflow/tensorflow/python/distribute/distribute_lib.py,2794,method,Whether initialization is needed.
2476,should_checkpoint,tensorflow/tensorflow/python/distribute/distribute_lib.py,2799,method,Whether checkpointing is needed.
2477,should_save_summary,tensorflow/tensorflow/python/distribute/distribute_lib.py,2804,method,Whether saving summaries is needed.
2478,ReplicaContext,tensorflow/tensorflow/python/distribute/distribute_lib.py,2829,class,"A class with a collection of APIs that can be called in a replica context.
You can use `tf.distribute.get_replica_context` to get an instance of
`ReplicaContext`, which can only be called inside the function passed to
`tf.distribute.Strategy.run`.
>>> strategy = tf.distribute.MirroredStrategy(['GPU:0', 'GPU:1'])
>>> def func():
... replica_context = tf.distribute.get_replica_context()
... return replica_context.replica_id_in_sync_group
>>> strategy.run(func)
PerReplica:{
0: <tf.Tensor: shape=(), dtype=int32, numpy=0>,
1: <tf.Tensor: shape=(), dtype=int32, numpy=1>
}"
2479,merge_call,tensorflow/tensorflow/python/distribute/distribute_lib.py,2874,method,"Merge args across replicas and run `merge_fn` in a cross-replica context.
This allows communication and coordination when there are multiple calls
to the step_fn triggered by a call to `strategy.run(step_fn, ...)`.
See `tf.distribute.Strategy.run` for an explanation.
If not inside a distributed scope, this is equivalent to:
```
strategy = tf.distribute.get_strategy()
with cross-replica-context(strategy):
return merge_fn(strategy, *args, **kwargs)
```
Args:
merge_fn: Function that joins arguments from threads that are given as
PerReplica. It accepts `tf.distribute.Strategy` object as
the first argument.
args: List or tuple with positional per-thread arguments for `merge_fn`.
kwargs: Dict with keyword per-thread arguments for `merge_fn`.
Returns:
The return value of `merge_fn`, except for `PerReplica` values which are
unpacked."
2480,num_replicas_in_sync,tensorflow/tensorflow/python/distribute/distribute_lib.py,2918,method,Returns number of replicas that are kept in sync.
2481,replica_id_in_sync_group,tensorflow/tensorflow/python/distribute/distribute_lib.py,2923,method,"Returns the id of the replica.
This identifies the replica among all replicas that are kept in sync. The
value of the replica id can range from 0 to
`tf.distribute.ReplicaContext.num_replicas_in_sync` - 1.
NOTE: This is not guaranteed to be the same ID as the XLA replica ID use
for low-level operations such as collective_permute."
2482,strategy,tensorflow/tensorflow/python/distribute/distribute_lib.py,2937,method,The current `tf.distribute.Strategy` object.
2483,devices,tensorflow/tensorflow/python/distribute/distribute_lib.py,2943,method,"Returns the devices this replica is to be executed on, as a tuple of strings.
NOTE: For `tf.distribute.MirroredStrategy` and
`tf.distribute.experimental.MultiWorkerMirroredStrategy`, this returns a
nested
list of device strings, e.g, [[""GPU:0""]]."
2484,all_reduce,tensorflow/tensorflow/python/distribute/distribute_lib.py,2954,method,"All-reduces the given `value Tensor` nest across replicas.
If `all_reduce` is called in any replica, it must be called in all replicas.
The nested structure and `Tensor` shapes must be identical in all replicas.
IMPORTANT: The ordering of communications must be identical in all replicas.
Example with two replicas:
Replica 0 `value`: {'a': 1, 'b': [40, 1]}
Replica 1 `value`: {'a': 3, 'b': [ 2, 98]}
If `reduce_op` == `SUM`:
Result (on all replicas): {'a': 4, 'b': [42, 99]}
If `reduce_op` == `MEAN`:
Result (on all replicas): {'a': 2, 'b': [21, 49.5]}
Args:
reduce_op: Reduction type, an instance of `tf.distribute.ReduceOp` enum.
value: The nested structure of `Tensor`s to all-reduce. The structure must
be compatible with `tf.nest`.
experimental_hints: A `tf.distrbute.experimental.CollectiveHints`. Hints
to perform collective operations.
Returns:
A `Tensor` nest with the reduced `value`s from each replica."
2485,replica_id_is_zero,tensorflow/tensorflow/python/distribute/distribute_lib.py,2858,method,
2486,batch_all_reduce,tensorflow/tensorflow/python/distribute/distribute_lib.py,2987,method,
2487,grad_wrapper,tensorflow/tensorflow/python/distribute/distribute_lib.py,2995,method,
2488,regroup,tensorflow/tensorflow/python/distribute/distribute_utils.py,32,function,"Makes a nest per-replica into a nest of PerReplica/Mirrored values.
Args:
values: Values to regroup
wrap_class: Class that `values` be wrapped in.
always_wrap: Always wrap the `values` in `wrap_class` even if the values
are the same except for DistributeVariable.
Returns:
Wrapped `values`."
2489,select_replica,tensorflow/tensorflow/python/distribute/distribute_utils.py,127,function,Specialize a nest of regular & per-replica values for one replica.
2490,select_replica_mirrored,tensorflow/tensorflow/python/distribute/distribute_utils.py,143,function,Specialize a nest of regular & mirrored values for one replica.
2491,update_regroup,tensorflow/tensorflow/python/distribute/distribute_utils.py,162,function,"Regroup for an update, with dependencies to ensure all updates execute."
2492,value_container,tensorflow/tensorflow/python/distribute/distribute_utils.py,195,function,"Returns the container that this per-replica `value` belongs to.
Args:
val: A value returned by `call_for_each_replica()` or a variable created in
`scope()`.
Returns:
A container that `value` belongs to.
If value does not belong to any container (including the case of
container having been destroyed), returns the value itself."
2493,is_distributed_variable,tensorflow/tensorflow/python/distribute/distribute_utils.py,217,function,Determine if a variable is ds variable or TPU mirrored variable.
2494,validate_colocate_distributed_variable,tensorflow/tensorflow/python/distribute/distribute_utils.py,231,function,
2495,validate_colocate,tensorflow/tensorflow/python/distribute/distribute_utils.py,239,function,
2496,create_mirrored_variable,tensorflow/tensorflow/python/distribute/distribute_utils.py,248,function,
2497,write_dirpath,tensorflow/tensorflow/python/distribute/distributed_file_utils.py,74,function,"Returns the writing dir that should be used to save file distributedly.
`dirpath` would be created if it doesn't exist.
Args:
dirpath: Original dirpath that would be used without distribution.
strategy: The tf.distribute strategy object currently used.
Returns:
The writing dir path that should be used to save with distribution."
2498,remove_temp_dirpath,tensorflow/tensorflow/python/distribute/distributed_file_utils.py,102,function,"Removes the temp path after writing is finished.
Args:
dirpath: Original dirpath that would be used without distribution.
strategy: The tf.distribute strategy object currently used."
2499,write_filepath,tensorflow/tensorflow/python/distribute/distributed_file_utils.py,125,function,"Returns the writing file path to be used to save file distributedly.
Directory to contain `filepath` would be created if it doesn't exist.
Args:
filepath: Original filepath that would be used without distribution.
strategy: The tf.distribute strategy object currently used.
Returns:
The writing filepath that should be used to save file with distribution."
2500,remove_temp_dir_with_filepath,tensorflow/tensorflow/python/distribute/distributed_file_utils.py,142,function,"Removes the temp path for file after writing is finished.
Args:
filepath: Original filepath that would be used without distribution.
strategy: The tf.distribute strategy object currently used."
2501,get_replica_context,tensorflow/tensorflow/python/distribute/distribution_strategy_context.py,92,function,"Returns the current `tf.distribute.ReplicaContext` or `None`.
Returns `None` if in a cross-replica context.
Note that execution:
1. starts in the default (single-replica) replica context (this function
will return the default `ReplicaContext` object);
2. switches to cross-replica context (in which case this will return
`None`) when entering a `with tf.distribute.Strategy.scope():` block;
3. switches to a (non-default) replica context inside `strategy.run(fn, ...)`;
4. if `fn` calls `get_replica_context().merge_call(merge_fn, ...)`, then
inside `merge_fn` you are back in the cross-replica context (and again
this function will return `None`).
Most `tf.distribute.Strategy` methods may only be executed in
a cross-replica context, in a replica context you should use the
API of the `tf.distribute.ReplicaContext` object returned by this
method instead.
```
assert tf.distribute.get_replica_context() is not None # default
with strategy.scope():
assert tf.distribute.get_replica_context() is None
def f():
replica_context = tf.distribute.get_replica_context() # for strategy
assert replica_context is not None
tf.print(""Replica id: "", replica_context.replica_id_in_sync_group,
"" of "", replica_context.num_replicas_in_sync)
strategy.run(f)
```
Returns:
The current `tf.distribute.ReplicaContext` object when in a replica context
scope, else `None`.
Within a particular block, exactly one of these two things will be true:
* `get_replica_context()` returns non-`None`, or
* `tf.distribute.is_cross_replica_context()` returns True."
2502,get_cross_replica_context,tensorflow/tensorflow/python/distribute/distribution_strategy_context.py,139,function,"Returns the current tf.distribute.Strategy if in a cross-replica context.
DEPRECATED: Please use `in_cross_replica_context()` and
`get_strategy()` instead.
Returns:
Returns the current `tf.distribute.Strategy` object in a cross-replica
context, or `None`.
Exactly one of `get_replica_context()` and `get_cross_replica_context()`
will return `None` in a particular block."
2503,in_cross_replica_context,tensorflow/tensorflow/python/distribute/distribution_strategy_context.py,156,function,"Returns `True` if in a cross-replica context.
See `tf.distribute.get_replica_context` for details.
```
assert not tf.distribute.in_cross_replica_context()
with strategy.scope():
assert tf.distribute.in_cross_replica_context()
def f():
assert not tf.distribute.in_cross_replica_context()
strategy.run(f)
```
Returns:
`True` if in a cross-replica context (`get_replica_context()` returns
`None`), or `False` if in a replica context (`get_replica_context()` returns
non-`None`)."
2504,get_strategy,tensorflow/tensorflow/python/distribute/distribution_strategy_context.py,181,function,"Returns the current `tf.distribute.Strategy` object.
Typically only used in a cross-replica context:
```
if tf.distribute.in_cross_replica_context():
strategy = tf.distribute.get_strategy()
...
```
Returns:
A `tf.distribute.Strategy` object. Inside a `with strategy.scope()` block,
it returns `strategy`, otherwise it returns the default (single-replica)
`tf.distribute.Strategy` object."
2505,has_strategy,tensorflow/tensorflow/python/distribute/distribution_strategy_context.py,201,function,"Return if there is a current non-default `tf.distribute.Strategy`.
```
assert not tf.distribute.has_strategy()
with strategy.scope():
assert tf.distribute.has_strategy()
```
Returns:
True if inside a `with strategy.scope():`."
2506,get_strategy_and_replica_context,tensorflow/tensorflow/python/distribute/distribution_strategy_context.py,216,function,
2507,experimental_set_strategy,tensorflow/tensorflow/python/distribute/distribution_strategy_context.py,222,function,"Set a `tf.distribute.Strategy` as current without `with strategy.scope()`.
```
tf.distribute.experimental_set_strategy(strategy1)
f()
tf.distribute.experimental_set_strategy(strategy2)
g()
tf.distribute.experimental_set_strategy(None)
h()
```
is equivalent to:
```
with strategy1.scope():
f()
with strategy2.scope():
g()
h()
```
In general, you should use the `with strategy.scope():` API, but this
alternative may be convenient in notebooks where you would have to put
each cell in a `with strategy.scope():` block.
Note: This should only be called outside of any TensorFlow scope to
avoid improper nesting.
Args:
strategy: A `tf.distribute.Strategy` object or None.
Raises:
RuntimeError: If called inside a `with strategy.scope():`."
2508,enter_or_assert_strategy,tensorflow/tensorflow/python/distribute/distribution_strategy_context.py,275,function,
2509,init_run_config,tensorflow/tensorflow/python/distribute/estimator_training.py,127,function,Initializes RunConfig for distribution strategies.
2510,should_run_distribute_coordinator,tensorflow/tensorflow/python/distribute/estimator_training.py,181,function,Checks the config to see whether to run distribute coordinator.
2511,train_and_evaluate,tensorflow/tensorflow/python/distribute/estimator_training.py,203,function,"Run distribute coordinator for Estimator's `train_and_evaluate`.
Args:
estimator: An `Estimator` instance to train and evaluate.
train_spec: A `TrainSpec` instance to specify the training specification.
eval_spec: A `EvalSpec` instance to specify the evaluation and export
specification.
executor_cls: the evaluation executor class of Estimator.
Raises:
ValueError: if `distribute_coordinator_mode` is None in RunConfig."
2512,estimator_train,tensorflow/tensorflow/python/distribute/estimator_training.py,295,function,Run distribute coordinator for Estimator's `train` method.
2513,estimator_evaluate,tensorflow/tensorflow/python/distribute/estimator_training.py,344,function,Run distribute coordinator for Estimator's `evaluate` method.
2514,get_distributed_dataset,tensorflow/tensorflow/python/distribute/input_lib.py,61,function,"Returns a distributed dataset from the given tf.data.Dataset instance.
This is a common function that is used by all strategies to return a
distributed dataset. The distributed dataset instance returned is different
depending on if we are in a TF 1 or TF 2 context. The distributed dataset
instances returned differ from each other in the APIs supported by each of
them.
Args:
dataset: a tf.data.Dataset instance.
input_workers: an InputWorkers object which specifies devices on which
iterators should be created.
strategy: a `tf.distribute.Strategy` object, used to run all-reduce to
handle last partial batch.
split_batch_by: Optional integer. If present, we ""split"" each batch of the
dataset by `split_batch_by` value.
input_context: `InputContext` for sharding. Only pass this in for between
graph multi-worker cases where there is only one `input_worker`. In
these cases, we will shard based on the `input_pipeline_id` and
`num_input_pipelines` in the `InputContext`.
Returns:
A distributed dataset instance."
2515,get_distributed_datasets_from_function,tensorflow/tensorflow/python/distribute/input_lib.py,106,function,"Returns a distributed dataset from the given input function.
This is a common function that is used by all strategies to return a
distributed dataset. The distributed dataset instance returned is different
depending on if we are in a TF 1 or TF 2 context. The distributed dataset
instances returned differ from each other in the APIs supported by each of
them.
Args:
dataset_fn: a function that returns a tf.data.Dataset instance.
input_workers: an InputWorkers object which specifies devices on which
iterators should be created.
input_contexts: A list of `InputContext` instances to be passed to call(s)
to `dataset_fn`. Length and order should match worker order in
`worker_device_pairs`.
strategy: a `tf.distribute.Strategy` object, used to run all-reduce to
handle last partial batch.
Returns:
A distributed dataset instance."
2516,DistributedIteratorInterface,tensorflow/tensorflow/python/distribute/input_lib.py,146,class,"An iterator over `tf.distribute.DistributedDataset`.
`tf.distribute.DistributedIterator` is the primary mechanism for enumerating
elements of a `tf.distribute.DistributedDataset`. It supports the Python
Iterator protocol, which means it can be iterated over using a for-loop or by
fetching individual elements explicitly via `get_next()`.
You can create a `tf.distribute.DistributedIterator` by calling `iter` on
a `tf.distribute.DistributedDataset` or creating a python loop over a
`tf.distribute.DistributedDataset`.
Visit the [tutorial](https://www.tensorflow.org/tutorials/distribute/input)
on distributed input for more examples and caveats."
2517,get_next,tensorflow/tensorflow/python/distribute/input_lib.py,163,method,"Returns the next input from the iterator for all replicas.
Example use:
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> dataset = tf.data.Dataset.range(100).batch(2)
>>> dist_dataset = strategy.experimental_distribute_dataset(dataset)
>>> dist_dataset_iterator = iter(dist_dataset)
>>> @tf.function
... def one_step(input):
... return input
>>> step_num = 5
>>> for _ in range(step_num):
... strategy.run(one_step, args=(dist_dataset_iterator.get_next(),))
>>> strategy.experimental_local_results(dist_dataset_iterator.get_next())
(<tf.Tensor: shape=(1,), dtype=int64, numpy=array([10])>,
<tf.Tensor: shape=(1,), dtype=int64, numpy=array([11])>)
Returns:
A single `tf.Tensor` or a `tf.distribute.DistributedValues` which contains
the next input for all replicas.
Raises:
`tf.errors.OutOfRangeError`: If the end of the iterator has been reached."
2518,element_spec,tensorflow/tensorflow/python/distribute/input_lib.py,193,method,"The type specification of an element of `tf.distribute.DistributedIterator`.
Example usage:
>>> global_batch_size = 16
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> dataset = tf.data.Dataset.from_tensors(([1.],[2])).repeat(100).batch(global_batch_size)
>>> distributed_iterator = iter(strategy.experimental_distribute_dataset(dataset))
>>> distributed_iterator.element_spec
(PerReplicaSpec(TensorSpec(shape=(None, 1), dtype=tf.float32, name=None),
TensorSpec(shape=(None, 1), dtype=tf.float32, name=None)),
PerReplicaSpec(TensorSpec(shape=(None, 1), dtype=tf.int32, name=None),
TensorSpec(shape=(None, 1), dtype=tf.int32, name=None)))
Returns:
A nested structure of `tf.TypeSpec` objects matching the structure of an
element of this `tf.distribute.DistributedIterator`. This returned value
is typically a `tf.distribute.DistributedValues` object and specifies the
`tf.TensorSpec` of individual components."
2519,get_next_as_optional,tensorflow/tensorflow/python/distribute/input_lib.py,218,method,"Returns a `tf.experimental.Optional` that contains the next value for all replicas.
If the `tf.distribute.DistributedIterator` has reached the end of the
sequence, the returned `tf.experimental.Optional` will have no value.
Example usage:
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> global_batch_size = 2
>>> steps_per_loop = 2
>>> dataset = tf.data.Dataset.range(10).batch(global_batch_size)
>>> distributed_iterator = iter(
... strategy.experimental_distribute_dataset(dataset))
>>> def step_fn(x):
... # train the model with inputs
... return x
>>> @tf.function
... def train_fn(distributed_iterator):
... for _ in tf.range(steps_per_loop):
... optional_data = distributed_iterator.get_next_as_optional()
... if not optional_data.has_value():
... break
... per_replica_results = strategy.run(step_fn, args=(optional_data.get_value(),))
... tf.print(strategy.experimental_local_results(per_replica_results))
>>> train_fn(distributed_iterator)
... # ([0 1], [2 3])
... # ([4], [])
Returns:
An `tf.experimental.Optional` object representing the next value from the
`tf.distribute.DistributedIterator` (if it has one) or no value."
2520,DistributedDatasetInterface,tensorflow/tensorflow/python/distribute/input_lib.py,258,class,"Represents a dataset distributed among devices and machines.
A `tf.distribute.DistributedDataset` could be thought of as a ""distributed""
dataset. When you use `tf.distribute` API to scale training to multiple
devices or machines, you also need to distribute the input data, which leads
to a `tf.distribute.DistributedDataset` instance, instead of a
`tf.data.Dataset` instance in the non-distributed case. In TF 2.x,
`tf.distribute.DistributedDataset` objects are Python iterables.
Note: `tf.distribute.DistributedDataset` instances are *not* of type
`tf.data.Dataset`. It only supports two usages we will mention below:
iteration and `element_spec`. We don't support any other APIs to transform or
inspect the dataset.
There are two APIs to create a `tf.distribute.DistributedDataset` object:
`tf.distribute.Strategy.experimental_distribute_dataset(dataset)`and
`tf.distribute.Strategy.experimental_distribute_datasets_from_function(dataset_fn)`.
*When to use which?* When you have a `tf.data.Dataset` instance, and the
regular batch splitting (i.e. re-batch the input `tf.data.Dataset` instance
with a new batch size that is equal to the global batch size divided by the
number of replicas in sync) and autosharding (i.e. the
`tf.data.experimental.AutoShardPolicy` options) work for you, use the former
API. Otherwise, if you are *not* using a canonical `tf.data.Dataset` instance,
or you would like to customize the batch splitting or sharding, you can wrap
these logic in a `dataset_fn` and use the latter API. Both API handles
prefetch to device for the user. For more details and examples, follow the
links to the APIs.
There are two main usages of a `DistributedDataset` object:
1. Iterate over it to generate the input for a single device or multiple
devices, which is a `tf.distribute.DistributedValues` instance. To do this,
you can:
* use a pythonic for-loop construct:
>>> global_batch_size = 4
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> dataset = tf.data.Dataset.from_tensors(([1.],[1.])).repeat(4).batch(global_batch_size)
>>> dist_dataset = strategy.experimental_distribute_dataset(dataset)
>>> @tf.function
... def train_step(input):
... features, labels = input
... return labels - 0.3 * features
>>> for x in dist_dataset:
... # train_step trains the model using the dataset elements
... loss = strategy.run(train_step, args=(x,))
... print(""Loss is"", loss)
Loss is PerReplica:{
0: tf.Tensor(
[[0.7]
[0.7]], shape=(2, 1), dtype=float32),
1: tf.Tensor(
[[0.7]
[0.7]], shape=(2, 1), dtype=float32)
}
Placing the loop inside a `tf.function` will give a performance boost.
However `break` and `return` are currently not supported if the loop is
placed inside a `tf.function`. We also don't support placing the loop
inside a `tf.function` when using
`tf.distribute.experimental.MultiWorkerMirroredStrategy` or
`tf.distribute.experimental.TPUStrategy` with multiple workers.
* use `__iter__` to create an explicit iterator, which is of type
`tf.distribute.DistributedIterator`
>>> global_batch_size = 4
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> train_dataset = tf.data.Dataset.from_tensors(([1.],[1.])).repeat(50).batch(global_batch_size)
>>> train_dist_dataset = strategy.experimental_distribute_dataset(train_dataset)
>>> @tf.function
... def distributed_train_step(dataset_inputs):
... def train_step(input):
... loss = tf.constant(0.1)
... return loss
... per_replica_losses = strategy.run(train_step, args=(dataset_inputs,))
... return strategy.reduce(tf.distribute.ReduceOp.SUM, per_replica_losses,axis=None)
>>> EPOCHS = 2
>>> STEPS = 3
>>> for epoch in range(EPOCHS):
... total_loss = 0.0
... num_batches = 0
... dist_dataset_iterator = iter(train_dist_dataset)
... for _ in range(STEPS):
... total_loss += distributed_train_step(next(dist_dataset_iterator))
... num_batches += 1
... average_train_loss = total_loss / num_batches
... template = (""Epoch {}, Loss: {:.4f}"")
... print (template.format(epoch+1, average_train_loss))
Epoch 1, Loss: 0.2000
Epoch 2, Loss: 0.2000
To achieve a performance improvement, you can also wrap the `strategy.run`
call with a `tf.range` inside a `tf.function`. This runs multiple steps in a
`tf.function`. Autograph will convert it to a `tf.while_loop` on the worker.
However, it is less flexible comparing with running a single step inside
`tf.function`. For example, you cannot run things eagerly or arbitrary
python code within the steps.
2. Inspect the `tf.TypeSpec` of the data generated by `DistributedDataset`.
`tf.distribute.DistributedDataset` generates
`tf.distribute.DistributedValues` as input to the devices. If you pass the
input to a `tf.function` and would like to specify the shape and type of
each Tensor argument to the function, you can pass a `tf.TypeSpec` object to
the `input_signature` argument of the `tf.function`. To get the
`tf.TypeSpec` of the input, you can use the `element_spec` property of the
`tf.distribute.DistributedDataset` or `tf.distribute.DistributedIterator`
object.
For example:
>>> global_batch_size = 4
>>> epochs = 1
>>> steps_per_epoch = 1
>>> mirrored_strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> dataset = tf.data.Dataset.from_tensors(([2.])).repeat(100).batch(global_batch_size)
>>> dist_dataset = mirrored_strategy.experimental_distribute_dataset(dataset)
>>> @tf.function(input_signature=[dist_dataset.element_spec])
... def train_step(per_replica_inputs):
... def step_fn(inputs):
... return tf.square(inputs)
... return mirrored_strategy.run(step_fn, args=(per_replica_inputs,))
>>> for _ in range(epochs):
... iterator = iter(dist_dataset)
... for _ in range(steps_per_epoch):
... output = train_step(next(iterator))
... print(output)
PerReplica:{
0: tf.Tensor(
[[4.]
[4.]], shape=(2, 1), dtype=float32),
1: tf.Tensor(
[[4.]
[4.]], shape=(2, 1), dtype=float32)
}
Visit the [tutorial](https://www.tensorflow.org/tutorials/distribute/input)
on distributed input for more examples and caveats."
2521,element_spec,tensorflow/tensorflow/python/distribute/input_lib.py,432,method,"The type specification of an element of this `tf.distribute.DistributedDataset`.
Example usage:
>>> global_batch_size = 16
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> dataset = tf.data.Dataset.from_tensors(([1.],[2])).repeat(100).batch(global_batch_size)
>>> dist_dataset = strategy.experimental_distribute_dataset(dataset)
>>> dist_dataset.element_spec
(PerReplicaSpec(TensorSpec(shape=(None, 1), dtype=tf.float32, name=None),
TensorSpec(shape=(None, 1), dtype=tf.float32, name=None)),
PerReplicaSpec(TensorSpec(shape=(None, 1), dtype=tf.int32, name=None),
TensorSpec(shape=(None, 1), dtype=tf.int32, name=None)))
Returns:
A nested structure of `tf.TypeSpec` objects matching the structure of an
element of this `tf.distribute.DistributedDataset`. This returned value is
typically a `tf.distribute.DistributedValues` object and specifies the
`tf.TensorSpec` of individual components."
2522,reduce,tensorflow/tensorflow/python/distribute/input_lib.py,457,method,
2523,InputWorkers,tensorflow/tensorflow/python/distribute/input_lib.py,462,class,A 1-to-many mapping from input worker devices to compute devices.
2524,num_workers,tensorflow/tensorflow/python/distribute/input_lib.py,478,method,
2525,worker_devices,tensorflow/tensorflow/python/distribute/input_lib.py,482,method,
2526,compute_devices_for_worker,tensorflow/tensorflow/python/distribute/input_lib.py,485,method,
2527,serialize,tensorflow/tensorflow/python/distribute/input_lib.py,495,method,
2528,deserialize,tensorflow/tensorflow/python/distribute/input_lib.py,498,method,
2529,DistributedIteratorBase,tensorflow/tensorflow/python/distribute/input_lib.py,583,class,Common implementation for all input iterators.
2530,next,tensorflow/tensorflow/python/distribute/input_lib.py,612,method,
2531,get_next_as_optional,tensorflow/tensorflow/python/distribute/input_lib.py,624,method,
2532,get_next,tensorflow/tensorflow/python/distribute/input_lib.py,649,method,Returns the next input from the iterator for all replicas.
2533,return_none,tensorflow/tensorflow/python/distribute/input_lib.py,627,method,
2534,return_value,tensorflow/tensorflow/python/distribute/input_lib.py,630,method,Wraps the inputs for replicas in an `tf.experimental.Optional`.
2535,out_of_range_fn,tensorflow/tensorflow/python/distribute/input_lib.py,666,method,This function will throw an OutOfRange error.
2536,DistributedIteratorV1,tensorflow/tensorflow/python/distribute/input_lib.py,699,class,Input Iterator for a distributed dataset.
2537,initialize,tensorflow/tensorflow/python/distribute/input_lib.py,713,method,"Initialize underlying iterators.
Returns:
A list of any initializer ops that should be run."
2538,initializer,tensorflow/tensorflow/python/distribute/input_lib.py,722,method,Returns a list of ops that initialize the iterator.
2539,output_classes,tensorflow/tensorflow/python/distribute/input_lib.py,728,method,
2540,output_shapes,tensorflow/tensorflow/python/distribute/input_lib.py,733,method,
2541,output_types,tensorflow/tensorflow/python/distribute/input_lib.py,738,method,
2542,get_iterator,tensorflow/tensorflow/python/distribute/input_lib.py,742,method,
2543,element_spec,tensorflow/tensorflow/python/distribute/input_lib.py,749,method,The type specification of an element of this iterator.
2544,DistributedIteratorSpec,tensorflow/tensorflow/python/distribute/input_lib.py,754,class,Type specification for `DistributedIterator`.
2545,value_type,tensorflow/tensorflow/python/distribute/input_lib.py,772,method,
2546,most_specific_compatible_type,tensorflow/tensorflow/python/distribute/input_lib.py,786,method,"Returns the most specific TypeSpec compatible with `self` and `other`.
Args:
other: A `TypeSpec`.
Raises:
ValueError: If there is no TypeSpec that is compatible with both `self`
and `other`."
2547,from_value,tensorflow/tensorflow/python/distribute/input_lib.py,836,method,
2548,DistributedIterator,tensorflow/tensorflow/python/distribute/input_lib.py,849,class,Input Iterator for a distributed dataset.
2549,element_spec,tensorflow/tensorflow/python/distribute/input_lib.py,885,method,
2550,DistributedDataset,tensorflow/tensorflow/python/distribute/input_lib.py,935,class,Distributed dataset that supports prefetching to multiple devices.
2551,element_spec,tensorflow/tensorflow/python/distribute/input_lib.py,1038,method,The type specification of an element of this dataset.
2552,DistributedDatasetV1,tensorflow/tensorflow/python/distribute/input_lib.py,1043,class,Distributed dataset that supports prefetching to multiple devices.
2553,make_one_shot_iterator,tensorflow/tensorflow/python/distribute/input_lib.py,1060,method,"Get a one time use iterator for DistributedDatasetV1.
Note: This API is deprecated. Please use `for ... in dataset:` to iterate
over the dataset or `iter` to create an iterator.
Returns:
A DistributedIteratorV1 instance."
2554,make_initializable_iterator,tensorflow/tensorflow/python/distribute/input_lib.py,1081,method,"Get an initializable iterator for DistributedDatasetV1.
Note: This API is deprecated. Please use
`tf.compat.v1.data.make_initializable_iterator(dataset)` to create an
initializable iterator.
Returns:
A DistributedIteratorV1 instance."
2555,DistributedDatasetsFromFunction,tensorflow/tensorflow/python/distribute/input_lib.py,1121,class,Inputs created from dataset function.
2556,element_spec,tensorflow/tensorflow/python/distribute/input_lib.py,1182,method,The type specification of an element of this dataset.
2557,DistributedDatasetsFromFunctionV1,tensorflow/tensorflow/python/distribute/input_lib.py,1187,class,Inputs created from dataset function.
2558,InputFunctionIterator,tensorflow/tensorflow/python/distribute/input_lib.py,1229,class,Iterator created from input function.
2559,DatasetIterator,tensorflow/tensorflow/python/distribute/input_lib.py,1278,class,Iterator created from input dataset.
2560,MultiStepContext,tensorflow/tensorflow/python/distribute/input_lib.py,1798,class,"A context object that can be used to capture things when running steps.
This context object is useful when running multiple steps at a time using the
`experimental_run_steps_on_iterator` API. For e.g. it allows the user's step
function to specify which outputs to emit at what frequency. Currently it
supports capturing output from the last step, as well as capturing non tensor
outputs. In the future it will be augmented to support other use cases such
as output each N steps."
2561,last_step_outputs,tensorflow/tensorflow/python/distribute/input_lib.py,1820,method,"A dictionary consisting of outputs to be captured on last step.
Keys in the dictionary are names of tensors to be captured, as specified
when `set_last_step_output` is called.
Values in the dictionary are the tensors themselves. If
`set_last_step_output` was called with a `reduce_op` for this output,
then the value is the reduced value.
Returns:
A dictionary with last step outputs."
2562,set_last_step_output,tensorflow/tensorflow/python/distribute/input_lib.py,1840,method,"Set `output` with `name` to be outputted from the last step.
Args:
name: String, name to identify the output. Doesn't need to match tensor
name.
output: The tensors that should be outputted with `name`. See below for
actual types supported.
reduce_op: Reduction method to use to reduce outputs from multiple
replicas. Required if `set_last_step_output` is called in a replica
context. Optional in cross_replica_context.
When present, the outputs from all the replicas are reduced using the
current distribution strategy's `reduce` method. Hence, the type of
`output` must be what's supported by the corresponding `reduce` method.
For e.g. if using MirroredStrategy and reduction is set, output
must be a `PerReplica` value.
The reduce method is also recorded in a dictionary
`_last_step_outputs_reduce_ops` for later interpreting of the
outputs as already reduced or not."
2563,non_tensor_outputs,tensorflow/tensorflow/python/distribute/input_lib.py,1882,method,A dictionary consisting of any non tensor outputs to be captured.
2564,set_non_tensor_output,tensorflow/tensorflow/python/distribute/input_lib.py,1886,method,Set `output` with `name` to be captured as a non tensor output.
2565,merge_fn,tensorflow/tensorflow/python/distribute/input_lib.py,1870,method,
2566,merge_fn,tensorflow/tensorflow/python/distribute/input_lib.py,1891,method,
2567,auto_shard_dataset,tensorflow/tensorflow/python/distribute/input_ops.py,30,function,"Shard the input pipeline by sharding the underlying list of files.
Args:
dataset: A `tf.data.Dataset` instance, typically the result of a bunch of
dataset transformations.
num_shards: A `tf.int64` scalar `tf.Tensor`, representing the number of
shards operating in parallel. Same usage as in `tf.data.Dataset.shard`.
index: A `tf.int64` scalar `tf.Tensor`, representing the worker index.
Same usage as in `tf.data.Dataset.shard`.
Returns:
A modified `Dataset` obtained by updating the pipeline sharded by the
files. The input dataset will be returned if we cannot automatically
determine a good way to shard the input dataset."
2568,all_combinations,tensorflow/tensorflow/python/distribute/metrics_v1_test.py,75,function,
2569,tpu_combinations,tensorflow/tensorflow/python/distribute/metrics_v1_test.py,86,function,
2570,MirroredFunctionStrategy,tensorflow/tensorflow/python/distribute/mirrored_function_strategy.py,61,class,"Mirrors vars to distribute across multiple devices and machines.
This strategy uses one replica per device and sync replication for its
multi-GPU version. Unlike `tf.distribute.MirroredStrategy`, it creates a
function for a single replica, and calls that function repeatedly instead of
recording the operations for each replica separately."
2571,MirroredFunctionExtended,tensorflow/tensorflow/python/distribute/mirrored_function_strategy.py,82,class,Implementation of MirroredFunctionStrategy.
2572,FnMergedValue,tensorflow/tensorflow/python/distribute/mirrored_function_strategy.py,146,class,
2573,MirroredFunctionReplicaContext,tensorflow/tensorflow/python/distribute/mirrored_function_strategy.py,164,class,ReplicaContext used in MirroredFunctionStrategy.
2574,devices,tensorflow/tensorflow/python/distribute/mirrored_function_strategy.py,198,method,
2575,call_for_each_replica,tensorflow/tensorflow/python/distribute/mirrored_run.py,45,function,"Call `fn` on each worker devices(replica).
It's highly recommended to wrap the call to this function inside a
`tf.function`, otherwise the performance is poor.
Args:
strategy: `tf.distribute.Strategy`.
fn: function to call on each worker devices.
args: positional arguments to `fn`.
kwargs: keyword arguments to `fn`.
Returns:
Wrapped returned value of `fn` from all replicas."
2576,all_local_devices,tensorflow/tensorflow/python/distribute/mirrored_strategy.py,171,function,
2577,all_devices,tensorflow/tensorflow/python/distribute/mirrored_strategy.py,178,function,
2578,MirroredStrategy,tensorflow/tensorflow/python/distribute/mirrored_strategy.py,188,class,"Synchronous training across multiple replicas on one machine.
This strategy is typically used for training on one
machine with multiple GPUs. For TPUs, use
`tf.distribute.TPUStrategy`. To use `MirroredStrategy` with multiple workers,
please refer to `tf.distribute.experimental.MultiWorkerMirroredStrategy`.
For example, a variable created under a `MirroredStrategy` is a
`MirroredVariable`. If no devices are specified in the constructor argument of
the strategy then it will use all the available GPUs. If no GPUs are found, it
will use the available CPUs. Note that TensorFlow treats all CPUs on a
machine as a single device, and uses threads internally for parallelism.
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> with strategy.scope():
... x = tf.Variable(1.)
>>> x
MirroredVariable:{
0: <tf.Variable ... shape=() dtype=float32, numpy=1.0>,
1: <tf.Variable ... shape=() dtype=float32, numpy=1.0>
}
While using distribution strategies, all the variable creation should be done
within the strategy's scope. This will replicate the variables across all the
replicas and keep them in sync using an all-reduce algorithm.
Variables created inside a `MirroredStrategy` which is wrapped with a
`tf.function` are still `MirroredVariables`.
>>> x = []
>>> @tf.function # Wrap the function with tf.function.
... def create_variable():
... if not x:
... x.append(tf.Variable(1.))
... return x[0]
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> with strategy.scope():
... _ = create_variable()
... print(x[0])
MirroredVariable:{
0: <tf.Variable ... shape=() dtype=float32, numpy=1.0>,
1: <tf.Variable ... shape=() dtype=float32, numpy=1.0>
}
`experimental_distribute_dataset` can be used to distribute the dataset across
the replicas when writing your own training loop. If you are using `.fit` and
`.compile` methods available in `tf.keras`, then `tf.keras` will handle the
distribution for you.
For example:
```python
my_strategy = tf.distribute.MirroredStrategy()
with my_strategy.scope():
@tf.function
def distribute_train_epoch(dataset):
def replica_fn(input):
# process input and return result
return result
total_result = 0
for x in dataset:
per_replica_result = my_strategy.run(replica_fn, args=(x,))
total_result += my_strategy.reduce(tf.distribute.ReduceOp.SUM,
per_replica_result, axis=None)
return total_result
dist_dataset = my_strategy.experimental_distribute_dataset(dataset)
for _ in range(EPOCHS):
train_result = distribute_train_epoch(dist_dataset)
```
Args:
devices: a list of device strings such as `['/gpu:0', '/gpu:1']`. If
`None`, all available GPUs are used. If no GPUs are found, CPU is used.
cross_device_ops: optional, a descedant of `CrossDeviceOps`. If this is not
set, `NcclAllReduce()` will be used by default. One would customize this
if NCCL isn't available or if a special implementation that exploits
the particular hardware is available."
2579,MirroredStrategyV1,tensorflow/tensorflow/python/distribute/mirrored_strategy.py,279,class,
2580,MirroredExtended,tensorflow/tensorflow/python/distribute/mirrored_strategy.py,292,class,Implementation of MirroredStrategy.
2581,read_var,tensorflow/tensorflow/python/distribute/mirrored_strategy.py,687,method,Read the aggregate value of a replica-local variable.
2582,value_container,tensorflow/tensorflow/python/distribute/mirrored_strategy.py,699,method,
2583,worker_devices,tensorflow/tensorflow/python/distribute/mirrored_strategy.py,707,method,
2584,worker_devices_by_replica,tensorflow/tensorflow/python/distribute/mirrored_strategy.py,711,method,
2585,parameter_devices,tensorflow/tensorflow/python/distribute/mirrored_strategy.py,715,method,
2586,experimental_between_graph,tensorflow/tensorflow/python/distribute/mirrored_strategy.py,719,method,
2587,experimental_should_init,tensorflow/tensorflow/python/distribute/mirrored_strategy.py,723,method,
2588,should_checkpoint,tensorflow/tensorflow/python/distribute/mirrored_strategy.py,727,method,
2589,should_save_summary,tensorflow/tensorflow/python/distribute/mirrored_strategy.py,731,method,
2590,non_slot_devices,tensorflow/tensorflow/python/distribute/mirrored_strategy.py,734,method,
2591,body,tensorflow/tensorflow/python/distribute/mirrored_strategy.py,537,method,A wrapper around `fn` to create the while loop body.
2592,initial_value_fn,tensorflow/tensorflow/python/distribute/mirrored_strategy.py,416,method,
2593,one_device_combinations,tensorflow/tensorflow/python/distribute/mirrored_strategy_test.py,278,function,
2594,MockModel,tensorflow/tensorflow/python/distribute/mirrored_strategy_test.py,1015,class,
2595,RemoteSingleWorkerMirroredStrategyGraph,tensorflow/tensorflow/python/distribute/mirrored_strategy_test.py,1256,class,
2596,Process,tensorflow/tensorflow/python/distribute/multi_process_lib.py,35,class,A process simulating a worker for testing multi-worker training.
2597,initialized,tensorflow/tensorflow/python/distribute/multi_process_lib.py,50,function,Returns whether the module is initialized.
2598,MultiProcessRunner,tensorflow/tensorflow/python/distribute/multi_process_runner.py,101,class,"A utility class to start multiple processes to simulate a cluster.
We need to use multiple processes to simulate a cluster in TF 2.0 tests
because TF 2.0 has some process-global data structures that have to be
separated by processes. We also need child processes to test out our fault
tolerance because shutting down a standard TensorFlow server within its
process is not supported.
Note: the main test program that uses this runner class must run main program
via `test_main` defined in this file. Using this runner in non-test binaries
is not supported yet.
This class is not thread-safe. Child processes will inherit TF2 behavior flag."
2599,set_args,tensorflow/tensorflow/python/distribute/multi_process_runner.py,221,method,
2600,start,tensorflow/tensorflow/python/distribute/multi_process_runner.py,294,method,"Starts processes, one for each task in `cluster_spec`.
Note that this is best effort by the applicable multiprocessing library,
and it may take up to seconds for a subprocess to be successfully started."
2601,start_in_process_as,tensorflow/tensorflow/python/distribute/multi_process_runner.py,317,method,"Start the processes, with the specified task run in main process.
This is similar to `start()` except that the task with task_type
`as_task_type` and task_id `as_task_id` is run in the main process.
This method is particularly useful when debugging tool such as `pdb` is
needed in some specific task. Note that since this method is blocking until
that specific task exits, additional actions would need a thread to be
called:
```python
def proc_func():
# user code to be run
import pdb; pdb.set_trace()
def follow_ups():
time.sleep(5)
mpr.start_single_process(
task_type='evaluator',
task_id=0)
mpr = multi_process_runner.MultiProcessRunner(
proc_func,
multi_worker_test_base.create_cluster_spec(
has_chief=True, num_workers=1))
threading.Thread(target=follow_ups).start()
mpr.start_in_process_as(as_task_type='chief', as_task_id=0)
mpr.join()
```
Note that if `list_stdout=True`, the logs/stdout by task
run by the main process is not available in result.stdout.
Args:
as_task_type: The task type to be run in the main process.
as_task_id: The task id to be run in the main process."
2602,start_single_process,tensorflow/tensorflow/python/distribute/multi_process_runner.py,365,method,"Starts a single process.
This starts a process in the cluster with the task type, task id, and the
process function (`proc_func`). If process function is `None`, the function
provided at `__init__` will be used. If `cluster_spec` is `None`, the
cluster spec provided at `__init__` will be used.
TODO(rchao): It is meant that all subprocesses will be updated with the new
cluster spec, but this has yet to be implemented. At this time only the
newly started subprocess picks up this updated cluster spec.
Args:
task_type: The task type.
task_id: The task id.
cluster_spec: The cluster spec to be used on the newly started
process. If `None`, the cluster spec provided at `__init__` will be
used.
proc_func: The process function to be run on the newly started
process. If specified, specify `args` and `kwargs` as well. If `None`,
the function provided at `__init__` will be used.
args: Optional positional arguments to be supplied in `proc_func`.
kwargs: Optional keyword arguments to be supplied in `proc_func`."
2603,get_process_id,tensorflow/tensorflow/python/distribute/multi_process_runner.py,414,method,Returns the subprocess id given the task type and task id.
2604,get_process_exit_code,tensorflow/tensorflow/python/distribute/multi_process_runner.py,419,method,"Returns the subprocess exit code given the task type and task id.
Args:
task_type: The task type.
task_id: The task id.
Returns:
The subprocess exit code; `None` if the subprocess has not exited yet.
Raises:
KeyError: If the corresponding subprocess is not found with `task_type`
and `task_id`."
2605,join,tensorflow/tensorflow/python/distribute/multi_process_runner.py,450,method,"Joins all the processes with timeout.
If any of the subprocesses does not exit approximately after `timeout`
seconds has passed after `join` call, this raises a
`SubprocessTimeoutError`.
Note: At timeout, it uses SIGTERM to terminate the subprocesses, in order to
log the stack traces of the subprocesses when they exit. However, this
results in timeout when the test runs with tsan (thread sanitizer); if tsan
is being run on the test targets that rely on timeout to assert information,
`MultiProcessRunner.terminate_all()` must be called after `join()`, before
the test exits, so the subprocesses are terminated with SIGKILL, and data
race is removed.
Args:
timeout: if set and not all processes report status within roughly
`timeout` seconds, a `SubprocessTimeoutError` exception will be raised.
Returns:
A MultiProcessRunnerResult object, which has two attributes,
`return_value` and `stdout`. `return_value` always contains the return
values from the subprocesses. If `list_stdout` argument is True at
`__init__`, `stdout` is available that contains a list of all messages
from subprocesses' stdout and stderr.
Raises:
SubprocessTimeoutError: if not all processes report status approximately
within `timeout` seconds. When this is raised, a
`MultiProcessRunnerResult` object can be retrieved by
`SubprocessTimeoutError`'s mpr_result attribute, which has the same
structure as above 'Returns' section describes.
UnexpectedSubprocessExitError: If any of the subprocesses did not exit
properly (for example, they exit on SIGTERM or SIGKILL signal). When
this is raised, a `MultiProcessRunnerResult` object can be retrieved by
`UnexpectedSubprocessExitError`'s mpr_result attribute, which has the
same structure as above 'Returns' section describes. If `max_run_time`
is not `None`, it is expected that some subprocesses may be
force-killed when `max_run_time` is up, and this is raised in those
cases.
Exception: if there is an Exception propagated from any subprocess."
2606,terminate,tensorflow/tensorflow/python/distribute/multi_process_runner.py,547,method,Terminates the process with `task_type` and `task_id`.
2607,terminate_all,tensorflow/tensorflow/python/distribute/multi_process_runner.py,556,method,Terminates all subprocesses.
2608,get_manager,tensorflow/tensorflow/python/distribute/multi_process_runner.py,571,method,"Returns the multiprocessing manager object for concurrency tools.
The manager object is useful as it controls a server process that holds
the python objects that can be shared across processes. This can be used
for parent-subprocess communication:
```python
mpr = multi_process_runner.MultiProcessRunner(...)
manager = mpr.get_manager()
some_event_happening_in_subprocess = manager.Event()
mpr.set_args(args=(some_event_happening_in_subprocess,))
mpr.start()
some_event_happening_in_subprocess.wait()
# Do something that only should after some event happens in subprocess.
```
Note that the user of multi_process_runner should not create additional
`multiprocessing.Manager()` objects; doing so can result in segfault in
some cases."
2609,handler,tensorflow/tensorflow/python/distribute/multi_process_runner.py,310,method,
2610,MultiProcessPoolRunner,tensorflow/tensorflow/python/distribute/multi_process_runner.py,730,class,"A utility class to start a process pool to simulate a cluster.
It's similar to MultiProcessRunner, but uses a pool of processes to avoid the
expensive initialization cost of Tensorflow."
2611,shutdown,tensorflow/tensorflow/python/distribute/multi_process_runner.py,760,method,Shuts down the worker pool.
2612,run,tensorflow/tensorflow/python/distribute/multi_process_runner.py,802,method,"Runs `proc_func` with `args` and `kwargs` on all jobs.
Args:
proc_func: The function to be run.
args: Optional positional arguments to be supplied in `proc_func`.
kwargs: Optional keyword arguments to be supplied in `proc_func`.
Returns:
A list of return values."
2613,SubprocessTimeoutError,tensorflow/tensorflow/python/distribute/multi_process_runner.py,908,class,"An error that indicates there is at least one subprocess timing out.
When this is raised, a `MultiProcessRunnerResult` object can be retrieved by
`SubprocessTimeoutError`'s mpr_result attribute. See
`MultiProcessRunner.join()` for more information."
2614,UnexpectedSubprocessExitError,tensorflow/tensorflow/python/distribute/multi_process_runner.py,921,class,"An error indicating there is at least one subprocess with unexpected exit.
When this is raised, a `MultiProcessRunnerResult` object can be retrieved by
`UnexpectedSubprocessExitError`'s mpr_result attribute. See
`MultiProcessRunner.join()` for more information."
2615,run,tensorflow/tensorflow/python/distribute/multi_process_runner.py,948,function,"Runs functions in local child processes.
It is a convenience method that creates a `MultiProcessRunner` object and
invokes `start` and `join` method. Please see these methods for detailed
documentations.
Returns:
A MultiProcessRunnerResult object returned from `MultiProcessRunner.join()`."
2616,barrier,tensorflow/tensorflow/python/distribute/multi_process_runner.py,985,function,
2617,proc_func_that_adds_task_type_in_return_data,tensorflow/tensorflow/python/distribute/multi_process_runner_test.py,36,function,
2618,proc_func_that_errors,tensorflow/tensorflow/python/distribute/multi_process_runner_test.py,40,function,
2619,proc_func_that_does_nothing,tensorflow/tensorflow/python/distribute/multi_process_runner_test.py,44,function,
2620,proc_func_that_adds_simple_return_data,tensorflow/tensorflow/python/distribute/multi_process_runner_test.py,48,function,
2621,proc_func_that_returns_args_and_kwargs,tensorflow/tensorflow/python/distribute/multi_process_runner_test.py,52,function,
2622,proc_func_with_barrier,tensorflow/tensorflow/python/distribute/multi_process_runner_test.py,56,function,
2623,proc_func_that_returns_pid,tensorflow/tensorflow/python/distribute/multi_process_runner_test.py,60,function,
2624,proc_func_that_sets_global,tensorflow/tensorflow/python/distribute/multi_process_runner_test.py,67,function,
2625,pick_unused_port,tensorflow/tensorflow/python/distribute/multi_worker_test_base.py,64,function,Returns an unused and unassigned local port.
2626,create_in_process_cluster,tensorflow/tensorflow/python/distribute/multi_worker_test_base.py,148,function,Create an in-process cluster that consists of only standard server.
2627,create_cluster_spec,tensorflow/tensorflow/python/distribute/multi_worker_test_base.py,205,function,Create a cluster spec with tasks with unused local ports.
2628,skip_if_grpc_server_cant_be_started,tensorflow/tensorflow/python/distribute/multi_worker_test_base.py,233,function,
2629,DummySession,tensorflow/tensorflow/python/distribute/multi_worker_test_base.py,387,class,
2630,MockOsEnv,tensorflow/tensorflow/python/distribute/multi_worker_test_base.py,396,class,A class that allows per-thread TF_CONFIG.
2631,get,tensorflow/tensorflow/python/distribute/multi_worker_test_base.py,404,method,
2632,get_tf_config_task,tensorflow/tensorflow/python/distribute/multi_worker_test_base.py,638,function,
2633,get_tf_config_cluster_spec,tensorflow/tensorflow/python/distribute/multi_worker_test_base.py,642,function,
2634,get_task_type,tensorflow/tensorflow/python/distribute/multi_worker_test_base.py,646,function,
2635,get_task_index,tensorflow/tensorflow/python/distribute/multi_worker_test_base.py,650,function,
2636,is_chief,tensorflow/tensorflow/python/distribute/multi_worker_test_base.py,654,function,
2637,normalize_cluster_spec,tensorflow/tensorflow/python/distribute/multi_worker_util.py,26,function,"Makes `cluster_spec` into a `ClusterSpec` object.
Args:
cluster_spec: a dict, ClusterDef or ClusterSpec object specifying the
cluster configurations.
Returns:
a `ClusterSpec` object.
Raises:
ValueError: if `cluster_spec` is not a dict or a `ClusterSpec` or a
`ClusterDef`."
2638,is_chief,tensorflow/tensorflow/python/distribute/multi_worker_util.py,97,function,"Returns whether the given task is chief in the cluster.
Since there is at most one evaluator and the evaluator itself should be
independent of the training cluster, the evaluator job is also a chief job on
its own.
If this is currently running under a `_WorkerContext` of distribute
coordinator, the arguments can be omitted as the result is already available.
Args:
cluster_spec: a dict, `ClusterDef` or `ClusterSpec` object specifying the
cluster configurations.
task_type: the task type in the cluster.
task_id: the task id in the cluster.
Returns:
a boolean indicating whether the given task is chief.
Raises:
ValueError: if `task_type` is not in the `cluster_spec` or `task_id` exceeds
the maximum id of the `task_type`."
2639,collective_leader,tensorflow/tensorflow/python/distribute/multi_worker_util.py,137,function,"Return the job name for the leader of for collective ops.
Args:
cluster_spec: a dict, `ClusterDef` or `ClusterSpec` object specifying the
cluster configurations.
task_type: the task type in the cluster.
task_id: the task id in the cluster.
Returns:
a string indicating the leader job name or empty string if no need to set
leader job."
2640,worker_count,tensorflow/tensorflow/python/distribute/multi_worker_util.py,171,function,Returns the number of workers in the cluster.
2641,id_in_cluster,tensorflow/tensorflow/python/distribute/multi_worker_util.py,192,function,"Returns a unique id for the task in the `task_type`'s cluster.
It returns an id ranging from [0, `worker_count(task_type, task_id)`).
Note: this function assumes that ""evaluate"" job is in its own cluster or its
own partition of a cluster.
Args:
cluster_spec: a dict, `ClusterDef` or `ClusterSpec` object to be validated.
task_type: string indicating the type of the task.
task_id: the id of the `task_type` in this cluster.
Returns:
an int indicating the unique id.
Throws:
ValueError: if `task_type` is not ""chief"", ""worker"" or ""evaluator""."
2642,should_save_checkpoint,tensorflow/tensorflow/python/distribute/multi_worker_util.py,230,function,"Returns whether the current worker should save checkpoints.
In multi-worker training, if saving checkpoint is requested by user, or needed
for fault-tolerance, the cluster should save checkpoint but not necessarily
every worker in the cluster should.
TODO(rchao): Consider generalizing this util to be `should_save_file` as there
can be other files to save such as summary.
Returns:
Whether this particular worker in the cluster should save checkpoints."
2643,should_load_checkpoint,tensorflow/tensorflow/python/distribute/multi_worker_util.py,246,function,"Returns whether the current worker should load checkpoints.
In multi-worker training, if loading checkpoint is requested by user, or
needed for fault-tolerance, the cluster should load checkpoint but not
necessarily every worker in the cluster should.
Returns:
Whether this particular worker in the cluster should load checkpoints."
2644,wait_for_other_workers,tensorflow/tensorflow/python/distribute/multi_worker_util.py,259,function,Waits for other workers to reach the same call to this method.
2645,has_worker_context,tensorflow/tensorflow/python/distribute/multi_worker_util.py,264,function,Returns whether a worker context has been entered.
2646,init_var_from_numpy,tensorflow/tensorflow/python/distribute/numpy_dataset.py,32,function,Initialize `input_var` to `numpy_input` using `session` in graph mode.
2647,one_host_numpy_dataset,tensorflow/tensorflow/python/distribute/numpy_dataset.py,76,function,Create a dataset on `colocate_with` from `numpy_input`.
2648,SingleDevice,tensorflow/tensorflow/python/distribute/numpy_dataset.py,94,class,Used with `colocate_with` to create a non-mirrored variable.
2649,OneDeviceStrategy,tensorflow/tensorflow/python/distribute/one_device_strategy.py,41,class,"A distribution strategy for running on a single device.
Using this strategy will place any variables created in its scope on the
specified device. Input distributed through this strategy will be
prefetched to the specified device. Moreover, any functions called via
`strategy.run` will also be placed on the specified device
as well.
Typical usage of this strategy could be testing your code with the
tf.distribute.Strategy API before switching to other strategies which
actually distribute to multiple devices/machines.
For example:
```
strategy = tf.distribute.OneDeviceStrategy(device=""/gpu:0"")
with strategy.scope():
v = tf.Variable(1.0)
print(v.device) # /job:localhost/replica:0/task:0/device:GPU:0
def step_fn(x):
return x * 2
result = 0
for i in range(10):
result += strategy.run(step_fn, args=(i,))
print(result) # 90
```"
2650,experimental_distribute_dataset,tensorflow/tensorflow/python/distribute/one_device_strategy.py,84,method,"Distributes a tf.data.Dataset instance provided via dataset.
In this case, there is only one device, so this is only a thin wrapper
around the input dataset. It will, however, prefetch the input data to the
specified device. The returned distributed dataset can be iterated over
similar to how regular datasets can.
NOTE: Currently, the user cannot add any more transformations to a
distributed dataset.
Example:
```
strategy = tf.distribute.OneDeviceStrategy()
dataset = tf.data.Dataset.range(10).batch(2)
dist_dataset = strategy.experimental_distribute_dataset(dataset)
for x in dist_dataset:
print(x) # [0, 1], [2, 3],...
```
Args:
dataset: `tf.data.Dataset` to be prefetched to device.
options: `tf.distribute.InputOptions` used to control options on how this
dataset is distributed.
Returns:
A ""distributed `Dataset`"" that the caller can iterate over."
2651,experimental_distribute_datasets_from_function,tensorflow/tensorflow/python/distribute/one_device_strategy.py,113,method,"Distributes `tf.data.Dataset` instances created by calls to `dataset_fn`.
`dataset_fn` will be called once for each worker in the strategy. In this
case, we only have one worker and one device so `dataset_fn` is called
once.
The `dataset_fn` should take an `tf.distribute.InputContext` instance where
information about batching and input replication can be accessed:
```
def dataset_fn(input_context):
batch_size = input_context.get_per_replica_batch_size(global_batch_size)
d = tf.data.Dataset.from_tensors([[1.]]).repeat().batch(batch_size)
return d.shard(
input_context.num_input_pipelines, input_context.input_pipeline_id)
inputs = strategy.experimental_distribute_datasets_from_function(dataset_fn)
for batch in inputs:
replica_results = strategy.run(replica_fn, args=(batch,))
```
IMPORTANT: The `tf.data.Dataset` returned by `dataset_fn` should have a
per-replica batch size, unlike `experimental_distribute_dataset`, which uses
the global batch size. This may be computed using
`input_context.get_per_replica_batch_size`.
Args:
dataset_fn: A function taking a `tf.distribute.InputContext` instance and
returning a `tf.data.Dataset`.
options: `tf.distribute.InputOptions` used to control options on how this
dataset is distributed.
Returns:
A ""distributed `Dataset`"", which the caller can iterate over like regular
datasets."
2652,experimental_local_results,tensorflow/tensorflow/python/distribute/one_device_strategy.py,156,method,"Returns the list of all local per-replica values contained in `value`.
In `OneDeviceStrategy`, the `value` is always expected to be a single
value, so the result is just the value in a tuple.
Args:
value: A value returned by `experimental_run()`, `run()`,
`extended.call_for_each_replica()`, or a variable created in `scope`.
Returns:
A tuple of values contained in `value`. If `value` represents a single
value, this returns `(value,).`"
2653,run,tensorflow/tensorflow/python/distribute/one_device_strategy.py,172,method,"Run `fn` on each replica, with the given arguments.
In `OneDeviceStrategy`, `fn` is simply called within a device scope for the
given device, with the provided arguments.
Args:
fn: The function to run. The output must be a `tf.nest` of `Tensor`s.
args: (Optional) Positional arguments to `fn`.
kwargs: (Optional) Keyword arguments to `fn`.
options: (Optional) An instance of `tf.distribute.RunOptions` specifying
the options to run `fn`.
Returns:
Return value from running `fn`."
2654,reduce,tensorflow/tensorflow/python/distribute/one_device_strategy.py,190,method,"Reduce `value` across replicas.
In `OneDeviceStrategy`, there is only one replica, so if axis=None, value
is simply returned. If axis is specified as something other than None,
such as axis=0, value is reduced along that axis and returned.
Example:
```
t = tf.range(10)
result = strategy.reduce(tf.distribute.ReduceOp.SUM, t, axis=None).numpy()
# result: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
result = strategy.reduce(tf.distribute.ReduceOp.SUM, t, axis=0).numpy()
# result: 45
```
Args:
reduce_op: A `tf.distribute.ReduceOp` value specifying how values should
be combined.
value: A ""per replica"" value, e.g. returned by `run` to
be combined into a single tensor.
axis: Specifies the dimension to reduce along within each
replica's tensor. Should typically be set to the batch dimension, or
`None` to only reduce across replicas (e.g. if the tensor has no batch
dimension).
Returns:
A `Tensor`."
2655,scope,tensorflow/tensorflow/python/distribute/one_device_strategy.py,223,method,"Returns a context manager selecting this Strategy as current.
Inside a `with strategy.scope():` code block, this thread
will use a variable creator set by `strategy`, and will
enter its ""cross-replica context"".
In `OneDeviceStrategy`, all variables created inside `strategy.scope()`
will be on `device` specified at strategy construction time.
See example in the docs for this class.
Returns:
A context manager to use for creating variables with this strategy."
2656,OneDeviceStrategyV1,tensorflow/tensorflow/python/distribute/one_device_strategy.py,241,class,
2657,OneDeviceExtended,tensorflow/tensorflow/python/distribute/one_device_strategy.py,255,class,Implementation of OneDeviceStrategy.
2658,read_var,tensorflow/tensorflow/python/distribute/one_device_strategy.py,401,method,Read the aggregate value of a replica-local variable.
2659,value_container,tensorflow/tensorflow/python/distribute/one_device_strategy.py,408,method,
2660,worker_devices,tensorflow/tensorflow/python/distribute/one_device_strategy.py,420,method,
2661,parameter_devices,tensorflow/tensorflow/python/distribute/one_device_strategy.py,424,method,
2662,non_slot_devices,tensorflow/tensorflow/python/distribute/one_device_strategy.py,427,method,
2663,experimental_should_init,tensorflow/tensorflow/python/distribute/one_device_strategy.py,432,method,
2664,experimental_between_graph,tensorflow/tensorflow/python/distribute/one_device_strategy.py,436,method,
2665,should_checkpoint,tensorflow/tensorflow/python/distribute/one_device_strategy.py,440,method,
2666,should_save_summary,tensorflow/tensorflow/python/distribute/one_device_strategy.py,444,method,
2667,body,tensorflow/tensorflow/python/distribute/one_device_strategy.py,342,method,A wrapper around `fn` to create the while loop body.
2668,PackedDistributedVariable,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,28,class,"A variable which packs multiple variables distributed across devices.
It's only supported when eager execution is enabled.
For op-by-op execution, use an unpacked handle on the current device; for
function execution, use the packed handle to reduce the overhead of function
calls."
2669,devices,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,84,method,
2670,on_device,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,87,method,
2671,get_var_on_device,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,90,method,
2672,get_var_on_current_device,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,96,method,
2673,initial_value,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,100,method,Returns the Tensor used as the initial value for the variable.
2674,handle,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,105,method,
2675,packed_handle,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,112,method,
2676,value,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,121,method,
2677,is_initialized,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,124,method,
2678,assign_sub,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,151,method,
2679,assign_add,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,160,method,
2680,assign,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,169,method,
2681,scatter_sub,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,178,method,
2682,scatter_add,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,186,method,
2683,scatter_mul,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,194,method,
2684,scatter_div,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,202,method,
2685,scatter_min,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,210,method,
2686,scatter_max,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,218,method,
2687,scatter_update,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,226,method,
2688,PackedVarAndDevice,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,247,class,Holds a packed distributed variable and a device.
2689,var,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,257,method,
2690,value,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,260,method,
2691,read_value,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,264,method,
2692,initial_value,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,269,method,
2693,initialized_value,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,272,method,
2694,device,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,277,method,
2695,handle,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,281,method,
2696,op,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,286,method,
2697,assign_sub,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,290,method,
2698,assign_add,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,294,method,
2699,assign,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,298,method,
2700,scatter_sub,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,302,method,
2701,scatter_add,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,306,method,
2702,scatter_mul,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,310,method,
2703,scatter_div,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,314,method,
2704,scatter_min,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,318,method,
2705,scatter_max,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,322,method,
2706,scatter_update,tensorflow/tensorflow/python/distribute/packed_distributed_variable.py,326,method,
2707,ParameterServerStrategy,tensorflow/tensorflow/python/distribute/parameter_server_strategy.py,52,class,"An asynchronous multi-worker parameter server tf.distribute strategy.
This strategy requires two roles: workers and parameter servers. Variables and
updates to those variables will be assigned to parameter servers and other
operations are assigned to workers.
When each worker has more than one GPU, operations will be replicated on all
GPUs. Even though operations may be replicated, variables are not and each
worker shares a common view for which parameter server a variable is assigned
to.
By default it uses `TFConfigClusterResolver` to detect configurations for
multi-worker training. This requires a 'TF_CONFIG' environment variable and
the 'TF_CONFIG' must have a cluster spec.
This class assumes each worker is running the same code independently, but
parameter servers are running a standard server. This means that while each
worker will synchronously compute a single gradient update across all GPUs,
updates between workers proceed asynchronously. Operations that occur only on
the first replica (such as incrementing the global step), will occur on the
first replica *of every worker*.
It is expected to call `call_for_each_replica(fn, ...)` for any
operations which potentially can be replicated across replicas (i.e. multiple
GPUs) even if there is only CPU or one GPU. When defining the `fn`, extra
caution needs to be taken:
1) It is generally not recommended to open a device scope under the strategy's
scope. A device scope (i.e. calling `tf.device`) will be merged with or
override the device for operations but will not change the device for
variables.
2) It is also not recommended to open a colocation scope (i.e. calling
`tf.compat.v1.colocate_with`) under the strategy's scope. For colocating
variables, use `strategy.extended.colocate_vars_with` instead. Colocation of
ops will possibly create device assignment conflicts.
Note: This strategy only works with the Estimator API. Pass an instance of
this strategy to the `experimental_distribute` argument when you create the
`RunConfig`. This instance of `RunConfig` should then be passed to the
`Estimator` instance on which `train_and_evaluate` is called.
For Example:
```
strategy = tf.distribute.experimental.ParameterServerStrategy()
run_config = tf.estimator.RunConfig(
experimental_distribute.train_distribute=strategy)
estimator = tf.estimator.Estimator(config=run_config)
tf.estimator.train_and_evaluate(estimator,...)
```"
2708,experimental_distribute_dataset,tensorflow/tensorflow/python/distribute/parameter_server_strategy.py,125,method,
2709,experimental_distribute_datasets_from_function,tensorflow/tensorflow/python/distribute/parameter_server_strategy.py,131,method,
2710,run,tensorflow/tensorflow/python/distribute/parameter_server_strategy.py,138,method,
2711,scope,tensorflow/tensorflow/python/distribute/parameter_server_strategy.py,143,method,
2712,ParameterServerStrategyV1,tensorflow/tensorflow/python/distribute/parameter_server_strategy.py,154,class,
2713,ParameterServerStrategyExtended,tensorflow/tensorflow/python/distribute/parameter_server_strategy.py,172,class,Implementation of ParameterServerStrategy and CentralStorageStrategy.
2714,value_container,tensorflow/tensorflow/python/distribute/parameter_server_strategy.py,586,method,
2715,read_var,tensorflow/tensorflow/python/distribute/parameter_server_strategy.py,594,method,
2716,worker_devices,tensorflow/tensorflow/python/distribute/parameter_server_strategy.py,663,method,
2717,worker_devices_by_replica,tensorflow/tensorflow/python/distribute/parameter_server_strategy.py,667,method,
2718,parameter_devices,tensorflow/tensorflow/python/distribute/parameter_server_strategy.py,671,method,
2719,non_slot_devices,tensorflow/tensorflow/python/distribute/parameter_server_strategy.py,674,method,
2720,experimental_between_graph,tensorflow/tensorflow/python/distribute/parameter_server_strategy.py,678,method,
2721,experimental_should_init,tensorflow/tensorflow/python/distribute/parameter_server_strategy.py,683,method,
2722,should_checkpoint,tensorflow/tensorflow/python/distribute/parameter_server_strategy.py,687,method,
2723,should_save_summary,tensorflow/tensorflow/python/distribute/parameter_server_strategy.py,691,method,
2724,var_creator,tensorflow/tensorflow/python/distribute/parameter_server_strategy.py,450,method,Create an AggregatingVariable and fix up collections.
2725,AggregatingVariable,tensorflow/tensorflow/python/distribute/ps_values.py,35,class,A wrapper around a variable that aggregates updates across replicas.
2726,get,tensorflow/tensorflow/python/distribute/ps_values.py,46,method,
2727,distribute_strategy,tensorflow/tensorflow/python/distribute/ps_values.py,50,method,
2728,assign_sub,tensorflow/tensorflow/python/distribute/ps_values.py,100,method,
2729,assign_add,tensorflow/tensorflow/python/distribute/ps_values.py,104,method,
2730,assign,tensorflow/tensorflow/python/distribute/ps_values.py,108,method,
2731,initializer,tensorflow/tensorflow/python/distribute/ps_values.py,113,method,
2732,initialized_value,tensorflow/tensorflow/python/distribute/ps_values.py,116,method,
2733,initial_value,tensorflow/tensorflow/python/distribute/ps_values.py,120,method,
2734,op,tensorflow/tensorflow/python/distribute/ps_values.py,124,method,
2735,read_value,tensorflow/tensorflow/python/distribute/ps_values.py,127,method,
2736,eval,tensorflow/tensorflow/python/distribute/ps_values.py,130,method,
2737,graph,tensorflow/tensorflow/python/distribute/ps_values.py,134,method,
2738,device,tensorflow/tensorflow/python/distribute/ps_values.py,138,method,
2739,shape,tensorflow/tensorflow/python/distribute/ps_values.py,142,method,
2740,aggregation,tensorflow/tensorflow/python/distribute/ps_values.py,146,method,
2741,synchronization,tensorflow/tensorflow/python/distribute/ps_values.py,150,method,
2742,name,tensorflow/tensorflow/python/distribute/ps_values.py,154,method,
2743,trainable,tensorflow/tensorflow/python/distribute/ps_values.py,158,method,
2744,dtype,tensorflow/tensorflow/python/distribute/ps_values.py,162,method,
2745,merge_fn,tensorflow/tensorflow/python/distribute/ps_values.py,80,method,
2746,ReduceOp,tensorflow/tensorflow/python/distribute/reduce_util.py,28,class,"Indicates how a set of values should be reduced.
* `SUM`: Add all the values.
* `MEAN`: Take the arithmetic mean (""average"") of the values."
2747,from_variable_aggregation,tensorflow/tensorflow/python/distribute/reduce_util.py,41,method,
2748,get_gpus,tensorflow/tensorflow/python/distribute/remote_mirrored_strategy_eager_test.py,29,function,
2749,RemoteSingleWorkerMirroredStrategyEager,tensorflow/tensorflow/python/distribute/remote_mirrored_strategy_eager_test.py,48,class,
2750,ShardedVariable,tensorflow/tensorflow/python/distribute/sharded_variable.py,30,class,"A container for `Variables` that should be treated as shards.
Variables that are too large to fit on a single device (e.g., large
embeddings)
may need to be sharded over multiple devices. This class maintains a list of
smaller variables that can be independently stored on separate devices (eg,
multiple parameter servers), and saves and restores those variables as if they
were a single larger variable.
Objects of this class can be saved with a given number of shards and then
restored from a checkpoint into a different number of shards.
Objects of this class can be saved to SavedModel format using
`tf.saved_model.save`. The SavedModel can be used by programs like TF serving
APIs. It is not yet supported to load the SavedModel with
`tf.saved_model.load`.
Since `ShardedVariable` can be saved and then restored to different number of
shards depending on the restore environments, for example, TF serving APIs
would restore to one shard for serving efficiency, when using
`ShardedVariable` in a tf.function, one should generally not assume it has the
same number of shards across save and load.
Sharding is only supported along the first dimension.
>>> class Model(tf.Module):
... def __init__(self):
... self.sharded_variable = ShardedVariable([
... tf.Variable([3.0], dtype=tf.float32),
... tf.Variable([2.0], dtype=tf.float32)
... ])
...
... @tf.function(input_signature=[tf.TensorSpec([], dtype=tf.int32)])
... def fn(self, x):
... return tf.nn.embedding_lookup(self.sharded_variable.variables, x)
...
... @tf.function(input_signature=[tf.TensorSpec([], dtype=tf.int32)])
... def serve_fn(self, x):
... return tf.nn.embedding_lookup(self.sharded_variable.variables, x)
>>>
>>> model = Model()
>>> model.fn(1).numpy()
2.0
>>> tf.saved_model.save(model, export_dir='/tmp/saved_model',
... signatures=model.serve_fn)"
2751,variables,tensorflow/tensorflow/python/distribute/sharded_variable.py,144,method,The list of `Variable`s that make up the shards of this object.
2752,name,tensorflow/tensorflow/python/distribute/sharded_variable.py,151,method,The name of this object. Used for checkpointing.
2753,dtype,tensorflow/tensorflow/python/distribute/sharded_variable.py,156,method,The dtype of all `Variable`s in this object.
2754,shape,tensorflow/tensorflow/python/distribute/sharded_variable.py,161,method,"The overall shape, combining all shards along axis `0`."
2755,make_fn,tensorflow/tensorflow/python/distribute/shared_variable_creator.py,38,function,"Construct the variable creator function for device `device_id`.
Constructs custom variable creator functions for the given device.
On first device (device_id == 0), it creates the variable using the
`next_creator`, and stores it in the provided `shared_variable_store`.
On all other devices (device_id > 0), it tries to re-use the variable
already created with the same name. If no such variable exists, it throws an
error.
Additionally, we de-uniquify variable names before checking for matches. This
helps re-use variables which are intended to be the same but have different
names due to variable uniquification happening upstream. Since this might
mean we may have multiple variables with the same canonical name, we store
them in a list per canonical name and return them in the same order as well.
Args:
shared_variable_store: A dictionary that we will use to store variables
created on the first device, and re-used by creators for other devices.
device_id: Integer index of the device whose creator should be
constructed.
Returns:
An appropriate creator function based on device_id."
2756,single_loss_example,tensorflow/tensorflow/python/distribute/single_loss_example.py,32,function,Build a very simple network to use in tests and examples.
2757,minimize_loss_example,tensorflow/tensorflow/python/distribute/single_loss_example.py,54,function,Example of non-distribution-aware legacy code.
2758,batchnorm_example,tensorflow/tensorflow/python/distribute/single_loss_example.py,82,function,Example of non-distribution-aware legacy code with batch normalization.
2759,Step,tensorflow/tensorflow/python/distribute/step_fn.py,25,class,Interface for performing each step of a training algorithm.
2760,distribution,tensorflow/tensorflow/python/distribute/step_fn.py,32,method,
2761,initialize,tensorflow/tensorflow/python/distribute/step_fn.py,35,method,
2762,StandardInputStep,tensorflow/tensorflow/python/distribute/step_fn.py,45,class,"Step with a standard implementation of input handling.
Args:
dataset_fn: a function that returns a tf.data Dataset that produces the
input for the model."
2763,initialize,tensorflow/tensorflow/python/distribute/step_fn.py,57,method,
2764,StandardSingleLossStep,tensorflow/tensorflow/python/distribute/step_fn.py,61,class,"A step function that implements a training step for a feed forward network.
An instance of this class is intended to be used as a callable:
```python
...
step = step_fn.StandardSingleLossStep(
dataset, loss_fn, optimizer, distribution)
# Run a single training step on a given DistributionStrategy:
step(distribution)
...
```
Args:
dataset_fn: a function that returns a tf.data Dataset that produces the
input for the model.
loss_fn: a function that takes a context and inputs as arguments. It returns
the loss for those inputs. `context` is an instance of
`values.MultiStepContext` that will be passed when `loss_fn` is run.
`context` can be used to specify the outputs to be returned from
`loss_fn`, among other things.
optimizer: an optimizer that implements an update rule.
distribution: a `DistributionStrategy` object."
2765,step_fn,tensorflow/tensorflow/python/distribute/step_fn.py,97,method,Function to run one iteration with one input.
2766,set_virtual_cpus_to_at_least,tensorflow/tensorflow/python/distribute/strategy_combinations.py,253,function,Create virtual CPU devices if they haven't yet been created.
2767,strategy_minus_tpu_combinations,tensorflow/tensorflow/python/distribute/strategy_combinations.py,332,function,
2768,tpu_strategy_combinations,tensorflow/tensorflow/python/distribute/strategy_combinations.py,337,function,
2769,all_strategy_combinations,tensorflow/tensorflow/python/distribute/strategy_combinations.py,341,function,
2770,all_strategy_minus_default_and_tpu_combinations,tensorflow/tensorflow/python/distribute/strategy_combinations.py,345,function,
2771,all_strategy_combinations_minus_default,tensorflow/tensorflow/python/distribute/strategy_combinations.py,354,function,
2772,create_variable_like_keras_layer,tensorflow/tensorflow/python/distribute/strategy_test_lib.py,126,function,Utitlity for create variables that works like variable in keras layer.
2773,is_optimizer_v2_instance,tensorflow/tensorflow/python/distribute/strategy_test_lib.py,134,function,
2774,RemoteSingleWorkerMirroredStrategyBase,tensorflow/tensorflow/python/distribute/strategy_test_lib.py,721,class,Tests for a Remote single worker.
2775,fn,tensorflow/tensorflow/python/distribute/strategy_test_lib.py,770,method,
2776,skip_summary,tensorflow/tensorflow/python/distribute/summary_op_util.py,27,function,"Determines if summary should be skipped.
If using multiple replicas in distributed strategy, skip summaries on all
replicas except the first one (replica_id=0).
Returns:
True if the summary is skipped; False otherwise."
2777,gather,tensorflow/tensorflow/python/distribute/test_util.py,33,function,"Gathers value from all workers.
This is intended for tests before we implement an official all-gather API.
Args:
strategy: a `tf.distribute.Strategy`.
value: a nested structure of n-dim `tf.distribute.DistributedValue` of
`tf.Tensor`, or of a `tf.Tensor` if the strategy only has one replica.
Cannot contain tf.sparse.SparseTensor.
Returns:
a (n+1)-dim `tf.Tensor`."
2778,maybe_init_scope,tensorflow/tensorflow/python/distribute/tpu_strategy.py,71,function,
2779,validate_run_function,tensorflow/tensorflow/python/distribute/tpu_strategy.py,79,function,Validate the function passed into strategy.run.
2780,TPUStrategyV2,tensorflow/tensorflow/python/distribute/tpu_strategy.py,105,class,"Synchronous training on TPUs and TPU Pods.
To construct a TPUStrategy object, you need to run the
initialization code as below:
>>> resolver = tf.distribute.cluster_resolver.TPUClusterResolver(tpu='')
>>> tf.config.experimental_connect_to_cluster(resolver)
>>> tf.tpu.experimental.initialize_tpu_system(resolver)
>>> strategy = tf.distribute.TPUStrategy(resolver)
While using distribution strategies, the variables created within the
strategy's scope will be replicated across all the replicas and can be kept in
sync using all-reduce algorithms.
To run TF2 programs on TPUs, you can either use `.compile` and
`.fit` APIs in `tf.keras` with TPUStrategy, or write your own customized
training loop by calling `strategy.run` directly. Note that
TPUStrategy doesn't support pure eager execution, so please make sure the
function passed into `strategy.run` is a `tf.function` or
`strategy.run` is called inside a `tf.function` if eager
behavior is enabled. See more details in https://www.tensorflow.org/guide/tpu.
`experimental_distribute_datasets_from_function` and
`experimental_distribute_dataset` APIs can be used to distribute the dataset
across the TPU workers when writing your own training loop. If you are using
`fit` and `compile` methods available in `tf.keras.Model`, then Keras will
handle the distribution for you.
An example of writing customized training loop on TPUs:
>>> with strategy.scope():
... model = tf.keras.Sequential([
... tf.keras.layers.Dense(2, input_shape=(5,)),
... ])
... optimizer = tf.keras.optimizers.SGD(learning_rate=0.1)
>>> def dataset_fn(ctx):
... x = np.random.random((2, 5)).astype(np.float32)
... y = np.random.randint(2, size=(2, 1))
... dataset = tf.data.Dataset.from_tensor_slices((x, y))
... return dataset.repeat().batch(1, drop_remainder=True)
>>> dist_dataset = strategy.experimental_distribute_datasets_from_function(
... dataset_fn)
>>> iterator = iter(dist_dataset)
>>> @tf.function()
... def train_step(iterator):
...
... def step_fn(inputs):
... features, labels = inputs
... with tf.GradientTape() as tape:
... logits = model(features, training=True)
... loss = tf.keras.losses.sparse_categorical_crossentropy(
... labels, logits)
...
... grads = tape.gradient(loss, model.trainable_variables)
... optimizer.apply_gradients(zip(grads, model.trainable_variables))
...
... strategy.run(step_fn, args=(next(iterator),))
>>> train_step(iterator)
For the advanced use cases like model parallelism, you can set
`experimental_device_assignment` argument when creating TPUStrategy to specify
number of replicas and number of logical devices. Below is an example to
initialize TPU system with 2 logical devices and 1 replica.
>>> resolver = tf.distribute.cluster_resolver.TPUClusterResolver(tpu='')
>>> tf.config.experimental_connect_to_cluster(resolver)
>>> topology = tf.tpu.experimental.initialize_tpu_system(resolver)
>>> device_assignment = tf.tpu.experimental.DeviceAssignment.build(
... topology,
... computation_shape=[1, 1, 1, 2],
... num_replicas=1)
>>> strategy = tf.distribute.TPUStrategy(
... resolver, experimental_device_assignment=device_assignment)
Then you can run a `tf.add` operation only on logical device 0.
>>> @tf.function()
... def step_fn(inputs):
... features, _ = inputs
... output = tf.add(features, features)
...
... # Add operation will be executed on logical device 0.
... output = strategy.experimental_assign_to_logical_device(output, 0)
... return output
>>> dist_dataset = strategy.experimental_distribute_datasets_from_function(
... dataset_fn)
>>> iterator = iter(dist_dataset)
>>> strategy.run(step_fn, args=(next(iterator),))"
2781,run,tensorflow/tensorflow/python/distribute/tpu_strategy.py,225,method,"Run the computation defined by `fn` on each TPU replica.
Executes ops specified by `fn` on each replica. If `args` or `kwargs` have
`tf.distribute.DistributedValues`, such as those produced by a
`tf.distribute.DistributedDataset` from
`tf.distribute.Strategy.experimental_distribute_dataset` or
`tf.distribute.Strategy.experimental_distribute_datasets_from_function`,
when `fn` is executed on a particular replica, it will be executed with the
component of `tf.distribute.DistributedValues` that correspond to that
replica.
`fn` may call `tf.distribute.get_replica_context()` to access members such
as `all_reduce`.
All arguments in `args` or `kwargs` should either be nest of tensors or
`tf.distribute.DistributedValues` containing tensors or composite tensors.
Example usage:
>>> resolver = tf.distribute.cluster_resolver.TPUClusterResolver(tpu='')
>>> tf.config.experimental_connect_to_cluster(resolver)
>>> tf.tpu.experimental.initialize_tpu_system(resolver)
>>> strategy = tf.distribute.TPUStrategy(resolver)
>>> @tf.function
... def run():
... def value_fn(value_context):
... return value_context.num_replicas_in_sync
... distributed_values = (
... strategy.experimental_distribute_values_from_function(value_fn))
... def replica_fn(input):
... return input * 2
... return strategy.run(replica_fn, args=(distributed_values,))
>>> result = run()
Args:
fn: The function to run. The output must be a `tf.nest` of `Tensor`s.
args: (Optional) Positional arguments to `fn`.
kwargs: (Optional) Keyword arguments to `fn`.
options: (Optional) An instance of `tf.distribute.RunOptions` specifying
the options to run `fn`.
Returns:
Merged return value of `fn` across replicas. The structure of the return
value is the same as the return value from `fn`. Each element in the
structure can either be `tf.distribute.DistributedValues`, `Tensor`
objects, or `Tensor`s (for example, if running on a single replica)."
2782,TPUStrategy,tensorflow/tensorflow/python/distribute/tpu_strategy.py,284,class,"Synchronous training on TPUs and TPU Pods.
To construct a TPUStrategy object, you need to run the
initialization code as below:
>>> resolver = tf.distribute.cluster_resolver.TPUClusterResolver(tpu='')
>>> tf.config.experimental_connect_to_cluster(resolver)
>>> tf.tpu.experimental.initialize_tpu_system(resolver)
>>> strategy = tf.distribute.experimental.TPUStrategy(resolver)
While using distribution strategies, the variables created within the
strategy's scope will be replicated across all the replicas and can be kept in
sync using all-reduce algorithms.
To run TF2 programs on TPUs, you can either use `.compile` and
`.fit` APIs in `tf.keras` with TPUStrategy, or write your own customized
training loop by calling `strategy.run` directly. Note that
TPUStrategy doesn't support pure eager execution, so please make sure the
function passed into `strategy.run` is a `tf.function` or
`strategy.run` is called inside a `tf.function` if eager
behavior is enabled."
2783,run,tensorflow/tensorflow/python/distribute/tpu_strategy.py,338,method,See base class.
2784,cluster_resolver,tensorflow/tensorflow/python/distribute/tpu_strategy.py,349,method,"Returns the cluster resolver associated with this strategy.
`tf.distribute.experimental.TPUStrategy` provides the
associated `tf.distribute.cluster_resolver.ClusterResolver`. If the user
provides one in `__init__`, that instance is returned; if the user does
not, a default
`tf.distribute.cluster_resolver.TPUClusterResolver` is provided."
2785,TPUStrategyV1,tensorflow/tensorflow/python/distribute/tpu_strategy.py,362,class,TPU distribution strategy implementation.
2786,steps_per_run,tensorflow/tensorflow/python/distribute/tpu_strategy.py,396,method,DEPRECATED: use .extended.steps_per_run instead.
2787,run,tensorflow/tensorflow/python/distribute/tpu_strategy.py,403,method,"Run `fn` on each replica, with the given arguments.
Executes ops specified by `fn` on each replica. If `args` or `kwargs` have
""per-replica"" values, such as those produced by a ""distributed `Dataset`"",
when `fn` is executed on a particular replica, it will be executed with the
component of those ""per-replica"" values that correspond to that replica.
`fn` may call `tf.distribute.get_replica_context()` to access members such
as `all_reduce`.
All arguments in `args` or `kwargs` should either be nest of tensors or
per-replica objects containing tensors or composite tensors.
Users can pass strategy specific options to `options` argument. An example
to enable bucketizing dynamic shapes in `TPUStrategy.run`
is:
>>> resolver = tf.distribute.cluster_resolver.TPUClusterResolver(tpu='')
>>> tf.config.experimental_connect_to_cluster(resolver)
>>> tf.tpu.experimental.initialize_tpu_system(resolver)
>>> strategy = tf.distribute.experimental.TPUStrategy(resolver)
>>> options = tf.distribute.RunOptions(
... experimental_bucketizing_dynamic_shape=True)
>>> dataset = tf.data.Dataset.range(
... strategy.num_replicas_in_sync, output_type=dtypes.float32).batch(
... strategy.num_replicas_in_sync, drop_remainder=True)
>>> input_iterator = iter(strategy.experimental_distribute_dataset(dataset))
>>> @tf.function()
... def step_fn(inputs):
... output = tf.reduce_sum(inputs)
... return output
>>> strategy.run(step_fn, args=(next(input_iterator),), options=options)
Args:
fn: The function to run. The output must be a `tf.nest` of `Tensor`s.
args: (Optional) Positional arguments to `fn`.
kwargs: (Optional) Keyword arguments to `fn`.
options: (Optional) An instance of `tf.distribute.RunOptions` specifying
the options to run `fn`.
Returns:
Merged return value of `fn` across replicas. The structure of the return
value is the same as the return value from `fn`. Each element in the
structure can either be ""per-replica"" `Tensor` objects or `Tensor`s
(for example, if running on a single replica)."
2788,TPUExtended,tensorflow/tensorflow/python/distribute/tpu_strategy.py,462,class,Implementation of TPUStrategy.
2789,experimental_logical_device,tensorflow/tensorflow/python/distribute/tpu_strategy.py,750,method,Places variables and ops on the specified logical device.
2790,read_var,tensorflow/tensorflow/python/distribute/tpu_strategy.py,969,method,
2791,value_container,tensorflow/tensorflow/python/distribute/tpu_strategy.py,979,method,
2792,num_hosts,tensorflow/tensorflow/python/distribute/tpu_strategy.py,1005,method,
2793,num_replicas_per_host,tensorflow/tensorflow/python/distribute/tpu_strategy.py,1013,method,
2794,experimental_between_graph,tensorflow/tensorflow/python/distribute/tpu_strategy.py,1033,method,
2795,experimental_should_init,tensorflow/tensorflow/python/distribute/tpu_strategy.py,1037,method,
2796,should_checkpoint,tensorflow/tensorflow/python/distribute/tpu_strategy.py,1041,method,
2797,should_save_summary,tensorflow/tensorflow/python/distribute/tpu_strategy.py,1045,method,
2798,worker_devices,tensorflow/tensorflow/python/distribute/tpu_strategy.py,1049,method,
2799,parameter_devices,tensorflow/tensorflow/python/distribute/tpu_strategy.py,1053,method,
2800,non_slot_devices,tensorflow/tensorflow/python/distribute/tpu_strategy.py,1056,method,
2801,tpu_run,tensorflow/tensorflow/python/distribute/tpu_strategy.py,1097,method,
2802,run_fn,tensorflow/tensorflow/python/distribute/tpu_strategy.py,658,method,Single step on the TPU device.
2803,rewrite_fn,tensorflow/tensorflow/python/distribute/tpu_strategy.py,677,method,The rewritten step fn running on TPU.
2804,tpu_function,tensorflow/tensorflow/python/distribute/tpu_strategy.py,1107,method,TF Function used to replicate the user computation.
2805,async_wait,tensorflow/tensorflow/python/distribute/tpu_strategy.py,542,method,
2806,replicated_fn,tensorflow/tensorflow/python/distribute/tpu_strategy.py,1125,method,Wraps user function to provide replica ID and `Tensor` inputs.
2807,get_tpu_cluster_resolver,tensorflow/tensorflow/python/distribute/tpu_strategy_compilation_test.py,36,function,
2808,get_tpu_strategy,tensorflow/tensorflow/python/distribute/tpu_strategy_compilation_test.py,45,function,
2809,get_tpu_cluster_resolver,tensorflow/tensorflow/python/distribute/tpu_strategy_test.py,70,function,
2810,get_tpu_strategy,tensorflow/tensorflow/python/distribute/tpu_strategy_test.py,79,function,
2811,TPUVariableMixin,tensorflow/tensorflow/python/distribute/tpu_values.py,77,class,Mixin for TPU variables.
2812,get,tensorflow/tensorflow/python/distribute/tpu_values.py,97,method,
2813,handle,tensorflow/tensorflow/python/distribute/tpu_values.py,112,method,The handle by which this variable can be accessed.
2814,device,tensorflow/tensorflow/python/distribute/tpu_values.py,129,method,
2815,read_value,tensorflow/tensorflow/python/distribute/tpu_values.py,145,method,
2816,value,tensorflow/tensorflow/python/distribute/tpu_values.py,151,method,
2817,op,tensorflow/tensorflow/python/distribute/tpu_values.py,164,method,
2818,enclosing_tpu_context,tensorflow/tensorflow/python/distribute/tpu_values.py,183,function,"Returns the TPUReplicateContext, which exists inside a tpu.rewrite()."
2819,TPUMirroredVariable,tensorflow/tensorflow/python/distribute/tpu_values.py,200,class,Holds a map from replica to TPU variables whose values are kept in sync.
2820,assign_sub,tensorflow/tensorflow/python/distribute/tpu_values.py,203,method,
2821,assign_add,tensorflow/tensorflow/python/distribute/tpu_values.py,223,method,
2822,assign,tensorflow/tensorflow/python/distribute/tpu_values.py,243,method,
2823,scatter_sub,tensorflow/tensorflow/python/distribute/tpu_values.py,262,method,
2824,scatter_add,tensorflow/tensorflow/python/distribute/tpu_values.py,265,method,
2825,scatter_max,tensorflow/tensorflow/python/distribute/tpu_values.py,268,method,
2826,scatter_min,tensorflow/tensorflow/python/distribute/tpu_values.py,271,method,
2827,scatter_mul,tensorflow/tensorflow/python/distribute/tpu_values.py,274,method,
2828,scatter_div,tensorflow/tensorflow/python/distribute/tpu_values.py,277,method,
2829,scatter_update,tensorflow/tensorflow/python/distribute/tpu_values.py,280,method,
2830,TPUSyncOnReadVariable,tensorflow/tensorflow/python/distribute/tpu_values.py,287,class,Holds a map from replica to variables whose values are reduced on save.
2831,assign_sub,tensorflow/tensorflow/python/distribute/tpu_values.py,290,method,
2832,assign_add,tensorflow/tensorflow/python/distribute/tpu_values.py,298,method,
2833,assign,tensorflow/tensorflow/python/distribute/tpu_values.py,306,method,
2834,DistributedValues,tensorflow/tensorflow/python/distribute/values.py,76,class,"Base class for representing distributed values.
A subclass instance of `tf.distribute.DistributedValues` is created when
creating variables within a distribution strategy, iterating a
`tf.distribute.DistributedDataset` or through `tf.distribute.Strategy.run`.
This base class should never be instantiated directly.
`tf.distribute.DistributedValues` contains a value per replica. Depending on
the subclass, the values could either be synced on update, synced on demand,
or never synced.
`tf.distribute.DistributedValues` can be reduced to obtain single value across
replicas, as input into `tf.distribute.Strategy.run` or the per-replica values
inspected using `tf.distribute.Strategy.experimental_local_results`.
Example usage:
1. Created from a `tf.distribute.DistributedDataset`:
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> dataset = tf.data.Dataset.from_tensor_slices([5., 6., 7., 8.]).batch(2)
>>> dataset_iterator = iter(strategy.experimental_distribute_dataset(dataset))
>>> distributed_values = next(dataset_iterator)
2. Returned by `run`:
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> @tf.function
... def run():
... ctx = tf.distribute.get_replica_context()
... return ctx.replica_id_in_sync_group
>>> distributed_values = strategy.run(run)
3. As input into `run`:
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> dataset = tf.data.Dataset.from_tensor_slices([5., 6., 7., 8.]).batch(2)
>>> dataset_iterator = iter(strategy.experimental_distribute_dataset(dataset))
>>> distributed_values = next(dataset_iterator)
>>> @tf.function
... def run(input):
... return input + 1.0
>>> updated_value = strategy.run(run, args=(distributed_values,))
4. Reduce value:
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> dataset = tf.data.Dataset.from_tensor_slices([5., 6., 7., 8.]).batch(2)
>>> dataset_iterator = iter(strategy.experimental_distribute_dataset(dataset))
>>> distributed_values = next(dataset_iterator)
>>> reduced_value = strategy.reduce(tf.distribute.ReduceOp.SUM,
... distributed_values,
... axis = 0)
5. Inspect local replica values:
>>> strategy = tf.distribute.MirroredStrategy([""GPU:0"", ""GPU:1""])
>>> dataset = tf.data.Dataset.from_tensor_slices([5., 6., 7., 8.]).batch(2)
>>> dataset_iterator = iter(strategy.experimental_distribute_dataset(dataset))
>>> per_replica_values = strategy.experimental_local_results(
... distributed_values)
>>> per_replica_values
(<tf.Tensor: shape=(1,), dtype=float32, numpy=array([5.], dtype=float32)>,
<tf.Tensor: shape=(1,), dtype=float32, numpy=array([6.], dtype=float32)>)"
2835,DistributedDelegate,tensorflow/tensorflow/python/distribute/values.py,202,class,A map from device to values; acts as the same type as the values.
2836,values,tensorflow/tensorflow/python/distribute/values.py,232,method,Returns the per replica values.
2837,PerReplica,tensorflow/tensorflow/python/distribute/values.py,361,class,Holds a map from replica to unsynchronized values.
2838,values,tensorflow/tensorflow/python/distribute/values.py,370,method,Returns the per replica values.
2839,PerReplicaSpec,tensorflow/tensorflow/python/distribute/values.py,375,class,Type specification for a `PerReplica`.
2840,Mirrored,tensorflow/tensorflow/python/distribute/values.py,407,class,Holds a map from replica to values which are kept in sync.
2841,DistributedVarOp,tensorflow/tensorflow/python/distribute/values.py,421,class,A class that looks like `tf.Operation`.
2842,DistributedVariable,tensorflow/tensorflow/python/distribute/values.py,440,class,Holds a map from replica to variables.
2843,is_initialized,tensorflow/tensorflow/python/distribute/values.py,480,method,"Identifies if all the component variables are initialized.
Args:
name: Name of the final `logical_and` op.
Returns:
The op that evaluates to True or False depending on if all the
component variables are initialized."
2844,initializer,tensorflow/tensorflow/python/distribute/values.py,504,method,
2845,initialized_value,tensorflow/tensorflow/python/distribute/values.py,514,method,
2846,initial_value,tensorflow/tensorflow/python/distribute/values.py,518,method,
2847,constraint,tensorflow/tensorflow/python/distribute/values.py,522,method,
2848,graph,tensorflow/tensorflow/python/distribute/values.py,526,method,
2849,name,tensorflow/tensorflow/python/distribute/values.py,543,method,
2850,dtype,tensorflow/tensorflow/python/distribute/values.py,547,method,
2851,shape,tensorflow/tensorflow/python/distribute/values.py,551,method,
2852,synchronization,tensorflow/tensorflow/python/distribute/values.py,555,method,
2853,aggregation,tensorflow/tensorflow/python/distribute/values.py,559,method,
2854,handle,tensorflow/tensorflow/python/distribute/values.py,569,method,
2855,eval,tensorflow/tensorflow/python/distribute/values.py,579,method,
2856,device,tensorflow/tensorflow/python/distribute/values.py,594,method,
2857,trainable,tensorflow/tensorflow/python/distribute/values.py,598,method,
2858,distribute_strategy,tensorflow/tensorflow/python/distribute/values.py,602,method,
2859,get_shape,tensorflow/tensorflow/python/distribute/values.py,605,method,
2860,to_proto,tensorflow/tensorflow/python/distribute/values.py,608,method,
2861,op,tensorflow/tensorflow/python/distribute/values.py,612,method,
2862,read_value,tensorflow/tensorflow/python/distribute/values.py,652,method,
2863,value,tensorflow/tensorflow/python/distribute/values.py,656,method,
2864,numpy,tensorflow/tensorflow/python/distribute/values.py,661,method,
2865,assign_sub,tensorflow/tensorflow/python/distribute/values.py,668,method,
2866,assign_add,tensorflow/tensorflow/python/distribute/values.py,679,method,
2867,assign,tensorflow/tensorflow/python/distribute/values.py,690,method,
2868,scatter_sub,tensorflow/tensorflow/python/distribute/values.py,701,method,
2869,scatter_add,tensorflow/tensorflow/python/distribute/values.py,708,method,
2870,scatter_mul,tensorflow/tensorflow/python/distribute/values.py,715,method,
2871,scatter_div,tensorflow/tensorflow/python/distribute/values.py,722,method,
2872,scatter_min,tensorflow/tensorflow/python/distribute/values.py,729,method,
2873,scatter_max,tensorflow/tensorflow/python/distribute/values.py,736,method,
2874,scatter_update,tensorflow/tensorflow/python/distribute/values.py,743,method,
2875,MirroredVariable,tensorflow/tensorflow/python/distribute/values.py,915,class,Holds a map from replica to variables whose values are kept in sync.
2876,scatter_min,tensorflow/tensorflow/python/distribute/values.py,921,method,
2877,scatter_max,tensorflow/tensorflow/python/distribute/values.py,928,method,
2878,scatter_update,tensorflow/tensorflow/python/distribute/values.py,935,method,
2879,SyncOnReadVariable,tensorflow/tensorflow/python/distribute/values.py,1016,class,Holds a map from replica to variables whose values are reduced on save.
2880,assign_sub,tensorflow/tensorflow/python/distribute/values.py,1024,method,
2881,assign_add,tensorflow/tensorflow/python/distribute/values.py,1033,method,
2882,assign,tensorflow/tensorflow/python/distribute/values.py,1042,method,
2883,scatter_sub,tensorflow/tensorflow/python/distribute/values.py,1056,method,
2884,scatter_add,tensorflow/tensorflow/python/distribute/values.py,1059,method,
2885,scatter_mul,tensorflow/tensorflow/python/distribute/values.py,1062,method,
2886,scatter_div,tensorflow/tensorflow/python/distribute/values.py,1065,method,
2887,scatter_min,tensorflow/tensorflow/python/distribute/values.py,1068,method,
2888,scatter_max,tensorflow/tensorflow/python/distribute/values.py,1071,method,
2889,scatter_update,tensorflow/tensorflow/python/distribute/values.py,1074,method,
2890,value,tensorflow/tensorflow/python/distribute/values.py,1077,method,
2891,VariablePolicy,tensorflow/tensorflow/python/distribute/values.py,1168,class,"Policy defining synchronization and aggregation of a distributed variable.
Given `synchronization` and `aggregation` parameters set on a `tf.Variable`
during variable creation within `tf.distribute` scope, `tf.distribute` creates
an appropriate policy object and assigns it to the distributed variable. All
variable operations are delegated to the respective policy object."
2892,value,tensorflow/tensorflow/python/distribute/values.py,1180,method,
2893,OnReadPolicy,tensorflow/tensorflow/python/distribute/values.py,1201,class,"Policy defined for `tf.VariableSynchronization.ON_READ` synchronization.
This policy is created when `synchronization` is set to
`tf.VariableSynchronization.ON_READ` and `aggregation` is set to any of the
values allowed by the `tf.VariableAggregation` enum such as `NONE`, `SUM`,
`MEAN` or `ONLY_FIRST_REPLICA`when creating a `tf.Variable` in `tf.distribute`
scope."
2894,value,tensorflow/tensorflow/python/distribute/values.py,1214,method,
2895,assign_sub,tensorflow/tensorflow/python/distribute/values.py,1245,method,Subtracts a value from this variable.
2896,assign_add,tensorflow/tensorflow/python/distribute/values.py,1257,method,Adds a value to this variable.
2897,assign,tensorflow/tensorflow/python/distribute/values.py,1269,method,
2898,scatter_sub,tensorflow/tensorflow/python/distribute/values.py,1280,method,
2899,scatter_add,tensorflow/tensorflow/python/distribute/values.py,1284,method,
2900,scatter_mul,tensorflow/tensorflow/python/distribute/values.py,1288,method,
2901,scatter_div,tensorflow/tensorflow/python/distribute/values.py,1292,method,
2902,scatter_min,tensorflow/tensorflow/python/distribute/values.py,1296,method,
2903,scatter_max,tensorflow/tensorflow/python/distribute/values.py,1300,method,
2904,scatter_update,tensorflow/tensorflow/python/distribute/values.py,1304,method,
2905,get_saveable,tensorflow/tensorflow/python/distribute/values.py,1308,method,Create a saveable object for the given variable.
2906,get_restore_ops,tensorflow/tensorflow/python/distribute/values.py,1326,method,Restore the same value into all variables.
2907,tensor,tensorflow/tensorflow/python/distribute/values.py,1313,method,
2908,AutoPolicy,tensorflow/tensorflow/python/distribute/values.py,1340,class,"Policy defined for `tf.VariableSynchronization.AUTO` synchronization.
This policy is created when `synchronization` is set to
`tf.VariableSynchronization.AUTO` and `aggregation` is set to
`tf.VariableAggregation.NONE` when creating a `tf.Variable` in `tf.distribute`
scope."
2909,value,tensorflow/tensorflow/python/distribute/values.py,1352,method,
2910,assign,tensorflow/tensorflow/python/distribute/values.py,1366,method,
2911,assign_add,tensorflow/tensorflow/python/distribute/values.py,1370,method,
2912,assign_sub,tensorflow/tensorflow/python/distribute/values.py,1375,method,
2913,scatter_sub,tensorflow/tensorflow/python/distribute/values.py,1380,method,
2914,scatter_add,tensorflow/tensorflow/python/distribute/values.py,1384,method,
2915,scatter_mul,tensorflow/tensorflow/python/distribute/values.py,1388,method,
2916,scatter_div,tensorflow/tensorflow/python/distribute/values.py,1392,method,
2917,scatter_min,tensorflow/tensorflow/python/distribute/values.py,1396,method,
2918,scatter_max,tensorflow/tensorflow/python/distribute/values.py,1404,method,
2919,scatter_update,tensorflow/tensorflow/python/distribute/values.py,1412,method,
2920,get_saveable,tensorflow/tensorflow/python/distribute/values.py,1421,method,
2921,get_restore_ops,tensorflow/tensorflow/python/distribute/values.py,1425,method,
2922,OnWritePolicy,tensorflow/tensorflow/python/distribute/values.py,1432,class,"Policy defined for `tf.VariableSynchronization.ON_WRITE` synchronization.
This policy is created when the following `synchronization` and
`aggregation` parameters are specified when creating a `tf.Variable` in
`tf.distribute` scope:
* `synchronization` is equal to `tf.VariableSynchronization.AUTO` and
aggregation can be any of the following `tf.VariableAggregation` enum
values such as `SUM`, `MEAN` or `ONLY_FIRST_REPLICA`.
* `synchronization` is equal to `tf.VariableSynchronization.ON_WRITE` and
aggregation can be any of the following `tf.VariableAggregation` enum
values such as `NONE`, `SUM`, `MEAN` or `ONLY_FIRST_REPLICA`."
2923,mirrored_and_tpu_strategy_combinations,tensorflow/tensorflow/python/distribute/values_test.py,395,function,
2924,strategy_and_run_tf_function_combinations,tensorflow/tensorflow/python/distribute/values_test.py,1358,function,
2925,on_write_assign,tensorflow/tensorflow/python/distribute/values_util.py,31,function,
2926,on_write_assign_add,tensorflow/tensorflow/python/distribute/values_util.py,41,function,
2927,on_write_assign_sub,tensorflow/tensorflow/python/distribute/values_util.py,52,function,
2928,assign_on_each_device,tensorflow/tensorflow/python/distribute/values_util.py,63,function,Update the variable on each replica with the given assign_func and value.
2929,on_read_assign_sub_cross_replica,tensorflow/tensorflow/python/distribute/values_util.py,78,function,
2930,on_read_assign_add_cross_replica,tensorflow/tensorflow/python/distribute/values_util.py,90,function,
2931,on_read_assign_cross_replica,tensorflow/tensorflow/python/distribute/values_util.py,102,function,Return the value of the variable in cross replica context.
2932,scatter_sub,tensorflow/tensorflow/python/distribute/values_util.py,118,function,
2933,scatter_add,tensorflow/tensorflow/python/distribute/values_util.py,127,function,
2934,scatter_mul,tensorflow/tensorflow/python/distribute/values_util.py,136,function,
2935,scatter_div,tensorflow/tensorflow/python/distribute/values_util.py,145,function,
2936,scatter_min,tensorflow/tensorflow/python/distribute/values_util.py,154,function,
2937,scatter_max,tensorflow/tensorflow/python/distribute/values_util.py,163,function,
2938,scatter_update,tensorflow/tensorflow/python/distribute/values_util.py,172,function,
2939,get_current_replica_id_as_int,tensorflow/tensorflow/python/distribute/values_util.py,181,function,"Returns the current replica ID as an integer, or `None`."
2940,assign_on_device,tensorflow/tensorflow/python/distribute/values_util.py,193,function,
2941,assign_add_on_device,tensorflow/tensorflow/python/distribute/values_util.py,198,function,
2942,assign_sub_on_device,tensorflow/tensorflow/python/distribute/values_util.py,203,function,
2943,assert_replica_context,tensorflow/tensorflow/python/distribute/values_util.py,208,function,
2944,apply_aggregation,tensorflow/tensorflow/python/distribute/values_util.py,218,function,
2945,format_master_url,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,35,function,
2946,get_accelerator_devices,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,42,function,Returns accelerator devices given a master and a configuration.
2947,ClusterResolver,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,61,class,"Abstract class for all implementations of ClusterResolvers.
This defines the skeleton for all implementations of ClusterResolvers.
ClusterResolvers are a way for TensorFlow to communicate with various cluster
management systems (e.g. GCE, AWS, etc...) and gives TensorFlow necessary
information to set up distributed training.
By letting TensorFlow communicate with these systems, we will be able to
automatically discover and resolve IP addresses for various TensorFlow
workers. This will eventually allow us to automatically recover from
underlying machine failures and scale TensorFlow worker clusters up and down.
Note to Implementors of `tf.distribute.cluster_resolver.ClusterResolver`
subclass: In addition to these abstract methods, when task_type, task_id, and
rpc_layer attributes are applicable, you should also implement them either as
properties with getters or setters, or directly set the attributes
`self._task_type`, `self._task_id`, or `self._rpc_layer` so the base class'
getters and setters are used. See
`tf.distribute.cluster_resolver.SimpleClusterResolver.__init__` for an
example.
In general, multi-client tf.distribute strategies such as
`tf.distribute.experimental.MultiWorkerMirroredStrategy` require task_type and
task_id properties to be available in the `ClusterResolver` they are using. On
the other hand, these concepts are not applicable in single-client strategies,
such as `tf.distribute.experimental.TPUStrategy`, because the program is only
expected to be run on one task, so there should not be a need to have code
branches according to task type and task id.
- task_type is the name of the server's current named job (e.g. 'worker',
'ps' in a distributed parameterized training job).
- task_id is the ordinal index of the server within the task type.
- rpc_layer is the protocol used by TensorFlow to communicate with other
TensorFlow servers in a distributed environment."
2948,cluster_spec,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,99,method,"Retrieve the current state of the cluster and return a `tf.train.ClusterSpec`.
Returns:
A `tf.train.ClusterSpec` representing the state of the cluster at the
moment this function is called.
Implementors of this function must take care in ensuring that the
ClusterSpec returned is up-to-date at the time of calling this function.
This usually means retrieving the information from the underlying cluster
management system every time this function is invoked and reconstructing
a cluster_spec, rather than attempting to cache anything."
2949,master,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,115,method,"Retrieves the name or URL of the session master.
Note: this is only useful for TensorFlow 1.x.
Args:
task_type: (Optional) The type of the TensorFlow task of the master.
task_id: (Optional) The index of the TensorFlow task of the master.
rpc_layer: (Optional) The RPC protocol for the given cluster.
Returns:
The name or URL of the session master.
Implementors of this function must take care in ensuring that the master
returned is up-to-date at the time to calling this function. This usually
means retrieving the master every time this function is invoked."
2950,num_accelerators,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,134,method,"Returns the number of accelerator cores per worker.
This returns the number of accelerator cores (such as GPUs and TPUs)
available per worker.
Optionally, we allow callers to specify the task_type, and task_id, for
if they want to target a specific TensorFlow task to query
the number of accelerators. This is to support heterogenous environments,
where the number of accelerators cores per host is different.
Args:
task_type: (Optional) The type of the TensorFlow task of the machine we
want to query.
task_id: (Optional) The index of the TensorFlow task of the machine we
want to query.
config_proto: (Optional) Configuration for starting a new session to
query how many accelerator cores it has.
Returns:
A map of accelerator types to number of cores."
2951,environment,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,174,method,"Returns the current environment which TensorFlow is running in.
There are two possible return values, ""google"" (when TensorFlow is running
in a Google-internal environment) or an empty string (when TensorFlow is
running elsewhere).
If you are implementing a ClusterResolver that works in both the Google
environment and the open-source world (for instance, a TPU ClusterResolver
or similar), you will have to return the appropriate string depending on the
environment, which you will have to detect.
Otherwise, if you are implementing a ClusterResolver that will only work
in open-source TensorFlow, you do not need to implement this property."
2952,task_type,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,192,method,"Returns the task type this `ClusterResolver` indicates.
In TensorFlow distributed environment, each job may have an applicable
task type. Valid task types in TensorFlow include
'chief': a worker that is designated with more responsibility,
'worker': a regular worker for training/evaluation,
'ps': a parameter server, or
'evaluator': an evaluator that evaluates the checkpoints for metrics.
See [Multi-worker configuration](
https://www.tensorflow.org/tutorials/distribute/multi_worker_with_keras#multi-worker_configuration)
for more information about 'chief' and 'worker' task type, which are most
commonly used.
Having access to such information is useful when user needs to run specific
code according to task types. For example,
```python
cluster_spec = tf.train.ClusterSpec({
""ps"": [""localhost:2222"", ""localhost:2223""],
""worker"": [""localhost:2224"", ""localhost:2225"", ""localhost:2226""]
})
# SimpleClusterResolver is used here for illustration; other cluster
# resolvers may be used for other source of task type/id.
simple_resolver = SimpleClusterResolver(cluster_spec, task_type=""worker"",
task_id=1)
...
if cluster_resolver.task_type == 'worker':
# Perform something that's only applicable on workers. This block
# will run on this particular instance since we've specified this task to
# be a worker in above cluster resolver.
elif cluster_resolver.task_type == 'ps':
# Perform something that's only applicable on parameter servers. This
# block will not run on this particular instance.
```
Returns `None` if such information is not available or is not applicable
in the current distributed environment, such as training with
`tf.distribute.experimental.TPUStrategy`.
For more information, please see
`tf.distribute.cluster_resolver.ClusterResolver`'s class doc."
2953,task_id,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,242,method,"Returns the task id this `ClusterResolver` indicates.
In TensorFlow distributed environment, each job may have an applicable
task id, which is the index of the instance within its task type. This is
useful when user needs to run specific code according to task index. For
example,
```python
cluster_spec = tf.train.ClusterSpec({
""ps"": [""localhost:2222"", ""localhost:2223""],
""worker"": [""localhost:2224"", ""localhost:2225"", ""localhost:2226""]
})
# SimpleClusterResolver is used here for illustration; other cluster
# resolvers may be used for other source of task type/id.
simple_resolver = SimpleClusterResolver(cluster_spec, task_type=""worker"",
task_id=0)
...
if cluster_resolver.task_type == 'worker' and cluster_resolver.task_id == 0:
# Perform something that's only applicable on 'worker' type, id 0. This
# block will run on this particular instance since we've specified this
# task to be a 'worker', id 0 in above cluster resolver.
else:
# Perform something that's only applicable on other ids. This block will
# not run on this particular instance.
```
Returns `None` if such information is not available or is not applicable
in the current distributed environment, such as training with
`tf.distribute.cluster_resolver.TPUClusterResolver`.
For more information, please see
`tf.distribute.cluster_resolver.ClusterResolver`'s class docstring."
2954,task_type,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,282,method,Setter of `task_type` property. See `task_type` property doc.
2955,task_id,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,287,method,Setter of `task_id` property. See `task_type` property doc.
2956,SimpleClusterResolver,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,293,class,"Simple implementation of ClusterResolver that accepts all attributes.
Please see the base class for documentation of arguments of its constructor.
It is useful if you want to specify some or all attributes.
Usage example with `tf.distribute.Strategy`:
```Python
cluster = tf.train.ClusterSpec({""worker"": [""worker0.example.com:2222"",
""worker1.example.com:2222""]})
# On worker 0
cluster_resolver = SimpleClusterResolver(cluster, task_type=""worker"",
task_id=0,
num_accelerators={""GPU"": 8},
rpc_layer=""grpc"")
strategy = tf.distribute.experimental.MultiWorkerMirroredStrategy(
cluster_resolver=cluster_resolver)
# On worker 1
cluster_resolver = SimpleClusterResolver(cluster, task_type=""worker"",
task_id=1,
num_accelerators={""GPU"": 8},
rpc_layer=""grpc"")
strategy = tf.distribute.experimental.MultiWorkerMirroredStrategy(
cluster_resolver=cluster_resolver)
```"
2957,cluster_spec,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,345,method,Returns the ClusterSpec passed into the constructor.
2958,master,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,349,method,"Returns the master address to use when creating a session.
Note: this is only useful for TensorFlow 1.x.
Args:
task_type: (Optional) The type of the TensorFlow task of the master.
task_id: (Optional) The index of the TensorFlow task of the master.
rpc_layer: (Optional) The RPC used by distributed TensorFlow.
Returns:
The name or URL of the session master.
If a task_type and task_id is given, this will override the `master`
string passed into the initialization function."
2959,task_type,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,373,method,
2960,task_id,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,377,method,
2961,task_type,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,381,method,
2962,task_id,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,385,method,
2963,environment,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,389,method,
2964,num_accelerators,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,392,method,"Returns the number of accelerator cores per worker.
The SimpleClusterResolver does not do automatic detection of accelerators,
and thus all arguments are unused and we simply return the value provided
in the constructor.
Args:
task_type: Unused.
task_id: Unused.
config_proto: Unused."
2965,rpc_layer,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,414,method,
2966,rpc_layer,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,418,method,
2967,UnionClusterResolver,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,423,class,"Performs a union on underlying ClusterResolvers.
This class performs a union given two or more existing ClusterResolvers. It
merges the underlying ClusterResolvers, and returns one unified ClusterSpec
when cluster_spec is called. The details of the merge function is
documented in the cluster_spec function.
For additional ClusterResolver properties such as task type, task index,
rpc layer, environment, etc..., we will return the value from the first
ClusterResolver in the union.
An example to combine two cluster resolvers:
```Python
cluster_0 = tf.train.ClusterSpec({""worker"": [""worker0.example.com:2222"",
""worker1.example.com:2222""]})
cluster_resolver_0 = SimpleClusterResolver(cluster, task_type=""worker"",
task_id=0,
rpc_layer=""grpc"")
cluster_1 = tf.train.ClusterSpec({""ps"": [""ps0.example.com:2222"",
""ps1.example.com:2222""]})
cluster_resolver_1 = SimpleClusterResolver(cluster, task_type=""ps"",
task_id=0,
rpc_layer=""grpc"")
# Its task type would be ""worker"".
cluster_resolver = UnionClusterResolver(cluster_resolver_0,
cluster_resolver_1)
```
An example to override the number of GPUs in a TFConfigClusterResolver
instance:
```Python
tf_config = TFConfigClusterResolver()
gpu_override = SimpleClusterResolver(tf_config.cluster_spec(),
num_accelerators={""GPU"": 1})
cluster_resolver = UnionResolver(gpu_override, tf_config)
```"
2968,cluster_spec,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,499,method,"Returns a union of all the ClusterSpecs from the ClusterResolvers.
Returns:
A ClusterSpec containing host information merged from all the underlying
ClusterResolvers.
Raises:
KeyError: If there are conflicting keys detected when merging two or
more dictionaries, this exception is raised.
Note: If there are multiple ClusterResolvers exposing ClusterSpecs with the
same job name, we will merge the list/dict of workers.
If *all* underlying ClusterSpecs expose the set of workers as lists, we will
concatenate the lists of workers, starting with the list of workers from
the first ClusterResolver passed into the constructor.
If *any* of the ClusterSpecs expose the set of workers as a dict, we will
treat all the sets of workers as dicts (even if they are returned as lists)
and will only merge them into a dict if there is no conflicting keys. If
there is a conflicting key, we will raise a `KeyError`."
2969,master,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,573,method,"Returns the master address to use when creating a session.
This usually returns the master from the first ClusterResolver passed in,
but you can override this by specifying the task_type and task_id.
Note: this is only useful for TensorFlow 1.x.
Args:
task_type: (Optional) The type of the TensorFlow task of the master.
task_id: (Optional) The index of the TensorFlow task of the master.
rpc_layer: (Optional) The RPC protocol for the given cluster.
Returns:
The name or URL of the session master."
2970,task_type,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,596,method,
2971,task_id,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,600,method,
2972,task_type,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,604,method,
2973,task_id,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,608,method,
2974,environment,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,612,method,
2975,num_accelerators,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,615,method,
2976,rpc_layer,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,623,method,
2977,rpc_layer,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver.py,627,method,
2978,MockBaseClusterResolver,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver_test.py,34,class,
2979,cluster_spec,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver_test.py,36,method,
2980,master,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver_test.py,39,method,
2981,environment,tensorflow/tensorflow/python/distribute/cluster_resolver/cluster_resolver_test.py,42,method,
2982,GCEClusterResolver,tensorflow/tensorflow/python/distribute/cluster_resolver/gce_cluster_resolver.py,35,class,"ClusterResolver for Google Compute Engine.
This is an implementation of cluster resolvers for the Google Compute Engine
instance group platform. By specifying a project, zone, and instance group,
this will retrieve the IP address of all the instances within the instance
group and return a ClusterResolver object suitable for use for distributed
TensorFlow.
Note: this cluster resolver cannot retrieve `task_type`, `task_id` or
`rpc_layer`. To use it with some distribution strategies like
`tf.distribute.experimental.MultiWorkerMirroredStrategy`, you will need to
specify `task_type` and `task_id` in the constructor.
Usage example with tf.distribute.Strategy:
```Python
# On worker 0
cluster_resolver = GCEClusterResolver(""my-project"", ""us-west1"",
""my-instance-group"",
task_type=""worker"", task_id=0)
strategy = tf.distribute.experimental.MultiWorkerMirroredStrategy(
cluster_resolver=cluster_resolver)
# On worker 1
cluster_resolver = GCEClusterResolver(""my-project"", ""us-west1"",
""my-instance-group"",
task_type=""worker"", task_id=1)
strategy = tf.distribute.experimental.MultiWorkerMirroredStrategy(
cluster_resolver=cluster_resolver)
```"
2983,cluster_spec,tensorflow/tensorflow/python/distribute/cluster_resolver/gce_cluster_resolver.py,129,method,"Returns a ClusterSpec object based on the latest instance group info.
This returns a ClusterSpec object for use based on information from the
specified instance group. We will retrieve the information from the GCE APIs
every time this method is called.
Returns:
A ClusterSpec containing host information retrieved from GCE."
2984,master,tensorflow/tensorflow/python/distribute/cluster_resolver/gce_cluster_resolver.py,174,method,
2985,task_type,tensorflow/tensorflow/python/distribute/cluster_resolver/gce_cluster_resolver.py,188,method,
2986,task_id,tensorflow/tensorflow/python/distribute/cluster_resolver/gce_cluster_resolver.py,192,method,
2987,task_type,tensorflow/tensorflow/python/distribute/cluster_resolver/gce_cluster_resolver.py,196,method,
2988,task_id,tensorflow/tensorflow/python/distribute/cluster_resolver/gce_cluster_resolver.py,202,method,
2989,rpc_layer,tensorflow/tensorflow/python/distribute/cluster_resolver/gce_cluster_resolver.py,206,method,
2990,rpc_layer,tensorflow/tensorflow/python/distribute/cluster_resolver/gce_cluster_resolver.py,210,method,
2991,KubernetesClusterResolver,tensorflow/tensorflow/python/distribute/cluster_resolver/kubernetes_cluster_resolver.py,35,class,"ClusterResolver for Kubernetes.
This is an implementation of cluster resolvers for Kubernetes. When given the
the Kubernetes namespace and label selector for pods, we will retrieve the
pod IP addresses of all running pods matching the selector, and return a
ClusterSpec based on that information.
Note: it cannot retrieve `task_type`, `task_id` or `rpc_layer`. To use it
with some distribution strategies like
`tf.distribute.experimental.MultiWorkerMirroredStrategy`, you will need to
specify `task_type` and `task_id` by setting these attributes.
Usage example with tf.distribute.Strategy:
```Python
# On worker 0
cluster_resolver = KubernetesClusterResolver(
{""worker"": [""job-name=worker-cluster-a"", ""job-name=worker-cluster-b""]})
cluster_resolver.task_type = ""worker""
cluster_resolver.task_id = 0
strategy = tf.distribute.experimental.MultiWorkerMirroredStrategy(
cluster_resolver=cluster_resolver)
# On worker 1
cluster_resolver = KubernetesClusterResolver(
{""worker"": [""job-name=worker-cluster-a"", ""job-name=worker-cluster-b""]})
cluster_resolver.task_type = ""worker""
cluster_resolver.task_id = 1
strategy = tf.distribute.experimental.MultiWorkerMirroredStrategy(
cluster_resolver=cluster_resolver)
```"
2992,master,tensorflow/tensorflow/python/distribute/cluster_resolver/kubernetes_cluster_resolver.py,121,method,"Returns the master address to use when creating a session.
You must have set the task_type and task_id object properties before
calling this function, or pass in the `task_type` and `task_id`
parameters when using this function. If you do both, the function parameters
will override the object properties.
Note: this is only useful for TensorFlow 1.x.
Args:
task_type: (Optional) The type of the TensorFlow task of the master.
task_id: (Optional) The index of the TensorFlow task of the master.
rpc_layer: (Optional) The RPC protocol for the given cluster.
Returns:
The name or URL of the session master."
2993,cluster_spec,tensorflow/tensorflow/python/distribute/cluster_resolver/kubernetes_cluster_resolver.py,149,method,"Returns a ClusterSpec object based on the latest info from Kubernetes.
We retrieve the information from the Kubernetes master every time this
method is called.
Returns:
A ClusterSpec containing host information returned from Kubernetes.
Raises:
RuntimeError: If any of the pods returned by the master is not in the
`Running` phase."
2994,expand_hostlist,tensorflow/tensorflow/python/distribute/cluster_resolver/slurm_cluster_resolver.py,31,function,"Create a list of hosts out of a SLURM hostlist.
The order of nodes is preserved and no deduplication is done
Input: 'n[1-2],m5,o[3-4,6,7-9]')
Output: ['n1', 'n2', 'm5', 'o3', 'o4', 'o6', 'o7', 'o8', 'o9']"
2995,expand_tasks_per_node,tensorflow/tensorflow/python/distribute/cluster_resolver/slurm_cluster_resolver.py,89,function,"Expands the tasks per node expression from SLURM.
The order is preserved so it can be matched to the hostlist
Input: '3(x2),2,1'
Output: [3, 3, 2, 1]"
2996,get_num_gpus,tensorflow/tensorflow/python/distribute/cluster_resolver/slurm_cluster_resolver.py,159,function,"Returns the number of GPUs visible on the current node.
Currently only implemented for NVIDIA GPUs."
2997,SlurmClusterResolver,tensorflow/tensorflow/python/distribute/cluster_resolver/slurm_cluster_resolver.py,168,class,"ClusterResolver for system with Slurm workload manager.
This is an implementation of ClusterResolver for Slurm clusters. This allows
the specification of jobs and task counts, number of tasks per node, number
of GPUs on each node and number of GPUs for each task. It retrieves system
attributes by Slurm environment variables, resolves allocated computing node
names, constructs a cluster and returns a ClusterResolver object which can be
used for distributed TensorFlow."
2998,cluster_spec,tensorflow/tensorflow/python/distribute/cluster_resolver/slurm_cluster_resolver.py,304,method,"Returns a ClusterSpec object based on the latest instance group info.
This returns a ClusterSpec object for use based on information from the
specified initialization parameters and Slurm environment variables. The
cluster specification is resolved each time this function is called. The
resolver extract hostnames of nodes by scontrol and pack tasks in that
order until a node a has number of tasks that is equal to specification.
GPUs on nodes are allocated to tasks by specification through setting
CUDA_VISIBLE_DEVICES environment variable.
Returns:
A ClusterSpec containing host information retrieved from Slurm's
environment variables."
2999,get_task_info,tensorflow/tensorflow/python/distribute/cluster_resolver/slurm_cluster_resolver.py,359,method,"Returns job name and task_id for the process which calls this.
This returns the job name and task index for the process which calls this
function according to its rank and cluster specification. The job name and
task index are set after a cluster is constructed by cluster_spec otherwise
defaults to None.
Returns:
A string specifying job name the process belongs to and an integer
specifying the task index the process belongs to in that job."
3000,master,tensorflow/tensorflow/python/distribute/cluster_resolver/slurm_cluster_resolver.py,373,method,"Returns the master string for connecting to a TensorFlow master.
Args:
task_type: (Optional) Overrides the default auto-selected task type.
task_id: (Optional) Overrides the default auto-selected task index.
rpc_layer: (Optional) Overrides the default RPC protocol TensorFlow uses
to communicate across nodes.
Returns:
A connection string for connecting to a TensorFlow master."
3001,num_accelerators,tensorflow/tensorflow/python/distribute/cluster_resolver/slurm_cluster_resolver.py,395,method,
3002,format_master_url,tensorflow/tensorflow/python/distribute/cluster_resolver/tfconfig_cluster_resolver.py,35,function,
3003,TFConfigClusterResolver,tensorflow/tensorflow/python/distribute/cluster_resolver/tfconfig_cluster_resolver.py,52,class,"Implementation of a ClusterResolver which reads the TF_CONFIG EnvVar.
This is an implementation of cluster resolvers when using TF_CONFIG to set
information about the cluster. The cluster spec returned will be
initialized from the TF_CONFIG environment variable.
An example to set TF_CONFIG is:
```Python
os.environ['TF_CONFIG'] = json.dumps({
'cluster': {
'worker': [""localhost:12345"", ""localhost:23456""]
},
'task': {'type': 'worker', 'index': 0}
})
```
However, sometimes the container orchestration framework will set TF_CONFIG
for you. In this case, you can just create an instance without passing in any
arguments. You can find an example here to let Kuburnetes set TF_CONFIG for
you: https://github.com/tensorflow/ecosystem/tree/master/kubernetes. Then you
can use it with `tf.distribute.Strategy` as:
```Python
# `TFConfigClusterResolver` is already the default one in the following
# strategy.
strategy = tf.distribute.experimental.MultiWorkerMirroredStrategy(
cluster_resolver=TFConfigClusterResolver())
```"
3004,task_type,tensorflow/tensorflow/python/distribute/cluster_resolver/tfconfig_cluster_resolver.py,106,method,
3005,task_id,tensorflow/tensorflow/python/distribute/cluster_resolver/tfconfig_cluster_resolver.py,114,method,
3006,task_type,tensorflow/tensorflow/python/distribute/cluster_resolver/tfconfig_cluster_resolver.py,122,method,
3007,task_id,tensorflow/tensorflow/python/distribute/cluster_resolver/tfconfig_cluster_resolver.py,126,method,
3008,environment,tensorflow/tensorflow/python/distribute/cluster_resolver/tfconfig_cluster_resolver.py,130,method,
3009,rpc_layer,tensorflow/tensorflow/python/distribute/cluster_resolver/tfconfig_cluster_resolver.py,134,method,
3010,rpc_layer,tensorflow/tensorflow/python/distribute/cluster_resolver/tfconfig_cluster_resolver.py,141,method,
3011,num_accelerators,tensorflow/tensorflow/python/distribute/cluster_resolver/tfconfig_cluster_resolver.py,144,method,
3012,cluster_spec,tensorflow/tensorflow/python/distribute/cluster_resolver/tfconfig_cluster_resolver.py,153,method,"Returns a ClusterSpec based on the TF_CONFIG environment variable.
Returns:
A ClusterSpec with information from the TF_CONFIG environment variable."
3013,master,tensorflow/tensorflow/python/distribute/cluster_resolver/tfconfig_cluster_resolver.py,164,method,"Returns the master address to use when creating a TensorFlow session.
Note: this is only useful for TensorFlow 1.x.
Args:
task_type: (String, optional) Overrides and sets the task_type of the
master.
task_id: (Integer, optional) Overrides and sets the task id of the
master.
rpc_layer: (String, optional) Overrides and sets the protocol over which
TensorFlow nodes communicate with each other.
Returns:
The address of the master.
Raises:
RuntimeError: If the task_type or task_id is not specified and the
`TF_CONFIG` environment variable does not contain a task section."
3014,is_running_in_gce,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver.py,40,function,
3015,TPUClusterResolver,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver.py,51,class,"Cluster Resolver for Google Cloud TPUs.
This is an implementation of cluster resolvers for the Google Cloud TPU
service.
TPUClusterResolver supports the following distinct environments:
Google Compute Engine
Google Kubernetes Engine
Google internal
It can be passed into `tf.distribute.TPUStrategy` to support TF2 training on
Cloud TPUs."
3016,connect,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver.py,67,method,"Initializes TPU and returns a TPUClusterResolver.
This API will connect to remote TPU cluster and initialize the TPU
hardwares. Example usage:
>>> resolver = tf.distribute.cluster_resolver.TPUClusterResolver.connect(
... tpu='')
It can be viewed as convenient wrapper of the following code:
>>> resolver = tf.distribute.cluster_resolver.TPUClusterResolver(tpu='')
>>> tf.config.experimental_connect_to_cluster(resolver)
>>> tf.tpu.experimental.initialize_tpu_system(resolver)
Args:
tpu: A string corresponding to the TPU to use. It can be the TPU name or
TPU worker gRPC address. If not set, it will try automatically resolve
the TPU address on Cloud TPUs.
zone: Zone where the TPUs are located. If omitted or empty, we will assume
that the zone of the TPU is the same as the zone of the GCE VM, which we
will try to discover from the GCE metadata service.
project: Name of the GCP project containing Cloud TPUs. If omitted or
empty, we will try to discover the project name of the GCE VM from the
GCE metadata service.
Returns:
An instance of TPUClusterResolver object.
Raises:
NotFoundError: If no TPU devices found in eager mode."
3017,master,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver.py,215,method,"Get the Master string to be used for the session.
In the normal case, this returns the grpc path (grpc://1.2.3.4:8470) of
first instance in the ClusterSpec returned by the cluster_spec function.
If a non-TPU name is used when constructing a TPUClusterResolver, that will
be returned instead (e.g. If the tpus argument's value when constructing
this TPUClusterResolver was 'grpc://10.240.1.2:8470',
'grpc://10.240.1.2:8470' will be returned).
Args:
task_type: (Optional, string) The type of the TensorFlow task of the
master.
task_id: (Optional, integer) The index of the TensorFlow task of the
master.
rpc_layer: (Optional, string) The RPC protocol TensorFlow should use to
communicate with TPUs.
Returns:
string, the connection string to use when creating a session.
Raises:
ValueError: If none of the TPUs specified exists."
3018,get_master,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver.py,256,method,
3019,get_job_name,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver.py,259,method,
3020,get_tpu_system_metadata,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver.py,262,method,"Returns the metadata of the TPU system.
Users can call this method to get some facts of the TPU system, like
total number of cores, number of TPU workers and the devices. E.g.
```python
resolver = tf.distribute.cluster_resolver.TPUClusterResolver(tpu='')
tpu_system_medata = resolver.get_tpu_system_metadata()
num_hosts = tpu_system_medata.num_hosts
```
Returns:
A `tf.tpu.experimental.TPUSystemMetadata` object."
3021,cluster_spec,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver.py,287,method,"Returns a ClusterSpec object based on the latest TPU information.
We retrieve the information from the GCE APIs every time this method is
called.
Returns:
A ClusterSpec containing host information returned from Cloud TPUs,
or None.
Raises:
RuntimeError: If the provided TPU is not healthy."
3022,num_accelerators,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver.py,323,method,"Returns the number of TPU cores per worker.
Connects to the master and list all the devices present in the master,
and counts them up. Also verifies that the device counts per host in the
cluster is the same before returning the number of TPU cores per host.
Args:
task_type: Unused.
task_id: Unused.
config_proto: Used to create a connection to a TPU master in order to
retrieve the system metadata.
Raises:
RuntimeError: If we cannot talk to a TPU worker after retrying or if the
number of TPU devices per host is different."
3023,environment,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver.py,368,method,Returns the current environment which TensorFlow is running in.
3024,MockRequestClass,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver_test.py,47,class,
3025,execute,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver_test.py,53,method,
3026,MockNodeClass,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver_test.py,60,class,
3027,get,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver_test.py,65,method,
3028,mock_request_compute_metadata,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver_test.py,69,function,
3029,mock_is_running_in_gce,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver_test.py,80,function,
3030,mock_is_not_running_in_gce,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver_test.py,84,function,
3031,mock_running_in_gce_urlopen,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver_test.py,88,function,
3032,mock_not_running_in_gce_urlopen,tensorflow/tensorflow/python/distribute/cluster_resolver/tpu/tpu_cluster_resolver_test.py,95,function,
3033,ParallelDevice,tensorflow/tensorflow/python/distribute/parallel_device/parallel_device.py,42,class,A device which executes operations in parallel.
3034,pack,tensorflow/tensorflow/python/distribute/parallel_device/parallel_device.py,70,method,"Create a tensor on the parallel device from a sequence of tensors.
Args:
tensors: A flat list of tensors, one per device in `self.components`.
Returns:
A single tensor placed on `self.name`."
3035,unpack,tensorflow/tensorflow/python/distribute/parallel_device/parallel_device.py,82,method,"Unpack a parallel tensor into its components.
Args:
parallel_tensor: A tensor placed on `self.name`.
Returns:
A flat list of tensors, one per `self.components`."
3036,device_ids,tensorflow/tensorflow/python/distribute/parallel_device/parallel_device.py,96,method,"A parallel tensor with scalar integers numbering component devices.
Each device ID is placed on its corresponding device, in the same order as
the `components` constructor argument.
Returns:
A parallel tensor containing 0 on the first device, 1 on the second, etc."
3037,scope,tensorflow/tensorflow/python/distribute/parallel_device/parallel_device.py,112,method,"Runs ops in parallel, makes variables which save independent buffers."
3038,VariableWithFixedCheckpointing,tensorflow/tensorflow/python/distribute/parallel_device/saving.py,86,class,Overrides checkpointing behavior to save like a partitioned variable.
3039,independent_buffers,tensorflow/tensorflow/python/distribute/parallel_device/saving.py,117,function,"Context manager which saves parallel buffers independently.
Creates a ParallelDevice-aware variable subclass which saves buffers for each
device separately.
Args:
parallel_device: A ParallelDevice object on which variables are placed.
Yields:
Nothing."
3040,to_dlpack,tensorflow/tensorflow/python/dlpack/dlpack.py,27,function,"Returns the dlpack capsule representing the tensor.
This operation ensures the underlying data memory is ready when returns.
```python
a = tf.tensor([1, 10])
dlcapsule = tf.experimental.dlpack.to_dlpack(a)
# dlcapsule represents the dlpack data structure
```
Args:
tf_tensor: Tensorflow eager tensor, to be converted to dlpack capsule.
Returns:
A PyCapsule named as dltensor, which shares the underlying memory to other
framework. This PyCapsule can be consumed only once."
3041,from_dlpack,tensorflow/tensorflow/python/dlpack/dlpack.py,49,function,"Returns the Tensorflow eager tensor.
The returned tensor uses the memory shared by dlpack capsules from other
framework.
```python
a = tf.experimental.dlpack.from_dlpack(dlcapsule)
# `a` uses the memory shared by dlpack
```
Args:
dlcapsule: A PyCapsule named as dltensor
Returns:
A Tensorflow eager tensor"
3042,FormatShapeAndDtype,tensorflow/tensorflow/python/dlpack/dlpack_test.py,40,function,
3043,op_attr_type,tensorflow/tensorflow/python/eager/backprop.py,75,function,
3044,make_attr,tensorflow/tensorflow/python/eager/backprop.py,86,function,
3045,implicit_val_and_grad,tensorflow/tensorflow/python/eager/backprop.py,183,function,"Returns a function which differentiates f with respect to variables.
The wrapped function returns the value and the gradient of f when called with
the same arguments. The gradient is with respect to all trainable TFE
variables accessed by `f`.
This function is useful when the exact set of variables to differentiate with
is not known ahead of time.
Example:
```python
dense_layer = tf.compat.v1.layers.Dense(1)
def loss(x, y):
return tf.reduce_sum(tf.square(dense_layer(x) - y))
# Obtain the gradient function.
val_grad_fn = tfe.implicit_value_and_gradients(loss)
# Invoke the gradient function with concrete values of x and y.
x = tf.constant([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])
y = tf.constant([[10.0], [20.0]])
value, grads_and_vars = val_grad_fn(x, y)
print('Value of loss: %s' % value)
# Apply the gradients to Variables.
optimizer = tf.compat.v1.train.GradientDescentOptimizer(0.1)
optimizer.apply_gradients(grads_and_vars)
```
Args:
f: function to be differentiated. If `f` returns a scalar, this scalar will
be differentiated. If `f` returns a tensor or list of tensors, by default
a scalar will be computed by adding all their values to produce a single
scalar.
Returns:
A function which, when called, returns a tuple pair.
Its first element is the value to which the function evaluates.
Its second element is list of (gradient, variable) pairs.
Raises:
ValueError: if `f` returns None."
3046,implicit_grad,tensorflow/tensorflow/python/eager/backprop.py,262,function,"Returns a function which differentiates f with respect to variables.
The wrapped function returns the gradient of f when called with the same
arguments. The gradient is with respect to all trainable TFE variables
accessed by `f`.
This function is useful when the exact set of variables to differentiate with
is not known ahead of time.
Example:
```python
dense_layer = tf.compat.v1.layers.Dense(1)
def loss(x, y):
return tf.reduce_sum(tf.square(dense_layer(x) - y))
# Obtain the gradient function.
grad_fn = tfe.implicit_gradients(loss)
# Invoke the gradient function with concrete values of x and y.
x = tf.constant([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])
y = tf.constant([[10.0], [20.0]])
grads_and_vars = grad_fn(x, y)
# Apply the gradients to Variables.
optimizer = tf.compat.v1.train.GradientDescentOptimizer(0.1)
optimizer.apply_gradients(grads_and_vars)
```
Args:
f: function to be differentiated. If `f` returns a scalar, this scalar will
be differentiated. If `f` returns a tensor or list of tensors, by default
a scalar will be computed by adding all their values to produce a single
scalar.
Returns:
A function which, when called, returns a list of (gradient, variable) pairs."
3047,gradients_function,tensorflow/tensorflow/python/eager/backprop.py,340,function,"Returns a function which differentiates f with respect to params.
Example:
```python
# f(x, y) = (x ^ 3) * y - x * (y ^ 2)
# Therefore, the 1st order derivatives are:
# df / dx = 3 * (x ^ 2) * y - y ^ 2
# df / dy = x ^ 3 - 2 * x * y
# The 2nd order derivatives with respect to x is:
# d^2 f / (dx)^2 = 6 * x * y
def f(x, y):
return x * x * x * y - x * y * y
# Obtain a function that returns 1st order gradients.
grad_fn = tfe.gradients_function(f)
x = 2.0
y = 3.0
# Invoke the 1st order gradient function.
x_grad, y_grad = grad_fn(x, y)
assert x_grad.numpy() == 3 * (2 ** 2) * 3 - 3 ** 2
assert y_grad.numpy() == (2 ** 3) - 2 * 2 * 3
# Obtain a function that returns the 2nd order gradient with respect to x.
gradgrad_fn = tfe.gradients_function(lambda x, y: grad_fn(x, y)[0])
# Invoke the 2nd order gradient function.
x_gradgrad = gradgrad_fn(x, y)[0]
assert x_gradgrad.numpy() == 6 * 2 * 3
# To obtain a callable that returns the gradient(s) of `f` with respect to a
# subset of its inputs, use the `params` keyword argument with
# `gradients_function()`.
ygrad_fn = tfe.gradients_function(f, params=[1])
(y_grad,) = ygrad_fn(x, y)
assert y_grad.numpy() == (2 ** 3) - 2 * 2 * 3
```
Note that only tensors with real or complex dtypes are differentiable.
Args:
f: function to be differentiated. If `f` returns a scalar, this scalar will
be differentiated. If `f` returns a tensor or list of tensors, by default
a scalar will be computed by adding all their values to produce a single
scalar. If desired, the tensors can be elementwise multiplied by the
tensors passed as the `dy` keyword argument to the returned gradient
function.
params: list of parameter names of f or list of integers indexing the
parameters with respect to which we'll differentiate. Passing None
differentiates with respect to all parameters.
Returns:
function which, when called, returns the value of f and the gradient
of `f` with respect to all of `params`. The function takes an extra optional
keyword argument `dy`. Setting it allows computation of vector jacobian
products for vectors other than the vector of ones.
Raises:
ValueError: if the params are not all strings or all integers."
3048,val_and_grad_function,tensorflow/tensorflow/python/eager/backprop.py,445,function,"Returns a function that computes f and its derivative w.r.t. params.
Example:
```python
# f(x, y) = (x ^ 3) * y - x * (y ^ 2)
# Therefore, the 1st order derivatives are:
# df / dx = 3 * (x ^ 2) * y - y ^ 2
# df / dy = x ^ 3 - 2 * x * y
def f(x, y):
return x * x * x * y - x * y * y
# Obtain a function that returns the function value and the 1st order
# gradients.
val_grads_fn = tfe.value_and_gradients_function(f)
x = 2.0
y = 3.0
# Invoke the value-and-gradients function.
f_val, (x_grad, y_grad) = val_grads_fn(x, y)
assert f_val.numpy() == (2 ** 3) * 3 - 2 * (3 ** 2)
assert x_grad.numpy() == 3 * (2 ** 2) * 3 - 3 ** 2
assert y_grad.numpy() == (2 ** 3) - 2 * 2 * 3
# To obtain a callable that returns the value of `f` and the gradient(s) of
# `f` with respect to a subset of its inputs, use the `params` keyword
# argument with `value_and_gradients_function()`.
val_ygrad_fn = tfe.value_and_gradients_function(f, params=[1])
f_val, (y_grad,) = val_ygrad_fn(x, y)
assert f_val.numpy() == (2 ** 3) * 3 - 2 * (3 ** 2)
assert y_grad.numpy() == (2 ** 3) - 2 * 2 * 3
```
Args:
f: function to be differentiated. If `f` returns a scalar, this scalar will
be differentiated. If `f` returns a tensor or list of tensors, by default
a scalar will be computed by adding all their values to produce a single
scalar. If desired, the tensors can be elementwise multiplied by the
tensors passed as the `dy` keyword argument to the returned gradient
function.
params: list of parameter names of f or list of integers indexing the
parameters with respect to which we'll differentiate. Passing `None`
differentiates with respect to all parameters.
Returns:
function which, when called, returns the value of f and the gradient
of f with respect to all of `params`. The function takes an extra optional
keyword argument ""dy"". Setting it allows computation of vector jacobian
products for vectors other than the vector of ones.
Raises:
ValueError: if the params are not all strings or all integers."
3049,make_vjp,tensorflow/tensorflow/python/eager/backprop.py,513,function,"Returns a function that computes f and its vjp w.r.t.
params.
The term ""vjp"" here is an abbreviation for vector-jacobian product.
Args:
f: the function to be differentiated.
params: the parameters (numbers or names) to differentiate with respect to.
A value of None will differentiate with respect to all parameters.
persistent: Boolean controlling whether the VJP function can be re-used.
Must be True or False.
Returns:
A function, which when called, returns a tuple (value, vjp), where:
- value is the result of calling f.
- vjp is a function, which takes a vector as an argument and
returns the product of that vector with the Jacobian of f.
Providing no argument to vjp is equivalent to providing a
vector of ones.
For example,
```python
def f(x):
return x * x
wrapped_fn = tfe.make_vjp(f)
result, vjp = wrapped_fn(tf.constant(3.0))
# result is 9.0
vjp() # the vjp function returns 6.0
Raises:
ValueError: if `f` returns None."
3050,flatten_nested_indexed_slices,tensorflow/tensorflow/python/eager/backprop.py,589,function,
3051,aggregate_indexed_slices_gradients,tensorflow/tensorflow/python/eager/backprop.py,601,function,Aggregates gradients containing `IndexedSlices`s.
3052,GradientTape,tensorflow/tensorflow/python/eager/backprop.py,736,class,"Record operations for automatic differentiation.
Operations are recorded if they are executed within this context manager and
at least one of their inputs is being ""watched"".
Trainable variables (created by `tf.Variable` or `tf.compat.v1.get_variable`,
where `trainable=True` is default in both cases) are automatically watched.
Tensors can be manually watched by invoking the `watch` method on this context
manager.
For example, consider the function `y = x * x`. The gradient at `x = 3.0` can
be computed as:
>>> x = tf.constant(3.0)
>>> with tf.GradientTape() as g:
... g.watch(x)
... y = x * x
>>> dy_dx = g.gradient(y, x)
>>> print(dy_dx)
tf.Tensor(6.0, shape=(), dtype=float32)
GradientTapes can be nested to compute higher-order derivatives. For example,
>>> x = tf.constant(5.0)
>>> with tf.GradientTape() as g:
... g.watch(x)
... with tf.GradientTape() as gg:
... gg.watch(x)
... y = x * x
... dy_dx = gg.gradient(y, x) # dy_dx = 2 * x
>>> d2y_dx2 = g.gradient(dy_dx, x) # d2y_dx2 = 2
>>> print(dy_dx)
tf.Tensor(10.0, shape=(), dtype=float32)
>>> print(d2y_dx2)
tf.Tensor(2.0, shape=(), dtype=float32)
By default, the resources held by a GradientTape are released as soon as
GradientTape.gradient() method is called. To compute multiple gradients over
the same computation, create a persistent gradient tape. This allows multiple
calls to the gradient() method as resources are released when the tape object
is garbage collected. For example:
>>> x = tf.constant(3.0)
>>> with tf.GradientTape(persistent=True) as g:
... g.watch(x)
... y = x * x
... z = y * y
>>> dz_dx = g.gradient(z, x) # (4*x^3 at x = 3)
>>> print(dz_dx)
tf.Tensor(108.0, shape=(), dtype=float32)
>>> dy_dx = g.gradient(y, x)
>>> print(dy_dx)
tf.Tensor(6.0, shape=(), dtype=float32)
By default GradientTape will automatically watch any trainable variables that
are accessed inside the context. If you want fine grained control over which
variables are watched you can disable automatic tracking by passing
`watch_accessed_variables=False` to the tape constructor:
>>> x = tf.Variable(2.0)
>>> w = tf.Variable(5.0)
>>> with tf.GradientTape(
... watch_accessed_variables=False, persistent=True) as tape:
... tape.watch(x)
... y = x ** 2 # Gradients will be available for `x`.
... z = w ** 3 # No gradients will be available as `w` isn't being watched.
>>> dy_dx = tape.gradient(y, x)
>>> print(dy_dx)
tf.Tensor(4.0, shape=(), dtype=float32)
>>> # No gradients will be available as `w` isn't being watched.
>>> dz_dy = tape.gradient(z, w)
>>> print(dz_dy)
None
Note that when using models you should ensure that your variables exist when
using `watch_accessed_variables=False`. Otherwise it's quite easy to make your
first iteration not have any gradients:
```python
a = tf.keras.layers.Dense(32)
b = tf.keras.layers.Dense(32)
with tf.GradientTape(watch_accessed_variables=False) as tape:
tape.watch(a.variables) # Since `a.build` has not been called at this point
# `a.variables` will return an empty list and the
# tape will not be watching anything.
result = b(a(inputs))
tape.gradient(result, a.variables) # The result of this computation will be
# a list of `None`s since a's variables
# are not being watched.
```
Note that only tensors with real or complex dtypes are differentiable."
3053,watch,tensorflow/tensorflow/python/eager/backprop.py,894,method,"Ensures that `tensor` is being traced by this tape.
Args:
tensor: a Tensor or list of Tensors.
Raises:
ValueError: if it encounters something that is not a tensor."
3054,stop_recording,tensorflow/tensorflow/python/eager/backprop.py,920,method,"Temporarily stops recording operations on this tape.
Operations executed while this context manager is active will not be
recorded on the tape. This is useful for reducing the memory used by tracing
all computations.
For example:
>>> x = tf.constant(4.0)
>>> with tf.GradientTape() as tape:
... with tape.stop_recording():
... y = x ** 2
>>> dy_dx = tape.gradient(y, x)
>>> print(dy_dx)
None
Yields:
None
Raises:
RuntimeError: if the tape is not currently recording."
3055,reset,tensorflow/tensorflow/python/eager/backprop.py,951,method,"Clears all information stored in this tape.
Equivalent to exiting and reentering the tape context manager with a new
tape. For example, the two following code blocks are equivalent:
```
with tf.GradientTape() as t:
loss = loss_fn()
with tf.GradientTape() as t:
loss += other_loss_fn()
t.gradient(loss, ...) # Only differentiates other_loss_fn, not loss_fn
# The following is equivalent to the above
with tf.GradientTape() as t:
loss = loss_fn()
t.reset()
loss += other_loss_fn()
t.gradient(loss, ...) # Only differentiates other_loss_fn, not loss_fn
```
This is useful if you don't want to exit the context manager for the tape,
or can't because the desired reset point is inside a control flow construct:
```
with tf.GradientTape() as t:
loss = ...
if loss > k:
t.reset()
```"
3056,watched_variables,tensorflow/tensorflow/python/eager/backprop.py,987,method,Returns variables watched by this tape in order of construction.
3057,gradient,tensorflow/tensorflow/python/eager/backprop.py,993,method,"Computes the gradient using operations recorded in context of this tape.
Args:
target: a list or nested structure of Tensors or Variables to be
differentiated.
sources: a list or nested structure of Tensors or Variables. `target`
will be differentiated against elements in `sources`.
output_gradients: a list of gradients, one for each element of
target. Defaults to None.
unconnected_gradients: a value which can either hold 'none' or 'zero' and
alters the value which will be returned if the target and sources are
unconnected. The possible values and effects are detailed in
'UnconnectedGradients' and it defaults to 'none'.
Returns:
a list or nested structure of Tensors (or IndexedSlices, or None),
one for each element in `sources`. Returned structure is the same as
the structure of `sources`.
Raises:
RuntimeError: if called inside the context of the tape, or if called more
than once on a non-persistent tape.
ValueError: if the target is a variable or if unconnected gradients is
called with an unknown value."
3058,jacobian,tensorflow/tensorflow/python/eager/backprop.py,1096,method,"Computes the jacobian using operations recorded in context of this tape.
See[wikipedia article](http://en.wikipedia.org/wiki/jacobian_matrix_and_determinant)
for the definition of a Jacobian.
Example usage:
```python
with tf.GradientTape() as g:
x = tf.constant([1.0, 2.0])
g.watch(x)
y = x * x
jacobian = g.jacobian(y, x)
# jacobian value is [[2., 0.], [0., 4.]]
```
Args:
target: Tensor to be differentiated.
sources: a list or nested structure of Tensors or Variables. `target`
will be differentiated against elements in `sources`.
unconnected_gradients: a value which can either hold 'none' or 'zero' and
alters the value which will be returned if the target and sources are
unconnected. The possible values and effects are detailed in
'UnconnectedGradients' and it defaults to 'none'.
parallel_iterations: A knob to control how many iterations are dispatched
in parallel. This knob can be used to control the total memory usage.
experimental_use_pfor: If true, vectorizes the jacobian computation. Else
falls back to a sequential while_loop. Vectorization can sometimes fail
or lead to excessive memory usage. This option can be used to disable
vectorization in such cases.
Returns:
A list or nested structure of Tensors (or None), one for each element in
`sources`. Returned structure is the same as the structure of `sources`.
Note if any gradient is sparse (IndexedSlices), jacobian function
currently makes it dense and returns a Tensor instead. This may change in
the future.
Raises:
RuntimeError: If called on a non-persistent tape with eager execution
enabled and without enabling experimental_use_pfor.
ValueError: If vectorization of jacobian computation fails."
3059,batch_jacobian,tensorflow/tensorflow/python/eager/backprop.py,1206,method,"Computes and stacks per-example jacobians.
See [wikipedia article](http://en.wikipedia.org/wiki/jacobian_matrix_and_determinant)
for the definition of a Jacobian. This function is essentially an efficient
implementation of the following:
`tf.stack([self.jacobian(y[i], x[i]) for i in range(x.shape[0])])`.
Note that compared to `GradientTape.jacobian` which computes gradient of
each output value w.r.t each input value, this function is useful when
`target[i,...]` is independent of `source[j,...]` for `j != i`. This
assumption allows more efficient computation as compared to
`GradientTape.jacobian`. The output, as well as intermediate activations,
are lower dimensional and avoid a bunch of redundant zeros which would
result in the jacobian computation given the independence assumption.
Example usage:
```python
with tf.GradientTape() as g:
x = tf.constant([[1., 2.], [3., 4.]], dtype=tf.float32)
g.watch(x)
y = x * x
batch_jacobian = g.batch_jacobian(y, x)
# batch_jacobian is [[[2, 0], [0, 4]], [[6, 0], [0, 8]]]
```
Args:
target: A tensor with rank 2 or higher and with shape [b, y1, ..., y_n].
`target[i,...]` should only depend on `source[i,...]`.
source: A tensor with rank 2 or higher and with shape [b, x1, ..., x_m].
unconnected_gradients: a value which can either hold 'none' or 'zero' and
alters the value which will be returned if the target and sources are
unconnected. The possible values and effects are detailed in
'UnconnectedGradients' and it defaults to 'none'.
parallel_iterations: A knob to control how many iterations are dispatched
in parallel. This knob can be used to control the total memory usage.
experimental_use_pfor: If true, uses pfor for computing the Jacobian. Else
uses a tf.while_loop.
Returns:
A tensor `t` with shape [b, y_1, ..., y_n, x1, ..., x_m] where `t[i, ...]`
is the jacobian of `target[i, ...]` w.r.t. `source[i, ...]`, i.e. stacked
per-example jacobians.
Raises:
RuntimeError: If called on a non-persistent tape with eager execution
enabled and without enabling experimental_use_pfor.
ValueError: If vectorization of jacobian computation fails or if first
dimension of `target` and `source` do not match."
3060,loop_fn,tensorflow/tensorflow/python/eager/backprop.py,1159,method,
3061,loop_fn,tensorflow/tensorflow/python/eager/backprop.py,1297,method,
3062,IsTrainable,tensorflow/tensorflow/python/eager/backprop_util.py,25,function,
3063,c_tfe_py_fastpath_execute,tensorflow/tensorflow/python/eager/benchmarks_test.py,74,function,
3064,run_benchmark,tensorflow/tensorflow/python/eager/benchmarks_test.py,95,function,
3065,MicroBenchmarks,tensorflow/tensorflow/python/eager/benchmarks_test.py,112,class,
3066,benchmark_create_np_array,tensorflow/tensorflow/python/eager/benchmarks_test.py,155,method,
3067,benchmark_create_float_constant,tensorflow/tensorflow/python/eager/benchmarks_test.py,189,method,
3068,benchmark_create_float_constant_uncached,tensorflow/tensorflow/python/eager/benchmarks_test.py,192,method,
3069,benchmark_create_int32_constant,tensorflow/tensorflow/python/eager/benchmarks_test.py,195,method,
3070,benchmark_create_int32_constant_uncached,tensorflow/tensorflow/python/eager/benchmarks_test.py,201,method,
3071,benchmark_add_float_scalars,tensorflow/tensorflow/python/eager/benchmarks_test.py,225,method,
3072,benchmark_add_int32_scalars,tensorflow/tensorflow/python/eager/benchmarks_test.py,228,method,
3073,benchmark_add_float_scalar_tensor,tensorflow/tensorflow/python/eager/benchmarks_test.py,231,method,
3074,benchmark_add_float_scalar_tensor_overloaded_operator,tensorflow/tensorflow/python/eager/benchmarks_test.py,236,method,
3075,benchmark_add_int32_scalar_tensor,tensorflow/tensorflow/python/eager/benchmarks_test.py,241,method,
3076,benchmark_add_float_dense_tensor,tensorflow/tensorflow/python/eager/benchmarks_test.py,246,method,
3077,benchmark_add_int32_dense_tensor,tensorflow/tensorflow/python/eager/benchmarks_test.py,251,method,
3078,benchmark_create_float_tensor_from_list_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,257,method,
3079,benchmark_create_float_tensor_from_np_array_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,261,method,
3080,benchmark_create_int32_tensor_from_list_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,267,method,
3081,benchmark_create_int32_tensor_from_np_array_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,271,method,
3082,benchmark_create_float_tensor_from_list_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,276,method,
3083,benchmark_create_float_tensor_from_np_array_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,282,method,
3084,benchmark_create_int32_tensor_from_list_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,290,method,
3085,benchmark_create_int32_tensor_from_np_array_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,297,method,
3086,benchmark_index_tensor_with_literal,tensorflow/tensorflow/python/eager/benchmarks_test.py,305,method,
3087,benchmark_index_tensor_with_tensor,tensorflow/tensorflow/python/eager/benchmarks_test.py,310,method,
3088,benchmark_index_tensor_with_np_array,tensorflow/tensorflow/python/eager/benchmarks_test.py,315,method,
3089,benchmark_np_multiply,tensorflow/tensorflow/python/eager/benchmarks_test.py,337,method,
3090,benchmark_tf_multiply_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,340,method,
3091,benchmark_tf_multiply_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,346,method,
3092,benchmark_tf_multiply_op_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,353,method,
3093,benchmark_tf_multiply_op_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,359,method,
3094,benchmark_tf_conv2d_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,366,method,
3095,benchmark_tf_conv2d_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,373,method,
3096,benchmark_tf_identity,tensorflow/tensorflow/python/eager/benchmarks_test.py,381,method,
3097,benchmark_slowpath_tf_identity,tensorflow/tensorflow/python/eager/benchmarks_test.py,386,method,
3098,benchmark_tfe_py_execute_identity,tensorflow/tensorflow/python/eager/benchmarks_test.py,389,method,
3099,benchmark_tf_gradient_function_identity,tensorflow/tensorflow/python/eager/benchmarks_test.py,401,method,
3100,benchmark_tf_gradient_forward_identity,tensorflow/tensorflow/python/eager/benchmarks_test.py,409,method,
3101,benchmark_tf_gradient_tape_push_pop,tensorflow/tensorflow/python/eager/benchmarks_test.py,416,method,
3102,benchmark_tf_gradient_function_no_op,tensorflow/tensorflow/python/eager/benchmarks_test.py,425,method,
3103,benchmark_np_matmul_2_by_2,tensorflow/tensorflow/python/eager/benchmarks_test.py,535,method,
3104,benchmark_tf_matmul_2_by_2_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,539,method,
3105,benchmark_tf_matmul_2_by_2_CPU_async,tensorflow/tensorflow/python/eager/benchmarks_test.py,546,method,
3106,benchmark_gen_math_ops_matmul_2_by_2_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,555,method,
3107,benchmark_tfe_py_fastpath_execute_matmul_2_by_2_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,561,method,
3108,benchmark_tfe_py_execute_matmul_2_by_2_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,567,method,
3109,benchmark_defun_matmul_2_by_2_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,573,method,
3110,benchmark_defun_matmul_2_by_2_CPU_async,tensorflow/tensorflow/python/eager/benchmarks_test.py,580,method,
3111,benchmark_defun_matmul_forward_backward_2_by_2_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,589,method,
3112,benchmark_defun_matmul_forward_backward_2_by_2_CPU_async,tensorflow/tensorflow/python/eager/benchmarks_test.py,596,method,
3113,benchmark_tf_matmul_2_by_2_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,606,method,
3114,benchmark_tf_matmul_2_by_2_GPU_async,tensorflow/tensorflow/python/eager/benchmarks_test.py,615,method,
3115,benchmark_gen_math_ops_matmul_2_by_2_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,627,method,
3116,benchmark_tfe_py_execute_matmul_2_by_2_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,636,method,
3117,benchmark_defun_matmul_2_by_2_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,645,method,
3118,benchmark_defun_args_matmul_2_by_2_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,654,method,
3119,benchmark_defun_matmul_2_by_2_GPU_async,tensorflow/tensorflow/python/eager/benchmarks_test.py,662,method,
3120,benchmark_nested_defun_matmul_2_by_2,tensorflow/tensorflow/python/eager/benchmarks_test.py,673,method,
3121,benchmark_np_matmul_100_by_784,tensorflow/tensorflow/python/eager/benchmarks_test.py,679,method,
3122,benchmark_tf_matmul_100_by_784_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,685,method,
3123,benchmark_tf_matmul_100_by_784_CPU_async,tensorflow/tensorflow/python/eager/benchmarks_test.py,692,method,
3124,benchmark_gen_math_ops_matmul_100_by_784_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,701,method,
3125,benchmark_tfe_py_fastpath_execute_matmul_100_by_784_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,707,method,
3126,benchmark_tfe_py_execute_matmul_100_by_784_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,714,method,
3127,benchmark_defun_matmul_100_by_784_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,720,method,
3128,benchmark_tf_matmul_100_by_784_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,727,method,
3129,benchmark_tf_matmul_100_by_784_GPU_async,tensorflow/tensorflow/python/eager/benchmarks_test.py,736,method,
3130,benchmark_gen_math_ops_matmul_100_by_784_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,748,method,
3131,benchmark_tfe_py_execute_matmul_100_by_784_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,757,method,
3132,benchmark_defun_matmul_100_by_784_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,766,method,
3133,benchmark_nested_defun_matmul_100_by_784_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,775,method,
3134,benchmark_forwardprop_matmul_256_by_2096_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,847,method,
3135,benchmark_forwardprop_in_defun_matmul_256_by_2096_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,850,method,
3136,benchmark_forwardprop_in_defun_of_defun_matmul_256_by_2096_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,853,method,
3137,benchmark_forwardprop_of_defun_matmul_256_by_2096_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,856,method,
3138,benchmark_forwardprop_matmul_100_by_784_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,859,method,
3139,benchmark_forwardprop_in_defun_matmul_100_by_784_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,862,method,
3140,benchmark_forwardprop_in_defun_of_defun_matmul_100_by_784_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,865,method,
3141,benchmark_forwardprop_of_defun_matmul_100_by_784_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,868,method,
3142,benchmark_tf_reduce_logsumexp_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,887,method,
3143,benchmark_tf_reduce_logsumexp_CPU_async,tensorflow/tensorflow/python/eager/benchmarks_test.py,891,method,
3144,benchmark_tf_reduce_logsumexp_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,895,method,
3145,benchmark_tf_reduce_logsumexp_GPU_async,tensorflow/tensorflow/python/eager/benchmarks_test.py,899,method,
3146,benchmark_tf_reduce_logsumexp_CPU_defunc,tensorflow/tensorflow/python/eager/benchmarks_test.py,904,method,
3147,benchmark_tf_reduce_logsumexp_CPU_async_defun,tensorflow/tensorflow/python/eager/benchmarks_test.py,908,method,
3148,benchmark_tf_reduce_logsumexp_GPU_defun,tensorflow/tensorflow/python/eager/benchmarks_test.py,913,method,
3149,benchmark_tf_reduce_logsumexp_GPU_async_defun,tensorflow/tensorflow/python/eager/benchmarks_test.py,917,method,
3150,benchmark_tf_reduce_logsumexp_GPU_defun_compile,tensorflow/tensorflow/python/eager/benchmarks_test.py,922,method,
3151,benchmark_tf_reduce_logsumexp_GPU_async_defun_compile,tensorflow/tensorflow/python/eager/benchmarks_test.py,927,method,
3152,benchmark_tf_tensordot_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,939,method,
3153,benchmark_tf_tensordot_CPU_async,tensorflow/tensorflow/python/eager/benchmarks_test.py,943,method,
3154,benchmark_tf_tensordot_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,947,method,
3155,benchmark_tf_tensordot_GPU_async,tensorflow/tensorflow/python/eager/benchmarks_test.py,951,method,
3156,benchmark_tf_zeros_2_by_2_float32_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,960,method,
3157,benchmark_tf_zeros_2_by_2_bool_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,964,method,
3158,benchmark_tf_zeros_2_by_2_string_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,968,method,
3159,benchmark_tf_zeros_2_by_2_float32_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,972,method,
3160,benchmark_tf_zeros_2_by_2_bool_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,976,method,
3161,benchmark_tf_zeros_30_by_30_float32_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,980,method,
3162,benchmark_tf_zeros_30_by_30_bool_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,984,method,
3163,benchmark_tf_zeros_30_by_30_string_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,988,method,
3164,benchmark_tf_zeros_30_by_30_float32_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,992,method,
3165,benchmark_tf_zeros_30_by_30_bool_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,996,method,
3166,benchmark_tf_zeros_100_by_100_float32_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1000,method,
3167,benchmark_tf_zeros_100_by_100_bool_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1004,method,
3168,benchmark_tf_zeros_100_by_100_string_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1008,method,
3169,benchmark_tf_zeros_100_by_100_float32_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1012,method,
3170,benchmark_tf_zeros_100_by_100_bool_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1016,method,
3171,benchmark_tf_zeros_like_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1024,method,
3172,benchmark_tf_zeros_like_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1027,method,
3173,benchmark_tf_zeros_like_variable_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1030,method,
3174,benchmark_tf_zeros_like_variable_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1034,method,
3175,benchmark_tf_random_uniform_2_by_2_integer_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1049,method,
3176,benchmark_tf_random_uniform_2_by_2_integer_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1052,method,
3177,benchmark_tf_random_uniform_2_by_2_float_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1055,method,
3178,benchmark_tf_random_uniform_2_by_2_float_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1058,method,
3179,benchmark_tf_random_uniform_2_by_2_default_setting_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1062,method,
3180,benchmark_tf_random_uniform_2_by_2_default_setting_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1067,method,
3181,benchmark_tf_dropout_scalar_rate_2_by_2_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1088,method,
3182,benchmark_tf_dropout_scalar_rate_2_by_2_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1091,method,
3183,benchmark_tf_dropout_2_by_2_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1094,method,
3184,benchmark_tf_dropout_2_by_2_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1097,method,
3185,benchmark_tf_transpose_2_by_2_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1109,method,
3186,benchmark_tf_transpose_2_by_2_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1115,method,
3187,benchmark_tf_transpose_variable_2_by_2_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1120,method,
3188,benchmark_tf_transpose_variable_2_by_2_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1126,method,
3189,benchmark_defun_without_signature,tensorflow/tensorflow/python/eager/benchmarks_test.py,1131,method,
3190,benchmark_defun_without_signature_and_with_kwargs,tensorflow/tensorflow/python/eager/benchmarks_test.py,1142,method,
3191,benchmark_defun_with_signature,tensorflow/tensorflow/python/eager/benchmarks_test.py,1154,method,
3192,benchmark_defun_with_signature_and_kwargs,tensorflow/tensorflow/python/eager/benchmarks_test.py,1166,method,
3193,benchmark_matmul_read_variable_op_2_by_2_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1179,method,
3194,benchmark_matmul_read_variable_op_with_tape_2_by_2_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1184,method,
3195,benchmark_read_variable_op_2_by_2_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1190,method,
3196,benchmark_read_variable_op_2_by_2_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1196,method,
3197,benchmark_read_variable_op_with_tape_2_by_2_CPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1203,method,
3198,benchmark_read_variable_op_with_tape_2_by_2_GPU,tensorflow/tensorflow/python/eager/benchmarks_test.py,1210,method,
3199,benchmarkScan,tensorflow/tensorflow/python/eager/benchmarks_test.py,1218,method,
3200,benchmarkScanDefun,tensorflow/tensorflow/python/eager/benchmarks_test.py,1227,method,
3201,benchmark_fastpath_conversion_type_inference,tensorflow/tensorflow/python/eager/benchmarks_test.py,1237,method,
3202,benchmark_convert_tensor,tensorflow/tensorflow/python/eager/benchmarks_test.py,1245,method,
3203,benchmark_convert_python_int,tensorflow/tensorflow/python/eager/benchmarks_test.py,1269,method,
3204,benchmark_convert_python_int_uncached,tensorflow/tensorflow/python/eager/benchmarks_test.py,1272,method,
3205,benchmark_convert_python_float,tensorflow/tensorflow/python/eager/benchmarks_test.py,1275,method,
3206,benchmark_convert_python_float_uncached,tensorflow/tensorflow/python/eager/benchmarks_test.py,1278,method,
3207,benchmark_convert_numpy_int,tensorflow/tensorflow/python/eager/benchmarks_test.py,1281,method,
3208,benchmark_convert_numpy_int_uncached,tensorflow/tensorflow/python/eager/benchmarks_test.py,1284,method,
3209,benchmark_convert_numpy_float,tensorflow/tensorflow/python/eager/benchmarks_test.py,1287,method,
3210,benchmark_convert_numpy_float_uncached,tensorflow/tensorflow/python/eager/benchmarks_test.py,1290,method,
3211,benchmark_convert_3x_list_to_tensor,tensorflow/tensorflow/python/eager/benchmarks_test.py,1294,method,
3212,benchmark_convert_3x_array_to_tensor,tensorflow/tensorflow/python/eager/benchmarks_test.py,1299,method,
3213,benchmark_constant_40x2_list_to_tensor,tensorflow/tensorflow/python/eager/benchmarks_test.py,1303,method,
3214,benchmark_constant_40x2_array_to_tensor,tensorflow/tensorflow/python/eager/benchmarks_test.py,1308,method,
3215,benchmark_constant_40x_list_of_2x_arrays_to_tensor,tensorflow/tensorflow/python/eager/benchmarks_test.py,1312,method,
3216,benchmark_constant_20x20x20_double_list_to_float32_tensor,tensorflow/tensorflow/python/eager/benchmarks_test.py,1316,method,
3217,benchmark_constant_20x20x20_double_list_to_float64_tensor,tensorflow/tensorflow/python/eager/benchmarks_test.py,1320,method,
3218,benchmark_list_of_zeros_to_np_array,tensorflow/tensorflow/python/eager/benchmarks_test.py,1324,method,
3219,benchmarkFunctionWithFiveResourceInputs,tensorflow/tensorflow/python/eager/benchmarks_test.py,1341,method,
3220,benchmarkFunctionWithFiveHundredResourceInputs,tensorflow/tensorflow/python/eager/benchmarks_test.py,1344,method,
3221,benchmarkTenThousandResourceReadsInCondInInnerFunc,tensorflow/tensorflow/python/eager/benchmarks_test.py,1379,method,
3222,benchmarkHundredResourceReadsInCondInInnerFunc,tensorflow/tensorflow/python/eager/benchmarks_test.py,1383,method,
3223,benchmarkTenResourceReadsInCondInInnerFunc,tensorflow/tensorflow/python/eager/benchmarks_test.py,1387,method,
3224,benchmark_tf_name_scope,tensorflow/tensorflow/python/eager/benchmarks_test.py,1390,method,
3225,benchmark_tf_nest_map_structure,tensorflow/tensorflow/python/eager/benchmarks_test.py,1398,method,
3226,benchmark_tf_nest_pack_sequence_as,tensorflow/tensorflow/python/eager/benchmarks_test.py,1406,method,
3227,benchmark_tf_nn_convolution_overhead,tensorflow/tensorflow/python/eager/benchmarks_test.py,1415,method,
3228,benchmark_tf_tensor_shape_creation_overhead,tensorflow/tensorflow/python/eager/benchmarks_test.py,1424,method,
3229,func,tensorflow/tensorflow/python/eager/benchmarks_test.py,165,method,
3230,cached_func,tensorflow/tensorflow/python/eager/benchmarks_test.py,174,method,
3231,uncached_func,tensorflow/tensorflow/python/eager/benchmarks_test.py,177,method,
3232,func,tensorflow/tensorflow/python/eager/benchmarks_test.py,208,method,
3233,func,tensorflow/tensorflow/python/eager/benchmarks_test.py,217,method,
3234,f,tensorflow/tensorflow/python/eager/benchmarks_test.py,395,method,
3235,f,tensorflow/tensorflow/python/eager/benchmarks_test.py,418,method,
3236,func,tensorflow/tensorflow/python/eager/benchmarks_test.py,443,method,
3237,func,tensorflow/tensorflow/python/eager/benchmarks_test.py,451,method,
3238,func,tensorflow/tensorflow/python/eager/benchmarks_test.py,465,method,
3239,defun_matmul,tensorflow/tensorflow/python/eager/benchmarks_test.py,482,method,
3240,outer,tensorflow/tensorflow/python/eager/benchmarks_test.py,492,method,
3241,func,tensorflow/tensorflow/python/eager/benchmarks_test.py,508,method,
3242,func,tensorflow/tensorflow/python/eager/benchmarks_test.py,1133,method,
3243,func,tensorflow/tensorflow/python/eager/benchmarks_test.py,1144,method,
3244,cache_computation,tensorflow/tensorflow/python/eager/benchmarks_test.py,1150,method,
3245,func,tensorflow/tensorflow/python/eager/benchmarks_test.py,1156,method,
3246,func,tensorflow/tensorflow/python/eager/benchmarks_test.py,1168,method,
3247,signature_computation,tensorflow/tensorflow/python/eager/benchmarks_test.py,1175,method,
3248,scan,tensorflow/tensorflow/python/eager/benchmarks_test.py,1221,method,
3249,scan,tensorflow/tensorflow/python/eager/benchmarks_test.py,1231,method,
3250,fn,tensorflow/tensorflow/python/eager/benchmarks_test.py,1240,method,
3251,fn,tensorflow/tensorflow/python/eager/benchmarks_test.py,1248,method,
3252,cached_func,tensorflow/tensorflow/python/eager/benchmarks_test.py,1257,method,
3253,uncached_func,tensorflow/tensorflow/python/eager/benchmarks_test.py,1260,method,
3254,add_all,tensorflow/tensorflow/python/eager/benchmarks_test.py,1332,method,
3255,benchmark_fn,tensorflow/tensorflow/python/eager/benchmarks_test.py,1354,method,
3256,fn,tensorflow/tensorflow/python/eager/benchmarks_test.py,1392,method,
3257,fn,tensorflow/tensorflow/python/eager/benchmarks_test.py,1401,method,
3258,fn,tensorflow/tensorflow/python/eager/benchmarks_test.py,1410,method,
3259,fn,tensorflow/tensorflow/python/eager/benchmarks_test.py,1419,method,
3260,fn,tensorflow/tensorflow/python/eager/benchmarks_test.py,1432,method,
3261,func,tensorflow/tensorflow/python/eager/benchmarks_test.py,785,method,
3262,compiled_function,tensorflow/tensorflow/python/eager/benchmarks_test.py,798,method,
3263,compiled_function,tensorflow/tensorflow/python/eager/benchmarks_test.py,817,method,
3264,func,tensorflow/tensorflow/python/eager/benchmarks_test.py,837,method,
3265,func,tensorflow/tensorflow/python/eager/benchmarks_test.py,1044,method,
3266,func,tensorflow/tensorflow/python/eager/benchmarks_test.py,1082,method,
3267,fn_with_many_reads,tensorflow/tensorflow/python/eager/benchmarks_test.py,1357,method,
3268,fn_with_many_reads_inner,tensorflow/tensorflow/python/eager/benchmarks_test.py,1360,method,
3269,then_branch,tensorflow/tensorflow/python/eager/benchmarks_test.py,1362,method,
3270,else_branch,tensorflow/tensorflow/python/eager/benchmarks_test.py,1365,method,
3271,MicroBenchmarksBase,tensorflow/tensorflow/python/eager/benchmarks_test_base.py,32,class,"Run and report benchmark results.
The first run is without any profilng.
Second run is with xprof and python trace. Third run is with xprof without
python trace. Note: xprof runs are with fewer iterations."
3272,run_with_xprof,tensorflow/tensorflow/python/eager/benchmarks_test_base.py,40,method,
3273,run_report,tensorflow/tensorflow/python/eager/benchmarks_test_base.py,53,method,Run and report benchmark results.
3274,CancellationManager,tensorflow/tensorflow/python/eager/cancellation.py,24,class,A mechanism for cancelling blocking computation.
3275,is_cancelled,tensorflow/tensorflow/python/eager/cancellation.py,33,method,Returns `True` if `CancellationManager.start_cancel` has been called.
3276,start_cancel,tensorflow/tensorflow/python/eager/cancellation.py,37,method,Cancels blocking operations that have been registered with this object.
3277,get_cancelable_function,tensorflow/tensorflow/python/eager/cancellation.py,41,method,
3278,FunctionCallOptions,tensorflow/tensorflow/python/eager/context.py,106,class,"Options applied at call sites of eager functions.
Eager functions are functions decorated with tf.contrib.eager.defun."
3279,executor_type,tensorflow/tensorflow/python/eager/context.py,132,method,
3280,executor_type,tensorflow/tensorflow/python/eager/context.py,136,method,
3281,config_proto_serialized,tensorflow/tensorflow/python/eager/context.py,140,method,
3282,config_proto_serialized,tensorflow/tensorflow/python/eager/context.py,144,method,
3283,LogicalDevice,tensorflow/tensorflow/python/eager/context.py,252,class,"Abstraction for a logical device initialized by the runtime.
A `tf.config.LogicalDevice` corresponds to an initialized logical device on a
`tf.config.PhysicalDevice` or a remote device visible to the cluster. Tensors
and operations can be placed on a specific logical device by calling
`tf.device` with a specified `tf.config.LogicalDevice`.
Fields:
name: The fully qualified name of the device. Can be used for Op or function
placement.
device_type: String declaring the type of device such as ""CPU"" or ""GPU""."
3284,LogicalDeviceConfiguration,tensorflow/tensorflow/python/eager/context.py,271,class,"Configuration class for a logical devices.
The class specifies the parameters to configure a `tf.config.PhysicalDevice`
as it is initialized to a `tf.config.LogicalDevice` during runtime
initialization. Not all fields are valid for all device types.
See `tf.config.get_logical_device_configuration` and
`tf.config.set_logical_device_configuration` for usage examples.
Fields:
memory_limit: (optional) Maximum memory (in MB) to allocate on the virtual
device. Currently only supported for GPUs.
experimental_priority: (optional) Priority to assign to a virtual device.
Lower values have higher priorities and 0 is the default.
Within a physical GPU, the GPU scheduler will prioritize ops on virtual
devices with higher priority. Currently only supported for Nvidia GPUs."
3285,PhysicalDevice,tensorflow/tensorflow/python/eager/context.py,298,class,"Abstraction for a locally visible physical device.
TensorFlow can utilize various devices such as the CPU or multiple GPUs
for computation. Before initializing a local device for use, the user can
customize certain properties of the device such as it's visibility or memory
configuration.
Once a visible `tf.config.PhysicalDevice` is initialized one or more
`tf.config.LogicalDevice` objects are created. Use
`tf.config.set_visible_devices` to configure the visibility of a physical
device and `tf.config.set_logical_device_configuration` to configure multiple
`tf.config.LogicalDevice` objects for a `tf.config.PhysicalDevice`. This is
useful when separation between models is needed or to simulate a multi-device
environment.
Fields:
name: Unique identifier for device.
device_type: String declaring the type of device such as ""CPU"" or ""GPU""."
3286,is_tfrt_enabled,tensorflow/tensorflow/python/eager/context.py,358,function,
3287,Context,tensorflow/tensorflow/python/eager/context.py,370,class,Environment in which eager operations execute.
3288,ensure_initialized,tensorflow/tensorflow/python/eager/context.py,524,method,Initialize handle and devices if not already done so.
3289,get_server_def,tensorflow/tensorflow/python/eager/context.py,572,method,
3290,set_server_def,tensorflow/tensorflow/python/eager/context.py,575,method,"Allow setting a server_def on the context.
When a server def is replaced, it effectively clears a bunch of caches
within the context. If you attempt to use a tensor object that was pointing
to a tensor on the remote device, it will raise an error.
Args:
server_def: A tensorflow::ServerDef proto.
Enables execution on remote devices.
keep_alive_secs: Num. seconds after which the remote end will hang up.
As long as the client is still alive, the server state for the context
will be kept alive. If the client is killed (or there is some failure),
the server will clean up its context keep_alive_secs after the final RPC
it receives.
Raises:
ValueError: if server_def is None."
3291,update_server_def,tensorflow/tensorflow/python/eager/context.py,608,method,"Update a server_def on the context.
Args:
server_def: A tensorflow::ServerDef proto. Enables execution on remote
devices.
keep_alive_secs: Num. seconds after which the remote end will hang up. As
long as the client is still alive, the server state for the context will
be kept alive. If the client is killed (or there is some failure), the
server will clean up its context keep_alive_secs after the final RPC it
receives.
Raises:
ValueError: if server_def is None."
3292,check_alive,tensorflow/tensorflow/python/eager/context.py,636,method,"Checks whether a remote worker is alive or not.
Args:
worker_name: a string representing the remote worker. It must be a fully
specified name like ""/job:worker/replica:0/task:0"".
Returns:
a boolean indicating whether the remote worker is alive or not.
Raises:
ValueError: if context is not initialized."
3293,sync_executors,tensorflow/tensorflow/python/eager/context.py,655,method,"Sync both local executors and the ones on remote workers.
In async execution mode, local function calls can return before the
coresponding remote op/function execution requests are completed. Calling
this method creates a synchronization barrier for remote executors. It only
returns when all remote pending nodes are finished, potentially with errors
if any remote executors are in error state.
Raises:
ValueError: if context is not initialized."
3294,clear_executor_errors,tensorflow/tensorflow/python/eager/context.py,672,method,"Clear errors in both local executors and remote workers.
After receiving errors from remote workers, additional requests on the fly
could further taint the status on the remote workers due to the async nature
of remote execution. Calling this method block on waiting for all pending
nodes in remote executors to finish and clear their error statuses.
Raises:
ValueError: if context is not initialized."
3295,enable_collective_ops,tensorflow/tensorflow/python/eager/context.py,688,method,"Enable distributed collective ops with an appropriate server_def.
Args:
server_def: A tensorflow::ServerDef proto. Enables execution on remote
devices.
Raises:
ValueError: if server_def is None.
RuntimeError: if this method is not called at program startup."
3296,configure_collective_ops,tensorflow/tensorflow/python/eager/context.py,716,method,"Configure collective ops.
Collective group leader is necessary for collective ops to run, other
configurations are mainly for the purpose of performance.
Args:
collective_leader: a device string for collective leader, e.g.
""/job:worker/replica:0/task:0""; empty string means local execution of
collective ops.
scoped_allocator_enabled_ops: a tuple or a list of op names for scoped
allocator to run with.
use_nccl_communication: whether to use nccl communication for collective
ops.
device_filters: a tuple or a list of device strings. If set, corresponding
task can only see the devices filtered by these device filters.
Raises:
RuntimeError: if this method is not called at program startup."
3297,abort_collective_ops,tensorflow/tensorflow/python/eager/context.py,759,method,"Abort the collective ops.
This is intended to be used when a peer failure is detected, which allows
the user to handle the case instead of hanging. This aborts all on-going
collectives. After all subsequent collectives error immediately. The only
way to recovery now is to restart the program.
Args:
code: a `tf.errors` error code.
message: a string. The error message."
3298,executing_eagerly,tensorflow/tensorflow/python/eager/context.py,816,method,Returns True if current thread has eager executing enabled.
3299,ones_rank_cache,tensorflow/tensorflow/python/eager/context.py,820,method,Per-device cache for scalars.
3300,zeros_cache,tensorflow/tensorflow/python/eager/context.py,824,method,Per-device cache for scalars.
3301,scope_name,tensorflow/tensorflow/python/eager/context.py,829,method,Returns scope name for the current thread.
3302,scope_name,tensorflow/tensorflow/python/eager/context.py,834,method,Sets scope name for the current thread.
3303,device_name,tensorflow/tensorflow/python/eager/context.py,839,method,Returns the device name for the current thread.
3304,device_spec,tensorflow/tensorflow/python/eager/context.py,844,method,Returns the device spec for the current thread.
3305,device,tensorflow/tensorflow/python/eager/context.py,852,method,"Context-manager to force placement of operations and Tensors on a device.
Args:
name: Name of the device or None to get default placement.
Returns:
Context manager that forces device placement.
Raises:
ValueError: If name is not a string or is an invalid device name.
RuntimeError: If device scopes are not properly nested."
3306,devices,tensorflow/tensorflow/python/eager/context.py,871,method,List of the names of devices available to execute operations.
3307,host_address_space,tensorflow/tensorflow/python/eager/context.py,875,method,
3308,execution_mode,tensorflow/tensorflow/python/eager/context.py,884,method,Gets execution mode for current thread.
3309,execution_mode,tensorflow/tensorflow/python/eager/context.py,889,method,Sets execution mode for current thread.
3310,is_async,tensorflow/tensorflow/python/eager/context.py,910,method,
3311,executor,tensorflow/tensorflow/python/eager/context.py,917,method,
3312,executor,tensorflow/tensorflow/python/eager/context.py,923,method,
3313,config,tensorflow/tensorflow/python/eager/context.py,928,method,Return the ConfigProto with all runtime deltas applied.
3314,function_call_options,tensorflow/tensorflow/python/eager/context.py,1082,method,"Returns function call options for current thread.
Note that the returned object is still referenced by the eager context.
Returns: the FunctionCallOptions for current thread."
3315,function_call_options,tensorflow/tensorflow/python/eager/context.py,1101,method,Returns function call options for current thread.
3316,num_gpus,tensorflow/tensorflow/python/eager/context.py,1105,method,The number of GPUs available to execute operations.
3317,add_function,tensorflow/tensorflow/python/eager/context.py,1110,method,"Add a function definition to the context.
Once added, the function (identified by its name) can be executed like any
other operation.
Args:
fn: A wrapped TF_Function (returned from TF_GraphToFunction_wrapper)."
3318,add_function_def,tensorflow/tensorflow/python/eager/context.py,1122,method,"Add a function definition to the context.
Once added, the function (identified by its name) can be executed like any
other operation.
Args:
fdef: A FunctionDef protocol buffer message."
3319,get_function_def,tensorflow/tensorflow/python/eager/context.py,1136,method,"Get a function definition from the context.
Args:
name: function signature name.
Returns:
The requested FunctionDef.
Raises:
tf.errors.NotFoundError: if name is not the name of a registered function."
3320,register_custom_device,tensorflow/tensorflow/python/eager/context.py,1156,method,Calls TFE_RegisterCustomDevice. See the non-member function.
3321,pack_eager_tensors,tensorflow/tensorflow/python/eager/context.py,1163,method,"Pack multiple `EagerTensor`s of the same dtype and shape.
Args:
tensors: a list of EagerTensors to pack.
Returns:
A packed EagerTensor."
3322,remove_function,tensorflow/tensorflow/python/eager/context.py,1179,method,"Remove a function from the context.
Once removed, the function cannot be executed anymore.
Args:
name: function signature name."
3323,has_function,tensorflow/tensorflow/python/eager/context.py,1190,method,Check if a function `name` is registered.
3324,add_op_callback,tensorflow/tensorflow/python/eager/context.py,1195,method,"Add a post-op callback to the context.
A post-op callback is invoked immediately after an eager operation or
function has finished execution or after a op has been added to a graph,
providing access to the op's type, name input and output tensors. Multiple
op callbacks can be added, in which case the callbacks will be invoked in
the order in which they are added.
Args:
callback: a callable of the signature
`f(op_type, inputs, attrs, outputs, op_name=None, graph=None)`.
See doc strings in `op_callbacks.py` for details on the function
signature and its semantics."
3325,remove_op_callback,tensorflow/tensorflow/python/eager/context.py,1213,method,"Remove an already-registered op callback.
Args:
callback: The op callback to be removed.
Raises:
KeyError: If `callback` is not already registered."
3326,op_callbacks,tensorflow/tensorflow/python/eager/context.py,1230,method,
3327,invoking_op_callbacks,tensorflow/tensorflow/python/eager/context.py,1234,method,
3328,invoking_op_callbacks,tensorflow/tensorflow/python/eager/context.py,1238,method,
3329,list_physical_devices,tensorflow/tensorflow/python/eager/context.py,1270,method,"List local devices visible to the system.
This API allows a client to query the devices before they have been
initialized by the eager runtime. Additionally a user can filter by device
type, to get only CPUs or GPUs.
Args:
device_type: Optional device type to limit results to
Returns:
List of PhysicalDevice objects."
3330,get_device_details,tensorflow/tensorflow/python/eager/context.py,1290,method,"Returns details about a physical devices.
Args:
device: A `tf.config.PhysicalDevice` returned by
`tf.config.list_physical_devices` or `tf.config.get_visible_devices`.
Returns:
A dict with string keys."
3331,list_logical_devices,tensorflow/tensorflow/python/eager/context.py,1367,method,Return logical devices.
3332,get_visible_devices,tensorflow/tensorflow/python/eager/context.py,1375,method,Get the list of visible devices.
3333,set_visible_devices,tensorflow/tensorflow/python/eager/context.py,1386,method,Set the list of visible devices.
3334,get_memory_growth,tensorflow/tensorflow/python/eager/context.py,1416,method,Get if memory growth is enabled for a PhysicalDevice.
3335,set_memory_growth,tensorflow/tensorflow/python/eager/context.py,1425,method,Set if memory growth should be enabled for a PhysicalDevice.
3336,get_logical_device_configuration,tensorflow/tensorflow/python/eager/context.py,1448,method,Get the virtual device configuration for a PhysicalDevice.
3337,set_logical_device_configuration,tensorflow/tensorflow/python/eager/context.py,1457,method,Set the virtual device configuration for a PhysicalDevice.
3338,enable_mlir_bridge,tensorflow/tensorflow/python/eager/context.py,1491,method,
3339,enable_mlir_graph_optimization,tensorflow/tensorflow/python/eager/context.py,1495,method,
3340,enable_mlir_bridge,tensorflow/tensorflow/python/eager/context.py,1499,method,
3341,enable_mlir_graph_optimization,tensorflow/tensorflow/python/eager/context.py,1504,method,
3342,optimizer_jit,tensorflow/tensorflow/python/eager/context.py,1509,method,
3343,optimizer_jit,tensorflow/tensorflow/python/eager/context.py,1515,method,
3344,get_optimizer_experimental_options,tensorflow/tensorflow/python/eager/context.py,1520,method,"Get experimental options for the optimizer.
Returns:
Dictionary of current option values"
3345,set_optimizer_experimental_options,tensorflow/tensorflow/python/eager/context.py,1558,method,"Set experimental options for the optimizer.
Args:
options: Dictionary of options to modify"
3346,intra_op_parallelism_threads,tensorflow/tensorflow/python/eager/context.py,1569,method,
3347,intra_op_parallelism_threads,tensorflow/tensorflow/python/eager/context.py,1573,method,
3348,inter_op_parallelism_threads,tensorflow/tensorflow/python/eager/context.py,1584,method,
3349,inter_op_parallelism_threads,tensorflow/tensorflow/python/eager/context.py,1588,method,
3350,soft_device_placement,tensorflow/tensorflow/python/eager/context.py,1599,method,
3351,soft_device_placement,tensorflow/tensorflow/python/eager/context.py,1603,method,
3352,log_device_placement,tensorflow/tensorflow/python/eager/context.py,1611,method,
3353,log_device_placement,tensorflow/tensorflow/python/eager/context.py,1615,method,
3354,device_policy,tensorflow/tensorflow/python/eager/context.py,1623,method,
3355,device_policy,tensorflow/tensorflow/python/eager/context.py,1631,method,
3356,mirroring_policy,tensorflow/tensorflow/python/eager/context.py,1644,method,
3357,mirroring_policy,tensorflow/tensorflow/python/eager/context.py,1652,method,
3358,lazy_remote_inputs_copy,tensorflow/tensorflow/python/eager/context.py,1665,method,
3359,lazy_remote_inputs_copy,tensorflow/tensorflow/python/eager/context.py,1669,method,Sets whether to copy remote inputs lazily for functions.
3360,use_tfrt,tensorflow/tensorflow/python/eager/context.py,1681,method,
3361,use_tfrt,tensorflow/tensorflow/python/eager/context.py,1685,method,Sets whether to use TFRT.
3362,enable_run_metadata,tensorflow/tensorflow/python/eager/context.py,1695,method,"Enables tracing of op execution via RunMetadata.
To retrieve the accumulated metadata call context.export_run_metadata()
and to stop tracing call context.disable_run_metadata()."
3363,disable_run_metadata,tensorflow/tensorflow/python/eager/context.py,1704,method,Disables tracing of op execution via RunMetadata.
3364,enable_graph_collection,tensorflow/tensorflow/python/eager/context.py,1710,method,"Enables graph collection of executed functions.
To retrieve the accumulated graphs call context.export_run_metadata()
and to stop collecting graphs call context.disable_graph_collection()."
3365,disable_graph_collection,tensorflow/tensorflow/python/eager/context.py,1719,method,Disables graph collection of executed functions.
3366,export_run_metadata,tensorflow/tensorflow/python/eager/context.py,1725,method,"Returns a RunMetadata proto with accumulated information.
The returned protocol buffer contains information since the most recent call
to either enable_run_metadata or export_run_metadata.
Returns:
A RunMetadata protocol buffer. Or None if not enabled."
3367,context_switches,tensorflow/tensorflow/python/eager/context.py,1744,method,Returns a stack of context switches.
3368,start_step,tensorflow/tensorflow/python/eager/context.py,1748,method,
3369,end_step,tensorflow/tensorflow/python/eager/context.py,1751,method,
3370,rewriter_toggle,tensorflow/tensorflow/python/eager/context.py,959,method,
3371,rewriter_bool,tensorflow/tensorflow/python/eager/context.py,969,method,
3372,rewriter_toggle,tensorflow/tensorflow/python/eager/context.py,1529,method,
3373,rewriter_bool,tensorflow/tensorflow/python/eager/context.py,1534,method,
3374,context,tensorflow/tensorflow/python/eager/context.py,1846,function,Returns a singleton context object.
3375,context_safe,tensorflow/tensorflow/python/eager/context.py,1853,function,Returns current context (or None if one hasn't been initialized).
3376,ensure_initialized,tensorflow/tensorflow/python/eager/context.py,1858,function,Initialize the context.
3377,set_global_seed,tensorflow/tensorflow/python/eager/context.py,1863,function,Sets the eager mode seed.
3378,global_seed,tensorflow/tensorflow/python/eager/context.py,1868,function,Returns the eager mode seed.
3379,internal_operation_seed,tensorflow/tensorflow/python/eager/context.py,1873,function,Returns the operation seed generated based on global seed.
3380,executing_eagerly,tensorflow/tensorflow/python/eager/context.py,1879,function,"Checks whether the current thread has eager execution enabled.
Eager execution is enabled by default and this API returns `True`
in most of cases. However, this API might return `False` in the following use
cases.
* Executing inside `tf.function`, unless under `tf.init_scope` or
`tf.config.run_functions_eagerly(True)` is previously called.
* Executing inside a transformation function for `tf.dataset`.
* `tf.compat.v1.disable_eager_execution()` is called.
General case:
>>> print(tf.executing_eagerly())
True
Inside `tf.function`:
>>> @tf.function
... def fn():
... with tf.init_scope():
... print(tf.executing_eagerly())
... print(tf.executing_eagerly())
>>> fn()
True
False
Inside `tf.function` after `tf.config.run_functions_eagerly(True)` is called:
>>> tf.config.run_functions_eagerly(True)
>>> @tf.function
... def fn():
... with tf.init_scope():
... print(tf.executing_eagerly())
... print(tf.executing_eagerly())
>>> fn()
True
True
>>> tf.config.run_functions_eagerly(False)
Inside a transformation function for `tf.dataset`:
>>> def data_fn(x):
... print(tf.executing_eagerly())
... return x
>>> dataset = tf.data.Dataset.range(100)
>>> dataset = dataset.map(data_fn)
False
Returns:
`True` if the current thread has eager execution enabled."
3381,executing_eagerly_v1,tensorflow/tensorflow/python/eager/context.py,1940,function,"Checks whether the current thread has eager execution enabled.
Eager execution is typically enabled via
`tf.compat.v1.enable_eager_execution`, but may also be enabled within the
context of a Python function via tf.contrib.eager.py_func.
When eager execution is enabled, returns `True` in most cases. However,
this API might return `False` in the following use cases.
* Executing inside `tf.function`, unless under `tf.init_scope` or
`tf.config.run_functions_eagerly(True)` is previously called.
* Executing inside a transformation function for `tf.dataset`.
* `tf.compat.v1.disable_eager_execution()` is called.
>>> tf.compat.v1.enable_eager_execution()
General case:
>>> print(tf.executing_eagerly())
True
Inside `tf.function`:
>>> @tf.function
... def fn():
... with tf.init_scope():
... print(tf.executing_eagerly())
... print(tf.executing_eagerly())
>>> fn()
True
False
Inside `tf.function`
after `tf.config.run_functions_eagerly(True)` is called:
>>> tf.config.run_functions_eagerly(True)
>>> @tf.function
... def fn():
... with tf.init_scope():
... print(tf.executing_eagerly())
... print(tf.executing_eagerly())
>>> fn()
True
True
>>> tf.config.run_functions_eagerly(False)
Inside a transformation function for `tf.dataset`:
>>> def data_fn(x):
... print(tf.executing_eagerly())
... return x
>>> dataset = tf.data.Dataset.range(100)
>>> dataset = dataset.map(data_fn)
False
Returns:
`True` if the current thread has eager execution enabled."
3382,in_eager_mode,tensorflow/tensorflow/python/eager/context.py,2002,function,Use executing_eagerly() instead. This function will be removed.
3383,shared_name,tensorflow/tensorflow/python/eager/context.py,2007,function,"Returns the anonymous shared name GUID if no shared name is specified.
In eager mode we need to use a unique shared name to avoid spurious sharing
issues. The runtime generates a unique name on our behalf when the reserved
GUID is used as a shared name.
Args:
name: Optional shared name
Returns:
Eager compatible shared name."
3384,graph_mode,tensorflow/tensorflow/python/eager/context.py,2028,function,Context-manager to disable eager execution for the current thread.
3385,eager_mode,tensorflow/tensorflow/python/eager/context.py,2033,function,Context-manager to enable eager execution for the current thread.
3386,scope_name,tensorflow/tensorflow/python/eager/context.py,2038,function,Name of the current scope.
3387,device,tensorflow/tensorflow/python/eager/context.py,2043,function,"Context-manager to force placement of operations and Tensors on a device.
Example:
```python
with tf.device('gpu:0'):
with tf.device('cpu:0'):
shape = tf.constant([], dtype=tf.int32)
x = tf.random.truncated_normal(shape, tf.float32)
```
will ensure that the `shape` Tensor is on CPU but the `truncated_normal`
operation runs on GPU 0.
Args:
name: Name of the device (see context().devices()), or None to
perform automatic placement.
Returns:
Context manager for setting the device."
3388,get_log_device_placement,tensorflow/tensorflow/python/eager/context.py,2068,function,"Get if device placements are logged.
Returns:
If device placements are logged."
3389,set_log_device_placement,tensorflow/tensorflow/python/eager/context.py,2078,function,"Set if device placements should be logged.
Args:
enabled: Whether to enabled device placement logging."
3390,device_policy,tensorflow/tensorflow/python/eager/context.py,2088,function,Context manager for setting device placement policy for current thread.
3391,mirroring_policy,tensorflow/tensorflow/python/eager/context.py,2100,function,Context manager for setting mirroring policy for current thread.
3392,set_execution_mode,tensorflow/tensorflow/python/eager/context.py,2111,function,Sets execution mode for the current thread.
3393,execution_mode,tensorflow/tensorflow/python/eager/context.py,2118,function,Context manager for setting execution mode for current thread.
3394,executor_scope,tensorflow/tensorflow/python/eager/context.py,2136,function,"Context manager for changing executor for current thread.
Args:
e: A Executor to execute eager ops under this scope. Setting it to None will
switch back to use the default executor for the context.
Yields:
Context manager for setting the executor for current thread."
3395,function_executor_type,tensorflow/tensorflow/python/eager/context.py,2157,function,"Context manager for setting the executor of eager defined functions.
Eager defined functions are functions decorated by tf.contrib.eager.defun.
Args:
executor_type: a string for the name of the executor to be used to execute
functions defined by tf.contrib.eager.defun.
Yields:
Context manager for setting the executor of eager defined functions."
3396,is_async,tensorflow/tensorflow/python/eager/context.py,2178,function,Returns true if current thread is in async mode.
3397,num_gpus,tensorflow/tensorflow/python/eager/context.py,2183,function,"Get the number of available GPU devices.
Returns:
The number of available GPU devices."
3398,enable_run_metadata,tensorflow/tensorflow/python/eager/context.py,2192,function,"Enables tracing of op execution via RunMetadata.
To retrieve the accumulated metadata call context.export_run_metadata()
and to stop tracing call context.disable_run_metadata()."
3399,disable_run_metadata,tensorflow/tensorflow/python/eager/context.py,2201,function,Disables tracing of op execution via RunMetadata.
3400,enable_graph_collection,tensorflow/tensorflow/python/eager/context.py,2206,function,"Enables graph collection of executed functions.
To retrieve the accumulated graphs call context.export_run_metadata()
and to stop collecting graphs call context.disable_graph_collection()."
3401,disable_graph_collection,tensorflow/tensorflow/python/eager/context.py,2215,function,Disables graph collection of executed functions.
3402,export_run_metadata,tensorflow/tensorflow/python/eager/context.py,2220,function,"Returns a RunMetadata proto with accumulated information.
The returned protocol buffer contains information since the most recent call
to either enable_run_metadata or export_run_metadata.
Returns:
A RunMetadata protocol buffer."
3403,collect_graphs,tensorflow/tensorflow/python/eager/context.py,2233,function,"Collects a flat list of pre- or post-optimization graphs.
The collected graphs include device placements, which can be useful for
testing.
Usage:
```
@def_function.function
def f(x):
return x + constant_op.constant(1.)
with context.collect_graphs() as graphs:
with ops.device(""CPU:0""):
f(constant_op.constant(1.))
graph, = graphs # `graph` contains a single GraphDef for inspection
```
Args:
optimized: whether to collect optimized graphs or non-optimized graphs
Yields:
A list of GraphDefs, populated when the context manager exits."
3404,get_server_def,tensorflow/tensorflow/python/eager/context.py,2273,function,
3405,set_server_def,tensorflow/tensorflow/python/eager/context.py,2277,function,
3406,update_server_def,tensorflow/tensorflow/python/eager/context.py,2281,function,
3407,check_alive,tensorflow/tensorflow/python/eager/context.py,2285,function,
3408,async_scope,tensorflow/tensorflow/python/eager/context.py,2291,function,"Context manager for grouping async operations.
Ops/function calls inside the scope can return before finishing the actual
execution. When exiting the async scope, a synchronization barrier will be
automatically added to ensure the completion of all async op and function
execution, potentially raising exceptions if async execution results in
an error state.
Users may write the following code to asynchronuously invoke `train_step_fn`
and log the `loss` metric for every `num_steps` steps in a training loop.
`train_step_fn` internally consumes data using `iterator.get_next()`, and may
throw OutOfRangeError when running out of data. In the case:
```
try:
with tf.experimental.async_scope():
for _ in range(num_steps):
# Step function updates the metric `loss` internally
train_step_fn()
except tf.errors.OutOfRangeError:
tf.experimental.async_clear_error()
logging.info('loss =', loss.numpy())
```
Yields:
Context manager for grouping async operations."
3409,async_wait,tensorflow/tensorflow/python/eager/context.py,2337,function,"Sync all async operations and raise any errors during execution.
In async execution mode, an op/function call can return before finishing the
actual execution. Calling this method creates a synchronization barrier for
all async op and function execution. It only returns when all pending nodes
are finished, potentially raising exceptions if async execution results in
an error state."
3410,async_clear_error,tensorflow/tensorflow/python/eager/context.py,2350,function,"Clear pending operations and error statuses in async execution.
In async execution mode, an error in op/function execution can lead to errors
in subsequent ops/functions that are scheduled but not yet executed. Calling
this method clears all pending operations and reset the async execution state.
Example:
```
while True:
try:
# Step function updates the metric `loss` internally
train_step_fn()
except tf.errors.OutOfRangeError:
tf.experimental.async_clear_error()
break
logging.info('loss =', loss.numpy())
```"
3411,add_function,tensorflow/tensorflow/python/eager/context.py,2373,function,Add a function definition to the context.
3412,remove_function,tensorflow/tensorflow/python/eager/context.py,2378,function,Remove a function from the context.
3413,get_function_def,tensorflow/tensorflow/python/eager/context.py,2383,function,
3414,register_custom_device,tensorflow/tensorflow/python/eager/context.py,2387,function,"Calls TFE_RegisterCustomDevice to register a custom device with Python.
Enables using C extensions specifying a custom device from Python. See the
experimental eager C API in tensorflow/c/eager/c_api_experimental.h for
details.
Note that custom devices are not currently supported inside `tf.function`s.
Args:
device_capsule: A PyCapsule with the name set to 'TFE_CustomDevice'
containing a pointer to a TFE_CustomDevice struct. The capsule retains
ownership of the memory.
device_name: A string indicating the name to register the custom device
under, e.g. '/job:localhost/replica:0/task:0/device:CUSTOM:0'. It may
subsequently be passed to `with tf.device(...):`.
device_info_capsule: A PyCapsule with the name set to
'TFE_CustomDevice_DeviceInfo' containing a pointer to a device-specific
struct with the initial state of the custom device (the void* device_info
argument to TFE_RegisterCustomDevice). This method takes ownership of the
memory and clears the capsule destructor."
3415,execute,tensorflow/tensorflow/python/eager/core_test.py,51,function,
3416,truncated_normal,tensorflow/tensorflow/python/eager/core_test.py,56,function,
3417,current_device,tensorflow/tensorflow/python/eager/core_test.py,65,function,
3418,configure_virtual_cpus,tensorflow/tensorflow/python/eager/core_test.py,69,function,
3419,UnliftedInitializerVariable,tensorflow/tensorflow/python/eager/def_function.py,130,class,"Variable which does not lift its initializer out of function context.
Instances of this variable, when created, build a graph which runs their
initializer inside a tf.cond(is_initialized) block.
This can only be created inside a defun called from (eventually) eager
mode. That is, non-function-building graphs are not supported."
3420,assign_fn,tensorflow/tensorflow/python/eager/def_function.py,287,method,
3421,not_assign_fn,tensorflow/tensorflow/python/eager/def_function.py,295,method,
3422,experimental_run_functions_eagerly,tensorflow/tensorflow/python/eager/def_function.py,318,function,"Enables / disables eager execution of `tf.function`s.
Calling `tf.config.experimental_run_functions_eagerly(True)` will make all
invocations of `tf.function` run eagerly instead of running as a traced graph
function.
This can be useful for debugging or profiling. For example, let's say you
implemented a simple iterative sqrt function, and you want to collect the
intermediate values and plot the convergence. Appending the values to a list
in `@tf.function` normally wouldn't work since it will just record the Tensors
being traced, not the values. Instead, you can do the following.
>>> ys = []
>>>
>>> @tf.function
... def sqrt(x):
... y = x / 2
... d = y
... for _ in range(10):
... d /= 2
... if y * y < x:
... y += d
... else:
... y -= d
... ys.append(y.numpy())
... return y
>>>
>>> tf.config.experimental_run_functions_eagerly(True)
>>> sqrt(tf.constant(2.))
<tf.Tensor: shape=(), dtype=float32, numpy=1.4150391>
>>> ys
[1.5, 1.25, 1.375, 1.4375, 1.40625, 1.421875, 1.4140625, 1.4179688, 1.4160156,
1.4150391]
>>> tf.config.experimental_run_functions_eagerly(False)
Calling `tf.config.experimental_run_functions_eagerly(False)` will undo this
behavior.
Note: This flag has no effect on functions passed into tf.data transformations
as arguments. tf.data functions are never executed eagerly and are always
executed as a compiled Tensorflow Graph.
Args:
run_eagerly: Boolean. Whether to run functions eagerly."
3423,run_functions_eagerly,tensorflow/tensorflow/python/eager/def_function.py,368,function,"Enables / disables eager execution of `tf.function`s.
Calling `tf.config.run_functions_eagerly(True)` will make all
invocations of `tf.function` run eagerly instead of running as a traced graph
function.
This can be useful for debugging or profiling. For example, let's say you
implemented a simple iterative sqrt function, and you want to collect the
intermediate values and plot the convergence. Appending the values to a list
in `@tf.function` normally wouldn't work since it will just record the Tensors
being traced, not the values. Instead, you can do the following.
>>> ys = []
>>>
>>> @tf.function
... def sqrt(x):
... y = x / 2
... d = y
... for _ in range(10):
... d /= 2
... if y * y < x:
... y += d
... else:
... y -= d
... ys.append(y.numpy())
... return y
>>>
>>> tf.config.run_functions_eagerly(True)
>>> sqrt(tf.constant(2.))
<tf.Tensor: shape=(), dtype=float32, numpy=1.4150391>
>>> ys
[1.5, 1.25, 1.375, 1.4375, 1.40625, 1.421875, 1.4140625, 1.4179688, 1.4160156,
1.4150391]
>>> tf.config.run_functions_eagerly(False)
Calling `tf.config.run_functions_eagerly(False)` will undo this
behavior.
Note: This flag has no effect on functions passed into tf.data transformations
as arguments. tf.data functions are never executed eagerly and are always
executed as a compiled Tensorflow Graph.
Args:
run_eagerly: Boolean. Whether to run functions eagerly."
3424,experimental_functions_run_eagerly,tensorflow/tensorflow/python/eager/def_function.py,422,function,Returns the value of the `experimental_run_functions_eagerly` setting.
3425,functions_run_eagerly,tensorflow/tensorflow/python/eager/def_function.py,428,function,Returns the value of the `run_functions_eagerly` setting.
3426,FunctionDeleter,tensorflow/tensorflow/python/eager/def_function.py,433,class,
3427,Function,tensorflow/tensorflow/python/eager/def_function.py,448,class,"Wrapper class for the graph functions defined for a Python function.
See the documentation for `tf.function` for more information on the semantics
of defined functions.
`Function` is thread-compatible."
3428,python_function,tensorflow/tensorflow/python/eager/def_function.py,923,method,The python function wrapped in this tf.function.
3429,input_signature,tensorflow/tensorflow/python/eager/def_function.py,928,method,
3430,function_spec,tensorflow/tensorflow/python/eager/def_function.py,932,method,
3431,pretty_printed_concrete_signatures,tensorflow/tensorflow/python/eager/def_function.py,935,method,
3432,get_initialization_function,tensorflow/tensorflow/python/eager/def_function.py,983,method,"Returns a `ConcreteFunction` which initializes this function's variables.
Requires that this function hasn't been accessed yet through either calling
it or calling get_concrete_function. Fails if we cannot build an initializer
function which does not depend on the concrete values of the inputs to this
function.
Note that running this function will overwrite any values currently assigned
to variables, for example restores from a checkpoint.
Args:
*args: arguments to the underlying python callable.
**kwargs: keyword arguments to the python callable.
Returns:
A `ConcreteFunction` object which initializes the variables of this
function.
Raises:
RuntimeError: if called after the variables have been initialized."
3433,get_concrete_function,tensorflow/tensorflow/python/eager/def_function.py,1107,method,"Returns a `ConcreteFunction` specialized to inputs and execution context.
If this `Function` was created with an `input_signature`, `args` and
`kwargs` may be omitted. With an input signature there is only one
concrete function associated with this `Function`.
If there is no fixed `input_signature` associated with this
`Function`, positional and keyword arguments to `get_concrete_function`
follow the same rules as input signature specification, with `tf.TensorSpec`
objects describing `tf.Tensor`s which will be passed to the concrete
function.
Each `tf.Tensor` argument to the concrete function must have a unique name,
either because it is the only one associated with a named argument of the
Python function or because an explicit `name=` was passed to its
`tf.TensorSpec` object. These names become the argument names for the
concrete function.
Arguments to the concrete function may always be specified as keyword
arguments, naming the Tensor input. Positional arguments may be used instead
when each preceding argument to the Python function is a Tensor.
```python
@tf.function
def f(x):
return x
f_concrete = f.get_concrete_function(tf.TensorSpec([], tf.float64))
f_concrete(tf.constant(1.))
f_concrete(x=tf.constant(1.))
```
Nested structures containing Tensors may be specified when retrieving
concrete functions. Structures with multiple Tensors are expanded into
multiple arguments of the concrete function. Since multiple concrete
function arguments are associated with one argument to the original
function, these Tensors must be named explicitly. Tensors in nested
structures may not be passed using positional arguments when calling the
concrete function.
```python
f_concrete2 = f.get_concrete_function(
(tf.TensorSpec(None, tf.float64, name=""first""),
tf.TensorSpec([], tf.float32, name=""second"")))
# Keyword arguments are required when identifying Tensors in nested
# structures.
f_concrete2(first=tf.constant([1.]), second=tf.constant(0.))
```
Functions with fixed input signatures have only one concrete function
associated with them, which can be retrieved without specifying any
arguments. As before Tensors must have unique names, either inferred from
the argument names in the original Python function or specified
explicitly.
```python
@tf.function(input_signature=(tf.TensorSpec(None, tf.float32)))
def f_sig(y):
return y
f_sig_concrete = f.get_concrete_function()
f_sig_concrete(tf.constant(1.))
f_sig_concrete(y=tf.constant(1.))
```
Args:
*args: inputs to specialize on.
**kwargs: inputs to specialize on.
Returns:
A TensorFlow function which takes exactly one `tf.Tensor` per argument.
Raises:
ValueError: if this object has not yet been called on concrete values."
3434,wrapped_fn,tensorflow/tensorflow/python/eager/def_function.py,597,method,Wraps `self._python_function` in a variable creator scope.
3435,variable_capturing_scope,tensorflow/tensorflow/python/eager/def_function.py,696,method,Creates UnliftedInitializerVariables and saves references to them.
3436,invalid_creator_scope,tensorflow/tensorflow/python/eager/def_function.py,714,method,Disables variable creation.
3437,fn_with_cond,tensorflow/tensorflow/python/eager/def_function.py,864,method,Conditionally runs initialization if it's needed.
3438,initialize_variables,tensorflow/tensorflow/python/eager/def_function.py,952,method,
3439,initialize_variables,tensorflow/tensorflow/python/eager/def_function.py,1017,method,
3440,function,tensorflow/tensorflow/python/eager/def_function.py,1216,function,"Compiles a function into a callable TensorFlow graph.
`tf.function` constructs a callable that executes a TensorFlow graph
(`tf.Graph`) created by trace-compiling the TensorFlow operations in `func`,
effectively executing `func` as a TensorFlow graph.
Example usage:
>>> @tf.function
... def f(x, y):
... return x ** 2 + y
>>> x = tf.constant([2, 3])
>>> y = tf.constant([3, -2])
>>> f(x, y)
<tf.Tensor: ... numpy=array([7, 7], ...)>
_Features_
`func` may use data-dependent control flow, including `if`, `for`, `while`
`break`, `continue` and `return` statements:
>>> @tf.function
... def f(x):
... if tf.reduce_sum(x) > 0:
... return x * x
... else:
... return -x // 2
>>> f(tf.constant(-2))
<tf.Tensor: ... numpy=1>
`func`'s closure may include `tf.Tensor` and `tf.Variable` objects:
>>> @tf.function
... def f():
... return x ** 2 + y
>>> x = tf.constant([-2, -3])
>>> y = tf.Variable([3, -2])
>>> f()
<tf.Tensor: ... numpy=array([7, 7], ...)>
`func` may also use ops with side effects, such as `tf.print`, `tf.Variable`
and others:
>>> v = tf.Variable(1)
>>> @tf.function
... def f(x):
... for i in tf.range(x):
... v.assign_add(i)
>>> f(3)
>>> v
<tf.Variable ... numpy=4>
Important: Any Python side-effects (appending to a list, printing with
`print`, etc) will only happen once, when `func` is traced. To have
side-effects executed into your `tf.function` they need to be written
as TF ops:
>>> l = []
>>> @tf.function
... def f(x):
... for i in x:
... l.append(i + 1) # Caution! Will only happen once when tracing
>>> f(tf.constant([1, 2, 3]))
>>> l
[<tf.Tensor ...>]
Instead, use TensorFlow collections like `tf.TensorArray`:
>>> @tf.function
... def f(x):
... ta = tf.TensorArray(dtype=tf.int32, size=0, dynamic_size=True)
... for i in range(len(x)):
... ta = ta.write(i, x[i] + 1)
... return ta.stack()
>>> f(tf.constant([1, 2, 3]))
<tf.Tensor: ..., numpy=array([2, 3, 4], ...)>
_`tf.function` is polymorphic_
Internally, `tf.function` can build more than one graph, to support arguments
with different data types or shapes, since TensorFlow can build more
efficient graphs that are specialized on shapes and dtypes. `tf.function`
also treats any pure Python value as opaque objects, and builds a separate
graph for each set of Python arguments that it encounters.
To obtain an individual graph, use the `get_concrete_function` method of
the callable created by `tf.function`. It can be called with the same
arguments as `func` and returns a special `tf.Graph` object:
>>> @tf.function
... def f(x):
... return x + 1
>>> isinstance(f.get_concrete_function(1).graph, tf.Graph)
True
Caution: Passing python scalars or lists as arguments to `tf.function` will
always build a new graph. To avoid this, pass numeric arguments as Tensors
whenever possible:
>>> @tf.function
... def f(x):
... return tf.abs(x)
>>> f1 = f.get_concrete_function(1)
>>> f2 = f.get_concrete_function(2) # Slow - builds new graph
>>> f1 is f2
False
>>> f1 = f.get_concrete_function(tf.constant(1))
>>> f2 = f.get_concrete_function(tf.constant(2)) # Fast - reuses f1
>>> f1 is f2
True
Python numerical arguments should only be used when they take few distinct
values, such as hyperparameters like the number of layers in a neural network.
_Input signatures_
For Tensor arguments, `tf.function` instantiates a separate graph for every
unique set of input shapes and datatypes. The example below creates two
separate graphs, each specialized to a different shape:
>>> @tf.function
... def f(x):
... return x + 1
>>> vector = tf.constant([1.0, 1.0])
>>> matrix = tf.constant([[3.0]])
>>> f.get_concrete_function(vector) is f.get_concrete_function(matrix)
False
An ""input signature"" can be optionally provided to `tf.function` to control
the graphs traced. The input signature specifies the shape and type of each
Tensor argument to the function using a `tf.TensorSpec` object. More general
shapes can be used. This is useful to avoid creating multiple graphs when
Tensors have dynamic shapes. It also restricts the shape and datatype of
Tensors that can be used:
>>> @tf.function(
... input_signature=[tf.TensorSpec(shape=None, dtype=tf.float32)])
... def f(x):
... return x + 1
>>> vector = tf.constant([1.0, 1.0])
>>> matrix = tf.constant([[3.0]])
>>> f.get_concrete_function(vector) is f.get_concrete_function(matrix)
True
_Variables may only be created once_
`tf.function` only allows creating new `tf.Variable` objects when it is called
for the first time:
>>> class MyModule(tf.Module):
... def __init__(self):
... self.v = None
...
... @tf.function
... def __call__(self, x):
... if self.v is None:
... self.v = tf.Variable(tf.ones_like(x))
... return self.v * x
In general, it is recommended to create stateful objects like `tf.Variable`
outside of `tf.function` and passing them as arguments.
_Using type annotations to improve performance_
'experimental_follow_type_hints` can be used along with type annotations to
improve performance by reducing the number of expensive graph retracings.
For example, an argument annotated with `tf.Tensor` is converted to Tensor
even when the input is a non-Tensor value.
>>> @tf.function(experimental_follow_type_hints=True)
... def f_with_hints(x: tf.Tensor):
... print('Tracing')
... return x
>>> @tf.function(experimental_follow_type_hints=False)
... def f_no_hints(x: tf.Tensor):
... print('Tracing')
... return x
>>> f_no_hints(1)
Tracing
<tf.Tensor: shape=(), dtype=int32, numpy=1>
>>> f_no_hints(2)
Tracing
<tf.Tensor: shape=(), dtype=int32, numpy=2>
>>> f_with_hints(1)
Tracing
<tf.Tensor: shape=(), dtype=int32, numpy=1>
>>> f_with_hints(2)
<tf.Tensor: shape=(), dtype=int32, numpy=2>
Args:
func: the function to be compiled. If `func` is None, `tf.function` returns
a decorator that can be invoked with a single argument - `func`. In other
words, `tf.function(input_signature=...)(func)` is equivalent to
`tf.function(func, input_signature=...)`. The former can be used as
decorator.
input_signature: A possibly nested sequence of `tf.TensorSpec` objects
specifying the shapes and dtypes of the Tensors that will be supplied to
this function. If `None`, a separate function is instantiated for each
inferred input signature. If input_signature is specified, every input to
`func` must be a `Tensor`, and `func` cannot accept `**kwargs`.
autograph: Whether autograph should be applied on `func` before tracing a
graph. Data-dependent control flow requires `autograph=True`. For more
information, see the [tf.function and AutoGraph guide](
https://www.tensorflow.org/guide/function).
experimental_implements: If provided, contains a name of a ""known"" function
this implements. For example ""mycompany.my_recurrent_cell"".
This is stored as an attribute in inference function,
which can then be detected when processing serialized function.
See [standardizing composite ops](https://github.com/tensorflow/community/blob/master/rfcs/20190610-standardizing-composite_ops.md) # pylint: disable=line-too-long
for details. For an example of utilizing this attribute see this
[example](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/compiler/mlir/lite/transforms/prepare_composite_functions_tf.cc)
The code above automatically detects and substitutes function that
implements ""embedded_matmul"" and allows TFLite to substitute its own
implementations. For instance, a tensorflow user can use this
attribute to mark that their function also implements
`embedded_matmul` (perhaps more efficiently!)
by specifying it using this parameter:
`@tf.function(experimental_implements=""embedded_matmul"")`
This can either be specified as just the string name of the function or
a NameAttrList corresponding to a list of key-value attributes associated
with the function name. The name of the function will be in the 'name'
field of the NameAttrList.
experimental_autograph_options: Optional tuple of
`tf.autograph.experimental.Feature` values.
experimental_relax_shapes: When True, `tf.function` may generate fewer,
graphs that are less specialized on input shapes.
experimental_compile: If True, the function is always compiled by
[XLA](https://www.tensorflow.org/xla). XLA may be more efficient in some
cases (e.g. TPU, XLA_GPU, dense tensor computations).
experimental_follow_type_hints: When True, the function may use type
annotations from `func` to optimize the tracing performance. For example,
arguments annotated with `tf.Tensor` will automatically be converted
to a Tensor.
Returns:
If `func` is not None, returns a callable that will execute the compiled
function (and return zero or more `tf.Tensor` objects).
If `func` is None, returns a decorator that, when invoked with a single
`func` argument, returns a callable equivalent to the case above.
Raises:
ValueError when attempting to use experimental_compile, but XLA support is
not enabled."
3441,undecorated_function,tensorflow/tensorflow/python/eager/def_function_test.py,52,function,
3442,quick_execute,tensorflow/tensorflow/python/eager/execute.py,33,function,"Execute a TensorFlow operation.
Args:
op_name: Name of the TensorFlow operation (see REGISTER_OP in C++ code) to
execute.
num_outputs: The number of outputs of the operation to fetch.
(Explicitly provided instead of being inferred for performance
reasons).
inputs: A list of inputs to the operation. Each entry should be a Tensor, or
a value which can be passed to the Tensor constructor to create one.
attrs: A tuple with alternating string attr names and attr values for this
operation.
ctx: The value of context.context().
name: Customized name for the operation.
Returns:
List of output Tensor objects. The list is empty if there are no outputs
Raises:
An exception on error."
3443,execute_with_cancellation,tensorflow/tensorflow/python/eager/execute.py,80,function,"Execute a TensorFlow operation.
Args:
op_name: Name of the TensorFlow operation (see REGISTER_OP in C++ code) to
execute.
num_outputs: The number of outputs of the operation to fetch. (Explicitly
provided instead of being inferred for performance reasons).
inputs: A list of inputs to the operation. Each entry should be a Tensor, or
a value which can be passed to the Tensor constructor to create one.
attrs: A tuple with alternating string attr names and attr values for this
operation.
ctx: The value of context.context().
cancellation_manager: a `CancellationManager` object that can be used to
cancel the operation.
name: Customized name for the operation.
Returns:
List of output Tensor objects. The list is empty if there are no outputs
Raises:
An exception on error."
3444,execute_with_callbacks,tensorflow/tensorflow/python/eager/execute.py,136,function,Monkey-patch to execute to enable execution callbacks.
3445,must_record_gradient,tensorflow/tensorflow/python/eager/execute.py,148,function,Import backprop if you want gradients recorded.
3446,record_gradient,tensorflow/tensorflow/python/eager/execute.py,153,function,Import backprop if you want gradients recorded.
3447,make_float,tensorflow/tensorflow/python/eager/execute.py,159,function,
3448,make_int,tensorflow/tensorflow/python/eager/execute.py,166,function,
3449,make_str,tensorflow/tensorflow/python/eager/execute.py,177,function,
3450,make_bool,tensorflow/tensorflow/python/eager/execute.py,184,function,
3451,make_type,tensorflow/tensorflow/python/eager/execute.py,191,function,
3452,make_shape,tensorflow/tensorflow/python/eager/execute.py,201,function,Convert v into a list.
3453,make_tensor,tensorflow/tensorflow/python/eager/execute.py,223,function,Ensure v is a TensorProto.
3454,args_to_matching_eager,tensorflow/tensorflow/python/eager/execute.py,236,function,Convert sequence `l` to eager same-type Tensors.
3455,convert_to_mixed_eager_tensors,tensorflow/tensorflow/python/eager/execute.py,294,function,
3456,args_to_mixed_eager_tensors,tensorflow/tensorflow/python/eager/execute.py,300,function,Converts a list of same-length lists of values to eager tensors.
3457,Executor,tensorflow/tensorflow/python/eager/executor.py,24,class,"A class for handling eager execution.
The default behavior for asynchronous execution is to serialize all ops on
a single thread. Having different `Executor` objects in different threads
enables executing ops asynchronously in parallel:
```python
def thread_function():
executor = executor.Executor(enable_async=True):
context.set_executor(executor)
a = threading.Thread(target=thread_function)
a.start()
b = threading.Thread(target=thread_function)
b.start()
```"
3458,is_async,tensorflow/tensorflow/python/eager/executor.py,61,method,
3459,handle,tensorflow/tensorflow/python/eager/executor.py,64,method,
3460,wait,tensorflow/tensorflow/python/eager/executor.py,67,method,Waits for ops dispatched in this executor to finish.
3461,clear_error,tensorflow/tensorflow/python/eager/executor.py,71,method,Clears errors raised in this executor during execution.
3462,new_executor,tensorflow/tensorflow/python/eager/executor.py,76,function,
3463,ForwardAccumulator,tensorflow/tensorflow/python/eager/forwardprop.py,221,class,"Computes Jacobian-vector products (""JVP""s) using forward-mode autodiff.
Compare to `tf.GradientTape` which computes vector-Jacobian products (""VJP""s)
using reverse-mode autodiff (backprop). Reverse mode is more attractive when
computing gradients of a scalar-valued function with respect to many inputs
(e.g. a neural network with many parameters and a scalar loss). Forward mode
works best on functions with many outputs and few inputs. Since it does not
hold on to intermediate activations, it is much more memory efficient than
backprop where it is applicable.
Consider a simple linear regression:
>>> x = tf.constant([[2.0, 3.0], [1.0, 4.0]])
>>> dense = tf.keras.layers.Dense(1)
>>> dense.build([None, 2])
>>> with tf.autodiff.ForwardAccumulator(
... primals=dense.kernel,
... tangents=tf.constant([[1.], [0.]])) as acc:
... loss = tf.reduce_sum((dense(x) - tf.constant([1., -1.])) ** 2.)
>>> acc.jvp(loss)
<tf.Tensor: shape=(), dtype=float32, numpy=...>
The example has two variables containing parameters, `dense.kernel` (2
parameters) and `dense.bias` (1 parameter). Considering the training data `x`
as a constant, this means the Jacobian matrix for the function mapping from
parameters to loss has one row and three columns.
With forwardprop, we specify a length-three vector in advance which multiplies
the Jacobian. The `primals` constructor argument is the parameter (a
`tf.Tensor` or `tf.Variable`) we're specifying a vector for, and the
`tangents` argument is the ""vector"" in Jacobian-vector product. If our goal is
to compute the entire Jacobian matrix, forwardprop computes one column at a
time while backprop computes one row at a time. Since the Jacobian in the
linear regression example has only one row, backprop requires fewer
invocations:
>>> x = tf.constant([[2.0, 3.0], [1.0, 4.0]])
>>> dense = tf.keras.layers.Dense(1)
>>> dense.build([None, 2])
>>> loss_fn = lambda: tf.reduce_sum((dense(x) - tf.constant([1., -1.])) ** 2.)
>>> kernel_fprop = []
>>> with tf.autodiff.ForwardAccumulator(
... dense.kernel, tf.constant([[1.], [0.]])) as acc:
... kernel_fprop.append(acc.jvp(loss_fn()))
>>> with tf.autodiff.ForwardAccumulator(
... dense.kernel, tf.constant([[0.], [1.]])) as acc:
... kernel_fprop.append(acc.jvp(loss_fn()))
>>> with tf.autodiff.ForwardAccumulator(dense.bias, tf.constant([1.])) as acc:
... bias_fprop = acc.jvp(loss_fn())
>>> with tf.GradientTape() as tape:
... loss = loss_fn()
>>> kernel_grad, bias_grad = tape.gradient(loss, (dense.kernel, dense.bias))
>>> np.testing.assert_allclose(
... kernel_grad, tf.stack(kernel_fprop)[:, tf.newaxis])
>>> np.testing.assert_allclose(bias_grad, bias_fprop[tf.newaxis])
Implicit in the `tape.gradient` call is a length-one vector which
left-multiplies the Jacobian, a vector-Jacobian product.
`ForwardAccumulator` maintains JVPs corresponding primal tensors it is
watching, derived from the original `primals` specified in the constructor. As
soon as a primal tensor is deleted, `ForwardAccumulator` deletes the
corresponding JVP.
`acc.jvp(x)` retrieves `acc`'s JVP corresponding to the primal tensor `x`. It
does not perform any computation. `acc.jvp` calls can be repeated as long as
`acc` is accessible, whether the context manager is active or not. New JVPs
are only computed while the context manager is active.
Note that `ForwardAccumulator`s are always applied in the order their context
managers were entered, so inner accumulators will not see JVP computation from
outer accumulators. Take higher-order JVPs from outer accumulators:
>>> primal = tf.constant(1.1)
>>> with tf.autodiff.ForwardAccumulator(primal, tf.constant(1.)) as outer:
... with tf.autodiff.ForwardAccumulator(primal, tf.constant(1.)) as inner:
... primal_out = primal ** tf.constant(3.5)
>>> inner_jvp = inner.jvp(primal_out)
>>> inner_jvp # 3.5 * 1.1 ** 2.5
<tf.Tensor: shape=(), dtype=float32, numpy=4.4417057>
>>> outer.jvp(inner_jvp) # 3.5 * 2.5 * 1.1 ** 1.5
<tf.Tensor: shape=(), dtype=float32, numpy=10.094786>
Reversing the collection in the last line to instead retrieve
`inner.jvp(outer.jvp(primal_out))` will not work.
Strict nesting also applies to combinations of `ForwardAccumulator` and
`tf.GradientTape`. More deeply nested `GradientTape` objects will ignore the
products of outer `ForwardAccumulator` objects. This allows (for example)
memory-efficient forward-over-backward computation of Hessian-vector products,
where the inner `GradientTape` would otherwise hold on to all intermediate
JVPs:
>>> v = tf.Variable([1., 2.])
>>> with tf.autodiff.ForwardAccumulator(
... v,
... # The ""vector"" in Hessian-vector product.
... tf.constant([1., 0.])) as acc:
... with tf.GradientTape() as tape:
... y = tf.reduce_sum(v ** 3.)
... backward = tape.gradient(y, v)
>>> backward # gradient from backprop
<tf.Tensor: shape=(2,), dtype=float32, numpy=array([ 3., 12.], dtype=float32)>
>>> acc.jvp(backward) # forward-over-backward Hessian-vector product
<tf.Tensor: shape=(2,), dtype=float32, numpy=array([6., 0.], dtype=float32)>"
3464,jvp,tensorflow/tensorflow/python/eager/forwardprop.py,411,method,"Fetches the Jacobian-vector product computed for `primals`.
Note that this method performs no computation, and simply looks up a JVP
that was already computed (unlike backprop using a `tf.GradientTape`, where
the computation happens on the call to `tape.gradient`).
Args:
primals: A watched Tensor or structure of Tensors to fetch the JVPs for.
unconnected_gradients: A value which can either hold 'none' or 'zero' and
alters the value which will be returned if no JVP was computed for
`primals`. The possible values and effects are detailed in
'tf.UnconnectedGradients' and it defaults to 'none'.
Returns:
Tensors with the same shapes and dtypes as `primals`, or None if no JVP
is available."
3465,TangentInfo,tensorflow/tensorflow/python/eager/forwardprop_util.py,30,class,Packed forward accumulator state. The return value of `pack_tangents`.
3466,pack_tangents,tensorflow/tensorflow/python/eager/forwardprop_util.py,42,function,"Packs forward accumulator state into a TangentInfo tuple.
Args:
tensors: A flat list of Tensors to pack forward accumulator state for.
Returns:
A tuple of (indices, tangents):
indices: A sequence of sequences of two-element tuples. Each forward
accumulator is represented as a sequence of tuples with (primal_index,
jvp_index). Both integers index into the concatenated `tensors + jvps`
array.
tangents: A flat list of Tensors. Best interpreted as a sequence to be
appended to `tensors`."
3467,push_forwardprop_state,tensorflow/tensorflow/python/eager/forwardprop_util.py,61,function,"Temporarily push or pop transient state for accumulators in the active set.
Allows an accumulator which is currently processing an operation to
temporarily reset its state. This is useful when building forwardprop versions
of functions, where an accumulator will trigger function building and then
must process captured symbolic tensors while building it. Without pushing and
popping, accumulators ignore operations executed as a direct result of their
own jvp computations.
Yields:
None (used for its side effect)."
3468,common_shape,tensorflow/tensorflow/python/eager/function.py,189,function,Find a `TensorShape` that is compatible with both `x` and `y`.
3469,is_same_structure,tensorflow/tensorflow/python/eager/function.py,214,function,"Check two structures for equality, optionally of types and of values."
3470,add_function_callback,tensorflow/tensorflow/python/eager/function.py,309,function,"Add a callback function for Function creation.
The callback function has the signature:
`def function_callback(function):`
wherein `function` is the just-created _EagerDefinedFunction.
The callback is invoked immediately after a new `_EagerDefinedFunction`
is created. The return value(s) of the callback function (if any) is ignored.
Repeated registration of the same callback function is idempotent.
After a callback is added, it can be removed with the
`remove_function_callback()` method.
Args:
function_callback: The callback to add."
3471,remove_function_callback,tensorflow/tensorflow/python/eager/function.py,330,function,"Remove an already-added function callback.
See the doc string of `add_function_callback()` for more information.
Args:
function_callback: The callback to remove."
3472,clear_function_callbacks,tensorflow/tensorflow/python/eager/function.py,341,function,"Clear all function callbacks, if any have been regisered."
3473,ConcreteFunction,tensorflow/tensorflow/python/eager/function.py,1464,class,"Callable object encapsulating a function definition and its gradient.
`ConcreteFunction` is a callable that encapsulates a function definition and
is differentiable under `tf.GradientTape` objects."
3474,variables,tensorflow/tensorflow/python/eager/function.py,1606,method,Sequence of variables for this function.
3475,trainable_variables,tensorflow/tensorflow/python/eager/function.py,1611,method,Sequence of trainable variables for this function.
3476,name,tensorflow/tensorflow/python/eager/function.py,1970,method,`ConcreteFunction` name.
3477,graph,tensorflow/tensorflow/python/eager/function.py,1975,method,Returns the graph from which this function was constructed.
3478,inputs,tensorflow/tensorflow/python/eager/function.py,1980,method,Returns tensors in `self.graph` corresponding to arguments.
3479,structured_input_signature,tensorflow/tensorflow/python/eager/function.py,1985,method,"Returns structured signature for this concrete function.
Returns:
A tuple `(args, kwargs)`, where:
* `args` is a tuple that specifies the expected type or value each for
positional argument.
* `kwargs` is a dictionary that specifies the expected type or value
for each keyword-only argument.
The type or value for each argument is specified using one of the
following:
* A `tf.TypeSpec`, indicating that a Tensor or other TensorFlow-native
value is expected.
* A Python value, such as an integer, indicating that an equal value
is expected.
* A nested structure of `tf.TypeSpec`s and Python values, indicating
that a corresponding nested structure is expected."
3480,outputs,tensorflow/tensorflow/python/eager/function.py,2009,method,Returns tensors in `self.graph` corresponding to returned tensors.
3481,structured_outputs,tensorflow/tensorflow/python/eager/function.py,2014,method,Returns outputs in `self.graph` as returned by the original function.
3482,captured_inputs,tensorflow/tensorflow/python/eager/function.py,2019,method,"Returns external Tensors captured by this function.
self.__call__(*args) passes `args + self.captured_inputs` to the function."
3483,function_def,tensorflow/tensorflow/python/eager/function.py,2029,method,Returns a `FunctionDef` object representing this function.
3484,output_shapes,tensorflow/tensorflow/python/eager/function.py,2034,method,The function's output shapes.
3485,output_dtypes,tensorflow/tensorflow/python/eager/function.py,2043,method,
3486,add_to_graph,tensorflow/tensorflow/python/eager/function.py,2051,method,"Registers the function, adds it to the graph g or default graph.
Args:
g: If specified, registers the function with this graph. Defaults to the
current context (either the default graph or the eager context)."
3487,add_gradient_functions_to_graph,tensorflow/tensorflow/python/eager/function.py,2067,method,Add forward/backward functions to graph `g` or the current context.
3488,pretty_printed_signature,tensorflow/tensorflow/python/eager/function.py,2246,method,Returns a string summarizing the signature of this concrete function.
3489,cancellable_call,tensorflow/tensorflow/python/eager/function.py,1963,method,
3490,pretty_print_spec,tensorflow/tensorflow/python/eager/function.py,2251,method,Returns a string describing the spec for a single argument.
3491,FunctionSpec,tensorflow/tensorflow/python/eager/function.py,2323,class,Specification of how to bind arguments to a function.
3492,from_function_and_signature,tensorflow/tensorflow/python/eager/function.py,2327,method,"Create a FunctionSpec instance given a python function and signature.
Args:
python_function: a function to inspect
input_signature: a signature of the function (None, if variable)
is_pure: if True all input arguments (including variables and constants)
will be converted to tensors and no variable changes allowed.
experimental_follow_type_hints: see `tf.function`
Returns:
instance of FunctionSpec"
3493,fullargspec,tensorflow/tensorflow/python/eager/function.py,2486,method,
3494,is_method,tensorflow/tensorflow/python/eager/function.py,2490,method,
3495,args_to_indices,tensorflow/tensorflow/python/eager/function.py,2494,method,
3496,kwargs_to_include,tensorflow/tensorflow/python/eager/function.py,2498,method,
3497,input_signature,tensorflow/tensorflow/python/eager/function.py,2502,method,
3498,flat_input_signature,tensorflow/tensorflow/python/eager/function.py,2506,method,
3499,is_pure,tensorflow/tensorflow/python/eager/function.py,2510,method,
3500,arg_names,tensorflow/tensorflow/python/eager/function.py,2514,method,
3501,vararg_name,tensorflow/tensorflow/python/eager/function.py,2518,method,
3502,varkw_name,tensorflow/tensorflow/python/eager/function.py,2522,method,
3503,signature_summary,tensorflow/tensorflow/python/eager/function.py,2525,method,"Returns a string summarizing this function's signature.
Args:
default_values: If true, then include default values in the signature.
Returns:
A `string`."
3504,canonicalize_function_inputs,tensorflow/tensorflow/python/eager/function.py,2589,method,"Canonicalizes `args` and `kwargs`.
Canonicalize the inputs to the Python function using a `FunctionSpec`
instance. In particular, we parse the varags and kwargs that the
original function was called with into a tuple corresponding to the
Python function's positional (named) arguments and a dictionary
corresponding to its kwargs. Missing default arguments are added.
If this `FunctionSpec` has an input signature, then it is used to convert
arguments to tensors; otherwise, any inputs containing numpy arrays are
converted to tensors.
Additionally, any inputs containing numpy arrays are converted to Tensors.
Args:
*args: The varargs this object was called with.
**kwargs: The keyword args this function was called with.
Returns:
A canonicalized ordering of the inputs representened by a tuple in the
form (args, kwargs). Here: `args` is a full list of bound arguments, and
`kwargs` contains only true keyword arguments, as opposed to named
arguments called in a keyword-like fashion.
Raises:
ValueError: If a keyword in `kwargs` cannot be matched with a positional
argument when an input signature is specified, or when the inputs
do not conform to the input signature."
3505,FunctionCache,tensorflow/tensorflow/python/eager/function.py,2802,class,"A lightweight container for cached functions.
"
3506,all_values,tensorflow/tensorflow/python/eager/function.py,2831,method,A set of all `ConcreteFunction` instances held by this cache.
3507,Function,tensorflow/tensorflow/python/eager/function.py,2836,class,"Wrapper class for the graph functions defined for a Python function.
See the documentation for `defun` for more information on the semantics of
defined functions.
`Function` class is thread-compatible meaning that minimal usage of defuns
(defining and calling) is thread-safe, but if users call other methods or
invoke the base `python_function` themselves, external synchronization is
necessary.
In addition, Function is not reentrant, so recursive functions need to call
the wrapped function, not the wrapper."
3508,python_function,tensorflow/tensorflow/python/eager/function.py,2925,method,Returns the wrapped Python function.
3509,function_spec,tensorflow/tensorflow/python/eager/function.py,2930,method,
3510,input_signature,tensorflow/tensorflow/python/eager/function.py,2934,method,Returns the input signature.
3511,flat_input_signature,tensorflow/tensorflow/python/eager/function.py,2939,method,Returns the flattened input signature.
3512,get_concrete_function,tensorflow/tensorflow/python/eager/function.py,3023,method,"Returns a `ConcreteFunction` specialized to inputs and execution context.
Args:
*args: inputs to specialize on.
**kwargs: inputs to specialize on."
3513,register,tensorflow/tensorflow/python/eager/function.py,3328,function,"Register a specialization of a `Function` into the graph.
This won't actually call the function with the inputs, and only put the
function definition into graph. Register function with different input param
will result into multiple version of functions registered in graph.
Args:
func: the `Function` instance that generated by a @defun
*args: input arguments for the Python function.
**kwargs: input keyword arguments for the Python function.
Returns:
a `ConcreteFunction` object specialized to inputs and execution context.
Raises:
ValueError: When the input function is not a defun wrapped python function."
3514,validate_signature,tensorflow/tensorflow/python/eager/function.py,3355,function,
3515,defun,tensorflow/tensorflow/python/eager/function.py,3363,function,"Compiles a Python function into a callable TensorFlow graph.
`defun` (short for ""define function"") compiles a Python function
composed of TensorFlow operations into a callable that executes a `tf.Graph`
containing those operations. The callable produced by `defun` contains only
the subgraph of TensorFlow operations that were executed when the Python
function was called with a particular input signature, defined as a list
of the shapes and dtypes of the Python function's Tensor-valued arguments and
the values of its non-Tensor Python objects.
When eager execution is enabled, the ability to create graphs from Python
functions makes it possible to incrementally trade off debuggability and
interactivity for performance. Functions compiled with `defun` cannot be
inspected with `pdb`; however, executing a graph
generated by `defun` sometimes takes less time and memory than eagerly
executing the corresponding Python function, since specifying computations as
graphs allows for optimizations like automatic buffer reuse and
parallelization among ops. Note that executing a `defun`-compiled function
incurs a small constant overhead, so eagerly executing sufficiently small
Python functions might take less time than executing their corresponding
`defun`-generated graphs.
For a Python function to be compatible with `defun`, all of its arguments must
be hashable Python objects or lists thereof. The function itself may not
modify the list/map structure of its arguments. Additionally, it must return
zero or more `tf.Tensor` objects. If the Python function returns
a `tf.Variable`, its compiled version will return the value of that variable
as a `tf.Tensor`.
Executing a graph generated by `defun` respects device annotations (i.e.,
all `with tf.device` directives present in a Python function will also be
present in its corresponding graph), but it is not yet possible to execute the
generated graphs across multiple machines.
_Example Usage_
```python
import tensorflow as tf
tf.compat.v1.enable_eager_execution()
# A simple example.
def f(x, y):
return tf.reduce_mean(tf.multiply(x ** 2, 3) + y)
g = tf.contrib.eager.defun(f)
x = tf.constant([[2.0, 3.0]])
y = tf.constant([[3.0, -2.0]])
# `f` and `g` will return the same value, but `g` will be executed as a
# TensorFlow graph.
assert f(x, y).numpy() == g(x, y).numpy()
# `defun` is capable of compiling Python functions that close over Python
# objects, including Tensors and Variables.
@tf.contrib.eager.defun
def h():
return f(x, y)
assert (h().numpy() == f(x, y).numpy()).all()
# `defun` automatically lifts variables out of the graphs it creates,
# allowing you to compile the `call` methods of `tf.keras.layers.Layer` and
# `tf.keras.Model` objects.
class MyModel(tf.keras.Model):
def __init__(self, keep_probability=0.2):
super(MyModel, self).__init__()
self.dense1 = tf.keras.layers.Dense(4, activation=tf.nn.relu)
self.dense2 = tf.keras.layers.Dense(5, activation=tf.nn.softmax)
self.keep_probability = keep_probability
@tf.contrib.eager.defun
def call(self, inputs, training=True):
x = self.dense2(self.dense1(inputs))
if training:
return tf.nn.dropout(x, self.keep_probability)
else:
return x
model = MyModel()
model(x, training=True) # executes a graph, with dropout
model(x, training=False) # executes a graph, without dropout
# `defun`-compiled functions are differentiable.
optimizer = tf.compat.v1.train.GradientDescentOptimizer(learning_rate=0.01)
with tf.GradientTape() as tape:
outputs = model(x)
gradient = tape.gradient(outputs, model.trainable_variables)
optimizer.apply_gradients((grad, var) for grad, var in zip(gradient,
model.trainable_variables))
```
When using `defun`, there are subtleties regarding inputs, Python control
flow, and variable creation that one should be aware of. For concreteness, let
`f` be a Python function that returns zero or more `tf.Tensor` objects and
let `F = defun(f)`. `F` builds a graph for each unique input signature it
sees, Python control flow is baked into graphs, and operations related to
variable initialization are automatically lifted out of the graphs that `F`
generates and placed in the eager context if executing eagerly or into an
outer graph otherwise.
_Input Signatures_
By default, `F = tf.contrib.eager.defun(f)` instantiates a separate graph
for every unique sequence of the shapes and dtypes of Tensor arguments and
the values of Python objects it is invoked with. For example, calling
`F(tf.random.uniform([2])` will execute a different graph than
`F(tf.random.uniform([3])` because the two inputs have different shapes.
The first time that `F(*args, **kwargs)` is called with a particular sequence
of Tensor shapes and dtypes and Python values, it constructs a graph by
tracing the execution of `f(*args, **kwargs)`; this graph is bound to an
input signature inferred from `(*args, **kwargs)` and cached for future reuse.
NumPy arrays passed as inputs to `F` are converted to `tf.Tensor` objects
before being passed to `f`, and are treated as Tensors for caching. This
allows a function to be called multiple times with NumPy arrays having
different values but the same shape and dtype without re-tracing each time.
`tf.contrib.eager.defun` caches graphs for your convenience, letting you
define TensorFlow functions without explicitly specifying their signatures.
However, this policy is conservative and potentially expensive; for example,
when different invocations of your function have differently-shaped Tensor
inputs, this policy might generate more graph functions than necessary. To
eliminate such costs, `tf.contrib.eager.defun` allows you to supply an
optional `input_signature` argument specifying the shapes and dtypes of the
inputs. In particular, the shapes may be partially unspecified, with `None`s
in the unknown dimensions. When an input signature is provided,
`tf.contrib.eager.defun` will only instantiate a single graph for the
decorated Python function. The following is an example:
```python
import tensorflow as tf
# The first `TensorSpec` below describes the shape and dtype of `words`,
# and the second describes the shape and dtype of `another_tensor`. Note that
# the last dimension of the `words` `TensorSpec` is left unspecified.
@tf.contrib.eager.defun(input_signature=[
tf.contrib.eager.TensorSpec(shape=[50, 300, None], dtype=tf.float32),
tf.contrib.eager.TensorSpec(shape=[300, 100], dtype=tf.float32)
])
def my_sequence_model(words, another_tensor):
...
# Note how the third dimension of the first input can vary freely.
words = tf.random.uniform(([50, 300, 10])
second_input = tf.random.uniform([300, 100])
my_sequence_model(words, second_input)
words = tf.random.uniform(([50, 300, 20])
my_sequence_model(words, second_input)
# Passing an input with an incompatible shape will raise an error.
words = tf.random.uniform(([50, 100, 20])
my_sequence_model(words, second_input) # <---- This will raise an error.
```
Python functions that are compiled with an `input_signature` must only accept
Tensors as arguments and must not take unnamed keyword arguments (**kwargs).
_Tracing_
Be aware that because `F` only logs TensorFlow operations, all the other
Python code that `f` executes will only shape the _construction_ of the graphs
that `F` executes: the Python code won't be executed when the graphs
themselves are executed, though it will be executed every time the Python
function is traced (and a given Python function might be traced multiple
times, once for each input signature it is invoked with). For example, whereas
the Python function
```python
import tensorflow as tf
import numpy as np
tf.compat.v1.enable_eager_execution()
def add_noise():
return tf.eye(5) + np.random.randn(5, 5)
```
will return a different output everytime it is invoked, the compiled function
`compiled = tf.contrib.eager.defun(add_noise)` will return the same value
every time it is called, since a particular random offset generated by NumPy
will be inserted into the graph as a TensorFlow constant. The solution is to
replace the call to `np.random.randn` with `tf.random.normal((5, 5))`.
_Python Side-Effects_
A corollary of the previous discussion on tracing is the following: If a
Python function `f` has Python side-effects, then executing `f` multiple times
will not necessarily be semantically equivalent to executing `F =
tf.contrib.eager.defun(f)` multiple times; this difference is due to the fact
that `defun` only captures the subgraph of TensorFlow operations that is
constructed when `f` is called in a graph-building context.
_Python Control Flow_
The structure of many machine learning computations depend upon whether one is
training or validating, and it is common to nest specialized logic under `if
training:` blocks. By mapping each input signature to a unique graph, `defun`
lets users transparently compile such code, as the following code snippet
demonstrates:
```python
import tensorflow as tf
tf.compat.v1.enable_eager_execution()
@tf.contrib.eager.defun
def lossy_matmul(W, x, training=True):
outputs = tf.matmul(W, x)
if training:
outputs = tf.nn.dropout(outputs, keep_probability=0.2)
return outputs
W = tf.random.normal((3, 5))
x = tf.random.normal((5, 1))
# Executes a graph that applies dropout.
lossy_outputs = lossy_matmul(W, x, training=True)
# Executes a graph that does not apply dropout.
exact_outputs = lossy_matmul(W, x, training=False)
```
_TensorFlow Control Flow_
When `autograph` is `True`, data-dependent control flow is allowed as well.
Control flow statements that depend on `Tensor` values are staged into
corresponding TensorFlow ops. For example, the following code will work as
expected:
```python
@tf.contrib.eager.defun
def dynamic_rnn_loop(cell, seq):
state, output = cell.zero_state()
for input in seq:
state, output = cell(input, state)
return output
```
For more information see `tf.autograph`.
_Variables_
TensorFlow operations related to variable creation and initialization are
automatically lifted out of the graphs generated by `defun`. In practice, this
implies that variable creation and initialization only happen the first time
`F` is called, and that variables are reused every time thereafter. Many
TensorFlow APIs, like `tf.keras.layers.Layer` objects, create variables the
first time they are called and reuse them thereafter. Automatic variable
lifting makes it possible to compile these APIs without extra effort, at the
cost of introducing a discrepancy between the semantics of executing Python
functions and their corresponding compiled functions. For example:
```python
import tensorflow as tf
tf.compat.v1.enable_eager_execution()
def fn():
x = tf.Variable(0.0)
x.assign_add(1.0)
return x.read_value()
# `fn` is a Python function, so x is created, initialized, and destroyed upon
# every invocation
assert fn().numpy() == fn().numpy() == 1.0
compiled = tf.contrib.eager.defun(fn)
# Compiling `fn` with `defun` hoists all variables outside of the generated
# graph, so initialization happens exactly once.
assert compiled().numpy() == 1.0
assert compiled().numpy() == 2.0
```
Finally, because each input signature is bound to a unique graph, if your
Python function constructs `tf.Variable` objects, then each graph constructed
for that Python function will reference a unique set of variables. To
circumvent this problem, we recommend against compiling Python functions that
create `tf.Variable` objects. Instead, Python functions should either
lexically close over `tf.Variable` objects or accept them as arguments,
preferably encapsulated in an object-oriented container. If you must create
variables inside your Python function and you want each graph generated for it
to reference the same set of variables, add logic to your Python function that
ensures that variables are only created the first time it is called and are
reused for every subsequent invocation; note that this is precisely what
`tf.keras.layers.Layer` objects do, so we recommend using them to represent
variable-bearing computations whenever possible.
Args:
func: function to be compiled. If `func` is None, returns a
decorator that can be invoked with a single argument - `func`. The
end result is equivalent to providing all the arguments up front.
In other words, defun(input_signature=...)(func) is equivalent to
defun(func, input_signature=...). The former allows
the following use case:
@tf.contrib.eager.defun(input_signature=...)
def foo(...):
...
input_signature: A possibly nested sequence of
`tf.contrib.eager.TensorSpec` objects specifying the shapes and dtypes of
the Tensors that will be supplied to this function. If `None`, a separate
function is instantiated for each inferred input signature. If a
signature is specified, every input to `func` must be a `Tensor`, and
`func` cannot accept `**kwargs`.
autograph: Whether `func` should be compiled before
constructing the graph. See https://www.tensorflow.org/guide/autograph
for more information.
experimental_autograph_options: Experimental knobs (in the form of a tuple
of tensorflow.autograph.Feature values) to control behavior when
autograph=True.
experimental_relax_shapes: When true, argument shapes may be relaxed to
avoid unnecessary retracing.
Returns:
If `func` is not None, returns a callable that will execute the compiled
function (and return zero or more `tf.Tensor` objects).
If `func` is None, returns a decorator that, when invoked with a single
`func` argument, returns a callable equivalent to the case above.
Raises:
TypeError: If `input_signature` is neither `None` nor a sequence of
`tf.contrib.eager.TensorSpec` objects."
3516,defun_with_attributes,tensorflow/tensorflow/python/eager/function.py,3705,function,"Compiles a Python function into a callable TensorFlow graph.
This function supports adding extra function attributes. See detailed
documentation in defun(). Currently this is not exposed in public API since we
don't expect user to directly use attributes, and attribute won't work by
itself. This assumption might change in future.
Args:
func: function to be compiled.
input_signature: same as defun()'s input_signature.
attributes: A dictionary of arguments which will be added to function def as
attributes. Currently only support primitive types as value, and only
allowlisted attribute name is allowed. Unallowlisted attribute name or
unsupported value will result into ValueError. `func_name` is also one of
the allowlisted argument which is a python string, and sets the name for
this `ConcreteFunction` in the graph.
autograph: same as defun()'s autograph.
experimental_autograph_options: same as defun()'s
experimental_autograph_options.
experimental_compile: same as defun()'s experimental_compile.
experimental_relax_shapes: same as defun()'s experimental_relax_shapes
experimental_follow_type_hints: see `tf.function`.
Returns:
Same as the return value of defun, with attributes added to the function in
graph."
3517,TfMethodTarget,tensorflow/tensorflow/python/eager/function.py,3784,class,Binding target for methods replaced by function and defun.
3518,target,tensorflow/tensorflow/python/eager/function.py,3794,method,
3519,target_class,tensorflow/tensorflow/python/eager/function.py,3798,method,
3520,call,tensorflow/tensorflow/python/eager/function.py,3806,method,
3521,class_method_to_instance_method,tensorflow/tensorflow/python/eager/function.py,3813,function,Constructs a new `Function` with `self` bound.
3522,ConcreteFunctionGarbageCollector,tensorflow/tensorflow/python/eager/function.py,3889,class,Cleans up reference cycles when a `ConcreteFunction` goes out of scope.
3523,release,tensorflow/tensorflow/python/eager/function.py,3897,method,Call off the FuncGraph deletion.
3524,total_function_cache,tensorflow/tensorflow/python/eager/function_test.py,93,function,
3525,get_entries,tensorflow/tensorflow/python/eager/gradient_input_output_exclusions.py,293,function,"Returns the dict of entries.
Each entry is of the form {op_name, {true|false, indices}}
true: All values are unused.
false: `indices` are the only unused indices.
Note: ops for which all values are used are not printed.
Args:
attr_name: inputs or outputs.
Returns:
A dict from op_type to formatted entry in the dict."
3526,get_function,tensorflow/tensorflow/python/eager/gradient_input_output_exclusions.py,338,function,Generates lookup function with given name and lookup table entries.
3527,get_contents,tensorflow/tensorflow/python/eager/gradient_input_output_exclusions.py,363,function,Returns contents for the generated file.
3528,graph_placeholder,tensorflow/tensorflow/python/eager/graph_only_ops.py,29,function,"Graph-only version of tf.compat.v1.placeholder(), for internal use only."
3529,imperative_grad,tensorflow/tensorflow/python/eager/imperative_grad.py,33,function,"Computes gradients from the imperatively defined tape on top of the stack.
Works by filtering the tape, computing how many downstream usages are of each
tensor and entry, and repeatedly applying backward functions until we have
gradients for all sources.
Args:
tape: the gradient tape which stores the trace.
target: either a Tensor or list of Tensors to be differentiated.
sources: list of Tensors for which we want gradients
output_gradients: if not None, a list of gradient provided for each Target,
or None if we are to use the target's computed downstream gradient.
sources_raw: if not None, a list of the source python objects from which the
sources were generated. Should have the same length as sources. Only needs
to be populated if unconnected_gradients is 'zero'.
unconnected_gradients: determines the value returned if the target and
sources are unconnected. When 'none' the value returned is None wheras when
'zero' a zero tensor in the same shape as the sources is returned.
Returns:
the gradient wrt each of the sources.
Raises:
ValueError: if the arguments are invalid.
RuntimeError: if something goes wrong."
3530,lift_to_graph,tensorflow/tensorflow/python/eager/lift_to_graph.py,205,function,"Copies the tensor and all its inputs recursively to the outer graph.
Args:
tensors: The Tensors to lift.
graph: The graph to lift to.
sources: Optional sequence of nodes to start from. If omitted the whole
subgraph which feeds into `init_tensor` is lifted.
disallowed_placeholders: An optional set of ops which may not appear in the
lifted graph. Defaults to all placeholders.
add_sources: A boolean indicating whether placeholders which are not in
sources should be allowed.
handle_captures: A boolean indicating whether to re-capture s in the new
graph or simply create a vanilla placeholder.
base_graph: The graph from which to lift ops. This will be inferred if not
specified.
op_map: A map contains all the existing nodes that have been lifted to the
destination graph, so they won't be lifted and copied again.
Returns:
A mapping from ops in the current default graph to ops in `graph`.
Raises:
UnliftableError: If a placeholder blocks lifting."
3531,Metric,tensorflow/tensorflow/python/eager/monitoring.py,104,class,The base class of metric.
3532,get_cell,tensorflow/tensorflow/python/eager/monitoring.py,138,method,Retrieves the cell.
3533,CounterCell,tensorflow/tensorflow/python/eager/monitoring.py,147,class,CounterCell stores each value of a Counter.
3534,increase_by,tensorflow/tensorflow/python/eager/monitoring.py,160,method,"Atomically increments the value.
Args:
value: non-negative value."
3535,value,tensorflow/tensorflow/python/eager/monitoring.py,168,method,Retrieves the current value.
3536,Counter,tensorflow/tensorflow/python/eager/monitoring.py,173,class,"A stateful class for updating a cumulative integer metric.
This class encapsulates a set of values (or a single value for a label-less
metric). Each value is identified by a tuple of labels. The class allows the
user to increment each value."
3537,get_cell,tensorflow/tensorflow/python/eager/monitoring.py,194,method,Retrieves the cell.
3538,IntGaugeCell,tensorflow/tensorflow/python/eager/monitoring.py,199,class,A single integer value stored in an `IntGauge`.
3539,set,tensorflow/tensorflow/python/eager/monitoring.py,212,method,"Atomically set the value.
Args:
value: integer value."
3540,value,tensorflow/tensorflow/python/eager/monitoring.py,220,method,Retrieves the current value.
3541,IntGauge,tensorflow/tensorflow/python/eager/monitoring.py,225,class,"A stateful class for updating a gauge-like integer metric.
This class encapsulates a set of integer values (or a single value for a
label-less metric). Each value is identified by a tuple of labels. The class
allows the user to set each value."
3542,get_cell,tensorflow/tensorflow/python/eager/monitoring.py,246,method,Retrieves the cell.
3543,StringGaugeCell,tensorflow/tensorflow/python/eager/monitoring.py,251,class,A single string value stored in an `StringGauge`.
3544,set,tensorflow/tensorflow/python/eager/monitoring.py,264,method,"Atomically set the value.
Args:
value: string value."
3545,value,tensorflow/tensorflow/python/eager/monitoring.py,272,method,Retrieves the current value.
3546,StringGauge,tensorflow/tensorflow/python/eager/monitoring.py,280,class,"A stateful class for updating a gauge-like string metric.
This class encapsulates a set of string values (or a single value for a
label-less metric). Each value is identified by a tuple of labels. The class
allows the user to set each value."
3547,get_cell,tensorflow/tensorflow/python/eager/monitoring.py,301,method,Retrieves the cell.
3548,BoolGaugeCell,tensorflow/tensorflow/python/eager/monitoring.py,306,class,A single boolean value stored in an `BoolGauge`.
3549,set,tensorflow/tensorflow/python/eager/monitoring.py,319,method,"Atomically set the value.
Args:
value: bool value."
3550,value,tensorflow/tensorflow/python/eager/monitoring.py,327,method,Retrieves the current value.
3551,BoolGauge,tensorflow/tensorflow/python/eager/monitoring.py,332,class,"A stateful class for updating a gauge-like bool metric.
This class encapsulates a set of boolean values (or a single value for a
label-less metric). Each value is identified by a tuple of labels. The class
allows the user to set each value."
3552,get_cell,tensorflow/tensorflow/python/eager/monitoring.py,353,method,Retrieves the cell.
3553,SamplerCell,tensorflow/tensorflow/python/eager/monitoring.py,358,class,SamplerCell stores each value of a Sampler.
3554,add,tensorflow/tensorflow/python/eager/monitoring.py,371,method,"Atomically add a sample.
Args:
value: float value."
3555,value,tensorflow/tensorflow/python/eager/monitoring.py,379,method,"Retrieves the current distribution of samples.
Returns:
A HistogramProto describing the distribution of samples."
3556,Buckets,tensorflow/tensorflow/python/eager/monitoring.py,393,class,Bucketing strategies for the samplers.
3557,ExponentialBuckets,tensorflow/tensorflow/python/eager/monitoring.py,410,class,"Exponential bucketing strategy.
Sets up buckets of the form:
[-DBL_MAX, ..., scale * growth^i,
scale * growth_factor^(i + 1), ..., DBL_MAX]."
3558,Sampler,tensorflow/tensorflow/python/eager/monitoring.py,433,class,"A stateful class for updating a cumulative histogram metric.
This class encapsulates a set of histograms (or a single histogram for a
label-less metric) configured with a list of increasing bucket boundaries.
Each histogram is identified by a tuple of labels. The class allows the
user to add a sample to each histogram value."
3559,get_cell,tensorflow/tensorflow/python/eager/monitoring.py,456,method,Retrieves the cell.
3560,MonitoredTimer,tensorflow/tensorflow/python/eager/monitoring.py,461,class,A context manager to measure the walltime and increment a Counter cell.
3561,monitored_timer,tensorflow/tensorflow/python/eager/monitoring.py,484,function,"A function decorator for adding MonitoredTimer support.
Arguments:
cell: the cell associated with the time metric that will be inremented.
Returns:
A decorator that measure the function runtime and increment the specified
counter cell."
3562,ProfilerAlreadyRunningError,tensorflow/tensorflow/python/eager/profiler.py,58,class,
3563,ProfilerNotRunningError,tensorflow/tensorflow/python/eager/profiler.py,62,class,
3564,start,tensorflow/tensorflow/python/eager/profiler.py,67,function,"Start profiling.
Raises:
ProfilerAlreadyRunningError: If another profiling session is running."
3565,stop,tensorflow/tensorflow/python/eager/profiler.py,90,function,"Stop current profiling session and return its result.
Returns:
A binary string of tensorflow.tpu.Trace. User can write the string
to file for offline analysis by tensorboard.
Raises:
ProfilerNotRunningError: If there is no active profiling session."
3566,maybe_create_event_file,tensorflow/tensorflow/python/eager/profiler.py,118,function,"Create an empty event file if not already exists.
This event file indicates that we have a plugins/profile/ directory in the
current logdir.
Args:
logdir: log directory."
3567,save,tensorflow/tensorflow/python/eager/profiler.py,140,function,"Save profile result to TensorBoard logdir.
Args:
logdir: log directory read by TensorBoard.
result: profiling result returned by stop()."
3568,start_profiler_server,tensorflow/tensorflow/python/eager/profiler.py,157,function,"Start a profiler grpc server that listens to given port.
The profiler server will keep the program running even the training finishes.
Please shutdown the server with CTRL-C. It can be used in both eager mode and
graph mode. The service defined in
tensorflow/core/profiler/profiler_service.proto. Please use
tensorflow/contrib/tpu/profiler/capture_tpu_profile to capture tracable
file following https://cloud.google.com/tpu/docs/cloud-tpu-tools#capture_trace
Args:
port: port profiler server listens to."
3569,Profiler,tensorflow/tensorflow/python/eager/profiler.py,176,class,"Context-manager eager profiler api.
Example usage:
```python
with Profiler(""/path/to/logdir""):
# do some work
```"
3570,start_tracing,tensorflow/tensorflow/python/eager/profiler_client.py,26,function,"Sends grpc requests to profiler server to perform on-demand profiling.
This method will block caller thread until receives tracing result.
Args:
service_addr: Address of profiler service e.g. localhost:6009.
logdir: Path of TensorBoard log directory e.g. /tmp/tb_log.
duration_ms: Duration of tracing or monitoring in ms.
worker_list: The list of worker TPUs that we are about to profile in the
current session. (TPU only)
include_dataset_ops: Set to false to profile longer traces.
num_tracing_attempts: Automatically retry N times when no trace event is
collected.
Raises:
UnavailableError: If no trace event is collected."
3571,monitor,tensorflow/tensorflow/python/eager/profiler_client.py,54,function,"Sends grpc requests to profiler server to perform on-demand monitoring.
This method will block caller thread until receives monitoring result.
Args:
service_addr: Address of profiler service e.g. localhost:6009.
duration_ms: Duration of tracing or monitoring in ms.
monitoring_level: Choose a monitoring level between 1 and 2 to monitor your
job. Level 2 is more verbose than level 1 and shows more metrics.
display_timestamp: Set to true to display timestamp in monitoring result.
Returns:
A string of monitoring output."
3572,connect_to_remote_host,tensorflow/tensorflow/python/eager/remote.py,42,function,"Connects to a single machine to enable remote execution on it.
Will make devices on the remote host available to use. Note that calling this
more than once will work, but will invalidate any tensor handles on the old
remote devices.
Using the default job_name of worker, you can schedule ops to run remotely as
follows:
```python
# When eager execution is enabled, connect to the remote host.
tf.config.experimental_connect_to_host(""exampleaddr.com:9876"")
with ops.device(""job:worker/replica:0/task:1/device:CPU:0""):
# The following tensors should be resident on the remote device, and the op
# will also execute remotely.
x1 = array_ops.ones([2, 2])
x2 = array_ops.ones([2, 2])
y = math_ops.matmul(x1, x2)
```
Args:
remote_host: a single or a list the remote server addr in host-port format.
job_name: The job name under which the new server will be accessible.
Raises:
ValueError: if remote_host is None."
3573,connect_to_cluster,tensorflow/tensorflow/python/eager/remote.py,81,function,"Connects to the given cluster.
Will make devices on the cluster available to use. Note that calling this more
than once will work, but will invalidate any tensor handles on the old remote
devices.
If the given local job name is not present in the cluster specification, it
will be automatically added, using an unused port on the localhost.
Device filters can be specified to isolate groups of remote tasks to avoid
undesired accesses between workers. Workers accessing resources or launching
ops / functions on filtered remote devices will result in errors (unknown
devices). For any remote task, if no device filter is present, all cluster
devices will be visible; if any device filter is specified, it can only
see devices matching at least one filter. Devices on the task itself are
always visible. Device filters can be particially specified.
For example, for a cluster set up for parameter server training, the following
device filters might be specified:
```python
cdf = tf.config.experimental.ClusterDeviceFilters()
# For any worker, only the devices on PS nodes and itself are visible
for i in range(num_workers):
cdf.set_device_filters('worker', i, ['/job:ps'])
# Similarly for any ps, only the devices on workers and itself are visible
for i in range(num_ps):
cdf.set_device_filters('ps', i, ['/job:worker'])
tf.config.experimental_connect_to_cluster(cluster_def,
cluster_device_filters=cdf)
```
Args:
cluster_spec_or_resolver: A `ClusterSpec` or `ClusterResolver` describing
the cluster.
job_name: The name of the local job.
task_index: The local task index.
protocol: The communication protocol, such as `""grpc""`. If unspecified, will
use the default from `python/platform/remote_utils.py`.
make_master_device_default: If True and a cluster resolver is passed, will
automatically enter the master task device scope, which indicates the
master becomes the default device to run ops. It won't do anything if
a cluster spec is passed. Will throw an error if the caller is currently
already in some device scope.
cluster_device_filters: an instance of
`tf.train.experimental/ClusterDeviceFilters` that specify device filters
to the remote tasks in cluster."
3574,run_benchmark,tensorflow/tensorflow/python/eager/remote_benchmarks_test.py,45,function,
3575,Foo,tensorflow/tensorflow/python/eager/remote_benchmarks_test.py,62,class,
3576,RemoteWorkerMicroBenchmarks,tensorflow/tensorflow/python/eager/remote_benchmarks_test.py,78,class,
3577,benchmark_send_mirroring_off,tensorflow/tensorflow/python/eager/remote_benchmarks_test.py,95,method,
3578,benchmark_send_mirroring_on,tensorflow/tensorflow/python/eager/remote_benchmarks_test.py,115,method,
3579,benchmark_worker_mirroring_off,tensorflow/tensorflow/python/eager/remote_benchmarks_test.py,135,method,
3580,benchmark_worker_mirroring_on,tensorflow/tensorflow/python/eager/remote_benchmarks_test.py,157,method,
3581,benchmark_create_vars_inside_function,tensorflow/tensorflow/python/eager/remote_benchmarks_test.py,179,method,
3582,remote_func,tensorflow/tensorflow/python/eager/remote_benchmarks_test.py,101,method,
3583,func,tensorflow/tensorflow/python/eager/remote_benchmarks_test.py,104,method,
3584,remote_func,tensorflow/tensorflow/python/eager/remote_benchmarks_test.py,121,method,
3585,func,tensorflow/tensorflow/python/eager/remote_benchmarks_test.py,124,method,
3586,remote_func,tensorflow/tensorflow/python/eager/remote_benchmarks_test.py,143,method,
3587,func,tensorflow/tensorflow/python/eager/remote_benchmarks_test.py,146,method,
3588,remote_func,tensorflow/tensorflow/python/eager/remote_benchmarks_test.py,165,method,
3589,func,tensorflow/tensorflow/python/eager/remote_benchmarks_test.py,168,method,
3590,func,tensorflow/tensorflow/python/eager/remote_benchmarks_test.py,182,method,
3591,remote_func,tensorflow/tensorflow/python/eager/remote_benchmarks_test.py,187,method,
3592,get_server_def,tensorflow/tensorflow/python/eager/remote_cluster_test.py,46,function,Returns a server def with a single job + multiple tasks.
3593,get_server_def,tensorflow/tensorflow/python/eager/remote_execution_test.py,46,function,Returns a server def with a single job + multiple tasks.
3594,Tape,tensorflow/tensorflow/python/eager/tape.py,35,class,Represents a gradient propagation trace.
3595,watched_variables,tensorflow/tensorflow/python/eager/tape.py,43,method,
3596,push_new_tape,tensorflow/tensorflow/python/eager/tape.py,47,function,Pushes a new tape onto the tape stack.
3597,push_tape,tensorflow/tensorflow/python/eager/tape.py,53,function,Pushes an existing tape onto the tape stack.
3598,watch,tensorflow/tensorflow/python/eager/tape.py,58,function,Marks this tensor to be watched by the given tape.
3599,VariableWatcher,tensorflow/tensorflow/python/eager/tape.py,63,class,"A scope that tracks all trainable variable accesses within it.
This explicitly ignores variables that are not marked as trainable.
Sample usage:
var = tf.Variable(0.0)
with VariableWatcher() as variable_watcher:
var.assign_add(1.0)
assert variable_watcher.watched_variables == [var]"
3600,watched_variables,tensorflow/tensorflow/python/eager/tape.py,89,method,Returns a tuple of variables accessed under this scope.
3601,watch_variable,tensorflow/tensorflow/python/eager/tape.py,95,function,Marks this variable to be watched by the given tape.
3602,variable_accessed,tensorflow/tensorflow/python/eager/tape.py,108,function,"Notifies all tapes in the stack that a variable has been accessed.
Args:
variable: variable to be watched."
3603,variables_accessed,tensorflow/tensorflow/python/eager/tape.py,125,function,"Notifies all tapes in the stack that variables have been accessed.
Only trainable variables are marked as accessed.
Args:
variables: iterable of variables to mark as accessed."
3604,pop_tape,tensorflow/tensorflow/python/eager/tape.py,149,function,Pops the given tape in the stack.
3605,stop_recording,tensorflow/tensorflow/python/eager/tape.py,155,function,Stop all gradient recording (backprop and forwardprop).
3606,should_record_backprop,tensorflow/tensorflow/python/eager/tape.py,167,function,"Returns true if any tape in the stack watches any of these tensors.
Only takes GradientTapes into account, not forward accumulators.
Args:
tensors: Tensors to check, typically inputs to an operation.
Returns:
Boolean, whether any tape watches any of `tensors`."
3607,record_operation,tensorflow/tensorflow/python/eager/tape.py,181,function,Records the operation on all tapes in the stack.
3608,record_operation_backprop_only,tensorflow/tensorflow/python/eager/tape.py,189,function,Records the operation on all backward tapes in the stack.
3609,record_operation_forwardprop_only,tensorflow/tensorflow/python/eager/tape.py,197,function,"Records the operation on all forward accumulators in the stack.
Args:
op_type: a string for the operation type, used in the backprop code
output_tensors: a list of Python Tensor objects output by the operation
input_tensors: a list of input Tensors to the recorded operation
backward_function: the function to be called to, given the gradients of the
output tensors, produce the gradients of the input tensors. This function
is automatically transposed to produce output gradients given input
gradients.
forwardprop_output_indices: indicates any output_tensors which contain JVPs.
Typically these will have come from TFE_Py_PackForwardGradients. May be
None or an empty sequence if there are no JVP outputs from the operation."
3610,delete_trace,tensorflow/tensorflow/python/eager/tape.py,219,function,Deletes traces for this Tensor from all tapes in the stack.
3611,could_possibly_record,tensorflow/tensorflow/python/eager/tape.py,224,function,Returns True if any tape is active.
3612,two_outputs,tensorflow/tensorflow/python/eager/tape_test.py,39,function,
3613,gradient_is_constant,tensorflow/tensorflow/python/eager/tape_test.py,55,function,
3614,VariableHolder,tensorflow/tensorflow/python/eager/wrap_function.py,46,class,Holds variables for a python function.
3615,variables,tensorflow/tensorflow/python/eager/wrap_function.py,56,method,
3616,variable_creator_scope,tensorflow/tensorflow/python/eager/wrap_function.py,59,method,Creates variables & adds them to collections to match legacy code.
3617,call_with_variable_creator_scope,tensorflow/tensorflow/python/eager/wrap_function.py,89,method,
3618,wrapped,tensorflow/tensorflow/python/eager/wrap_function.py,91,method,
3619,WrappedFunction,tensorflow/tensorflow/python/eager/wrap_function.py,220,class,Wraps a tf V1 piece of code in a function.
3620,prune,tensorflow/tensorflow/python/eager/wrap_function.py,249,method,"Extract a subgraph of this function's underlying graph.
Wraps the subgraph in a new `WrappedFunction` object.
Args:
feeds: Input tensors to the subgraph to extract, as `Tensor` objects.
fetches: Possibly-nested Python data structure containing information
about outputs of the target subgraph. Each entry can either be a
`Tensor` object (for data outputs), an `Operation` object (for control
outputs), or a `TensorInfo` proto. Any additional shape/dtype
information provided in a `TensorInfo` and not present in the original
graph will be added to the returned subgraph.
name: (optional) Name to give to the underlying `FuncGraph` of the
returned object. If no name is provided, the graph's name will be
`""pruned""`.
input_signature: (optional) possibly-nested Python data structure
containing `TensorSpec` objects, with which to populate the returned
functions's `FuncGraph`'s `structured_input_signature` field.
Returns:
A new `WrappedFunction` object containing a copy of the portion of this
object's graph that goes from `feeds` to `fetches`."
3621,WrappedGraph,tensorflow/tensorflow/python/eager/wrap_function.py,409,class,"Class for wrapping multiple TF 1.X functions in a single graph.
Maintains a dictionary mapping names to wrapped functions. See
`tf.compat.v1.wrap_function` to learn more about wrapping V1 functions.
Functions wrapped using this class have access to variables and collections
created in other wrapped functions, using the standard TF 1.X API (
`tf.compat.v1.get_variable` or
`tf.compat.v1.get_default_graph().get_collection(...)`)
Outside a function, variables and collections may be accessed using the
`variables` and `graph` properties.
Example:
```
def add_v1(x):
with tf.compat.v1.variable_scope('vars', reuse=tf.compat.v1.AUTO_REUSE):
v = tf.compat.v1.get_variable('v', shape=[], dtype=tf.int32)
return v + x
def increment_var_v1(x):
with tf.compat.v1.variable_scope('vars', reuse=tf.compat.v1.AUTO_REUSE):
v = tf.compat.v1.get_variable('v', shape=[], dtype=tf.int32)
return v.assign_add(x)
g = WrappedGraph()
add = g.wrap_function(add_v1, [tf.TensorSpec([], tf.int32)])
increment_var = g.wrap_function(increment_var_v1,
[tf.TensorSpec([], tf.int32)])
assert len(g.variables) == 1
assert g.variables[0].numpy() == 0
increment_var(tf.constant(5))
assert g.variables[0].numpy() == 5
```"
3622,functions,tensorflow/tensorflow/python/eager/wrap_function.py,463,method,
3623,variables,tensorflow/tensorflow/python/eager/wrap_function.py,467,method,
3624,wrap_function,tensorflow/tensorflow/python/eager/wrap_function.py,470,method,"Wraps a TF 1.X function and returns an eager-compatible function.
All functions wrapped in the same `WrappedGraph` will have access to the
same graph (`tf.compat.v1.get_default_graph` to get the graph object
within a function, or `WrappedGraph.graph` to get the graph outside a
function). Variables created within the function will be added to the
`variables` list.
Function inputs: All inputs to the function must be tensors (nested ok),
with their shapes and dtypes defined in the `signature` argument.
Function outputs:
* The 1.X function may return tensors, variables, and ops. The wrapped
eager-compatible function will always return tensors in the same nested
structure.
* Variables are replaced with a tensor containing the latest read values.
* Returned ops are executed, and replaced with None.
* The order of op execution and variable reads in the return is
nondeterministic. For example:
```
def update_var(x):
v = tf.Variable(0)
op = tf.compat.v1.assign(v, x).op
return v, op
g = WrappedGraph()
fn = g.wrap_function(update_var)
read_value, _ = fn(tf.constant(3))
print(read_value.numpy()) # could be 0 or 3
print(g.variables[0].numpy()) # always 3
```
To ensure that ops in the function are executed (e.g. ops added to the
`tf.GraphKeys.UPDATE_OPS` collection), include them in the function returns.
Args:
fn: a 1.X tensorflow function.
signature: a possibly nested sequence of `TensorSpecs` specifying the
shapes and dtypes of the arguments.
name: an optional string name for the function. The function will be saved
with key `name` in the `functions` dictionary.
Returns:
An eager-compatible function."
3625,wrap_function,tensorflow/tensorflow/python/eager/wrap_function.py,560,function,"Wraps the TF 1.x function fn into a graph function.
The python function `fn` will be called once with symbolic arguments specified
in the `signature`, traced, and turned into a graph function. Any variables
created by `fn` will be owned by the object returned by `wrap_function`. The
resulting graph function can be called with tensors which match the
signature.
```python
def f(x, do_add):
v = tf.Variable(5.0)
if do_add:
op = v.assign_add(x)
else:
op = v.assign_sub(x)
with tf.control_dependencies([op]):
return v.read_value()
f_add = tf.compat.v1.wrap_function(f, [tf.TensorSpec((), tf.float32), True])
assert float(f_add(1.0)) == 6.0
assert float(f_add(1.0)) == 7.0
# Can call tf.compat.v1.wrap_function again to get a new trace, a new set
# of variables, and possibly different non-template arguments.
f_sub= tf.compat.v1.wrap_function(f, [tf.TensorSpec((), tf.float32), False])
assert float(f_sub(1.0)) == 4.0
assert float(f_sub(1.0)) == 3.0
```
Both `tf.compat.v1.wrap_function` and `tf.function` create a callable
TensorFlow graph. But while `tf.function` runs all stateful operations
(e.g. `tf.print`) and sequences operations to provide the same semantics as
eager execution, `wrap_function` is closer to the behavior of `session.run` in
TensorFlow 1.x. It will not run any operations unless they are required to
compute the function's outputs, either through a data dependency or a control
dependency. Nor will it sequence operations.
Unlike `tf.function`, `wrap_function` will only trace the Python function
once. As with placeholders in TF 1.x, shapes and dtypes must be provided to
`wrap_function`'s `signature` argument.
Since it is only traced once, variables and state may be created inside the
function and owned by the function wrapper object.
Args:
fn: python function to be wrapped
signature: the placeholder and python arguments to be passed to the wrapped
function
name: Optional. The name of the function.
Returns:
the wrapped graph function."
3626,function_from_graph_def,tensorflow/tensorflow/python/eager/wrap_function.py,633,function,"Creates a ConcreteFunction from a GraphDef.
Args:
graph_def: A GraphDef to make a function out of.
inputs: A Tensor name or nested structure of names in `graph_def` which
should be inputs to the function.
outputs: A Tensor name or nested structure of names in `graph_def` which
should be outputs of the function.
Returns:
A ConcreteFunction."
3627,HVPBenchmarks,tensorflow/tensorflow/python/eager/benchmarks/resnet50/hvp_test.py,108,class,
3628,benchmark_forward_over_backward_hvp_eager,tensorflow/tensorflow/python/eager/benchmarks/resnet50/hvp_test.py,144,method,
3629,benchmark_forward_over_backward_hvp_function,tensorflow/tensorflow/python/eager/benchmarks/resnet50/hvp_test.py,149,method,
3630,benchmark_tf_gradients_forward_over_backward_hvp_function,tensorflow/tensorflow/python/eager/benchmarks/resnet50/hvp_test.py,154,method,
3631,benchmark_backward_over_backward_hvp_eager,tensorflow/tensorflow/python/eager/benchmarks/resnet50/hvp_test.py,159,method,
3632,benchmark_backward_over_backward_hvp_function,tensorflow/tensorflow/python/eager/benchmarks/resnet50/hvp_test.py,164,method,
3633,benchmark_backward_over_forward_hvp_eager,tensorflow/tensorflow/python/eager/benchmarks/resnet50/hvp_test.py,169,method,
3634,benchmark_backward_over_forward_hvp_function,tensorflow/tensorflow/python/eager/benchmarks/resnet50/hvp_test.py,174,method,
3635,ResNet50,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50.py,168,class,"Instantiates the ResNet50 architecture.
Args:
data_format: format for the image. Either 'channels_first' or
'channels_last'. 'channels_first' is typically faster on GPUs while
'channels_last' is typically faster on CPUs. See
https://www.tensorflow.org/performance/performance_guide#data_formats
name: Prefix applied to names of variables created in the model.
trainable: Is the model trainable? If true, performs backward
and optimization after call() method.
include_top: whether to include the fully-connected layer at the top of the
network.
pooling: Optional pooling mode for feature extraction when `include_top`
is `False`.
- `None` means that the output of the model will be the 4D tensor
output of the last convolutional layer.
- `avg` means that global average pooling will be applied to the output of
the last convolutional layer, and thus the output of the model will be
a 2D tensor.
- `max` means that global max pooling will be applied.
block3_strides: whether to add a stride of 2 to block3 to make it compatible
with tf.slim ResNet implementation.
average_pooling: whether to do average pooling of block4 features before
global pooling.
classes: optional number of classes to classify images into, only to be
specified if `include_top` is True.
Raises:
ValueError: in case of invalid argument for data_format."
3636,call,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50.py,297,method,"Call the ResNet50 model.
Args:
inputs: Images to compute features for.
training: Whether model is in training phase.
intermediates_dict: `None` or dictionary. If not None, accumulate feature
maps from intermediate blocks into the dictionary.
""""
Returns:
Tensor with featuremap."
3637,conv_block,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50.py,220,method,
3638,id_block,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50.py,229,method,
3639,data_format,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_graph_test.py,29,function,
3640,image_shape,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_graph_test.py,33,function,
3641,random_batch,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_graph_test.py,39,function,
3642,ResNet50Benchmarks,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_graph_test.py,69,class,
3643,benchmark_graph_apply,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_graph_test.py,79,method,
3644,benchmark_graph_train,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_graph_test.py,102,method,
3645,compute_gradients,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test.py,37,function,
3646,apply_gradients,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test.py,54,function,
3647,events_from_logdir,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test.py,76,function,"Returns all events in the single eventfile in logdir.
Args:
logdir: The directory in which the single event file is sought.
Returns:
A list of all tf.compat.v1.Event protos from the single event file.
Raises:
AssertionError: If logdir does not contain exactly one file."
3648,MockIterator,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test.py,260,class,
3649,next,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test.py,265,method,
3650,ResNet50Benchmarks,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test.py,269,class,
3651,benchmark_eager_apply_sync,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test.py,331,method,
3652,benchmark_eager_apply_async,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test.py,336,method,
3653,benchmark_eager_apply_with_defun,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test.py,344,method,
3654,benchmark_eager_train_sync,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test.py,392,method,
3655,benchmark_eager_train_async,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test.py,397,method,
3656,benchmark_eager_train_with_defun,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test.py,405,method,
3657,benchmark_eager_train_datasets,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test.py,410,method,
3658,benchmark_eager_train_datasets_with_defun,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test.py,424,method,
3659,make_iterator,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test.py,412,method,
3660,make_iterator,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test.py,426,method,
3661,device_and_data_format,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test_util.py,26,function,
3662,random_batch,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test_util.py,32,function,Create synthetic resnet50 images and labels for testing.
3663,report,tensorflow/tensorflow/python/eager/benchmarks/resnet50/resnet50_test_util.py,50,function,
3664,assert_no_leak,tensorflow/tensorflow/python/eager/memory_tests/memory_test_util.py,48,function,Assert memory usage doesn't increase beyond given threshold for f.
3665,memory_profiler_is_available,tensorflow/tensorflow/python/eager/memory_tests/memory_test_util.py,78,function,
3666,input_layer,tensorflow/tensorflow/python/feature_column/feature_column.py,234,function,"Returns a dense `Tensor` as input layer based on given `feature_columns`.
Generally a single example in training data is described with FeatureColumns.
At the first layer of the model, this column oriented data should be converted
to a single `Tensor`.
Example:
```python
price = numeric_column('price')
keywords_embedded = embedding_column(
categorical_column_with_hash_bucket(""keywords"", 10K), dimensions=16)
columns = [price, keywords_embedded, ...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
dense_tensor = input_layer(features, columns)
for units in [128, 64, 32]:
dense_tensor = tf.compat.v1.layers.dense(dense_tensor, units, tf.nn.relu)
prediction = tf.compat.v1.layers.dense(dense_tensor, 1)
```
Args:
features: A mapping from key to tensors. `_FeatureColumn`s look up via these
keys. For example `numeric_column('price')` will look at 'price' key in
this dict. Values can be a `SparseTensor` or a `Tensor` depends on
corresponding `_FeatureColumn`.
feature_columns: An iterable containing the FeatureColumns to use as inputs
to your model. All items should be instances of classes derived from
`_DenseColumn` such as `numeric_column`, `embedding_column`,
`bucketized_column`, `indicator_column`. If you have categorical features,
you can wrap them with an `embedding_column` or `indicator_column`.
weight_collections: A list of collection names to which the Variable will be
added. Note that variables will also be added to collections
`tf.GraphKeys.GLOBAL_VARIABLES` and `ops.GraphKeys.MODEL_VARIABLES`.
trainable: If `True` also add the variable to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see `tf.Variable`).
cols_to_vars: If not `None`, must be a dictionary that will be filled with a
mapping from `_FeatureColumn` to list of `Variable`s. For example, after
the call, we might have cols_to_vars =
{_EmbeddingColumn(
categorical_column=_HashedCategoricalColumn(
key='sparse_feature', hash_bucket_size=5, dtype=tf.string),
dimension=10): [<tf.Variable 'some_variable:0' shape=(5, 10),
<tf.Variable 'some_variable:1' shape=(5, 10)]}
If a column creates no variables, its value will be an empty list.
cols_to_output_tensors: If not `None`, must be a dictionary that will be
filled with a mapping from '_FeatureColumn' to the associated
output `Tensor`s.
Returns:
A `Tensor` which represents input layer of a model. Its shape
is (batch_size, first_layer_dimension) and its dtype is `float32`.
first_layer_dimension is determined based on given `feature_columns`.
Raises:
ValueError: if an item in `feature_columns` is not a `_DenseColumn`."
3667,InputLayer,tensorflow/tensorflow/python/feature_column/feature_column.py,309,class,An object-oriented version of `input_layer` that reuses variables.
3668,name,tensorflow/tensorflow/python/feature_column/feature_column.py,340,method,
3669,non_trainable_variables,tensorflow/tensorflow/python/feature_column/feature_column.py,344,method,
3670,non_trainable_weights,tensorflow/tensorflow/python/feature_column/feature_column.py,348,method,
3671,trainable_variables,tensorflow/tensorflow/python/feature_column/feature_column.py,352,method,
3672,trainable_weights,tensorflow/tensorflow/python/feature_column/feature_column.py,356,method,
3673,variables,tensorflow/tensorflow/python/feature_column/feature_column.py,360,method,
3674,weights,tensorflow/tensorflow/python/feature_column/feature_column.py,364,method,
3675,linear_model,tensorflow/tensorflow/python/feature_column/feature_column.py,369,function,"Returns a linear prediction `Tensor` based on given `feature_columns`.
This function generates a weighted sum based on output dimension `units`.
Weighted sum refers to logits in classification problems. It refers to the
prediction itself for linear regression problems.
Note on supported columns: `linear_model` treats categorical columns as
`indicator_column`s. To be specific, assume the input as `SparseTensor` looks
like:
```python
shape = [2, 2]
{
[0, 0]: ""a""
[1, 0]: ""b""
[1, 1]: ""c""
}
```
`linear_model` assigns weights for the presence of ""a"", ""b"", ""c' implicitly,
just like `indicator_column`, while `input_layer` explicitly requires wrapping
each of categorical columns with an `embedding_column` or an
`indicator_column`.
Example of usage:
```python
price = numeric_column('price')
price_buckets = bucketized_column(price, boundaries=[0., 10., 100., 1000.])
keywords = categorical_column_with_hash_bucket(""keywords"", 10K)
keywords_price = crossed_column('keywords', price_buckets, ...)
columns = [price_buckets, keywords, keywords_price ...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
prediction = linear_model(features, columns)
```
The `sparse_combiner` argument works as follows
For example, for two features represented as the categorical columns:
```python
# Feature 1
shape = [2, 2]
{
[0, 0]: ""a""
[0, 1]: ""b""
[1, 0]: ""c""
}
# Feature 2
shape = [2, 3]
{
[0, 0]: ""d""
[1, 0]: ""e""
[1, 1]: ""f""
[1, 2]: ""f""
}
```
with `sparse_combiner` as ""mean"", the linear model outputs consequently
are:
```
y_0 = 1.0 / 2.0 * ( w_a + w_b ) + w_d + b
y_1 = w_c + 1.0 / 3.0 * ( w_e + 2.0 * w_f ) + b
```
where `y_i` is the output, `b` is the bias, and `w_x` is the weight
assigned to the presence of `x` in the input features.
Args:
features: A mapping from key to tensors. `_FeatureColumn`s look up via these
keys. For example `numeric_column('price')` will look at 'price' key in
this dict. Values are `Tensor` or `SparseTensor` depending on
corresponding `_FeatureColumn`.
feature_columns: An iterable containing the FeatureColumns to use as inputs
to your model. All items should be instances of classes derived from
`_FeatureColumn`s.
units: An integer, dimensionality of the output space. Default value is 1.
sparse_combiner: A string specifying how to reduce if a categorical column
is multivalent. Except `numeric_column`, almost all columns passed to
`linear_model` are considered as categorical columns. It combines each
categorical column independently. Currently ""mean"", ""sqrtn"" and ""sum"" are
supported, with ""sum"" the default for linear model. ""sqrtn"" often achieves
good accuracy, in particular with bag-of-words columns.
* ""sum"": do not normalize features in the column
* ""mean"": do l1 normalization on features in the column
* ""sqrtn"": do l2 normalization on features in the column
weight_collections: A list of collection names to which the Variable will be
added. Note that, variables will also be added to collections
`tf.GraphKeys.GLOBAL_VARIABLES` and `ops.GraphKeys.MODEL_VARIABLES`.
trainable: If `True` also add the variable to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see `tf.Variable`).
cols_to_vars: If not `None`, must be a dictionary that will be filled with a
mapping from `_FeatureColumn` to associated list of `Variable`s. For
example, after the call, we might have cols_to_vars = {
_NumericColumn(
key='numeric_feature1', shape=(1,):
[<tf.Variable 'linear_model/price2/weights:0' shape=(1, 1)>],
'bias': [<tf.Variable 'linear_model/bias_weights:0' shape=(1,)>],
_NumericColumn(
key='numeric_feature2', shape=(2,)):
[<tf.Variable 'linear_model/price1/weights:0' shape=(2, 1)>]}
If a column creates no variables, its value will be an empty list. Note
that cols_to_vars will also contain a string key 'bias' that maps to a
list of Variables.
Returns:
A `Tensor` which represents predictions/logits of a linear model. Its shape
is (batch_size, units) and its dtype is `float32`.
Raises:
ValueError: if an item in `feature_columns` is neither a `_DenseColumn`
nor `_CategoricalColumn`."
3676,make_parse_example_spec,tensorflow/tensorflow/python/feature_column/feature_column.py,761,function,"Creates parsing spec dictionary from input feature_columns.
The returned dictionary can be used as arg 'features' in
`tf.io.parse_example`.
Typical usage example:
```python
# Define features and transformations
feature_a = categorical_column_with_vocabulary_file(...)
feature_b = numeric_column(...)
feature_c_bucketized = bucketized_column(numeric_column(""feature_c""), ...)
feature_a_x_feature_c = crossed_column(
columns=[""feature_a"", feature_c_bucketized], ...)
feature_columns = set(
[feature_b, feature_c_bucketized, feature_a_x_feature_c])
features = tf.io.parse_example(
serialized=serialized_examples,
features=make_parse_example_spec(feature_columns))
```
For the above example, make_parse_example_spec would return the dict:
```python
{
""feature_a"": parsing_ops.VarLenFeature(tf.string),
""feature_b"": parsing_ops.FixedLenFeature([1], dtype=tf.float32),
""feature_c"": parsing_ops.FixedLenFeature([1], dtype=tf.float32)
}
```
Args:
feature_columns: An iterable containing all feature columns. All items
should be instances of classes derived from `_FeatureColumn`.
Returns:
A dict mapping each feature key to a `FixedLenFeature` or `VarLenFeature`
value.
Raises:
ValueError: If any of the given `feature_columns` is not a `_FeatureColumn`
instance."
3677,get_linear_model_bias,tensorflow/tensorflow/python/feature_column/feature_column_test.py,1297,function,
3678,get_linear_model_column_var,tensorflow/tensorflow/python/feature_column/feature_column_test.py,1302,function,
3679,get_keras_linear_model_predictions,tensorflow/tensorflow/python/feature_column/feature_column_test.py,1307,function,
3680,StateManager,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,176,class,"Manages the state associated with FeatureColumns.
Some `FeatureColumn`s create variables or resources to assist their
computation. The `StateManager` is responsible for creating and storing these
objects since `FeatureColumn`s are supposed to be stateless configuration
only."
3681,create_variable,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,185,method,"Creates a new variable.
Args:
feature_column: A `FeatureColumn` object this variable corresponds to.
name: variable name.
shape: variable shape.
dtype: The type of the variable. Defaults to `self.dtype` or `float32`.
trainable: Whether this variable is trainable or not.
use_resource: If true, we use resource variables. Otherwise we use
RefVariable.
initializer: initializer instance (callable).
Returns:
The created variable."
3682,add_variable,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,211,method,"Adds an existing variable to the state.
Args:
feature_column: A `FeatureColumn` object to associate this variable with.
var: The variable."
3683,get_variable,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,221,method,"Returns an existing variable.
Args:
feature_column: A `FeatureColumn` object this variable corresponds to.
name: variable name."
3684,add_resource,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,231,method,"Creates a new resource.
Resources can be things such as tables, variables, trackables, etc.
Args:
feature_column: A `FeatureColumn` object this resource corresponds to.
name: Name of the resource.
resource: The resource.
Returns:
The created resource."
3685,has_resource,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,247,method,"Returns true iff a resource with same name exists.
Resources can be things such as tables, variables, trackables, etc.
Args:
feature_column: A `FeatureColumn` object this variable corresponds to.
name: Name of the resource."
3686,get_resource,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,259,method,"Returns an already created resource.
Resources can be things such as tables, variables, trackables, etc.
Args:
feature_column: A `FeatureColumn` object this variable corresponds to.
name: Name of the resource."
3687,make_parse_example_spec_v2,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,429,function,"Creates parsing spec dictionary from input feature_columns.
The returned dictionary can be used as arg 'features' in
`tf.io.parse_example`.
Typical usage example:
```python
# Define features and transformations
feature_a = tf.feature_column.categorical_column_with_vocabulary_file(...)
feature_b = tf.feature_column.numeric_column(...)
feature_c_bucketized = tf.feature_column.bucketized_column(
tf.feature_column.numeric_column(""feature_c""), ...)
feature_a_x_feature_c = tf.feature_column.crossed_column(
columns=[""feature_a"", feature_c_bucketized], ...)
feature_columns = set(
[feature_b, feature_c_bucketized, feature_a_x_feature_c])
features = tf.io.parse_example(
serialized=serialized_examples,
features=tf.feature_column.make_parse_example_spec(feature_columns))
```
For the above example, make_parse_example_spec would return the dict:
```python
{
""feature_a"": parsing_ops.VarLenFeature(tf.string),
""feature_b"": parsing_ops.FixedLenFeature([1], dtype=tf.float32),
""feature_c"": parsing_ops.FixedLenFeature([1], dtype=tf.float32)
}
```
Args:
feature_columns: An iterable containing all feature columns. All items
should be instances of classes derived from `FeatureColumn`.
Returns:
A dict mapping each feature key to a `FixedLenFeature` or `VarLenFeature`
value.
Raises:
ValueError: If any of the given `feature_columns` is not a `FeatureColumn`
instance."
3688,embedding_column,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,491,function,"`DenseColumn` that converts from sparse, categorical input.
Use this when your inputs are sparse, but you want to convert them to a dense
representation (e.g., to feed to a DNN).
Inputs must be a `CategoricalColumn` created by any of the
`categorical_column_*` function. Here is an example of using
`embedding_column` with `DNNClassifier`:
```python
video_id = categorical_column_with_identity(
key='video_id', num_buckets=1000000, default_value=0)
columns = [embedding_column(video_id, 9),...]
estimator = tf.estimator.DNNClassifier(feature_columns=columns, ...)
label_column = ...
def input_fn():
features = tf.io.parse_example(
..., features=make_parse_example_spec(columns + [label_column]))
labels = features.pop(label_column.name)
return features, labels
estimator.train(input_fn=input_fn, steps=100)
```
Here is an example using `embedding_column` with model_fn:
```python
def model_fn(features, ...):
video_id = categorical_column_with_identity(
key='video_id', num_buckets=1000000, default_value=0)
columns = [embedding_column(video_id, 9),...]
dense_tensor = input_layer(features, columns)
# Form DNN layers, calculate loss, and return EstimatorSpec.
...
```
Args:
categorical_column: A `CategoricalColumn` created by a
`categorical_column_with_*` function. This column produces the sparse IDs
that are inputs to the embedding lookup.
dimension: An integer specifying dimension of the embedding, must be > 0.
combiner: A string specifying how to reduce if there are multiple entries in
a single row. Currently 'mean', 'sqrtn' and 'sum' are supported, with
'mean' the default. 'sqrtn' often achieves good accuracy, in particular
with bag-of-words columns. Each of this can be thought as example level
normalizations on the column. For more information, see
`tf.embedding_lookup_sparse`.
initializer: A variable initializer function to be used in embedding
variable initialization. If not specified, defaults to
`truncated_normal_initializer` with mean `0.0` and
standard deviation `1/sqrt(dimension)`.
ckpt_to_load_from: String representing checkpoint name/pattern from which to
restore column weights. Required if `tensor_name_in_ckpt` is not `None`.
tensor_name_in_ckpt: Name of the `Tensor` in `ckpt_to_load_from` from which
to restore the column weights. Required if `ckpt_to_load_from` is not
`None`.
max_norm: If not `None`, embedding values are l2-normalized to this value.
trainable: Whether or not the embedding is trainable. Default is True.
use_safe_embedding_lookup: If true, uses safe_embedding_lookup_sparse
instead of embedding_lookup_sparse. safe_embedding_lookup_sparse ensures
there are no empty rows and all weights and ids are positive at the
expense of extra compute cost. This only applies to rank 2 (NxM) shaped
input tensors. Defaults to true, consider turning off if the above checks
are not needed. Note that having empty rows will not trigger any error
though the output result might be 0 or omitted.
Returns:
`DenseColumn` that converts from sparse input.
Raises:
ValueError: if `dimension` not > 0.
ValueError: if exactly one of `ckpt_to_load_from` and `tensor_name_in_ckpt`
is specified.
ValueError: if `initializer` is specified and is not callable.
RuntimeError: If eager execution is enabled."
3689,shared_embedding_columns,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,605,function,"List of dense columns that convert from sparse, categorical input.
This is similar to `embedding_column`, except that it produces a list of
embedding columns that share the same embedding weights.
Use this when your inputs are sparse and of the same type (e.g. watched and
impression video IDs that share the same vocabulary), and you want to convert
them to a dense representation (e.g., to feed to a DNN).
Inputs must be a list of categorical columns created by any of the
`categorical_column_*` function. They must all be of the same type and have
the same arguments except `key`. E.g. they can be
categorical_column_with_vocabulary_file with the same vocabulary_file. Some or
all columns could also be weighted_categorical_column.
Here is an example embedding of two features for a DNNClassifier model:
```python
watched_video_id = categorical_column_with_vocabulary_file(
'watched_video_id', video_vocabulary_file, video_vocabulary_size)
impression_video_id = categorical_column_with_vocabulary_file(
'impression_video_id', video_vocabulary_file, video_vocabulary_size)
columns = shared_embedding_columns(
[watched_video_id, impression_video_id], dimension=10)
estimator = tf.estimator.DNNClassifier(feature_columns=columns, ...)
label_column = ...
def input_fn():
features = tf.io.parse_example(
..., features=make_parse_example_spec(columns + [label_column]))
labels = features.pop(label_column.name)
return features, labels
estimator.train(input_fn=input_fn, steps=100)
```
Here is an example using `shared_embedding_columns` with model_fn:
```python
def model_fn(features, ...):
watched_video_id = categorical_column_with_vocabulary_file(
'watched_video_id', video_vocabulary_file, video_vocabulary_size)
impression_video_id = categorical_column_with_vocabulary_file(
'impression_video_id', video_vocabulary_file, video_vocabulary_size)
columns = shared_embedding_columns(
[watched_video_id, impression_video_id], dimension=10)
dense_tensor = input_layer(features, columns)
# Form DNN layers, calculate loss, and return EstimatorSpec.
...
```
Args:
categorical_columns: List of categorical columns created by a
`categorical_column_with_*` function. These columns produce the sparse IDs
that are inputs to the embedding lookup. All columns must be of the same
type and have the same arguments except `key`. E.g. they can be
categorical_column_with_vocabulary_file with the same vocabulary_file.
Some or all columns could also be weighted_categorical_column.
dimension: An integer specifying dimension of the embedding, must be > 0.
combiner: A string specifying how to reduce if there are multiple entries in
a single row. Currently 'mean', 'sqrtn' and 'sum' are supported, with
'mean' the default. 'sqrtn' often achieves good accuracy, in particular
with bag-of-words columns. Each of this can be thought as example level
normalizations on the column. For more information, see
`tf.embedding_lookup_sparse`.
initializer: A variable initializer function to be used in embedding
variable initialization. If not specified, defaults to
`truncated_normal_initializer` with mean `0.0` and
standard deviation `1/sqrt(dimension)`.
shared_embedding_collection_name: Optional name of the collection where
shared embedding weights are added. If not given, a reasonable name will
be chosen based on the names of `categorical_columns`. This is also used
in `variable_scope` when creating shared embedding weights.
ckpt_to_load_from: String representing checkpoint name/pattern from which to
restore column weights. Required if `tensor_name_in_ckpt` is not `None`.
tensor_name_in_ckpt: Name of the `Tensor` in `ckpt_to_load_from` from which
to restore the column weights. Required if `ckpt_to_load_from` is not
`None`.
max_norm: If not `None`, each embedding is clipped if its l2-norm is larger
than this value, before combining.
trainable: Whether or not the embedding is trainable. Default is True.
use_safe_embedding_lookup: If true, uses safe_embedding_lookup_sparse
instead of embedding_lookup_sparse. safe_embedding_lookup_sparse ensures
there are no empty rows and all weights and ids are positive at the
expense of extra compute cost. This only applies to rank 2 (NxM) shaped
input tensors. Defaults to true, consider turning off if the above checks
are not needed. Note that having empty rows will not trigger any error
though the output result might be 0 or omitted.
Returns:
A list of dense columns that converts from sparse input. The order of
results follows the ordering of `categorical_columns`.
Raises:
ValueError: if `dimension` not > 0.
ValueError: if any of the given `categorical_columns` is of different type
or has different arguments than the others.
ValueError: if exactly one of `ckpt_to_load_from` and `tensor_name_in_ckpt`
is specified.
ValueError: if `initializer` is specified and is not callable.
RuntimeError: if eager execution is enabled."
3690,shared_embedding_columns_v2,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,789,function,"List of dense columns that convert from sparse, categorical input.
This is similar to `embedding_column`, except that it produces a list of
embedding columns that share the same embedding weights.
Use this when your inputs are sparse and of the same type (e.g. watched and
impression video IDs that share the same vocabulary), and you want to convert
them to a dense representation (e.g., to feed to a DNN).
Inputs must be a list of categorical columns created by any of the
`categorical_column_*` function. They must all be of the same type and have
the same arguments except `key`. E.g. they can be
categorical_column_with_vocabulary_file with the same vocabulary_file. Some or
all columns could also be weighted_categorical_column.
Here is an example embedding of two features for a DNNClassifier model:
```python
watched_video_id = categorical_column_with_vocabulary_file(
'watched_video_id', video_vocabulary_file, video_vocabulary_size)
impression_video_id = categorical_column_with_vocabulary_file(
'impression_video_id', video_vocabulary_file, video_vocabulary_size)
columns = shared_embedding_columns(
[watched_video_id, impression_video_id], dimension=10)
estimator = tf.estimator.DNNClassifier(feature_columns=columns, ...)
label_column = ...
def input_fn():
features = tf.io.parse_example(
..., features=make_parse_example_spec(columns + [label_column]))
labels = features.pop(label_column.name)
return features, labels
estimator.train(input_fn=input_fn, steps=100)
```
Here is an example using `shared_embedding_columns` with model_fn:
```python
def model_fn(features, ...):
watched_video_id = categorical_column_with_vocabulary_file(
'watched_video_id', video_vocabulary_file, video_vocabulary_size)
impression_video_id = categorical_column_with_vocabulary_file(
'impression_video_id', video_vocabulary_file, video_vocabulary_size)
columns = shared_embedding_columns(
[watched_video_id, impression_video_id], dimension=10)
dense_tensor = input_layer(features, columns)
# Form DNN layers, calculate loss, and return EstimatorSpec.
...
```
Args:
categorical_columns: List of categorical columns created by a
`categorical_column_with_*` function. These columns produce the sparse IDs
that are inputs to the embedding lookup. All columns must be of the same
type and have the same arguments except `key`. E.g. they can be
categorical_column_with_vocabulary_file with the same vocabulary_file.
Some or all columns could also be weighted_categorical_column.
dimension: An integer specifying dimension of the embedding, must be > 0.
combiner: A string specifying how to reduce if there are multiple entries
in a single row. Currently 'mean', 'sqrtn' and 'sum' are supported, with
'mean' the default. 'sqrtn' often achieves good accuracy, in particular
with bag-of-words columns. Each of this can be thought as example level
normalizations on the column. For more information, see
`tf.embedding_lookup_sparse`.
initializer: A variable initializer function to be used in embedding
variable initialization. If not specified, defaults to
`truncated_normal_initializer` with mean `0.0` and standard
deviation `1/sqrt(dimension)`.
shared_embedding_collection_name: Optional collective name of these columns.
If not given, a reasonable name will be chosen based on the names of
`categorical_columns`.
ckpt_to_load_from: String representing checkpoint name/pattern from which to
restore column weights. Required if `tensor_name_in_ckpt` is not `None`.
tensor_name_in_ckpt: Name of the `Tensor` in `ckpt_to_load_from` from
which to restore the column weights. Required if `ckpt_to_load_from` is
not `None`.
max_norm: If not `None`, each embedding is clipped if its l2-norm is
larger than this value, before combining.
trainable: Whether or not the embedding is trainable. Default is True.
use_safe_embedding_lookup: If true, uses safe_embedding_lookup_sparse
instead of embedding_lookup_sparse. safe_embedding_lookup_sparse ensures
there are no empty rows and all weights and ids are positive at the
expense of extra compute cost. This only applies to rank 2 (NxM) shaped
input tensors. Defaults to true, consider turning off if the above checks
are not needed. Note that having empty rows will not trigger any error
though the output result might be 0 or omitted.
Returns:
A list of dense columns that converts from sparse input. The order of
results follows the ordering of `categorical_columns`.
Raises:
ValueError: if `dimension` not > 0.
ValueError: if any of the given `categorical_columns` is of different type
or has different arguments than the others.
ValueError: if exactly one of `ckpt_to_load_from` and `tensor_name_in_ckpt`
is specified.
ValueError: if `initializer` is specified and is not callable.
RuntimeError: if eager execution is enabled."
3691,numeric_column,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,964,function,"Represents real valued or numerical features.
Example:
```python
price = numeric_column('price')
columns = [price, ...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
dense_tensor = input_layer(features, columns)
# or
bucketized_price = bucketized_column(price, boundaries=[...])
columns = [bucketized_price, ...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
linear_prediction = linear_model(features, columns)
```
Args:
key: A unique string identifying the input feature. It is used as the
column name and the dictionary key for feature parsing configs, feature
`Tensor` objects, and feature columns.
shape: An iterable of integers specifies the shape of the `Tensor`. An
integer can be given which means a single dimension `Tensor` with given
width. The `Tensor` representing the column will have the shape of
[batch_size] + `shape`.
default_value: A single value compatible with `dtype` or an iterable of
values compatible with `dtype` which the column takes on during
`tf.Example` parsing if data is missing. A default value of `None` will
cause `tf.io.parse_example` to fail if an example does not contain this
column. If a single value is provided, the same value will be applied as
the default value for every item. If an iterable of values is provided,
the shape of the `default_value` should be equal to the given `shape`.
dtype: defines the type of values. Default value is `tf.float32`. Must be a
non-quantized, real integer or floating point type.
normalizer_fn: If not `None`, a function that can be used to normalize the
value of the tensor after `default_value` is applied for parsing.
Normalizer function takes the input `Tensor` as its argument, and returns
the output `Tensor`. (e.g. lambda x: (x - 3.0) / 4.2). Please note that
even though the most common use case of this function is normalization, it
can be used for any kind of Tensorflow transformations.
Returns:
A `NumericColumn`.
Raises:
TypeError: if any dimension in shape is not an int
ValueError: if any dimension in shape is not a positive integer
TypeError: if `default_value` is an iterable but not compatible with `shape`
TypeError: if `default_value` is not compatible with `dtype`.
ValueError: if `dtype` is not convertible to `tf.float32`."
3692,bucketized_column,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,1041,function,"Represents discretized dense input bucketed by `boundaries`.
Buckets include the left boundary, and exclude the right boundary. Namely,
`boundaries=[0., 1., 2.]` generates buckets `(-inf, 0.)`, `[0., 1.)`,
`[1., 2.)`, and `[2., +inf)`.
For example, if the inputs are
```python
boundaries = [0, 10, 100]
input tensor = [[-5, 10000]
[150, 10]
[5, 100]]
```
then the output will be
```python
output = [[0, 3]
[3, 2]
[1, 3]]
```
Example:
```python
price = tf.feature_column.numeric_column('price')
bucketized_price = tf.feature_column.bucketized_column(
price, boundaries=[...])
columns = [bucketized_price, ...]
features = tf.io.parse_example(
..., features=tf.feature_column.make_parse_example_spec(columns))
dense_tensor = tf.keras.layers.DenseFeatures(columns)(features)
```
A `bucketized_column` can also be crossed with another categorical column
using `crossed_column`:
```python
price = tf.feature_column.numeric_column('price')
# bucketized_column converts numerical feature to a categorical one.
bucketized_price = tf.feature_column.bucketized_column(
price, boundaries=[...])
# 'keywords' is a string feature.
price_x_keywords = tf.feature_column.crossed_column(
[bucketized_price, 'keywords'], 50K)
columns = [price_x_keywords, ...]
features = tf.io.parse_example(
..., features=tf.feature_column.make_parse_example_spec(columns))
dense_tensor = tf.keras.layers.DenseFeatures(columns)(features)
linear_model = tf.keras.experimental.LinearModel(units=...)(dense_tensor)
```
Args:
source_column: A one-dimensional dense column which is generated with
`numeric_column`.
boundaries: A sorted list or tuple of floats specifying the boundaries.
Returns:
A `BucketizedColumn`.
Raises:
ValueError: If `source_column` is not a numeric column, or if it is not
one-dimensional.
ValueError: If `boundaries` is not a sorted list or tuple."
3693,categorical_column_with_hash_bucket,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,1127,function,"Represents sparse feature where ids are set by hashing.
Use this when your sparse features are in string or integer format, and you
want to distribute your inputs into a finite number of buckets by hashing.
output_id = Hash(input_feature_string) % bucket_size for string type input.
For int type input, the value is converted to its string representation first
and then hashed by the same formula.
For input dictionary `features`, `features[key]` is either `Tensor` or
`SparseTensor`. If `Tensor`, missing values can be represented by `-1` for int
and `''` for string, which will be dropped by this feature column.
Example:
```python
keywords = categorical_column_with_hash_bucket(""keywords"", 10K)
columns = [keywords, ...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
linear_prediction = linear_model(features, columns)
# or
keywords_embedded = embedding_column(keywords, 16)
columns = [keywords_embedded, ...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
dense_tensor = input_layer(features, columns)
```
Args:
key: A unique string identifying the input feature. It is used as the
column name and the dictionary key for feature parsing configs, feature
`Tensor` objects, and feature columns.
hash_bucket_size: An int > 1. The number of buckets.
dtype: The type of features. Only string and integer types are supported.
Returns:
A `HashedCategoricalColumn`.
Raises:
ValueError: `hash_bucket_size` is not greater than 1.
ValueError: `dtype` is neither string nor integer."
3694,categorical_column_with_vocabulary_file,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,1186,function,"A `CategoricalColumn` with a vocabulary file.
Use this when your inputs are in string or integer format, and you have a
vocabulary file that maps each value to an integer ID. By default,
out-of-vocabulary values are ignored. Use either (but not both) of
`num_oov_buckets` and `default_value` to specify how to include
out-of-vocabulary values.
For input dictionary `features`, `features[key]` is either `Tensor` or
`SparseTensor`. If `Tensor`, missing values can be represented by `-1` for int
and `''` for string, which will be dropped by this feature column.
Example with `num_oov_buckets`:
File '/us/states.txt' contains 50 lines, each with a 2-character U.S. state
abbreviation. All inputs with values in that file are assigned an ID 0-49,
corresponding to its line number. All other values are hashed and assigned an
ID 50-54.
```python
states = categorical_column_with_vocabulary_file(
key='states', vocabulary_file='/us/states.txt', vocabulary_size=50,
num_oov_buckets=5)
columns = [states, ...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
linear_prediction = linear_model(features, columns)
```
Example with `default_value`:
File '/us/states.txt' contains 51 lines - the first line is 'XX', and the
other 50 each have a 2-character U.S. state abbreviation. Both a literal 'XX'
in input, and other values missing from the file, will be assigned ID 0. All
others are assigned the corresponding line number 1-50.
```python
states = categorical_column_with_vocabulary_file(
key='states', vocabulary_file='/us/states.txt', vocabulary_size=51,
default_value=0)
columns = [states, ...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
linear_prediction, _, _ = linear_model(features, columns)
```
And to make an embedding with either:
```python
columns = [embedding_column(states, 3),...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
dense_tensor = input_layer(features, columns)
```
Args:
key: A unique string identifying the input feature. It is used as the
column name and the dictionary key for feature parsing configs, feature
`Tensor` objects, and feature columns.
vocabulary_file: The vocabulary file name.
vocabulary_size: Number of the elements in the vocabulary. This must be no
greater than length of `vocabulary_file`, if less than length, later
values are ignored. If None, it is set to the length of `vocabulary_file`.
num_oov_buckets: Non-negative integer, the number of out-of-vocabulary
buckets. All out-of-vocabulary inputs will be assigned IDs in the range
`[vocabulary_size, vocabulary_size+num_oov_buckets)` based on a hash of
the input value. A positive `num_oov_buckets` can not be specified with
`default_value`.
default_value: The integer ID value to return for out-of-vocabulary feature
values, defaults to `-1`. This can not be specified with a positive
`num_oov_buckets`.
dtype: The type of features. Only string and integer types are supported.
Returns:
A `CategoricalColumn` with a vocabulary file.
Raises:
ValueError: `vocabulary_file` is missing or cannot be opened.
ValueError: `vocabulary_size` is missing or < 1.
ValueError: `num_oov_buckets` is a negative integer.
ValueError: `num_oov_buckets` and `default_value` are both specified.
ValueError: `dtype` is neither string nor integer."
3695,categorical_column_with_vocabulary_file_v2,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,1277,function,"A `CategoricalColumn` with a vocabulary file.
Use this when your inputs are in string or integer format, and you have a
vocabulary file that maps each value to an integer ID. By default,
out-of-vocabulary values are ignored. Use either (but not both) of
`num_oov_buckets` and `default_value` to specify how to include
out-of-vocabulary values.
For input dictionary `features`, `features[key]` is either `Tensor` or
`SparseTensor`. If `Tensor`, missing values can be represented by `-1` for int
and `''` for string, which will be dropped by this feature column.
Example with `num_oov_buckets`:
File `'/us/states.txt'` contains 50 lines, each with a 2-character U.S. state
abbreviation. All inputs with values in that file are assigned an ID 0-49,
corresponding to its line number. All other values are hashed and assigned an
ID 50-54.
```python
states = categorical_column_with_vocabulary_file(
key='states', vocabulary_file='/us/states.txt', vocabulary_size=50,
num_oov_buckets=5)
columns = [states, ...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
linear_prediction = linear_model(features, columns)
```
Example with `default_value`:
File `'/us/states.txt'` contains 51 lines - the first line is `'XX'`, and the
other 50 each have a 2-character U.S. state abbreviation. Both a literal
`'XX'` in input, and other values missing from the file, will be assigned
ID 0. All others are assigned the corresponding line number 1-50.
```python
states = categorical_column_with_vocabulary_file(
key='states', vocabulary_file='/us/states.txt', vocabulary_size=51,
default_value=0)
columns = [states, ...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
linear_prediction, _, _ = linear_model(features, columns)
```
And to make an embedding with either:
```python
columns = [embedding_column(states, 3),...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
dense_tensor = input_layer(features, columns)
```
Args:
key: A unique string identifying the input feature. It is used as the
column name and the dictionary key for feature parsing configs, feature
`Tensor` objects, and feature columns.
vocabulary_file: The vocabulary file name.
vocabulary_size: Number of the elements in the vocabulary. This must be no
greater than length of `vocabulary_file`, if less than length, later
values are ignored. If None, it is set to the length of `vocabulary_file`.
dtype: The type of features. Only string and integer types are supported.
default_value: The integer ID value to return for out-of-vocabulary feature
values, defaults to `-1`. This can not be specified with a positive
`num_oov_buckets`.
num_oov_buckets: Non-negative integer, the number of out-of-vocabulary
buckets. All out-of-vocabulary inputs will be assigned IDs in the range
`[vocabulary_size, vocabulary_size+num_oov_buckets)` based on a hash of
the input value. A positive `num_oov_buckets` can not be specified with
`default_value`.
Returns:
A `CategoricalColumn` with a vocabulary file.
Raises:
ValueError: `vocabulary_file` is missing or cannot be opened.
ValueError: `vocabulary_size` is missing or < 1.
ValueError: `num_oov_buckets` is a negative integer.
ValueError: `num_oov_buckets` and `default_value` are both specified.
ValueError: `dtype` is neither string nor integer."
3696,categorical_column_with_vocabulary_list,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,1397,function,"A `CategoricalColumn` with in-memory vocabulary.
Use this when your inputs are in string or integer format, and you have an
in-memory vocabulary mapping each value to an integer ID. By default,
out-of-vocabulary values are ignored. Use either (but not both) of
`num_oov_buckets` and `default_value` to specify how to include
out-of-vocabulary values.
For input dictionary `features`, `features[key]` is either `Tensor` or
`SparseTensor`. If `Tensor`, missing values can be represented by `-1` for int
and `''` for string, which will be dropped by this feature column.
Example with `num_oov_buckets`:
In the following example, each input in `vocabulary_list` is assigned an ID
0-3 corresponding to its index (e.g., input 'B' produces output 2). All other
inputs are hashed and assigned an ID 4-5.
```python
colors = categorical_column_with_vocabulary_list(
key='colors', vocabulary_list=('R', 'G', 'B', 'Y'),
num_oov_buckets=2)
columns = [colors, ...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
linear_prediction, _, _ = linear_model(features, columns)
```
Example with `default_value`:
In the following example, each input in `vocabulary_list` is assigned an ID
0-4 corresponding to its index (e.g., input 'B' produces output 3). All other
inputs are assigned `default_value` 0.
```python
colors = categorical_column_with_vocabulary_list(
key='colors', vocabulary_list=('X', 'R', 'G', 'B', 'Y'), default_value=0)
columns = [colors, ...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
linear_prediction, _, _ = linear_model(features, columns)
```
And to make an embedding with either:
```python
columns = [embedding_column(colors, 3),...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
dense_tensor = input_layer(features, columns)
```
Args:
key: A unique string identifying the input feature. It is used as the column
name and the dictionary key for feature parsing configs, feature `Tensor`
objects, and feature columns.
vocabulary_list: An ordered iterable defining the vocabulary. Each feature
is mapped to the index of its value (if present) in `vocabulary_list`.
Must be castable to `dtype`.
dtype: The type of features. Only string and integer types are supported. If
`None`, it will be inferred from `vocabulary_list`.
default_value: The integer ID value to return for out-of-vocabulary feature
values, defaults to `-1`. This can not be specified with a positive
`num_oov_buckets`.
num_oov_buckets: Non-negative integer, the number of out-of-vocabulary
buckets. All out-of-vocabulary inputs will be assigned IDs in the range
`[len(vocabulary_list), len(vocabulary_list)+num_oov_buckets)` based on a
hash of the input value. A positive `num_oov_buckets` can not be specified
with `default_value`.
Returns:
A `CategoricalColumn` with in-memory vocabulary.
Raises:
ValueError: if `vocabulary_list` is empty, or contains duplicate keys.
ValueError: `num_oov_buckets` is a negative integer.
ValueError: `num_oov_buckets` and `default_value` are both specified.
ValueError: if `dtype` is not integer or string."
3697,categorical_column_with_identity,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,1514,function,"A `CategoricalColumn` that returns identity values.
Use this when your inputs are integers in the range `[0, num_buckets)`, and
you want to use the input value itself as the categorical ID. Values outside
this range will result in `default_value` if specified, otherwise it will
fail.
Typically, this is used for contiguous ranges of integer indexes, but
it doesn't have to be. This might be inefficient, however, if many of IDs
are unused. Consider `categorical_column_with_hash_bucket` in that case.
For input dictionary `features`, `features[key]` is either `Tensor` or
`SparseTensor`. If `Tensor`, missing values can be represented by `-1` for int
and `''` for string, which will be dropped by this feature column.
In the following examples, each input in the range `[0, 1000000)` is assigned
the same value. All other inputs are assigned `default_value` 0. Note that a
literal 0 in inputs will result in the same default ID.
Linear model:
```python
video_id = categorical_column_with_identity(
key='video_id', num_buckets=1000000, default_value=0)
columns = [video_id, ...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
linear_prediction, _, _ = linear_model(features, columns)
```
Embedding for a DNN model:
```python
columns = [embedding_column(video_id, 9),...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
dense_tensor = input_layer(features, columns)
```
Args:
key: A unique string identifying the input feature. It is used as the
column name and the dictionary key for feature parsing configs, feature
`Tensor` objects, and feature columns.
num_buckets: Range of inputs and outputs is `[0, num_buckets)`.
default_value: If set, values outside of range `[0, num_buckets)` will
be replaced with this value. If not set, values >= num_buckets will
cause a failure while values < 0 will be dropped.
Returns:
A `CategoricalColumn` that returns identity values.
Raises:
ValueError: if `num_buckets` is less than one.
ValueError: if `default_value` is not in range `[0, num_buckets)`."
3698,indicator_column,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,1582,function,"Represents multi-hot representation of given categorical column.
- For DNN model, `indicator_column` can be used to wrap any
`categorical_column_*` (e.g., to feed to DNN). Consider to Use
`embedding_column` if the number of buckets/unique(values) are large.
- For Wide (aka linear) model, `indicator_column` is the internal
representation for categorical column when passing categorical column
directly (as any element in feature_columns) to `linear_model`. See
`linear_model` for details.
```python
name = indicator_column(categorical_column_with_vocabulary_list(
'name', ['bob', 'george', 'wanda']))
columns = [name, ...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
dense_tensor = input_layer(features, columns)
dense_tensor == [[1, 0, 0]] # If ""name"" bytes_list is [""bob""]
dense_tensor == [[1, 0, 1]] # If ""name"" bytes_list is [""bob"", ""wanda""]
dense_tensor == [[2, 0, 0]] # If ""name"" bytes_list is [""bob"", ""bob""]
```
Args:
categorical_column: A `CategoricalColumn` which is created by
`categorical_column_with_*` or `crossed_column` functions.
Returns:
An `IndicatorColumn`.
Raises:
ValueError: If `categorical_column` is not CategoricalColumn type."
3699,weighted_categorical_column,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,1625,function,"Applies weight values to a `CategoricalColumn`.
Use this when each of your sparse inputs has both an ID and a value. For
example, if you're representing text documents as a collection of word
frequencies, you can provide 2 parallel sparse input features ('terms' and
'frequencies' below).
Example:
Input `tf.Example` objects:
```proto
[
features {
feature {
key: ""terms""
value {bytes_list {value: ""very"" value: ""model""}}
}
feature {
key: ""frequencies""
value {float_list {value: 0.3 value: 0.1}}
}
},
features {
feature {
key: ""terms""
value {bytes_list {value: ""when"" value: ""course"" value: ""human""}}
}
feature {
key: ""frequencies""
value {float_list {value: 0.4 value: 0.1 value: 0.2}}
}
}
]
```
```python
categorical_column = categorical_column_with_hash_bucket(
column_name='terms', hash_bucket_size=1000)
weighted_column = weighted_categorical_column(
categorical_column=categorical_column, weight_feature_key='frequencies')
columns = [weighted_column, ...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
linear_prediction, _, _ = linear_model(features, columns)
```
This assumes the input dictionary contains a `SparseTensor` for key
'terms', and a `SparseTensor` for key 'frequencies'. These 2 tensors must have
the same indices and dense shape.
Args:
categorical_column: A `CategoricalColumn` created by
`categorical_column_with_*` functions.
weight_feature_key: String key for weight values.
dtype: Type of weights, such as `tf.float32`. Only float and integer weights
are supported.
Returns:
A `CategoricalColumn` composed of two sparse features: one represents id,
the other represents weight (value) of the id feature in that example.
Raises:
ValueError: if `dtype` is not convertible to float."
3700,crossed_column,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,1701,function,"Returns a column for performing crosses of categorical features.
Crossed features will be hashed according to `hash_bucket_size`. Conceptually,
the transformation can be thought of as:
Hash(cartesian product of features) % `hash_bucket_size`
For example, if the input features are:
* SparseTensor referred by first key:
```python
shape = [2, 2]
{
[0, 0]: ""a""
[1, 0]: ""b""
[1, 1]: ""c""
}
```
* SparseTensor referred by second key:
```python
shape = [2, 1]
{
[0, 0]: ""d""
[1, 0]: ""e""
}
```
then crossed feature will look like:
```python
shape = [2, 2]
{
[0, 0]: Hash64(""d"", Hash64(""a"")) % hash_bucket_size
[1, 0]: Hash64(""e"", Hash64(""b"")) % hash_bucket_size
[1, 1]: Hash64(""e"", Hash64(""c"")) % hash_bucket_size
}
```
Here is an example to create a linear model with crosses of string features:
```python
keywords_x_doc_terms = crossed_column(['keywords', 'doc_terms'], 50K)
columns = [keywords_x_doc_terms, ...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
linear_prediction = linear_model(features, columns)
```
You could also use vocabulary lookup before crossing:
```python
keywords = categorical_column_with_vocabulary_file(
'keywords', '/path/to/vocabulary/file', vocabulary_size=1K)
keywords_x_doc_terms = crossed_column([keywords, 'doc_terms'], 50K)
columns = [keywords_x_doc_terms, ...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
linear_prediction = linear_model(features, columns)
```
If an input feature is of numeric type, you can use
`categorical_column_with_identity`, or `bucketized_column`, as in the example:
```python
# vertical_id is an integer categorical feature.
vertical_id = categorical_column_with_identity('vertical_id', 10K)
price = numeric_column('price')
# bucketized_column converts numerical feature to a categorical one.
bucketized_price = bucketized_column(price, boundaries=[...])
vertical_id_x_price = crossed_column([vertical_id, bucketized_price], 50K)
columns = [vertical_id_x_price, ...]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
linear_prediction = linear_model(features, columns)
```
To use crossed column in DNN model, you need to add it in an embedding column
as in this example:
```python
vertical_id_x_price = crossed_column([vertical_id, bucketized_price], 50K)
vertical_id_x_price_embedded = embedding_column(vertical_id_x_price, 10)
dense_tensor = input_layer(features, [vertical_id_x_price_embedded, ...])
```
Args:
keys: An iterable identifying the features to be crossed. Each element can
be either:
* string: Will use the corresponding feature which must be of string type.
* `CategoricalColumn`: Will use the transformed tensor produced by this
column. Does not support hashed categorical column.
hash_bucket_size: An int > 1. The number of buckets.
hash_key: Specify the hash_key that will be used by the `FingerprintCat64`
function to combine the crosses fingerprints on SparseCrossOp (optional).
Returns:
A `CrossedColumn`.
Raises:
ValueError: If `len(keys) < 2`.
ValueError: If any of the keys is neither a string nor `CategoricalColumn`.
ValueError: If any of the keys is `HashedCategoricalColumn`.
ValueError: If `hash_bucket_size < 1`."
3701,FeatureColumn,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,1829,class,"Represents a feature column abstraction.
WARNING: Do not subclass this layer unless you know what you are doing:
the API is subject to future changes.
To distinguish between the concept of a feature family and a specific binary
feature within a family, we refer to a feature family like ""country"" as a
feature column. For example, we can have a feature in a `tf.Example` format:
{key: ""country"", value: [ ""US"" ]}
In this example the value of feature is ""US"" and ""country"" refers to the
column of the feature.
This class is an abstract class. Users should not create instances of this."
3702,name,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,1846,method,Returns string. Used for naming.
3703,transform_feature,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,1913,method,"Returns intermediate representation (usually a `Tensor`).
Uses `transformation_cache` to create an intermediate representation
(usually a `Tensor`) that other feature columns can use.
Example usage of `transformation_cache`:
Let's say a Feature column depends on raw feature ('raw') and another
`FeatureColumn` (input_fc). To access corresponding `Tensor`s,
transformation_cache will be used as follows:
```python
raw_tensor = transformation_cache.get('raw', state_manager)
fc_tensor = transformation_cache.get(input_fc, state_manager)
```
Args:
transformation_cache: A `FeatureTransformationCache` object to access
features.
state_manager: A `StateManager` to create / access resources such as
lookup tables.
Returns:
Transformed feature `Tensor`."
3704,parse_example_spec,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,1941,method,"Returns a `tf.Example` parsing spec as dict.
It is used for get_parsing_spec for `tf.io.parse_example`. Returned spec is
a dict from keys ('string') to `VarLenFeature`, `FixedLenFeature`, and other
supported objects. Please check documentation of `tf.io.parse_example` for
all supported spec objects.
Let's say a Feature column depends on raw feature ('raw') and another
`FeatureColumn` (input_fc). One possible implementation of
parse_example_spec is as follows:
```python
spec = {'raw': tf.io.FixedLenFeature(...)}
spec.update(input_fc.parse_example_spec)
return spec
```"
3705,create_state,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,1961,method,"Uses the `state_manager` to create state for the FeatureColumn.
Args:
state_manager: A `StateManager` to create / access resources such as
lookup tables and variables."
3706,parents,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,1981,method,"Returns a list of immediate raw feature and FeatureColumn dependencies.
For example:
# For the following feature columns
a = numeric_column('f1')
c = crossed_column(a, 'f2')
# The expected parents are:
a.parents = ['f1']
c.parents = [a, 'f2']"
3707,get_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,1994,method,"Returns the config of the feature column.
A FeatureColumn config is a Python dictionary (serializable) containing the
configuration of a FeatureColumn. The same FeatureColumn can be
reinstantiated later from this configuration.
The config of a feature column does not include information about feature
columns depending on it nor the FeatureColumn class name.
Example with (de)serialization practices followed in this file:
```python
class SerializationExampleFeatureColumn(
FeatureColumn, collections.namedtuple(
'SerializationExampleFeatureColumn',
('dimension', 'parent', 'dtype', 'normalizer_fn'))):
def get_config(self):
# Create a dict from the namedtuple.
# Python attribute literals can be directly copied from / to the config.
# For example 'dimension', assuming it is an integer literal.
config = dict(zip(self._fields, self))
# (De)serialization of parent FeatureColumns should use the provided
# (de)serialize_feature_column() methods that take care of de-duping.
config['parent'] = serialize_feature_column(self.parent)
# Many objects provide custom (de)serialization e.g: for tf.DType
# tf.DType.name, tf.as_dtype() can be used.
config['dtype'] = self.dtype.name
# Non-trivial dependencies should be Keras-(de)serializable.
config['normalizer_fn'] = generic_utils.serialize_keras_object(
self.normalizer_fn)
return config
@classmethod
def from_config(cls, config, custom_objects=None, columns_by_name=None):
# This should do the inverse transform from `get_config` and construct
# the namedtuple.
kwargs = config.copy()
kwargs['parent'] = deserialize_feature_column(
config['parent'], custom_objects, columns_by_name)
kwargs['dtype'] = dtypes.as_dtype(config['dtype'])
kwargs['normalizer_fn'] = generic_utils.deserialize_keras_object(
config['normalizer_fn'], custom_objects=custom_objects)
return cls(**kwargs)
```
Returns:
A serializable Dict that can be used to deserialize the object with
from_config."
3708,from_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2054,method,"Creates a FeatureColumn from its config.
This method should be the reverse of `get_config`, capable of instantiating
the same FeatureColumn from the config dictionary. See `get_config` for an
example of common (de)serialization practices followed in this file.
TODO(b/118939620): This is a private method until consensus is reached on
supporting object deserialization deduping within Keras.
Args:
config: A Dict config acquired with `get_config`.
custom_objects: Optional dictionary mapping names (strings) to custom
classes or functions to be considered during deserialization.
columns_by_name: A Dict[String, FeatureColumn] of existing columns in
order to avoid duplication. Should be passed to any calls to
deserialize_feature_column().
Returns:
A FeatureColumn for the input config."
3709,DenseColumn,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2082,class,"Represents a column which can be represented as `Tensor`.
Some examples of this type are: numeric_column, embedding_column,
indicator_column."
3710,variable_shape,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2090,method,"`TensorShape` of `get_dense_tensor`, without batch dimension."
3711,get_dense_tensor,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2095,method,"Returns a `Tensor`.
The output of this function will be used by model-builder-functions. For
example the pseudo code of `input_layer` will be like:
```python
def input_layer(features, feature_columns, ...):
outputs = [fc.get_dense_tensor(...) for fc in feature_columns]
return tf.concat(outputs)
```
Args:
transformation_cache: A `FeatureTransformationCache` object to access
features.
state_manager: A `StateManager` to create / access resources such as
lookup tables.
Returns:
`Tensor` of shape [batch_size] + `variable_shape`."
3712,is_feature_column_v2,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2119,function,Returns True if all feature columns are V2.
3713,CategoricalColumn,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2157,class,"Represents a categorical feature.
A categorical feature typically handled with a `tf.sparse.SparseTensor` of
IDs."
3714,num_buckets,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2168,method,Returns number of buckets in this sparse feature.
3715,get_sparse_tensors,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2173,method,"Returns an IdWeightPair.
`IdWeightPair` is a pair of `SparseTensor`s which represents ids and
weights.
`IdWeightPair.id_tensor` is typically a `batch_size` x `num_buckets`
`SparseTensor` of `int64`. `IdWeightPair.weight_tensor` is either a
`SparseTensor` of `float` or `None` to indicate all weights should be
taken to be 1. If specified, `weight_tensor` must have exactly the same
shape and indices as `sp_ids`. Expected `SparseTensor` is same as parsing
output of a `VarLenFeature` which is a ragged matrix.
Args:
transformation_cache: A `FeatureTransformationCache` object to access
features.
state_manager: A `StateManager` to create / access resources such as
lookup tables."
3716,SequenceDenseColumn,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2243,class,Represents dense sequence data.
3717,get_sequence_dense_tensor,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2250,method,"Returns a `TensorSequenceLengthPair`.
Args:
transformation_cache: A `FeatureTransformationCache` object to access
features.
state_manager: A `StateManager` to create / access resources such as
lookup tables."
3718,FeatureTransformationCache,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2262,class,"Handles caching of transformations while building the model.
`FeatureColumn` specifies how to digest an input column to the network. Some
feature columns require data transformations. This class caches those
transformations.
Some features may be used in more than one place. For example, one can use a
bucketized feature by itself and a cross with it. In that case we
should create only one bucketization op instead of creating ops for each
feature column separately. To handle re-use of transformed columns,
`FeatureTransformationCache` caches all previously transformed columns.
Example:
We're trying to use the following `FeatureColumn`s:
```python
bucketized_age = fc.bucketized_column(fc.numeric_column(""age""), ...)
keywords = fc.categorical_column_with_hash_buckets(""keywords"", ...)
age_X_keywords = fc.crossed_column([bucketized_age, ""keywords""])
... = linear_model(features,
[bucketized_age, keywords, age_X_keywords]
```
If we transform each column independently, then we'll get duplication of
bucketization (one for cross, one for bucketization itself).
The `FeatureTransformationCache` eliminates this duplication."
3719,get,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2305,method,"Returns a `Tensor` for the given key.
A `str` key is used to access a base feature (not-transformed). When a
`FeatureColumn` is passed, the transformed feature is returned if it
already exists, otherwise the given `FeatureColumn` is asked to provide its
transformed output, which is then cached.
Args:
key: a `str` or a `FeatureColumn`.
state_manager: A StateManager object that holds the FeatureColumn state.
training: Boolean indicating whether to the column is being used in
training mode. This argument is passed to the transform_feature method
of any `FeatureColumn` that takes a `training` argument. For example, if
a `FeatureColumn` performed dropout, it could expose a `training`
argument to control whether the dropout should be applied.
Returns:
The transformed `Tensor` corresponding to the `key`.
Raises:
ValueError: if key is not found or a transformed `Tensor` cannot be
computed."
3720,expand_dims,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2383,method,
3721,NumericColumn,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2501,class,see `numeric_column`.
3722,name,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2514,method,See `FeatureColumn` base class.
3723,parse_example_spec,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2519,method,See `FeatureColumn` base class.
3724,transform_feature,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2548,method,"See `FeatureColumn` base class.
In this case, we apply the `normalizer_fn` to the input tensor.
Args:
transformation_cache: A `FeatureTransformationCache` object to access
features.
state_manager: A `StateManager` to create / access resources such as
lookup tables.
Returns:
Normalized input tensor.
Raises:
ValueError: If a SparseTensor is passed in."
3725,variable_shape,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2568,method,See `DenseColumn` base class.
3726,get_dense_tensor,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2578,method,"Returns dense `Tensor` representing numeric feature.
Args:
transformation_cache: A `FeatureTransformationCache` object to access
features.
state_manager: A `StateManager` to create / access resources such as
lookup tables.
Returns:
Dense `Tensor` created within `transform_feature`."
3727,parents,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2602,method,See 'FeatureColumn` base class.
3728,get_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2606,method,See 'FeatureColumn` base class.
3729,from_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2616,method,See 'FeatureColumn` base class.
3730,BucketizedColumn,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2628,class,See `bucketized_column`.
3731,name,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2643,method,See `FeatureColumn` base class.
3732,parse_example_spec,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2648,method,See `FeatureColumn` base class.
3733,transform_feature,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2667,method,Returns bucketized categorical `source_column` tensor.
3734,variable_shape,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2675,method,See `DenseColumn` base class.
3735,get_dense_tensor,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2693,method,Returns one hot encoded dense `Tensor`.
3736,num_buckets,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2707,method,See `CategoricalColumn` base class.
3737,get_sparse_tensors,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2745,method,Converts dense inputs to SparseTensor so downstream code can use it.
3738,parents,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2761,method,See 'FeatureColumn` base class.
3739,get_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2765,method,See 'FeatureColumn` base class.
3740,from_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2773,method,See 'FeatureColumn` base class.
3741,EmbeddingColumn,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2783,class,See `embedding_column`.
3742,name,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2823,method,See `FeatureColumn` base class.
3743,parse_example_spec,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2828,method,See `FeatureColumn` base class.
3744,transform_feature,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2838,method,Transforms underlying `categorical_column`.
3745,variable_shape,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2848,method,See `DenseColumn` base class.
3746,create_state,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2858,method,Creates the embedding lookup variable.
3747,get_dense_tensor,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2927,method,"Returns tensor after doing the embedding lookup.
Args:
transformation_cache: A `FeatureTransformationCache` object to access
features.
state_manager: A `StateManager` to create / access resources such as
lookup tables.
Returns:
Embedding lookup tensor.
Raises:
ValueError: `categorical_column` is SequenceCategoricalColumn."
3748,get_sequence_dense_tensor,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,2977,method,See `SequenceDenseColumn` base class.
3749,parents,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3023,method,See 'FeatureColumn` base class.
3750,get_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3027,method,See 'FeatureColumn` base class.
3751,from_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3038,method,See 'FeatureColumn` base class.
3752,SharedEmbeddingColumnCreator,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3061,class,
3753,embedding_weights,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3088,method,
3754,dimension,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3109,method,
3755,SharedEmbeddingColumn,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3113,class,See `embedding_column`.
3756,name,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3143,method,See `FeatureColumn` base class.
3757,parse_example_spec,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3148,method,See `FeatureColumn` base class.
3758,transform_feature,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3156,method,See `FeatureColumn` base class.
3759,variable_shape,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3164,method,See `DenseColumn` base class.
3760,get_dense_tensor,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3202,method,Returns the embedding lookup result.
3761,get_sequence_dense_tensor,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3219,method,See `SequenceDenseColumn` base class.
3762,parents,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3245,method,See 'FeatureColumn` base class.
3763,HashedCategoricalColumn,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3266,class,see `categorical_column_with_hash_bucket`.
3764,name,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3278,method,See `FeatureColumn` base class.
3765,parse_example_spec,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3283,method,See `FeatureColumn` base class.
3766,transform_feature,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3318,method,Hashes the values in the feature_column.
3767,num_buckets,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3331,method,Returns number of buckets in this sparse feature.
3768,get_sparse_tensors,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3341,method,See `CategoricalColumn` base class.
3769,parents,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3355,method,See 'FeatureColumn` base class.
3770,get_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3359,method,See 'FeatureColumn` base class.
3771,from_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3366,method,See 'FeatureColumn` base class.
3772,VocabularyFileCategoricalColumn,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3374,class,See `categorical_column_with_vocabulary_file`.
3773,name,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3387,method,See `FeatureColumn` base class.
3774,parse_example_spec,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3392,method,See `FeatureColumn` base class.
3775,transform_feature,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3437,method,Creates a lookup table for the vocabulary.
3776,num_buckets,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3450,method,Returns number of buckets in this sparse feature.
3777,get_sparse_tensors,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3460,method,See `CategoricalColumn` base class.
3778,parents,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3474,method,See 'FeatureColumn` base class.
3779,get_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3478,method,See 'FeatureColumn` base class.
3780,from_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3485,method,See 'FeatureColumn` base class.
3781,VocabularyListCategoricalColumn,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3493,class,See `categorical_column_with_vocabulary_list`.
3782,name,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3507,method,See `FeatureColumn` base class.
3783,parse_example_spec,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3512,method,See `FeatureColumn` base class.
3784,transform_feature,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3556,method,Creates a lookup table for the vocabulary list.
3785,num_buckets,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3569,method,Returns number of buckets in this sparse feature.
3786,get_sparse_tensors,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3579,method,See `CategoricalColumn` base class.
3787,parents,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3593,method,See 'FeatureColumn` base class.
3788,get_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3597,method,See 'FeatureColumn` base class.
3789,from_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3604,method,See 'FeatureColumn` base class.
3790,IdentityCategoricalColumn,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3612,class,See `categorical_column_with_identity`.
3791,name,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3625,method,See `FeatureColumn` base class.
3792,parse_example_spec,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3630,method,See `FeatureColumn` base class.
3793,transform_feature,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3668,method,Returns a SparseTensor with identity values.
3794,num_buckets,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3681,method,Returns number of buckets in this sparse feature.
3795,get_sparse_tensors,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3691,method,See `CategoricalColumn` base class.
3796,parents,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3705,method,See 'FeatureColumn` base class.
3797,get_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3709,method,See 'FeatureColumn` base class.
3798,from_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3714,method,See 'FeatureColumn` base class.
3799,WeightedCategoricalColumn,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3721,class,See `weighted_categorical_column`.
3800,name,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3735,method,See `FeatureColumn` base class.
3801,parse_example_spec,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3741,method,See `FeatureColumn` base class.
3802,num_buckets,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3762,method,See `DenseColumn` base class.
3803,transform_feature,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3788,method,Applies weights to tensor generated from `categorical_column`'.
3804,get_sparse_tensors,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3805,method,See `CategoricalColumn` base class.
3805,parents,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3820,method,See 'FeatureColumn` base class.
3806,get_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3824,method,See 'FeatureColumn` base class.
3807,from_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3834,method,See 'FeatureColumn` base class.
3808,CrossedColumn,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3845,class,See `crossed_column`.
3809,name,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3864,method,See `FeatureColumn` base class.
3810,parse_example_spec,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3875,method,See `FeatureColumn` base class.
3811,transform_feature,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3893,method,Generates a hashed sparse cross from the input tensors.
3812,num_buckets,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3939,method,Returns number of buckets in this sparse feature.
3813,get_sparse_tensors,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3949,method,See `CategoricalColumn` base class.
3814,parents,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3964,method,See 'FeatureColumn` base class.
3815,get_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3968,method,See 'FeatureColumn` base class.
3816,from_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,3976,method,See 'FeatureColumn` base class.
3817,IndicatorColumn,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4028,class,"Represents a one-hot column for use in deep networks.
Args:
categorical_column: A `CategoricalColumn` which is created by
`categorical_column_with_*` function."
3818,name,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4047,method,See `FeatureColumn` base class.
3819,transform_feature,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4079,method,"Returns dense `Tensor` representing feature.
Args:
transformation_cache: A `FeatureTransformationCache` object to access
features.
state_manager: A `StateManager` to create / access resources such as
lookup tables.
Returns:
Transformed feature `Tensor`.
Raises:
ValueError: if input rank is not known at graph building time."
3820,parse_example_spec,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4107,method,See `FeatureColumn` base class.
3821,variable_shape,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4118,method,Returns a `TensorShape` representing the shape of the dense `Tensor`.
3822,get_dense_tensor,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4131,method,"Returns dense `Tensor` representing feature.
Args:
transformation_cache: A `FeatureTransformationCache` object to access
features.
state_manager: A `StateManager` to create / access resources such as
lookup tables.
Returns:
Dense `Tensor` created within `transform_feature`.
Raises:
ValueError: If `categorical_column` is a `SequenceCategoricalColumn`."
3823,get_sequence_dense_tensor,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4181,method,See `SequenceDenseColumn` base class.
3824,parents,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4231,method,See 'FeatureColumn` base class.
3825,get_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4235,method,See 'FeatureColumn` base class.
3826,from_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4244,method,See 'FeatureColumn` base class.
3827,SequenceCategoricalColumn,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4281,class,Represents sequences of categorical data.
3828,name,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4294,method,See `FeatureColumn` base class.
3829,parse_example_spec,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4299,method,See `FeatureColumn` base class.
3830,transform_feature,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4309,method,See `FeatureColumn` base class.
3831,num_buckets,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4320,method,Returns number of buckets in this sparse feature.
3832,get_sparse_tensors,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4346,method,"Returns an IdWeightPair.
`IdWeightPair` is a pair of `SparseTensor`s which represents ids and
weights.
`IdWeightPair.id_tensor` is typically a `batch_size` x `num_buckets`
`SparseTensor` of `int64`. `IdWeightPair.weight_tensor` is either a
`SparseTensor` of `float` or `None` to indicate all weights should be
taken to be 1. If specified, `weight_tensor` must have exactly the same
shape and indices as `sp_ids`. Expected `SparseTensor` is same as parsing
output of a `VarLenFeature` which is a ragged matrix.
Args:
transformation_cache: A `FeatureTransformationCache` object to access
features.
state_manager: A `StateManager` to create / access resources such as
lookup tables."
3833,parents,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4377,method,See 'FeatureColumn` base class.
3834,get_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4381,method,See 'FeatureColumn` base class.
3835,from_config,tensorflow/tensorflow/python/feature_column/feature_column_v2.py,4390,method,See 'FeatureColumn` base class.
3836,get_linear_model_bias,tensorflow/tensorflow/python/feature_column/feature_column_v2_test.py,59,function,
3837,get_linear_model_column_var,tensorflow/tensorflow/python/feature_column/feature_column_v2_test.py,64,function,
3838,concatenate_context_input,tensorflow/tensorflow/python/feature_column/sequence_feature_column.py,41,function,"Replicates `context_input` across all timesteps of `sequence_input`.
Expands dimension 1 of `context_input` then tiles it `sequence_length` times.
This value is appended to `sequence_input` on dimension 2 and the result is
returned.
Args:
context_input: A `Tensor` of dtype `float32` and shape `[batch_size, d1]`.
sequence_input: A `Tensor` of dtype `float32` and shape `[batch_size,
padded_length, d0]`.
Returns:
A `Tensor` of dtype `float32` and shape `[batch_size, padded_length,
d0 + d1]`.
Raises:
ValueError: If `sequence_input` does not have rank 3 or `context_input` does
not have rank 2."
3839,sequence_categorical_column_with_identity,tensorflow/tensorflow/python/feature_column/sequence_feature_column.py,91,function,"Returns a feature column that represents sequences of integers.
Pass this to `embedding_column` or `indicator_column` to convert sequence
categorical data into dense representation for input to sequence NN, such as
RNN.
Example:
```python
watches = sequence_categorical_column_with_identity(
'watches', num_buckets=1000)
watches_embedding = embedding_column(watches, dimension=10)
columns = [watches_embedding]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
sequence_feature_layer = SequenceFeatures(columns)
sequence_input, sequence_length = sequence_feature_layer(features)
sequence_length_mask = tf.sequence_mask(sequence_length)
rnn_cell = tf.keras.layers.SimpleRNNCell(hidden_size)
rnn_layer = tf.keras.layers.RNN(rnn_cell)
outputs, state = rnn_layer(sequence_input, mask=sequence_length_mask)
```
Args:
key: A unique string identifying the input feature.
num_buckets: Range of inputs. Namely, inputs are expected to be in the
range `[0, num_buckets)`.
default_value: If `None`, this column's graph operations will fail for
out-of-range inputs. Otherwise, this value must be in the range
`[0, num_buckets)`, and will replace out-of-range inputs.
Returns:
A `SequenceCategoricalColumn`.
Raises:
ValueError: if `num_buckets` is less than one.
ValueError: if `default_value` is not in range `[0, num_buckets)`."
3840,sequence_categorical_column_with_hash_bucket,tensorflow/tensorflow/python/feature_column/sequence_feature_column.py,140,function,"A sequence of categorical terms where ids are set by hashing.
Pass this to `embedding_column` or `indicator_column` to convert sequence
categorical data into dense representation for input to sequence NN, such as
RNN.
Example:
```python
tokens = sequence_categorical_column_with_hash_bucket(
'tokens', hash_bucket_size=1000)
tokens_embedding = embedding_column(tokens, dimension=10)
columns = [tokens_embedding]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
sequence_feature_layer = SequenceFeatures(columns)
sequence_input, sequence_length = sequence_feature_layer(features)
sequence_length_mask = tf.sequence_mask(sequence_length)
rnn_cell = tf.keras.layers.SimpleRNNCell(hidden_size)
rnn_layer = tf.keras.layers.RNN(rnn_cell)
outputs, state = rnn_layer(sequence_input, mask=sequence_length_mask)
```
Args:
key: A unique string identifying the input feature.
hash_bucket_size: An int > 1. The number of buckets.
dtype: The type of features. Only string and integer types are supported.
Returns:
A `SequenceCategoricalColumn`.
Raises:
ValueError: `hash_bucket_size` is not greater than 1.
ValueError: `dtype` is neither string nor integer."
3841,sequence_categorical_column_with_vocabulary_file,tensorflow/tensorflow/python/feature_column/sequence_feature_column.py,186,function,"A sequence of categorical terms where ids use a vocabulary file.
Pass this to `embedding_column` or `indicator_column` to convert sequence
categorical data into dense representation for input to sequence NN, such as
RNN.
Example:
```python
states = sequence_categorical_column_with_vocabulary_file(
key='states', vocabulary_file='/us/states.txt', vocabulary_size=50,
num_oov_buckets=5)
states_embedding = embedding_column(states, dimension=10)
columns = [states_embedding]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
sequence_feature_layer = SequenceFeatures(columns)
sequence_input, sequence_length = sequence_feature_layer(features)
sequence_length_mask = tf.sequence_mask(sequence_length)
rnn_cell = tf.keras.layers.SimpleRNNCell(hidden_size)
rnn_layer = tf.keras.layers.RNN(rnn_cell)
outputs, state = rnn_layer(sequence_input, mask=sequence_length_mask)
```
Args:
key: A unique string identifying the input feature.
vocabulary_file: The vocabulary file name.
vocabulary_size: Number of the elements in the vocabulary. This must be no
greater than length of `vocabulary_file`, if less than length, later
values are ignored. If None, it is set to the length of `vocabulary_file`.
num_oov_buckets: Non-negative integer, the number of out-of-vocabulary
buckets. All out-of-vocabulary inputs will be assigned IDs in the range
`[vocabulary_size, vocabulary_size+num_oov_buckets)` based on a hash of
the input value. A positive `num_oov_buckets` can not be specified with
`default_value`.
default_value: The integer ID value to return for out-of-vocabulary feature
values, defaults to `-1`. This can not be specified with a positive
`num_oov_buckets`.
dtype: The type of features. Only string and integer types are supported.
Returns:
A `SequenceCategoricalColumn`.
Raises:
ValueError: `vocabulary_file` is missing or cannot be opened.
ValueError: `vocabulary_size` is missing or < 1.
ValueError: `num_oov_buckets` is a negative integer.
ValueError: `num_oov_buckets` and `default_value` are both specified.
ValueError: `dtype` is neither string nor integer."
3842,sequence_categorical_column_with_vocabulary_list,tensorflow/tensorflow/python/feature_column/sequence_feature_column.py,251,function,"A sequence of categorical terms where ids use an in-memory list.
Pass this to `embedding_column` or `indicator_column` to convert sequence
categorical data into dense representation for input to sequence NN, such as
RNN.
Example:
```python
colors = sequence_categorical_column_with_vocabulary_list(
key='colors', vocabulary_list=('R', 'G', 'B', 'Y'),
num_oov_buckets=2)
colors_embedding = embedding_column(colors, dimension=3)
columns = [colors_embedding]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
sequence_feature_layer = SequenceFeatures(columns)
sequence_input, sequence_length = sequence_feature_layer(features)
sequence_length_mask = tf.sequence_mask(sequence_length)
rnn_cell = tf.keras.layers.SimpleRNNCell(hidden_size)
rnn_layer = tf.keras.layers.RNN(rnn_cell)
outputs, state = rnn_layer(sequence_input, mask=sequence_length_mask)
```
Args:
key: A unique string identifying the input feature.
vocabulary_list: An ordered iterable defining the vocabulary. Each feature
is mapped to the index of its value (if present) in `vocabulary_list`.
Must be castable to `dtype`.
dtype: The type of features. Only string and integer types are supported.
If `None`, it will be inferred from `vocabulary_list`.
default_value: The integer ID value to return for out-of-vocabulary feature
values, defaults to `-1`. This can not be specified with a positive
`num_oov_buckets`.
num_oov_buckets: Non-negative integer, the number of out-of-vocabulary
buckets. All out-of-vocabulary inputs will be assigned IDs in the range
`[len(vocabulary_list), len(vocabulary_list)+num_oov_buckets)` based on a
hash of the input value. A positive `num_oov_buckets` can not be specified
with `default_value`.
Returns:
A `SequenceCategoricalColumn`.
Raises:
ValueError: if `vocabulary_list` is empty, or contains duplicate keys.
ValueError: `num_oov_buckets` is a negative integer.
ValueError: `num_oov_buckets` and `default_value` are both specified.
ValueError: if `dtype` is not integer or string."
3843,sequence_numeric_column,tensorflow/tensorflow/python/feature_column/sequence_feature_column.py,313,function,"Returns a feature column that represents sequences of numeric data.
Example:
```python
temperature = sequence_numeric_column('temperature')
columns = [temperature]
features = tf.io.parse_example(..., features=make_parse_example_spec(columns))
sequence_feature_layer = SequenceFeatures(columns)
sequence_input, sequence_length = sequence_feature_layer(features)
sequence_length_mask = tf.sequence_mask(sequence_length)
rnn_cell = tf.keras.layers.SimpleRNNCell(hidden_size)
rnn_layer = tf.keras.layers.RNN(rnn_cell)
outputs, state = rnn_layer(sequence_input, mask=sequence_length_mask)
```
Args:
key: A unique string identifying the input features.
shape: The shape of the input data per sequence id. E.g. if `shape=(2,)`,
each example must contain `2 * sequence_length` values.
default_value: A single value compatible with `dtype` that is used for
padding the sparse data into a dense `Tensor`.
dtype: The type of values.
normalizer_fn: If not `None`, a function that can be used to normalize the
value of the tensor after `default_value` is applied for parsing.
Normalizer function takes the input `Tensor` as its argument, and returns
the output `Tensor`. (e.g. lambda x: (x - 3.0) / 4.2). Please note that
even though the most common use case of this function is normalization, it
can be used for any kind of Tensorflow transformations.
Returns:
A `SequenceNumericColumn`.
Raises:
TypeError: if any dimension in shape is not an int.
ValueError: if any dimension in shape is not a positive integer.
ValueError: if `dtype` is not convertible to `tf.float32`."
3844,SequenceNumericColumn,tensorflow/tensorflow/python/feature_column/sequence_feature_column.py,387,class,Represents sequences of numeric data.
3845,name,tensorflow/tensorflow/python/feature_column/sequence_feature_column.py,399,method,See `FeatureColumn` base class.
3846,parse_example_spec,tensorflow/tensorflow/python/feature_column/sequence_feature_column.py,404,method,See `FeatureColumn` base class.
3847,transform_feature,tensorflow/tensorflow/python/feature_column/sequence_feature_column.py,408,method,"See `FeatureColumn` base class.
In this case, we apply the `normalizer_fn` to the input tensor.
Args:
transformation_cache: A `FeatureTransformationCache` object to access
features.
state_manager: A `StateManager` to create / access resources such as
lookup tables.
Returns:
Normalized input tensor."
3848,variable_shape,tensorflow/tensorflow/python/feature_column/sequence_feature_column.py,428,method,Returns a `TensorShape` representing the shape of sequence input.
3849,get_sequence_dense_tensor,tensorflow/tensorflow/python/feature_column/sequence_feature_column.py,432,method,"Returns a `TensorSequenceLengthPair`.
Args:
transformation_cache: A `FeatureTransformationCache` object to access
features.
state_manager: A `StateManager` to create / access resources such as
lookup tables."
3850,parents,tensorflow/tensorflow/python/feature_column/sequence_feature_column.py,465,method,See 'FeatureColumn` base class.
3851,get_config,tensorflow/tensorflow/python/feature_column/sequence_feature_column.py,469,method,See 'FeatureColumn` base class.
3852,from_config,tensorflow/tensorflow/python/feature_column/sequence_feature_column.py,476,method,See 'FeatureColumn` base class.
3853,serialize_feature_column,tensorflow/tensorflow/python/feature_column/serialization.py,41,function,"Serializes a FeatureColumn or a raw string key.
This method should only be used to serialize parent FeatureColumns when
implementing FeatureColumn.get_config(), else serialize_feature_columns()
is preferable.
This serialization also keeps information of the FeatureColumn class, so
deserialization is possible without knowing the class type. For example:
a = numeric_column('x')
a.get_config() gives:
{
'key': 'price',
'shape': (1,),
'default_value': None,
'dtype': 'float32',
'normalizer_fn': None
}
While serialize_feature_column(a) gives:
{
'class_name': 'NumericColumn',
'config': {
'key': 'price',
'shape': (1,),
'default_value': None,
'dtype': 'float32',
'normalizer_fn': None
}
}
Args:
fc: A FeatureColumn or raw feature key string.
Returns:
Keras serialization for FeatureColumns, leaves string keys unaffected.
Raises:
ValueError if called with input that is not string or FeatureColumn."
3854,deserialize_feature_column,tensorflow/tensorflow/python/feature_column/serialization.py,89,function,"Deserializes a `config` generated with `serialize_feature_column`.
This method should only be used to deserialize parent FeatureColumns when
implementing FeatureColumn.from_config(), else deserialize_feature_columns()
is preferable. Returns a FeatureColumn for this config.
TODO(b/118939620): Simplify code if Keras utils support object deduping.
Args:
config: A Dict with the serialization of feature columns acquired by
`serialize_feature_column`, or a string representing a raw column.
custom_objects: A Dict from custom_object name to the associated keras
serializable objects (FeatureColumns, classes or functions).
columns_by_name: A Dict[String, FeatureColumn] of existing columns in order
to avoid duplication.
Raises:
ValueError if `config` has invalid format (e.g: expected keys missing,
or refers to unknown classes).
Returns:
A FeatureColumn corresponding to the input `config`."
3855,serialize_feature_columns,tensorflow/tensorflow/python/feature_column/serialization.py,146,function,"Serializes a list of FeatureColumns.
Returns a list of Keras-style config dicts that represent the input
FeatureColumns and can be used with `deserialize_feature_columns` for
reconstructing the original columns.
Args:
feature_columns: A list of FeatureColumns.
Returns:
Keras serialization for the list of FeatureColumns.
Raises:
ValueError if called with input that is not a list of FeatureColumns."
3856,deserialize_feature_columns,tensorflow/tensorflow/python/feature_column/serialization.py,165,function,"Deserializes a list of FeatureColumns configs.
Returns a list of FeatureColumns given a list of config dicts acquired by
`serialize_feature_columns`.
Args:
configs: A list of Dicts with the serialization of feature columns acquired
by `serialize_feature_columns`.
custom_objects: A Dict from custom_object name to the associated keras
serializable objects (FeatureColumns, classes or functions).
Returns:
FeatureColumn objects corresponding to the input configs.
Raises:
ValueError if called with input that is not a list of FeatureColumns."
3857,sequence_length_from_sparse_tensor,tensorflow/tensorflow/python/feature_column/utils.py,30,function,Returns a [batch_size] Tensor with per-example sequence length.
3858,assert_string_or_int,tensorflow/tensorflow/python/feature_column/utils.py,55,function,
3859,assert_key_is_string,tensorflow/tensorflow/python/feature_column/utils.py,61,function,
3860,check_default_value,tensorflow/tensorflow/python/feature_column/utils.py,68,function,"Returns default value as tuple if it's valid, otherwise raises errors.
This function verifies that `default_value` is compatible with both `shape`
and `dtype`. If it is not compatible, it raises an error. If it is compatible,
it casts default_value to a tuple and returns it. `key` is used only
for error message.
Args:
shape: An iterable of integers specifies the shape of the `Tensor`.
default_value: If a single value is provided, the same value will be applied
as the default value for every item. If an iterable of values is
provided, the shape of the `default_value` should be equal to the given
`shape`.
dtype: defines the type of values. Default value is `tf.float32`. Must be a
non-quantized, real integer or floating point type.
key: Column name, used only for error messages.
Returns:
A tuple which will be used as default value.
Raises:
TypeError: if `default_value` is an iterable but not compatible with `shape`
TypeError: if `default_value` is not compatible with `dtype`.
ValueError: if `dtype` is not convertible to `tf.float32`."
3861,op_is_stateful,tensorflow/tensorflow/python/framework/auto_control_deps.py,126,function,
3862,ResourceType,tensorflow/tensorflow/python/framework/auto_control_deps.py,132,class,
3863,collective_manager_ids_from_op,tensorflow/tensorflow/python/framework/auto_control_deps.py,137,function,"Returns CollectiveManager ID from the op if one exists, else None.
CollectiveManager adds collective and no_op operations tagged with an ID,
unique to the manager object. This function extracts that ID, or None, if the
node was not generated by a CollectiveManager.
Args:
op: `Operation` to get the collective manager ID from.
Returns:
List of CollectiveManager IDs used by the op."
3864,AutomaticControlDependencies,tensorflow/tensorflow/python/framework/auto_control_deps.py,163,class,"Context manager to automatically add control dependencies.
Code under this context manager will act as if a sensible set of control
dependencies were present. More specifically:
1. All stateful ops in the scope will execute (with the exception of ops in
ASYNC_STATEFUL_OPS and LEGACY_RANDOM_OPS)
2. Stateful ops which modify the same resource will execute in program order
Note: creating variables in an automatic control dependencies context is not
supported (the value of the variables will never change as they will keep
getting reinitialized).
NOT THREAD SAFE"
3865,mark_as_return,tensorflow/tensorflow/python/framework/auto_control_deps.py,188,method,"Acts like identity but marks the `Tensor` as a return value.
This will possibly return a copy of the `Tensor`. Usage:
```
with AutomaticControlDependencies() as a:
...
t = a.mark_as_return(t)
_ = ...(t...) # i.e. it's safe to use t here
```
Args:
tensor: the `Tensor` to be marked
Returns:
a copy of the `Tensor`."
3866,register_acd_resource_resolver,tensorflow/tensorflow/python/framework/auto_control_deps.py,483,function,"Register a function for resolving resources touched by an op.
`f` is called for every Operation added in the ACD context with the op's
original resource reads and writes. `f` is expected to update the sets of
resource reads and writes in-place and return True if it updated either of the
sets, False otherwise.
Example:
@register_acd_resource_resolver
def ResolveIdentity(op, resource_reads, resource_writes):
# op: The `Operation` being processed by ACD currently.
# resource_reads: An `ObjectIdentitySet` of read-only resources.
# resource_writes: An `ObjectIdentitySet` of read-write resources.
if not resource_reads or resource_writes:
return False
def update(resource_inputs):
to_add = []
to_remove = []
for t in resource_inputs:
if t.op.type == ""Identity"":
to_remove.append(t)
to_add.append(t.op.inputs[0])
if not to_add and not to_remove:
return False
for t in to_remove:
resource_inputs.discard(t)
resource_inputs.update(to_add)
return True
return update(resource_reads) or update(resource_writes)
Args:
f: Python function with signature
(Operation, ObjectIdentitySet, ObjectIdentitySet) -> bool
Returns:
The function `f` after adding it to the registry."
3867,automatic_control_dependencies,tensorflow/tensorflow/python/framework/auto_control_deps.py,550,function,"Wraps f to automatically insert control dependencies.
The inserted dependencies ensure that:
1. All stateful ops in f run when the result of f runs
2. Updates to the same resources happen in order.
Args:
f: the function to be wrapped.
Returns:
The wrapped function."
3868,register_read_only_resource_op,tensorflow/tensorflow/python/framework/auto_control_deps_utils.py,31,function,Declares that `op_type` does not update its touched resource.
3869,get_read_only_resource_input_indices_graph,tensorflow/tensorflow/python/framework/auto_control_deps_utils.py,36,function,Returns sorted list of read-only resource indices in func_graph.inputs.
3870,get_read_write_resource_inputs,tensorflow/tensorflow/python/framework/auto_control_deps_utils.py,89,function,"Returns a tuple of resource reads, writes in op.inputs.
Args:
op: Operation
Returns:
A 2-tuple of ObjectIdentitySets, the first entry containing read-only
resource handles and the second containing read-write resource handles in
`op.inputs`."
3871,ScopedTFStatus,tensorflow/tensorflow/python/framework/c_api_util.py,29,class,Wrapper around TF_Status that handles deletion.
3872,ScopedTFGraph,tensorflow/tensorflow/python/framework/c_api_util.py,44,class,Wrapper around TF_Graph that handles deletion.
3873,ScopedTFImportGraphDefOptions,tensorflow/tensorflow/python/framework/c_api_util.py,61,class,Wrapper around TF_ImportGraphDefOptions that handles deletion.
3874,ScopedTFImportGraphDefResults,tensorflow/tensorflow/python/framework/c_api_util.py,76,class,Wrapper around TF_ImportGraphDefOptions that handles deletion.
3875,ScopedTFFunction,tensorflow/tensorflow/python/framework/c_api_util.py,91,class,Wrapper around TF_Function that handles deletion.
3876,ScopedTFBuffer,tensorflow/tensorflow/python/framework/c_api_util.py,110,class,An internal class to help manage the TF_Buffer lifetime.
3877,ApiDefMap,tensorflow/tensorflow/python/framework/c_api_util.py,122,class,"Wrapper around Tf_ApiDefMap that handles querying and deletion.
The OpDef protos are also stored in this class so that they could
be queried by op name."
3878,put_api_def,tensorflow/tensorflow/python/framework/c_api_util.py,150,method,
3879,get_api_def,tensorflow/tensorflow/python/framework/c_api_util.py,153,method,
3880,get_op_def,tensorflow/tensorflow/python/framework/c_api_util.py,162,method,
3881,op_names,tensorflow/tensorflow/python/framework/c_api_util.py,167,method,
3882,tf_buffer,tensorflow/tensorflow/python/framework/c_api_util.py,172,function,"Context manager that creates and deletes TF_Buffer.
Example usage:
with tf_buffer() as buf:
# get serialized graph def into buf
...
proto_data = c_api.TF_GetBuffer(buf)
graph_def.ParseFromString(compat.as_bytes(proto_data))
# buf has been deleted
with tf_buffer(some_string) as buf:
c_api.TF_SomeFunction(buf)
# buf has been deleted
Args:
data: An optional `bytes`, `str`, or `unicode` object. If not None, the
yielded buffer will contain this data.
Yields:
Created TF_Buffer"
3883,tf_output,tensorflow/tensorflow/python/framework/c_api_util.py,204,function,"Returns a wrapped TF_Output with specified operation and index.
Args:
c_op: wrapped TF_Operation
index: integer
Returns:
Wrapped TF_Output"
3884,tf_operations,tensorflow/tensorflow/python/framework/c_api_util.py,220,function,"Generator that yields every TF_Operation in `graph`.
Args:
graph: Graph
Yields:
wrapped TF_Operation"
3885,new_tf_operations,tensorflow/tensorflow/python/framework/c_api_util.py,238,function,"Generator that yields newly-added TF_Operations in `graph`.
Specifically, yields TF_Operations that don't have associated Operations in
`graph`. This is useful for processing nodes added by the C API.
Args:
graph: Graph
Yields:
wrapped TF_Operation"
3886,EagerGraphCombination,tensorflow/tensorflow/python/framework/combinations.py,33,class,"Run the test in Graph or Eager mode.
The optional `mode` parameter controls the test's execution mode. Its
accepted values are ""graph"" or ""eager"" literals."
3887,context_managers,tensorflow/tensorflow/python/framework/combinations.py,40,method,
3888,parameter_modifiers,tensorflow/tensorflow/python/framework/combinations.py,52,method,
3889,TFVersionCombination,tensorflow/tensorflow/python/framework/combinations.py,56,class,"Control the execution of the test in TF1.x and TF2.
If TF2 is enabled then a test with TF1 test is going to be skipped and vice
versa.
Test targets continuously run in TF2 thanks to the tensorflow.v2 TAP target.
A test can be run in TF2 with bazel by passing --test_env=TF2_BEHAVIOR=1."
3890,should_execute_combination,tensorflow/tensorflow/python/framework/combinations.py,66,method,
3891,parameter_modifiers,tensorflow/tensorflow/python/framework/combinations.py,74,method,
3892,is_broadcast_compatible,tensorflow/tensorflow/python/framework/common_shapes.py,73,function,"Returns True if `shape_x` and `shape_y` are broadcast compatible.
Args:
shape_x: A `TensorShape`
shape_y: A `TensorShape`
Returns:
True if a shape exists that both `shape_x` and `shape_y` can be broadcasted
to. False otherwise."
3893,broadcast_shape,tensorflow/tensorflow/python/framework/common_shapes.py,89,function,"Returns the broadcasted shape between `shape_x` and `shape_y`.
Args:
shape_x: A `TensorShape`
shape_y: A `TensorShape`
Returns:
A `TensorShape` representing the broadcasted shape.
Raises:
ValueError: If the two shapes can not be broadcasted."
3894,CompositeTensor,tensorflow/tensorflow/python/framework/composite_tensor.py,31,class,"Abstract base class for Tensor-like objects that are composed from Tensors.
Each `CompositeTensor` can be decomposed into a structured collection of
component `tf.Tensor`s, and reconstructed from those components.
The `tensorflow.python.util.nest` module has support for treating composite
tensors as structure, which makes it easy to flatten and reconstruct
composite tensors (or larger structures that contain composite tensors).
E.g.:
```python
ct = ... # Create a composite tensor.
flat_list_of_tensors = nest.flatten(ct, expand_composites=True)
transformed_list_of_tensors = ... # do something with the flat tensors.
result = nest.pack_sequence_as(ct, transformed_list_of_tensors,
expand_composites=True)
```"
3895,replace_composites_with_components,tensorflow/tensorflow/python/framework/composite_tensor.py,94,function,"Recursively replaces CompositeTensors with their components.
Args:
structure: A `nest`-compatible structure, possibly containing composite
tensors.
Returns:
A copy of `structure`, where each composite tensor has been replaced by
its components. The result will contain no composite tensors.
Note that `nest.flatten(replace_composites_with_components(structure))`
returns the same value as `nest.flatten(structure)`."
3896,CTSpec,tensorflow/tensorflow/python/framework/composite_tensor_test.py,40,class,"A generic CompositeTensor TypeSpec, used for constructing tests."
3897,CT,tensorflow/tensorflow/python/framework/composite_tensor_test.py,60,class,"A generic CompositeTensor, used for constructing tests."
3898,CTSpec2,tensorflow/tensorflow/python/framework/composite_tensor_test.py,87,class,
3899,CT2,tensorflow/tensorflow/python/framework/composite_tensor_test.py,91,class,
3900,is_composite_or_composite_value,tensorflow/tensorflow/python/framework/composite_tensor_utils.py,31,function,Returns true if 'tensor' is a CompositeTensor or a CT Value object.
3901,get_shape,tensorflow/tensorflow/python/framework/composite_tensor_utils.py,41,function,Returns the shape of the passed composite tensor.
3902,append_composite_tensor,tensorflow/tensorflow/python/framework/composite_tensor_utils.py,113,function,"Helper function to append composite tensors to each other in the 0 axis.
In order to support batching within a fit/evaluate/predict call, we need
to be able to aggregate within a CompositeTensor. Unfortunately, the CT
API currently does not make this easy - especially in V1 mode, where we're
working with CompositeTensor Value objects that have no connection with the
CompositeTensors that created them.
Arguments:
target: CompositeTensor or CompositeTensor value object that will be
appended to.
to_append: CompositeTensor or CompositeTensor value object to append to.
'target'.
Returns:
A CompositeTensor or CompositeTensor value object.
Raises:
RuntimeError: if concatenation is not possible."
3903,tensor_float_32_execution_allowed,tensorflow/tensorflow/python/framework/config.py,28,function,"Get if TensorFloat-32 operations are enabled on supported hardware.
Returns:
True if TensorFloat-32 execution is enabled and False otherwise."
3904,allow_tensor_float_32_execution,tensorflow/tensorflow/python/framework/config.py,38,function,"Allow use of TensorFloat-32 with float32 ops on supported hardware.
TensorFloat-32 is a math mode introduced with the NVIDIA Ampere architecture.
TensorFloat-32 kernels take float32 inputs and produce float32 outputs.
Internally, the inputs are cast to a custom representation with 10-bit
mantissa (similar to float16) and 8-bit exponent (similar to float32) and are
executed using TensorCores with float32 accumulation. For more information,
see https://blogs.nvidia.com/blog/2020/05/14/tensorfloat-32-precision-format/.
TensorFloat-32 execution is disabled by default, but this may change in a
future version.
Args:
allowed: whether to allow TensorFloat-32 execution"
3905,get_intra_op_parallelism_threads,tensorflow/tensorflow/python/framework/config.py,58,function,"Get number of threads used within an individual op for parallelism.
Certain operations like matrix multiplication and reductions can utilize
parallel threads for speed ups. A value of 0 means the system picks an
appropriate number.
Returns:
Number of parallel threads"
3906,set_intra_op_parallelism_threads,tensorflow/tensorflow/python/framework/config.py,72,function,"Set number of threads used within an individual op for parallelism.
Certain operations like matrix multiplication and reductions can utilize
parallel threads for speed ups. A value of 0 means the system picks an
appropriate number.
Args:
num_threads: Number of parallel threads"
3907,get_inter_op_parallelism_threads,tensorflow/tensorflow/python/framework/config.py,86,function,"Get number of threads used for parallelism between independent operations.
Determines the number of threads used by independent non-blocking operations.
0 means the system picks an appropriate number.
Returns:
Number of parallel threads"
3908,set_inter_op_parallelism_threads,tensorflow/tensorflow/python/framework/config.py,99,function,"Set number of threads used for parallelism between independent operations.
Determines the number of threads used by independent non-blocking operations.
0 means the system picks an appropriate number.
Args:
num_threads: Number of parallel threads"
3909,get_optimizer_jit,tensorflow/tensorflow/python/framework/config.py,112,function,"Get if JIT compilation is enabled.
Note that optimizations are only applied to code that is compiled into a
graph. In eager mode, which is the TF2 API default, that means only code that
is defined under a tf.function decorator.
Returns:
If JIT compilation is enabled."
3910,set_optimizer_jit,tensorflow/tensorflow/python/framework/config.py,126,function,"Set if JIT compilation is enabled.
Note that optimizations are only applied to code that is compiled into a
graph. In eager mode, which is the TF2 API default, that means only code that
is defined under a tf.function decorator.
Args:
enabled: Whether to enable JIT compilation."
3911,get_optimizer_experimental_options,tensorflow/tensorflow/python/framework/config.py,140,function,"Get experimental optimizer options.
Refer to tf.config.optimizer.set_experimental_options for a list of current
options.
Note that optimizations are only applied in graph mode, (within tf.function).
In addition, as these are experimental options, the list is subject to change.
Returns:
Dictionary of configured experimental optimizer options"
3912,set_optimizer_experimental_options,tensorflow/tensorflow/python/framework/config.py,156,function,"Set experimental optimizer options.
Note that optimizations are only applied in graph mode, (within tf.function).
In addition, as these are experimental options, the list is subject to change.
Args:
options: Dictionary of experimental optimizer options to configure.
Valid keys:
- layout_optimizer: Optimize tensor layouts
e.g. This will try to use NCHW layout on GPU which is faster.
- constant_folding: Fold constants
Statically infer the value of tensors when possible, and materialize the
result using constants.
- shape_optimization: Simplify computations made on shapes.
- remapping: Remap subgraphs onto more efficient implementations.
- arithmetic_optimization: Simplify arithmetic ops with common
sub-expression elimination and arithmetic simplification.
- dependency_optimization: Control dependency optimizations. Remove
redundant control dependencies, which may enable other optimization.
This optimizer is also essential for pruning Identity and NoOp nodes.
- loop_optimization: Loop optimizations.
- function_optimization: Function optimizations and inlining.
- debug_stripper: Strips debug-related nodes from the graph.
- disable_model_pruning: Disable removal of unnecessary ops from the graph
- scoped_allocator_optimization: Try to allocate some independent Op
outputs contiguously in order to merge or eliminate downstream Ops.
- pin_to_host_optimization: Force small ops onto the CPU.
- implementation_selector: Enable the swap of kernel implementations based
on the device placement.
- auto_mixed_precision: Change certain float32 ops to float16 on Volta
GPUs and above. Without the use of loss scaling, this can cause
numerical underflow (see
`keras.mixed_precision.experimental.LossScaleOptimizer`).
- disable_meta_optimizer: Disable the entire meta optimizer.
- min_graph_nodes: The minimum number of nodes in a graph to optimizer.
For smaller graphs, optimization is skipped."
3913,get_soft_device_placement,tensorflow/tensorflow/python/framework/config.py,198,function,"Get if soft device placement is enabled.
If enabled, an op will be placed on CPU if any of the following are true
1. there's no GPU implementation for the OP
2. no GPU devices are known or registered
3. need to co-locate with reftype input(s) which are from CPU
Returns:
If soft placement is enabled."
3914,set_soft_device_placement,tensorflow/tensorflow/python/framework/config.py,213,function,"Set if soft device placement is enabled.
If enabled, an op will be placed on CPU if any of the following are true
1. there's no GPU implementation for the OP
2. no GPU devices are known or registered
3. need to co-locate with reftype input(s) which are from CPU
Args:
enabled: Whether to enable soft placement."
3915,get_device_policy,tensorflow/tensorflow/python/framework/config.py,228,function,"Gets the current device policy.
The device policy controls how operations requiring inputs on a specific
device (e.g., on GPU:0) handle inputs on a different device (e.g. GPU:1).
This function only gets the device policy for the current thread. Any
subsequently started thread will again use the default policy.
Returns:
Current thread device policy"
3916,set_device_policy,tensorflow/tensorflow/python/framework/config.py,254,function,"Sets the current thread device policy.
The device policy controls how operations requiring inputs on a specific
device (e.g., on GPU:0) handle inputs on a different device (e.g. GPU:1).
When using the default, an appropriate policy will be picked automatically.
The default policy may change over time.
This function only sets the device policy for the current thread. Any
subsequently started thread will again use the default policy.
Args:
device_policy: A device policy.
Valid values:
- None: Switch to a system default.
- 'warn': Copies the tensors which are not on the right device and logs
a warning.
- 'explicit': Raises an error if the placement is not as required.
- 'silent': Silently copies the tensors. Note that this may hide
performance problems as there is no notification provided when
operations are blocked on the tensor being copied between devices.
- 'silent_for_int32': silently copies `int32` tensors, raising errors on
the other ones.
Raises:
ValueError: If an invalid `device_policy` is passed."
3917,get_synchronous_execution,tensorflow/tensorflow/python/framework/config.py,297,function,"Gets whether operations are executed synchronously or asynchronously.
TensorFlow can execute operations synchronously or asynchronously. If
asynchronous execution is enabled, operations may return ""non-ready"" handles.
Returns:
Current thread execution mode"
3918,set_synchronous_execution,tensorflow/tensorflow/python/framework/config.py,310,function,"Specifies whether operations are executed synchronously or asynchronously.
TensorFlow can execute operations synchronously or asynchronously. If
asynchronous execution is enabled, operations may return ""non-ready"" handles.
When `enable` is set to None, an appropriate value will be picked
automatically. The value picked may change between TensorFlow releases.
Args:
enable: Whether operations should be dispatched synchronously.
Valid values:
- None: sets the system default.
- True: executes each operation synchronously.
- False: executes each operation asynchronously."
3919,list_physical_devices,tensorflow/tensorflow/python/framework/config.py,338,function,"Return a list of physical devices visible to the host runtime.
Physical devices are hardware devices present on the host machine. By default
all discovered CPU and GPU devices are considered visible.
This API allows querying the physical hardware resources prior to runtime
initialization. Thus, giving an opportunity to call any additional
configuration APIs. This is in contrast to `tf.config.list_logical_devices`,
which triggers runtime initialization in order to list the configured devices.
The following example lists the number of visible GPUs on the host.
>>> physical_devices = tf.config.list_physical_devices('GPU')
>>> print(""Num GPUs:"", len(physical_devices))
Num GPUs: ...
However, the number of GPUs available to the runtime may change during runtime
initialization due to marking certain devices as not visible or configuring
multiple logical devices.
Args:
device_type: (optional string) Only include devices matching this device
type. For example ""CPU"" or ""GPU"".
Returns:
List of discovered `tf.config.PhysicalDevice` objects"
3920,list_logical_devices,tensorflow/tensorflow/python/framework/config.py,373,function,"Return a list of logical devices created by runtime.
Logical devices may correspond to physical devices or remote devices in the
cluster. Operations and tensors may be placed on these devices by using the
`name` of the `tf.config.LogicalDevice`.
Calling `tf.config.list_logical_devices` triggers the runtime to configure any
`tf.config.PhysicalDevice` visible to the runtime, thereby preventing
further configuration. To avoid runtime initialization, call
`tf.config.list_physical_devices` instead.
For example:
>>> logical_devices = tf.config.list_logical_devices('GPU')
>>> if len(logical_devices) > 0:
... # Allocate on GPU:0
... with tf.device(logical_devices[0].name):
... one = tf.constant(1)
... # Allocate on GPU:1
... with tf.device(logical_devices[1].name):
... two = tf.constant(2)
Args:
device_type: (optional string) Only include devices matching this device
type. For example ""CPU"" or ""GPU"".
Returns:
List of initialized `LogicalDevice`s"
3921,get_visible_devices,tensorflow/tensorflow/python/framework/config.py,410,function,"Get the list of visible physical devices.
Returns the list of `PhysicalDevice`s currently marked as visible to the
runtime. A visible device will have at least one `LogicalDevice` associated
with it once the runtime is initialized.
The following example verifies all visible GPUs have been disabled:
>>> physical_devices = tf.config.list_physical_devices('GPU')
>>> try:
... # Disable all GPUS
... tf.config.set_visible_devices([], 'GPU')
... visible_devices = tf.config.get_visible_devices()
... for device in visible_devices:
... assert device.device_type != 'GPU'
... except:
... # Invalid device or cannot modify virtual devices once initialized.
... pass
Args:
device_type: (optional string) Only include devices matching this device
type. For example ""CPU"" or ""GPU"".
Returns:
List of visible `PhysicalDevice`s"
3922,set_visible_devices,tensorflow/tensorflow/python/framework/config.py,444,function,"Set the list of visible devices.
Specifies which `PhysicalDevice` objects are visible to the runtime.
TensorFlow will only allocate memory and place operations on visible
physical devices, as otherwise no `LogicalDevice` will be created on them.
By default all discovered devices are marked as visible.
The following example demonstrates disabling the first GPU on the machine.
>>> physical_devices = tf.config.list_physical_devices('GPU')
>>> try:
... # Disable first GPU
... tf.config.set_visible_devices(physical_devices[1:], 'GPU')
... logical_devices = tf.config.list_logical_devices('GPU')
... # Logical device was not created for first GPU
... assert len(logical_devices) == len(physical_devices) - 1
... except:
... # Invalid device or cannot modify virtual devices once initialized.
... pass
Args:
devices: List of `PhysicalDevice`s to make visible
device_type: (optional) Only configure devices matching this device type.
For example ""CPU"" or ""GPU"". Other devices will be left unaltered.
Raises:
ValueError: If argument validation fails.
RuntimeError: Runtime is already initialized."
3923,get_memory_growth,tensorflow/tensorflow/python/framework/config.py,478,function,"Get if memory growth is enabled for a `PhysicalDevice`.
If memory growth is enabled for a `PhysicalDevice`, the runtime initialization
will not allocate all memory on the device.
For example:
>>> physical_devices = tf.config.list_physical_devices('GPU')
>>> try:
... tf.config.experimental.set_memory_growth(physical_devices[0], True)
... assert tf.config.experimental.get_memory_growth(physical_devices[0])
... except:
... # Invalid device or cannot modify virtual devices once initialized.
... pass
Args:
device: `PhysicalDevice` to query
Returns:
A boolean indicating the memory growth setting for the `PhysicalDevice`.
Raises:
ValueError: Invalid `PhysicalDevice` specified."
3924,set_memory_growth,tensorflow/tensorflow/python/framework/config.py,507,function,"Set if memory growth should be enabled for a `PhysicalDevice`.
If memory growth is enabled for a `PhysicalDevice`, the runtime initialization
will not allocate all memory on the device. Memory growth cannot be configured
on a `PhysicalDevice` with virtual devices configured.
For example:
>>> physical_devices = tf.config.list_physical_devices('GPU')
>>> try:
... tf.config.experimental.set_memory_growth(physical_devices[0], True)
... except:
... # Invalid device or cannot modify virtual devices once initialized.
... pass
Args:
device: `PhysicalDevice` to configure
enable: (Boolean) Whether to enable or disable memory growth
Raises:
ValueError: Invalid `PhysicalDevice` specified.
RuntimeError: Runtime is already initialized."
3925,get_device_details,tensorflow/tensorflow/python/framework/config.py,535,function,"Returns details about a physical devices.
This API takes in a `tf.config.PhysicalDevice` returned by
`tf.config.list_physical_devices`. It returns a dict with string keys
containing various details about the device. Each key is only supported by a
subset of devices, so you should not assume the returned dict will have any
particular key.
>>> gpu_devices = tf.config.list_physical_devices('GPU')
>>> if gpu_devices:
... details = tf.config.experimental.get_device_details(gpu_devices[0])
... details.get('device_name', 'Unknown GPU')
Currently, details are only returned for GPUs. This function returns an
empty dict if passed a non-GPU device.
The returned dict may have the following keys:
* `'device_name'`: A human-readable name of the device as a string, e.g.
""Titan V"". Unlike `tf.config.PhysicalDevice.name`, this will be the same for
multiple devices if each device is the same model. Currently only available
for GPUs.
* `'compute_capability'`: The
[compute capability](https://developer.nvidia.com/cuda-gpus) of the device
as a tuple of two ints, in the form `(major_version, minor_version)`. Only
available for NVIDIA GPUs
Note: This is similar to `tf.sysconfig.get_build_info` in that both functions
can return information relating to GPUs. However, this function returns
run-time information about a specific device (such as a GPU's compute
capability), while `tf.sysconfig.get_build_info` returns compile-time
information about how TensorFlow was built (such as what version of CUDA
TensorFlow was built for).
Args:
device: A `tf.config.PhysicalDevice` returned by
`tf.config.list_physical_devices` or `tf.config.get_visible_devices`.
Returns:
A dict with string keys."
3926,get_logical_device_configuration,tensorflow/tensorflow/python/framework/config.py,583,function,"Get the virtual device configuration for a `tf.config.PhysicalDevice`.
Returns the list of `tf.config.LogicalDeviceConfiguration`
objects previously configured by a call to
`tf.config.set_logical_device_configuration`.
For example:
>>> physical_devices = tf.config.list_physical_devices('CPU')
>>> assert len(physical_devices) == 1, ""No CPUs found""
>>> configs = tf.config.get_logical_device_configuration(
... physical_devices[0])
>>> try:
... assert configs is None
... tf.config.set_logical_device_configuration(
... physical_devices[0],
... [tf.config.LogicalDeviceConfiguration(),
... tf.config.LogicalDeviceConfiguration()])
... configs = tf.config.get_logical_device_configuration(
... physical_devices[0])
... assert len(configs) == 2
... except:
... # Cannot modify virtual devices once initialized.
... pass
Args:
device: `PhysicalDevice` to query
Returns:
List of `tf.config.LogicalDeviceConfiguration` objects or
`None` if no virtual device configuration has been set for this physical
device."
3927,set_logical_device_configuration,tensorflow/tensorflow/python/framework/config.py,624,function,"Set the logical device configuration for a `tf.config.PhysicalDevice`.
A visible `tf.config.PhysicalDevice` will by default have a single
`tf.config.LogicalDevice` associated with it once the runtime is initialized.
Specifying a list of `tf.config.LogicalDeviceConfiguration` objects allows
multiple devices to be created on the same `tf.config.PhysicalDevice`.
The following example splits the CPU into 2 logical devices:
>>> physical_devices = tf.config.list_physical_devices('CPU')
>>> assert len(physical_devices) == 1, ""No CPUs found""
>>> # Specify 2 virtual CPUs. Note currently memory limit is not supported.
>>> try:
... tf.config.set_logical_device_configuration(
... physical_devices[0],
... [tf.config.LogicalDeviceConfiguration(),
... tf.config.LogicalDeviceConfiguration()])
... logical_devices = tf.config.list_logical_devices('CPU')
... assert len(logical_devices) == 2
...
... tf.config.set_logical_device_configuration(
... physical_devices[0],
... [tf.config.LogicalDeviceConfiguration(),
... tf.config.LogicalDeviceConfiguration(),
... tf.config.LogicalDeviceConfiguration(),
... tf.config.LogicalDeviceConfiguration()])
... except:
... # Cannot modify logical devices once initialized.
... pass
The following example splits the GPU into 2 logical devices with 100 MB each:
>>> physical_devices = tf.config.list_physical_devices('GPU')
>>> try:
... tf.config.set_logical_device_configuration(
... physical_devices[0],
... [tf.config.LogicalDeviceConfiguration(memory_limit=100),
... tf.config.LogicalDeviceConfiguration(memory_limit=100)])
...
... logical_devices = tf.config.list_logical_devices('GPU')
... assert len(logical_devices) == len(physical_devices) + 1
...
... tf.config.set_logical_device_configuration(
... physical_devices[0],
... [tf.config.LogicalDeviceConfiguration(memory_limit=10),
... tf.config.LogicalDeviceConfiguration(memory_limit=10)])
... except:
... # Invalid device or cannot modify logical devices once initialized.
... pass
Args:
device: The `PhysicalDevice` to configure.
logical_devices: (optional) List of `tf.config.LogicalDeviceConfiguration`
objects to allocate for the specified `PhysicalDevice`. If None, the
default configuration will be used.
Raises:
ValueError: If argument validation fails.
RuntimeError: Runtime is already initialized."
3928,enable_mlir_bridge,tensorflow/tensorflow/python/framework/config.py,689,function,"Enables experimental MLIR-Based TensorFlow Compiler Bridge.
DO NOT USE, DEV AND TESTING ONLY AT THE MOMENT.
NOTE: MLIR-Based TensorFlow Compiler is under active development and has
missing features, please refrain from using. This API exists for development
and testing only.
TensorFlow Compiler Bridge (TF Bridge) is responsible for translating parts
of TensorFlow graph into a form that can be accepted as an input by a backend
compiler such as XLA."
3929,enable_mlir_graph_optimization,tensorflow/tensorflow/python/framework/config.py,706,function,"Enables experimental MLIR-Based TensorFlow Compiler Optimizations.
DO NOT USE, DEV AND TESTING ONLY AT THE MOMENT.
NOTE: MLIR-Based TensorFlow Compiler is under active development and has
missing features, please refrain from using. This API exists for development
and testing only.
TensorFlow Compiler Optimizations are responsible general graph level
optimizations that in the current stack mostly done by Grappler graph
optimizers."
3930,disable_mlir_bridge,tensorflow/tensorflow/python/framework/config.py,723,function,Disables experimental MLIR-Based TensorFlow Compiler Bridge.
3931,disable_mlir_graph_optimization,tensorflow/tensorflow/python/framework/config.py,729,function,Disables experimental MLIR-Based TensorFlow Compiler Optimizations.
3932,reset_eager,tensorflow/tensorflow/python/framework/config_test.py,42,function,
3933,convert_to_eager_tensor,tensorflow/tensorflow/python/framework/constant_op.py,70,function,"Converts the given `value` to an `EagerTensor`.
Note that this function could return cached copies of created constants for
performance reasons.
Args:
value: value to convert to EagerTensor.
ctx: value of context.context().
dtype: optional desired dtype of the converted EagerTensor.
Returns:
EagerTensor created from value.
Raises:
TypeError: if `dtype` is not compatible with the type of t."
3934,constant_v1,tensorflow/tensorflow/python/framework/constant_op.py,102,function,"Creates a constant tensor.
The resulting tensor is populated with values of type `dtype`, as
specified by arguments `value` and (optionally) `shape` (see examples
below).
The argument `value` can be a constant value, or a list of values of type
`dtype`. If `value` is a list, then the length of the list must be less
than or equal to the number of elements implied by the `shape` argument (if
specified). In the case where the list length is less than the number of
elements specified by `shape`, the last element in the list will be used
to fill the remaining entries.
The argument `shape` is optional. If present, it specifies the dimensions of
the resulting tensor. If not present, the shape of `value` is used.
If the argument `dtype` is not specified, then the type is inferred from
the type of `value`.
For example:
```python
# Constant 1-D Tensor populated with value list.
tensor = tf.constant([1, 2, 3, 4, 5, 6, 7]) => [1 2 3 4 5 6 7]
# Constant 2-D tensor populated with scalar value -1.
tensor = tf.constant(-1.0, shape=[2, 3]) => [[-1. -1. -1.]
[-1. -1. -1.]]
```
`tf.constant` differs from `tf.fill` in a few ways:
* `tf.constant` supports arbitrary constants, not just uniform scalar
Tensors like `tf.fill`.
* `tf.constant` creates a `Const` node in the computation graph with the
exact value at graph construction time. On the other hand, `tf.fill`
creates an Op in the graph that is expanded at runtime.
* Because `tf.constant` only embeds constant values in the graph, it does
not support dynamic shapes based on other runtime Tensors, whereas
`tf.fill` does.
Args:
value: A constant value (or list) of output type `dtype`.
dtype: The type of the elements of the resulting tensor.
shape: Optional dimensions of resulting tensor.
name: Optional name for the tensor.
verify_shape: Boolean that enables verification of a shape of values.
Returns:
A Constant Tensor.
Raises:
TypeError: if shape is incorrectly specified or unsupported."
3935,constant,tensorflow/tensorflow/python/framework/constant_op.py,167,function,"Creates a constant tensor from a tensor-like object.
Note: All eager `tf.Tensor` values are immutable (in contrast to
`tf.Variable`). There is nothing especially _constant_ about the value
returned from `tf.constant`. This function it is not fundamentally different
from `tf.convert_to_tensor`. The name `tf.constant` comes from the `value`
being embeded in a `Const` node in the `tf.Graph`. `tf.constant` is useful
for asserting that the value can be embedded that way.
If the argument `dtype` is not specified, then the type is inferred from
the type of `value`.
>>> # Constant 1-D Tensor from a python list.
>>> tf.constant([1, 2, 3, 4, 5, 6])
<tf.Tensor: shape=(6,), dtype=int32,
numpy=array([1, 2, 3, 4, 5, 6], dtype=int32)>
>>> # Or a numpy array
>>> a = np.array([[1, 2, 3], [4, 5, 6]])
>>> tf.constant(a)
<tf.Tensor: shape=(2, 3), dtype=int64, numpy=
array([[1, 2, 3],
[4, 5, 6]])>
If `dtype` is specified the resulting tensor values are cast to the requested
`dtype`.
>>> tf.constant([1, 2, 3, 4, 5, 6], dtype=tf.float64)
<tf.Tensor: shape=(6,), dtype=float64,
numpy=array([1., 2., 3., 4., 5., 6.])>
If `shape` is set, the `value` is reshaped to match. Scalars are expanded to
fill the `shape`:
>>> tf.constant(0, shape=(2, 3))
<tf.Tensor: shape=(2, 3), dtype=int32, numpy=
array([[0, 0, 0],
[0, 0, 0]], dtype=int32)>
>>> tf.constant([1, 2, 3, 4, 5, 6], shape=[2, 3])
<tf.Tensor: shape=(2, 3), dtype=int32, numpy=
array([[1, 2, 3],
[4, 5, 6]], dtype=int32)>
`tf.constant` has no effect if an eager Tensor is passed as the `value`, it
even transmits gradients:
>>> v = tf.Variable([0.0])
>>> with tf.GradientTape() as g:
... loss = tf.constant(v + v)
>>> g.gradient(loss, v).numpy()
array([2.], dtype=float32)
But, since `tf.constant` embeds the value in the `tf.Graph` this fails for
symbolic tensors:
>>> with tf.compat.v1.Graph().as_default():
... i = tf.compat.v1.placeholder(shape=[None, None], dtype=tf.float32)
... t = tf.constant(i)
Traceback (most recent call last):
...
TypeError: ...
`tf.constant` will _always_ create CPU (host) tensors. In order to create
tensors on other devices, use `tf.identity`. (If the `value` is an eager
Tensor, however, the tensor will be returned unmodified as mentioned above.)
Related Ops:
* `tf.convert_to_tensor` is similar but:
* It has no `shape` argument.
* Symbolic tensors are allowed to pass through.
>>> with tf.compat.v1.Graph().as_default():
... i = tf.compat.v1.placeholder(shape=[None, None], dtype=tf.float32)
... t = tf.convert_to_tensor(i)
* `tf.fill`: differs in a few ways:
* `tf.constant` supports arbitrary constants, not just uniform scalar
Tensors like `tf.fill`.
* `tf.fill` creates an Op in the graph that is expanded at runtime, so it
can efficiently represent large tensors.
* Since `tf.fill` does not embed the value, it can produce dynamically
sized outputs.
Args:
value: A constant value (or list) of output type `dtype`.
dtype: The type of the elements of the resulting tensor.
shape: Optional dimensions of resulting tensor.
name: Optional name for the tensor.
Returns:
A Constant Tensor.
Raises:
TypeError: if shape is incorrectly specified or unsupported.
ValueError: if called on a symbolic tensor."
3936,is_constant,tensorflow/tensorflow/python/framework/constant_op.py,328,function,
3937,disable_lower_using_switch_merge,tensorflow/tensorflow/python/framework/convert_to_constants.py,881,function,"Set '_lower_using_switch_merge' attributes to False.
Sets the attribute to False in the NodeDefs in the main graph and the NodeDefs
in each function's graph.
Args:
graph_def: GraphDef proto.
Returns:
GraphDef"
3938,convert_variables_to_constants_v2,tensorflow/tensorflow/python/framework/convert_to_constants.py,1042,function,"Replaces all the variables in a graph with constants of the same values.
TensorFlow 2.0 function for converting all Variable ops into Const ops holding
the same values. This makes it possible to describe the network fully with a
single GraphDef file, and allows the removal of a lot of ops related to
loading and saving the variables. This function runs Grappler's function
inlining optimization in order to return a single subgraph.
The current implementation only works for graphs that do not contain any
control flow or embedding related ops.
Args:
func: ConcreteFunction.
lower_control_flow: Boolean indicating whether or not to lower control flow
ops such as If and While. (default True)
aggressive_inlining: Boolean indicating whether or not to to aggressive
function inlining (might be unsafe if function has stateful ops, not
properly connected to control outputs). (default False)
Returns:
ConcreteFunction containing a simplified version of the original."
3939,convert_variables_to_constants_v2_as_graph,tensorflow/tensorflow/python/framework/convert_to_constants.py,1080,function,"Replaces all the variables in a graph with constants of the same values.
This function works as same as convert_variables_to_constants_v2, but it
returns the intermediate `GraphDef` as well. This `GraphDef` contains all the
debug information after all the transformations in the frozen phase.
Args:
func: ConcreteFunction.
lower_control_flow: Boolean indicating whether or not to lower control flow
ops such as If and While. (default True)
aggressive_inlining: Boolean indicating whether or not to to aggressive
function inlining (might be unsafe if function has stateful ops, not
properly connected to control outputs).
Returns:
ConcreteFunction containing a simplified version of the original, and also
the intermediate GraphDef containing the node debug information for the
transformations in the frozen phase."
3940,convert_variables_to_constants_from_session_graph,tensorflow/tensorflow/python/framework/convert_to_constants.py,1115,function,"Replaces all the variables in a graph with constants of the same values.
This function works similarly to convert_variables_to_constants_v2, but it
retrieves the constant values from a Session instead of from a
ConcreteFunction. This is useful when converting graphs generated from
TensorFlow V1, where ConcreteFunctions are not available. This also differs
from graph_util.convert_variables_to_constants in that it supports resource
variables when V2 control flow constructions are present.
Args:
session: Active TensorFlow session containing the variables.
graph_def: A GraphDef to convert.
output_node_names: List of name strings for the result nodes of the graph.
variable_names_allowlist: The set of variable names to convert (by default,
all variables are converted).
variable_names_denylist: The set of variable names to omit converting to
constants.
Returns:
An optimized GraphDef."
3941,check_valid,tensorflow/tensorflow/python/framework/device.py,32,function,"Check that a device spec is valid.
Args:
spec: a string.
Raises:
An exception if the spec is invalid."
3942,is_device_spec,tensorflow/tensorflow/python/framework/device.py,45,function,Abstract away the fact that DeviceSpecV2 is the base class.
3943,canonical_name,tensorflow/tensorflow/python/framework/device.py,50,function,Returns a canonical name for the given `DeviceSpec` or device name.
3944,merge_device,tensorflow/tensorflow/python/framework/device.py,67,function,"Returns a device function that merges devices specifications.
This can be used to merge partial specifications of devices. The
innermost setting for a device field takes precedence. For example:
with tf.device(merge_device(""/device:GPU:0""))
# Nodes created here have device ""/device:GPU:0""
with tf.device(merge_device(""/job:worker"")):
# Nodes created here have device ""/job:worker/device:GPU:0""
with tf.device(merge_device(""/device:CPU:0"")):
# Nodes created here have device ""/job:worker/device:CPU:0""
with tf.device(merge_device(""/job:ps"")):
# Nodes created here have device ""/job:ps/device:CPU:0""
Args:
spec: A `DeviceSpec` or a device spec string (partially) describing the
device that should be used for all nodes created in the scope of
the returned device function's with block.
Returns:
A MergeDevice object with the above-described behavior.
Raises:
ValueError: if the spec was not valid."
3945,MergeDevice,tensorflow/tensorflow/python/framework/device.py,107,class,"Wraps a device specification (DeviceSpec or str) with merge functionality.
When called, this class will merge a node_def with its own spec. It also
exposes a `shortcut_string_merge` method which can significantly improve
performance of device placement."
3946,shortcut_string_merge,tensorflow/tensorflow/python/framework/device.py,135,method,"Merge a node def without materializing a full DeviceSpec object.
Often a device merge is invoked in order to generate a string which can be
passed into the c api. In such a case, we can cache the
node_def.device -> merge_result_string
map, and in most cases avoid:
- Materializing a copy of self._spec (In the case of DeviceSpecV1)
- Materializing a DeviceSpec for node_def.device
- A DeviceSpec.merge_from invocation
In practice the cache hit rate for this function is very high, because the
number of invocations when iterating through the device stack is much
larger than the number of devices.
Args:
node_def: An Operation (or Operation-like) to merge device constraints
with self._spec
Returns:
A string containing the merged device specification."
3947,is_null_merge,tensorflow/tensorflow/python/framework/device.py,175,method,"Indicate whether the wrapped spec is empty.
In the degenerate case where self._spec is an empty specification, a caller
may wish to skip a merge step entirely. (However this class does not have
enough information to make that determination.)
Returns:
A boolean indicating whether a device merge will be trivial."
3948,DeviceSpecV2,tensorflow/tensorflow/python/framework/device_spec.py,51,class,"Represents a (possibly partial) specification for a TensorFlow device.
`DeviceSpec`s are used throughout TensorFlow to describe where state is stored
and computations occur. Using `DeviceSpec` allows you to parse device spec
strings to verify their validity, merge them or compose them programmatically.
Example:
```python
# Place the operations on device ""GPU:0"" in the ""ps"" job.
device_spec = DeviceSpec(job=""ps"", device_type=""GPU"", device_index=0)
with tf.device(device_spec.to_string()):
# Both my_var and squared_var will be placed on /job:ps/device:GPU:0.
my_var = tf.Variable(..., name=""my_variable"")
squared_var = tf.square(my_var)
```
With eager execution disabled (by default in TensorFlow 1.x and by calling
disable_eager_execution() in TensorFlow 2.x), the following syntax
can be used:
```python
tf.compat.v1.disable_eager_execution()
# Same as previous
device_spec = DeviceSpec(job=""ps"", device_type=""GPU"", device_index=0)
# No need of .to_string() method.
with tf.device(device_spec):
my_var = tf.Variable(..., name=""my_variable"")
squared_var = tf.square(my_var)
```
If a `DeviceSpec` is partially specified, it will be merged with other
`DeviceSpec`s according to the scope in which it is defined. `DeviceSpec`
components defined in inner scopes take precedence over those defined in
outer scopes.
```python
gpu0_spec = DeviceSpec(job=""ps"", device_type=""GPU"", device_index=0)
with tf.device(DeviceSpec(job=""train"").to_string()):
with tf.device(gpu0_spec.to_string()):
# Nodes created here will be assigned to /job:ps/device:GPU:0.
with tf.device(DeviceSpec(device_type=""GPU"", device_index=1).to_string()):
# Nodes created here will be assigned to /job:train/device:GPU:1.
```
A `DeviceSpec` consists of 5 components -- each of
which is optionally specified:
* Job: The job name.
* Replica: The replica index.
* Task: The task index.
* Device type: The device type string (e.g. ""CPU"" or ""GPU"").
* Device index: The device index."
3949,to_string,tensorflow/tensorflow/python/framework/device_spec.py,133,method,"Return a string representation of this `DeviceSpec`.
Returns:
a string of the form
/job:<name>/replica:<id>/task:<id>/device:<device_type>:<id>."
3950,from_string,tensorflow/tensorflow/python/framework/device_spec.py,143,method,"Construct a `DeviceSpec` from a string.
Args:
spec: a string of the form
/job:<name>/replica:<id>/task:<id>/device:CPU:<id>
or
/job:<name>/replica:<id>/task:<id>/device:GPU:<id>
as cpu and gpu are mutually exclusive.
All entries are optional.
Returns:
A DeviceSpec."
3951,parse_from_string,tensorflow/tensorflow/python/framework/device_spec.py,159,method,"Parse a `DeviceSpec` name into its components.
2.x behavior change:
In TensorFlow 1.x, this function mutates its own state and returns itself.
In 2.x, DeviceSpecs are immutable, and this function will return a
DeviceSpec which contains the spec.
Recommended:
```
# my_spec and my_updated_spec are unrelated.
my_spec = tf.DeviceSpec.from_string(""/CPU:0"")
my_updated_spec = tf.DeviceSpec.from_string(""/GPU:0"")
with tf.device(my_updated_spec):
...
```
Will work in 1.x and 2.x (though deprecated in 2.x):
```
my_spec = tf.DeviceSpec.from_string(""/CPU:0"")
my_updated_spec = my_spec.parse_from_string(""/GPU:0"")
with tf.device(my_updated_spec):
...
```
Will NOT work in 2.x:
```
my_spec = tf.DeviceSpec.from_string(""/CPU:0"")
my_spec.parse_from_string(""/GPU:0"") # <== Will not update my_spec
with tf.device(my_spec):
...
```
In general, `DeviceSpec.from_string` should completely replace
`DeviceSpec.parse_from_string`, and `DeviceSpec.replace` should
completely replace setting attributes directly.
Args:
spec: an optional string of the form
/job:<name>/replica:<id>/task:<id>/device:CPU:<id>
or
/job:<name>/replica:<id>/task:<id>/device:GPU:<id>
as cpu and gpu are mutually exclusive.
All entries are optional.
Returns:
The `DeviceSpec`.
Raises:
ValueError: if the spec was not valid."
3952,make_merged_spec,tensorflow/tensorflow/python/framework/device_spec.py,212,method,"Returns a new DeviceSpec which incorporates `dev`.
When combining specs, `dev` will take precedence over the current spec.
So for instance:
```
first_spec = tf.DeviceSpec(job=0, device_type=""CPU"")
second_spec = tf.DeviceSpec(device_type=""GPU"")
combined_spec = first_spec.make_merged_spec(second_spec)
```
is equivalent to:
```
combined_spec = tf.DeviceSpec(job=0, device_type=""GPU"")
```
Args:
dev: a `DeviceSpec`
Returns:
A new `DeviceSpec` which combines `self` and `dev`"
3953,replace,tensorflow/tensorflow/python/framework/device_spec.py,236,method,"Convenience method for making a new DeviceSpec by overriding fields.
For instance:
```
my_spec = DeviceSpec=(job=""my_job"", device=""CPU"")
my_updated_spec = my_spec.replace(device=""GPU"")
my_other_spec = my_spec.replace(device=None)
```
Args:
**kwargs: This method takes the same args as the DeviceSpec constructor
Returns:
A DeviceSpec with the fields specified in kwargs overridden."
3954,job,tensorflow/tensorflow/python/framework/device_spec.py,261,method,
3955,replica,tensorflow/tensorflow/python/framework/device_spec.py,265,method,
3956,task,tensorflow/tensorflow/python/framework/device_spec.py,269,method,
3957,device_type,tensorflow/tensorflow/python/framework/device_spec.py,273,method,
3958,device_index,tensorflow/tensorflow/python/framework/device_spec.py,277,method,
3959,DeviceSpecV1,tensorflow/tensorflow/python/framework/device_spec.py,397,class,
3960,job,tensorflow/tensorflow/python/framework/device_spec.py,402,method,
3961,replica,tensorflow/tensorflow/python/framework/device_spec.py,407,method,
3962,task,tensorflow/tensorflow/python/framework/device_spec.py,412,method,
3963,device_type,tensorflow/tensorflow/python/framework/device_spec.py,417,method,
3964,device_index,tensorflow/tensorflow/python/framework/device_spec.py,422,method,
3965,to_string,tensorflow/tensorflow/python/framework/device_spec.py,431,method,
3966,parse_from_string,tensorflow/tensorflow/python/framework/device_spec.py,438,method,
3967,merge_from,tensorflow/tensorflow/python/framework/device_spec.py,444,method,"Merge the properties of ""dev"" into this `DeviceSpec`.
Note: Will be removed in TensorFlow 2.x since DeviceSpecs will become
immutable.
Args:
dev: a `DeviceSpec`."
3968,DType,tensorflow/tensorflow/python/framework/dtypes.py,37,class,"Represents the type of the elements in a `Tensor`.
The following `DType` objects are defined:
* `tf.float16`: 16-bit half-precision floating-point.
* `tf.float32`: 32-bit single-precision floating-point.
* `tf.float64`: 64-bit double-precision floating-point.
* `tf.bfloat16`: 16-bit truncated floating-point.
* `tf.complex64`: 64-bit single-precision complex.
* `tf.complex128`: 128-bit double-precision complex.
* `tf.int8`: 8-bit signed integer.
* `tf.uint8`: 8-bit unsigned integer.
* `tf.uint16`: 16-bit unsigned integer.
* `tf.uint32`: 32-bit unsigned integer.
* `tf.uint64`: 64-bit unsigned integer.
* `tf.int16`: 16-bit signed integer.
* `tf.int32`: 32-bit signed integer.
* `tf.int64`: 64-bit signed integer.
* `tf.bool`: Boolean.
* `tf.string`: String.
* `tf.qint8`: Quantized 8-bit signed integer.
* `tf.quint8`: Quantized 8-bit unsigned integer.
* `tf.qint16`: Quantized 16-bit signed integer.
* `tf.quint16`: Quantized 16-bit unsigned integer.
* `tf.qint32`: Quantized 32-bit signed integer.
* `tf.resource`: Handle to a mutable resource.
* `tf.variant`: Values of arbitrary types.
The `tf.as_dtype()` function converts numpy types and string type
names to a `DType` object."
3969,base_dtype,tensorflow/tensorflow/python/framework/dtypes.py,85,method,Returns a non-reference `DType` based on this `DType`.
3970,real_dtype,tensorflow/tensorflow/python/framework/dtypes.py,93,method,Returns the `DType` corresponding to this `DType`'s real part.
3971,as_numpy_dtype,tensorflow/tensorflow/python/framework/dtypes.py,104,method,Returns a Python `type` object based on this `DType`.
3972,min,tensorflow/tensorflow/python/framework/dtypes.py,109,method,"Returns the minimum representable value in this data type.
Raises:
TypeError: if this is a non-numeric, unordered, or quantized type."
3973,max,tensorflow/tensorflow/python/framework/dtypes.py,133,method,"Returns the maximum representable value in this data type.
Raises:
TypeError: if this is a non-numeric, unordered, or quantized type."
3974,limits,tensorflow/tensorflow/python/framework/dtypes.py,157,method,"Return intensity limits, i.e.
(min, max) tuple, of the dtype.
Args:
clip_negative : bool, optional If True, clip the negative range (i.e.
return 0 for min intensity) even if the image dtype allows negative
values. Returns
min, max : tuple Lower and upper intensity limits."
3975,is_compatible_with,tensorflow/tensorflow/python/framework/dtypes.py,172,method,"Returns True if the `other` DType will be converted to this DType.
The conversion rules are as follows:
```python
DType(T) .is_compatible_with(DType(T)) == True
```
Args:
other: A `DType` (or object that may be converted to a `DType`).
Returns:
True if a Tensor of the `other` `DType` will be implicitly converted to
this `DType`."
3976,as_dtype,tensorflow/tensorflow/python/framework/dtypes.py,607,function,"Converts the given `type_value` to a `DType`.
Note: `DType` values are interned. When passed a new `DType` object,
`as_dtype` always returns the interned value.
Args:
type_value: A value that can be converted to a `tf.DType` object. This may
currently be a `tf.DType` object, a [`DataType`
enum](https://www.tensorflow.org/code/tensorflow/core/framework/types.proto),
a string type name, or a `numpy.dtype`.
Returns:
A `DType` corresponding to `type_value`.
Raises:
TypeError: If `type_value` cannot be converted to a `DType`."
3977,parse_message,tensorflow/tensorflow/python/framework/error_interpolation.py,69,function,"Parses the message.
Splits the message into separators and tags. Tags are named tuples
representing the string {{type name}} and they are separated by
separators. For example, in ""123{{node Foo}}456{{node Bar}}789"", there are
two tags and three separators. The separators are the numeric characters.
Args:
message: String to parse
Returns:
(list of separator strings, list of _ParseTags).
For example, if message is ""123{{node Foo}}456"" then this function
returns ([""123"", ""456""], [_ParseTag(""node"", ""Foo"")])"
3978,create_graph_debug_info_def,tensorflow/tensorflow/python/framework/error_interpolation.py,284,function,"Construct and returns a `GraphDebugInfo` protocol buffer.
Args:
func_named_operations: An iterable of (func_name, op.Operation) tuples
where the Operation instances have a _traceback members. The func_name
should be the empty string for operations in the top-level Graph.
Returns:
GraphDebugInfo protocol buffer.
Raises:
TypeError: If the arguments are not of the correct proto buffer type."
3979,traceback_files_common_prefix,tensorflow/tensorflow/python/framework/error_interpolation.py,404,function,"Determines the common prefix from the paths of the stacktrace of 'all_ops'.
For example, if the paths are '/foo/bar/baz/' and '/foo/car', this would
return '/foo'.
Args:
all_ops: All the input nodes in the form of a list of lists of ops.
Returns:
The common prefix."
3980,interpolate,tensorflow/tensorflow/python/framework/error_interpolation.py,487,function,"Interpolates an error message.
The error message can contain tags of the form `{{type name}}` which will be
replaced. For example: ""{{node <name>}}"" would get expanded to:
""node <name>(defined at <path>)"".
Args:
error_message: A string to interpolate.
graph: ops.Graph object containing all nodes referenced in the error
message.
Returns:
The string with tags of the form {{type name}} interpolated."
3981,InaccessibleTensorError,tensorflow/tensorflow/python/framework/errors_impl.py,47,class,
3982,OperatorNotAllowedInGraphError,tensorflow/tensorflow/python/framework/errors_impl.py,52,class,"An error is raised for unsupported operator in Graph execution.
For example, using a `tf.Tensor` as a Python `bool` in Graph execution
is not allowed."
3983,OpError,tensorflow/tensorflow/python/framework/errors_impl.py,63,class,"A generic error that is raised when TensorFlow execution fails.
Whenever possible, the session will raise a more specific subclass
of `OpError` from the `tf.errors` module."
3984,message,tensorflow/tensorflow/python/framework/errors_impl.py,93,method,The error message that describes the error.
3985,op,tensorflow/tensorflow/python/framework/errors_impl.py,98,method,"The operation that failed, if known.
*N.B.* If the failed op was synthesized at runtime, e.g. a `Send`
or `Recv` op, there will be no corresponding
`tf.Operation`
object. In that case, this will return `None`, and you should
instead use the `tf.errors.OpError.node_def` to
discover information about the op.
Returns:
The `Operation` that failed, or None."
3986,error_code,tensorflow/tensorflow/python/framework/errors_impl.py,114,method,The integer error code that describes the error.
3987,node_def,tensorflow/tensorflow/python/framework/errors_impl.py,119,method,The `NodeDef` proto representing the op that failed.
3988,CancelledError,tensorflow/tensorflow/python/framework/errors_impl.py,217,class,"Raised when an operation or step is cancelled.
For example, a long-running operation (e.g.
`tf.QueueBase.enqueue` may be
cancelled by running another operation (e.g.
`tf.QueueBase.close`,
or by `tf.Session.close`.
A step that is running such a long-running operation will fail by raising
`CancelledError`.
@@__init__"
3989,UnknownError,tensorflow/tensorflow/python/framework/errors_impl.py,238,class,"Unknown error.
An example of where this error may be returned is if a Status value
received from another address space belongs to an error-space that
is not known to this address space. Also, errors raised by APIs that
do not return enough error information may be converted to this
error.
@@__init__"
3990,InvalidArgumentError,tensorflow/tensorflow/python/framework/errors_impl.py,256,class,"Raised when an operation receives an invalid argument.
This may occur, for example, if an operation receives an input
tensor that has an invalid value or shape. For example, the
`tf.matmul` op will raise this
error if it receives an input that is not a matrix, and the
`tf.reshape` op will raise
this error if the new shape does not match the number of elements in the input
tensor.
@@__init__"
3991,DeadlineExceededError,tensorflow/tensorflow/python/framework/errors_impl.py,277,class,"Raised when a deadline expires before an operation could complete.
This exception is not currently used.
@@__init__"
3992,NotFoundError,tensorflow/tensorflow/python/framework/errors_impl.py,292,class,"Raised when a requested entity (e.g., a file or directory) was not found.
For example, running the
`tf.WholeFileReader.read`
operation could raise `NotFoundError` if it receives the name of a file that
does not exist.
@@__init__"
3993,AlreadyExistsError,tensorflow/tensorflow/python/framework/errors_impl.py,309,class,"Raised when an entity that we attempted to create already exists.
For example, running an operation that saves a file
(e.g. `tf.train.Saver.save`)
could potentially raise this exception if an explicit filename for an
existing file was passed.
@@__init__"
3994,PermissionDeniedError,tensorflow/tensorflow/python/framework/errors_impl.py,327,class,"Raised when the caller does not have permission to run an operation.
For example, running the
`tf.WholeFileReader.read`
operation could raise `PermissionDeniedError` if it receives the name of a
file for which the user does not have the read file permission.
@@__init__"
3995,UnauthenticatedError,tensorflow/tensorflow/python/framework/errors_impl.py,345,class,"The request does not have valid authentication credentials.
This exception is not currently used.
@@__init__"
3996,ResourceExhaustedError,tensorflow/tensorflow/python/framework/errors_impl.py,360,class,"Some resource has been exhausted.
For example, this error might be raised if a per-user quota is
exhausted, or perhaps the entire file system is out of space.
@@__init__"
3997,FailedPreconditionError,tensorflow/tensorflow/python/framework/errors_impl.py,376,class,"Operation was rejected because the system is not in a state to execute it.
This exception is most commonly raised when running an operation
that reads a `tf.Variable`
before it has been initialized.
@@__init__"
3998,AbortedError,tensorflow/tensorflow/python/framework/errors_impl.py,393,class,"The operation was aborted, typically due to a concurrent action.
For example, running a
`tf.QueueBase.enqueue`
operation may raise `AbortedError` if a
`tf.QueueBase.close` operation
previously ran.
@@__init__"
3999,OutOfRangeError,tensorflow/tensorflow/python/framework/errors_impl.py,411,class,"Raised when an operation iterates past the valid input range.
This exception is raised in ""end-of-file"" conditions, such as when a
`tf.QueueBase.dequeue`
operation is blocked on an empty queue, and a
`tf.QueueBase.close`
operation executes.
@@__init__"
4000,UnimplementedError,tensorflow/tensorflow/python/framework/errors_impl.py,430,class,"Raised when an operation has not been implemented.
Some operations may raise this error when passed otherwise-valid
arguments that it does not currently support. For example, running
the `tf.nn.max_pool2d` operation
would raise this error if pooling was requested on the batch dimension,
because this is not yet supported.
@@__init__"
4001,InternalError,tensorflow/tensorflow/python/framework/errors_impl.py,449,class,"Raised when the system experiences an internal error.
This exception is raised when some invariant expected by the runtime
has been broken. Catching this exception is not recommended.
@@__init__"
4002,UnavailableError,tensorflow/tensorflow/python/framework/errors_impl.py,464,class,"Raised when the runtime is currently unavailable.
This exception is not currently used.
@@__init__"
4003,DataLossError,tensorflow/tensorflow/python/framework/errors_impl.py,479,class,"Raised when unrecoverable data loss or corruption is encountered.
For example, this may be raised by running a
`tf.WholeFileReader.read`
operation, if the file is truncated while it is being read.
@@__init__"
4004,exception_type_from_error_code,tensorflow/tensorflow/python/framework/errors_impl.py,520,function,
4005,error_code_from_exception_type,tensorflow/tensorflow/python/framework/errors_impl.py,525,function,
4006,raise_exception_on_not_ok_status,tensorflow/tensorflow/python/framework/errors_impl.py,547,class,Context manager to check for C API status.
4007,UnknownArgument,tensorflow/tensorflow/python/framework/func_graph.py,64,class,Signifies an argument which is not currently handled.
4008,convert_structure_to_signature,tensorflow/tensorflow/python/framework/func_graph.py,69,function,"Convert a potentially nested structure to a signature.
Args:
structure: Structure to convert, where top level collection is a list or a
tuple.
arg_names: Optional list of arguments that has equal number of elements as
`structure` and is used for naming corresponding TensorSpecs.
Returns:
Identical structure that has TensorSpec objects instead of Tensors and
UnknownArgument instead of any unsupported types."
4009,FuncGraph,tensorflow/tensorflow/python/framework/func_graph.py,134,class,"Graph representing a function body.
Attributes:
name: The name of the function.
inputs: Placeholder tensors representing the inputs to this function. The
tensors are in this FuncGraph. This represents ""regular"" inputs as well as
captured inputs (i.e. the values of self.captures), with the regular
inputs coming first.
outputs: Tensors that will be returned by this function. The tensors are in
this FuncGraph.
control_outputs: Operations that must be executed before the function
represented by this graph can be said to have been executed.
structured_input_signature: A tuple of (args, kwargs), which are both
possibly-nested python objects that were received by this function. Note
that these structures might contain Python `None`s.
structured_outputs: A possibly-nested python object which will be returned
by this function. The Tensors in this structure are the same as those of
self.outputs. Note that this structure might contain Python `None`s.
variables: Variables that should be watched during function execution.
outer_graph: The graph this function is defined in. May be another FuncGraph
or the global default Graph.
captures: Maps external tensor -> internal tensor (i.e. input placeholder).
The entries are in the order they were captured.
control_captures: Set of external ops on which this graph has a control
dependency.
seed: The graph-level random seed.
capture_by_value: If True, the func graph will capture Variables by value
instead of reference."
4010,watch_variable,tensorflow/tensorflow/python/framework/func_graph.py,265,method,Marks the variable v as accessed while building this graph.
4011,capture_call_time_value,tensorflow/tensorflow/python/framework/func_graph.py,271,method,"Creates a placeholder which at call time has the value closure().
Useful, for example, to respect TensorFlow context managers, which are often
dynamically scoped.
Args:
closure: function which takes no arguments, to be evaluated at function
call time, returning a nest of tensors compatible with `spec`.
spec: nest of TypeSpec for the value to capture.
key: optional. If not None, multiple calls to lazy_capture with the same
key in the same graph will return the same placeholder, and the
first closure will be used at function call time.
Returns:
Nest of placeholders which, at function call time, will be fed with the
result of calling closure().
Raises:
ValueError: at function call time, if the return value of closure() is
not compatible with `spec`."
4012,control_dependencies,tensorflow/tensorflow/python/framework/func_graph.py,321,method,"Handles control dependencies.
FuncGraph wraps Graph's control_dependencies logic by first filtering out
any external tensors / operations and storing them in the graph's
control_captures member. Any consumers of this function graph must then
decide how to handle the control captures.
Args:
control_inputs: A list of `Operation` or `Tensor` objects which
must be executed or computed before running the operations
defined in the context. Can also be `None` to clear the control
dependencies.
Returns:
A context manager that specifies control dependencies for all
operations constructed within the context.
Raises:
TypeError: If `control_inputs` is not a list of `Operation` or
`Tensor` objects."
4013,as_default,tensorflow/tensorflow/python/framework/func_graph.py,362,method,
4014,outer_graph,tensorflow/tensorflow/python/framework/func_graph.py,428,method,"The Graph this FuncGraph is nested in.
Functions may capture Tensors from graphs they are nested in (transitive).
Returns:
A Graph object. Initially set to the current default graph when the
FuncGraph was created. If the previous `outer_graph` was deleted because
the function that owns it was deleted, `outer_graph` is reset to the
outermost default graph active when the FuncGraph was created. This
FuncGraph won't have captured anything from the new `outer_graph` (and
likely not from the previous setting, since that would have created a
strong reference), but it is returned so that FuncGraphs always have a
parent."
4015,output_types,tensorflow/tensorflow/python/framework/func_graph.py,449,method,
4016,output_shapes,tensorflow/tensorflow/python/framework/func_graph.py,453,method,
4017,trainable_variables,tensorflow/tensorflow/python/framework/func_graph.py,457,method,"A sequence of trainable variables accessed by this FuncGraph.
Note that functions keep only weak references to variables. Calling the
function after a variable it accesses has been deleted is an error.
Returns:
Sequence of trainable variables for this func graph."
4018,variables,tensorflow/tensorflow/python/framework/func_graph.py,469,method,"A sequence of variables accessed by this FuncGraph.
Note that functions keep only weak references to variables. Calling the
function after a variable it accesses has been deleted is an error.
Returns:
Sequence of variables for this func graph."
4019,variables,tensorflow/tensorflow/python/framework/func_graph.py,492,method,
4020,capture,tensorflow/tensorflow/python/framework/func_graph.py,595,method,"Captures `tensor` if it's external to this graph.
If `tensor` is from a different graph, returns a placeholder for it.
`tensor` and the placeholder will appear in self.captures, and the
placeholder will appear in self.inputs. Multiple calls to this method with
the same `tensor` argument will return the same placeholder. If `tensor` is
from this graph, returns `tensor`.
Args:
tensor: Tensor. May be from this FuncGraph or a different graph.
name: Optional name if a placeholder is created.
shape: Optional shape if a placeholder is created.
Returns:
Tensor from this FuncGraph.
Raises:
InaccessibleTensorError: if any tensors are accessed in a manner that
bypasses the mechanisms required for the data dependencies to be correctly
wired."
4021,captures,tensorflow/tensorflow/python/framework/func_graph.py,665,method,Order list of tuples containing external and internal captures.
4022,add_capture,tensorflow/tensorflow/python/framework/func_graph.py,669,method,"Capture a specific tensor and utilize the provided placeholder.
Args:
tensor: Tensor to captures.
placeholder: Provided placeholder for the tensor."
4023,replace_capture,tensorflow/tensorflow/python/framework/func_graph.py,679,method,Replace already existing capture.
4024,reset_captures,tensorflow/tensorflow/python/framework/func_graph.py,683,method,Set the captures with the provided list of captures & placeholder.
4025,pop_capture,tensorflow/tensorflow/python/framework/func_graph.py,689,method,Remove the capture and return the generated placeholder.
4026,clear_captures,tensorflow/tensorflow/python/framework/func_graph.py,697,method,
4027,capture_distributed_variable,tensorflow/tensorflow/python/framework/func_graph.py,707,method,Add given distributed variable to captures with given placeholder.
4028,capture_eager_tensor,tensorflow/tensorflow/python/framework/func_graph.py,714,method,
4029,captured,tensorflow/tensorflow/python/framework/func_graph.py,731,method,Check if the specified tensor has been captured.
4030,external_captures,tensorflow/tensorflow/python/framework/func_graph.py,736,method,External tensors captured by this function.
4031,internal_captures,tensorflow/tensorflow/python/framework/func_graph.py,741,method,Placeholders in this function corresponding captured tensors.
4032,deferred_external_captures,tensorflow/tensorflow/python/framework/func_graph.py,746,method,Ordered nest of tensors whose placeholders will be fed at call time.
4033,deferred_internal_captures,tensorflow/tensorflow/python/framework/func_graph.py,751,method,List of nest of placeholders which at call time will be fed.
4034,variable_captures,tensorflow/tensorflow/python/framework/func_graph.py,756,method,Map of python object ids of variables to variables which are captured.
4035,mark_as_unsaveable,tensorflow/tensorflow/python/framework/func_graph.py,764,method,"Marks this FuncGraph as unsaveable.
Any attempts to export this FuncGraph will raise an error with the specified
message.
Args:
error_message: List or string containing the error message to be raised
when saving this FuncGraph to SavedModel."
4036,saveable,tensorflow/tensorflow/python/framework/func_graph.py,780,method,Returns whether this FuncGraph is saveable.
4037,saving_errors,tensorflow/tensorflow/python/framework/func_graph.py,785,method,Returns set of errors preventing this FuncGraph from being saved.
4038,inner_cm,tensorflow/tensorflow/python/framework/func_graph.py,366,method,Context manager for copying distribute.Strategy scope information.
4039,deref,tensorflow/tensorflow/python/framework/func_graph.py,478,method,
4040,convert_to_placeholder,tensorflow/tensorflow/python/framework/func_graph.py,297,method,
4041,wrapped_closure,tensorflow/tensorflow/python/framework/func_graph.py,307,method,
4042,func_graph_from_py_func,tensorflow/tensorflow/python/framework/func_graph.py,801,function,"Returns a `FuncGraph` generated from `python_func`.
Args:
name: an identifier for the function.
python_func: the Python function to trace.
args: the positional args with which the Python function should be called;
ignored if a signature is provided.
kwargs: the keyword args with which the Python function should be called;
ignored if a signature is provided.
signature: a possibly nested sequence of `TensorSpecs` specifying the shapes
and dtypes of the arguments. When a signature is provided, `args` and
`kwargs` are ignored, and `python_func` is traced with Tensors conforming
to `signature`. If `None`, the shapes and dtypes are inferred from the
inputs.
func_graph: Optional. An instance of FuncGraph. If provided, we will use
this graph else a new one is built and returned.
autograph: whether to use autograph to compile `python_func`.
See https://www.tensorflow.org/guide/autograph for more information.
autograph_options: additional knobs to control when `autograph=True`.
See https://www.tensorflow.org/guide/autograph for more information.
add_control_dependencies: If True, automatically adds control dependencies
to ensure program order matches execution order and stateful ops always
execute.
arg_names: Optional list of argument names, used to give input placeholders
recognizable names.
op_return_value: Optional. A Tensor. If set and `python_func` returns
Operations, those return values will be replaced with this value. If not
set, returning an Operation triggers an error.
collections: a dictionary of collections this FuncGraph should start
with. If not specified (None), the FuncGraph will read (but not write to)
the outer graph's collections that are not allowlisted, and both
read and write to the outer graph's collections that are allowlisted.
The current allowlisted collections are the global variables, the
local variables, and the trainable variables.
Defaults to None.
capture_by_value: An optional boolean. If True, the func graph will capture
Variables by value instead of reference. By default inherit from outer
graphs, and failing that will default to False.
override_flat_arg_shapes: An optional list of instances that are either
`None` or `TensorShape`. The length must match that of
`nest.flatten((args, kwargs), expand_composites=True)`. The entries
containing value `None` must match entries in flattened arguments
containing non-tensors, while entries containing a `TensorShape` must
match entries in the flattened arguments containing tensors.
Returns:
A FuncGraph.
Raises:
TypeError: If any of `python_func`'s return values is neither `None` nor a
`Tensor`.
ValueError: If both `signature` and `override_flat_arg_shapes` are
passed in."
4043,maybe_captured,tensorflow/tensorflow/python/framework/func_graph.py,1037,function,"If t is a captured value placeholder, returns the original captured value.
Args:
tensor: Tensor.
Returns:
A tensor, potentially from a different Graph/FuncGraph."
4044,device_stack_has_callable,tensorflow/tensorflow/python/framework/func_graph.py,1055,function,Checks whether a device stack contains a callable.
4045,check_mutation,tensorflow/tensorflow/python/framework/func_graph.py,1061,function,Check if two list of arguments are exactly the same.
4046,flatten,tensorflow/tensorflow/python/framework/func_graph.py,1081,function,"Like nest.flatten w/ expand_composites, but returns flow for TensorArrays.
Args:
sequence: A nested structure of Tensors, CompositeTensors, and
TensorArrays.
Returns:
A list of tensors."
4047,pack_sequence_as,tensorflow/tensorflow/python/framework/func_graph.py,1098,function,"Like `nest.pack_sequence_as` but also builds TensorArrays from flows.
Args:
structure: The structure to pack into. May contain Tensors,
CompositeTensors, or TensorArrays.
flat_sequence: An iterable containing tensors.
Returns:
A nested structure.
Raises:
AssertionError if `structure` and `flat_sequence` are not compatible."
4048,dismantle_func_graph,tensorflow/tensorflow/python/framework/func_graph.py,1269,function,"Removes reference cycles in `func_graph` FuncGraph.
Helpful for making sure the garbage collector doesn't need to run when
the FuncGraph goes out of scope, e.g. in tests using defun with
@test_util.run_in_graph_and_eager_modes(assert_no_eager_garbage=True).
Args:
func_graph: A `FuncGraph` object to destroy. `func_graph` is unusable
after this function."
4049,override_func_graph_name_scope,tensorflow/tensorflow/python/framework/func_graph.py,1284,function,
4050,Defun,tensorflow/tensorflow/python/framework/function.py,45,class,"Decorator used to define TensorFlow functions.
Use this decorator to make a Python function usable directly as a TensorFlow
function.
The decorated function must add ops to the default graph and return zero or
more `Tensor` objects. Call the decorator with named arguments, one for each
argument of the function to decorate, with the expected type of the argument
as value.
For example if the function to decorate accepts two `tf.float32` arguments
named `x` and `y`, call the decorator with:
@Defun(tf.float32, tf.float32)
def foo(x, y):
...
When you call the decorated function, it adds the `call` ops to the
default graph. In addition, it adds the definition of the function into the
default graph. Because the addition of the function into the graph
is deferred, the decorator can be used anywhere in the program.
Any variables created inside of the function are hoisted into the outer graph.
Note that the variables are created in the variable scope that was active
during the first call to the function. Subsequent function calls will refer to
the same set of variables.
Definitions of functions in a graph are frozen as soon as the graph is used to
create a session. However, new functions and new calls to existing functions
may be added to the graph, with the new functions themselves becoming
immediately frozen.
Example, but also see the [How To on functions](link_needed).
```python
# Defining the function.
@tf.Defun(tf.float32, tf.float32)
def MyFunc(x, y):
return x + y, x - y
# Building the graph.
a = tf.constant([1.0])
b = tf.constant([2.0])
c, d = MyFunc(a, b, name='mycall')
```"
4051,func_graph_from_py_func,tensorflow/tensorflow/python/framework/function.py,907,function,"Returns a _FuncGraph generated from `func`.
Args:
func: A Python callable which constructs a TF function body. The arguments
must correspond to `arg_types`. Returns a value or list/tuple of values.
No returned value can be None.
arg_names: A sequence of strings for the function argument names.
arg_types: A sequence of the function's argument types.
name: The function name. If None, the name is derived from `func`.
capture_by_value: boolean. If True, captured values will be copied into the
function body.
device: device name or function.
colocation_stack: A colocation stack (list) the _FuncGraph should use.
container: A container name the _FuncGraph should start with.
collections_ref: A reference to a collections dict the _FuncGraph should
use internally.
arg_shapes: A sequence of the function's argument shapes.
allowlisted_stateful_ops: A set of ops that if stateful we ignore and
re-create.
capture_resource_var_by_value: Boolean (defaults to True). If False,
captured resource variable returns the handle instead of value.
Returns:
A _FuncGraph.
Raises:
ValueError: if func returns None."
4052,from_library,tensorflow/tensorflow/python/framework/function.py,1138,function,"Creates _DefinedFunctions initialized from a FunctionDefLibrary proto.
This method handles assigning the correct gradient functions to each
function.
Args:
lib: a FunctionDefLibrary
Returns:
A list of _DefinedFunctions
Raises:
ValueError: `lib` is invalid"
4053,get_extra_vars,tensorflow/tensorflow/python/framework/function.py,1267,function,"Returns the captured variables by the function.
Returns:
If the default graph is being used to define a function, the
returned list of variables are those created inside the function
body so far. Otherwise, returns an empty list."
4054,get_extra_inputs,tensorflow/tensorflow/python/framework/function.py,1282,function,"Returns the captured input tensors by the function.
Returns:
If the default graph is being used to define a function, the
returned list of tensors are those accessed inside the function body
but defined outside the function body so far. Otherwise, returns an
empty list."
4055,get_extra_args,tensorflow/tensorflow/python/framework/function.py,1298,function,"Returns the corresponding function arguments for the captured inputs.
Returns:
If the default graph is being used to define a function, the
returned list of place holders are those used inside the function
body corresponding those returned by get_extra_inputs(). Otherwise,
returns an empty list."
4056,function_def_from_tf_function,tensorflow/tensorflow/python/framework/function.py,1346,function,Converts a SWIG-wrapped TF_Function* to a FunctionDef proto.
4057,function_def_to_graph,tensorflow/tensorflow/python/framework/function_def_to_graph.py,33,function,"Converts a FunctionDef to a FuncGraph (sub-class Graph).
The returned FuncGraph's `name`, `inputs` and `outputs` fields will be set.
The input tensors are represented as placeholders.
Note: `FuncGraph.inputs` and `FuncGraph.captures` are not set and may be set
by the caller.
Args:
fdef: FunctionDef.
input_shapes: Optional. A list of TensorShape objects of the shapes of
function inputs. Defaults to the function's ""_input_shapes"" attribute. If
specified, its length must match length of `fdef.signature.input_arg`. If
a shape is None, the corresponding input placeholder will have unknown
shape.
Returns:
A FuncGraph."
4058,is_function,tensorflow/tensorflow/python/framework/function_def_to_graph.py,107,function,Checks for a function definition with `fname` in the current context.
4059,function_def_to_graph_def,tensorflow/tensorflow/python/framework/function_def_to_graph.py,122,function,"Convert a FunctionDef to a GraphDef.
Steps:
1. Creates placeholder nodes corresponding to inputs in
`FunctionDef.signature.input_arg`.
2. Adds NodeDefs in `FunctionDef.node_def` to `GraphDef.node`.
3. Renames inputs of all nodes to use the convention of GraphDef instead of
FunctionDef. See comment on `FunctionDef.node_def` on how the tensor naming
in FunctionDefs is different from GraphDefs.
Args:
fdef: FunctionDef.
input_shapes: Optional. A list of TensorShape objects of the shapes of
function inputs. If specified, its length must match length of
`fdef.signature.input_arg`. If a shape is None, the corresponding input
placeholder will have unknown shape.
Returns:
A tuple of (GraphDef, dict<string, string>). The dict contains a mapping
from nested tensor names (in FunctionDef) to flattened names (in GraphDef).
Raises:
ValueError: If the length of input_shapes does not match the number of
input_args or if the FunctionDef is invalid."
4060,FunctionsFromProtos,tensorflow/tensorflow/python/framework/function_test.py,1168,class,
4061,expectFunctionsEqual,tensorflow/tensorflow/python/framework/function_test.py,1170,method,
4062,Foo,tensorflow/tensorflow/python/framework/function_test.py,1188,method,
4063,G,tensorflow/tensorflow/python/framework/function_test.py,1196,method,
4064,F,tensorflow/tensorflow/python/framework/function_test.py,1200,method,
4065,Foo,tensorflow/tensorflow/python/framework/function_test.py,1209,method,
4066,Outer,tensorflow/tensorflow/python/framework/function_test.py,1226,method,
4067,G1,tensorflow/tensorflow/python/framework/function_test.py,1242,method,
4068,G2,tensorflow/tensorflow/python/framework/function_test.py,1246,method,
4069,F1,tensorflow/tensorflow/python/framework/function_test.py,1251,method,
4070,F2,tensorflow/tensorflow/python/framework/function_test.py,1255,method,
4071,F3,tensorflow/tensorflow/python/framework/function_test.py,1260,method,
4072,F4,tensorflow/tensorflow/python/framework/function_test.py,1265,method,
4073,CheckNewFunc,tensorflow/tensorflow/python/framework/function_test.py,1281,method,
4074,G1,tensorflow/tensorflow/python/framework/function_test.py,1300,method,
4075,F1,tensorflow/tensorflow/python/framework/function_test.py,1304,method,
4076,F1,tensorflow/tensorflow/python/framework/function_test.py,1334,method,
4077,F2,tensorflow/tensorflow/python/framework/function_test.py,1338,method,
4078,FunctionWithStrAttr,tensorflow/tensorflow/python/framework/function_test.py,1363,method,
4079,FunctionWithIntAttr,tensorflow/tensorflow/python/framework/function_test.py,1367,method,
4080,FunctionWithFloatAttr,tensorflow/tensorflow/python/framework/function_test.py,1371,method,
4081,FunctionWithBoolAttr,tensorflow/tensorflow/python/framework/function_test.py,1375,method,
4082,FunctionWithStrAttr,tensorflow/tensorflow/python/framework/function_test.py,1395,method,
4083,Inner,tensorflow/tensorflow/python/framework/function_test.py,1229,method,
4084,compute_capability_from_device_desc,tensorflow/tensorflow/python/framework/gpu_util.py,35,function,"Returns the GpuInfo given a DeviceAttributes proto.
Args:
device_attrs: A DeviceAttributes proto.
Returns
A gpu_info tuple. Both fields are None if `device_attrs` does not have a
valid physical_device_desc field."
4085,run_benchmark,tensorflow/tensorflow/python/framework/graph_building_benchmark.py,44,function,
4086,SingleOpBenchmarks,tensorflow/tensorflow/python/framework/graph_building_benchmark.py,52,class,Benchmark for graph building time of ops.
4087,benchmarkAddScalars,tensorflow/tensorflow/python/framework/graph_building_benchmark.py,65,method,
4088,benchmarkAddBatchedMatrices,tensorflow/tensorflow/python/framework/graph_building_benchmark.py,75,method,
4089,benchmarkMatMul,tensorflow/tensorflow/python/framework/graph_building_benchmark.py,87,method,
4090,bench,tensorflow/tensorflow/python/framework/graph_building_benchmark.py,70,method,
4091,bench,tensorflow/tensorflow/python/framework/graph_building_benchmark.py,82,method,
4092,bench,tensorflow/tensorflow/python/framework/graph_building_benchmark.py,94,method,
4093,write_graph,tensorflow/tensorflow/python/framework/graph_io.py,31,function,"Writes a graph proto to a file.
The graph is written as a text proto unless `as_text` is `False`.
```python
v = tf.Variable(0, name='my_variable')
sess = tf.compat.v1.Session()
tf.io.write_graph(sess.graph_def, '/tmp/my-model', 'train.pbtxt')
```
or
```python
v = tf.Variable(0, name='my_variable')
sess = tf.compat.v1.Session()
tf.io.write_graph(sess.graph, '/tmp/my-model', 'train.pbtxt')
```
Args:
graph_or_graph_def: A `Graph` or a `GraphDef` protocol buffer.
logdir: Directory where to write the graph. This can refer to remote
filesystems, such as Google Cloud Storage (GCS).
name: Filename for the graph.
as_text: If `True`, writes the graph as an ASCII proto.
Returns:
The path of the output proto file."
4094,graph_to_function_def,tensorflow/tensorflow/python/framework/graph_to_function_def.py,122,function,"Returns `graph` as a `FunctionDef` protocol buffer.
This method creates a [`FunctionDef`](
https://www.tensorflow.org/code/tensorflow/core/framework/function.proto)
protocol buffer that contains all the ops in `operations`. The
operations become the body of the function.
The arguments `inputs` and `outputs` will be listed as the inputs
and outputs tensors of the function. They must be lists of
tensors present in the graph. The lists can optionally be empty.
Args:
graph: Graph.
operations: the operations to put in the function. Must be a subset of
the operations in the graph.
inputs: List of tensors. Inputs to the function.
outputs: List of tensors. Outputs of the function.
out_names: Optional list of string names for the outputs.
Returns:
A FunctionDef protocol buffer.
Raises:
ValueError: if out_names is specified and the wrong length."
4095,must_run_on_cpu,tensorflow/tensorflow/python/framework/graph_util_impl.py,71,function,"Returns True if the given node_def must run on CPU, otherwise False.
Args:
node: The node to be assigned to a device. Could be either an ops.Operation
or NodeDef.
pin_variables_on_cpu: If True, this function will return False if node_def
represents a variable-related op.
Returns:
True if the given node must run on CPU, otherwise False."
4096,extract_sub_graph,tensorflow/tensorflow/python/framework/graph_util_impl.py,187,function,"Extract the subgraph that can reach any of the nodes in 'dest_nodes'.
Args:
graph_def: A graph_pb2.GraphDef proto.
dest_nodes: A list of strings specifying the destination node names.
Returns:
The GraphDef of the sub-graph.
Raises:
TypeError: If 'graph_def' is not a graph_pb2.GraphDef proto."
4097,tensor_shape_from_node_def_name,tensorflow/tensorflow/python/framework/graph_util_impl.py,229,function,Convenience function to get a shape from a NodeDef's input string.
4098,convert_variables_to_constants,tensorflow/tensorflow/python/framework/graph_util_impl.py,247,function,"Replaces all the variables in a graph with constants of the same values.
If you have a trained graph containing Variable ops, it can be convenient to
convert them all to Const ops holding the same values. This makes it possible
to describe the network fully with a single GraphDef file, and allows the
removal of a lot of ops related to loading and saving the variables.
Args:
sess: Active TensorFlow session containing the variables.
input_graph_def: GraphDef object holding the network.
output_node_names: List of name strings for the result nodes of the graph.
variable_names_whitelist: The set of variable names to convert (by default,
all variables are converted).
variable_names_blacklist: The set of variable names to omit converting
to constants.
Returns:
GraphDef containing a simplified version of the original.
Raises:
RuntimeError: if a DT_RESOURCE op is found whose ancestor Variables are both
denylisted AND whitelisted for freezing."
4099,remove_training_nodes,tensorflow/tensorflow/python/framework/graph_util_impl.py,291,function,"Prunes out nodes that aren't needed for inference.
There are nodes like Identity and CheckNumerics that are only useful
during training, and can be removed in graphs that will be used for
nothing but inference. Here we identify and remove them, returning an
equivalent graph. To be specific, CheckNumerics nodes are always removed, and
Identity nodes that aren't involved in control edges are spliced out so that
their input and outputs are directly connected.
Args:
input_graph: Model to analyze and prune.
protected_nodes: An optional list of names of nodes to be kept
unconditionally. This is for example useful to preserve Identity output
nodes.
Returns:
A list of nodes with the unnecessary ones removed."
4100,import_graph_def,tensorflow/tensorflow/python/framework/importer.py,351,function,"Imports the graph from `graph_def` into the current default `Graph`.
This function provides a way to import a serialized TensorFlow
[`GraphDef`](https://www.tensorflow.org/code/tensorflow/core/framework/graph.proto)
protocol buffer, and extract individual objects in the `GraphDef` as
`tf.Tensor` and `tf.Operation` objects. Once extracted,
these objects are placed into the current default `Graph`. See
`tf.Graph.as_graph_def` for a way to create a `GraphDef`
proto.
Args:
graph_def: A `GraphDef` proto containing operations to be imported into
the default graph.
input_map: A dictionary mapping input names (as strings) in `graph_def`
to `Tensor` objects. The values of the named input tensors in the
imported graph will be re-mapped to the respective `Tensor` values.
return_elements: A list of strings containing operation names in
`graph_def` that will be returned as `Operation` objects; and/or
tensor names in `graph_def` that will be returned as `Tensor` objects.
name: (Optional.) A prefix that will be prepended to the names in
`graph_def`. Note that this does not apply to imported function names.
Defaults to `""import""`.
op_dict: (Optional.) Deprecated, do not use.
producer_op_list: (Optional.) An `OpList` proto with the (possibly stripped)
list of `OpDef`s used by the producer of the graph. If provided,
unrecognized attrs for ops in `graph_def` that have their default value
according to `producer_op_list` will be removed. This will allow some more
`GraphDef`s produced by later binaries to be accepted by earlier binaries.
Returns:
A list of `Operation` and/or `Tensor` objects from the imported graph,
corresponding to the names in `return_elements`,
and None if `returns_elements` is None.
Raises:
TypeError: If `graph_def` is not a `GraphDef` proto,
`input_map` is not a dictionary mapping strings to `Tensor` objects,
or `return_elements` is not a list of strings.
ValueError: If `input_map`, or `return_elements` contains names that
do not appear in `graph_def`, or `graph_def` is not well-formed (e.g.
it refers to an unknown tensor)."
4101,import_graph_def_for_function,tensorflow/tensorflow/python/framework/importer.py,408,function,Like import_graph_def but does not validate colocation constraints.
4102,IndexedSlices,tensorflow/tensorflow/python/framework/indexed_slices.py,61,class,"A sparse representation of a set of tensor slices at given indices.
This class is a simple wrapper for a pair of `Tensor` objects:
* `values`: A `Tensor` of any dtype with shape `[D0, D1, ..., Dn]`.
* `indices`: A 1-D integer `Tensor` with shape `[D0]`.
An `IndexedSlices` is typically used to represent a subset of a larger
tensor `dense` of shape `[LARGE0, D1, .. , DN]` where `LARGE0 >> D0`.
The values in `indices` are the indices in the first dimension of
the slices that have been extracted from the larger tensor.
The dense tensor `dense` represented by an `IndexedSlices` `slices` has
```python
dense[slices.indices[i], :, :, :, ...] = slices.values[i, :, :, :, ...]
```
The `IndexedSlices` class is used principally in the definition of
gradients for operations that have sparse gradients
(e.g. `tf.gather`).
Contrast this representation with
`tf.sparse.SparseTensor`,
which uses multi-dimensional indices and scalar values."
4103,values,tensorflow/tensorflow/python/framework/indexed_slices.py,96,method,A `Tensor` containing the values of the slices.
4104,indices,tensorflow/tensorflow/python/framework/indexed_slices.py,101,method,A 1-D `Tensor` containing the indices of the slices.
4105,dense_shape,tensorflow/tensorflow/python/framework/indexed_slices.py,106,method,A 1-D `Tensor` containing the shape of the corresponding dense tensor.
4106,shape,tensorflow/tensorflow/python/framework/indexed_slices.py,111,method,"Gets the `tf.TensorShape` representing the shape of the dense tensor.
Returns:
A `tf.TensorShape` object."
4107,name,tensorflow/tensorflow/python/framework/indexed_slices.py,123,method,The name of this `IndexedSlices`.
4108,device,tensorflow/tensorflow/python/framework/indexed_slices.py,128,method,"The name of the device on which `values` will be produced, or `None`."
4109,op,tensorflow/tensorflow/python/framework/indexed_slices.py,133,method,The `Operation` that produces `values` as an output.
4110,dtype,tensorflow/tensorflow/python/framework/indexed_slices.py,138,method,The `DType` of elements in this tensor.
4111,graph,tensorflow/tensorflow/python/framework/indexed_slices.py,143,method,"The `Graph` that contains the values, indices, and shape tensors."
4112,consumers,tensorflow/tensorflow/python/framework/indexed_slices.py,184,method,
4113,IndexedSlicesSpec,tensorflow/tensorflow/python/framework/indexed_slices.py,193,class,Type specification for a `tf.IndexedSlices`.
4114,convert_to_tensor_or_indexed_slices,tensorflow/tensorflow/python/framework/indexed_slices.py,260,function,"Converts the given object to a `Tensor` or an `IndexedSlices`.
If `value` is an `IndexedSlices` or `SparseTensor` it is returned
unmodified. Otherwise, it is converted to a `Tensor` using
`convert_to_tensor()`.
Args:
value: An `IndexedSlices`, `SparseTensor`, or an object that can be consumed
by `convert_to_tensor()`.
dtype: (Optional.) The required `DType` of the returned `Tensor` or
`IndexedSlices`.
name: (Optional.) A name to use if a new `Tensor` is created.
Returns:
A `Tensor`, `IndexedSlices`, or `SparseTensor` based on `value`.
Raises:
ValueError: If `dtype` does not match the element type of `value`."
4115,internal_convert_to_tensor_or_indexed_slices,tensorflow/tensorflow/python/framework/indexed_slices.py,284,function,"Converts the given object to a `Tensor` or an `IndexedSlices`.
If `value` is an `IndexedSlices` or `SparseTensor` it is returned
unmodified. Otherwise, it is converted to a `Tensor` using
`convert_to_tensor()`.
Args:
value: An `IndexedSlices`, `SparseTensor`, or an object that can be consumed
by `convert_to_tensor()`.
dtype: (Optional.) The required `DType` of the returned `Tensor` or
`IndexedSlices`.
name: (Optional.) A name to use if a new `Tensor` is created.
as_ref: True if the caller wants the results as ref tensors.
Returns:
A `Tensor`, `IndexedSlices`, or `SparseTensor` based on `value`.
Raises:
ValueError: If `dtype` does not match the element type of `value`."
4116,internal_convert_n_to_tensor_or_indexed_slices,tensorflow/tensorflow/python/framework/indexed_slices.py,321,function,"Converts `values` to a list of `Tensor` or `IndexedSlices` objects.
Any `IndexedSlices` or `SparseTensor` objects in `values` are returned
unmodified.
Args:
values: An iterable of `None`, `IndexedSlices`, `SparseTensor`, or objects
that can be consumed by `convert_to_tensor()`.
dtype: (Optional.) The required `DType` of the returned `Tensor` or
`IndexedSlices`.
name: (Optional.) A name prefix to used when a new `Tensor` is created, in
which case element `i` will be given the name `name + '_' + i`.
as_ref: True if the caller wants the results as ref tensors.
Returns:
A list of `Tensor`, `IndexedSlices`, `SparseTensor` and/or `None` objects.
Raises:
TypeError: If no conversion function is registered for an element in
`values`.
RuntimeError: If a registered conversion function returns an invalid
value."
4117,convert_n_to_tensor_or_indexed_slices,tensorflow/tensorflow/python/framework/indexed_slices.py,362,function,"Converts `values` to a list of `Output` or `IndexedSlices` objects.
Any `IndexedSlices` or `SparseTensor` objects in `values` are returned
unmodified.
Args:
values: A list of `None`, `IndexedSlices`, `SparseTensor`, or objects that
can be consumed by `convert_to_tensor()`.
dtype: (Optional.) The required `DType` of the returned `Tensor`
`IndexedSlices`.
name: (Optional.) A name prefix to used when a new `Tensor` is created, in
which case element `i` will be given the name `name + '_' + i`.
Returns:
A list of `Tensor`, `IndexedSlices`, and/or `SparseTensor` objects.
Raises:
TypeError: If no conversion function is registered for an element in
`values`.
RuntimeError: If a registered conversion function returns an invalid
value."
4118,is_mlir_bridge_enabled,tensorflow/tensorflow/python/framework/is_mlir_bridge_test_true.py,28,function,Returns true to if MLIR bridge should be enabled for tests.
4119,is_tfrt_enabled,tensorflow/tensorflow/python/framework/is_tfrt_test_true.py,28,function,Returns true to state TFRT should be enabled for Tensorflow tests.
4120,is_xla_enabled,tensorflow/tensorflow/python/framework/is_xla_test_true.py,27,function,Returns true to state XLA should be enabled for Tensorflow tests.
4121,get_all_registered_kernels,tensorflow/tensorflow/python/framework/kernels.py,26,function,"Returns a KernelList proto of all registered kernels.
"
4122,get_registered_kernels_for_op,tensorflow/tensorflow/python/framework/kernels.py,36,function,"Returns a KernelList proto of registered kernels for a given op.
Args:
name: A string representing the name of the op whose kernels to retrieve."
4123,GetRegisteredKernelsForOp,tensorflow/tensorflow/python/framework/kernels_test.py,32,class,
4124,load_op_library,tensorflow/tensorflow/python/framework/load_library.py,36,function,"Loads a TensorFlow plugin, containing custom ops and kernels.
Pass ""library_filename"" to a platform-specific mechanism for dynamically
loading a library. The rules for determining the exact location of the
library are platform-specific and are not documented here. When the
library is loaded, ops and kernels registered in the library via the
`REGISTER_*` macros are made available in the TensorFlow process. Note
that ops with the same name as an existing op are rejected and not
registered with the process.
Args:
library_filename: Path to the plugin.
Relative or absolute filesystem path to a dynamic library file.
Returns:
A python module containing the Python wrappers for Ops defined in
the plugin.
Raises:
RuntimeError: when unable to load the library or get the python wrappers."
4125,load_file_system_library,tensorflow/tensorflow/python/framework/load_library.py,83,function,"Loads a TensorFlow plugin, containing file system implementation.
Pass `library_filename` to a platform-specific mechanism for dynamically
loading a library. The rules for determining the exact location of the
library are platform-specific and are not documented here.
Args:
library_filename: Path to the plugin.
Relative or absolute filesystem path to a dynamic library file.
Returns:
None.
Raises:
RuntimeError: when unable to load the library."
4126,load_library,tensorflow/tensorflow/python/framework/load_library.py,124,function,"Loads a TensorFlow plugin.
""library_location"" can be a path to a specific shared object, or a folder.
If it is a folder, all shared objects that are named ""libtfkernel*"" will be
loaded. When the library is loaded, kernels registered in the library via the
`REGISTER_*` macros are made available in the TensorFlow process.
Args:
library_location: Path to the plugin or the folder of plugins.
Relative or absolute filesystem path to a dynamic library file or folder.
Returns:
None
Raises:
OSError: When the file to be loaded is not found.
RuntimeError: when unable to load the library."
4127,MemoryChecker,tensorflow/tensorflow/python/framework/memory_checker.py,47,class,"Memory leak detection class.
This is a utility class to detect Python and C++ memory leaks. It's intended
for both testing and debugging. Basic usage:
>>> # MemoryChecker() context manager tracks memory status inside its scope.
>>> with MemoryChecker() as memory_checker:
>>> tensors = []
>>> for _ in range(10):
>>> # Simulating `tf.constant(1)` object leak every iteration.
>>> tensors.append(tf.constant(1))
>>>
>>> # Take a memory snapshot for later analysis.
>>> memory_checker.record_snapshot()
>>>
>>> # `report()` generates a html graph file showing allocations over
>>> # snapshots per every stack trace.
>>> memory_checker.report()
>>>
>>> # This assertion will detect `tf.constant(1)` object leak.
>>> memory_checker.assert_no_leak_if_all_possibly_except_one()
`record_snapshot()` must be called once every iteration at the same location.
This is because the detection algorithm relies on the assumption that if there
is a leak, it's happening similarly on every snapshot."
4128,record_snapshot,tensorflow/tensorflow/python/framework/memory_checker.py,91,method,"Take a memory snapshot for later analysis.
`record_snapshot()` must be called once every iteration at the same
location. This is because the detection algorithm relies on the assumption
that if there is a leak, it's happening similarly on every snapshot.
The recommended number of `record_snapshot()` call depends on the testing
code complexity and the allcoation pattern."
4129,report,tensorflow/tensorflow/python/framework/memory_checker.py,106,method,"Generates a html graph file showing allocations over snapshots.
It create a temporary directory and put all the output files there.
If this is running under Google internal testing infra, it will use the
directory provided the infra instead."
4130,assert_no_leak_if_all_possibly_except_one,tensorflow/tensorflow/python/framework/memory_checker.py,118,method,"Raises an exception if a leak is detected.
This algorithm classifies a series of allocations as a leak if it's the same
type(Python) orit happens at the same stack trace(C++) at every snapshot,
but possibly except one snapshot."
4131,assert_no_new_python_objects,tensorflow/tensorflow/python/framework/memory_checker.py,131,method,"Raises an exception if there are new Python objects created.
It computes the number of new Python objects per type using the first and
the last snapshots.
Args:
threshold: A dictionary of [Type name string], [count] pair. It won't
raise an exception if the new Python objects are under this threshold."
4132,ops_used_by_graph_def,tensorflow/tensorflow/python/framework/meta_graph.py,138,function,"Collect the list of ops used by a graph.
Does not validate that the ops are all registered.
Args:
graph_def: A `GraphDef` proto, as from `graph.as_graph_def()`.
Returns:
A list of strings, each naming an op used by the graph."
4133,stripped_op_list_for_graph,tensorflow/tensorflow/python/framework/meta_graph.py,179,function,"Collect the stripped OpDefs for ops used by a graph.
This function computes the `stripped_op_list` field of `MetaGraphDef` and
similar protos. The result can be communicated from the producer to the
consumer, which can then use the C++ function
`RemoveNewDefaultAttrsFromGraphDef` to improve forwards compatibility.
Args:
graph_def: A `GraphDef` proto, as from `graph.as_graph_def()`.
Returns:
An `OpList` of ops used by the graph."
4134,add_collection_def,tensorflow/tensorflow/python/framework/meta_graph.py,374,function,"Adds a collection to MetaGraphDef protocol buffer.
Args:
meta_graph_def: MetaGraphDef protocol buffer.
key: One of the GraphKeys or user-defined string.
graph: The `Graph` from which to get collections.
export_scope: Optional `string`. Name scope to remove.
exclude_nodes: An iterable of nodes or `string` node names to omit from the
collection, or None.
override_contents: An iterable of values to place in the collection,
ignoring the current values (if set)."
4135,strip_graph_default_valued_attrs,tensorflow/tensorflow/python/framework/meta_graph.py,462,function,"Strips default valued attributes for node defs in given MetaGraphDef.
This method also sets `meta_info_def.stripped_default_attrs` in the given
`MetaGraphDef` proto to True.
Args:
meta_graph_def: `MetaGraphDef` protocol buffer
Returns:
None."
4136,create_meta_graph_def,tensorflow/tensorflow/python/framework/meta_graph.py,509,function,"Construct and returns a `MetaGraphDef` protocol buffer.
Args:
meta_info_def: `MetaInfoDef` protocol buffer.
graph_def: `GraphDef` protocol buffer.
saver_def: `SaverDef` protocol buffer.
collection_list: List of string keys to collect.
graph: The `Graph` to create `MetaGraphDef` out of.
export_scope: Optional `string`. Name scope to remove.
exclude_nodes: An iterable of nodes or `string` node names to omit from all
collection, or None.
clear_extraneous_savers: Remove any preexisting SaverDefs from the SAVERS
collection. Note this method does not alter the graph, so any
extraneous Save/Restore ops should have been removed already, as needed.
strip_default_attrs: Boolean. If `True`, default-valued attributes will be
removed from the NodeDefs. For a detailed guide, see
[Stripping Default-Valued Attributes](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/saved_model/README.md#stripping-default-valued-attributes).
Returns:
MetaGraphDef protocol buffer.
Raises:
TypeError: If the arguments are not of the correct proto buffer type."
4137,read_meta_graph_file,tensorflow/tensorflow/python/framework/meta_graph.py,616,function,"Reads a file containing `MetaGraphDef` and returns the protocol buffer.
Args:
filename: `meta_graph_def` filename including the path.
Returns:
A `MetaGraphDef` protocol buffer.
Raises:
IOError: If the file doesn't exist, or cannot be successfully parsed."
4138,import_scoped_meta_graph,tensorflow/tensorflow/python/framework/meta_graph.py,648,function,"Recreates a `Graph` saved in a `MetaGraphDef` proto.
This function takes a `MetaGraphDef` protocol buffer as input. If
the argument is a file containing a `MetaGraphDef` protocol buffer ,
it constructs a protocol buffer from the file content. The function
then adds all the nodes from the `graph_def` field to the
current graph, recreates the desired collections, and returns a dictionary of
all the Variables imported into the name scope.
In combination with `export_scoped_meta_graph()`, this function can be used to
* Serialize a graph along with other Python objects such as `QueueRunner`,
`Variable` into a `MetaGraphDef`.
* Restart training from a saved graph and checkpoints.
* Run inference from a saved graph and checkpoints.
Args:
meta_graph_or_file: `MetaGraphDef` protocol buffer or filename (including
the path) containing a `MetaGraphDef`.
clear_devices: Boolean which controls whether to clear device information
from graph_def. Default false.
graph: The `Graph` to import into. If `None`, use the default graph.
import_scope: Optional `string`. Name scope into which to import the
subgraph. If `None`, the graph is imported to the root name scope.
input_map: A dictionary mapping input names (as strings) in `graph_def` to
`Tensor` objects. The values of the named input tensors in the imported
graph will be re-mapped to the respective `Tensor` values.
unbound_inputs_col_name: Collection name for looking up unbound inputs.
restore_collections_predicate: a predicate on collection names. A collection
named c (i.e whose key is c) will be restored iff
1) `restore_collections_predicate(c)` is True, and
2) `c != unbound_inputs_col_name`.
Returns:
A dictionary of all the `Variables` imported into the name scope.
Raises:
ValueError: If the graph_def contains unbound inputs."
4139,import_scoped_meta_graph_with_return_elements,tensorflow/tensorflow/python/framework/meta_graph.py,701,function,"Imports graph from `MetaGraphDef` and returns vars and return elements.
This function takes a `MetaGraphDef` protocol buffer as input. If
the argument is a file containing a `MetaGraphDef` protocol buffer ,
it constructs a protocol buffer from the file content. The function
then adds all the nodes from the `graph_def` field to the
current graph, recreates the desired collections, and returns a dictionary of
all the Variables imported into the name scope.
In combination with `export_scoped_meta_graph()`, this function can be used to
* Serialize a graph along with other Python objects such as `QueueRunner`,
`Variable` into a `MetaGraphDef`.
* Restart training from a saved graph and checkpoints.
* Run inference from a saved graph and checkpoints.
Args:
meta_graph_or_file: `MetaGraphDef` protocol buffer or filename (including
the path) containing a `MetaGraphDef`.
clear_devices: Boolean which controls whether to clear device information
from graph_def. Default false.
graph: The `Graph` to import into. If `None`, use the default graph.
import_scope: Optional `string`. Name scope into which to import the
subgraph. If `None`, the graph is imported to the root name scope.
input_map: A dictionary mapping input names (as strings) in `graph_def` to
`Tensor` objects. The values of the named input tensors in the imported
graph will be re-mapped to the respective `Tensor` values.
unbound_inputs_col_name: Collection name for looking up unbound inputs.
restore_collections_predicate: a predicate on collection names. A collection
named c (i.e whose key is c) will be restored iff
1) `restore_collections_predicate(c)` is True, and
2) `c != unbound_inputs_col_name`.
return_elements: A list of strings containing operation names in the
`MetaGraphDef` that will be returned as `Operation` objects; and/or
tensor names in `MetaGraphDef` that will be returned as `Tensor` objects.
Returns:
A tuple of (
dictionary of all the `Variables` imported into the name scope,
list of `Operation` or `Tensor` objects from the `return_elements` list).
Raises:
ValueError: If the graph_def contains unbound inputs."
4140,export_scoped_meta_graph,tensorflow/tensorflow/python/framework/meta_graph.py,908,function,"Returns `MetaGraphDef` proto. Optionally writes it to filename.
This function exports the graph, saver, and collection objects into
`MetaGraphDef` protocol buffer with the intention of it being imported
at a later time or location to restart training, run inference, or be
a subgraph.
Args:
filename: Optional filename including the path for writing the
generated `MetaGraphDef` protocol buffer.
graph_def: `GraphDef` protocol buffer.
graph: The `Graph` to export. If `None`, use the default graph.
export_scope: Optional `string`. Name scope under which to extract
the subgraph. The scope name will be stripped from the node definitions
for easy import later into new name scopes. If `None`, the whole graph
is exported.
as_text: If `True`, writes the `MetaGraphDef` as an ASCII proto.
unbound_inputs_col_name: Optional `string`. If provided, a string collection
with the given name will be added to the returned `MetaGraphDef`,
containing the names of tensors that must be remapped when importing the
`MetaGraphDef`.
clear_devices: Boolean which controls whether to clear device information
before exporting the graph.
saver_def: `SaverDef` protocol buffer.
clear_extraneous_savers: Remove any Saver-related information from the
graph (both Save/Restore ops and SaverDefs) that are not associated
with the provided SaverDef.
strip_default_attrs: Set to true if default valued attributes must be
removed while exporting the GraphDef.
save_debug_info: If `True`, save the GraphDebugInfo to a separate file,
which in the same directory of filename and with `_debug` added before the
file extension.
**kwargs: Optional keyed arguments, including meta_info_def and
collection_list.
Returns:
A `MetaGraphDef` proto and dictionary of `Variables` in the exported
name scope.
Raises:
ValueError: When the `GraphDef` is larger than 2GB.
ValueError: When executing in Eager mode and either `graph_def` or `graph`
is undefined."
4141,copy_scoped_meta_graph,tensorflow/tensorflow/python/framework/meta_graph.py,1073,function,"Copies a sub-meta_graph from one scope to another.
Args:
from_scope: `String` name scope containing the subgraph to be copied.
to_scope: `String` name scope under which the copied subgraph will reside.
from_graph: Optional `Graph` from which to copy the subgraph. If `None`, the
default graph is use.
to_graph: Optional `Graph` to which to copy the subgraph. If `None`, the
default graph is used.
Returns:
A dictionary of `Variables` that has been copied into `to_scope`.
Raises:
ValueError: If `from_scope` and `to_scope` are the same while
`from_graph` and `to_graph` are also the same."
4142,add_op_callback,tensorflow/tensorflow/python/framework/op_callbacks.py,25,function,"Add a thread-local callback that intercepts op execution and op creation.
The `callback_fn` will be invoked immediately after any of the three types
of events:
- The execution of an TensorFlow operation (""op"" for short hereafter)
under eager mode,
- The execution of a FuncGraph under eager mode,
- The creation of an op during graph construction (e.g., in
@tf.function-decorated Python functions).
Known limitations:
1. Under graph mode, overriding the output tensors of control-flow ops,
including ""If"", ""StatelessIf"" and ""While"", may cause errors
(b/139668453). Overriding other tensors in a graph consisting of such
control-flow ops is okay.
2. Under eager mode, calling eager ops from the callback function itself
may lead to recursion stack overflow. This can be prevented by
returning from the callback function immediately on encountering the
op type involved (b/140334369).
Args:
callback_fn: A callback_fn that has the following signature:
def callback_fn(op_type,
inputs,
attrs,
outputs,
op_name=None,
graph=None):
# op_type: The type of the op, as a string. E.g., ""MatMul"".
# For the special case of FuncGraph execution, op_type
# takes the name of the graph name, e.g.,
# ""__inference_my_func_24"".
# inputs: (`tuple` of `Tensor`s) Input tensors to the op or the
# FuncGraph.
# - In eager execution, these are `EagerTensor`s.
# - In graph construction, these are non-eager `Tensor`s
# that form the inputs to the just-created op.
# attrs: The attributes of the op or FuncGraph of which the execution
# or creation caused the current invocation of the callback.
# This is applicable to both eager- and graph-based execution,
# as well as graph construction.
# This is a tuple of alternating attribute keys and attribute
# values. E.g., `('adjoint_a', False, 'adjoint_b', False)`.
# outputs: (`tuple of `Tensor`s) Output tensors from the op or
# FuncGraph.
# In eager execution, these are `EagerTensor`s.
# In graph construction, these are non-eager `Tensor`s that
# are the outputs of the just-created op.
# op_name: Name of the op.
# - If the current invocation of the callback is due to the
# eager execution of an op or FuncGraph, this will be
# `None`, as op names are meaningless in eager execution.
# - In graph construction, this is the name of the op, e.g.,
# ""MatMul_2"".
# graph: The graph that the op belongs to (if any).
# - In eager execution of an op or FuncGraph, this is `None`.
# - In graph construction, this is the op's enclosing graph
# as a `tf.Graph` object.
#
# Return values:
# This callback function is expected to return `None` or
# a `list` or `tuple` of `Tensor`s with its length matching
# `len(outputs)`, in the order that corresponds to that of the
# `outputs` argument.
# If the return value is `None`, downstream execution or graph
# construction will be unaffected.
# However, if the return value is a `list` or `tuple` of `Tensor`s,
# - In eager execution, these returned `Tensor`s should be
# `EagerTensor`s. Their values will replace the original values of
# `outputs` for downstream eager execution. (*Not implemented yet*).
# - In graph construction, these returned `Tensor`s should be
# non-eager `Tensor`s. Their values will replace the original
# `outputs` for downstream graph construction.
Raises:
ValueEror: If `callback_fn` is `None` or not callable."
4143,should_invoke_op_callbacks,tensorflow/tensorflow/python/framework/op_callbacks.py,118,function,"Determine if op callbacks are present and should be invoked.
Returns:
A thread-local result (boolean) indicating whether any op callback(s) exist
and should be invoked."
4144,remove_op_callback,tensorflow/tensorflow/python/framework/op_callbacks.py,129,function,"Remove an already-added op callback.
Args:
op_callback: The op callback to be removed.
Raises:
KeyError: If `op_callback` has not been registered using `add_op_callback()`
before."
4145,clear_op_callbacks,tensorflow/tensorflow/python/framework/op_callbacks.py,146,function,Clear all op callbacks registered in the current thread.
4146,invoke_op_callbacks,tensorflow/tensorflow/python/framework/op_callbacks.py,152,function,"Invoke the callbacks that exist in the current scope (if any).
If no callbacks are present in the current scope, this method returns
immediately.
Args:
op_type: Type of the operation (e.g., ""MatMul"").
inputs: Input tensors to the op. These are `EagerTensor`s in the case of
eager execution of ops or `FuncGraph`s, and are non-eager `Tensor`s in the
case of graph construction.
attrs: Attributes of the op, as `tuple` of alternating keys and values.
outputs: Output tensors from the op. These are `EagerTensor`s in the case of
eager execution and are non-eager `Tensor`s in the case of graph
construction.
op_name: Name of the op. Applicable if and only if this method is invoked
due to the graph construction of an op or the eager execution of of a
`FuncGraph`.
graph: The graph involved (if any).
- In the case if the eager execution of an op or FuncGraph, this is
`None`.
- In the case of the graph construction of an op, this is the `tf.Graph`
object being built.
Returns:
`None`, or a `list` or `tuple` of output tenors that will override the
original (input) `outputs`."
4147,apply_op,tensorflow/tensorflow/python/framework/op_def_library.py,255,function,"Add a node invoking a registered Op to a graph.
Example usage:
# input1 and input2 can be Tensors or anything ops.convert_to_tensor()
# will convert to a Tensor.
op_def_library.apply_op(""op"", input1=input1, input2=input2)
# Can specify a node name.
op_def_library.apply_op(""op"", input1=input1, name=""node_name"")
# Must use keyword arguments, with the names specified in the OpDef.
op_def_library.apply_op(""op"", input_name=input, attr_name=attr)
All attrs must either be inferred from an input or specified.
(If inferred, the attr must not be specified.) If an attr has a default
value specified in the Op's OpDef, then you may pass None as the value
of that attr to get the default.
Args:
op_type_name: string. Must match the name field of a registered Op.
name: string. Optional name of the created op.
**keywords: input Tensor and attr arguments specified by name,
and optional parameters to pass when constructing the Operation.
Returns:
The Tensor(s) representing the output of the operation, or the Operation
itself if there are no outputs.
Raises:
RuntimeError: On some errors.
TypeError: On some errors.
ValueError: On some errors."
4148,get,tensorflow/tensorflow/python/framework/op_def_registry.py,34,function,Returns an OpDef for a given `name` or None if the lookup fails.
4149,sync,tensorflow/tensorflow/python/framework/op_def_registry.py,59,function,No-op. Used to synchronize the contents of the Python registry with C++.
4150,tensor_id,tensorflow/tensorflow/python/framework/ops.py,114,function,Returns a unique identifier for this Tensor.
4151,NullContextmanager,tensorflow/tensorflow/python/framework/ops.py,166,class,
4152,is_dense_tensor_like,tensorflow/tensorflow/python/framework/ops.py,213,function,
4153,uid,tensorflow/tensorflow/python/framework/ops.py,217,function,A unique (within this program execution) integer.
4154,numpy_text,tensorflow/tensorflow/python/framework/ops.py,222,function,Human readable representation of a tensor's numpy value.
4155,enable_tensor_equality,tensorflow/tensorflow/python/framework/ops.py,235,function,"Compare Tensors with element-wise comparison and thus be unhashable.
Comparing tensors with element-wise allows comparisons such as
tf.Variable(1.0) == 1.0. Element-wise equality implies that tensors are
unhashable. Thus tensors can no longer be directly used in sets or as a key in
a dictionary."
4156,disable_tensor_equality,tensorflow/tensorflow/python/framework/ops.py,248,function,"Compare Tensors by their id and be hashable.
This is a legacy behaviour of TensorFlow and is highly discouraged."
4157,Tensor,tensorflow/tensorflow/python/framework/ops.py,259,class,"A tensor is a multidimensional array of elements represented by a
`tf.Tensor` object. All elements are of a single known data type.
When writing a TensorFlow program, the main object that is
manipulated and passed around is the `tf.Tensor`.
A `tf.Tensor` has the following properties:
* a single data type (float32, int32, or string, for example)
* a shape
TensorFlow supports eager execution and graph execution. In eager
execution, operations are evaluated immediately. In graph
execution, a computational graph is constructed for later
evaluation.
TensorFlow defaults to eager execution. In the example below, the
matrix multiplication results are calculated immediately.
>>> # Compute some values using a Tensor
>>> c = tf.constant([[1.0, 2.0], [3.0, 4.0]])
>>> d = tf.constant([[1.0, 1.0], [0.0, 1.0]])
>>> e = tf.matmul(c, d)
>>> print(e)
tf.Tensor(
[[1. 3.]
[3. 7.]], shape=(2, 2), dtype=float32)
Note that during eager execution, you may discover your `Tensors` are actually
of type `EagerTensor`. This is an internal detail, but it does give you
access to a useful function, `numpy`:
>>> type(e)
<class '...ops.EagerTensor'>
>>> print(e.numpy())
[[1. 3.]
[3. 7.]]
In TensorFlow, `tf.function`s are a common way to define graph execution.
A Tensor's shape (that is, the rank of the Tensor and the size of
each dimension) may not always be fully known. In `tf.function`
definitions, the shape may only be partially known.
Most operations produce tensors of fully-known shapes if the shapes of their
inputs are also fully known, but in some cases it's only possible to find the
shape of a tensor at execution time.
A number of specialized tensors are available: see `tf.Variable`,
`tf.constant`, `tf.placeholder`, `tf.sparse.SparseTensor`, and
`tf.RaggedTensor`.
For more on Tensors, see the [guide](https://tensorflow.org/guide/tensor)."
4158,op,tensorflow/tensorflow/python/framework/ops.py,394,method,The `Operation` that produces this tensor as an output.
4159,dtype,tensorflow/tensorflow/python/framework/ops.py,399,method,The `DType` of elements in this tensor.
4160,graph,tensorflow/tensorflow/python/framework/ops.py,404,method,The `Graph` that contains this tensor.
4161,name,tensorflow/tensorflow/python/framework/ops.py,409,method,The string name of this tensor.
4162,device,tensorflow/tensorflow/python/framework/ops.py,418,method,"The name of the device on which this tensor will be produced, or None."
4163,shape,tensorflow/tensorflow/python/framework/ops.py,423,method,"Returns a `tf.TensorShape` that represents the shape of this tensor.
>>> t = tf.constant([1,2,3,4,5])
>>> t.shape
TensorShape([5])
`tf.Tensor.shape` is equivalent to `tf.Tensor.get_shape()`.
In a `tf.function` or when building a model using
`tf.keras.Input`, they return the build-time shape of the
tensor, which may be partially unknown.
A `tf.TensorShape` is not a tensor. Use `tf.shape(t)` to get a tensor
containing the shape, calculated at runtime.
See `tf.Tensor.get_shape()`, and `tf.TensorShape` for details and examples."
4164,get_shape,tensorflow/tensorflow/python/framework/ops.py,536,method,"Returns a `tf.TensorShape` that represents the shape of this tensor.
In eager execution the shape is always fully-known.
>>> a = tf.constant([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])
>>> print(a.shape)
(2, 3)
`tf.Tensor.get_shape()` is equivalent to `tf.Tensor.shape`.
When executing in a `tf.function` or building a model using
`tf.keras.Input`, `Tensor.shape` may return a partial shape (including
`None` for unknown dimensions). See `tf.TensorShape` for more details.
>>> inputs = tf.keras.Input(shape = [10])
>>> # Unknown batch size
>>> print(inputs.shape)
(None, 10)
The shape is computed using shape inference functions that are
registered for each `tf.Operation`.
The returned `tf.TensorShape` is determined at *build* time, without
executing the underlying kernel. It is not a `tf.Tensor`. If you need a
shape *tensor*, either convert the `tf.TensorShape` to a `tf.constant`, or
use the `tf.shape(tensor)` function, which returns the tensor's shape at
*execution* time.
This is useful for debugging and providing early errors. For
example, when tracing a `tf.function`, no ops are being executed, shapes
may be unknown (See the [Concrete Functions
Guide](https://www.tensorflow.org/guide/concrete_function) for details).
>>> @tf.function
... def my_matmul(a, b):
... result = a@b
... # the `print` executes during tracing.
... print(""Result shape: "", result.shape)
... return result
The shape inference functions propagate shapes to the extent possible:
>>> f = my_matmul.get_concrete_function(
... tf.TensorSpec([None,3]),
... tf.TensorSpec([3,5]))
Result shape: (None, 5)
Tracing may fail if a shape missmatch can be detected:
>>> cf = my_matmul.get_concrete_function(
... tf.TensorSpec([None,3]),
... tf.TensorSpec([4,5]))
Traceback (most recent call last):
...
ValueError: Dimensions must be equal, but are 3 and 4 for 'matmul' (op:
'MatMul') with input shapes: [?,3], [4,5].
In some cases, the inferred shape may have unknown dimensions. If
the caller has additional information about the values of these
dimensions, `Tensor.set_shape()` can be used to augment the
inferred shape.
>>> @tf.function
... def my_fun(a):
... a.set_shape([5, 5])
... # the `print` executes during tracing.
... print(""Result shape: "", a.shape)
... return a
>>> cf = my_fun.get_concrete_function(
... tf.TensorSpec([None, None]))
Result shape: (5, 5)
Returns:
A `tf.TensorShape` representing the shape of this tensor."
4165,set_shape,tensorflow/tensorflow/python/framework/ops.py,617,method,"Updates the shape of this tensor.
With eager execution this operates as a shape assertion.
Here the shapes match:
>>> t = tf.constant([[1,2,3]])
>>> t.set_shape([1, 3])
Passing a `None` in the new shape allows any value for that axis:
>>> t.set_shape([1,None])
An error is raised if an incompatible shape is passed.
>>> t.set_shape([1,5])
Traceback (most recent call last):
...
ValueError: Tensor's shape (1, 3) is not compatible with supplied
shape [1, 5]
When executing in a `tf.function`, or building a model using
`tf.keras.Input`, `Tensor.set_shape` will *merge* the given `shape` with
the current shape of this tensor, and set the tensor's shape to the
merged value (see `tf.TensorShape.merge_with` for details):
>>> t = tf.keras.Input(shape=[None, None, 3])
>>> print(t.shape)
(None, None, None, 3)
Dimensions set to `None` are not updated:
>>> t.set_shape([None, 224, 224, None])
>>> print(t.shape)
(None, 224, 224, 3)
The main use case for this is to provide additional shape information
that cannot be inferred from the graph alone.
For example if you know all the images in a dataset have shape [28,28,3] you
can set it with `tf.set_shape`:
>>> @tf.function
... def load_image(filename):
... raw = tf.io.read_file(filename)
... image = tf.image.decode_png(raw, channels=3)
... # the `print` executes during tracing.
... print(""Initial shape: "", image.shape)
... image.set_shape([28, 28, 3])
... print(""Final shape: "", image.shape)
... return image
Trace the function, see the [Concrete Functions
Guide](https://www.tensorflow.org/guide/concrete_function) for details.
>>> cf = load_image.get_concrete_function(
... tf.TensorSpec([], dtype=tf.string))
Initial shape: (None, None, 3)
Final shape: (28, 28, 3)
Similarly the `tf.io.parse_tensor` function could return a tensor with
any shape, even the `tf.rank` is unknown. If you know that all your
serialized tensors will be 2d, set it with `set_shape`:
>>> @tf.function
... def my_parse(string_tensor):
... result = tf.io.parse_tensor(string_tensor, out_type=tf.float32)
... # the `print` executes during tracing.
... print(""Initial shape: "", result.shape)
... result.set_shape([None, None])
... print(""Final shape: "", result.shape)
... return result
Trace the function
>>> concrete_parse = my_parse.get_concrete_function(
... tf.TensorSpec([], dtype=tf.string))
Initial shape: <unknown>
Final shape: (None, None)
Make sure it works:
>>> t = tf.ones([5,3], dtype=tf.float32)
>>> serialized = tf.io.serialize_tensor(t)
>>> print(serialized.dtype)
<dtype: 'string'>
>>> print(serialized.shape)
()
>>> t2 = concrete_parse(serialized)
>>> print(t2.shape)
(5, 3)
Caution: `set_shape` ensures that the applied shape is compatible with
the existing shape, but it does not check at runtime. Setting
incorrect shapes can result in inconsistencies between the
statically-known graph and the runtime value of tensors. For runtime
validation of the shape, use `tf.ensure_shape` instead. It also modifies
the `shape` of the tensor.
>>> # Serialize a rank-3 tensor
>>> t = tf.ones([5,5,5], dtype=tf.float32)
>>> serialized = tf.io.serialize_tensor(t)
>>> # The function still runs, even though it `set_shape([None,None])`
>>> t2 = concrete_parse(serialized)
>>> print(t2.shape)
(5, 5, 5)
Args:
shape: A `TensorShape` representing the shape of this tensor, a
`TensorShapeProto`, a list, a tuple, or None.
Raises:
ValueError: If `shape` is not compatible with the current shape of
this tensor."
4166,value_index,tensorflow/tensorflow/python/framework/ops.py,759,method,The index of this tensor in the outputs of its `Operation`.
4167,consumers,tensorflow/tensorflow/python/framework/ops.py,763,method,"Returns a list of `Operation`s that consume this tensor.
Returns:
A list of `Operation`s."
4168,eval,tensorflow/tensorflow/python/framework/ops.py,891,method,"Evaluates this tensor in a `Session`.
Note: If you are not using `compat.v1` libraries, you should not need this,
(or `feed_dict` or `Session`). In eager execution (or within `tf.function`)
you do not need to call `eval`.
Calling this method will execute all preceding operations that
produce the inputs needed for the operation that produces this
tensor.
*N.B.* Before invoking `Tensor.eval()`, its graph must have been
launched in a session, and either a default session must be
available, or `session` must be specified explicitly.
Args:
feed_dict: A dictionary that maps `Tensor` objects to feed values. See
`tf.Session.run` for a description of the valid feed values.
session: (Optional.) The `Session` to be used to evaluate this tensor. If
none, the default session will be used.
Returns:
A numpy array corresponding to the value of this tensor."
4169,experimental_ref,tensorflow/tensorflow/python/framework/ops.py,918,method,
4170,ref,tensorflow/tensorflow/python/framework/ops.py,921,method,"Returns a hashable reference object to this Tensor.
The primary use case for this API is to put tensors in a set/dictionary.
We can't put tensors in a set/dictionary as `tensor.__hash__()` is no longer
available starting Tensorflow 2.0.
The following will raise an exception starting 2.0
>>> x = tf.constant(5)
>>> y = tf.constant(10)
>>> z = tf.constant(10)
>>> tensor_set = {x, y, z}
Traceback (most recent call last):
...
TypeError: Tensor is unhashable. Instead, use tensor.ref() as the key.
>>> tensor_dict = {x: 'five', y: 'ten'}
Traceback (most recent call last):
...
TypeError: Tensor is unhashable. Instead, use tensor.ref() as the key.
Instead, we can use `tensor.ref()`.
>>> tensor_set = {x.ref(), y.ref(), z.ref()}
>>> x.ref() in tensor_set
True
>>> tensor_dict = {x.ref(): 'five', y.ref(): 'ten', z.ref(): 'ten'}
>>> tensor_dict[y.ref()]
'ten'
Also, the reference object provides `.deref()` function that returns the
original Tensor.
>>> x = tf.constant(5)
>>> x.ref().deref()
<tf.Tensor: shape=(), dtype=int32, numpy=5>"
4171,convert_to_tensor_v1_with_dispatch,tensorflow/tensorflow/python/framework/ops.py,1263,function,"Converts the given `value` to a `Tensor`.
This function converts Python objects of various types to `Tensor`
objects. It accepts `Tensor` objects, numpy arrays, Python lists,
and Python scalars. For example:
```python
import numpy as np
def my_func(arg):
arg = tf.convert_to_tensor(arg, dtype=tf.float32)
return tf.matmul(arg, arg) + arg
# The following calls are equivalent.
value_1 = my_func(tf.constant([[1.0, 2.0], [3.0, 4.0]]))
value_2 = my_func([[1.0, 2.0], [3.0, 4.0]])
value_3 = my_func(np.array([[1.0, 2.0], [3.0, 4.0]], dtype=np.float32))
```
This function can be useful when composing a new operation in Python
(such as `my_func` in the example above). All standard Python op
constructors apply this function to each of their Tensor-valued
inputs, which allows those ops to accept numpy arrays, Python lists,
and scalars in addition to `Tensor` objects.
Note: This function diverges from default Numpy behavior for `float` and
`string` types when `None` is present in a Python list or scalar. Rather
than silently converting `None` values, an error will be thrown.
Args:
value: An object whose type has a registered `Tensor` conversion function.
dtype: Optional element type for the returned tensor. If missing, the type
is inferred from the type of `value`.
name: Optional name to use if a new `Tensor` is created.
preferred_dtype: Optional element type for the returned tensor, used when
dtype is None. In some cases, a caller may not have a dtype in mind when
converting to a tensor, so preferred_dtype can be used as a soft
preference. If the conversion to `preferred_dtype` is not possible, this
argument has no effect.
dtype_hint: same meaning as preferred_dtype, and overrides it.
Returns:
A `Tensor` based on `value`.
Raises:
TypeError: If no conversion function is registered for `value` to `dtype`.
RuntimeError: If a registered conversion function returns an invalid value.
ValueError: If the `value` is a tensor not of given `dtype` in graph mode."
4172,convert_to_tensor_v1,tensorflow/tensorflow/python/framework/ops.py,1323,function,Converts the given `value` to a `Tensor` (with the TF1 API).
4173,convert_to_tensor_v2_with_dispatch,tensorflow/tensorflow/python/framework/ops.py,1336,function,"Converts the given `value` to a `Tensor`.
This function converts Python objects of various types to `Tensor`
objects. It accepts `Tensor` objects, numpy arrays, Python lists,
and Python scalars.
For example:
>>> import numpy as np
>>> def my_func(arg):
... arg = tf.convert_to_tensor(arg, dtype=tf.float32)
... return arg
>>> # The following calls are equivalent.
...
>>> value_1 = my_func(tf.constant([[1.0, 2.0], [3.0, 4.0]]))
>>> print(value_1)
tf.Tensor(
[[1. 2.]
[3. 4.]], shape=(2, 2), dtype=float32)
>>> value_2 = my_func([[1.0, 2.0], [3.0, 4.0]])
>>> print(value_2)
tf.Tensor(
[[1. 2.]
[3. 4.]], shape=(2, 2), dtype=float32)
>>> value_3 = my_func(np.array([[1.0, 2.0], [3.0, 4.0]], dtype=np.float32))
>>> print(value_3)
tf.Tensor(
[[1. 2.]
[3. 4.]], shape=(2, 2), dtype=float32)
This function can be useful when composing a new operation in Python
(such as `my_func` in the example above). All standard Python op
constructors apply this function to each of their Tensor-valued
inputs, which allows those ops to accept numpy arrays, Python lists,
and scalars in addition to `Tensor` objects.
Note: This function diverges from default Numpy behavior for `float` and
`string` types when `None` is present in a Python list or scalar. Rather
than silently converting `None` values, an error will be thrown.
Args:
value: An object whose type has a registered `Tensor` conversion function.
dtype: Optional element type for the returned tensor. If missing, the type
is inferred from the type of `value`.
dtype_hint: Optional element type for the returned tensor, used when dtype
is None. In some cases, a caller may not have a dtype in mind when
converting to a tensor, so dtype_hint can be used as a soft preference.
If the conversion to `dtype_hint` is not possible, this argument has no
effect.
name: Optional name to use if a new `Tensor` is created.
Returns:
A `Tensor` based on `value`.
Raises:
TypeError: If no conversion function is registered for `value` to `dtype`.
RuntimeError: If a registered conversion function returns an invalid value.
ValueError: If the `value` is a tensor not of given `dtype` in graph mode."
4174,convert_to_tensor_v2,tensorflow/tensorflow/python/framework/ops.py,1402,function,Converts the given `value` to a `Tensor`.
4175,pack_eager_tensors,tensorflow/tensorflow/python/framework/ops.py,1416,function,"Pack multiple `EagerTensor`s of the same dtype and shape.
Args:
tensors: a list of EagerTensors to pack.
ctx: context.context().
Returns:
A packed EagerTensor."
4176,convert_to_tensor,tensorflow/tensorflow/python/framework/ops.py,1475,function,Implementation of the public convert_to_tensor.
4177,internal_convert_n_to_tensor,tensorflow/tensorflow/python/framework/ops.py,1550,function,"Converts `values` to a list of `Tensor` objects.
Args:
values: A list of objects that can be consumed by `tf.convert_to_tensor()`.
dtype: (Optional.) The required `DType` of the returned `Tensor` objects.
name: (Optional.) A name prefix to used when a new `Tensor` is created, in
which case element `i` will be given the name `name + '_' + i`.
as_ref: True if the caller wants the results as ref tensors.
preferred_dtype: Optional element type for the returned tensors, used when
dtype is None. In some cases, a caller may not have a dtype in mind when
converting to a tensor, so preferred_dtype can be used as a soft
preference. If the conversion to `preferred_dtype` is not possible, this
argument has no effect.
ctx: The value of context.context().
Returns:
A list of `Tensor` and/or `IndexedSlices` objects.
Raises:
TypeError: If no conversion function is registered for an element in
`values`.
RuntimeError: If a registered conversion function returns an invalid
value."
4178,convert_n_to_tensor,tensorflow/tensorflow/python/framework/ops.py,1598,function,"Converts `values` to a list of `Tensor` objects.
Args:
values: A list of objects that can be consumed by `tf.convert_to_tensor()`.
dtype: (Optional.) The required `DType` of the returned `Tensor` objects.
name: (Optional.) A name prefix to used when a new `Tensor` is created, in
which case element `i` will be given the name `name + '_' + i`.
preferred_dtype: Optional element type for the returned tensors, used when
dtype is None. In some cases, a caller may not have a dtype in mind when
converting to a tensor, so preferred_dtype can be used as a soft
preference. If the conversion to `preferred_dtype` is not possible, this
argument has no effect.
Returns:
A list of `Tensor` and/or `IndexedSlices` objects.
Raises:
TypeError: If no conversion function is registered for an element in
`values`.
RuntimeError: If a registered conversion function returns an invalid
value."
4179,convert_to_tensor_or_composite,tensorflow/tensorflow/python/framework/ops.py,1629,function,"Converts the given object to a `Tensor` or `CompositeTensor`.
If `value` is a `CompositeTensor` it is returned unmodified. Otherwise, it
is converted to a `Tensor` using `convert_to_tensor()`.
Args:
value: A `CompositeTensor` or an object that can be consumed by
`convert_to_tensor()`.
dtype: (Optional.) The required `DType` of the returned `Tensor` or
`CompositeTensor`.
name: (Optional.) A name to use if a new `Tensor` is created.
Returns:
A `Tensor` or `CompositeTensor`, based on `value`.
Raises:
ValueError: If `dtype` does not match the element type of `value`."
4180,internal_convert_to_tensor_or_composite,tensorflow/tensorflow/python/framework/ops.py,1652,function,"Converts the given object to a `Tensor` or `CompositeTensor`.
If `value` is a `CompositeTensor` it is returned unmodified. Otherwise, it
is converted to a `Tensor` using `convert_to_tensor()`.
Args:
value: A `CompositeTensor`, or an object that can be consumed by
`convert_to_tensor()`.
dtype: (Optional.) The required `DType` of the returned `Tensor` or
`CompositeTensor`.
name: (Optional.) A name to use if a new `Tensor` is created.
as_ref: True if the caller wants the results as ref tensors.
Returns:
A `Tensor` or `CompositeTensor`, based on `value`.
Raises:
ValueError: If `dtype` does not match the element type of `value`."
4181,internal_convert_n_to_tensor_or_composite,tensorflow/tensorflow/python/framework/ops.py,1691,function,"Converts `values` to a list of `Tensor` or `CompositeTensor` objects.
Any `CompositeTensor` objects in `values` are returned unmodified.
Args:
values: A list of `None`, `CompositeTensor`, or objects that can be consumed
by `convert_to_tensor()`.
dtype: (Optional.) The required `DType` of the returned `Tensor`s or
`CompositeTensor`s.
name: (Optional.) A name prefix to used when a new `Tensor` is created, in
which case element `i` will be given the name `name + '_' + i`.
as_ref: True if the caller wants the results as ref tensors.
Returns:
A list of `Tensor`, `CompositeTensor`, and/or `None` objects.
Raises:
TypeError: If no conversion function is registered for an element in
`values`.
RuntimeError: If a registered conversion function returns an invalid
value."
4182,convert_n_to_tensor_or_composite,tensorflow/tensorflow/python/framework/ops.py,1731,function,"Converts `values` to a list of `Output` or `CompositeTensor` objects.
Any `CompositeTensor` objects in `values` are returned unmodified.
Args:
values: A list of `None`, `CompositeTensor``, or objects that can be
consumed by `convert_to_tensor()`.
dtype: (Optional.) The required `DType` of the returned `Tensor`s or
`CompositeTensor`s.
name: (Optional.) A name prefix to used when a new `Tensor` is created, in
which case element `i` will be given the name `name + '_' + i`.
Returns:
A list of `Tensor` and/or `CompositeTensor` objects.
Raises:
TypeError: If no conversion function is registered for an element in
`values`.
RuntimeError: If a registered conversion function returns an invalid
value."
4183,Operation,tensorflow/tensorflow/python/framework/ops.py,1848,class,"Represents a graph node that performs computation on tensors.
An `Operation` is a node in a `tf.Graph` that takes zero or more `Tensor`
objects as input, and produces zero or more `Tensor` objects as output.
Objects of type `Operation` are created by calling a Python op constructor
(such as `tf.matmul`) within a `tf.function` or under a `tf.Graph.as_default`
context manager.
For example, within a `tf.function`, `c = tf.matmul(a, b)` creates an
`Operation` of type ""MatMul"" that takes tensors `a` and `b` as input, and
produces `c` as output.
If a `tf.compat.v1.Session` is used, an `Operation` of a `tf.Graph` can be
executed by passing it to `tf.Session.run`. `op.run()` is a shortcut for
calling `tf.compat.v1.get_default_session().run(op)`."
4184,colocation_groups,tensorflow/tensorflow/python/framework/ops.py,2040,method,Returns the list of colocation groups of the op.
4185,values,tensorflow/tensorflow/python/framework/ops.py,2059,method,DEPRECATED: Use outputs.
4186,name,tensorflow/tensorflow/python/framework/ops.py,2080,method,The full name of this operation.
4187,device,tensorflow/tensorflow/python/framework/ops.py,2090,method,"The name of the device to which this op has been assigned, if any.
Returns:
The string name of the device to which this op has been
assigned, or an empty string if it has not been assigned to a
device."
4188,outputs,tensorflow/tensorflow/python/framework/ops.py,2335,method,The list of `Tensor` objects representing the outputs of this op.
4189,inputs,tensorflow/tensorflow/python/framework/ops.py,2340,method,The sequence of `Tensor` objects representing the data inputs of this op.
4190,control_inputs,tensorflow/tensorflow/python/framework/ops.py,2361,method,"The `Operation` objects on which this op has a control dependency.
Before this op is executed, TensorFlow will ensure that the
operations in `self.control_inputs` have finished executing. This
mechanism can be used to run ops sequentially for performance
reasons, or to ensure that the side effects of an op are observed
in the correct order.
Returns:
A list of `Operation` objects."
4191,type,tensorflow/tensorflow/python/framework/ops.py,2404,method,"The type of the op (e.g. `""MatMul""`)."
4192,graph,tensorflow/tensorflow/python/framework/ops.py,2409,method,The `Graph` that contains this operation.
4193,node_def,tensorflow/tensorflow/python/framework/ops.py,2414,method,"Returns the `NodeDef` representation of this operation.
Returns:
A
[`NodeDef`](https://www.tensorflow.org/code/tensorflow/core/framework/node_def.proto)
protocol buffer."
4194,op_def,tensorflow/tensorflow/python/framework/ops.py,2432,method,"Returns the `OpDef` proto that represents the type of this op.
Returns:
An
[`OpDef`](https://www.tensorflow.org/code/tensorflow/core/framework/op_def.proto)
protocol buffer."
4195,traceback,tensorflow/tensorflow/python/framework/ops.py,2445,method,Returns the call stack from when this operation was constructed.
4196,get_attr,tensorflow/tensorflow/python/framework/ops.py,2498,method,"Returns the value of the attr of this op with the given `name`.
Args:
name: The name of the attr to fetch.
Returns:
The value of the attr, as a Python object.
Raises:
ValueError: If this op does not have an attr with the given `name`."
4197,run,tensorflow/tensorflow/python/framework/ops.py,2562,method,"Runs this operation in a `Session`.
Calling this method will execute all preceding operations that
produce the inputs needed for this operation.
*N.B.* Before invoking `Operation.run()`, its graph must have been
launched in a session, and either a default session must be
available, or `session` must be specified explicitly.
Args:
feed_dict: A dictionary that maps `Tensor` objects to feed values. See
`tf.Session.run` for a description of the valid feed values.
session: (Optional.) The `Session` to be used to run to this operation. If
none, the default session will be used."
4198,RegisterGradient,tensorflow/tensorflow/python/framework/ops.py,2584,class,"A decorator for registering the gradient function for an op type.
This decorator is only used when defining a new op type. For an op
with `m` inputs and `n` outputs, the gradient function is a function
that takes the original `Operation` and `n` `Tensor` objects
(representing the gradients with respect to each output of the op),
and returns `m` `Tensor` objects (representing the partial gradients
with respect to each input of the op).
For example, assuming that operations of type `""Sub""` take two
inputs `x` and `y`, and return a single output `x - y`, the
following gradient function would be registered:
```python
@tf.RegisterGradient(""Sub"")
def _sub_grad(unused_op, grad):
return grad, tf.negative(grad)
```
The decorator argument `op_type` is the string type of an
operation. This corresponds to the `OpDef.name` field for the proto
that defines the operation."
4199,no_gradient,tensorflow/tensorflow/python/framework/ops.py,2633,function,"Specifies that ops of type `op_type` is not differentiable.
This function should *not* be used for operations that have a
well-defined gradient that is not yet implemented.
This function is only used when defining a new op type. It may be
used for ops such as `tf.size()` that are not differentiable. For
example:
```python
tf.no_gradient(""Size"")
```
The gradient computed for 'op_type' will then propagate zeros.
For ops that have a well-defined gradient but are not yet implemented,
no declaration should be made, and an error *must* be thrown if
an attempt to request its gradient is made.
Args:
op_type: The string type of an operation. This corresponds to the
`OpDef.name` field for the proto that defines the operation.
Raises:
TypeError: If `op_type` is not a string."
4200,get_gradient_function,tensorflow/tensorflow/python/framework/ops.py,2671,function,"Returns the function that computes gradients for ""op""."
4201,set_shape_and_handle_data_for_outputs,tensorflow/tensorflow/python/framework/ops.py,2687,function,No op. TODO(b/74620627): Remove this.
4202,OpStats,tensorflow/tensorflow/python/framework/ops.py,2692,class,"A holder for statistics about an operator.
This class holds information about the resource requirements for an op,
including the size of its weight parameters on-disk and how many FLOPS it
requires to execute forward inference.
If you define a new operation, you can create a function that will return a
set of information about its usage of the CPU and disk space when serialized.
The function itself takes a Graph object that's been set up so you can call
methods like get_tensor_by_name to help calculate the results, and a NodeDef
argument."
4203,statistic_type,tensorflow/tensorflow/python/framework/ops.py,2715,method,
4204,statistic_type,tensorflow/tensorflow/python/framework/ops.py,2719,method,
4205,value,tensorflow/tensorflow/python/framework/ops.py,2723,method,
4206,value,tensorflow/tensorflow/python/framework/ops.py,2727,method,
4207,RegisterStatistics,tensorflow/tensorflow/python/framework/ops.py,2744,class,"A decorator for registering the statistics function for an op type.
This decorator can be defined for an op type so that it gives a
report on the resources used by an instance of an operator, in the
form of an OpStats object.
Well-known types of statistics include these so far:
- flops: When running a graph, the bulk of the computation happens doing
numerical calculations like matrix multiplications. This type allows a node
to return how many floating-point operations it takes to complete. The
total number of FLOPs for a graph is a good guide to its expected latency.
You can add your own statistics just by picking a new type string, registering
functions for the ops you care about, and then calling get_stats_for_node_def.
If a statistic for an op is registered multiple times, a KeyError will be
raised.
Since the statistics is counted on a per-op basis. It is not suitable for
model parameters (capacity), which is expected to be counted only once, even
if it is shared by multiple ops. (e.g. RNN)
For example, you can define a new metric called doohickey for a Foo operation
by placing this in your code:
```python
@ops.RegisterStatistics(""Foo"", ""doohickey"")
def _calc_foo_bojangles(unused_graph, unused_node_def):
return ops.OpStats(""doohickey"", 20)
```
Then in client code you can retrieve the value by making this call:
```python
doohickey = ops.get_stats_for_node_def(graph, node_def, ""doohickey"")
```
If the NodeDef is for an op with a registered doohickey function, you'll get
back the calculated amount in doohickey.value, or None if it's not defined."
4208,get_stats_for_node_def,tensorflow/tensorflow/python/framework/ops.py,2809,function,"Looks up the node's statistics function in the registry and calls it.
This function takes a Graph object and a NodeDef from a GraphDef, and if
there's an associated statistics method, calls it and returns a result. If no
function has been registered for the particular node type, it returns an empty
statistics object.
Args:
graph: A Graph object that's been set up with the node's graph.
node: A NodeDef describing the operator.
statistic_type: A string identifying the statistic we're interested in.
Returns:
An OpStats object containing information about resource usage."
4209,name_from_scope_name,tensorflow/tensorflow/python/framework/ops.py,2834,function,"Returns the name of an op given the name of its scope.
Args:
name: the name of the scope.
Returns:
the name of the op (equal to scope name minus any trailing slash)."
4210,Graph,tensorflow/tensorflow/python/framework/ops.py,2851,class,"A TensorFlow computation, represented as a dataflow graph.
Graphs are used by `tf.function`s to represent the function's computations.
Each graph contains a set of `tf.Operation` objects, which represent units of
computation; and `tf.Tensor` objects, which represent the units of data that
flow between operations.
### Using graphs directly (deprecated)
A `tf.Graph` can be constructed and used directly without a `tf.function`, as
was required in TensorFlow 1, but this is deprecated and it is recommended to
use a `tf.function` instead. If a graph is directly used, other deprecated
TensorFlow 1 classes are also required to execute the graph, such as a
`tf.compat.v1.Session`.
A default graph can be registered with the `tf.Graph.as_default` context
manager. Then, operations will be added to the graph instead of being executed
eagerly. For example:
```python
g = tf.Graph()
with g.as_default():
# Define operations and tensors in `g`.
c = tf.constant(30.0)
assert c.graph is g
```
`tf.compat.v1.get_default_graph()` can be used to obtain the default graph.
Important note: This class *is not* thread-safe for graph construction. All
operations should be created from a single thread, or external
synchronization must be provided. Unless otherwise specified, all methods
are not thread-safe.
A `Graph` instance supports an arbitrary number of ""collections""
that are identified by name. For convenience when building a large
graph, collections can store groups of related objects: for
example, the `tf.Variable` uses a collection (named
`tf.GraphKeys.GLOBAL_VARIABLES`) for
all variables that are created during the construction of a graph. The caller
may define additional collections by specifying a new name."
4211,version,tensorflow/tensorflow/python/framework/ops.py,3111,method,"Returns a version number that increases as ops are added to the graph.
Note that this is unrelated to the
`tf.Graph.graph_def_versions`.
Returns:
An integer version that increases as ops are added to the graph."
4212,graph_def_versions,tensorflow/tensorflow/python/framework/ops.py,3127,method,"The GraphDef version information of this graph.
For details on the meaning of each version, see
[`GraphDef`](https://www.tensorflow.org/code/tensorflow/core/framework/graph.proto).
Returns:
A `VersionDef`."
4213,seed,tensorflow/tensorflow/python/framework/ops.py,3146,method,The graph-level random seed of this graph.
4214,seed,tensorflow/tensorflow/python/framework/ops.py,3151,method,
4215,finalized,tensorflow/tensorflow/python/framework/ops.py,3155,method,True if this graph has been finalized.
4216,finalize,tensorflow/tensorflow/python/framework/ops.py,3159,method,"Finalizes this graph, making it read-only.
After calling `g.finalize()`, no new operations can be added to
`g`. This method is used to ensure that no operations are added
to a graph when it is shared between multiple threads, for example
when using a `tf.compat.v1.train.QueueRunner`."
4217,as_graph_def,tensorflow/tensorflow/python/framework/ops.py,3301,method,"Returns a serialized `GraphDef` representation of this graph.
The serialized `GraphDef` can be imported into another `Graph`
(using `tf.import_graph_def`) or used with the
[C++ Session API](../../api_docs/cc/index.md).
This method is thread-safe.
Args:
from_version: Optional. If this is set, returns a `GraphDef` containing
only the nodes that were added to this graph since its `version`
property had the given value.
add_shapes: If true, adds an ""_output_shapes"" list attr to each node with
the inferred shapes of each of its outputs.
Returns:
A
[`GraphDef`](https://www.tensorflow.org/code/tensorflow/core/framework/graph.proto)
protocol buffer.
Raises:
ValueError: If the `graph_def` would be too large."
4218,building_function,tensorflow/tensorflow/python/framework/ops.py,3388,method,Returns True iff this graph represents a function.
4219,create_op,tensorflow/tensorflow/python/framework/ops.py,3396,method,"Creates an `Operation` in this graph.
This is a low-level interface for creating an `Operation`. Most
programs will not call this method directly, and instead use the
Python op constructors, such as `tf.constant()`, which add ops to
the default graph.
Args:
op_type: The `Operation` type to create. This corresponds to the
`OpDef.name` field for the proto that defines the operation.
inputs: A list of `Tensor` objects that will be inputs to the `Operation`.
dtypes: (Optional) A list of `DType` objects that will be the types of the
tensors that the operation produces.
input_types: (Optional.) A list of `DType`s that will be the types of the
tensors that the operation consumes. By default, uses the base `DType`
of each input in `inputs`. Operations that expect reference-typed inputs
must specify `input_types` explicitly.
name: (Optional.) A string name for the operation. If not specified, a
name is generated based on `op_type`.
attrs: (Optional.) A dictionary where the key is the attribute name (a
string) and the value is the respective `attr` attribute of the
`NodeDef` proto that will represent the operation (an `AttrValue`
proto).
op_def: (Optional.) The `OpDef` proto that describes the `op_type` that
the operation will have.
compute_shapes: (Optional.) Deprecated. Has no effect (shapes are always
computed).
compute_device: (Optional.) If True, device functions will be executed to
compute the device property of the Operation.
Raises:
TypeError: if any of the inputs is not a `Tensor`.
ValueError: if colocation conflicts with existing device assignment.
Returns:
An `Operation` object."
4220,as_graph_element,tensorflow/tensorflow/python/framework/ops.py,3674,method,"Returns the object referred to by `obj`, as an `Operation` or `Tensor`.
This function validates that `obj` represents an element of this
graph, and gives an informative error message if it is not.
This function is the canonical way to get/validate an object of
one of the allowed types from an external argument reference in the
Session API.
This method may be called concurrently from multiple threads.
Args:
obj: A `Tensor`, an `Operation`, or the name of a tensor or operation. Can
also be any object with an `_as_graph_element()` method that returns a
value of one of these types. Note: `_as_graph_element` will be called
inside the graph's lock and so may not modify the graph.
allow_tensor: If true, `obj` may refer to a `Tensor`.
allow_operation: If true, `obj` may refer to an `Operation`.
Returns:
The `Tensor` or `Operation` in the Graph corresponding to `obj`.
Raises:
TypeError: If `obj` is not a type we support attempting to convert
to types.
ValueError: If `obj` is of an appropriate type but invalid. For
example, an invalid string.
KeyError: If `obj` is not an object in the graph."
4221,get_operations,tensorflow/tensorflow/python/framework/ops.py,3799,method,"Return the list of operations in the graph.
You can modify the operations in place, but modifications
to the list such as inserts/delete have no effect on the
list of operations known to the graph.
This method may be called concurrently from multiple threads.
Returns:
A list of Operations."
4222,get_operation_by_name,tensorflow/tensorflow/python/framework/ops.py,3817,method,"Returns the `Operation` with the given `name`.
This method may be called concurrently from multiple threads.
Args:
name: The name of the `Operation` to return.
Returns:
The `Operation` with the given `name`.
Raises:
TypeError: If `name` is not a string.
KeyError: If `name` does not correspond to an operation in this graph."
4223,get_tensor_by_name,tensorflow/tensorflow/python/framework/ops.py,3865,method,"Returns the `Tensor` with the given `name`.
This method may be called concurrently from multiple threads.
Args:
name: The name of the `Tensor` to return.
Returns:
The `Tensor` with the given `name`.
Raises:
TypeError: If `name` is not a string.
KeyError: If `name` does not correspond to a tensor in this graph."
4224,as_default,tensorflow/tensorflow/python/framework/ops.py,3924,method,"Returns a context manager that makes this `Graph` the default graph.
This method should be used if you want to create multiple graphs
in the same process. For convenience, a global default graph is
provided, and all ops will be added to this graph if you do not
create a new graph explicitly.
Use this method with the `with` keyword to specify that ops created within
the scope of a block should be added to this graph. In this case, once
the scope of the `with` is exited, the previous default graph is set again
as default. There is a stack, so it's ok to have multiple nested levels
of `as_default` calls.
The default graph is a property of the current thread. If you
create a new thread, and wish to use the default graph in that
thread, you must explicitly add a `with g.as_default():` in that
thread's function.
The following code examples are equivalent:
```python
# 1. Using Graph.as_default():
g = tf.Graph()
with g.as_default():
c = tf.constant(5.0)
assert c.graph is g
# 2. Constructing and making default:
with tf.Graph().as_default() as g:
c = tf.constant(5.0)
assert c.graph is g
```
If eager execution is enabled ops created under this context manager will be
added to the graph instead of executed eagerly.
Returns:
A context manager for using this graph as the default graph."
4225,collections,tensorflow/tensorflow/python/framework/ops.py,3967,method,Returns the names of the collections known to this graph.
4226,add_to_collection,tensorflow/tensorflow/python/framework/ops.py,3971,method,"Stores `value` in the collection with the given `name`.
Note that collections are not sets, so it is possible to add a value to
a collection several times.
Args:
name: The key for the collection. The `GraphKeys` class contains many
standard names for collections.
value: The value to add to the collection."
4227,add_to_collections,tensorflow/tensorflow/python/framework/ops.py,3989,method,"Stores `value` in the collections given by `names`.
Note that collections are not sets, so it is possible to add a value to
a collection several times. This function makes sure that duplicates in
`names` are ignored, but it will not check for pre-existing membership of
`value` in any of the collections in `names`.
`names` can be any iterable, but if `names` is a string, it is treated as a
single collection name.
Args:
names: The keys for the collections to add to. The `GraphKeys` class
contains many standard names for collections.
value: The value to add to the collections."
4228,get_collection_ref,tensorflow/tensorflow/python/framework/ops.py,4010,method,"Returns a list of values in the collection with the given `name`.
If the collection exists, this returns the list itself, which can
be modified in place to change the collection. If the collection does
not exist, it is created as an empty list and the list is returned.
This is different from `get_collection()` which always returns a copy of
the collection list if it exists and never creates an empty collection.
Args:
name: The key for the collection. For example, the `GraphKeys` class
contains many standard names for collections.
Returns:
The list of values in the collection with the given `name`, or an empty
list if no value has been added to that collection."
4229,get_collection,tensorflow/tensorflow/python/framework/ops.py,4035,method,"Returns a list of values in the collection with the given `name`.
This is different from `get_collection_ref()` which always returns the
actual collection list if it exists in that it returns a new list each time
it is called.
Args:
name: The key for the collection. For example, the `GraphKeys` class
contains many standard names for collections.
scope: (Optional.) A string. If supplied, the resulting list is filtered
to include only items whose `name` attribute matches `scope` using
`re.match`. Items without a `name` attribute are never returned if a
scope is supplied. The choice of `re.match` means that a `scope` without
special tokens filters by prefix.
Returns:
The list of values in the collection with the given `name`, or
an empty list if no value has been added to that collection. The
list contains the values in the order under which they were
collected."
4230,get_all_collection_keys,tensorflow/tensorflow/python/framework/ops.py,4075,method,Returns a list of collections used in this graph.
4231,clear_collection,tensorflow/tensorflow/python/framework/ops.py,4080,method,"Clears all values in a collection.
Args:
name: The key for the collection. The `GraphKeys` class contains many
standard names for collections."
4232,name_scope,tensorflow/tensorflow/python/framework/ops.py,4130,method,"Returns a context manager that creates hierarchical names for operations.
A graph maintains a stack of name scopes. A `with name_scope(...):`
statement pushes a new name onto the stack for the lifetime of the context.
The `name` argument will be interpreted as follows:
* A string (not ending with '/') will create a new name scope, in which
`name` is appended to the prefix of all operations created in the
context. If `name` has been used before, it will be made unique by
calling `self.unique_name(name)`.
* A scope previously captured from a `with g.name_scope(...) as
scope:` statement will be treated as an ""absolute"" name scope, which
makes it possible to re-enter existing scopes.
* A value of `None` or the empty string will reset the current name scope
to the top-level (empty) name scope.
For example:
```python
with tf.Graph().as_default() as g:
c = tf.constant(5.0, name=""c"")
assert c.op.name == ""c""
c_1 = tf.constant(6.0, name=""c"")
assert c_1.op.name == ""c_1""
# Creates a scope called ""nested""
with g.name_scope(""nested"") as scope:
nested_c = tf.constant(10.0, name=""c"")
assert nested_c.op.name == ""nested/c""
# Creates a nested scope called ""inner"".
with g.name_scope(""inner""):
nested_inner_c = tf.constant(20.0, name=""c"")
assert nested_inner_c.op.name == ""nested/inner/c""
# Create a nested scope called ""inner_1"".
with g.name_scope(""inner""):
nested_inner_1_c = tf.constant(30.0, name=""c"")
assert nested_inner_1_c.op.name == ""nested/inner_1/c""
# Treats `scope` as an absolute name scope, and
# switches to the ""nested/"" scope.
with g.name_scope(scope):
nested_d = tf.constant(40.0, name=""d"")
assert nested_d.op.name == ""nested/d""
with g.name_scope(""""):
e = tf.constant(50.0, name=""e"")
assert e.op.name == ""e""
```
The name of the scope itself can be captured by `with
g.name_scope(...) as scope:`, which stores the name of the scope
in the variable `scope`. This value can be used to name an
operation that represents the overall result of executing the ops
in a scope. For example:
```python
inputs = tf.constant(...)
with g.name_scope('my_layer') as scope:
weights = tf.Variable(..., name=""weights"")
biases = tf.Variable(..., name=""biases"")
affine = tf.matmul(inputs, weights) + biases
output = tf.nn.relu(affine, name=scope)
```
NOTE: This constructor validates the given `name`. Valid scope
names match one of the following regular expressions:
[A-Za-z0-9.][A-Za-z0-9_.\-/]* (for scopes at the root)
[A-Za-z0-9_.\-/]* (for other scopes)
Args:
name: A name for the scope.
Returns:
A context manager that installs `name` as a new name scope.
Raises:
ValueError: If `name` is not a valid scope name, according to the rules
above."
4233,unique_name,tensorflow/tensorflow/python/framework/ops.py,4244,method,"Return a unique operation name for `name`.
Note: You rarely need to call `unique_name()` directly. Most of
the time you just need to create `with g.name_scope()` blocks to
generate structured names.
`unique_name` is used to generate structured names, separated by
`""/""`, to help identify operations when debugging a graph.
Operation names are displayed in error messages reported by the
TensorFlow runtime, and in various visualization tools such as
TensorBoard.
If `mark_as_used` is set to `True`, which is the default, a new
unique name is created and marked as in use. If it's set to `False`,
the unique name is returned without actually being marked as used.
This is useful when the caller simply wants to know what the name
to be created will be.
Args:
name: The name for an operation.
mark_as_used: Whether to mark this name as being used.
Returns:
A string to be passed to `create_op()` that will be used
to name the operation being created."
4234,get_name_scope,tensorflow/tensorflow/python/framework/ops.py,4296,method,"Returns the current name scope.
For example:
```python
with tf.name_scope('scope1'):
with tf.name_scope('scope2'):
print(tf.compat.v1.get_default_graph().get_name_scope())
```
would print the string `scope1/scope2`.
Returns:
A string representing the current name scope."
4235,colocate_with,tensorflow/tensorflow/python/framework/ops.py,4327,method,"Returns a context manager that specifies an op to colocate with.
Note: this function is not for public use, only for internal libraries.
For example:
```python
a = tf.Variable([1.0])
with g.colocate_with(a):
b = tf.constant(1.0)
c = tf.add(a, b)
```
`b` and `c` will always be colocated with `a`, no matter where `a`
is eventually placed.
**NOTE** Using a colocation scope resets any existing device constraints.
If `op` is `None` then `ignore_existing` must be `True` and the new
scope resets all colocation and device constraints.
Args:
op: The op to colocate all created ops with, or `None`.
ignore_existing: If true, only applies colocation of this op within the
context, rather than applying all colocation properties on the stack.
If `op` is `None`, this value must be `True`.
Raises:
ValueError: if op is None but ignore_existing is False.
Yields:
A context manager that specifies the op with which to colocate
newly created ops."
4236,device,tensorflow/tensorflow/python/framework/ops.py,4413,method,"Returns a context manager that specifies the default device to use.
The `device_name_or_function` argument may either be a device name
string, a device function, or None:
* If it is a device name string, all operations constructed in
this context will be assigned to the device with that name, unless
overridden by a nested `device()` context.
* If it is a function, it will be treated as a function from
Operation objects to device name strings, and invoked each time
a new Operation is created. The Operation will be assigned to
the device with the returned name.
* If it is None, all `device()` invocations from the enclosing context
will be ignored.
For information about the valid syntax of device name strings, see
the documentation in
[`DeviceNameUtils`](https://www.tensorflow.org/code/tensorflow/core/util/device_name_utils.h).
For example:
```python
with g.device('/device:GPU:0'):
# All operations constructed in this context will be placed
# on GPU 0.
with g.device(None):
# All operations constructed in this context will have no
# assigned device.
# Defines a function from `Operation` to device string.
def matmul_on_gpu(n):
if n.type == ""MatMul"":
return ""/device:GPU:0""
else:
return ""/cpu:0""
with g.device(matmul_on_gpu):
# All operations of type ""MatMul"" constructed in this context
# will be placed on GPU 0; all other operations will be placed
# on CPU 0.
```
**N.B.** The device scope may be overridden by op wrappers or
other library code. For example, a variable assignment op
`v.assign()` must be colocated with the `tf.Variable` `v`, and
incompatible device scopes will be ignored.
Args:
device_name_or_function: The device name or function to use in the
context.
Yields:
A context manager that specifies the default device to use for newly
created ops.
Raises:
RuntimeError: If device scopes are not properly nested."
4237,container,tensorflow/tensorflow/python/framework/ops.py,4510,method,"Returns a context manager that specifies the resource container to use.
Stateful operations, such as variables and queues, can maintain their
states on devices so that they can be shared by multiple processes.
A resource container is a string name under which these stateful
operations are tracked. These resources can be released or cleared
with `tf.Session.reset()`.
For example:
```python
with g.container('experiment0'):
# All stateful Operations constructed in this context will be placed
# in resource container ""experiment0"".
v1 = tf.Variable([1.0])
v2 = tf.Variable([2.0])
with g.container(""experiment1""):
# All stateful Operations constructed in this context will be
# placed in resource container ""experiment1"".
v3 = tf.Variable([3.0])
q1 = tf.queue.FIFOQueue(10, tf.float32)
# All stateful Operations constructed in this context will be
# be created in the ""experiment0"".
v4 = tf.Variable([4.0])
q1 = tf.queue.FIFOQueue(20, tf.float32)
with g.container(""""):
# All stateful Operations constructed in this context will be
# be placed in the default resource container.
v5 = tf.Variable([5.0])
q3 = tf.queue.FIFOQueue(30, tf.float32)
# Resets container ""experiment0"", after which the state of v1, v2, v4, q1
# will become undefined (such as uninitialized).
tf.Session.reset(target, [""experiment0""])
```
Args:
container_name: container name string.
Returns:
A context manager for defining resource containers for stateful ops,
yields the container name."
4238,control_dependencies,tensorflow/tensorflow/python/framework/ops.py,4690,method,"Returns a context manager that specifies control dependencies.
Use with the `with` keyword to specify that all operations constructed
within the context should have control dependencies on
`control_inputs`. For example:
```python
with g.control_dependencies([a, b, c]):
# `d` and `e` will only run after `a`, `b`, and `c` have executed.
d = ...
e = ...
```
Multiple calls to `control_dependencies()` can be nested, and in
that case a new `Operation` will have control dependencies on the union
of `control_inputs` from all active contexts.
```python
with g.control_dependencies([a, b]):
# Ops constructed here run after `a` and `b`.
with g.control_dependencies([c, d]):
# Ops constructed here run after `a`, `b`, `c`, and `d`.
```
You can pass None to clear the control dependencies:
```python
with g.control_dependencies([a, b]):
# Ops constructed here run after `a` and `b`.
with g.control_dependencies(None):
# Ops constructed here run normally, not waiting for either `a` or `b`.
with g.control_dependencies([c, d]):
# Ops constructed here run after `c` and `d`, also not waiting
# for either `a` or `b`.
```
*N.B.* The control dependencies context applies *only* to ops that
are constructed within the context. Merely using an op or tensor
in the context does not add a control dependency. The following
example illustrates this point:
```python
# WRONG
def my_func(pred, tensor):
t = tf.matmul(tensor, tensor)
with tf.control_dependencies([pred]):
# The matmul op is created outside the context, so no control
# dependency will be added.
return t
# RIGHT
def my_func(pred, tensor):
with tf.control_dependencies([pred]):
# The matmul op is created in the context, so a control dependency
# will be added.
return tf.matmul(tensor, tensor)
```
Also note that though execution of ops created under this scope will trigger
execution of the dependencies, the ops created under this scope might still
be pruned from a normal tensorflow graph. For example, in the following
snippet of code the dependencies are never executed:
```python
loss = model.loss()
with tf.control_dependencies(dependencies):
loss = loss + tf.constant(1) # note: dependencies ignored in the
# backward pass
return tf.gradients(loss, model.variables)
```
This is because evaluating the gradient graph does not require evaluating
the constant(1) op created in the forward pass.
Args:
control_inputs: A list of `Operation` or `Tensor` objects which must be
executed or computed before running the operations defined in the
context. Can also be `None` to clear the control dependencies.
Returns:
A context manager that specifies control dependencies for all
operations constructed within the context.
Raises:
TypeError: If `control_inputs` is not a list of `Operation` or
`Tensor` objects."
4239,gradient_override_map,tensorflow/tensorflow/python/framework/ops.py,4944,method,"EXPERIMENTAL: A context manager for overriding gradient functions.
This context manager can be used to override the gradient function
that will be used for ops within the scope of the context.
For example:
```python
@tf.RegisterGradient(""CustomSquare"")
def _custom_square_grad(op, grad):
# ...
with tf.Graph().as_default() as g:
c = tf.constant(5.0)
s_1 = tf.square(c) # Uses the default gradient for tf.square.
with g.gradient_override_map({""Square"": ""CustomSquare""}):
s_2 = tf.square(s_2) # Uses _custom_square_grad to compute the
# gradient of s_2.
```
Args:
op_type_map: A dictionary mapping op type strings to alternative op type
strings.
Returns:
A context manager that sets the alternative op type to be used for one
or more ops created in that context.
Raises:
TypeError: If `op_type_map` is not a dictionary mapping strings to
strings."
4240,prevent_feeding,tensorflow/tensorflow/python/framework/ops.py,5006,method,Marks the given `tensor` as unfeedable in this graph.
4241,is_feedable,tensorflow/tensorflow/python/framework/ops.py,5010,method,Returns `True` if and only if `tensor` is feedable.
4242,prevent_fetching,tensorflow/tensorflow/python/framework/ops.py,5014,method,Marks the given `op` as unfetchable in this graph.
4243,is_fetchable,tensorflow/tensorflow/python/framework/ops.py,5018,method,Returns `True` if and only if `tensor_or_op` is fetchable.
4244,switch_to_thread_local,tensorflow/tensorflow/python/framework/ops.py,5025,method,"Make device, colocation and dependencies stacks thread-local.
Device, colocation and dependencies stacks are not thread-local be default.
If multiple threads access them, then the state is shared. This means that
one thread may affect the behavior of another thread.
After this method is called, the stacks become thread-local. If multiple
threads access them, then the state is not shared. Each thread uses its own
value; a thread doesn't affect other threads by mutating such a stack.
The initial value for every thread's stack is set to the current value
of the stack when `switch_to_thread_local()` was first called."
4245,control_inputs,tensorflow/tensorflow/python/framework/ops.py,4620,method,
4246,add_op,tensorflow/tensorflow/python/framework/ops.py,4623,method,
4247,op_in_group,tensorflow/tensorflow/python/framework/ops.py,4628,method,
4248,enable_auto_cast_variables,tensorflow/tensorflow/python/framework/ops.py,5212,class,"Enables the autocasting of `AutoCastVariable`s.
Under this context manager, `AutoCastVariable`s will be cast to `dtype` if
`dtype` is floating-point. Otherwise, `AutoCastVariable`s will not be cast."
4249,device,tensorflow/tensorflow/python/framework/ops.py,5249,function,"Wrapper for `Graph.device()` using the default graph.
See `tf.Graph.device` for more details.
Args:
device_name_or_function: The device name or function to use in the context.
Returns:
A context manager that specifies the default device to use for newly
created ops.
Raises:
RuntimeError: If eager execution is enabled and a function is passed in."
4250,device_v2,tensorflow/tensorflow/python/framework/ops.py,5285,function,"Specifies the device for ops created/executed in this context.
This function specifies the device to be used for ops created/executed in a
particular context. Nested contexts will inherit and also create/execute
their ops on the specified device. If a specific device is not required,
consider not using this function so that a device can be automatically
assigned. In general the use of this function is optional. `device_name` can
be fully specified, as in ""/job:worker/task:1/device:cpu:0"", or partially
specified, containing only a subset of the ""/""-separated fields. Any fields
which are specified will override device annotations from outer scopes.
For example:
```python
with tf.device('/job:foo'):
# ops created here have devices with /job:foo
with tf.device('/job:bar/task:0/device:gpu:2'):
# ops created here have the fully specified device above
with tf.device('/device:gpu:1'):
# ops created here have the device '/job:foo/device:gpu:1'
```
Args:
device_name: The device name to use in the context.
Returns:
A context manager that specifies the default device to use for newly
created ops.
Raises:
RuntimeError: If a function is passed in."
4251,container,tensorflow/tensorflow/python/framework/ops.py,5324,function,"Wrapper for `Graph.container()` using the default graph.
Args:
container_name: The container string to use in the context.
Returns:
A context manager that specifies the default container to use for newly
created stateful ops."
4252,colocate_with,tensorflow/tensorflow/python/framework/ops.py,5360,function,
4253,control_dependencies,tensorflow/tensorflow/python/framework/ops.py,5372,function,"Wrapper for `Graph.control_dependencies()` using the default graph.
See `tf.Graph.control_dependencies`
for more details.
Note: *In TensorFlow 2 with eager and/or Autograph, you should not require
this method, as code executes in the expected order.* Only use
`tf.control_dependencies` when working with v1-style code or in a graph
context such as inside `Dataset.map`.
When eager execution is enabled, any callable object in the `control_inputs`
list will be called.
Args:
control_inputs: A list of `Operation` or `Tensor` objects which must be
executed or computed before running the operations defined in the context.
Can also be `None` to clear the control dependencies. If eager execution
is enabled, any callable object in the `control_inputs` list will be
called.
Returns:
A context manager that specifies control dependencies for all
operations constructed within the context."
4254,default_session,tensorflow/tensorflow/python/framework/ops.py,5455,function,"Python ""with"" handler for defining a default session.
This function provides a means of registering a session for handling
Tensor.eval() and Operation.run() calls. It is primarily intended for use
by session.Session, but can be used with any object that implements
the Session.run() interface.
Use with the ""with"" keyword to specify that Tensor.eval() and Operation.run()
invocations within the scope of a block should be executed by a particular
session.
The default session applies to the current thread only, so it is always
possible to inspect the call stack and determine the scope of a default
session. If you create a new thread, and wish to use the default session
in that thread, you must explicitly add a ""with ops.default_session(sess):""
block in that thread's function.
Example:
The following code examples are equivalent:
# 1. Using the Session object directly:
sess = ...
c = tf.constant(5.0)
sess.run(c)
# 2. Using default_session():
sess = ...
with ops.default_session(sess):
c = tf.constant(5.0)
result = c.eval()
# 3. Overriding default_session():
sess = ...
with ops.default_session(sess):
c = tf.constant(5.0)
with ops.default_session(...):
c.eval(session=sess)
Args:
session: The session to be installed as the default session.
Returns:
A context manager for the default session."
4255,get_default_session,tensorflow/tensorflow/python/framework/ops.py,5504,function,"Returns the default session for the current thread.
The returned `Session` will be the innermost session on which a
`Session` or `Session.as_default()` context has been entered.
NOTE: The default session is a property of the current thread. If you
create a new thread, and wish to use the default session in that
thread, you must explicitly add a `with sess.as_default():` in that
thread's function.
Returns:
The default `Session` being used in the current thread."
4256,init_scope,tensorflow/tensorflow/python/framework/ops.py,5687,function,"A context manager that lifts ops out of control-flow scopes and function-building graphs.
There is often a need to lift variable initialization ops out of control-flow
scopes, function-building graphs, and gradient tapes. Entering an
`init_scope` is a mechanism for satisfying these desiderata. In particular,
entering an `init_scope` has three effects:
(1) All control dependencies are cleared the moment the scope is entered;
this is equivalent to entering the context manager returned from
`control_dependencies(None)`, which has the side-effect of exiting
control-flow scopes like `tf.cond` and `tf.while_loop`.
(2) All operations that are created while the scope is active are lifted
into the lowest context on the `context_stack` that is not building a
graph function. Here, a context is defined as either a graph or an eager
context. Every context switch, i.e., every installation of a graph as
the default graph and every switch into eager mode, is logged in a
thread-local stack called `context_switches`; the log entry for a
context switch is popped from the stack when the context is exited.
Entering an `init_scope` is equivalent to crawling up
`context_switches`, finding the first context that is not building a
graph function, and entering it. A caveat is that if graph mode is
enabled but the default graph stack is empty, then entering an
`init_scope` will simply install a fresh graph as the default one.
(3) The gradient tape is paused while the scope is active.
When eager execution is enabled, code inside an init_scope block runs with
eager execution enabled even when tracing a `tf.function`. For example:
```python
tf.compat.v1.enable_eager_execution()
@tf.function
def func():
# A function constructs TensorFlow graphs,
# it does not execute eagerly.
assert not tf.executing_eagerly()
with tf.init_scope():
# Initialization runs with eager execution enabled
assert tf.executing_eagerly()
```
Raises:
RuntimeError: if graph state is incompatible with this initialization."
4257,executing_eagerly_outside_functions,tensorflow/tensorflow/python/framework/ops.py,5791,function,"Returns True if executing eagerly, even if inside a graph function.
This function will check the outermost context for the program and see if
it is in eager mode. It is useful comparing to `tf.executing_eagerly()`,
which checks the current context and will return `False` within a
`tf.function` body. It can be used to build library that behave differently
in eager runtime and v1 session runtime (deprecated).
Example:
>>> tf.compat.v1.enable_eager_execution()
>>> @tf.function
... def func():
... # A function constructs TensorFlow graphs, it does not execute eagerly,
... # but the outer most context is still eager.
... assert not tf.executing_eagerly()
... return tf.compat.v1.executing_eagerly_outside_functions()
>>> func()
<tf.Tensor: shape=(), dtype=bool, numpy=True>
Returns:
boolean, whether the outermost context is in eager mode."
4258,inside_function,tensorflow/tensorflow/python/framework/ops.py,5823,function,
4259,enable_eager_execution,tensorflow/tensorflow/python/framework/ops.py,5828,function,"Enables eager execution for the lifetime of this program.
Eager execution provides an imperative interface to TensorFlow. With eager
execution enabled, TensorFlow functions execute operations immediately (as
opposed to adding to a graph to be executed later in a `tf.compat.v1.Session`)
and
return concrete values (as opposed to symbolic references to a node in a
computational graph).
For example:
```python
tf.compat.v1.enable_eager_execution()
# After eager execution is enabled, operations are executed as they are
# defined and Tensor objects hold concrete values, which can be accessed as
# numpy.ndarray`s through the numpy() method.
assert tf.multiply(6, 7).numpy() == 42
```
Eager execution cannot be enabled after TensorFlow APIs have been used to
create or execute graphs. It is typically recommended to invoke this function
at program startup and not in a library (as most libraries should be usable
both with and without eager execution).
Args:
config: (Optional.) A `tf.compat.v1.ConfigProto` to use to configure the
environment in which operations are executed. Note that
`tf.compat.v1.ConfigProto` is also used to configure graph execution (via
`tf.compat.v1.Session`) and many options within `tf.compat.v1.ConfigProto`
are not implemented (or are irrelevant) when eager execution is enabled.
device_policy: (Optional.) Policy controlling how operations requiring
inputs on a specific device (e.g., a GPU 0) handle inputs on a different
device (e.g. GPU 1 or CPU). When set to None, an appropriate value will
be picked automatically. The value picked may change between TensorFlow
releases.
Valid values:
- tf.contrib.eager.DEVICE_PLACEMENT_EXPLICIT: raises an error if the
placement is not correct.
- tf.contrib.eager.DEVICE_PLACEMENT_WARN: copies the tensors which are not
on the right device but logs a warning.
- tf.contrib.eager.DEVICE_PLACEMENT_SILENT: silently copies the tensors.
Note that this may hide performance problems as there is no notification
provided when operations are blocked on the tensor being copied between
devices.
- tf.contrib.eager.DEVICE_PLACEMENT_SILENT_FOR_INT32: silently copies
int32 tensors, raising errors on the other ones.
execution_mode: (Optional.) Policy controlling how operations dispatched are
actually executed. When set to None, an appropriate value will be picked
automatically. The value picked may change between TensorFlow releases.
Valid values:
- tf.contrib.eager.SYNC: executes each operation synchronously.
- tf.contrib.eager.ASYNC: executes each operation asynchronously. These
operations may return ""non-ready"" handles.
Raises:
ValueError: If eager execution is enabled after creating/executing a
TensorFlow graph, or if options provided conflict with a previous call
to this function."
4260,disable_eager_execution,tensorflow/tensorflow/python/framework/ops.py,5900,function,"Disables eager execution.
This function can only be called before any Graphs, Ops, or Tensors have been
created. It can be used at the beginning of the program for complex migration
projects from TensorFlow 1.x to 2.x."
4261,enable_eager_execution_internal,tensorflow/tensorflow/python/framework/ops.py,5914,function,"Enables eager execution for the lifetime of this program.
Most of the doc string for enable_eager_execution is relevant here as well.
Args:
config: See enable_eager_execution doc string
device_policy: See enable_eager_execution doc string
execution_mode: See enable_eager_execution doc string
server_def: (Optional.) A tensorflow::ServerDef proto. Enables execution on
remote devices. GrpcServers need to be started by creating an identical
server_def to this, and setting the appropriate task_indexes, so that the
servers can communicate. It will then be possible to execute operations on
remote devices.
Raises:
ValueError"
4262,eager_run,tensorflow/tensorflow/python/framework/ops.py,5987,function,"Runs the program with an optional main function and argv list.
The program will run with eager execution enabled.
Example:
```python
import tensorflow as tf
# Import subject to future changes:
from tensorflow.contrib.eager.python import tfe
def main(_):
u = tf.constant(6.0)
v = tf.constant(7.0)
print(u * v)
if __name__ == ""__main__"":
tfe.run()
```
Args:
main: the main function to run.
argv: the arguments to pass to it."
4263,reset_default_graph,tensorflow/tensorflow/python/framework/ops.py,6016,function,"Clears the default graph stack and resets the global default graph.
NOTE: The default graph is a property of the current thread. This
function applies only to the current thread. Calling this function while
a `tf.compat.v1.Session` or `tf.compat.v1.InteractiveSession` is active will
result in undefined
behavior. Using any previously created `tf.Operation` or `tf.Tensor` objects
after calling this function will result in undefined behavior.
Raises:
AssertionError: If this function is called within a nested graph."
4264,get_default_graph,tensorflow/tensorflow/python/framework/ops.py,6036,function,"Returns the default graph for the current thread.
The returned graph will be the innermost graph on which a
`Graph.as_default()` context has been entered, or a global default
graph if none has been explicitly created.
NOTE: The default graph is a property of the current thread. If you
create a new thread, and wish to use the default graph in that
thread, you must explicitly add a `with g.as_default():` in that
thread's function.
Returns:
The default `Graph` being used in the current thread."
4265,has_default_graph,tensorflow/tensorflow/python/framework/ops.py,6054,function,Returns True if there is a default graph.
4266,get_name_scope,tensorflow/tensorflow/python/framework/ops.py,6059,function,"Returns the current name scope in the default_graph.
For example:
```python
with tf.name_scope('scope1'):
with tf.name_scope('scope2'):
print(tf.get_name_scope())
```
would print the string `scope1/scope2`.
Returns:
A string representing the current name scope."
4267,GraphKeys,tensorflow/tensorflow/python/framework/ops.py,6168,class,"Standard names to use for graph collections.
The standard library uses various well-known names to collect and
retrieve values associated with a graph. For example, the
`tf.Optimizer` subclasses default to optimizing the variables
collected under `tf.GraphKeys.TRAINABLE_VARIABLES` if none is
specified, but it is also possible to pass an explicit list of
variables.
The following standard keys are defined:
* `GLOBAL_VARIABLES`: the default collection of `Variable` objects, shared
across distributed environment (model variables are subset of these). See
`tf.compat.v1.global_variables`
for more details.
Commonly, all `TRAINABLE_VARIABLES` variables will be in `MODEL_VARIABLES`,
and all `MODEL_VARIABLES` variables will be in `GLOBAL_VARIABLES`.
* `LOCAL_VARIABLES`: the subset of `Variable` objects that are local to each
machine. Usually used for temporarily variables, like counters.
Note: use `tf.contrib.framework.local_variable` to add to this collection.
* `MODEL_VARIABLES`: the subset of `Variable` objects that are used in the
model for inference (feed forward). Note: use
`tf.contrib.framework.model_variable` to add to this collection.
* `TRAINABLE_VARIABLES`: the subset of `Variable` objects that will
be trained by an optimizer. See
`tf.compat.v1.trainable_variables`
for more details.
* `SUMMARIES`: the summary `Tensor` objects that have been created in the
graph. See
`tf.compat.v1.summary.merge_all`
for more details.
* `QUEUE_RUNNERS`: the `QueueRunner` objects that are used to
produce input for a computation. See
`tf.compat.v1.train.start_queue_runners`
for more details.
* `MOVING_AVERAGE_VARIABLES`: the subset of `Variable` objects that will also
keep moving averages. See
`tf.compat.v1.moving_average_variables`
for more details.
* `REGULARIZATION_LOSSES`: regularization losses collected during graph
construction.
The following standard keys are _defined_, but their collections are **not**
automatically populated as many of the others are:
* `WEIGHTS`
* `BIASES`
* `ACTIVATIONS`"
4268,VARIABLES,tensorflow/tensorflow/python/framework/ops.py,6309,method,
4269,dismantle_graph,tensorflow/tensorflow/python/framework/ops.py,6313,function,"Cleans up reference cycles from a `Graph`.
Helpful for making sure the garbage collector doesn't need to run after a
temporary `Graph` is no longer needed.
Args:
graph: A `Graph` object to destroy. Neither it nor any of its ops are usable
after this function runs."
4270,add_to_collection,tensorflow/tensorflow/python/framework/ops.py,6334,function,"Wrapper for `Graph.add_to_collection()` using the default graph.
See `tf.Graph.add_to_collection`
for more details.
Args:
name: The key for the collection. For example, the `GraphKeys` class
contains many standard names for collections.
value: The value to add to the collection.
@compatibility(eager)
Collections are only supported in eager when variables are created inside
an EagerVariableStore (e.g. as part of a layer or template).
@end_compatibility"
4271,add_to_collections,tensorflow/tensorflow/python/framework/ops.py,6354,function,"Wrapper for `Graph.add_to_collections()` using the default graph.
See `tf.Graph.add_to_collections`
for more details.
Args:
names: The key for the collections. The `GraphKeys` class contains many
standard names for collections.
value: The value to add to the collections.
@compatibility(eager)
Collections are only supported in eager when variables are created inside
an EagerVariableStore (e.g. as part of a layer or template).
@end_compatibility"
4272,get_collection_ref,tensorflow/tensorflow/python/framework/ops.py,6374,function,"Wrapper for `Graph.get_collection_ref()` using the default graph.
See `tf.Graph.get_collection_ref`
for more details.
Args:
key: The key for the collection. For example, the `GraphKeys` class contains
many standard names for collections.
Returns:
The list of values in the collection with the given `name`, or an empty
list if no value has been added to that collection. Note that this returns
the collection list itself, which can be modified in place to change the
collection.
@compatibility(eager)
Collections are not supported when eager execution is enabled.
@end_compatibility"
4273,get_collection,tensorflow/tensorflow/python/framework/ops.py,6398,function,"Wrapper for `Graph.get_collection()` using the default graph.
See `tf.Graph.get_collection`
for more details.
Args:
key: The key for the collection. For example, the `GraphKeys` class contains
many standard names for collections.
scope: (Optional.) If supplied, the resulting list is filtered to include
only items whose `name` attribute matches using `re.match`. Items without
a `name` attribute are never returned if a scope is supplied and the
choice or `re.match` means that a `scope` without special tokens filters
by prefix.
Returns:
The list of values in the collection with the given `name`, or
an empty list if no value has been added to that collection. The
list contains the values in the order under which they were
collected.
@compatibility(eager)
Collections are not supported when eager execution is enabled.
@end_compatibility"
4274,get_all_collection_keys,tensorflow/tensorflow/python/framework/ops.py,6426,function,Returns a list of collections used in the default graph.
4275,name_scope,tensorflow/tensorflow/python/framework/ops.py,6431,function,"Internal-only entry point for `name_scope*`.
Internal ops do not use the public API and instead rely on
`ops.name_scope` regardless of the execution mode. This function
dispatches to the correct `name_scope*` implementation based on
the arguments provided and the current mode. Specifically,
* if `values` contains a graph tensor `Graph.name_scope` is used;
* `name_scope_v1` is used in graph mode;
* `name_scope_v2` -- in eager mode.
Args:
name: The name argument that is passed to the op function.
default_name: The default name to use if the `name` argument is `None`.
values: The list of `Tensor` arguments that are passed to the op function.
skip_on_eager: Indicates to return NullContextmanager if executing eagerly.
By default this is True since naming tensors and operations in eager mode
have little use and cause unnecessary performance overhead. However, it is
important to preserve variable names since they are often useful for
debugging and saved models.
Returns:
`name_scope*` context manager."
4276,internal_name_scope_v1,tensorflow/tensorflow/python/framework/ops.py,6476,class,Graph-only version of `name_scope_v1`.
4277,name,tensorflow/tensorflow/python/framework/ops.py,6480,method,
4278,name_scope_v1,tensorflow/tensorflow/python/framework/ops.py,6552,class,"A context manager for use when defining a Python op.
This context manager validates that the given `values` are from the
same graph, makes that graph the default graph, and pushes a
name scope in that graph (see
`tf.Graph.name_scope`
for more details on that).
For example, to define a new Python op called `my_op`:
```python
def my_op(a, b, c, name=None):
with tf.name_scope(name, ""MyOp"", [a, b, c]) as scope:
a = tf.convert_to_tensor(a, name=""a"")
b = tf.convert_to_tensor(b, name=""b"")
c = tf.convert_to_tensor(c, name=""c"")
# Define some computation that uses `a`, `b`, and `c`.
return foo_op(..., name=scope)
```"
4279,name,tensorflow/tensorflow/python/framework/ops.py,6577,method,
4280,name_scope_v2,tensorflow/tensorflow/python/framework/ops.py,6603,class,"A context manager for use when defining a Python op.
This context manager pushes a name scope, which will make the name of all
operations added within it have a prefix.
For example, to define a new Python op called `my_op`:
```python
def my_op(a, b, c, name=None):
with tf.name_scope(""MyOp"") as scope:
a = tf.convert_to_tensor(a, name=""a"")
b = tf.convert_to_tensor(b, name=""b"")
c = tf.convert_to_tensor(c, name=""c"")
# Define some computation that uses `a`, `b`, and `c`.
return foo_op(..., name=scope)
```
When executed, the Tensors `a`, `b`, `c`, will have names `MyOp/a`, `MyOp/b`,
and `MyOp/c`.
Inside a `tf.function`, if the scope name already exists, the name will be
made unique by appending `_n`. For example, calling `my_op` the second time
will generate `MyOp_1/a`, etc."
4281,name,tensorflow/tensorflow/python/framework/ops.py,6646,method,
4282,strip_name_scope,tensorflow/tensorflow/python/framework/ops.py,6695,function,"Removes name scope from a name.
Args:
name: A `string` name.
export_scope: Optional `string`. Name scope to remove.
Returns:
Name with name scope removed, or the original name if export_scope
is None."
4283,prepend_name_scope,tensorflow/tensorflow/python/framework/ops.py,6723,function,"Prepends name scope to a name.
Args:
name: A `string` name.
import_scope: Optional `string`. Name scope to add.
Returns:
Name with name scope added, or the original name if import_scope
is None."
4284,op_scope,tensorflow/tensorflow/python/framework/ops.py,6754,function,"DEPRECATED. Same as name_scope above, just different argument order."
4285,register_proto_function,tensorflow/tensorflow/python/framework/ops.py,6765,function,"Registers `to_proto` and `from_proto` functions for collection_name.
`to_proto` function converts a Python object to the corresponding protocol
buffer, and returns the protocol buffer.
`from_proto` function converts protocol buffer into a Python object, and
returns the object..
Args:
collection_name: Name of the collection.
proto_type: Protobuf type, such as `saver_pb2.SaverDef`,
`variable_pb2.VariableDef`, `queue_runner_pb2.QueueRunnerDef`..
to_proto: Function that implements Python object to protobuf conversion.
from_proto: Function that implements protobuf to Python object conversion."
4286,get_collection_proto_type,tensorflow/tensorflow/python/framework/ops.py,6793,function,Returns the proto_type for collection_name.
4287,get_to_proto_function,tensorflow/tensorflow/python/framework/ops.py,6801,function,Returns the to_proto function for collection_name.
4288,get_from_proto_function,tensorflow/tensorflow/python/framework/ops.py,6809,function,Returns the from_proto function for collection_name.
4289,to_raw_op,tensorflow/tensorflow/python/framework/ops.py,6881,function,"Make a given op wrapper function `f` raw.
Raw op wrappers can only be called with keyword arguments.
Args:
f: An op wrapper function to make raw.
Returns:
Raw `f`."
4290,raise_from_not_ok_status,tensorflow/tensorflow/python/framework/ops.py,6899,function,
4291,add_exit_callback_to_default_func_graph,tensorflow/tensorflow/python/framework/ops.py,6906,function,"Add a callback to run when the default function graph goes out of scope.
Usage:
```python
@tf.function
def fn(x, v):
expensive = expensive_object(v)
add_exit_callback_to_default_func_graph(lambda: expensive.release())
return g(x, expensive)
fn(x=tf.constant(...), v=...)
# `expensive` has been released.
```
Args:
fn: A callable that takes no arguments and whose output is ignored.
To be executed when exiting func graph scope.
Raises:
RuntimeError: If executed when the current default graph is not a FuncGraph,
or not currently executing in function creation mode (e.g., if inside
an init_scope)."
4292,set_int_list_attr,tensorflow/tensorflow/python/framework/ops.py,6998,function,TF internal method used to set a list(int) attribute in the node_def.
4293,ObjectWithName,tensorflow/tensorflow/python/framework/ops_test.py,2057,class,
4294,name,tensorflow/tensorflow/python/framework/ops_test.py,2063,method,
4295,get_seed,tensorflow/tensorflow/python/framework/random_seed.py,43,function,"Returns the local seeds an operation should use given an op-specific seed.
Given operation-specific seed, `op_seed`, this helper function returns two
seeds derived from graph-level and op-level seeds. Many random operations
internally use the two seeds to allow user to change the seed globally for a
graph, or for only specific operations.
For details on how the graph-level seed interacts with op seeds, see
`tf.compat.v1.random.set_random_seed`.
Args:
op_seed: integer.
Returns:
A tuple of two integers that should be used for the local seed of this
operation."
4296,set_random_seed,tensorflow/tensorflow/python/framework/random_seed.py,93,function,"Sets the graph-level random seed for the default graph.
Operations that rely on a random seed actually derive it from two seeds:
the graph-level and operation-level seeds. This sets the graph-level seed.
Its interactions with operation-level seeds is as follows:
1. If neither the graph-level nor the operation seed is set:
A random seed is used for this op.
2. If the graph-level seed is set, but the operation seed is not:
The system deterministically picks an operation seed in conjunction with
the graph-level seed so that it gets a unique random sequence. Within the
same version of tensorflow and user code, this sequence is deterministic.
However across different versions, this sequence might change. If the
code depends on particular seeds to work, specify both graph-level
and operation-level seeds explicitly.
3. If the graph-level seed is not set, but the operation seed is set:
A default graph-level seed and the specified operation seed are used to
determine the random sequence.
4. If both the graph-level and the operation seed are set:
Both seeds are used in conjunction to determine the random sequence.
To illustrate the user-visible effects, consider these examples:
To generate different sequences across sessions, set neither
graph-level nor op-level seeds:
```python
a = tf.random.uniform([1])
b = tf.random.normal([1])
print(""Session 1"")
with tf.compat.v1.Session() as sess1:
print(sess1.run(a)) # generates 'A1'
print(sess1.run(a)) # generates 'A2'
print(sess1.run(b)) # generates 'B1'
print(sess1.run(b)) # generates 'B2'
print(""Session 2"")
with tf.compat.v1.Session() as sess2:
print(sess2.run(a)) # generates 'A3'
print(sess2.run(a)) # generates 'A4'
print(sess2.run(b)) # generates 'B3'
print(sess2.run(b)) # generates 'B4'
```
To generate the same repeatable sequence for an op across sessions, set the
seed for the op:
```python
a = tf.random.uniform([1], seed=1)
b = tf.random.normal([1])
# Repeatedly running this block with the same graph will generate the same
# sequence of values for 'a', but different sequences of values for 'b'.
print(""Session 1"")
with tf.compat.v1.Session() as sess1:
print(sess1.run(a)) # generates 'A1'
print(sess1.run(a)) # generates 'A2'
print(sess1.run(b)) # generates 'B1'
print(sess1.run(b)) # generates 'B2'
print(""Session 2"")
with tf.compat.v1.Session() as sess2:
print(sess2.run(a)) # generates 'A1'
print(sess2.run(a)) # generates 'A2'
print(sess2.run(b)) # generates 'B3'
print(sess2.run(b)) # generates 'B4'
```
To make the random sequences generated by all ops be repeatable across
sessions, set a graph-level seed:
```python
tf.compat.v1.random.set_random_seed(1234)
a = tf.random.uniform([1])
b = tf.random.normal([1])
# Repeatedly running this block with the same graph will generate the same
# sequences of 'a' and 'b'.
print(""Session 1"")
with tf.compat.v1.Session() as sess1:
print(sess1.run(a)) # generates 'A1'
print(sess1.run(a)) # generates 'A2'
print(sess1.run(b)) # generates 'B1'
print(sess1.run(b)) # generates 'B2'
print(""Session 2"")
with tf.compat.v1.Session() as sess2:
print(sess2.run(a)) # generates 'A1'
print(sess2.run(a)) # generates 'A2'
print(sess2.run(b)) # generates 'B1'
print(sess2.run(b)) # generates 'B2'
```
Args:
seed: integer."
4297,set_seed,tensorflow/tensorflow/python/framework/random_seed.py,199,function,"Sets the global random seed.
Operations that rely on a random seed actually derive it from two seeds:
the global and operation-level seeds. This sets the global seed.
Its interactions with operation-level seeds is as follows:
1. If neither the global seed nor the operation seed is set: A randomly
picked seed is used for this op.
2. If the graph-level seed is set, but the operation seed is not:
The system deterministically picks an operation seed in conjunction with
the graph-level seed so that it gets a unique random sequence. Within the
same version of tensorflow and user code, this sequence is deterministic.
However across different versions, this sequence might change. If the
code depends on particular seeds to work, specify both graph-level
and operation-level seeds explicitly.
3. If the operation seed is set, but the global seed is not set:
A default global seed and the specified operation seed are used to
determine the random sequence.
4. If both the global and the operation seed are set:
Both seeds are used in conjunction to determine the random sequence.
To illustrate the user-visible effects, consider these examples:
If neither the global seed nor the operation seed is set, we get different
results for every call to the random op and every re-run of the program:
```python
print(tf.random.uniform([1])) # generates 'A1'
print(tf.random.uniform([1])) # generates 'A2'
```
(now close the program and run it again)
```python
print(tf.random.uniform([1])) # generates 'A3'
print(tf.random.uniform([1])) # generates 'A4'
```
If the global seed is set but the operation seed is not set, we get different
results for every call to the random op, but the same sequence for every
re-run of the program:
```python
tf.random.set_seed(1234)
print(tf.random.uniform([1])) # generates 'A1'
print(tf.random.uniform([1])) # generates 'A2'
```
(now close the program and run it again)
```python
tf.random.set_seed(1234)
print(tf.random.uniform([1])) # generates 'A1'
print(tf.random.uniform([1])) # generates 'A2'
```
The reason we get 'A2' instead 'A1' on the second call of `tf.random.uniform`
above is because the second call uses a different operation seed.
Note that `tf.function` acts like a re-run of a program in this case. When
the global seed is set but operation seeds are not set, the sequence of random
numbers are the same for each `tf.function`. For example:
```python
tf.random.set_seed(1234)
@tf.function
def f():
a = tf.random.uniform([1])
b = tf.random.uniform([1])
return a, b
@tf.function
def g():
a = tf.random.uniform([1])
b = tf.random.uniform([1])
return a, b
print(f()) # prints '(A1, A2)'
print(g()) # prints '(A1, A2)'
```
If the operation seed is set, we get different results for every call to the
random op, but the same sequence for every re-run of the program:
```python
print(tf.random.uniform([1], seed=1)) # generates 'A1'
print(tf.random.uniform([1], seed=1)) # generates 'A2'
```
(now close the program and run it again)
```python
print(tf.random.uniform([1], seed=1)) # generates 'A1'
print(tf.random.uniform([1], seed=1)) # generates 'A2'
```
The reason we get 'A2' instead 'A1' on the second call of `tf.random.uniform`
above is because the same `tf.random.uniform` kernel (i.e. internal
representation) is used by TensorFlow for all calls of it with the same
arguments, and the kernel maintains an internal counter which is incremented
every time it is executed, generating different results.
Calling `tf.random.set_seed` will reset any such counters:
```python
tf.random.set_seed(1234)
print(tf.random.uniform([1], seed=1)) # generates 'A1'
print(tf.random.uniform([1], seed=1)) # generates 'A2'
tf.random.set_seed(1234)
print(tf.random.uniform([1], seed=1)) # generates 'A1'
print(tf.random.uniform([1], seed=1)) # generates 'A2'
```
When multiple identical random ops are wrapped in a `tf.function`, their
behaviors change because the ops no long share the same counter. For example:
```python
@tf.function
def foo():
a = tf.random.uniform([1], seed=1)
b = tf.random.uniform([1], seed=1)
return a, b
print(foo()) # prints '(A1, A1)'
print(foo()) # prints '(A2, A2)'
@tf.function
def bar():
a = tf.random.uniform([1])
b = tf.random.uniform([1])
return a, b
print(bar()) # prints '(A1, A2)'
print(bar()) # prints '(A3, A4)'
```
The second call of `foo` returns '(A2, A2)' instead of '(A1, A1)' because
`tf.random.uniform` maintains an internal counter. If you want `foo` to return
'(A1, A1)' every time, use the stateless random ops such as
`tf.random.stateless_uniform`. Also see `tf.random.experimental.Generator` for
a new set of stateful random ops that use external variables to manage their
states.
Args:
seed: integer."
4298,Registry,tensorflow/tensorflow/python/framework/registry.py,36,class,Provides a registry for saving objects.
4299,register,tensorflow/tensorflow/python/framework/registry.py,46,method,"Registers a Python object ""candidate"" for the given ""name"".
Args:
candidate: The candidate object to add to the registry.
name: An optional string specifying the registry key for the candidate.
If None, candidate.__name__ will be used.
Raises:
KeyError: If same name is used twice."
4300,list,tensorflow/tensorflow/python/framework/registry.py,76,method,"Lists registered items.
Returns:
A list of names of registered objects."
4301,lookup,tensorflow/tensorflow/python/framework/registry.py,84,method,"Looks up ""name"".
Args:
name: a string specifying the registry key for the candidate.
Returns:
Registered object if found
Raises:
LookupError: if ""name"" has not been registered."
4302,bar,tensorflow/tensorflow/python/framework/registry_test.py,28,function,
4303,smart_cond,tensorflow/tensorflow/python/framework/smart_cond.py,27,function,"Return either `true_fn()` if predicate `pred` is true else `false_fn()`.
If `pred` is a bool or has a constant value, we return either `true_fn()`
or `false_fn()`, otherwise we use `tf.cond` to dynamically route to both.
Arguments:
pred: A scalar determining whether to return the result of `true_fn` or
`false_fn`.
true_fn: The callable to be performed if pred is true.
false_fn: The callable to be performed if pred is false.
name: Optional name prefix when using `tf.cond`.
Returns:
Tensors returned by the call to either `true_fn` or `false_fn`.
Raises:
TypeError: If `true_fn` or `false_fn` is not callable."
4304,smart_constant_value,tensorflow/tensorflow/python/framework/smart_cond.py,62,function,"Return the bool value for `pred`, or None if `pred` had a dynamic value.
Arguments:
pred: A scalar, either a Python bool or tensor.
Returns:
True or False if `pred` has a constant boolean value, None otherwise.
Raises:
TypeError: If `pred` is not a Tensor or bool."
4305,smart_case,tensorflow/tensorflow/python/framework/smart_cond.py,93,function,"Like tf.case, except attempts to statically evaluate predicates.
If any predicate in `pred_fn_pairs` is a bool or has a constant value, the
associated callable will be called or omitted depending on its value.
Otherwise this functions like tf.case.
Args:
pred_fn_pairs: Dict or list of pairs of a boolean scalar tensor and a
callable which returns a list of tensors.
default: Optional callable that returns a list of tensors.
exclusive: True iff at most one predicate is allowed to evaluate to `True`.
name: A name for this operation (optional).
Returns:
The tensors returned by the first pair whose predicate evaluated to True, or
those returned by `default` if none does.
Raises:
TypeError: If `pred_fn_pairs` is not a list/dictionary.
TypeError: If `pred_fn_pairs` is a list but does not contain 2-tuples.
TypeError: If `fns[i]` is not callable for any i, or `default` is not
callable."
4306,raise_exception,tensorflow/tensorflow/python/framework/smart_cond_test.py,32,function,
4307,SparseTensor,tensorflow/tensorflow/python/framework/sparse_tensor.py,47,class,"Represents a sparse tensor.
TensorFlow represents a sparse tensor as three separate dense tensors:
`indices`, `values`, and `dense_shape`. In Python, the three tensors are
collected into a `SparseTensor` class for ease of use. If you have separate
`indices`, `values`, and `dense_shape` tensors, wrap them in a `SparseTensor`
object before passing to the ops below.
Concretely, the sparse tensor `SparseTensor(indices, values, dense_shape)`
comprises the following components, where `N` and `ndims` are the number
of values and number of dimensions in the `SparseTensor`, respectively:
* `indices`: A 2-D int64 tensor of shape `[N, ndims]`, which specifies the
indices of the elements in the sparse tensor that contain nonzero values
(elements are zero-indexed). For example, `indices=[[1,3], [2,4]]` specifies
that the elements with indexes of [1,3] and [2,4] have nonzero values.
* `values`: A 1-D tensor of any type and shape `[N]`, which supplies the
values for each element in `indices`. For example, given `indices=[[1,3],
[2,4]]`, the parameter `values=[18, 3.6]` specifies that element [1,3] of
the sparse tensor has a value of 18, and element [2,4] of the tensor has a
value of 3.6.
* `dense_shape`: A 1-D int64 tensor of shape `[ndims]`, which specifies the
dense_shape of the sparse tensor. Takes a list indicating the number of
elements in each dimension. For example, `dense_shape=[3,6]` specifies a
two-dimensional 3x6 tensor, `dense_shape=[2,3,4]` specifies a
three-dimensional 2x3x4 tensor, and `dense_shape=[9]` specifies a
one-dimensional tensor with 9 elements.
The corresponding dense tensor satisfies:
```python
dense.shape = dense_shape
dense[tuple(indices[i])] = values[i]
```
By convention, `indices` should be sorted in row-major order (or equivalently
lexicographic order on the tuples `indices[i]`). This is not enforced when
`SparseTensor` objects are constructed, but most ops assume correct ordering.
If the ordering of sparse tensor `st` is wrong, a fixed version can be
obtained by calling `tf.sparse.reorder(st)`.
Example: The sparse tensor
```python
SparseTensor(indices=[[0, 0], [1, 2]], values=[1, 2], dense_shape=[3, 4])
```
represents the dense tensor
```python
[[1, 0, 0, 0]
[0, 0, 2, 0]
[0, 0, 0, 0]]
```"
4308,from_value,tensorflow/tensorflow/python/framework/sparse_tensor.py,107,method,
4309,get_shape,tensorflow/tensorflow/python/framework/sparse_tensor.py,154,method,"Get the `TensorShape` representing the shape of the dense tensor.
Returns:
A `TensorShape` object."
4310,indices,tensorflow/tensorflow/python/framework/sparse_tensor.py,163,method,"The indices of non-zero values in the represented dense tensor.
Returns:
A 2-D Tensor of int64 with dense_shape `[N, ndims]`, where `N` is the
number of non-zero values in the tensor, and `ndims` is the rank."
4311,values,tensorflow/tensorflow/python/framework/sparse_tensor.py,173,method,"The non-zero values in the represented dense tensor.
Returns:
A 1-D Tensor of any data type."
4312,op,tensorflow/tensorflow/python/framework/sparse_tensor.py,182,method,The `Operation` that produces `values` as an output.
4313,dtype,tensorflow/tensorflow/python/framework/sparse_tensor.py,187,method,The `DType` of elements in this tensor.
4314,dense_shape,tensorflow/tensorflow/python/framework/sparse_tensor.py,192,method,A 1-D Tensor of int64 representing the shape of the dense tensor.
4315,shape,tensorflow/tensorflow/python/framework/sparse_tensor.py,197,method,"Get the `TensorShape` representing the shape of the dense tensor.
Returns:
A `TensorShape` object."
4316,graph,tensorflow/tensorflow/python/framework/sparse_tensor.py,206,method,"The `Graph` that contains the index, value, and dense_shape tensors."
4317,eval,tensorflow/tensorflow/python/framework/sparse_tensor.py,214,method,"Evaluates this sparse tensor in a `Session`.
Calling this method will execute all preceding operations that
produce the inputs needed for the operation that produces this
tensor.
*N.B.* Before invoking `SparseTensor.eval()`, its graph must have been
launched in a session, and either a default session must be
available, or `session` must be specified explicitly.
Args:
feed_dict: A dictionary that maps `Tensor` objects to feed values. See
`tf.Session.run` for a description of the valid feed values.
session: (Optional.) The `Session` to be used to evaluate this sparse
tensor. If none, the default session will be used.
Returns:
A `SparseTensorValue` object."
4318,consumers,tensorflow/tensorflow/python/framework/sparse_tensor.py,259,method,
4319,SparseTensorSpec,tensorflow/tensorflow/python/framework/sparse_tensor.py,270,class,Type specification for a `tf.sparse.SparseTensor`.
4320,dtype,tensorflow/tensorflow/python/framework/sparse_tensor.py,292,method,The `tf.dtypes.DType` specified by this type for the SparseTensor.
4321,shape,tensorflow/tensorflow/python/framework/sparse_tensor.py,297,method,The `tf.TensorShape` specified by this type for the SparseTensor.
4322,from_value,tensorflow/tensorflow/python/framework/sparse_tensor.py,393,method,
4323,convert_to_tensor_or_sparse_tensor,tensorflow/tensorflow/python/framework/sparse_tensor.py,415,function,"Converts value to a `SparseTensor` or `Tensor`.
Args:
value: A `SparseTensor`, `SparseTensorValue`, or an object whose type has a
registered `Tensor` conversion function.
dtype: Optional element type for the returned tensor. If missing, the type
is inferred from the type of `value`.
name: Optional name to use if a new `Tensor` is created.
Returns:
A `SparseTensor` or `Tensor` based on `value`.
Raises:
RuntimeError: If result type is incompatible with `dtype`."
4324,is_sparse,tensorflow/tensorflow/python/framework/sparse_tensor.py,443,function,"Check whether `x` is sparse.
Check whether an object is a `tf.sparse.SparseTensor` or
`tf.compat.v1.SparseTensorValue`.
Args:
x: A python object to check.
Returns:
`True` iff `x` is a `tf.sparse.SparseTensor` or
`tf.compat.v1.SparseTensorValue`."
4325,subscribe,tensorflow/tensorflow/python/framework/subscribe.py,313,function,"Subscribe to a tensor.
This method will attach side effect graphs to a given set
of tensors. Set of tensors follows from session.run and supports
single `Tensor`, `list`, nested `list`, `tuple`, `namedtuple`, or `dict`. It
returns the tensors in the same passed in structure, but as clones with
side effects applied. The supplied side effect graphs are specified
as a constructor function which takes the target tensor and
constructs a side effect graph and returns a list of ops that should
be control dependencies on fetching the tensor. It will append
'subscription' to the name scope of the tensor for every node in
the side effect graph. These control dependencies are what trigger
the side effects. Subscribe will construct the additions to your
graph and return the created identity tensor downstream of the control
dependencies. Use these tensors as you would normally in the rest of
your tensorflow code. If a given tensor has already been subscribed or a
tensor returned by a call to subscribe is passed, the previously created
identity tensor will be reused and the side effect graphs will be added to
the existing ones.
Args:
tensors: `Tensor` or set of tensors to subscribe to. Set of tensors format
follows from `Session.run` and supports single `Tensor`, `list`, nested
`list`, `tuple`, `namedtuple`, or `dict`.
side_effects: Function(s) that takes a `Tensor`, construct a subgraph, and
return a nonempty list of control dependencies. This can be a single
function or list of functions.
Returns:
Subscribed tensors, which are identity copies of the passed in tensors
in the same passed in structure, but the graph has been modified
such that these are downstream of the control dependencies for
the side effect graphs. Use these functionally equivalent tensors
instead of the passed in tensors for further construction or running."
4326,register_tensor_conversion_function,tensorflow/tensorflow/python/framework/tensor_conversion_registry.py,57,function,"Registers a function for converting objects of `base_type` to `Tensor`.
The conversion function must have the following signature:
```python
def conversion_func(value, dtype=None, name=None, as_ref=False):
# ...
```
It must return a `Tensor` with the given `dtype` if specified. If the
conversion function creates a new `Tensor`, it should use the given
`name` if specified. All exceptions will be propagated to the caller.
The conversion function may return `NotImplemented` for some
inputs. In this case, the conversion process will continue to try
subsequent conversion functions.
If `as_ref` is true, the function must return a `Tensor` reference,
such as a `Variable`.
NOTE: The conversion functions will execute in order of priority,
followed by order of registration. To ensure that a conversion function
`F` runs before another conversion function `G`, ensure that `F` is
registered with a smaller priority than `G`.
Args:
base_type: The base type or tuple of base types for all objects that
`conversion_func` accepts.
conversion_func: A function that converts instances of `base_type` to
`Tensor`.
priority: Optional integer that indicates the priority for applying this
conversion function. Conversion functions with smaller priority values run
earlier than conversion functions with larger priority values. Defaults to
100.
Raises:
TypeError: If the arguments do not have the appropriate type."
4327,get,tensorflow/tensorflow/python/framework/tensor_conversion_registry.py,114,function,"Get conversion function for objects of `cls`.
Args:
query: The type to query for.
Returns:
A list of conversion functions in increasing order of priority."
4328,enable_v2_tensorshape,tensorflow/tensorflow/python/framework/tensor_shape.py,35,function,"In TensorFlow 2.0, iterating over a TensorShape instance returns values.
This enables the new behavior.
Concretely, `tensor_shape[i]` returned a Dimension instance in V1, but
it V2 it returns either an integer, or None.
Examples:
```
#######################
# If you had this in V1:
value = tensor_shape[i].value
# Do this in V2 instead:
value = tensor_shape[i]
#######################
# If you had this in V1:
for dim in tensor_shape:
value = dim.value
print(value)
# Do this in V2 instead:
for value in tensor_shape:
print(value)
#######################
# If you had this in V1:
dim = tensor_shape[i]
dim.assert_is_compatible_with(other_shape) # or using any other shape method
# Do this in V2 instead:
if tensor_shape.rank is None:
dim = Dimension(None)
else:
dim = tensor_shape.dims[i]
dim.assert_is_compatible_with(other_shape) # or using any other shape method
# The V2 suggestion above is more explicit, which will save you from
# the following trap (present in V1):
# you might do in-place modifications to `dim` and expect them to be reflected
# in `tensor_shape[i]`, but they would not be.
```"
4329,disable_v2_tensorshape,tensorflow/tensorflow/python/framework/tensor_shape.py,87,function,"Disables the V2 TensorShape behavior and reverts to V1 behavior.
See docstring for `enable_v2_tensorshape` for details about the new behavior."
4330,dimension_value,tensorflow/tensorflow/python/framework/tensor_shape.py,99,function,"Compatibility utility required to allow for both V1 and V2 behavior in TF.
Until the release of TF 2.0, we need the legacy behavior of `TensorShape` to
coexist with the new behavior. This utility is a bridge between the two.
When accessing the value of a TensorShape dimension,
use this utility, like this:
```
# If you had this in your V1 code:
value = tensor_shape[i].value
# Use `dimension_value` as direct replacement compatible with both V1 & V2:
value = dimension_value(tensor_shape[i])
# This would be the V2 equivalent:
value = tensor_shape[i] # Warning: this will return the dim value in V2!
```
Arguments:
dimension: Either a `Dimension` instance, an integer, or None.
Returns:
A plain value, i.e. an integer or None."
4331,dimension_at_index,tensorflow/tensorflow/python/framework/tensor_shape.py,133,function,"Compatibility utility required to allow for both V1 and V2 behavior in TF.
Until the release of TF 2.0, we need the legacy behavior of `TensorShape` to
coexist with the new behavior. This utility is a bridge between the two.
If you want to retrieve the Dimension instance corresponding to a certain
index in a TensorShape instance, use this utility, like this:
```
# If you had this in your V1 code:
dim = tensor_shape[i]
# Use `dimension_at_index` as direct replacement compatible with both V1 & V2:
dim = dimension_at_index(tensor_shape, i)
# Another possibility would be this, but WARNING: it only works if the
# tensor_shape instance has a defined rank.
dim = tensor_shape.dims[i] # `dims` may be None if the rank is undefined!
# In native V2 code, we recommend instead being more explicit:
if tensor_shape.rank is None:
dim = Dimension(None)
else:
dim = tensor_shape.dims[i]
# Being more explicit will save you from the following trap (present in V1):
# you might do in-place modifications to `dim` and expect them to be reflected
# in `tensor_shape[i]`, but they would not be (as the Dimension object was
# instantiated on the fly.
```
Arguments:
shape: A TensorShape instance.
index: An integer index.
Returns:
A dimension object."
4332,Dimension,tensorflow/tensorflow/python/framework/tensor_shape.py,180,class,Represents the value of one dimension in a TensorShape.
4333,value,tensorflow/tensorflow/python/framework/tensor_shape.py,248,method,"The value of this dimension, or None if it is unknown."
4334,is_compatible_with,tensorflow/tensorflow/python/framework/tensor_shape.py,252,method,"Returns true if `other` is compatible with this Dimension.
Two known Dimensions are compatible if they have the same value.
An unknown Dimension is compatible with all other Dimensions.
Args:
other: Another Dimension.
Returns:
True if this Dimension and `other` are compatible."
4335,assert_is_compatible_with,tensorflow/tensorflow/python/framework/tensor_shape.py,268,method,"Raises an exception if `other` is not compatible with this Dimension.
Args:
other: Another Dimension.
Raises:
ValueError: If `self` and `other` are not compatible (see
is_compatible_with)."
4336,merge_with,tensorflow/tensorflow/python/framework/tensor_shape.py,282,method,"Returns a Dimension that combines the information in `self` and `other`.
Dimensions are combined as follows:
```python
tf.compat.v1.Dimension(n) .merge_with(tf.compat.v1.Dimension(n)) ==
tf.compat.v1.Dimension(n)
tf.compat.v1.Dimension(n) .merge_with(tf.compat.v1.Dimension(None)) ==
tf.compat.v1.Dimension(n)
tf.compat.v1.Dimension(None).merge_with(tf.compat.v1.Dimension(n)) ==
tf.compat.v1.Dimension(n)
# equivalent to tf.compat.v1.Dimension(None)
tf.compat.v1.Dimension(None).merge_with(tf.compat.v1.Dimension(None))
# raises ValueError for n != m
tf.compat.v1.Dimension(n) .merge_with(tf.compat.v1.Dimension(m))
```
Args:
other: Another Dimension.
Returns:
A Dimension containing the combined information of `self` and
`other`.
Raises:
ValueError: If `self` and `other` are not compatible (see
is_compatible_with)."
4337,as_dimension,tensorflow/tensorflow/python/framework/tensor_shape.py,704,function,"Converts the given value to a Dimension.
A Dimension input will be returned unmodified.
An input of `None` will be converted to an unknown Dimension.
An integer input will be converted to a Dimension with that value.
Args:
value: The value to be converted.
Returns:
A Dimension corresponding to the given value."
4338,TensorShape,tensorflow/tensorflow/python/framework/tensor_shape.py,724,class,"Represents the shape of a `Tensor`.
A `TensorShape` represents a possibly-partial shape specification for a
`Tensor`. It may be one of the following:
* *Fully-known shape:* has a known number of dimensions and a known size
for each dimension. e.g. `TensorShape([16, 256])`
* *Partially-known shape:* has a known number of dimensions, and an unknown
size for one or more dimension. e.g. `TensorShape([None, 256])`
* *Unknown shape:* has an unknown number of dimensions, and an unknown
size in all dimensions. e.g. `TensorShape(None)`
If a tensor is produced by an operation of type `""Foo""`, its shape
may be inferred if there is a registered shape function for
`""Foo""`. See [Shape
functions](https://tensorflow.org/extend/adding_an_op#shape_functions_in_c)
for details of shape functions and how to register them. Alternatively,
the shape may be set explicitly using `tf.Tensor.set_shape`."
4339,rank,tensorflow/tensorflow/python/framework/tensor_shape.py,820,method,"Returns the rank of this shape, or None if it is unspecified."
4340,dims,tensorflow/tensorflow/python/framework/tensor_shape.py,827,method,"Deprecated. Returns list of dimensions for this shape.
Suggest `TensorShape.as_list` instead.
Returns:
A list containing `tf.compat.v1.Dimension`s, or None if the shape is
unspecified."
4341,ndims,tensorflow/tensorflow/python/framework/tensor_shape.py,839,method,Deprecated accessor for `rank`.
4342,num_elements,tensorflow/tensorflow/python/framework/tensor_shape.py,916,method,"Returns the total number of elements, or none for incomplete shapes."
4343,merge_with,tensorflow/tensorflow/python/framework/tensor_shape.py,926,method,"Returns a `TensorShape` combining the information in `self` and `other`.
The dimensions in `self` and `other` are merged elementwise,
according to the rules defined for `Dimension.merge_with()`.
Args:
other: Another `TensorShape`.
Returns:
A `TensorShape` containing the combined information of `self` and
`other`.
Raises:
ValueError: If `self` and `other` are not compatible."
4344,concatenate,tensorflow/tensorflow/python/framework/tensor_shape.py,965,method,"Returns the concatenation of the dimension in `self` and `other`.
*N.B.* If either `self` or `other` is completely unknown,
concatenation will discard information about the other shape. In
future, we might support concatenation that preserves this
information for use with slicing.
Args:
other: Another `TensorShape`.
Returns:
A `TensorShape` whose dimensions are the concatenation of the
dimensions in `self` and `other`."
4345,assert_same_rank,tensorflow/tensorflow/python/framework/tensor_shape.py,988,method,"Raises an exception if `self` and `other` do not have compatible ranks.
Args:
other: Another `TensorShape`.
Raises:
ValueError: If `self` and `other` do not represent shapes with the
same rank."
4346,assert_has_rank,tensorflow/tensorflow/python/framework/tensor_shape.py,1004,method,"Raises an exception if `self` is not compatible with the given `rank`.
Args:
rank: An integer.
Raises:
ValueError: If `self` does not represent a shape with the given `rank`."
4347,with_rank,tensorflow/tensorflow/python/framework/tensor_shape.py,1016,method,"Returns a shape based on `self` with the given rank.
This method promotes a completely unknown shape to one with a
known rank.
Args:
rank: An integer.
Returns:
A shape that is at least as specific as `self` with the given rank.
Raises:
ValueError: If `self` does not represent a shape with the given `rank`."
4348,with_rank_at_least,tensorflow/tensorflow/python/framework/tensor_shape.py,1036,method,"Returns a shape based on `self` with at least the given rank.
Args:
rank: An integer.
Returns:
A shape that is at least as specific as `self` with at least the given
rank.
Raises:
ValueError: If `self` does not represent a shape with at least the given
`rank`."
4349,with_rank_at_most,tensorflow/tensorflow/python/framework/tensor_shape.py,1055,method,"Returns a shape based on `self` with at most the given rank.
Args:
rank: An integer.
Returns:
A shape that is at least as specific as `self` with at most the given
rank.
Raises:
ValueError: If `self` does not represent a shape with at most the given
`rank`."
4350,is_compatible_with,tensorflow/tensorflow/python/framework/tensor_shape.py,1074,method,"Returns True iff `self` is compatible with `other`.
Two possibly-partially-defined shapes are compatible if there
exists a fully-defined shape that both shapes can represent. Thus,
compatibility allows the shape inference code to reason about
partially-defined shapes. For example:
* TensorShape(None) is compatible with all shapes.
* TensorShape([None, None]) is compatible with all two-dimensional
shapes, such as TensorShape([32, 784]), and also TensorShape(None). It is
not compatible with, for example, TensorShape([None]) or
TensorShape([None, None, None]).
* TensorShape([32, None]) is compatible with all two-dimensional shapes
with size 32 in the 0th dimension, and also TensorShape([None, None])
and TensorShape(None). It is not compatible with, for example,
TensorShape([32]), TensorShape([32, None, 1]) or TensorShape([64, None]).
* TensorShape([32, 784]) is compatible with itself, and also
TensorShape([32, None]), TensorShape([None, 784]), TensorShape([None,
None]) and TensorShape(None). It is not compatible with, for example,
TensorShape([32, 1, 784]) or TensorShape([None]).
The compatibility relation is reflexive and symmetric, but not
transitive. For example, TensorShape([32, 784]) is compatible with
TensorShape(None), and TensorShape(None) is compatible with
TensorShape([4, 4]), but TensorShape([32, 784]) is not compatible with
TensorShape([4, 4]).
Args:
other: Another TensorShape.
Returns:
True iff `self` is compatible with `other`."
4351,assert_is_compatible_with,tensorflow/tensorflow/python/framework/tensor_shape.py,1121,method,"Raises exception if `self` and `other` do not represent the same shape.
This method can be used to assert that there exists a shape that both
`self` and `other` represent.
Args:
other: Another TensorShape.
Raises:
ValueError: If `self` and `other` do not represent the same shape."
4352,most_specific_compatible_shape,tensorflow/tensorflow/python/framework/tensor_shape.py,1136,method,"Returns the most specific TensorShape compatible with `self` and `other`.
* TensorShape([None, 1]) is the most specific TensorShape compatible with
both TensorShape([2, 1]) and TensorShape([5, 1]). Note that
TensorShape(None) is also compatible with above mentioned TensorShapes.
* TensorShape([1, 2, 3]) is the most specific TensorShape compatible with
both TensorShape([1, 2, 3]) and TensorShape([1, 2, 3]). There are more
less specific TensorShapes compatible with above mentioned TensorShapes,
e.g. TensorShape([1, 2, None]), TensorShape(None).
Args:
other: Another `TensorShape`.
Returns:
A `TensorShape` which is the most specific compatible shape of `self`
and `other`."
4353,is_fully_defined,tensorflow/tensorflow/python/framework/tensor_shape.py,1166,method,Returns True iff `self` is fully defined in every dimension.
4354,assert_is_fully_defined,tensorflow/tensorflow/python/framework/tensor_shape.py,1171,method,"Raises an exception if `self` is not fully defined in every dimension.
Raises:
ValueError: If `self` does not have a known value for every dimension."
4355,as_list,tensorflow/tensorflow/python/framework/tensor_shape.py,1180,method,"Returns a list of integers or `None` for each dimension.
Returns:
A list of integers or `None` for each dimension.
Raises:
ValueError: If `self` is an unknown shape with an unknown rank."
4356,as_proto,tensorflow/tensorflow/python/framework/tensor_shape.py,1193,method,Returns this shape as a `TensorShapeProto`.
4357,as_shape,tensorflow/tensorflow/python/framework/tensor_shape.py,1230,function,Converts the given object to a TensorShape.
4358,unknown_shape,tensorflow/tensorflow/python/framework/tensor_shape.py,1238,function,"Returns an unknown TensorShape, optionally with a known rank.
Args:
rank: (Optional) If specified, the number of dimensions in the shape.
**kwargs: For backwards compatibility.
Returns:
An unknown TensorShape.
Raises:
TypeError: In case of invalid arguments."
4359,DenseSpec,tensorflow/tensorflow/python/framework/tensor_spec.py,32,class,"Describes a dense object with shape, dtype, and name."
4360,from_spec,tensorflow/tensorflow/python/framework/tensor_spec.py,60,method,
4361,shape,tensorflow/tensorflow/python/framework/tensor_spec.py,64,method,Returns the `TensorShape` that represents the shape of the tensor.
4362,dtype,tensorflow/tensorflow/python/framework/tensor_spec.py,69,method,Returns the `dtype` of elements in the tensor.
4363,name,tensorflow/tensorflow/python/framework/tensor_spec.py,74,method,Returns the (optionally provided) name of the described tensor.
4364,is_compatible_with,tensorflow/tensorflow/python/framework/tensor_spec.py,78,method,
4365,most_specific_compatible_type,tensorflow/tensorflow/python/framework/tensor_spec.py,100,method,
4366,TensorSpec,tensorflow/tensorflow/python/framework/tensor_spec.py,121,class,"Describes a tf.Tensor.
Metadata for describing the `tf.Tensor` objects accepted or returned
by some TensorFlow APIs."
4367,is_compatible_with,tensorflow/tensorflow/python/framework/tensor_spec.py,130,method,"Returns True if spec_or_tensor is compatible with this TensorSpec.
Two tensors are considered compatible if they have the same dtype
and their shapes are compatible (see `tf.TensorShape.is_compatible_with`).
Args:
spec_or_tensor: A tf.TensorSpec or a tf.Tensor
Returns:
True if spec_or_tensor is compatible with self."
4368,from_tensor,tensorflow/tensorflow/python/framework/tensor_spec.py,145,method,
4369,BoundedTensorSpec,tensorflow/tensorflow/python/framework/tensor_spec.py,197,class,"A `TensorSpec` that specifies minimum and maximum values.
Example usage:
```python
spec = tensor_spec.BoundedTensorSpec((1, 2, 3), tf.float32, 0, (5, 5, 5))
tf_minimum = tf.convert_to_tensor(spec.minimum, dtype=spec.dtype)
tf_maximum = tf.convert_to_tensor(spec.maximum, dtype=spec.dtype)
```
Bounds are meant to be inclusive. This is especially important for
integer types. The following spec will be satisfied by tensors
with values in the set {0, 1, 2}:
```python
spec = tensor_spec.BoundedTensorSpec((3, 5), tf.int32, 0, 2)
```"
4370,from_spec,tensorflow/tensorflow/python/framework/tensor_spec.py,265,method,
4371,minimum,tensorflow/tensorflow/python/framework/tensor_spec.py,272,method,Returns a NumPy array specifying the minimum bounds (inclusive).
4372,maximum,tensorflow/tensorflow/python/framework/tensor_spec.py,277,method,Returns a NumPy array specifying the maximum bounds (inclusive).
4373,ExtractBitsFromFloat16,tensorflow/tensorflow/python/framework/tensor_util.py,46,function,
4374,SlowAppendFloat16ArrayToTensorProto,tensorflow/tensorflow/python/framework/tensor_util.py,50,function,
4375,ExtractBitsFromBFloat16,tensorflow/tensorflow/python/framework/tensor_util.py,62,function,
4376,SlowAppendBFloat16ArrayToTensorProto,tensorflow/tensorflow/python/framework/tensor_util.py,67,function,
4377,FastAppendBFloat16ArrayToTensorProto,tensorflow/tensorflow/python/framework/tensor_util.py,72,function,
4378,SlowAppendFloat32ArrayToTensorProto,tensorflow/tensorflow/python/framework/tensor_util.py,126,function,
4379,SlowAppendFloat64ArrayToTensorProto,tensorflow/tensorflow/python/framework/tensor_util.py,129,function,
4380,SlowAppendIntArrayToTensorProto,tensorflow/tensorflow/python/framework/tensor_util.py,132,function,
4381,SlowAppendInt64ArrayToTensorProto,tensorflow/tensorflow/python/framework/tensor_util.py,135,function,
4382,SlowAppendQIntArrayToTensorProto,tensorflow/tensorflow/python/framework/tensor_util.py,138,function,
4383,SlowAppendUInt32ArrayToTensorProto,tensorflow/tensorflow/python/framework/tensor_util.py,141,function,
4384,SlowAppendUInt64ArrayToTensorProto,tensorflow/tensorflow/python/framework/tensor_util.py,144,function,
4385,SlowAppendComplex64ArrayToTensorProto,tensorflow/tensorflow/python/framework/tensor_util.py,147,function,
4386,SlowAppendComplex128ArrayToTensorProto,tensorflow/tensorflow/python/framework/tensor_util.py,151,function,
4387,SlowAppendObjectArrayToTensorProto,tensorflow/tensorflow/python/framework/tensor_util.py,155,function,
4388,SlowAppendBoolArrayToTensorProto,tensorflow/tensorflow/python/framework/tensor_util.py,158,function,
4389,GetFromNumpyDTypeDict,tensorflow/tensorflow/python/framework/tensor_util.py,187,function,
4390,GetNumpyAppendFn,tensorflow/tensorflow/python/framework/tensor_util.py,195,function,
4391,TensorShapeProtoToList,tensorflow/tensorflow/python/framework/tensor_util.py,208,function,"Convert a TensorShape to a list.
Args:
shape: A TensorShapeProto.
Returns:
List of integers representing the dimensions of the tensor."
4392,make_tensor_proto,tensorflow/tensorflow/python/framework/tensor_util.py,362,function,"Create a TensorProto.
In TensorFlow 2.0, representing tensors as protos should no longer be a
common workflow. That said, this utility function is still useful for
generating TF Serving request protos:
```python
request = tensorflow_serving.apis.predict_pb2.PredictRequest()
request.model_spec.name = ""my_model""
request.model_spec.signature_name = ""serving_default""
request.inputs[""images""].CopyFrom(tf.make_tensor_proto(X_new))
```
`make_tensor_proto` accepts ""values"" of a python scalar, a python list, a
numpy ndarray, or a numpy scalar.
If ""values"" is a python scalar or a python list, make_tensor_proto
first convert it to numpy ndarray. If dtype is None, the
conversion tries its best to infer the right numpy data
type. Otherwise, the resulting numpy array has a compatible data
type with the given dtype.
In either case above, the numpy ndarray (either the caller provided
or the auto-converted) must have the compatible type with dtype.
`make_tensor_proto` then converts the numpy array to a tensor proto.
If ""shape"" is None, the resulting tensor proto represents the numpy
array precisely.
Otherwise, ""shape"" specifies the tensor's shape and the numpy array
can not have more elements than what ""shape"" specifies.
Args:
values: Values to put in the TensorProto.
dtype: Optional tensor_pb2 DataType value.
shape: List of integers representing the dimensions of tensor.
verify_shape: Boolean that enables verification of a shape of values.
allow_broadcast: Boolean that enables allowing scalars and 1 length vector
broadcasting. Cannot be true when verify_shape is true.
Returns:
A `TensorProto`. Depending on the type, it may contain data in the
""tensor_content"" attribute, which is not directly useful to Python programs.
To access the values you should convert the proto back to a numpy ndarray
with `tf.make_ndarray(proto)`.
If `values` is a `TensorProto`, it is immediately returned; `dtype` and
`shape` are ignored.
Raises:
TypeError: if unsupported types are provided.
ValueError: if arguments have inappropriate values or if verify_shape is
True and shape of values is not equals to a shape from the argument."
4393,MakeNdarray,tensorflow/tensorflow/python/framework/tensor_util.py,571,function,"Create a numpy ndarray from a tensor.
Create a numpy ndarray with the same shape and data as the tensor.
For example:
```python
# Tensor a has shape (2,3)
a = tf.constant([[1,2,3],[4,5,6]])
proto_tensor = tf.make_tensor_proto(a) # convert `tensor a` to a proto tensor
tf.make_ndarray(proto_tensor) # output: array([[1, 2, 3],
# [4, 5, 6]], dtype=int32)
# output has shape (2,3)
```
Args:
tensor: A TensorProto.
Returns:
A numpy array with the tensor contents.
Raises:
TypeError: if tensor has unsupported type."
4394,ShapeEquals,tensorflow/tensorflow/python/framework/tensor_util.py,651,function,"Returns True if ""tensor_proto"" has the given ""shape"".
Args:
tensor_proto: A TensorProto.
shape: A tensor shape, expressed as a TensorShape, list, or tuple.
Returns:
True if ""tensor_proto"" has the given ""shape"", otherwise False.
Raises:
TypeError: If ""tensor_proto"" is not a TensorProto, or shape is not a
TensorShape, list, or tuple."
4395,constant_value,tensorflow/tensorflow/python/framework/tensor_util.py,806,function,"Returns the constant value of the given tensor, if efficiently calculable.
This function attempts to partially evaluate the given tensor, and
returns its value as a numpy ndarray if this succeeds.
Compatibility(V1): If `constant_value(tensor)` returns a non-`None` result, it
will no longer be possible to feed a different value for `tensor`. This allows
the result of this function to influence the graph that is constructed, and
permits static shape optimizations.
Args:
tensor: The Tensor to be evaluated.
partial: If True, the returned numpy array is allowed to have partially
evaluated values. Values that can't be evaluated will be None.
Returns:
A numpy ndarray containing the constant value of the given `tensor`,
or None if it cannot be calculated.
Raises:
TypeError: if tensor is not an ops.Tensor."
4396,constant_value_as_shape,tensorflow/tensorflow/python/framework/tensor_util.py,848,function,"A version of `constant_value()` that returns a `TensorShape`.
This version should be used when a constant tensor value is
interpreted as a (possibly partial) shape, e.g. in the shape
function for `tf.reshape()`. By explicitly requesting a
`TensorShape` as the return value, it is possible to represent
unknown dimensions; by contrast, `constant_value()` is
all-or-nothing.
Args:
tensor: The rank-0 or rank-1 Tensor to be evaluated.
Returns:
A `TensorShape` based on the constant value of the given `tensor`.
Raises:
ValueError: If the shape is rank-0 and is not statically known to be -1."
4397,is_tensor,tensorflow/tensorflow/python/framework/tensor_util.py,993,function,"Checks whether `x` is a TF-native type that can be passed to many TF ops.
Use is_tensor to differentiate types that can ingested by TensorFlow ops
without any conversion (e.g., `tf.Tensor`, `tf.SparseTensor`, and
`tf.RaggedTensor`) from types that need to be converted into tensors before
they are ingested (e.g., numpy `ndarray` and Python scalars).
For example, in the following code block:
```python
if not tf.is_tensor(t):
t = tf.convert_to_tensor(t)
return t.dtype
```
we check to make sure that `t` is a tensor (and convert it if not) before
accessing its `shape` and `dtype`.
Args:
x: A python object to check.
Returns:
`True` if `x` is a tensor or ""tensor-like"", `False` if not."
4398,shape_tensor,tensorflow/tensorflow/python/framework/tensor_util.py,1023,function,"Convert to an int32 or int64 tensor, defaulting to int32 if empty."
4399,maybe_set_static_shape,tensorflow/tensorflow/python/framework/tensor_util.py,1042,function,"Sets the shape of `tensor` to the `shape`'s constant value, if inferrable.
This is a temporary workaround to fix shape inference across functional op
boundaries. E.g.
```python
shape = tf.constant([3])
@tf.function
def f():
u = tf.random_uniform(shape)
return u
```
If we were to rely solely on C++ shape inference, the shape of `u` inside
`f` would be unknown because C++ shape inference is not aware of the outer
graph and all it sees is a Placeholder node when backtracing the captured
tensor for `shape`. `maybe_set_static_shape` computes the static shape value
of `shape` by traversing the `FuncGraph` boundaries and sets the correct
shape.
A longer term solution would be to fix C++ shape inference.
Args:
tensor: A tensor.
shape: A shape tensor."
4400,ParameterModifier,tensorflow/tensorflow/python/framework/test_combinations.py,115,class,Customizes the behavior of a particular parameter.
4401,modified_arguments,tensorflow/tensorflow/python/framework/test_combinations.py,133,method,"Replace user-provided arguments before they are passed to a test.
This makes it possible to adjust user-provided arguments before passing
them to the test method.
Arguments:
kwargs: The combined arguments for the test.
requested_parameters: The set of parameters that are defined in the
signature of the test method.
Returns:
A dictionary with updates to `kwargs`. Keys with values set to
`ParameterModifier.DO_NOT_PASS_TO_THE_TEST` are going to be deleted and
not passed to the test."
4402,OptionalParameter,tensorflow/tensorflow/python/framework/test_combinations.py,172,class,A parameter that is optional in `combine()` and in the test signature.
4403,modified_arguments,tensorflow/tensorflow/python/framework/test_combinations.py,175,method,
4404,generate,tensorflow/tensorflow/python/framework/test_combinations.py,182,function,"A decorator for generating combinations of a test method or a test class.
Parameters of the test method must match by name to get the corresponding
value of the combination. Tests must accept all parameters that are passed
other than the ones that are `OptionalParameter`.
Args:
combinations: a list of dictionaries created using combine() and times().
test_combinations: a tuple of `TestCombination` instances that customize
the execution of generated tests.
Returns:
a decorator that will cause the test method or the test class to be run
under the specified conditions.
Raises:
ValueError: if any parameters were not accepted by the test method"
4405,combine,tensorflow/tensorflow/python/framework/test_combinations.py,319,function,"Generate combinations based on its keyword arguments.
Two sets of returned combinations can be concatenated using +. Their product
can be computed using `times()`.
Args:
**kwargs: keyword arguments of form `option=[possibilities, ...]`
or `option=the_only_possibility`.
Returns:
a list of dictionaries for each combination. Keys in the dictionaries are
the keyword argument names. Each key has one value - one of the
corresponding keyword argument values."
4406,times,tensorflow/tensorflow/python/framework/test_combinations.py,356,function,"Generate a product of N sets of combinations.
times(combine(a=[1,2]), combine(b=[3,4])) == combine(a=[1,2], b=[3,4])
Args:
*combined: N lists of dictionaries that specify combinations.
Returns:
a list of dictionaries for each combination.
Raises:
ValueError: if some of the inputs have overlapping keys."
4407,NamedObject,tensorflow/tensorflow/python/framework/test_combinations.py,389,class,A class that translates an object into a good test name.
4408,is_xla_enabled,tensorflow/tensorflow/python/framework/test_util.py,93,function,
4409,is_mlir_bridge_enabled,tensorflow/tensorflow/python/framework/test_util.py,104,function,
4410,is_tfrt_enabled,tensorflow/tensorflow/python/framework/test_util.py,115,function,
4411,gpu_device_name,tensorflow/tensorflow/python/framework/test_util.py,129,function,Returns the name of a GPU device if available or the empty string.
4412,assert_ops_in_graph,tensorflow/tensorflow/python/framework/test_util.py,137,function,"Assert all expected operations are found.
Args:
expected_ops: `dict<string, string>` of op name to op type.
graph: Graph to check.
Returns:
`dict<string, node>` of node name to node.
Raises:
ValueError: If the expected ops are not present in the graph."
4413,assert_equal_graph_def_v2,tensorflow/tensorflow/python/framework/test_util.py,165,function,"Asserts that two `GraphDef`s are (mostly) the same.
Compares two `GraphDef` protos for equality, ignoring versions and ordering of
nodes, attrs, and control inputs. Node names are used to match up nodes
between the graphs, so the naming of nodes must be consistent. This function
ignores randomized attribute values that may appear in V2 checkpoints.
Args:
expected: The `GraphDef` we expected.
actual: The `GraphDef` we have.
Raises:
AssertionError: If the `GraphDef`s do not match.
TypeError: If either argument is not a `GraphDef`."
4414,assert_equal_graph_def_v1,tensorflow/tensorflow/python/framework/test_util.py,186,function,"Asserts that two `GraphDef`s are (mostly) the same.
Compares two `GraphDef` protos for equality, ignoring versions and ordering of
nodes, attrs, and control inputs. Node names are used to match up nodes
between the graphs, so the naming of nodes must be consistent.
Args:
actual: The `GraphDef` we have.
expected: The `GraphDef` we expected.
checkpoint_v2: boolean determining whether to ignore randomized attribute
values that appear in V2 checkpoints.
hash_table_shared_name: boolean determining whether to ignore randomized
shared_names that appear in HashTableV2 op defs.
Raises:
AssertionError: If the `GraphDef`s do not match.
TypeError: If either argument is not a `GraphDef`."
4415,assert_equal_graph_def,tensorflow/tensorflow/python/framework/test_util.py,210,function,
4416,assert_meta_graph_protos_equal,tensorflow/tensorflow/python/framework/test_util.py,233,function,Compares MetaGraphDefs `a` and `b` in unit test class `tester`.
4417,IsGoogleCudaEnabled,tensorflow/tensorflow/python/framework/test_util.py,304,function,
4418,IsBuiltWithROCm,tensorflow/tensorflow/python/framework/test_util.py,308,function,
4419,IsBuiltWithXLA,tensorflow/tensorflow/python/framework/test_util.py,312,function,
4420,IsBuiltWithNvcc,tensorflow/tensorflow/python/framework/test_util.py,316,function,
4421,GpuSupportsHalfMatMulAndConv,tensorflow/tensorflow/python/framework/test_util.py,320,function,
4422,IsMklEnabled,tensorflow/tensorflow/python/framework/test_util.py,324,function,
4423,InstallStackTraceHandler,tensorflow/tensorflow/python/framework/test_util.py,328,function,
4424,NHWCToNCHW,tensorflow/tensorflow/python/framework/test_util.py,332,function,"Converts the input from the NHWC format to NCHW.
Args:
input_tensor: a 4- or 5-D tensor, or an array representing shape
Returns:
converted tensor or shape array"
4425,NHWCToNCHW_VECT_C,tensorflow/tensorflow/python/framework/test_util.py,351,function,"Transforms the input from the NHWC layout to NCHW_VECT_C layout.
Note: Does not include quantization or type conversion steps, which should
be applied afterwards.
Args:
input_shape_or_tensor: a 4- or 5-D tensor, or an array representing shape
Returns:
tensor or shape array transformed into NCHW_VECT_C
Raises:
ValueError: if last dimension of `input_shape_or_tensor` is not evenly
divisible by 4."
4426,NCHW_VECT_CToNHWC,tensorflow/tensorflow/python/framework/test_util.py,386,function,"Transforms the input from the NCHW_VECT_C layout to NHWC layout.
Note: Does not include de-quantization or type conversion steps, which should
be applied beforehand.
Args:
input_shape_or_tensor: a 5- or 6-D tensor, or an array representing shape
Returns:
tensor or shape array transformed into NHWC
Raises:
ValueError: if last dimension of `input_shape_or_tensor` is not 4."
4427,NCHWToNHWC,tensorflow/tensorflow/python/framework/test_util.py,418,function,"Converts the input from the NCHW format to NHWC.
Args:
input_tensor: a 4- or 5-D tensor, or an array representing shape
Returns:
converted tensor or shape array"
4428,skip_if,tensorflow/tensorflow/python/framework/test_util.py,437,function,"Skips the decorated function if condition is or evaluates to True.
Args:
condition: Either an expression that can be used in ""if not condition""
statement, or a callable whose result should be a boolean.
Returns:
The wrapped function"
4429,skip_if_error,tensorflow/tensorflow/python/framework/test_util.py,464,function,"Context manager to skip cases not considered failures by the tests.
Note that this does not work if used in setUpClass/tearDownClass.
Usage in setUp/tearDown works fine just like regular test methods.
Args:
test_obj: A test object provided as `self` in the test methods; this object
is usually an instance of `unittest.TestCase`'s subclass and should have
`skipTest` method.
error_type: The error type to skip. Note that if `messages` are given, both
`error_type` and `messages` need to match for the test to be skipped.
messages: Optional, a string or list of strings. If `None`, the test will be
skipped if `error_type` matches what is raised; otherwise, the test is
skipped if any of the `messages` is contained in the message of the error
raised, and `error_type` matches the error raised.
Yields:
Nothing."
4430,enable_c_shapes,tensorflow/tensorflow/python/framework/test_util.py,495,function,No-op. TODO(b/74620627): Remove this.
4431,with_c_shapes,tensorflow/tensorflow/python/framework/test_util.py,500,function,No-op. TODO(b/74620627): Remove this.
4432,enable_control_flow_v2,tensorflow/tensorflow/python/framework/test_util.py,505,function,"Decorator for enabling CondV2 and WhileV2 on a test.
Note this enables using CondV2 and WhileV2 after running the test class's
setup/teardown methods.
In addition to this, callers must import the while_v2 module in order to set
the _while_v2 module in control_flow_ops.
Args:
fn: the function to be wrapped
Returns:
The wrapped function"
4433,with_control_flow_v2,tensorflow/tensorflow/python/framework/test_util.py,532,function,"Adds methods that call original methods with WhileV2 and CondV2 enabled.
Note this enables CondV2 and WhileV2 in new methods after running the test
class's setup method.
In addition to this, callers must import the while_v2 module in order to set
the _while_v2 module in control_flow_ops.
If a test function has _disable_control_flow_v2 attr set to True (using the
@disable_control_flow_v2 decorator), the v2 function is not generated for it.
Example:
@test_util.with_control_flow_v2
class ControlFlowTest(test.TestCase):
def testEnabledForV2(self):
...
@test_util.disable_control_flow_v2(""b/xyzabc"")
def testDisabledForV2(self):
...
Generated class:
class ControlFlowTest(test.TestCase):
def testEnabledForV2(self):
...
def testEnabledForV2WithControlFlowV2(self):
// Enable V2 flags.
testEnabledForV2(self)
// Restore V2 flags.
def testDisabledForV2(self):
...
Args:
cls: class to decorate
Returns:
cls with new test methods added"
4434,disable_control_flow_v2,tensorflow/tensorflow/python/framework/test_util.py,587,function,"Decorator for a function in a with_control_flow_v2 enabled test class.
Blocks the function from being run with v2 control flow ops.
Args:
unused_msg: Reason for disabling.
Returns:
The wrapped function with _disable_control_flow_v2 attr set to True."
4435,enable_output_all_intermediates,tensorflow/tensorflow/python/framework/test_util.py,606,function,"Force-enable outputing all intermediates from functional control flow ops.
Args:
fn: the function to be wrapped
Returns:
The wrapped function"
4436,assert_no_new_pyobjects_executing_eagerly,tensorflow/tensorflow/python/framework/test_util.py,629,function,"Decorator for asserting that no new Python objects persist after a test.
Runs the test multiple times executing eagerly, first as a warmup and then to
let objects accumulate. The warmup helps ignore caches which do not grow as
the test is run repeatedly.
Useful for checking that there are no missing Py_DECREFs in the C exercised by
a bit of Python.
Args:
func: The function to test.
warmup_iters: The numer of warmup iterations, excluded from measuring.
Returns:
The wrapped function performing the test."
4437,assert_no_new_tensors,tensorflow/tensorflow/python/framework/test_util.py,712,function,"Decorator for asserting that no new Tensors persist after a test.
Mainly useful for checking that code using the Python C API has correctly
manipulated reference counts.
Clears the caches that it knows about, runs the garbage collector, then checks
that there are no Tensor or Tensor-like objects still around. This includes
Tensors to which something still has a reference (e.g. from missing
Py_DECREFs) and uncollectable cycles (i.e. Python reference cycles where one
of the objects has __del__ defined).
Args:
f: The test case to run.
Returns:
The decorated test case."
4438,assert_no_garbage_created,tensorflow/tensorflow/python/framework/test_util.py,877,function,"Test method decorator to assert that no garbage has been created.
Note that this decorator sets DEBUG_SAVEALL, which in some Python interpreters
cannot be un-set (i.e. will disable garbage collection for any other unit
tests in the same file/shard).
Args:
f: The function to decorate.
Returns:
The decorated function."
4439,run_all_in_graph_and_eager_modes,tensorflow/tensorflow/python/framework/test_util.py,1010,function,Execute all test methods in the given class with and without eager.
4440,build_as_function_and_v1_graph,tensorflow/tensorflow/python/framework/test_util.py,1025,function,"Run a test case in v1 graph mode and inside tf.function in eager mode.
WARNING: This decorator can only be used in test cases that statically checks
generated graph. Attempting to evaluate graph or function results via.
session.run() or self.evaluate() will fail.
WARNING: This decorator can only be used for test cases that inherit from
absl.testing.parameterized.TestCase.
Args:
func: Test case function to be decorated.
Returns:
Decorated test case function."
4441,run_in_async_and_sync_mode,tensorflow/tensorflow/python/framework/test_util.py,1077,function,Execute the test in async mode and sync mode.
4442,eager_lazy_remote_copy_on_and_off,tensorflow/tensorflow/python/framework/test_util.py,1092,function,Execute the test method w/o lazy tensor copy for function remote inputs.
4443,run_in_graph_and_eager_modes,tensorflow/tensorflow/python/framework/test_util.py,1107,function,"Execute the decorated test with and without enabling eager execution.
This function returns a decorator intended to be applied to test methods in
a `tf.test.TestCase` class. Doing so will cause the contents of the test
method to be executed twice - once normally, and once with eager execution
enabled. This allows unittests to confirm the equivalence between eager
and graph execution (see `tf.compat.v1.enable_eager_execution`).
For example, consider the following unittest:
```python
class MyTests(tf.test.TestCase):
@run_in_graph_and_eager_modes
def test_foo(self):
x = tf.constant([1, 2])
y = tf.constant([3, 4])
z = tf.add(x, y)
self.assertAllEqual([4, 6], self.evaluate(z))
if __name__ == ""__main__"":
tf.test.main()
```
This test validates that `tf.add()` has the same behavior when computed with
eager execution enabled as it does when constructing a TensorFlow graph and
executing the `z` tensor in a session.
`deprecated_graph_mode_only`, `run_v1_only`, `run_v2_only`, and
`run_in_graph_and_eager_modes` are available decorators for different
v1/v2/eager/graph combinations.
Args:
func: function to be annotated. If `func` is None, this method returns a
decorator the can be applied to a function. If `func` is not None this
returns the decorator applied to `func`.
config: An optional config_pb2.ConfigProto to use to configure the session
when executing graphs.
use_gpu: If True, attempt to run as many operations as possible on GPU.
assert_no_eager_garbage: If True, sets DEBUG_SAVEALL on the garbage
collector and asserts that no extra garbage has been created when running
the test with eager execution enabled. This will fail if there are
reference cycles (e.g. a = []; a.append(a)). Off by default because some
tests may create garbage for legitimate reasons (e.g. they define a class
which inherits from `object`), and because DEBUG_SAVEALL is sticky in some
Python interpreters (meaning that tests which rely on objects being
collected elsewhere in the unit test file will not work). Additionally,
checks that nothing still has a reference to Tensors that the test
allocated.
Returns:
Returns a decorator that will run the decorated test method twice:
once by constructing and executing a graph in a session and once with
eager execution enabled."
4444,py_func_if_in_function,tensorflow/tensorflow/python/framework/test_util.py,1214,function,
4445,also_run_as_tf_function,tensorflow/tensorflow/python/framework/test_util.py,1238,function,"Runs the decorated test twice--once as is, once inside a tf.function.
This allows you to run a test both in eager execution and inside a
tf.function, exercising the two execution modes supported in tf 2.0. The test
assertions are automatically done inside tf.py_funcs, and tf.function ensures
that they run in the proper order and with the proper side effects.
Currently variable creation is not supported in tests annotated with this
decorator since it's tricky to ensure the variable doesn't get repeatedly
created when retracing the tf.function.
Args:
f: the test method to be decorated
Returns:
The decorated test method, which will run both in eager and inside a
tf.function."
4446,deprecated_graph_mode_only,tensorflow/tensorflow/python/framework/test_util.py,1273,function,"Execute the decorated test in graph mode.
This function returns a decorator intended to be applied to tests that are not
compatible with eager mode. When this decorator is applied, the test body will
be run in an environment where API calls construct graphs instead of executing
eagerly.
`deprecated_graph_mode_only`, `run_v1_only`, `run_v2_only`, and
`run_in_graph_and_eager_modes` are available decorators for different
v1/v2/eager/graph combinations.
Args:
func: function to be annotated. If `func` is None, this method returns a
decorator the can be applied to a function. If `func` is not None this
returns the decorator applied to `func`.
Returns:
Returns a decorator that will run the decorated test method in graph mode."
4447,run_all_in_deprecated_graph_mode_only,tensorflow/tensorflow/python/framework/test_util.py,1325,function,Execute all tests in a class in graph mode.
4448,run_v1_only,tensorflow/tensorflow/python/framework/test_util.py,1338,function,"Execute the decorated test only if running in v1 mode.
This function is intended to be applied to tests that exercise v1 only
functionality. If the test is run in v2 mode it will simply be skipped.
`deprecated_graph_mode_only`, `run_v1_only`, `run_v2_only`, and
`run_in_graph_and_eager_modes` are available decorators for different
v1/v2/eager/graph combinations.
Args:
reason: string giving a reason for limiting the test to v1 only.
func: function to be annotated. If `func` is None, this method returns a
decorator the can be applied to a function. If `func` is not None this
returns the decorator applied to `func`.
Returns:
Returns a decorator that will conditionally skip the decorated test method."
4449,run_v2_only,tensorflow/tensorflow/python/framework/test_util.py,1391,function,"Execute the decorated test only if running in v2 mode.
This function is intended to be applied to tests that exercise v2 only
functionality. If the test is run in v1 mode it will simply be skipped.
`deprecated_graph_mode_only`, `run_v1_only`, `run_v2_only`, and
`run_in_graph_and_eager_modes` are available decorators for different
v1/v2/eager/graph combinations.
Args:
func: function to be annotated. If `func` is None, this method returns a
decorator the can be applied to a function. If `func` is not None this
returns the decorator applied to `func`.
Returns:
Returns a decorator that will conditionally skip the decorated test method."
4450,run_gpu_only,tensorflow/tensorflow/python/framework/test_util.py,1428,function,"Execute the decorated test only if a GPU is available.
This function is intended to be applied to tests that require the presence
of a GPU. If a GPU is absent, it will simply be skipped.
Args:
func: function to be annotated. If `func` is None, this method returns a
decorator the can be applied to a function. If `func` is not None this
returns the decorator applied to `func`.
Returns:
Returns a decorator that will conditionally skip the decorated test method."
4451,run_cuda_only,tensorflow/tensorflow/python/framework/test_util.py,1461,function,"Execute the decorated test only if a GPU is available.
This function is intended to be applied to tests that require the presence
of a CUDA GPU. If a CUDA GPU is absent, it will simply be skipped.
Args:
func: function to be annotated. If `func` is None, this method returns a
decorator the can be applied to a function. If `func` is not None this
returns the decorator applied to `func`.
Returns:
Returns a decorator that will conditionally skip the decorated test method."
4452,with_forward_compatibility_horizons,tensorflow/tensorflow/python/framework/test_util.py,1494,function,"Executes the decorated test with the specified forward-compat horizons.
Args:
*horizons: A list of (year, month, day) tuples. If the list includes
`None`, then the test will also be run with no forward-compatibility
horizon set.
Returns:
A decorator that will execute the test with the specified horizons."
4453,is_gpu_available,tensorflow/tensorflow/python/framework/test_util.py,1532,function,"Returns whether TensorFlow can access a GPU.
Warning: if a non-GPU version of the package is installed, the function would
also return False. Use `tf.test.is_built_with_cuda` to validate if TensorFlow
was build with CUDA support.
Args:
cuda_only: limit the search to CUDA GPUs.
min_cuda_compute_capability: a (major,minor) pair that indicates the minimum
CUDA compute capability required, or None if no requirement.
Note that the keyword arg name ""cuda_only"" is misleading (since routine will
return true when a GPU device is available irrespective of whether TF was
built with CUDA support or ROCm support. However no changes here because
++ Changing the name ""cuda_only"" to something more generic would break
backward compatibility
++ Adding an equivalent ""rocm_only"" would require the implementation check
the build type. This in turn would require doing the same for CUDA and thus
potentially break backward compatibility
++ Adding a new ""cuda_or_rocm_only"" would not break backward compatibility,
but would require most (if not all) callers to update the call to use
""cuda_or_rocm_only"" instead of ""cuda_only""
Returns:
True if a GPU device of the requested kind is available."
4454,device,tensorflow/tensorflow/python/framework/test_util.py,1581,function,Uses gpu when requested and available.
4455,use_gpu,tensorflow/tensorflow/python/framework/test_util.py,1592,function,Uses gpu when requested and available.
4456,force_gpu,tensorflow/tensorflow/python/framework/test_util.py,1599,function,Force the gpu to be used.
4457,force_cpu,tensorflow/tensorflow/python/framework/test_util.py,1606,function,Force the cpu to be used.
4458,CapturedWrites,tensorflow/tensorflow/python/framework/test_util.py,1612,class,A utility class to load the captured writes made to a stream.
4459,contents,tensorflow/tensorflow/python/framework/test_util.py,1618,method,Get the captured writes as a single string.
4460,FakeEagerSession,tensorflow/tensorflow/python/framework/test_util.py,1625,class,"Fake session so tests that conditionally use placeholders can use eager.
There are a number of tests that conditionally use placeholders for shape
inference. The pattern is demonstrated here:
```python
with self.cached_session() as sess:
if static_shape:
y = math_ops.matmul(x, ...)
feed_dict = {}
else:
x_ph = array_ops.placeholder(...)
y = math_ops.matmul(x_ph, ...)
feed_dict = {x_ph: x}
val = sess.run(y, feed_dict=feed_dict)
```
Since the feed_dict is empty when not using placeholders we should be able to
call self.evaluate(), however this requires rewriting the test case.
This class should be considered a stop-gap solution to get tests running with
eager with minimal changes to the actual test."
4461,run,tensorflow/tensorflow/python/framework/test_util.py,1652,method,"Evaluate `fetches`.
Fail if additional args are specified.
Args:
fetches: A Tensor or a nested list/tuple of Tensors.
*args: Positional arguments
**kwargs: Keyword arguments
Raises:
RuntimeError: If args or kwargs are specified.
Returns:
Tensors as numpy values."
4462,ErrorLoggingSession,tensorflow/tensorflow/python/framework/test_util.py,1682,class,Wrapper around a Session that logs errors in run().
4463,run,tensorflow/tensorflow/python/framework/test_util.py,1685,method,
4464,disable_cudnn_autotune,tensorflow/tensorflow/python/framework/test_util.py,1697,function,"Disable autotuning during the call to this function.
Some tests want to base assertions on a graph being isomorphic with a copy.
To ensure this, this decorator disables autotuning.
Args:
func: Function to run with CuDNN autotuning turned off.
Returns:
Decorated function."
4465,enable_tf_xla_constant_folding,tensorflow/tensorflow/python/framework/test_util.py,1743,function,
4466,disable_xla,tensorflow/tensorflow/python/framework/test_util.py,1802,function,Execute the test method only if xla is not enabled.
4467,disable_mlir_bridge,tensorflow/tensorflow/python/framework/test_util.py,1809,function,Execute the test method only if MLIR bridge is not enabled.
4468,disable_tfrt,tensorflow/tensorflow/python/framework/test_util.py,1816,function,
4469,no_xla_auto_jit,tensorflow/tensorflow/python/framework/test_util.py,1873,function,This test is not intended to be run with XLA auto jit enabled.
4470,xla_allow_fallback,tensorflow/tensorflow/python/framework/test_util.py,1880,function,
4471,EagerSessionWarner,tensorflow/tensorflow/python/framework/test_util.py,1909,class,
4472,create_local_cluster,tensorflow/tensorflow/python/framework/test_util.py,3164,function,"Create and start local servers and return the associated `Server` objects.
""PS"" stands for ""parameter server"": a task responsible for storing and
updating the model's parameters. Other tasks send updates to these parameters
as they work on optimizing the parameters. This particular division of labor
between tasks is not required, but is common for distributed training.
Read more at https://www.tensorflow.org/guide/extend/architecture
![components](https://www.tensorflow.org/images/diag1.svg ""components"")
Figure illustrates the interaction of these components.
""/job:worker/task:0"" and ""/job:ps/task:0"" are both tasks with worker services.
Example:
```python
workers, _ = tf.test.create_local_cluster(num_workers=2, num_ps=2)
worker_sessions = [tf.compat.v1.Session(w.target) for w in workers]
with tf.device(""/job:ps/task:0""):
...
with tf.device(""/job:ps/task:1""):
...
with tf.device(""/job:worker/task:0""):
...
with tf.device(""/job:worker/task:1""):
...
worker_sessions[0].run(...)
```
Args:
num_workers: Number of worker servers to start.
num_ps: Number of PS servers to start.
protocol: Communication protocol. Allowed values are documented in the
documentation of `tf.distribute.Server`.
worker_config: (optional) `tf.ConfigProto` to initialize workers. Can be
used to instantiate multiple devices etc.
ps_config: (optional) `tf.ConfigProto` to initialize PS servers.
Returns:
A tuple `(worker_servers, ps_servers)`. `worker_servers` is a list
of `num_workers` objects of type `tf.distribute.Server` (all running
locally);
and `ps_servers` is a list of `num_ps` objects of similar type.
Raises:
ImportError: if portpicker module was not found at load time"
4473,get_node_def_from_graph,tensorflow/tensorflow/python/framework/test_util.py,3252,function,"Returns the `NodeDef` instance for given node name in the graph def.
This method explores only the NodeDefs in `graph_def.node`.
Args:
node_name: Name of the NodeDef to search for.
graph_def: An instance of `GraphDef` proto.
Returns:
the `NodeDef` instance whose name field matches the given node_name or None."
4474,set_producer_version,tensorflow/tensorflow/python/framework/test_util.py,3270,function,Sets graph.graph_def_versions.producer to `producer_version`.
4475,GraphAndEagerNoVariableSharing,tensorflow/tensorflow/python/framework/test_util_test.py,874,class,
4476,setUp,tensorflow/tensorflow/python/framework/test_util_test.py,876,method,
4477,set_environ,tensorflow/tensorflow/python/framework/tf2_test.py,30,function,
4478,unset_environ,tensorflow/tensorflow/python/framework/tf2_test.py,34,function,
4479,EnablingTF2Behavior,tensorflow/tensorflow/python/framework/tf2_test.py,38,class,
4480,setUp,tensorflow/tensorflow/python/framework/tf2_test.py,40,method,
4481,state,tensorflow/tensorflow/python/framework/tf2_test.py,54,method,"Returns bool tuple (tf2_enabled, force_enabled, force_disabled)."
4482,TraceableObject,tensorflow/tensorflow/python/framework/traceable_stack.py,24,class,Wrap an object together with its the code definition location.
4483,set_filename_and_line_from_caller,tensorflow/tensorflow/python/framework/traceable_stack.py,35,method,"Set filename and line using the caller's stack frame.
If the requested stack information is not available, a heuristic may
be applied and self.HEURISTIC USED will be returned. If the heuristic
fails then no change will be made to the filename and lineno members
(None by default) and self.FAILURE will be returned.
Args:
offset: Integer. If 0, the caller's stack frame is used. If 1,
the caller's caller's stack frame is used. Larger values are
permissible but if out-of-range (larger than the number of stack
frames available) the outermost stack frame will be used.
Returns:
TraceableObject.SUCCESS if appropriate stack information was found,
TraceableObject.HEURISTIC_USED if the offset was larger than the stack,
and TraceableObject.FAILURE if the stack was empty."
4484,copy_metadata,tensorflow/tensorflow/python/framework/traceable_stack.py,75,method,"Return a TraceableObject like this one, but without the object."
4485,TraceableStack,tensorflow/tensorflow/python/framework/traceable_stack.py,80,class,A stack of TraceableObjects.
4486,push_obj,tensorflow/tensorflow/python/framework/traceable_stack.py,92,method,"Add object to the stack and record its filename and line information.
Args:
obj: An object to store on the stack.
offset: Integer. If 0, the caller's stack frame is used. If 1,
the caller's caller's stack frame is used.
Returns:
TraceableObject.SUCCESS if appropriate stack information was found,
TraceableObject.HEURISTIC_USED if the stack was smaller than expected,
and TraceableObject.FAILURE if the stack was empty."
4487,pop_obj,tensorflow/tensorflow/python/framework/traceable_stack.py,111,method,"Remove last-inserted object and return it, without filename/line info."
4488,peek_top_obj,tensorflow/tensorflow/python/framework/traceable_stack.py,115,method,Return the most recent stored object.
4489,peek_objs,tensorflow/tensorflow/python/framework/traceable_stack.py,119,method,Return iterator over stored objects ordered newest to oldest.
4490,peek_traceable_objs,tensorflow/tensorflow/python/framework/traceable_stack.py,123,method,Return iterator over stored TraceableObjects ordered newest to oldest.
4491,copy,tensorflow/tensorflow/python/framework/traceable_stack.py,131,method,"Return a copy of self referencing the same objects but in a new list.
This method is implemented to support thread-local stacks.
Returns:
TraceableStack with a new list that holds existing objects."
4492,TypeSpec,tensorflow/tensorflow/python/framework/type_spec.py,49,class,"Specifies a TensorFlow value type.
A `tf.TypeSpec` provides metadata describing an object accepted or returned
by TensorFlow APIs. Concrete subclasses, such as `tf.TensorSpec` and
`tf.RaggedTensorSpec`, are used to describe different value types.
For example, `tf.function`'s `input_signature` argument accepts a list
(or nested structure) of `TypeSpec`s.
Creating new subclasses of TypeSpec (outside of TensorFlow core) is not
currently supported. In particular, we may make breaking changes to the
private methods and properties defined by this base class."
4493,value_type,tensorflow/tensorflow/python/framework/type_spec.py,85,method,"The Python type for values that are compatible with this TypeSpec.
In particular, all values that are compatible with this TypeSpec must be an
instance of this type."
4494,is_compatible_with,tensorflow/tensorflow/python/framework/type_spec.py,93,method,Returns true if `spec_or_value` is compatible with this TypeSpec.
4495,most_specific_compatible_type,tensorflow/tensorflow/python/framework/type_spec.py,110,method,"Returns the most specific TypeSpec compatible with `self` and `other`.
Args:
other: A `TypeSpec`.
Raises:
ValueError: If there is no TypeSpec that is compatible with both `self`
and `other`."
4496,relax,tensorflow/tensorflow/python/framework/type_spec.py,148,method,
4497,BatchableTypeSpec,tensorflow/tensorflow/python/framework/type_spec.py,459,class,"TypeSpec with a batchable tensor encoding.
The batchable tensor encoding is a list of `tf.Tensor`s that supports
batching and unbatching. In particular, stacking (or unstacking)
values with the same `TypeSpec` must be equivalent to stacking (or
unstacking) each of their tensor lists. Unlike the component encoding
(returned by `self._to_components)`, the batchable tensor encoding
may require using encoding/decoding ops.
If a subclass's batchable tensor encoding is not simply a flattened version
of the component encoding, then the subclass must override `_to_tensor_list`,
`_from_tensor_list`, and _flat_tensor_specs`."
4498,type_spec_from_value,tensorflow/tensorflow/python/framework/type_spec.py,507,function,"Returns a `tf.TypeSpec` that represents the given `value`.
Examples:
>>> tf.type_spec_from_value(tf.constant([1, 2, 3]))
TensorSpec(shape=(3,), dtype=tf.int32, name=None)
>>> tf.type_spec_from_value(np.array([4.0, 5.0], np.float64))
TensorSpec(shape=(2,), dtype=tf.float64, name=None)
>>> tf.type_spec_from_value(tf.ragged.constant([[1, 2], [3, 4, 5]]))
RaggedTensorSpec(TensorShape([2, None]), tf.int32, 1, tf.int64)
>>> example_input = tf.ragged.constant([[1, 2], [3]])
>>> @tf.function(input_signature=[tf.type_spec_from_value(example_input)])
... def f(x):
... return tf.reduce_sum(x, axis=1)
Args:
value: A value that can be accepted or returned by TensorFlow APIs.
Accepted types for `value` include `tf.Tensor`, any value that can be
converted to `tf.Tensor` using `tf.convert_to_tensor`, and any subclass
of `CompositeTensor` (such as `tf.RaggedTensor`).
Returns:
A `TypeSpec` that is compatible with `value`.
Raises:
TypeError: If a TypeSpec cannot be built for `value`, because its type
is not supported."
4499,register_type_spec_from_value_converter,tensorflow/tensorflow/python/framework/type_spec.py,590,function,"Registers a function for converting values with a given type to TypeSpecs.
If multiple registered `type_object`s match a value, then the most recent
registration takes precedence. Custom converters should not be defined for
`CompositeTensor`s; use `CompositeTensor._type_spec` instead.
Args:
type_object: A Python `type` object representing the type of values
accepted by `converter_fn`.
converter_fn: A function that takes one argument (an instance of the
type represented by `type_object`) and returns a `TypeSpec`.
allow_subclass: If true, then use `isinstance(value, type_object)` to
check for matches. If false, then use `type(value) is type_object`."
4500,TwoTensors,tensorflow/tensorflow/python/framework/type_spec_test.py,35,class,"A simple value type to test TypeSpec.
Contains two tensors (x, y) and a string (color). The color value is a
stand-in for any extra type metadata we might need to store."
4501,TwoTensorsSpec,tensorflow/tensorflow/python/framework/type_spec_test.py,49,class,A TypeSpec for the TwoTensors value type.
4502,from_value,tensorflow/tensorflow/python/framework/type_spec_test.py,76,method,
4503,Cluster,tensorflow/tensorflow/python/grappler/cluster.py,29,class,Grappler Clusters.
4504,Shutdown,tensorflow/tensorflow/python/grappler/cluster.py,59,method,
4505,tf_cluster,tensorflow/tensorflow/python/grappler/cluster.py,68,method,
4506,ListDevices,tensorflow/tensorflow/python/grappler/cluster.py,71,method,Returns a list of available hardware devices.
4507,ListAvailableOps,tensorflow/tensorflow/python/grappler/cluster.py,78,method,Returns a list of all available operations (sorted alphabetically).
4508,GetSupportedDevices,tensorflow/tensorflow/python/grappler/cluster.py,82,method,
4509,EstimatePerformance,tensorflow/tensorflow/python/grappler/cluster.py,85,method,
4510,MeasureCosts,tensorflow/tensorflow/python/grappler/cluster.py,88,method,"Returns the cost of running the specified item.
Args:
item: The item for which to measure the costs.
Returns: The triplet op_perfs, runtime, step_stats."
4511,DeterminePeakMemoryUsage,tensorflow/tensorflow/python/grappler/cluster.py,103,method,"Returns a snapshot of the peak memory usage.
Args:
item: The item for which to measure the costs.
Returns: A hashtable indexed by device name."
4512,Provision,tensorflow/tensorflow/python/grappler/cluster.py,115,function,
4513,GenerateCostReport,tensorflow/tensorflow/python/grappler/cost_analyzer.py,26,function,"Analyze the cost of each TensorFlow op and node in the provided metagraph.
Args:
metagraph: A TensorFlow MetaGraphDef.
per_node_report: by default the report contains stats aggregated on a per op
type basis, setting per_node_report to True adds results for each
individual node to the report.
verbose: Prints out the entire operation proto instead of a summary table.
cluster: Analyze the costs using the specified cluster, or the local machine
if no cluster was specified.
Returns:
A string of cost report."
4514,GenerateMemoryReport,tensorflow/tensorflow/python/grappler/cost_analyzer.py,52,function,"Analyze the peak memory usage for the provided metagraph.
Args:
metagraph: A TensorFlow MetaGraphDef.
detailed_report: print the live tensors in addition to the peak memory
usage.
cluster: Analyze the memory using the specified cluster, or the local
machine if no cluster was specified.
Returns:
A string with the formatted memory usage."
4515,get_metagraph,tensorflow/tensorflow/python/grappler/cost_analyzer_tool.py,40,function,Constructs and returns a MetaGraphDef from the input file.
4516,Item,tensorflow/tensorflow/python/grappler/item.py,26,class,GrapplerItem.
4517,IdentifyImportantOps,tensorflow/tensorflow/python/grappler/item.py,52,method,
4518,GetOpProperties,tensorflow/tensorflow/python/grappler/item.py,55,method,Get Op properties.
4519,GetColocationGroups,tensorflow/tensorflow/python/grappler/item.py,69,method,"Return a list of hard colocation constraints.
All the nodes in a colocation tuple must be placed on the same device for
the model to work.
Returns:
A list of colocation tuples."
4520,metagraph,tensorflow/tensorflow/python/grappler/item.py,81,method,
4521,tf_item,tensorflow/tensorflow/python/grappler/item.py,85,method,
4522,GenerateModelReport,tensorflow/tensorflow/python/grappler/model_analyzer.py,24,function,"Report what's known statically about each node in the provided metagraph.
Args:
metagraph: A TensorFlow MetaGraphDef.
assume_valid_feeds: If True, assume that the shape of the fed nodes is valid
debug: Add some information useful for debugging.
Returns:
A string containing the report."
4523,OptimizeGraph,tensorflow/tensorflow/python/grappler/tf_optimizer.py,27,function,"Optimize the provided metagraph.
For best results, the signature_def field in `metagraph` should be populated
with information about input (feed) and output (fetch) tensors.
Args:
config_proto: a ConfigProto protobuf.
metagraph: a MetagraphDef protobuf.
verbose: whether to log optimization results.
graph_id: a string identifying this graph.
cluster: a grappler cluster object representing hardware resources
available to run this graph.
strip_default_attributes: whether graph node attributes having default
values should be removed after all the optimization passes. This
option is useful if the resulting graph will be executed by an older
process that might not know some of the recently added attributes."
4524,softmax,tensorflow/tensorflow/python/keras/activations.py,46,function,"Softmax converts a real vector to a vector of categorical probabilities.
The elements of the output vector are in range (0, 1) and sum to 1.
Each vector is handled independently. The `axis` argument sets which axis
of the input the function is applied along.
Softmax is often used as the activation for the last
layer of a classification network because the result could be interpreted as
a probability distribution.
The softmax of each vector x is computed as
`exp(x) / tf.reduce_sum(exp(x))`.
The input values in are the log-odds of the resulting probability.
Arguments:
x : Input tensor.
axis: Integer, axis along which the softmax normalization is applied.
Returns:
Tensor, output of softmax transformation (all values are non-negative
and sum to 1).
Raises:
ValueError: In case `dim(x) == 1`."
4525,elu,tensorflow/tensorflow/python/keras/activations.py,88,function,"Exponential Linear Unit.
The exponential linear unit (ELU) with `alpha > 0` is:
`x` if `x > 0` and
`alpha * (exp(x) - 1)` if `x < 0`
The ELU hyperparameter `alpha` controls the value to which an
ELU saturates for negative net inputs. ELUs diminish the
vanishing gradient effect.
ELUs have negative values which pushes the mean of the activations
closer to zero.
Mean activations that are closer to zero enable faster learning as they
bring the gradient closer to the natural gradient.
ELUs saturate to a negative value when the argument gets smaller.
Saturation means a small derivative which decreases the variation
and the information that is propagated to the next layer.
Example Usage:
>>> import tensorflow as tf
>>> model = tf.keras.Sequential()
>>> model.add(tf.keras.layers.Conv2D(32, (3, 3), activation='elu',
... input_shape=(28, 28, 1)))
>>> model.add(tf.keras.layers.MaxPooling2D((2, 2)))
>>> model.add(tf.keras.layers.Conv2D(64, (3, 3), activation='elu'))
>>> model.add(tf.keras.layers.MaxPooling2D((2, 2)))
>>> model.add(tf.keras.layers.Conv2D(64, (3, 3), activation='elu'))
<tensorflow.python.keras.engine.sequential.Sequential object ...>
Arguments:
x: Input tensor.
alpha: A scalar, slope of negative section. `alpha` controls the value to
which an ELU saturates for negative net inputs.
Returns:
The exponential linear unit (ELU) activation function: `x` if `x > 0` and
`alpha * (exp(x) - 1)` if `x < 0`.
Reference:
[Fast and Accurate Deep Network Learning by Exponential Linear Units
(ELUs) (Clevert et al, 2016)](https://arxiv.org/abs/1511.07289)"
4526,selu,tensorflow/tensorflow/python/keras/activations.py,138,function,"Scaled Exponential Linear Unit (SELU).
The Scaled Exponential Linear Unit (SELU) activation function is defined as:
- `if x > 0: return scale * x`
- `if x < 0: return scale * alpha * (exp(x) - 1)`
where `alpha` and `scale` are pre-defined constants
(`alpha=1.67326324` and `scale=1.05070098`).
Basically, the SELU activation function multiplies `scale` (> 1) with the
output of the `tf.keras.activations.elu` function to ensure a slope larger
than one for positive inputs.
The values of `alpha` and `scale` are
chosen so that the mean and variance of the inputs are preserved
between two consecutive layers as long as the weights are initialized
correctly (see `tf.keras.initializers.LecunNormal` initializer)
and the number of input units is ""large enough""
(see reference paper for more information).
Example Usage:
>>> num_classes = 10 # 10-class problem
>>> model = tf.keras.Sequential()
>>> model.add(tf.keras.layers.Dense(64, kernel_initializer='lecun_normal',
... activation='selu'))
>>> model.add(tf.keras.layers.Dense(32, kernel_initializer='lecun_normal',
... activation='selu'))
>>> model.add(tf.keras.layers.Dense(16, kernel_initializer='lecun_normal',
... activation='selu'))
>>> model.add(tf.keras.layers.Dense(num_classes, activation='softmax'))
Arguments:
x: A tensor or variable to compute the activation function for.
Returns:
The scaled exponential unit activation: `scale * elu(x, alpha)`.
Notes:
- To be used together with the
`tf.keras.initializers.LecunNormal` initializer.
- To be used together with the dropout variant
`tf.keras.layers.AlphaDropout` (not regular dropout).
References:
- [Klambauer et al., 2017](https://arxiv.org/abs/1706.02515)"
4527,softplus,tensorflow/tensorflow/python/keras/activations.py,192,function,"Softplus activation function, `softplus(x) = log(exp(x) + 1)`.
Example Usage:
>>> a = tf.constant([-20, -1.0, 0.0, 1.0, 20], dtype = tf.float32)
>>> b = tf.keras.activations.softplus(a)
>>> b.numpy()
array([2.0611537e-09, 3.1326166e-01, 6.9314718e-01, 1.3132616e+00,
2.0000000e+01], dtype=float32)
Arguments:
x: Input tensor.
Returns:
The softplus activation: `log(exp(x) + 1)`."
4528,softsign,tensorflow/tensorflow/python/keras/activations.py,214,function,"Softsign activation function, `softsign(x) = x / (abs(x) + 1)`.
Example Usage:
>>> a = tf.constant([-1.0, 0.0, 1.0], dtype = tf.float32)
>>> b = tf.keras.activations.softsign(a)
>>> b.numpy()
array([-0.5, 0. , 0.5], dtype=float32)
Arguments:
x: Input tensor.
Returns:
The softsign activation: `x / (abs(x) + 1)`."
4529,swish,tensorflow/tensorflow/python/keras/activations.py,235,function,"Swish activation function, `swish(x) = x * sigmoid(x)`.
Swish activation function which returns `x*sigmoid(x)`.
It is a smooth, non-monotonic function that consistently matches
or outperforms ReLU on deep networks, it is unbounded above and
bounded below.
Example Usage:
>>> a = tf.constant([-20, -1.0, 0.0, 1.0, 20], dtype = tf.float32)
>>> b = tf.keras.activations.swish(a)
>>> b.numpy()
array([-4.1223075e-08, -2.6894143e-01, 0.0000000e+00, 7.3105860e-01,
2.0000000e+01], dtype=float32)
Arguments:
x: Input tensor.
Returns:
The swish activation applied to `x` (see reference paper for details).
Reference:
- [Ramachandran et al., 2017](https://arxiv.org/abs/1710.05941)"
4530,relu,tensorflow/tensorflow/python/keras/activations.py,266,function,"Applies the rectified linear unit activation function.
With default values, this returns the standard ReLU activation:
`max(x, 0)`, the element-wise maximum of 0 and the input tensor.
Modifying default parameters allows you to use non-zero thresholds,
change the max value of the activation,
and to use a non-zero multiple of the input for values below the threshold.
For example:
>>> foo = tf.constant([-10, -5, 0.0, 5, 10], dtype = tf.float32)
>>> tf.keras.activations.relu(foo).numpy()
array([ 0., 0., 0., 5., 10.], dtype=float32)
>>> tf.keras.activations.relu(foo, alpha=0.5).numpy()
array([-5. , -2.5, 0. , 5. , 10. ], dtype=float32)
>>> tf.keras.activations.relu(foo, max_value=5).numpy()
array([0., 0., 0., 5., 5.], dtype=float32)
>>> tf.keras.activations.relu(foo, threshold=5).numpy()
array([-0., -0., 0., 0., 10.], dtype=float32)
Arguments:
x: Input `tensor` or `variable`.
alpha: A `float` that governs the slope for values lower than the
threshold.
max_value: A `float` that sets the saturation threshold (the largest value
the function will return).
threshold: A `float` giving the threshold value of the activation function
below which values will be damped or set to zero.
Returns:
A `Tensor` representing the input tensor,
transformed by the relu activation function.
Tensor will be of the same shape and dtype of input `x`."
4531,tanh,tensorflow/tensorflow/python/keras/activations.py,307,function,"Hyperbolic tangent activation function.
For example:
>>> a = tf.constant([-3.0,-1.0, 0.0,1.0,3.0], dtype = tf.float32)
>>> b = tf.keras.activations.tanh(a)
>>> b.numpy()
array([-0.9950547, -0.7615942, 0., 0.7615942, 0.9950547], dtype=float32)
Arguments:
x: Input tensor.
Returns:
Tensor of same shape and dtype of input `x`, with tanh activation:
`tanh(x) = sinh(x)/cosh(x) = ((exp(x) - exp(-x))/(exp(x) + exp(-x)))`."
4532,sigmoid,tensorflow/tensorflow/python/keras/activations.py,329,function,"Sigmoid activation function, `sigmoid(x) = 1 / (1 + exp(-x))`.
Applies the sigmoid activation function. For small values (<-5),
`sigmoid` returns a value close to zero, and for large values (>5)
the result of the function gets close to 1.
Sigmoid is equivalent to a 2-element Softmax, where the second element is
assumed to be zero. The sigmoid function always returns a value between
0 and 1.
For example:
>>> a = tf.constant([-20, -1.0, 0.0, 1.0, 20], dtype = tf.float32)
>>> b = tf.keras.activations.sigmoid(a)
>>> b.numpy()
array([2.0611537e-09, 2.6894143e-01, 5.0000000e-01, 7.3105860e-01,
1.0000000e+00], dtype=float32)
Arguments:
x: Input tensor.
Returns:
Tensor with the sigmoid activation: `1 / (1 + exp(-x))`."
4533,exponential,tensorflow/tensorflow/python/keras/activations.py,359,function,"Exponential activation function.
For example:
>>> a = tf.constant([-3.0,-1.0, 0.0,1.0,3.0], dtype = tf.float32)
>>> b = tf.keras.activations.exponential(a)
>>> b.numpy()
array([0.04978707, 0.36787945, 1., 2.7182817 , 20.085537], dtype=float32)
Arguments:
x: Input tensor.
Returns:
Tensor with exponential activation: `exp(x)`."
4534,hard_sigmoid,tensorflow/tensorflow/python/keras/activations.py,380,function,"Hard sigmoid activation function.
A faster approximation of the sigmoid activation.
For example:
>>> a = tf.constant([-3.0,-1.0, 0.0,1.0,3.0], dtype = tf.float32)
>>> b = tf.keras.activations.hard_sigmoid(a)
>>> b.numpy()
array([0. , 0.3, 0.5, 0.7, 1. ], dtype=float32)
Arguments:
x: Input tensor.
Returns:
The hard sigmoid activation, defined as:
- `if x < -2.5: return 0`
- `if x > 2.5: return 1`
- `if -2.5 <= x <= 2.5: return 0.2 * x + 0.5`"
4535,linear,tensorflow/tensorflow/python/keras/activations.py,407,function,"Linear activation function (pass-through).
For example:
>>> a = tf.constant([-3.0,-1.0, 0.0,1.0,3.0], dtype = tf.float32)
>>> b = tf.keras.activations.linear(a)
>>> b.numpy()
array([-3., -1., 0., 1., 3.], dtype=float32)
Arguments:
x: Input tensor.
Returns:
The input, unmodified."
4536,serialize,tensorflow/tensorflow/python/keras/activations.py,428,function,"Returns the string identifier of an activation function.
Arguments:
activation : Function object.
Returns:
String denoting the name attribute of the input function
For example:
>>> tf.keras.activations.serialize(tf.keras.activations.tanh)
'tanh'
>>> tf.keras.activations.serialize(tf.keras.activations.sigmoid)
'sigmoid'
>>> tf.keras.activations.serialize('abcd')
Traceback (most recent call last):
...
ValueError: ('Cannot serialize', 'abcd')
Raises:
ValueError: The input function is not a valid one."
4537,deserialize,tensorflow/tensorflow/python/keras/activations.py,459,function,"Returns activation function given a string identifier.
Arguments:
x : String identifier.
Returns:
Corresponding activation function.
For example:
>>> tf.keras.activations.deserialize('linear')
<function linear at 0x1239596a8>
>>> tf.keras.activations.deserialize('sigmoid')
<function sigmoid at 0x123959510>
>>> tf.keras.activations.deserialize('abcd')
Traceback (most recent call last):
...
ValueError: Unknown activation function:abcd
Args:
name: The name of the activation function.
custom_objects: Optional `{function_name: function_obj}`
dictionary listing user-provided activation functions.
Raises:
ValueError: `Unknown activation function` if the input string does not
denote any defined Tensorflow activation function."
4538,get,tensorflow/tensorflow/python/keras/activations.py,497,function,"Returns function.
Arguments:
identifier: Function or string
Returns:
Function corresponding to the input string or input function.
For example:
>>> tf.keras.activations.get('softmax')
<function softmax at 0x1222a3d90>
>>> tf.keras.activations.get(tf.keras.activations.softmax)
<function softmax at 0x1222a3d90>
>>> tf.keras.activations.get(None)
<function linear at 0x1239596a8>
>>> tf.keras.activations.get(abs)
<built-in function abs>
>>> tf.keras.activations.get('abcd')
Traceback (most recent call last):
...
ValueError: Unknown activation function:abcd
Raises:
ValueError: Input is an unknown function or string, i.e., the input does
not denote any defined function."
4539,backend,tensorflow/tensorflow/python/keras/backend.py,167,function,"Publicly accessible method for determining the current backend.
Only exists for API compatibility with multi-backend Keras.
Returns:
The string ""tensorflow""."
4540,cast_to_floatx,tensorflow/tensorflow/python/keras/backend.py,180,function,"Cast a Numpy array to the default Keras float type.
Arguments:
x: Numpy array or TensorFlow tensor.
Returns:
The same array (Numpy array if `x` was a Numpy array, or TensorFlow tensor
if `x` was a tensor), cast to its new type.
Example:
>>> tf.keras.backend.floatx()
'float32'
>>> arr = np.array([1.0, 2.0], dtype='float64')
>>> arr.dtype
dtype('float64')
>>> new_arr = cast_to_floatx(arr)
>>> new_arr
array([1., 2.], dtype=float32)
>>> new_arr.dtype
dtype('float32')"
4541,get_uid,tensorflow/tensorflow/python/keras/backend.py,218,function,"Associates a string prefix with an integer counter in a TensorFlow graph.
Arguments:
prefix: String prefix to index.
Returns:
Unique integer ID.
Example:
>>> get_uid('dense')
1
>>> get_uid('dense')
2"
4542,reset_uids,tensorflow/tensorflow/python/keras/backend.py,244,function,"Resets graph identifiers.
"
4543,clear_session,tensorflow/tensorflow/python/keras/backend.py,252,function,"Resets all state generated by Keras.
Keras manages a global state, which it uses to implement the Functional
model-building API and to uniquify autogenerated layer names.
If you are creating many models in a loop, this global state will consume
an increasing amount of memory over time, and you may want to clear it.
Calling `clear_session()` releases the global state: this helps avoid clutter
from old models and layers, especially when memory is limited.
Example 1: calling `clear_session()` when creating models in a loop
```python
for _ in range(100):
# Without `clear_session()`, each iteration of this loop will
# slightly increase the size of the global state managed by Keras
model = tf.keras.Sequential([tf.keras.layers.Dense(10) for _ in range(10)])
for _ in range(100):
# With `clear_session()` called at the beginning,
# Keras starts with a blank state at each iteration
# and memory consumption is constant over time.
tf.keras.backend.clear_session()
model = tf.keras.Sequential([tf.keras.layers.Dense(10) for _ in range(10)])
```
Example 2: resetting the layer name generation counter
>>> import tensorflow as tf
>>> layers = [tf.keras.layers.Dense(10) for _ in range(10)]
>>> new_layer = tf.keras.layers.Dense(10)
>>> print(new_layer.name)
dense_10
>>> tf.keras.backend.set_learning_phase(1)
>>> print(tf.keras.backend.learning_phase())
1
>>> tf.keras.backend.clear_session()
>>> new_layer = tf.keras.layers.Dense(10)
>>> print(new_layer.name)
dense"
4544,manual_variable_initialization,tensorflow/tensorflow/python/keras/backend.py,314,function,"Sets the manual variable initialization flag.
This boolean flag determines whether
variables should be initialized
as they are instantiated (default), or if
the user should handle the initialization
(e.g. via `tf.compat.v1.initialize_all_variables()`).
Arguments:
value: Python boolean."
4545,learning_phase,tensorflow/tensorflow/python/keras/backend.py,331,function,"Returns the learning phase flag.
The learning phase flag is a bool tensor (0 = test, 1 = train)
to be passed as input to any Keras function
that uses a different behavior at train time and test time.
Returns:
Learning phase (scalar integer tensor or Python integer)."
4546,global_learning_phase_is_set,tensorflow/tensorflow/python/keras/backend.py,360,function,
4547,symbolic_learning_phase,tensorflow/tensorflow/python/keras/backend.py,383,function,
4548,set_learning_phase,tensorflow/tensorflow/python/keras/backend.py,402,function,"Sets the learning phase to a fixed value.
The backend learning phase affects any code that calls
`backend.learning_phase()`
In particular, all Keras built-in layers use the learning phase as the default
for the `training` arg to `Layer.__call__`.
User-written layers and models can achieve the same behavior with code that
looks like:
```python
def call(self, inputs, training=None):
if training is None:
training = backend.learning_phase()
```
Arguments:
value: Learning phase value, either 0 or 1 (integers).
0 = test, 1 = train
Raises:
ValueError: if `value` is neither `0` nor `1`."
4549,deprecated_internal_set_learning_phase,tensorflow/tensorflow/python/keras/backend.py,429,function,"A deprecated internal implementation of set_learning_phase.
This method is an internal-only version of `set_learning_phase` that
does not raise a deprecation error. It is required because
saved_model needs to keep working with user code that uses the deprecated
learning phase methods until those apis are fully removed from the public api.
Specifically SavedModel saving needs to make sure the learning phase is 0
during tracing even if users overwrote it to a different value.
But, we don't want to raise deprecation warnings for users when savedmodel
sets learning phase just for compatibility with code that relied on
explicitly setting the learning phase for other values.
Arguments:
value: Learning phase value, either 0 or 1 (integers). 0 = test, 1 = train
Raises:
ValueError: if `value` is neither `0` nor `1`."
4550,learning_phase_scope,tensorflow/tensorflow/python/keras/backend.py,467,function,"Provides a scope within which the learning phase is equal to `value`.
The learning phase gets restored to its original value upon exiting the scope.
Arguments:
value: Learning phase value, either 0 or 1 (integers).
0 = test, 1 = train
Yields:
None.
Raises:
ValueError: if `value` is neither `0` nor `1`."
4551,deprecated_internal_learning_phase_scope,tensorflow/tensorflow/python/keras/backend.py,490,function,"An internal-only version of `learning_phase_scope`.
Unlike the public method, this method does not raise a deprecation warning.
This is needed because saved model saving needs to set learning phase
to maintain compatibility
with code that sets/gets the learning phase, but saved model
saving itself shouldn't raise a deprecation warning.
We can get rid of this method and its usages when the public api is
removed.
Arguments:
value: Learning phase value, either 0 or 1 (integers). 0 = test, 1 = train
Yields:
None.
Raises:
ValueError: if `value` is neither `0` nor `1`."
4552,eager_learning_phase_scope,tensorflow/tensorflow/python/keras/backend.py,544,function,"Internal scope that sets the learning phase in eager / tf.function only.
Arguments:
value: Learning phase value, either 0 or 1 (integers).
0 = test, 1 = train
Yields:
None.
Raises:
ValueError: if `value` is neither `0` nor `1`."
4553,get_session,tensorflow/tensorflow/python/keras/backend.py,605,function,"Returns the TF session to be used by the backend.
If a default TensorFlow session is available, we will return it.
Else, we will return the global Keras session assuming it matches
the current graph.
If no global Keras session exists at this point:
we will create a new global session.
Note that you can manually set the global session
via `K.set_session(sess)`.
Arguments:
op_input_list: An option sequence of tensors or ops, which will be used
to determine the current graph. Otherwise the default graph will be
used.
Returns:
A TensorFlow session."
4554,get_graph,tensorflow/tensorflow/python/keras/backend.py,639,function,
4555,set_session,tensorflow/tensorflow/python/keras/backend.py,686,function,"Sets the global TensorFlow session.
Arguments:
session: A TF Session."
4556,get_default_session_config,tensorflow/tensorflow/python/keras/backend.py,696,function,
4557,get_default_graph_uid_map,tensorflow/tensorflow/python/keras/backend.py,708,function,
4558,is_sparse,tensorflow/tensorflow/python/keras/backend.py,839,function,"Returns whether a tensor is a sparse tensor.
Arguments:
tensor: A tensor instance.
Returns:
A boolean.
Example:
>>> a = tf.keras.backend.placeholder((2, 2), sparse=False)
>>> print(tf.keras.backend.is_sparse(a))
False
>>> b = tf.keras.backend.placeholder((2, 2), sparse=True)
>>> print(tf.keras.backend.is_sparse(b))
True"
4559,to_dense,tensorflow/tensorflow/python/keras/backend.py,867,function,"Converts a sparse tensor into a dense tensor and returns it.
Arguments:
tensor: A tensor instance (potentially sparse).
Returns:
A dense tensor.
Examples:
>>> b = tf.keras.backend.placeholder((2, 2), sparse=True)
>>> print(tf.keras.backend.is_sparse(b))
True
>>> c = tf.keras.backend.to_dense(b)
>>> print(tf.keras.backend.is_sparse(c))
False"
4560,name_scope,tensorflow/tensorflow/python/keras/backend.py,894,function,"A context manager for use when defining a Python op.
This context manager pushes a name scope, which will make the name of all
operations added within it have a prefix.
For example, to define a new Python op called `my_op`:
def my_op(a):
with tf.name_scope(""MyOp"") as scope:
a = tf.convert_to_tensor(a, name=""a"")
# Define some computation that uses `a`.
return foo_op(..., name=scope)
When executed, the Tensor `a` will have the name `MyOp/a`.
Args:
name: The prefix to use on all names created within the name scope.
Returns:
Name scope context manager."
4561,variable,tensorflow/tensorflow/python/keras/backend.py,925,function,"Instantiates a variable and returns it.
Arguments:
value: Numpy array, initial value of the tensor.
dtype: Tensor type.
name: Optional name string for the tensor.
constraint: Optional projection function to be
applied to the variable after an optimizer update.
Returns:
A variable instance (with Keras metadata included).
Examples:
>>> val = np.array([[1, 2], [3, 4]])
>>> kvar = tf.keras.backend.variable(value=val, dtype='float64',
... name='example_var')
>>> tf.keras.backend.dtype(kvar)
'float64'
>>> print(kvar)
<tf.Variable 'example_var:...' shape=(2, 2) dtype=float64, numpy=
array([[1., 2.],
[3., 4.]])>"
4562,track_tf_optimizer,tensorflow/tensorflow/python/keras/backend.py,974,function,Tracks the given TF optimizer for initialization of its variables.
4563,track_variable,tensorflow/tensorflow/python/keras/backend.py,982,function,Tracks the given variable for initialization.
4564,unique_object_name,tensorflow/tensorflow/python/keras/backend.py,990,function,"Makes a object name (or arbitrary string) unique within a TensorFlow graph.
Arguments:
name: String name to make unique.
name_uid_map: An optional defaultdict(int) to use when creating unique
names. If None (default), uses a per-Graph dictionary.
avoid_names: An optional set or dict with names which should not be used. If
None (default) does not avoid any names.
namespace: Gets a name which is unique within the (graph, namespace). Layers
which are not Networks use a blank namespace and so get graph-global
names.
zero_based: If True, name sequences start with no suffix (e.g. ""dense"",
""dense_1""). If False, naming is one-based (""dense_1"", ""dense_2"").
Returns:
Unique string name.
Example:
unique_object_name('dense') # dense_1
unique_object_name('dense') # dense_2"
4565,constant,tensorflow/tensorflow/python/keras/backend.py,1076,function,"Creates a constant tensor.
Arguments:
value: A constant value (or list)
dtype: The type of the elements of the resulting tensor.
shape: Optional dimensions of resulting tensor.
name: Optional name for the tensor.
Returns:
A Constant Tensor."
4566,is_keras_tensor,tensorflow/tensorflow/python/keras/backend.py,1095,function,"Returns whether `x` is a Keras tensor.
A ""Keras tensor"" is a tensor that was returned by a Keras layer,
(`Layer` class) or by `Input`.
Arguments:
x: A candidate tensor.
Returns:
A boolean: Whether the argument is a Keras tensor.
Raises:
ValueError: In case `x` is not a symbolic tensor.
Examples:
>>> np_var = np.array([1, 2])
>>> # A numpy array is not a symbolic tensor.
>>> tf.keras.backend.is_keras_tensor(np_var)
Traceback (most recent call last):
...
ValueError: Unexpectedly found an instance of type `<class 'numpy.ndarray'>`.
Expected a symbolic tensor instance.
>>> keras_var = tf.keras.backend.variable(np_var)
>>> # A variable created with the keras backend is not a Keras tensor.
>>> tf.keras.backend.is_keras_tensor(keras_var)
False
>>> keras_placeholder = tf.keras.backend.placeholder(shape=(2, 4, 5))
>>> # A placeholder is a Keras tensor.
>>> tf.keras.backend.is_keras_tensor(keras_placeholder)
True
>>> keras_input = tf.keras.layers.Input([10])
>>> # An Input is a Keras tensor.
>>> tf.keras.backend.is_keras_tensor(keras_input)
True
>>> keras_layer_output = tf.keras.layers.Dense(10)(keras_input)
>>> # Any Keras layer output is a Keras tensor.
>>> tf.keras.backend.is_keras_tensor(keras_layer_output)
True"
4567,placeholder,tensorflow/tensorflow/python/keras/backend.py,1149,function,"Instantiates a placeholder tensor and returns it.
Arguments:
shape: Shape of the placeholder
(integer tuple, may include `None` entries).
ndim: Number of axes of the tensor.
At least one of {`shape`, `ndim`} must be specified.
If both are specified, `shape` is used.
dtype: Placeholder type.
sparse: Boolean, whether the placeholder should have a sparse type.
name: Optional name string for the placeholder.
ragged: Boolean, whether the placeholder should have a ragged type.
In this case, values of 'None' in the 'shape' argument represent
ragged dimensions. For more information about RaggedTensors, see this
[guide](https://www.tensorflow.org/guide/ragged_tensors).
Raises:
ValueError: If called with eager execution
ValueError: If called with sparse = True and ragged = True.
Returns:
Tensor instance (with Keras metadata included).
Examples:
>>> input_ph = tf.keras.backend.placeholder(shape=(2, 4, 5))
>>> input_ph
<tf.Tensor 'Placeholder_...' shape=(2, 4, 5) dtype=float32>"
4568,is_placeholder,tensorflow/tensorflow/python/keras/backend.py,1244,function,"Returns whether `x` is a placeholder.
Arguments:
x: A candidate placeholder.
Returns:
Boolean."
4569,shape,tensorflow/tensorflow/python/keras/backend.py,1267,function,"Returns the symbolic shape of a tensor or variable.
Arguments:
x: A tensor or variable.
Returns:
A symbolic shape (which is itself a tensor).
Examples:
>>> val = np.array([[1, 2], [3, 4]])
>>> kvar = tf.keras.backend.variable(value=val)
>>> tf.keras.backend.shape(kvar)
<tf.Tensor: shape=(2,), dtype=int32, numpy=array([2, 2], dtype=int32)>
>>> input = tf.keras.backend.placeholder(shape=(2, 4, 5))
>>> tf.keras.backend.shape(input)
<tf.Tensor 'Shape_...' shape=(3,) dtype=int32>"
4570,int_shape,tensorflow/tensorflow/python/keras/backend.py,1291,function,"Returns the shape of tensor or variable as a tuple of int or None entries.
Arguments:
x: Tensor or variable.
Returns:
A tuple of integers (or None entries).
Examples:
>>> input = tf.keras.backend.placeholder(shape=(2, 4, 5))
>>> tf.keras.backend.int_shape(input)
(2, 4, 5)
>>> val = np.array([[1, 2], [3, 4]])
>>> kvar = tf.keras.backend.variable(value=val)
>>> tf.keras.backend.int_shape(kvar)
(2, 2)"
4571,ndim,tensorflow/tensorflow/python/keras/backend.py,1321,function,"Returns the number of axes in a tensor, as an integer.
Arguments:
x: Tensor or variable.
Returns:
Integer (scalar), number of axes.
Examples:
>>> input = tf.keras.backend.placeholder(shape=(2, 4, 5))
>>> val = np.array([[1, 2], [3, 4]])
>>> kvar = tf.keras.backend.variable(value=val)
>>> tf.keras.backend.ndim(input)
3
>>> tf.keras.backend.ndim(kvar)
2"
4572,dtype,tensorflow/tensorflow/python/keras/backend.py,1350,function,"Returns the dtype of a Keras tensor or variable, as a string.
Arguments:
x: Tensor or variable.
Returns:
String, dtype of `x`.
Examples:
>>> tf.keras.backend.dtype(tf.keras.backend.placeholder(shape=(2,4,5)))
'float32'
>>> tf.keras.backend.dtype(tf.keras.backend.placeholder(shape=(2,4,5),
... dtype='float32'))
'float32'
>>> tf.keras.backend.dtype(tf.keras.backend.placeholder(shape=(2,4,5),
... dtype='float64'))
'float64'
>>> kvar = tf.keras.backend.variable(np.array([[1, 2], [3, 4]]))
>>> tf.keras.backend.dtype(kvar)
'float32'
>>> kvar = tf.keras.backend.variable(np.array([[1, 2], [3, 4]]),
... dtype='float32')
>>> tf.keras.backend.dtype(kvar)
'float32'"
4573,eval,tensorflow/tensorflow/python/keras/backend.py,1382,function,"Evaluates the value of a variable.
Arguments:
x: A variable.
Returns:
A Numpy array.
Examples:
>>> kvar = tf.keras.backend.variable(np.array([[1, 2], [3, 4]]),
... dtype='float32')
>>> tf.keras.backend.eval(kvar)
array([[1., 2.],
[3., 4.]], dtype=float32)"
4574,zeros,tensorflow/tensorflow/python/keras/backend.py,1404,function,"Instantiates an all-zeros variable and returns it.
Arguments:
shape: Tuple or list of integers, shape of returned Keras variable
dtype: data type of returned Keras variable
name: name of returned Keras variable
Returns:
A variable (including Keras metadata), filled with `0.0`.
Note that if `shape` was symbolic, we cannot return a variable,
and will return a dynamically-shaped tensor instead.
Example:
>>> kvar = tf.keras.backend.zeros((3,4))
>>> tf.keras.backend.eval(kvar)
array([[0., 0., 0., 0.],
[0., 0., 0., 0.],
[0., 0., 0., 0.]], dtype=float32)
>>> A = tf.constant([1,2,3])
>>> kvar2 = tf.keras.backend.zeros(A.shape) # [0., 0., 0.]
>>> tf.keras.backend.eval(kvar2)
array([0., 0., 0.], dtype=float32)
>>> kvar3 = tf.keras.backend.zeros(A.shape,dtype=tf.int32)
>>> tf.keras.backend.eval(kvar3)
array([0, 0, 0], dtype=int32)
>>> kvar4 = tf.keras.backend.zeros([2,3])
>>> tf.keras.backend.eval(kvar4)
array([[0., 0., 0.],
[0., 0., 0.]], dtype=float32)"
4575,ones,tensorflow/tensorflow/python/keras/backend.py,1449,function,"Instantiates an all-ones variable and returns it.
Arguments:
shape: Tuple of integers, shape of returned Keras variable.
dtype: String, data type of returned Keras variable.
name: String, name of returned Keras variable.
Returns:
A Keras variable, filled with `1.0`.
Note that if `shape` was symbolic, we cannot return a variable,
and will return a dynamically-shaped tensor instead.
Example:
>>> kvar = tf.keras.backend.ones((3,4))
>>> tf.keras.backend.eval(kvar)
array([[1., 1., 1., 1.],
[1., 1., 1., 1.],
[1., 1., 1., 1.]], dtype=float32)"
4576,eye,tensorflow/tensorflow/python/keras/backend.py,1484,function,"Instantiate an identity matrix and returns it.
Arguments:
size: Integer, number of rows/columns.
dtype: String, data type of returned Keras variable.
name: String, name of returned Keras variable.
Returns:
A Keras variable, an identity matrix.
Example:
>>> kvar = tf.keras.backend.eye(3)
>>> tf.keras.backend.eval(kvar)
array([[1., 0., 0.],
[0., 1., 0.],
[0., 0., 1.]], dtype=float32)"
4577,zeros_like,tensorflow/tensorflow/python/keras/backend.py,1513,function,"Instantiates an all-zeros variable of the same shape as another tensor.
Arguments:
x: Keras variable or Keras tensor.
dtype: dtype of returned Keras variable.
`None` uses the dtype of `x`.
name: name for the variable to create.
Returns:
A Keras variable with the shape of `x` filled with zeros.
Example:
from tensorflow.keras import backend as K
kvar = K.variable(np.random.random((2,3)))
kvar_zeros = K.zeros_like(kvar)
K.eval(kvar_zeros)
# array([[ 0., 0., 0.], [ 0., 0., 0.]], dtype=float32)"
4578,ones_like,tensorflow/tensorflow/python/keras/backend.py,1541,function,"Instantiates an all-ones variable of the same shape as another tensor.
Arguments:
x: Keras variable or tensor.
dtype: String, dtype of returned Keras variable.
None uses the dtype of x.
name: String, name for the variable to create.
Returns:
A Keras variable with the shape of x filled with ones.
Example:
>>> kvar = tf.keras.backend.variable(np.random.random((2,3)))
>>> kvar_ones = tf.keras.backend.ones_like(kvar)
>>> tf.keras.backend.eval(kvar_ones)
array([[1., 1., 1.],
[1., 1., 1.]], dtype=float32)"
4579,identity,tensorflow/tensorflow/python/keras/backend.py,1565,function,"Returns a tensor with the same content as the input tensor.
Arguments:
x: The input tensor.
name: String, name for the variable to create.
Returns:
A tensor of the same shape, type and content."
4580,random_uniform_variable,tensorflow/tensorflow/python/keras/backend.py,1579,function,"Instantiates a variable with values drawn from a uniform distribution.
Arguments:
shape: Tuple of integers, shape of returned Keras variable.
low: Float, lower boundary of the output interval.
high: Float, upper boundary of the output interval.
dtype: String, dtype of returned Keras variable.
name: String, name of returned Keras variable.
seed: Integer, random seed.
Returns:
A Keras variable, filled with drawn samples.
Example:
>>> kvar = tf.keras.backend.random_uniform_variable(shape=(2,3),
... low=0.0, high=1.0)
>>> kvar
<tf.Variable 'Variable:0' shape=(2, 3) dtype=float32, numpy=...,
dtype=float32)>"
4581,random_normal_variable,tensorflow/tensorflow/python/keras/backend.py,1613,function,"Instantiates a variable with values drawn from a normal distribution.
Arguments:
shape: Tuple of integers, shape of returned Keras variable.
mean: Float, mean of the normal distribution.
scale: Float, standard deviation of the normal distribution.
dtype: String, dtype of returned Keras variable.
name: String, name of returned Keras variable.
seed: Integer, random seed.
Returns:
A Keras variable, filled with drawn samples.
Example:
>>> kvar = tf.keras.backend.random_normal_variable(shape=(2,3),
... mean=0.0, scale=1.0)
>>> kvar
<tf.Variable 'Variable:0' shape=(2, 3) dtype=float32, numpy=...,
dtype=float32)>"
4582,count_params,tensorflow/tensorflow/python/keras/backend.py,1648,function,"Returns the static number of elements in a variable or tensor.
Arguments:
x: Variable or tensor.
Returns:
Integer, the number of scalars in `x`.
Example:
>>> kvar = tf.keras.backend.zeros((2,3))
>>> tf.keras.backend.count_params(kvar)
6
>>> tf.keras.backend.eval(kvar)
array([[0., 0., 0.],
[0., 0., 0.]], dtype=float32)"
4583,cast,tensorflow/tensorflow/python/keras/backend.py,1672,function,"Casts a tensor to a different dtype and returns it.
You can cast a Keras variable but it still returns a Keras tensor.
Arguments:
x: Keras tensor (or variable).
dtype: String, either (`'float16'`, `'float32'`, or `'float64'`).
Returns:
Keras tensor with dtype `dtype`.
Examples:
Cast a float32 variable to a float64 tensor
>>> input = tf.keras.backend.ones(shape=(1,3))
>>> print(input)
<tf.Variable 'Variable:0' shape=(1, 3) dtype=float32,
numpy=array([[1., 1., 1.]], dtype=float32)>
>>> cast_input = tf.keras.backend.cast(input, dtype='float64')
>>> print(cast_input)
tf.Tensor([[1. 1. 1.]], shape=(1, 3), dtype=float64)"
4584,update,tensorflow/tensorflow/python/keras/backend.py,1703,function,
4585,update_add,tensorflow/tensorflow/python/keras/backend.py,1708,function,"Update the value of `x` by adding `increment`.
Arguments:
x: A Variable.
increment: A tensor of same shape as `x`.
Returns:
The variable `x` updated."
4586,update_sub,tensorflow/tensorflow/python/keras/backend.py,1722,function,"Update the value of `x` by subtracting `decrement`.
Arguments:
x: A Variable.
decrement: A tensor of same shape as `x`.
Returns:
The variable `x` updated."
4587,moving_average_update,tensorflow/tensorflow/python/keras/backend.py,1736,function,"Compute the exponential moving average of a value.
The moving average 'x' is updated with 'value' following:
```
x = x * momentum + value * (1 - momentum)
```
For example:
>>> x = tf.Variable(0.0)
>>> momentum=0.9
>>> moving_average_update(x, value = 2.0, momentum=momentum).numpy()
>>> x.numpy()
0.2
The result will be biased towards the initial value of the variable.
If the variable was initialized to zero, you can divide by
`1 - momentum ** num_updates` to debias it (Section 3 of
[Kingma et al., 2015](https://arxiv.org/abs/1412.6980)):
>>> num_updates = 1.0
>>> x_zdb = x/(1 - momentum**num_updates)
>>> x_zdb.numpy()
2.0
Arguments:
x: A Variable, the moving average.
value: A tensor with the same shape as `x`, the new value to be
averaged in.
momentum: The moving average momentum.
Returns:
The updated variable."
4588,dot,tensorflow/tensorflow/python/keras/backend.py,1783,function,"Multiplies 2 tensors (and/or variables) and returns a tensor.
Arguments:
x: Tensor or variable.
y: Tensor or variable.
Returns:
A tensor, dot product of `x` and `y`.
Examples:
>>> x = tf.keras.backend.placeholder(shape=(2, 3))
>>> y = tf.keras.backend.placeholder(shape=(3, 4))
>>> xy = tf.keras.backend.dot(x, y)
>>> xy
<tf.Tensor ... shape=(2, 4) dtype=float32>
>>> x = tf.keras.backend.placeholder(shape=(32, 28, 3))
>>> y = tf.keras.backend.placeholder(shape=(3, 4))
>>> xy = tf.keras.backend.dot(x, y)
>>> xy
<tf.Tensor ... shape=(32, 28, 4) dtype=float32>
>>> x = tf.keras.backend.random_uniform_variable(shape=(2, 3), low=0, high=1)
>>> y = tf.keras.backend.ones((4, 3, 5))
>>> xy = tf.keras.backend.dot(x, y)
>>> tf.keras.backend.int_shape(xy)
(2, 4, 5)"
4589,batch_dot,tensorflow/tensorflow/python/keras/backend.py,1844,function,"Batchwise dot product.
`batch_dot` is used to compute dot product of `x` and `y` when
`x` and `y` are data in batch, i.e. in a shape of
`(batch_size, :)`.
`batch_dot` results in a tensor or variable with less dimensions
than the input. If the number of dimensions is reduced to 1,
we use `expand_dims` to make sure that ndim is at least 2.
Arguments:
x: Keras tensor or variable with `ndim >= 2`.
y: Keras tensor or variable with `ndim >= 2`.
axes: Tuple or list of integers with target dimensions, or single integer.
The sizes of `x.shape[axes[0]]` and `y.shape[axes[1]]` should be equal.
Returns:
A tensor with shape equal to the concatenation of `x`'s shape
(less the dimension that was summed over) and `y`'s shape
(less the batch dimension and the dimension that was summed over).
If the final rank is 1, we reshape it to `(batch_size, 1)`.
Examples:
>>> x_batch = tf.keras.backend.ones(shape=(32, 20, 1))
>>> y_batch = tf.keras.backend.ones(shape=(32, 30, 20))
>>> xy_batch_dot = tf.keras.backend.batch_dot(x_batch, y_batch, axes=(1, 2))
>>> tf.keras.backend.int_shape(xy_batch_dot)
(32, 1, 30)
Shape inference:
Let `x`'s shape be `(100, 20)` and `y`'s shape be `(100, 30, 20)`.
If `axes` is (1, 2), to find the output shape of resultant tensor,
loop through each dimension in `x`'s shape and `y`'s shape:
* `x.shape[0]` : 100 : append to output shape
* `x.shape[1]` : 20 : do not append to output shape,
dimension 1 of `x` has been summed over. (`dot_axes[0]` = 1)
* `y.shape[0]` : 100 : do not append to output shape,
always ignore first dimension of `y`
* `y.shape[1]` : 30 : append to output shape
* `y.shape[2]` : 20 : do not append to output shape,
dimension 2 of `y` has been summed over. (`dot_axes[1]` = 2)
`output_shape` = `(100, 30)`"
4590,transpose,tensorflow/tensorflow/python/keras/backend.py,2033,function,"Transposes a tensor and returns it.
Arguments:
x: Tensor or variable.
Returns:
A tensor.
Examples:
>>> var = tf.keras.backend.variable([[1, 2, 3], [4, 5, 6]])
>>> tf.keras.backend.eval(var)
array([[1., 2., 3.],
[4., 5., 6.]], dtype=float32)
>>> var_transposed = tf.keras.backend.transpose(var)
>>> tf.keras.backend.eval(var_transposed)
array([[1., 4.],
[2., 5.],
[3., 6.]], dtype=float32)
>>> input = tf.keras.backend.placeholder((2, 3))
>>> input
<tf.Tensor 'Placeholder_...' shape=(2, 3) dtype=float32>
>>> input_transposed = tf.keras.backend.transpose(input)
>>> input_transposed
<tf.Tensor 'Transpose_...' shape=(3, 2) dtype=float32>"
4591,gather,tensorflow/tensorflow/python/keras/backend.py,2065,function,"Retrieves the elements of indices `indices` in the tensor `reference`.
Arguments:
reference: A tensor.
indices: An integer tensor of indices.
Returns:
A tensor of same type as `reference`.
Examples:
>>> var = tf.keras.backend.variable([[1, 2, 3], [4, 5, 6]])
>>> tf.keras.backend.eval(var)
array([[1., 2., 3.],
[4., 5., 6.]], dtype=float32)
>>> var_gathered = tf.keras.backend.gather(var, [0])
>>> tf.keras.backend.eval(var_gathered)
array([[1., 2., 3.]], dtype=float32)
>>> var_gathered = tf.keras.backend.gather(var, [1])
>>> tf.keras.backend.eval(var_gathered)
array([[4., 5., 6.]], dtype=float32)
>>> var_gathered = tf.keras.backend.gather(var, [0,1,0])
>>> tf.keras.backend.eval(var_gathered)
array([[1., 2., 3.],
[4., 5., 6.],
[1., 2., 3.]], dtype=float32)"
4592,max,tensorflow/tensorflow/python/keras/backend.py,2101,function,"Maximum value in a tensor.
Arguments:
x: A tensor or variable.
axis: An integer, the axis to find maximum values.
keepdims: A boolean, whether to keep the dimensions or not.
If `keepdims` is `False`, the rank of the tensor is reduced
by 1. If `keepdims` is `True`,
the reduced dimension is retained with length 1.
Returns:
A tensor with maximum values of `x`."
4593,min,tensorflow/tensorflow/python/keras/backend.py,2120,function,"Minimum value in a tensor.
Arguments:
x: A tensor or variable.
axis: An integer, the axis to find minimum values.
keepdims: A boolean, whether to keep the dimensions or not.
If `keepdims` is `False`, the rank of the tensor is reduced
by 1. If `keepdims` is `True`,
the reduced dimension is retained with length 1.
Returns:
A tensor with minimum values of `x`."
4594,sum,tensorflow/tensorflow/python/keras/backend.py,2139,function,"Sum of the values in a tensor, alongside the specified axis.
Arguments:
x: A tensor or variable.
axis: An integer, the axis to sum over.
keepdims: A boolean, whether to keep the dimensions or not.
If `keepdims` is `False`, the rank of the tensor is reduced
by 1. If `keepdims` is `True`,
the reduced dimension is retained with length 1.
Returns:
A tensor with sum of `x`."
4595,prod,tensorflow/tensorflow/python/keras/backend.py,2158,function,"Multiplies the values in a tensor, alongside the specified axis.
Arguments:
x: A tensor or variable.
axis: An integer, the axis to compute the product.
keepdims: A boolean, whether to keep the dimensions or not.
If `keepdims` is `False`, the rank of the tensor is reduced
by 1. If `keepdims` is `True`,
the reduced dimension is retained with length 1.
Returns:
A tensor with the product of elements of `x`."
4596,cumsum,tensorflow/tensorflow/python/keras/backend.py,2177,function,"Cumulative sum of the values in a tensor, alongside the specified axis.
Arguments:
x: A tensor or variable.
axis: An integer, the axis to compute the sum.
Returns:
A tensor of the cumulative sum of values of `x` along `axis`."
4597,cumprod,tensorflow/tensorflow/python/keras/backend.py,2192,function,"Cumulative product of the values in a tensor, alongside the specified axis.
Arguments:
x: A tensor or variable.
axis: An integer, the axis to compute the product.
Returns:
A tensor of the cumulative product of values of `x` along `axis`."
4598,var,tensorflow/tensorflow/python/keras/backend.py,2206,function,"Variance of a tensor, alongside the specified axis.
Arguments:
x: A tensor or variable.
axis: An integer, the axis to compute the variance.
keepdims: A boolean, whether to keep the dimensions or not.
If `keepdims` is `False`, the rank of the tensor is reduced
by 1. If `keepdims` is `True`,
the reduced dimension is retained with length 1.
Returns:
A tensor with the variance of elements of `x`."
4599,std,tensorflow/tensorflow/python/keras/backend.py,2227,function,"Standard deviation of a tensor, alongside the specified axis.
It is an alias to `tf.math.reduce_std`.
Arguments:
x: A tensor or variable. It should have numerical dtypes. Boolean type
inputs will be converted to float.
axis: An integer, the axis to compute the standard deviation. If `None`
(the default), reduces all dimensions. Must be in the range
`[-rank(x), rank(x))`.
keepdims: A boolean, whether to keep the dimensions or not.
If `keepdims` is `False`, the rank of the tensor is reduced
by 1. If `keepdims` is `True`, the reduced dimension is retained with
length 1.
Returns:
A tensor with the standard deviation of elements of `x` with same dtype.
Boolean type input will be converted to float."
4600,mean,tensorflow/tensorflow/python/keras/backend.py,2254,function,"Mean of a tensor, alongside the specified axis.
Arguments:
x: A tensor or variable.
axis: A list of integer. Axes to compute the mean.
keepdims: A boolean, whether to keep the dimensions or not.
If `keepdims` is `False`, the rank of the tensor is reduced
by 1 for each entry in `axis`. If `keepdims` is `True`,
the reduced dimensions are retained with length 1.
Returns:
A tensor with the mean of elements of `x`."
4601,any,tensorflow/tensorflow/python/keras/backend.py,2275,function,"Bitwise reduction (logical OR).
Arguments:
x: Tensor or variable.
axis: axis along which to perform the reduction.
keepdims: whether the drop or broadcast the reduction axes.
Returns:
A uint8 tensor (0s and 1s)."
4602,all,tensorflow/tensorflow/python/keras/backend.py,2292,function,"Bitwise reduction (logical AND).
Arguments:
x: Tensor or variable.
axis: axis along which to perform the reduction.
keepdims: whether the drop or broadcast the reduction axes.
Returns:
A uint8 tensor (0s and 1s)."
4603,argmax,tensorflow/tensorflow/python/keras/backend.py,2309,function,"Returns the index of the maximum value along an axis.
Arguments:
x: Tensor or variable.
axis: axis along which to perform the reduction.
Returns:
A tensor."
4604,argmin,tensorflow/tensorflow/python/keras/backend.py,2324,function,"Returns the index of the minimum value along an axis.
Arguments:
x: Tensor or variable.
axis: axis along which to perform the reduction.
Returns:
A tensor."
4605,square,tensorflow/tensorflow/python/keras/backend.py,2339,function,"Element-wise square.
Arguments:
x: Tensor or variable.
Returns:
A tensor."
4606,abs,tensorflow/tensorflow/python/keras/backend.py,2353,function,"Element-wise absolute value.
Arguments:
x: Tensor or variable.
Returns:
A tensor."
4607,sqrt,tensorflow/tensorflow/python/keras/backend.py,2367,function,"Element-wise square root.
Arguments:
x: Tensor or variable.
Returns:
A tensor."
4608,exp,tensorflow/tensorflow/python/keras/backend.py,2384,function,"Element-wise exponential.
Arguments:
x: Tensor or variable.
Returns:
A tensor."
4609,log,tensorflow/tensorflow/python/keras/backend.py,2398,function,"Element-wise log.
Arguments:
x: Tensor or variable.
Returns:
A tensor."
4610,logsumexp,tensorflow/tensorflow/python/keras/backend.py,2410,function,"Computes log(sum(exp(elements across dimensions of a tensor))).
This function is more numerically stable than log(sum(exp(x))).
It avoids overflows caused by taking the exp of large inputs and
underflows caused by taking the log of small inputs.
Arguments:
x: A tensor or variable.
axis: An integer, the axis to reduce over.
keepdims: A boolean, whether to keep the dimensions or not.
If `keepdims` is `False`, the rank of the tensor is reduced
by 1. If `keepdims` is `True`, the reduced dimension is
retained with length 1.
Returns:
The reduced tensor."
4611,round,tensorflow/tensorflow/python/keras/backend.py,2433,function,"Element-wise rounding to the closest integer.
In case of tie, the rounding mode used is ""half to even"".
Arguments:
x: Tensor or variable.
Returns:
A tensor."
4612,sign,tensorflow/tensorflow/python/keras/backend.py,2449,function,"Element-wise sign.
Arguments:
x: Tensor or variable.
Returns:
A tensor."
4613,pow,tensorflow/tensorflow/python/keras/backend.py,2463,function,"Element-wise exponentiation.
Arguments:
x: Tensor or variable.
a: Python integer.
Returns:
A tensor."
4614,clip,tensorflow/tensorflow/python/keras/backend.py,2478,function,"Element-wise value clipping.
Arguments:
x: Tensor or variable.
min_value: Python float, integer, or tensor.
max_value: Python float, integer, or tensor.
Returns:
A tensor."
4615,equal,tensorflow/tensorflow/python/keras/backend.py,2502,function,"Element-wise equality between two tensors.
Arguments:
x: Tensor or variable.
y: Tensor or variable.
Returns:
A bool tensor."
4616,not_equal,tensorflow/tensorflow/python/keras/backend.py,2517,function,"Element-wise inequality between two tensors.
Arguments:
x: Tensor or variable.
y: Tensor or variable.
Returns:
A bool tensor."
4617,greater,tensorflow/tensorflow/python/keras/backend.py,2532,function,"Element-wise truth value of (x > y).
Arguments:
x: Tensor or variable.
y: Tensor or variable.
Returns:
A bool tensor."
4618,greater_equal,tensorflow/tensorflow/python/keras/backend.py,2547,function,"Element-wise truth value of (x >= y).
Arguments:
x: Tensor or variable.
y: Tensor or variable.
Returns:
A bool tensor."
4619,less,tensorflow/tensorflow/python/keras/backend.py,2562,function,"Element-wise truth value of (x < y).
Arguments:
x: Tensor or variable.
y: Tensor or variable.
Returns:
A bool tensor."
4620,less_equal,tensorflow/tensorflow/python/keras/backend.py,2577,function,"Element-wise truth value of (x <= y).
Arguments:
x: Tensor or variable.
y: Tensor or variable.
Returns:
A bool tensor."
4621,maximum,tensorflow/tensorflow/python/keras/backend.py,2592,function,"Element-wise maximum of two tensors.
Arguments:
x: Tensor or variable.
y: Tensor or variable.
Returns:
A tensor with the element wise maximum value(s) of `x` and `y`.
Examples:
>>> x = tf.Variable([[1, 2], [3, 4]])
>>> y = tf.Variable([[2, 1], [0, -1]])
>>> m = tf.keras.backend.maximum(x, y)
>>> m
<tf.Tensor: shape=(2, 2), dtype=int32, numpy=
array([[2, 2],
[3, 4]], dtype=int32)>"
4622,minimum,tensorflow/tensorflow/python/keras/backend.py,2617,function,"Element-wise minimum of two tensors.
Arguments:
x: Tensor or variable.
y: Tensor or variable.
Returns:
A tensor."
4623,sin,tensorflow/tensorflow/python/keras/backend.py,2632,function,"Computes sin of x element-wise.
Arguments:
x: Tensor or variable.
Returns:
A tensor."
4624,cos,tensorflow/tensorflow/python/keras/backend.py,2646,function,"Computes cos of x element-wise.
Arguments:
x: Tensor or variable.
Returns:
A tensor."
4625,normalize_batch_in_training,tensorflow/tensorflow/python/keras/backend.py,2761,function,"Computes mean and std for batch then apply batch_normalization on batch.
Arguments:
x: Input tensor or variable.
gamma: Tensor by which to scale the input.
beta: Tensor with which to center the input.
reduction_axes: iterable of integers,
axes over which to normalize.
epsilon: Fuzz factor.
Returns:
A tuple length of 3, `(normalized_tensor, mean, variance)`."
4626,batch_normalization,tensorflow/tensorflow/python/keras/backend.py,2792,function,"Applies batch normalization on x given mean, var, beta and gamma.
I.e. returns:
`output = (x - mean) / (sqrt(var) + epsilon) * gamma + beta`
Arguments:
x: Input tensor or variable.
mean: Mean of batch.
var: Variance of batch.
beta: Tensor with which to center the input.
gamma: Tensor by which to scale the input.
axis: Integer, the axis that should be normalized.
(typically the features axis).
epsilon: Fuzz factor.
Returns:
A tensor."
4627,concatenate,tensorflow/tensorflow/python/keras/backend.py,2855,function,"Concatenates a list of tensors alongside the specified axis.
Arguments:
tensors: list of tensors to concatenate.
axis: concatenation axis.
Returns:
A tensor.
Example:
>>> a = tf.constant([[1, 2, 3], [4, 5, 6], [7, 8, 9]])
>>> b = tf.constant([[10, 20, 30], [40, 50, 60], [70, 80, 90]])
>>> tf.keras.backend.concatenate((a, b), axis=-1)
<tf.Tensor: shape=(3, 6), dtype=int32, numpy=
array([[ 1, 2, 3, 10, 20, 30],
[ 4, 5, 6, 40, 50, 60],
[ 7, 8, 9, 70, 80, 90]], dtype=int32)>"
4628,reshape,tensorflow/tensorflow/python/keras/backend.py,2893,function,"Reshapes a tensor to the specified shape.
Arguments:
x: Tensor or variable.
shape: Target shape tuple.
Returns:
A tensor.
Example:
>>> a = tf.constant([[1, 2, 3], [4, 5, 6], [7, 8, 9], [10, 11, 12]])
>>> a
<tf.Tensor: shape=(4, 3), dtype=int32, numpy=
array([[ 1, 2, 3],
[ 4, 5, 6],
[ 7, 8, 9],
[10, 11, 12]], dtype=int32)>
>>> tf.keras.backend.reshape(a, shape=(2, 6))
<tf.Tensor: shape=(2, 6), dtype=int32, numpy=
array([[ 1, 2, 3, 4, 5, 6],
[ 7, 8, 9, 10, 11, 12]], dtype=int32)>"
4629,permute_dimensions,tensorflow/tensorflow/python/keras/backend.py,2923,function,"Permutes axes in a tensor.
Arguments:
x: Tensor or variable.
pattern: A tuple of
dimension indices, e.g. `(0, 2, 1)`.
Returns:
A tensor.
Example:
>>> a = tf.constant([[1, 2, 3], [4, 5, 6], [7, 8, 9], [10, 11, 12]])
>>> a
<tf.Tensor: shape=(4, 3), dtype=int32, numpy=
array([[ 1, 2, 3],
[ 4, 5, 6],
[ 7, 8, 9],
[10, 11, 12]], dtype=int32)>
>>> tf.keras.backend.permute_dimensions(a, pattern=(1, 0))
<tf.Tensor: shape=(3, 4), dtype=int32, numpy=
array([[ 1, 4, 7, 10],
[ 2, 5, 8, 11],
[ 3, 6, 9, 12]], dtype=int32)>"
4630,resize_images,tensorflow/tensorflow/python/keras/backend.py,2955,function,"Resizes the images contained in a 4D tensor.
Arguments:
x: Tensor or variable to resize.
height_factor: Positive integer.
width_factor: Positive integer.
data_format: One of `""channels_first""`, `""channels_last""`.
interpolation: A string, one of `nearest` or `bilinear`.
Returns:
A tensor.
Raises:
ValueError: in case of incorrect value for
`data_format` or `interpolation`."
4631,resize_volumes,tensorflow/tensorflow/python/keras/backend.py,3019,function,"Resizes the volume contained in a 5D tensor.
Arguments:
x: Tensor or variable to resize.
depth_factor: Positive integer.
height_factor: Positive integer.
width_factor: Positive integer.
data_format: One of `""channels_first""`, `""channels_last""`.
Returns:
A tensor.
Raises:
ValueError: if `data_format` is neither
`channels_last` or `channels_first`."
4632,repeat_elements,tensorflow/tensorflow/python/keras/backend.py,3052,function,"Repeats the elements of a tensor along an axis, like `np.repeat`.
If `x` has shape `(s1, s2, s3)` and `axis` is `1`, the output
will have shape `(s1, s2 * rep, s3)`.
Arguments:
x: Tensor or variable.
rep: Python integer, number of times to repeat.
axis: Axis along which to repeat.
Returns:
A tensor.
Example:
>>> b = tf.constant([1, 2, 3])
>>> tf.keras.backend.repeat_elements(b, rep=2, axis=0)
<tf.Tensor: shape=(6,), dtype=int32,
numpy=array([1, 1, 2, 2, 3, 3], dtype=int32)>"
4633,repeat,tensorflow/tensorflow/python/keras/backend.py,3114,function,"Repeats a 2D tensor.
if `x` has shape (samples, dim) and `n` is `2`,
the output will have shape `(samples, 2, dim)`.
Arguments:
x: Tensor or variable.
n: Python integer, number of times to repeat.
Returns:
A tensor.
Example:
>>> b = tf.constant([[1, 2], [3, 4]])
>>> b
<tf.Tensor: shape=(2, 2), dtype=int32, numpy=
array([[1, 2],
[3, 4]], dtype=int32)>
>>> tf.keras.backend.repeat(b, n=2)
<tf.Tensor: shape=(2, 2, 2), dtype=int32, numpy=
array([[[1, 2],
[1, 2]],
[[3, 4],
[3, 4]]], dtype=int32)>"
4634,arange,tensorflow/tensorflow/python/keras/backend.py,3150,function,"Creates a 1D tensor containing a sequence of integers.
The function arguments use the same convention as
Theano's arange: if only one argument is provided,
it is in fact the ""stop"" argument and ""start"" is 0.
The default type of the returned tensor is `'int32'` to
match TensorFlow's default.
Arguments:
start: Start value.
stop: Stop value.
step: Difference between two successive values.
dtype: Integer dtype to use.
Returns:
An integer tensor.
Example:
>>> tf.keras.backend.arange(start=0, stop=10, step=1.5)
<tf.Tensor: shape=(7,), dtype=float32,
numpy=array([0. , 1.5, 3. , 4.5, 6. , 7.5, 9. ], dtype=float32)>"
4635,tile,tensorflow/tensorflow/python/keras/backend.py,3189,function,"Creates a tensor by tiling `x` by `n`.
Arguments:
x: A tensor or variable
n: A list of integer. The length must be the same as the number of
dimensions in `x`.
Returns:
A tiled tensor."
4636,flatten,tensorflow/tensorflow/python/keras/backend.py,3207,function,"Flatten a tensor.
Arguments:
x: A tensor or variable.
Returns:
A tensor, reshaped into 1-D
Example:
>>> b = tf.constant([[1, 2], [3, 4]])
>>> b
<tf.Tensor: shape=(2, 2), dtype=int32, numpy=
array([[1, 2],
[3, 4]], dtype=int32)>
>>> tf.keras.backend.flatten(b)
<tf.Tensor: shape=(4,), dtype=int32,
numpy=array([1, 2, 3, 4], dtype=int32)>"
4637,batch_flatten,tensorflow/tensorflow/python/keras/backend.py,3233,function,"Turn a nD tensor into a 2D tensor with same 0th dimension.
In other words, it flattens each data samples of a batch.
Arguments:
x: A tensor or variable.
Returns:
A tensor.
Examples:
Flattening a 3D tensor to 2D by collapsing the last dimension.
>>> x_batch = tf.keras.backend.ones(shape=(2, 3, 4, 5))
>>> x_batch_flatten = batch_flatten(x_batch)
>>> tf.keras.backend.int_shape(x_batch_flatten)
(2, 60)"
4638,expand_dims,tensorflow/tensorflow/python/keras/backend.py,3259,function,"Adds a 1-sized dimension at index ""axis"".
Arguments:
x: A tensor or variable.
axis: Position where to add a new axis.
Returns:
A tensor with expanded dimensions."
4639,squeeze,tensorflow/tensorflow/python/keras/backend.py,3274,function,"Removes a 1-dimension from the tensor at index ""axis"".
Arguments:
x: A tensor or variable.
axis: Axis to drop.
Returns:
A tensor with the same data as `x` but reduced dimensions."
4640,temporal_padding,tensorflow/tensorflow/python/keras/backend.py,3289,function,"Pads the middle dimension of a 3D tensor.
Arguments:
x: Tensor or variable.
padding: Tuple of 2 integers, how many zeros to
add at the start and end of dim 1.
Returns:
A padded 3D tensor."
4641,spatial_2d_padding,tensorflow/tensorflow/python/keras/backend.py,3307,function,"Pads the 2nd and 3rd dimensions of a 4D tensor.
Arguments:
x: Tensor or variable.
padding: Tuple of 2 tuples, padding pattern.
data_format: One of `channels_last` or `channels_first`.
Returns:
A padded 4D tensor.
Raises:
ValueError: if `data_format` is neither
`channels_last` or `channels_first`."
4642,spatial_3d_padding,tensorflow/tensorflow/python/keras/backend.py,3339,function,"Pads 5D tensor with zeros along the depth, height, width dimensions.
Pads these dimensions with respectively
""padding[0]"", ""padding[1]"" and ""padding[2]"" zeros left and right.
For 'channels_last' data_format,
the 2nd, 3rd and 4th dimension will be padded.
For 'channels_first' data_format,
the 3rd, 4th and 5th dimension will be padded.
Arguments:
x: Tensor or variable.
padding: Tuple of 3 tuples, padding pattern.
data_format: One of `channels_last` or `channels_first`.
Returns:
A padded 5D tensor.
Raises:
ValueError: if `data_format` is neither
`channels_last` or `channels_first`."
4643,stack,tensorflow/tensorflow/python/keras/backend.py,3384,function,"Stacks a list of rank `R` tensors into a rank `R+1` tensor.
Arguments:
x: List of tensors.
axis: Axis along which to perform stacking.
Returns:
A tensor.
Example:
>>> a = tf.constant([[1, 2],[3, 4]])
>>> b = tf.constant([[10, 20],[30, 40]])
>>> tf.keras.backend.stack((a, b))
<tf.Tensor: shape=(2, 2, 2), dtype=int32, numpy=
array([[[ 1, 2],
[ 3, 4]],
[[10, 20],
[30, 40]]], dtype=int32)>"
4644,one_hot,tensorflow/tensorflow/python/keras/backend.py,3411,function,"Computes the one-hot representation of an integer tensor.
Arguments:
indices: nD integer tensor of shape
`(batch_size, dim1, dim2, ... dim(n-1))`
num_classes: Integer, number of classes to consider.
Returns:
(n + 1)D one hot representation of the input
with shape `(batch_size, dim1, dim2, ... dim(n-1), num_classes)`
Returns:
The one-hot tensor."
4645,reverse,tensorflow/tensorflow/python/keras/backend.py,3431,function,"Reverse a tensor along the specified axes.
Arguments:
x: Tensor to reverse.
axes: Integer or iterable of integers.
Axes to reverse.
Returns:
A tensor."
4646,get_value,tensorflow/tensorflow/python/keras/backend.py,3477,function,"Returns the value of a variable.
`backend.get_value` is the compliment of `backend.set_value`, and provides
a generic interface for reading from variables while abstracting away the
differences between TensorFlow 1.x and 2.x semantics.
{snippet}
Arguments:
x: input variable.
Returns:
A Numpy array."
4647,batch_get_value,tensorflow/tensorflow/python/keras/backend.py,3512,function,"Returns the value of more than one tensor variable.
Arguments:
tensors: list of ops to run.
Returns:
A list of Numpy arrays.
Raises:
RuntimeError: If this method is called inside defun."
4648,set_value,tensorflow/tensorflow/python/keras/backend.py,3535,function,"Sets the value of a variable, from a Numpy array.
`backend.set_value` is the compliment of `backend.get_value`, and provides
a generic interface for assigning to variables while abstracting away the
differences between TensorFlow 1.x and 2.x semantics.
{snippet}
Arguments:
x: Variable to set to a new value.
value: Value to set the tensor to, as a Numpy array
(of the same shape)."
4649,batch_set_value,tensorflow/tensorflow/python/keras/backend.py,3574,function,"Sets the values of many tensor variables at once.
Arguments:
tuples: a list of tuples `(tensor, value)`.
`value` should be a Numpy array."
4650,print_tensor,tensorflow/tensorflow/python/keras/backend.py,3617,function,"Prints `message` and the tensor value when evaluated.
Note that `print_tensor` returns a new tensor identical to `x`
which should be used in the following code. Otherwise the
print operation is not taken into account during evaluation.
Example:
>>> x = tf.constant([[1.0, 2.0], [3.0, 4.0]])
>>> tf.keras.backend.print_tensor(x)
<tf.Tensor: shape=(2, 2), dtype=float32, numpy=
array([[1., 2.],
[3., 4.]], dtype=float32)>
Arguments:
x: Tensor to print.
message: Message to print jointly with the tensor.
Returns:
The same tensor `x`, unchanged."
4651,GraphExecutionFunction,tensorflow/tensorflow/python/keras/backend.py,3651,class,"Runs a computation graph.
It's possible to pass arguments to `tf.Session.run()` via `session_kwargs`.
In particular additional operations via `fetches` argument and additional
tensor substitutions via `feed_dict` arguments. Note that given
substitutions are merged with substitutions from `inputs`. Even though
`feed_dict` is passed once in the constructor (called in `model.compile()`)
we can modify the values in the dictionary. Through this feed_dict we can
provide additional substitutions besides Keras inputs.
Arguments:
inputs: Feed placeholders to the computation graph.
outputs: Output tensors to fetch.
updates: Additional update ops to be run at function call.
name: A name to help users identify what this function does.
session_kwargs: Arguments to `tf.Session.run()`:
`fetches`, `feed_dict`, `options`, `run_metadata`."
4652,eval_in_eager_or_function,tensorflow/tensorflow/python/keras/backend.py,3847,function,"Method to evaluate a tensor in eager or in a tf.function.
In the case of a tf.function, it will lift the tensor out of the function
and try to evaluate that piece of the graph.
Warning: Do not add new usages of this function.
TODO(b/150169018): delete this function once _keras_history_helper is no
longer needed, after Keras switches to KerasTensors and op layers
work via dispatch.
Arguments:
outputs: tensors to fetch.
Returns:
The value of the tensors (as numpy arrays)."
4653,function,tensorflow/tensorflow/python/keras/backend.py,3918,function,"Instantiates a Keras function.
Arguments:
inputs: List of placeholder tensors.
outputs: List of output tensors.
updates: List of update ops.
name: String, name of function.
**kwargs: Passed to `tf.Session.run`.
Returns:
Output values as Numpy arrays.
Raises:
ValueError: if invalid kwargs are passed in or if in eager execution."
4654,gradients,tensorflow/tensorflow/python/keras/backend.py,3965,function,"Returns the gradients of `loss` w.r.t. `variables`.
Arguments:
loss: Scalar tensor to minimize.
variables: List of variables.
Returns:
A gradients tensor."
4655,stop_gradient,tensorflow/tensorflow/python/keras/backend.py,3981,function,"Returns `variables` but with zero gradient w.r.t. every other variable.
Arguments:
variables: Tensor or list of tensors to consider constant with respect
to any other variable.
Returns:
A single tensor or a list of tensors (depending on the passed argument)
that has no gradient with respect to any other variable."
4656,rnn,tensorflow/tensorflow/python/keras/backend.py,4003,function,"Iterates over the time dimension of a tensor.
Arguments:
step_function: RNN step function.
Args;
input; Tensor with shape `(samples, ...)` (no time dimension),
representing input for the batch of samples at a certain
time step.
states; List of tensors.
Returns;
output; Tensor with shape `(samples, output_dim)`
(no time dimension).
new_states; List of tensors, same length and shapes
as 'states'. The first state in the list must be the
output tensor at the previous timestep.
inputs: Tensor of temporal data of shape `(samples, time, ...)`
(at least 3D), or nested tensors, and each of which has shape
`(samples, time, ...)`.
initial_states: Tensor with shape `(samples, state_size)`
(no time dimension), containing the initial values for the states used
in the step function. In the case that state_size is in a nested
shape, the shape of initial_states will also follow the nested
structure.
go_backwards: Boolean. If True, do the iteration over the time
dimension in reverse order and return the reversed sequence.
mask: Binary tensor with shape `(samples, time, 1)`,
with a zero for every element that is masked.
constants: List of constant values passed at each step.
unroll: Whether to unroll the RNN or to use a symbolic `while_loop`.
input_length: An integer or a 1-D Tensor, depending on whether
the time dimension is fixed-length or not. In case of variable length
input, it is used for masking in case there's no mask specified.
time_major: Boolean. If true, the inputs and outputs will be in shape
`(timesteps, batch, ...)`, whereas in the False case, it will be
`(batch, timesteps, ...)`. Using `time_major = True` is a bit more
efficient because it avoids transposes at the beginning and end of the
RNN calculation. However, most TensorFlow data is batch-major, so by
default this function accepts input and emits output in batch-major
form.
zero_output_for_mask: Boolean. If True, the output for masked timestep
will be zeros, whereas in the False case, output from previous
timestep is returned.
Returns:
A tuple, `(last_output, outputs, new_states)`.
last_output: the latest output of the rnn, of shape `(samples, ...)`
outputs: tensor with shape `(samples, time, ...)` where each
entry `outputs[s, t]` is the output of the step function
at time `t` for sample `s`.
new_states: list of tensors, latest states returned by
the step function, of shape `(samples, ...)`.
Raises:
ValueError: if input dimension is less than 3.
ValueError: if `unroll` is `True` but input timestep is not a fixed
number.
ValueError: if `mask` is provided (not `None`) but states is not provided
(`len(states)` == 0)."
4657,switch,tensorflow/tensorflow/python/keras/backend.py,4398,function,"Switches between two operations depending on a scalar value.
Note that both `then_expression` and `else_expression`
should be symbolic tensors of the *same shape*.
Arguments:
condition: tensor (`int` or `bool`).
then_expression: either a tensor, or a callable that returns a tensor.
else_expression: either a tensor, or a callable that returns a tensor.
Returns:
The selected tensor.
Raises:
ValueError: If rank of `condition` is greater than rank of expressions."
4658,in_train_phase,tensorflow/tensorflow/python/keras/backend.py,4462,function,"Selects `x` in train phase, and `alt` otherwise.
Note that `alt` should have the *same shape* as `x`.
Arguments:
x: What to return in train phase
(tensor or callable that returns a tensor).
alt: What to return otherwise
(tensor or callable that returns a tensor).
training: Optional scalar tensor
(or Python boolean, or Python integer)
specifying the learning phase.
Returns:
Either `x` or `alt` based on the `training` flag.
the `training` flag defaults to `K.learning_phase()`."
4659,relu,tensorflow/tensorflow/python/keras/backend.py,4532,function,"Rectified linear unit.
With default values, it returns element-wise `max(x, 0)`.
Otherwise, it follows:
`f(x) = max_value` for `x >= max_value`,
`f(x) = x` for `threshold <= x < max_value`,
`f(x) = alpha * (x - threshold)` otherwise.
Arguments:
x: A tensor or variable.
alpha: A scalar, slope of negative section (default=`0.`).
max_value: float. Saturation threshold.
threshold: float. Threshold value for thresholded activation.
Returns:
A tensor."
4660,elu,tensorflow/tensorflow/python/keras/backend.py,4589,function,"Exponential linear unit.
Arguments:
x: A tensor or variable to compute the activation function for.
alpha: A scalar, slope of negative section.
Returns:
A tensor."
4661,softmax,tensorflow/tensorflow/python/keras/backend.py,4608,function,"Softmax of a tensor.
Arguments:
x: A tensor or variable.
axis: The dimension softmax would be performed on.
The default is -1 which indicates the last dimension.
Returns:
A tensor."
4662,softplus,tensorflow/tensorflow/python/keras/backend.py,4624,function,"Softplus of a tensor.
Arguments:
x: A tensor or variable.
Returns:
A tensor."
4663,softsign,tensorflow/tensorflow/python/keras/backend.py,4638,function,"Softsign of a tensor.
Arguments:
x: A tensor or variable.
Returns:
A tensor."
4664,categorical_crossentropy,tensorflow/tensorflow/python/keras/backend.py,4652,function,"Categorical crossentropy between an output tensor and a target tensor.
Arguments:
target: A tensor of the same shape as `output`.
output: A tensor resulting from a softmax
(unless `from_logits` is True, in which
case `output` is expected to be the logits).
from_logits: Boolean, whether `output` is the
result of a softmax, or is a tensor of logits.
axis: Int specifying the channels axis. `axis=-1` corresponds to data
format `channels_last', and `axis=1` corresponds to data format
`channels_first`.
Returns:
Output tensor.
Raises:
ValueError: if `axis` is neither -1 nor one of the axes of `output`.
Example:
>>> a = tf.constant([1., 0., 0., 0., 1., 0., 0., 0., 1.], shape=[3,3])
>>> print(a)
tf.Tensor(
[[1. 0. 0.]
[0. 1. 0.]
[0. 0. 1.]], shape=(3, 3), dtype=float32)
>>> b = tf.constant([.9, .05, .05, .05, .89, .06, .05, .01, .94], shape=[3,3])
>>> print(b)
tf.Tensor(
[[0.9 0.05 0.05]
[0.05 0.89 0.06]
[0.05 0.01 0.94]], shape=(3, 3), dtype=float32)
>>> loss = tf.keras.backend.categorical_crossentropy(a, b)
>>> print(np.around(loss, 5))
[0.10536 0.11653 0.06188]
>>> loss = tf.keras.backend.categorical_crossentropy(a, a)
>>> print(np.around(loss, 5))
[0. 0. 0.]"
4665,sparse_categorical_crossentropy,tensorflow/tensorflow/python/keras/backend.py,4723,function,"Categorical crossentropy with integer targets.
Arguments:
target: An integer tensor.
output: A tensor resulting from a softmax
(unless `from_logits` is True, in which
case `output` is expected to be the logits).
from_logits: Boolean, whether `output` is the
result of a softmax, or is a tensor of logits.
axis: Int specifying the channels axis. `axis=-1` corresponds to data
format `channels_last', and `axis=1` corresponds to data format
`channels_first`.
Returns:
Output tensor.
Raises:
ValueError: if `axis` is neither -1 nor one of the axes of `output`."
4666,binary_crossentropy,tensorflow/tensorflow/python/keras/backend.py,4807,function,"Binary crossentropy between an output tensor and a target tensor.
Arguments:
target: A tensor with the same shape as `output`.
output: A tensor.
from_logits: Whether `output` is expected to be a logits tensor.
By default, we consider that `output`
encodes a probability distribution.
Returns:
A tensor."
4667,sigmoid,tensorflow/tensorflow/python/keras/backend.py,4846,function,"Element-wise sigmoid.
Arguments:
x: A tensor or variable.
Returns:
A tensor."
4668,hard_sigmoid,tensorflow/tensorflow/python/keras/backend.py,4860,function,"Segment-wise linear approximation of sigmoid.
Faster than sigmoid.
Returns `0.` if `x < -2.5`, `1.` if `x > 2.5`.
In `-2.5 <= x <= 2.5`, returns `0.2 * x + 0.5`.
Arguments:
x: A tensor or variable.
Returns:
A tensor."
4669,tanh,tensorflow/tensorflow/python/keras/backend.py,4883,function,"Element-wise tanh.
Arguments:
x: A tensor or variable.
Returns:
A tensor."
4670,dropout,tensorflow/tensorflow/python/keras/backend.py,4897,function,"Sets entries in `x` to zero at random, while scaling the entire tensor.
Arguments:
x: tensor
level: fraction of the entries in the tensor
that will be set to 0.
noise_shape: shape for randomly generated keep/drop flags,
must be broadcastable to the shape of `x`
seed: random seed to ensure determinism.
Returns:
A tensor."
4671,l2_normalize,tensorflow/tensorflow/python/keras/backend.py,4918,function,"Normalizes a tensor wrt the L2 norm alongside the specified axis.
Arguments:
x: Tensor or variable.
axis: axis along which to perform normalization.
Returns:
A tensor."
4672,in_top_k,tensorflow/tensorflow/python/keras/backend.py,4933,function,"Returns whether the `targets` are in the top `k` `predictions`.
Arguments:
predictions: A tensor of shape `(batch_size, classes)` and type `float32`.
targets: A 1D tensor of length `batch_size` and type `int32` or `int64`.
k: An `int`, number of top elements to consider.
Returns:
A 1D tensor of length `batch_size` and type `bool`.
`output[i]` is `True` if `predictions[i, targets[i]]` is within top-`k`
values of `predictions[i]`."
4673,conv1d,tensorflow/tensorflow/python/keras/backend.py,5036,function,"1D convolution.
Arguments:
x: Tensor or variable.
kernel: kernel tensor.
strides: stride integer.
padding: string, `""same""`, `""causal""` or `""valid""`.
data_format: string, one of ""channels_last"", ""channels_first"".
dilation_rate: integer dilate rate.
Returns:
A tensor, result of 1D convolution.
Raises:
ValueError: if `data_format` is neither `channels_last` or
`channels_first`."
4674,conv2d,tensorflow/tensorflow/python/keras/backend.py,5087,function,"2D convolution.
Arguments:
x: Tensor or variable.
kernel: kernel tensor.
strides: strides tuple.
padding: string, `""same""` or `""valid""`.
data_format: `""channels_last""` or `""channels_first""`.
dilation_rate: tuple of 2 integers.
Returns:
A tensor, result of 2D convolution.
Raises:
ValueError: if `data_format` is neither `channels_last` or
`channels_first`."
4675,conv2d_transpose,tensorflow/tensorflow/python/keras/backend.py,5131,function,"2D deconvolution (i.e.
transposed convolution).
Arguments:
x: Tensor or variable.
kernel: kernel tensor.
output_shape: 1D int tensor for the output shape.
strides: strides tuple.
padding: string, `""same""` or `""valid""`.
data_format: string, `""channels_last""` or `""channels_first""`.
dilation_rate: Tuple of 2 integers.
Returns:
A tensor, result of transposed 2D convolution.
Raises:
ValueError: if `data_format` is neither `channels_last` or
`channels_first`."
4676,separable_conv1d,tensorflow/tensorflow/python/keras/backend.py,5203,function,"1D convolution with separable filters.
Arguments:
x: input tensor
depthwise_kernel: convolution kernel for the depthwise convolution.
pointwise_kernel: kernel for the 1x1 convolution.
strides: stride integer.
padding: string, `""same""` or `""valid""`.
data_format: string, `""channels_last""` or `""channels_first""`.
dilation_rate: integer dilation rate.
Returns:
Output tensor.
Raises:
ValueError: if `data_format` is neither `channels_last` or
`channels_first`."
4677,separable_conv2d,tensorflow/tensorflow/python/keras/backend.py,5272,function,"2D convolution with separable filters.
Arguments:
x: input tensor
depthwise_kernel: convolution kernel for the depthwise convolution.
pointwise_kernel: kernel for the 1x1 convolution.
strides: strides tuple (length 2).
padding: string, `""same""` or `""valid""`.
data_format: string, `""channels_last""` or `""channels_first""`.
dilation_rate: tuple of integers,
dilation rates for the separable convolution.
Returns:
Output tensor.
Raises:
ValueError: if `data_format` is neither `channels_last` or
`channels_first`.
ValueError: if `strides` is not a tuple of 2 integers."
4678,depthwise_conv2d,tensorflow/tensorflow/python/keras/backend.py,5330,function,"2D convolution with separable filters.
Arguments:
x: input tensor
depthwise_kernel: convolution kernel for the depthwise convolution.
strides: strides tuple (length 2).
padding: string, `""same""` or `""valid""`.
data_format: string, `""channels_last""` or `""channels_first""`.
dilation_rate: tuple of integers,
dilation rates for the separable convolution.
Returns:
Output tensor.
Raises:
ValueError: if `data_format` is neither `channels_last` or
`channels_first`."
4679,conv3d,tensorflow/tensorflow/python/keras/backend.py,5380,function,"3D convolution.
Arguments:
x: Tensor or variable.
kernel: kernel tensor.
strides: strides tuple.
padding: string, `""same""` or `""valid""`.
data_format: string, `""channels_last""` or `""channels_first""`.
dilation_rate: tuple of 3 integers.
Returns:
A tensor, result of 3D convolution.
Raises:
ValueError: if `data_format` is neither `channels_last` or
`channels_first`."
4680,conv3d_transpose,tensorflow/tensorflow/python/keras/backend.py,5422,function,"3D deconvolution (i.e.
transposed convolution).
Arguments:
x: input tensor.
kernel: kernel tensor.
output_shape: 1D int tensor for the output shape.
strides: strides tuple.
padding: string, ""same"" or ""valid"".
data_format: string, `""channels_last""` or `""channels_first""`.
Returns:
A tensor, result of transposed 3D convolution.
Raises:
ValueError: if `data_format` is neither `channels_last` or
`channels_first`."
4681,pool2d,tensorflow/tensorflow/python/keras/backend.py,5483,function,"2D Pooling.
Arguments:
x: Tensor or variable.
pool_size: tuple of 2 integers.
strides: tuple of 2 integers.
padding: string, `""same""` or `""valid""`.
data_format: string, `""channels_last""` or `""channels_first""`.
pool_mode: string, `""max""` or `""avg""`.
Returns:
A tensor, result of 2D pooling.
Raises:
ValueError: if `data_format` is neither `""channels_last""` or
`""channels_first""`.
ValueError: if `pool_size` is not a tuple of 2 integers.
ValueError: if `strides` is not a tuple of 2 integers.
ValueError: if `pool_mode` is neither `""max""` or `""avg""`."
4682,pool3d,tensorflow/tensorflow/python/keras/backend.py,5543,function,"3D Pooling.
Arguments:
x: Tensor or variable.
pool_size: tuple of 3 integers.
strides: tuple of 3 integers.
padding: string, `""same""` or `""valid""`.
data_format: string, `""channels_last""` or `""channels_first""`.
pool_mode: string, `""max""` or `""avg""`.
Returns:
A tensor, result of 3D pooling.
Raises:
ValueError: if `data_format` is neither `""channels_last""` or
`""channels_first""`.
ValueError: if `pool_mode` is neither `""max""` or `""avg""`."
4683,local_conv,tensorflow/tensorflow/python/keras/backend.py,5595,function,"Apply N-D convolution with un-shared weights.
Arguments:
inputs: (N+2)-D tensor with shape
(batch_size, channels_in, d_in1, ..., d_inN)
if data_format='channels_first', or
(batch_size, d_in1, ..., d_inN, channels_in)
if data_format='channels_last'.
kernel: the unshared weight for N-D convolution,
with shape (output_items, feature_dim, channels_out), where
feature_dim = np.prod(kernel_size) * channels_in,
output_items = np.prod(output_shape).
kernel_size: a tuple of N integers, specifying the
spatial dimensions of the N-D convolution window.
strides: a tuple of N integers, specifying the strides
of the convolution along the spatial dimensions.
output_shape: a tuple of (d_out1, ..., d_outN) specifying the spatial
dimensionality of the output.
data_format: string, ""channels_first"" or ""channels_last"".
Returns:
An (N+2)-D tensor with shape:
(batch_size, channels_out) + output_shape
if data_format='channels_first', or:
(batch_size,) + output_shape + (channels_out,)
if data_format='channels_last'.
Raises:
ValueError: if `data_format` is neither
`channels_last` nor `channels_first`."
4684,local_conv1d,tensorflow/tensorflow/python/keras/backend.py,5674,function,"Apply 1D conv with un-shared weights.
Arguments:
inputs: 3D tensor with shape:
(batch_size, steps, input_dim)
if data_format is ""channels_last"" or
(batch_size, input_dim, steps)
if data_format is ""channels_first"".
kernel: the unshared weight for convolution,
with shape (output_length, feature_dim, filters).
kernel_size: a tuple of a single integer,
specifying the length of the 1D convolution window.
strides: a tuple of a single integer,
specifying the stride length of the convolution.
data_format: the data format, channels_first or channels_last.
Returns:
A 3d tensor with shape:
(batch_size, output_length, filters)
if data_format='channels_first'
or 3D tensor with shape:
(batch_size, filters, output_length)
if data_format='channels_last'."
4685,local_conv2d,tensorflow/tensorflow/python/keras/backend.py,5710,function,"Apply 2D conv with un-shared weights.
Arguments:
inputs: 4D tensor with shape:
(batch_size, filters, new_rows, new_cols)
if data_format='channels_first'
or 4D tensor with shape:
(batch_size, new_rows, new_cols, filters)
if data_format='channels_last'.
kernel: the unshared weight for convolution,
with shape (output_items, feature_dim, filters).
kernel_size: a tuple of 2 integers, specifying the
width and height of the 2D convolution window.
strides: a tuple of 2 integers, specifying the strides
of the convolution along the width and height.
output_shape: a tuple with (output_row, output_col).
data_format: the data format, channels_first or channels_last.
Returns:
A 4D tensor with shape:
(batch_size, filters, new_rows, new_cols)
if data_format='channels_first'
or 4D tensor with shape:
(batch_size, new_rows, new_cols, filters)
if data_format='channels_last'."
4686,bias_add,tensorflow/tensorflow/python/keras/backend.py,5752,function,"Adds a bias vector to a tensor.
Arguments:
x: Tensor or variable.
bias: Bias tensor to add.
data_format: string, `""channels_last""` or `""channels_first""`.
Returns:
Output tensor.
Raises:
ValueError: In one of the two cases below:
1. invalid `data_format` argument.
2. invalid bias shape.
the bias should be either a vector or
a tensor with ndim(x) - 1 dimension"
4687,random_normal,tensorflow/tensorflow/python/keras/backend.py,5797,function,"Returns a tensor with normal distribution of values.
It is an alias to `tf.random.normal`.
Arguments:
shape: A tuple of integers, the shape of tensor to create.
mean: A float, the mean value of the normal distribution to draw samples.
Default to 0.0.
stddev: A float, the standard deviation of the normal distribution
to draw samples. Default to 1.0.
dtype: `tf.dtypes.DType`, dtype of returned tensor. Default to use Keras
backend dtype which is float32.
seed: Integer, random seed. Will use a random numpy integer when not
specified.
Returns:
A tensor with normal distribution of values.
Example:
>>> random_normal_tensor = tf.keras.backend.random_normal(shape=(2,3),
... mean=0.0, stddev=1.0)
>>> random_normal_tensor
<tf.Tensor: shape=(2, 3), dtype=float32, numpy=...,
dtype=float32)>"
4688,random_uniform,tensorflow/tensorflow/python/keras/backend.py,5834,function,"Returns a tensor with uniform distribution of values.
Arguments:
shape: A tuple of integers, the shape of tensor to create.
minval: A float, lower boundary of the uniform distribution
to draw samples.
maxval: A float, upper boundary of the uniform distribution
to draw samples.
dtype: String, dtype of returned tensor.
seed: Integer, random seed.
Returns:
A tensor.
Example:
>>> random_uniform_tensor = tf.keras.backend.random_uniform(shape=(2,3),
... minval=0.0, maxval=1.0)
>>> random_uniform_tensor
<tf.Tensor: shape=(2, 3), dtype=float32, numpy=...,
dtype=float32)>"
4689,random_binomial,tensorflow/tensorflow/python/keras/backend.py,5868,function,"Returns a tensor with random binomial distribution of values.
DEPRECATED, use `tf.keras.backend.random_bernoulli` instead.
The binomial distribution with parameters `n` and `p` is the probability
distribution of the number of successful Bernoulli process. Only supports
`n` = 1 for now.
Arguments:
shape: A tuple of integers, the shape of tensor to create.
p: A float, `0. <= p <= 1`, probability of binomial distribution.
dtype: String, dtype of returned tensor.
seed: Integer, random seed.
Returns:
A tensor.
Example:
>>> random_binomial_tensor = tf.keras.backend.random_binomial(shape=(2,3),
... p=0.5)
>>> random_binomial_tensor
<tf.Tensor: shape=(2, 3), dtype=float32, numpy=...,
dtype=float32)>"
4690,random_bernoulli,tensorflow/tensorflow/python/keras/backend.py,5905,function,"Returns a tensor with random bernoulli distribution of values.
Arguments:
shape: A tuple of integers, the shape of tensor to create.
p: A float, `0. <= p <= 1`, probability of bernoulli distribution.
dtype: String, dtype of returned tensor.
seed: Integer, random seed.
Returns:
A tensor."
4691,truncated_normal,tensorflow/tensorflow/python/keras/backend.py,5922,function,"Returns a tensor with truncated random normal distribution of values.
The generated values follow a normal distribution
with specified mean and standard deviation,
except that values whose magnitude is more than
two standard deviations from the mean are dropped and re-picked.
Arguments:
shape: A tuple of integers, the shape of tensor to create.
mean: Mean of the values.
stddev: Standard deviation of the values.
dtype: String, dtype of returned tensor.
seed: Integer, random seed.
Returns:
A tensor."
4692,ctc_label_dense_to_sparse,tensorflow/tensorflow/python/keras/backend.py,5957,function,"Converts CTC labels from dense to sparse.
Arguments:
labels: dense CTC labels.
label_lengths: length of the labels.
Returns:
A sparse tensor representation of the labels."
4693,ctc_batch_cost,tensorflow/tensorflow/python/keras/backend.py,6004,function,"Runs CTC loss algorithm on each batch element.
Arguments:
y_true: tensor `(samples, max_string_length)`
containing the truth labels.
y_pred: tensor `(samples, time_steps, num_categories)`
containing the prediction, or output of the softmax.
input_length: tensor `(samples, 1)` containing the sequence length for
each batch item in `y_pred`.
label_length: tensor `(samples, 1)` containing the sequence length for
each batch item in `y_true`.
Returns:
Tensor with shape (samples,1) containing the
CTC loss of each element."
4694,ctc_decode,tensorflow/tensorflow/python/keras/backend.py,6037,function,"Decodes the output of a softmax.
Can use either greedy search (also known as best path)
or a constrained dictionary search.
Arguments:
y_pred: tensor `(samples, time_steps, num_categories)`
containing the prediction, or output of the softmax.
input_length: tensor `(samples, )` containing the sequence length for
each batch item in `y_pred`.
greedy: perform much faster best-path search if `true`.
This does not use a dictionary.
beam_width: if `greedy` is `false`: a beam search decoder will be used
with a beam of this width.
top_paths: if `greedy` is `false`,
how many of the most probable paths will be returned.
Returns:
Tuple:
List: if `greedy` is `true`, returns a list of one element that
contains the decoded sequence.
If `false`, returns the `top_paths` most probable
decoded sequences.
Each decoded sequence has shape (samples, time_steps).
Important: blank labels are returned as `-1`.
Tensor `(top_paths, )` that contains
the log probability of each decoded sequence."
4695,map_fn,tensorflow/tensorflow/python/keras/backend.py,6093,function,"Map the function fn over the elements elems and return the outputs.
Arguments:
fn: Callable that will be called upon each element in elems
elems: tensor
name: A string name for the map node in the graph
dtype: Output data type.
Returns:
Tensor with dtype `dtype`."
4696,foldl,tensorflow/tensorflow/python/keras/backend.py,6109,function,"Reduce elems using fn to combine them from left to right.
Arguments:
fn: Callable that will be called upon each element in elems and an
accumulator, for instance `lambda acc, x: acc + x`
elems: tensor
initializer: The first value used (`elems[0]` in case of None)
name: A string name for the foldl node in the graph
Returns:
Tensor with same type and shape as `initializer`."
4697,foldr,tensorflow/tensorflow/python/keras/backend.py,6126,function,"Reduce elems using fn to combine them from right to left.
Arguments:
fn: Callable that will be called upon each element in elems and an
accumulator, for instance `lambda acc, x: acc + x`
elems: tensor
initializer: The first value used (`elems[-1]` in case of None)
name: A string name for the foldr node in the graph
Returns:
Same type and shape as initializer"
4698,configure_and_create_distributed_session,tensorflow/tensorflow/python/keras/backend.py,6190,function,Configure session config and create a session with it.
4699,is_tpu_strategy,tensorflow/tensorflow/python/keras/backend.py,6235,function,We're executing TPU Strategy.
4700,cast_variables_to_tensor,tensorflow/tensorflow/python/keras/backend.py,6241,function,
4701,convert_inputs_if_ragged,tensorflow/tensorflow/python/keras/backend.py,6255,function,Converts any ragged tensors to dense.
4702,maybe_convert_to_ragged,tensorflow/tensorflow/python/keras/backend.py,6278,function,Converts any ragged input back to its initial structure.
4703,ContextValueCache,tensorflow/tensorflow/python/keras/backend.py,6286,class,"Container that caches (possibly tensor) values based on the context.
This class is similar to defaultdict, where values may be produced by the
default factory specified during initialization. This class also has a default
value for the key (when key is `None`) -- the key is set to the the current
graph or eager context. The default factories for key and value are only used
in `__getitem__` and `setdefault`. The `.get()` behavior remains the same.
This object will return the value of the current graph or closest parent graph
if the current graph is a function. This is to reflect the fact that if a
tensor is created in eager/graph, child functions may capture that tensor.
The default factory method may accept keyword arguments (unlike defaultdict,
which only accepts callables with 0 arguments). To pass keyword arguments to
`default_factory`, use the `setdefault` method instead of `__getitem__`.
An example of how this class can be used in different contexts:
```
cache = ContextValueCache(int)
# Eager mode
cache[None] += 2
cache[None] += 4
assert cache[None] == 6
# Graph mode
with tf.Graph().as_default() as g:
cache[None] += 5
cache[g] += 3
assert cache[g] == 8
```
Example of a default factory with arguments:
```
cache = ContextValueCache(lambda x: x + 1)
g = tf.get_default_graph()
# Example with keyword argument.
value = cache.setdefault(key=g, kwargs={'x': 3})
assert cache[g] == 4
```"
4704,setdefault,tensorflow/tensorflow/python/keras/backend.py,6386,method,"Sets the default value if key is not in dict, and returns the value."
4705,epsilon,tensorflow/tensorflow/python/keras/backend_config.py,35,function,"Returns the value of the fuzz factor used in numeric expressions.
Returns:
A float.
Example:
>>> tf.keras.backend.epsilon()
1e-07"
4706,set_epsilon,tensorflow/tensorflow/python/keras/backend_config.py,49,function,"Sets the value of the fuzz factor used in numeric expressions.
Arguments:
value: float. New value of epsilon.
Example:
>>> tf.keras.backend.epsilon()
1e-07
>>> tf.keras.backend.set_epsilon(1e-5)
>>> tf.keras.backend.epsilon()
1e-05
>>> tf.keras.backend.set_epsilon(1e-7)"
4707,floatx,tensorflow/tensorflow/python/keras/backend_config.py,68,function,"Returns the default float type, as a string.
E.g. `'float16'`, `'float32'`, `'float64'`.
Returns:
String, the current default float type.
Example:
>>> tf.keras.backend.floatx()
'float32'"
4708,set_floatx,tensorflow/tensorflow/python/keras/backend_config.py,84,function,"Sets the default float type.
Note: It is not recommended to set this to float16 for training, as this will
likely cause numeric stability issues. Instead, mixed precision, which is
using a mix of float16 and float32, can be used by calling
`tf.keras.mixed_precision.experimental.set_policy('mixed_float16')`. See the
[mixed precision guide](
https://www.tensorflow.org/guide/keras/mixed_precision) for details.
Arguments:
value: String; `'float16'`, `'float32'`, or `'float64'`.
Example:
>>> tf.keras.backend.floatx()
'float32'
>>> tf.keras.backend.set_floatx('float64')
>>> tf.keras.backend.floatx()
'float64'
>>> tf.keras.backend.set_floatx('float32')
Raises:
ValueError: In case of invalid value."
4709,image_data_format,tensorflow/tensorflow/python/keras/backend_config.py,116,function,"Returns the default image data format convention.
Returns:
A string, either `'channels_first'` or `'channels_last'`
Example:
>>> tf.keras.backend.image_data_format()
'channels_last'"
4710,set_image_data_format,tensorflow/tensorflow/python/keras/backend_config.py,130,function,"Sets the value of the image data format convention.
Arguments:
data_format: string. `'channels_first'` or `'channels_last'`.
Example:
>>> tf.keras.backend.image_data_format()
'channels_last'
>>> tf.keras.backend.set_image_data_format('channels_first')
>>> tf.keras.backend.image_data_format()
'channels_first'
>>> tf.keras.backend.set_image_data_format('channels_last')
Raises:
ValueError: In case of invalid `data_format` value."
4711,compare_single_input_op_to_numpy,tensorflow/tensorflow/python/keras/backend_test.py,46,function,
4712,compare_two_inputs_op_to_numpy,tensorflow/tensorflow/python/keras/backend_test.py,74,function,
4713,configure_callbacks,tensorflow/tensorflow/python/keras/callbacks.py,71,function,"Configures callbacks for use in various training loops.
Arguments:
callbacks: List of Callbacks.
model: Model being trained.
do_validation: Whether or not validation loop will be run.
batch_size: Number of samples per batch.
epochs: Number of epoch to train.
steps_per_epoch: Number of batches to run per training epoch.
samples: Number of training samples.
verbose: int, 0 or 1. Keras logging verbosity to pass to ProgbarLogger.
count_mode: One of 'steps' or 'samples'. Per-batch or per-sample count.
mode: String. One of ModeKeys.TRAIN, ModeKeys.TEST, or ModeKeys.PREDICT.
Which loop mode to configure callbacks for.
Returns:
Instance of CallbackList used to control all Callbacks."
4714,set_callback_parameters,tensorflow/tensorflow/python/keras/callbacks.py,133,function,"Sets callback parameters.
Arguments:
callback_list: CallbackList instance.
model: Model being trained.
do_validation: Whether or not validation loop will be run.
batch_size: Number of samples per batch.
epochs: Number of epoch to train.
steps_per_epoch: Number of batches to run per training epoch.
samples: Number of training samples.
verbose: int, 0 or 1. Keras logging verbosity to pass to ProgbarLogger.
mode: String. One of ModeKeys.TRAIN, ModeKeys.TEST, or ModeKeys.PREDICT.
Which loop mode to configure callbacks for."
4715,make_logs,tensorflow/tensorflow/python/keras/callbacks.py,187,function,Computes logs for sending to `on_batch_end` methods.
4716,CallbackList,tensorflow/tensorflow/python/keras/callbacks.py,199,class,Container abstracting a list of callbacks.
4717,append,tensorflow/tensorflow/python/keras/callbacks.py,266,method,
4718,set_params,tensorflow/tensorflow/python/keras/callbacks.py,269,method,
4719,set_model,tensorflow/tensorflow/python/keras/callbacks.py,274,method,
4720,on_batch_begin,tensorflow/tensorflow/python/keras/callbacks.py,369,method,
4721,on_batch_end,tensorflow/tensorflow/python/keras/callbacks.py,373,method,
4722,on_epoch_begin,tensorflow/tensorflow/python/keras/callbacks.py,377,method,"Calls the `on_epoch_begin` methods of its callbacks.
This function should only be called during TRAIN mode.
Arguments:
epoch: Integer, index of epoch.
logs: Dict. Currently no data is passed to this argument for this method
but that may change in the future."
4723,on_epoch_end,tensorflow/tensorflow/python/keras/callbacks.py,397,method,"Calls the `on_epoch_end` methods of its callbacks.
This function should only be called during TRAIN mode.
Arguments:
epoch: Integer, index of epoch.
logs: Dict, metric results for this training epoch, and for the
validation epoch if validation is performed. Validation result keys
are prefixed with `val_`."
4724,on_train_batch_begin,tensorflow/tensorflow/python/keras/callbacks.py,418,method,"Calls the `on_train_batch_begin` methods of its callbacks.
Arguments:
batch: Integer, index of batch within the current epoch.
logs: Dict, contains the return value of `model.train_step`. Typically,
the values of the `Model`'s metrics are returned. Example:
`{'loss': 0.2, 'accuracy': 0.7}`."
4725,on_train_batch_end,tensorflow/tensorflow/python/keras/callbacks.py,430,method,"Calls the `on_train_batch_end` methods of its callbacks.
Arguments:
batch: Integer, index of batch within the current epoch.
logs: Dict. Aggregated metric results up until this batch."
4726,on_predict_batch_begin,tensorflow/tensorflow/python/keras/callbacks.py,462,method,"Calls the `on_predict_batch_begin` methods of its callbacks.
Arguments:
batch: Integer, index of batch within the current epoch.
logs: Dict, contains the return value of `model.predict_step`,
it typically returns a dict with a key 'outputs' containing
the model's outputs."
4727,on_predict_batch_end,tensorflow/tensorflow/python/keras/callbacks.py,474,method,"Calls the `on_predict_batch_end` methods of its callbacks.
Arguments:
batch: Integer, index of batch within the current epoch.
logs: Dict. Aggregated metric results up until this batch."
4728,on_train_begin,tensorflow/tensorflow/python/keras/callbacks.py,484,method,"Calls the `on_train_begin` methods of its callbacks.
Arguments:
logs: Dict. Currently no data is passed to this argument for this method
but that may change in the future."
4729,on_train_end,tensorflow/tensorflow/python/keras/callbacks.py,501,method,"Calls the `on_train_end` methods of its callbacks.
Arguments:
logs: Dict. Currently no data is passed to this argument for this method
but that may change in the future."
4730,on_predict_begin,tensorflow/tensorflow/python/keras/callbacks.py,552,method,"Calls the 'on_predict_begin` methods of its callbacks.
Arguments:
logs: Dict. Currently no data is passed to this argument for this method
but that may change in the future."
4731,on_predict_end,tensorflow/tensorflow/python/keras/callbacks.py,569,method,"Calls the `on_predict_end` methods of its callbacks.
Arguments:
logs: Dict. Currently no data is passed to this argument for this method
but that may change in the future."
4732,Callback,tensorflow/tensorflow/python/keras/callbacks.py,591,class,"Abstract base class used to build new callbacks.
Attributes:
params: Dict. Training parameters
(eg. verbosity, batch size, number of epochs...).
model: Instance of `keras.models.Model`.
Reference of the model being trained.
The `logs` dictionary that callback methods
take as argument will contain keys for quantities relevant to
the current batch or epoch (see method-specific docstrings)."
4733,set_params,tensorflow/tensorflow/python/keras/callbacks.py,614,method,
4734,set_model,tensorflow/tensorflow/python/keras/callbacks.py,617,method,
4735,on_batch_begin,tensorflow/tensorflow/python/keras/callbacks.py,622,method,A backwards compatibility alias for `on_train_batch_begin`.
4736,on_batch_end,tensorflow/tensorflow/python/keras/callbacks.py,627,method,A backwards compatibility alias for `on_train_batch_end`.
4737,on_epoch_begin,tensorflow/tensorflow/python/keras/callbacks.py,631,method,"Called at the start of an epoch.
Subclasses should override for any actions to run. This function should only
be called during TRAIN mode.
Arguments:
epoch: Integer, index of epoch.
logs: Dict. Currently no data is passed to this argument for this method
but that may change in the future."
4738,on_epoch_end,tensorflow/tensorflow/python/keras/callbacks.py,644,method,"Called at the end of an epoch.
Subclasses should override for any actions to run. This function should only
be called during TRAIN mode.
Arguments:
epoch: Integer, index of epoch.
logs: Dict, metric results for this training epoch, and for the
validation epoch if validation is performed. Validation result keys
are prefixed with `val_`."
4739,on_train_batch_begin,tensorflow/tensorflow/python/keras/callbacks.py,659,method,"Called at the beginning of a training batch in `fit` methods.
Subclasses should override for any actions to run.
Arguments:
batch: Integer, index of batch within the current epoch.
logs: Dict, contains the return value of `model.train_step`. Typically,
the values of the `Model`'s metrics are returned. Example:
`{'loss': 0.2, 'accuracy': 0.7}`."
4740,on_train_batch_end,tensorflow/tensorflow/python/keras/callbacks.py,675,method,"Called at the end of a training batch in `fit` methods.
Subclasses should override for any actions to run.
Arguments:
batch: Integer, index of batch within the current epoch.
logs: Dict. Aggregated metric results up until this batch."
4741,on_predict_batch_begin,tensorflow/tensorflow/python/keras/callbacks.py,721,method,"Called at the beginning of a batch in `predict` methods.
Subclasses should override for any actions to run.
Arguments:
batch: Integer, index of batch within the current epoch.
logs: Dict, contains the return value of `model.predict_step`,
it typically returns a dict with a key 'outputs' containing
the model's outputs."
4742,on_predict_batch_end,tensorflow/tensorflow/python/keras/callbacks.py,735,method,"Called at the end of a batch in `predict` methods.
Subclasses should override for any actions to run.
Arguments:
batch: Integer, index of batch within the current epoch.
logs: Dict. Aggregated metric results up until this batch."
4743,on_train_begin,tensorflow/tensorflow/python/keras/callbacks.py,746,method,"Called at the beginning of training.
Subclasses should override for any actions to run.
Arguments:
logs: Dict. Currently no data is passed to this argument for this method
but that may change in the future."
4744,on_train_end,tensorflow/tensorflow/python/keras/callbacks.py,757,method,"Called at the end of training.
Subclasses should override for any actions to run.
Arguments:
logs: Dict. Currently the output of the last call to `on_epoch_end()`
is passed to this argument for this method but that may change in
the future."
4745,on_predict_begin,tensorflow/tensorflow/python/keras/callbacks.py,792,method,"Called at the beginning of prediction.
Subclasses should override for any actions to run.
Arguments:
logs: Dict. Currently no data is passed to this argument for this method
but that may change in the future."
4746,on_predict_end,tensorflow/tensorflow/python/keras/callbacks.py,803,method,"Called at the end of prediction.
Subclasses should override for any actions to run.
Arguments:
logs: Dict. Currently no data is passed to this argument for this method
but that may change in the future."
4747,BaseLogger,tensorflow/tensorflow/python/keras/callbacks.py,832,class,"Callback that accumulates epoch averages of metrics.
This callback is automatically applied to every Keras model.
Arguments:
stateful_metrics: Iterable of string names of metrics that
should *not* be averaged over an epoch.
Metrics in this list will be logged as-is in `on_epoch_end`.
All others will be averaged in `on_epoch_end`."
4748,on_epoch_begin,tensorflow/tensorflow/python/keras/callbacks.py,848,method,
4749,on_batch_end,tensorflow/tensorflow/python/keras/callbacks.py,852,method,
4750,on_epoch_end,tensorflow/tensorflow/python/keras/callbacks.py,869,method,
4751,TerminateOnNaN,tensorflow/tensorflow/python/keras/callbacks.py,881,class,"Callback that terminates training when a NaN loss is encountered.
"
4752,on_batch_end,tensorflow/tensorflow/python/keras/callbacks.py,885,method,
4753,ProgbarLogger,tensorflow/tensorflow/python/keras/callbacks.py,895,class,"Callback that prints metrics to stdout.
Arguments:
count_mode: One of `""steps""` or `""samples""`.
Whether the progress bar should
count samples seen or steps (batches) seen.
stateful_metrics: Iterable of string names of metrics that
should *not* be averaged over an epoch.
Metrics in this list will be logged as-is.
All others will be averaged over time (e.g. loss, etc).
If not provided, defaults to the `Model`'s metrics.
Raises:
ValueError: In case of invalid `count_mode`."
4754,set_params,tensorflow/tensorflow/python/keras/callbacks.py,935,method,
4755,on_train_begin,tensorflow/tensorflow/python/keras/callbacks.py,954,method,
4756,on_predict_begin,tensorflow/tensorflow/python/keras/callbacks.py,963,method,
4757,on_epoch_begin,tensorflow/tensorflow/python/keras/callbacks.py,967,method,
4758,on_train_batch_end,tensorflow/tensorflow/python/keras/callbacks.py,973,method,
4759,on_predict_batch_end,tensorflow/tensorflow/python/keras/callbacks.py,980,method,
4760,on_epoch_end,tensorflow/tensorflow/python/keras/callbacks.py,984,method,
4761,on_predict_end,tensorflow/tensorflow/python/keras/callbacks.py,991,method,
4762,History,tensorflow/tensorflow/python/keras/callbacks.py,1054,class,"Callback that records events into a `History` object.
This callback is automatically applied to
every Keras model. The `History` object
gets returned by the `fit` method of models."
4763,on_train_begin,tensorflow/tensorflow/python/keras/callbacks.py,1066,method,
4764,on_epoch_end,tensorflow/tensorflow/python/keras/callbacks.py,1069,method,
4765,ModelCheckpoint,tensorflow/tensorflow/python/keras/callbacks.py,1081,class,"Callback to save the Keras model or model weights at some frequency.
`ModelCheckpoint` callback is used in conjunction with training using
`model.fit()` to save a model or weights (in a checkpoint file) at some
interval, so the model or weights can be loaded later to continue the training
from the state saved.
A few options this callback provides include:
- Whether to only keep the model that has achieved the ""best performance"" so
far, or whether to save the model at the end of every epoch regardless of
performance.
- Definition of 'best'; which quantity to monitor and whether it should be
maximized or minimized.
- The frequency it should save at. Currently, the callback supports saving at
the end of every epoch, or after a fixed number of training batches.
- Whether only weights are saved, or the whole model is saved.
Example:
```python
EPOCHS = 10
checkpoint_filepath = '/tmp/checkpoint'
model_checkpoint_callback = tf.keras.callbacks.ModelCheckpoint(
filepath=checkpoint_filepath,
save_weights_only=True,
monitor='val_acc',
mode='max',
save_best_only=True)
# Model weights are saved at the end of every epoch, if it's the best seen
# so far.
model.fit(epochs=EPOCHS, callbacks=[model_checkpoint_callback])
# The model weights (that are considered the best) are loaded into the model.
model.load_weights(checkpoint_filepath)
```
Arguments:
filepath: string or `PathLike`, path to save the model file. `filepath`
can contain named formatting options, which will be filled the value of
`epoch` and keys in `logs` (passed in `on_epoch_end`). For example: if
`filepath` is `weights.{epoch:02d}-{val_loss:.2f}.hdf5`, then the model
checkpoints will be saved with the epoch number and the validation loss
in the filename.
monitor: quantity to monitor.
verbose: verbosity mode, 0 or 1.
save_best_only: if `save_best_only=True`, the latest best model according
to the quantity monitored will not be overwritten.
If `filepath` doesn't contain formatting options like `{epoch}` then
`filepath` will be overwritten by each new better model.
mode: one of {auto, min, max}. If `save_best_only=True`, the decision to
overwrite the current save file is made based on either the maximization
or the minimization of the monitored quantity. For `val_acc`, this
should be `max`, for `val_loss` this should be `min`, etc. In `auto`
mode, the direction is automatically inferred from the name of the
monitored quantity.
save_weights_only: if True, then only the model's weights will be saved
(`model.save_weights(filepath)`), else the full model is saved
(`model.save(filepath)`).
save_freq: `'epoch'` or integer. When using `'epoch'`, the callback saves
the model after each epoch. When using integer, the callback saves the
model at end of this many batches. If the `Model` is compiled with
`experimental_steps_per_execution=N`, then the saving criteria will be
checked every Nth batch. Note that if the saving isn't aligned to
epochs, the monitored metric may potentially be less reliable (it
could reflect as little as 1 batch, since the metrics get reset every
epoch). Defaults to `'epoch'`.
options: Optional `tf.train.CheckpointOptions` object if
`save_weights_only` is true or optional `tf.saved_model.SavedOptions`
object if `save_weights_only` is false.
**kwargs: Additional arguments for backwards compatibility. Possible key
is `period`."
4766,set_model,tensorflow/tensorflow/python/keras/callbacks.py,1239,method,
4767,on_train_begin,tensorflow/tensorflow/python/keras/callbacks.py,1247,method,
4768,on_train_batch_end,tensorflow/tensorflow/python/keras/callbacks.py,1276,method,
4769,on_epoch_begin,tensorflow/tensorflow/python/keras/callbacks.py,1280,method,
4770,on_epoch_end,tensorflow/tensorflow/python/keras/callbacks.py,1283,method,
4771,BackupAndRestore,tensorflow/tensorflow/python/keras/callbacks.py,1484,class,"Callback to back up and restore the training state.
`BackupAndRestore` callback is intended to recover from interruptions that
happened in the middle of a model.fit execution by backing up the
training states in a temporary checkpoint file (based on TF CheckpointManager)
at the end of each epoch. If training restarted before completion, the
training state and model are restored to the most recently saved state at the
beginning of a new model.fit() run.
Note that user is responsible to bring jobs back up.
This callback is important for the backup and restore mechanism for fault
tolerance purpose. And the model to be restored from an previous checkpoint is
expected to be the same as the one used to back up. If user changes arguments
passed to compile or fit, the checkpoint saved for fault tolerance can become
invalid.
Note:
1. This callback is not compatible with disabling eager execution.
2. A checkpoint is saved at the end of each epoch, when restoring we'll redo
any partial work from an unfinished epoch in which the training got restarted
(so the work done before a interruption doesn't affect the final model state).
3. This works for both single worker and multi-worker mode, only
MirroredStrategy and MultiWorkerMirroredStrategy are supported for now.
Example:
>>> class InterruptingCallback(tf.keras.callbacks.Callback):
... def on_epoch_begin(self, epoch, logs=None):
... if epoch == 4:
... raise RuntimeError('Interrupting!')
>>> callback = tf.keras.callbacks.experimental.BackupAndRestore(
... backup_dir=""/tmp"")
>>> model = tf.keras.models.Sequential([tf.keras.layers.Dense(10)])
>>> model.compile(tf.keras.optimizers.SGD(), loss='mse')
>>> try:
... model.fit(np.arange(100).reshape(5, 20), np.zeros(5), epochs=10,
... batch_size=1, callbacks=[callback, InterruptingCallback()],
... verbose=0)
... except:
... pass
>>> history = model.fit(np.arange(100).reshape(5, 20), np.zeros(5), epochs=10,
... batch_size=1, callbacks=[callback], verbose=0)
>>> # Only 6 more epochs are run, since first trainning got interrupted at
>>> # zero-indexed epoch 4, second training will continue from 4 to 9.
>>> len(history.history['loss'])
6
Arguments:
backup_dir: String, path to save the model file. This is the directory in
which the system stores temporary files to recover the model from jobs
terminated unexpectedly. The directory cannot be reused elsewhere to
store other checkpoints, e.g. by BackupAndRestore callback of another
training, or by another callback (ModelCheckpoint) of the same training."
4772,set_model,tensorflow/tensorflow/python/keras/callbacks.py,1563,method,
4773,on_train_begin,tensorflow/tensorflow/python/keras/callbacks.py,1566,method,
4774,on_train_end,tensorflow/tensorflow/python/keras/callbacks.py,1581,method,
4775,on_epoch_end,tensorflow/tensorflow/python/keras/callbacks.py,1591,method,
4776,EarlyStopping,tensorflow/tensorflow/python/keras/callbacks.py,1597,class,"Stop training when a monitored metric has stopped improving.
Assuming the goal of a training is to minimize the loss. With this, the
metric to be monitored would be `'loss'`, and mode would be `'min'`. A
`model.fit()` training loop will check at end of every epoch whether
the loss is no longer decreasing, considering the `min_delta` and
`patience` if applicable. Once it's found no longer decreasing,
`model.stop_training` is marked True and the training terminates.
The quantity to be monitored needs to be available in `logs` dict.
To make it so, pass the loss or metrics at `model.compile()`.
Arguments:
monitor: Quantity to be monitored.
min_delta: Minimum change in the monitored quantity
to qualify as an improvement, i.e. an absolute
change of less than min_delta, will count as no
improvement.
patience: Number of epochs with no improvement
after which training will be stopped.
verbose: verbosity mode.
mode: One of `{""auto"", ""min"", ""max""}`. In `min` mode,
training will stop when the quantity
monitored has stopped decreasing; in `""max""`
mode it will stop when the quantity
monitored has stopped increasing; in `""auto""`
mode, the direction is automatically inferred
from the name of the monitored quantity.
baseline: Baseline value for the monitored quantity.
Training will stop if the model doesn't show improvement over the
baseline.
restore_best_weights: Whether to restore model weights from
the epoch with the best value of the monitored quantity.
If False, the model weights obtained at the last step of
training are used.
Example:
>>> callback = tf.keras.callbacks.EarlyStopping(monitor='loss', patience=3)
>>> # This callback will stop the training when there is no improvement in
>>> # the validation loss for three consecutive epochs.
>>> model = tf.keras.models.Sequential([tf.keras.layers.Dense(10)])
>>> model.compile(tf.keras.optimizers.SGD(), loss='mse')
>>> history = model.fit(np.arange(100).reshape(5, 20), np.zeros(5),
... epochs=10, batch_size=1, callbacks=[callback],
... verbose=0)
>>> len(history.history['loss']) # Only 4 epochs are run.
4"
4777,on_train_begin,tensorflow/tensorflow/python/keras/callbacks.py,1688,method,
4778,on_epoch_end,tensorflow/tensorflow/python/keras/callbacks.py,1698,method,
4779,on_train_end,tensorflow/tensorflow/python/keras/callbacks.py,1717,method,
4780,get_monitor_value,tensorflow/tensorflow/python/keras/callbacks.py,1721,method,
4781,RemoteMonitor,tensorflow/tensorflow/python/keras/callbacks.py,1732,class,"Callback used to stream events to a server.
Requires the `requests` library.
Events are sent to `root + '/publish/epoch/end/'` by default. Calls are
HTTP POST, with a `data` argument which is a
JSON-encoded dictionary of event data.
If `send_as_json=True`, the content type of the request will be
`""application/json""`.
Otherwise the serialized JSON will be sent within a form.
Arguments:
root: String; root url of the target server.
path: String; path relative to `root` to which the events will be sent.
field: String; JSON field under which the data will be stored.
The field is used only if the payload is sent within a form
(i.e. send_as_json is set to False).
headers: Dictionary; optional custom HTTP headers.
send_as_json: Boolean; whether the request should be
sent as `""application/json""`."
4782,on_epoch_end,tensorflow/tensorflow/python/keras/callbacks.py,1768,method,
4783,LearningRateScheduler,tensorflow/tensorflow/python/keras/callbacks.py,1795,class,"Learning rate scheduler.
At the beginning of every epoch, this callback gets the updated learning rate
value from `schedule` function provided at `__init__`, with the current epoch
and current learning rate, and applies the updated learning rate
on the optimizer.
Arguments:
schedule: a function that takes an epoch index (integer, indexed from 0)
and current learning rate (float) as inputs and returns a new
learning rate as output (float).
verbose: int. 0: quiet, 1: update messages.
Example:
>>> # This function keeps the initial learning rate for the first ten epochs
>>> # and decreases it exponentially after that.
>>> def scheduler(epoch, lr):
... if epoch < 10:
... return lr
... else:
... return lr * tf.math.exp(-0.1)
>>>
>>> model = tf.keras.models.Sequential([tf.keras.layers.Dense(10)])
>>> model.compile(tf.keras.optimizers.SGD(), loss='mse')
>>> round(model.optimizer.lr.numpy(), 5)
0.01
>>> callback = tf.keras.callbacks.LearningRateScheduler(scheduler)
>>> history = model.fit(np.arange(100).reshape(5, 20), np.zeros(5),
... epochs=15, callbacks=[callback], verbose=0)
>>> round(model.optimizer.lr.numpy(), 5)
0.00607"
4784,on_epoch_begin,tensorflow/tensorflow/python/keras/callbacks.py,1837,method,
4785,on_epoch_end,tensorflow/tensorflow/python/keras/callbacks.py,1855,method,
4786,TensorBoard,tensorflow/tensorflow/python/keras/callbacks.py,1861,class,"Enable visualizations for TensorBoard.
TensorBoard is a visualization tool provided with TensorFlow.
This callback logs events for TensorBoard, including:
* Metrics summary plots
* Training graph visualization
* Activation histograms
* Sampled profiling
If you have installed TensorFlow with pip, you should be able
to launch TensorBoard from the command line:
```
tensorboard --logdir=path_to_your_logs
```
You can find more information about TensorBoard
[here](https://www.tensorflow.org/get_started/summaries_and_tensorboard).
Example (Basic):
```python
tensorboard_callback = tf.keras.callbacks.TensorBoard(log_dir=""./logs"")
model.fit(x_train, y_train, epochs=2, callbacks=[tensorboard_callback])
# run the tensorboard command to view the visualizations.
```
Example (Profile):
```python
# profile a single batch, e.g. the 5th batch.
tensorboard_callback = tf.keras.callbacks.TensorBoard(log_dir='./logs',
profile_batch=5)
model.fit(x_train, y_train, epochs=2, callbacks=[tensorboard_callback])
# Now run the tensorboard command to view the visualizations (profile plugin).
# profile a range of batches, e.g. from 10 to 20.
tensorboard_callback = tf.keras.callbacks.TensorBoard(log_dir='./logs',
profile_batch='10,20')
model.fit(x_train, y_train, epochs=2, callbacks=[tensorboard_callback])
# Now run the tensorboard command to view the visualizations (profile plugin).
```
Arguments:
log_dir: the path of the directory where to save the log files to be
parsed by TensorBoard.
histogram_freq: frequency (in epochs) at which to compute activation and
weight histograms for the layers of the model. If set to 0, histograms
won't be computed. Validation data (or split) must be specified for
histogram visualizations.
write_graph: whether to visualize the graph in TensorBoard. The log file
can become quite large when write_graph is set to True.
write_images: whether to write model weights to visualize as image in
TensorBoard.
update_freq: `'batch'` or `'epoch'` or integer. When using `'batch'`,
writes the losses and metrics to TensorBoard after each batch. The same
applies for `'epoch'`. If using an integer, let's say `1000`, the
callback will write the metrics and losses to TensorBoard every 1000
batches. Note that writing too frequently to TensorBoard can slow down
your training.
profile_batch: Profile the batch(es) to sample compute characteristics.
profile_batch must be a non-negative integer or a tuple of integers.
A pair of positive integers signify a range of batches to profile.
By default, it will profile the second batch. Set profile_batch=0
to disable profiling.
embeddings_freq: frequency (in epochs) at which embedding layers will be
visualized. If set to 0, embeddings won't be visualized.
embeddings_metadata: a dictionary which maps layer name to a file name in
which metadata for this embedding layer is saved. See the
[details](
https://www.tensorflow.org/how_tos/embedding_viz/#metadata_optional)
about metadata files format. In case if the same metadata file is
used for all embedding layers, string can be passed.
Raises:
ValueError: If histogram_freq is set and no validation data is provided."
4787,set_model,tensorflow/tensorflow/python/keras/callbacks.py,2004,method,Sets Keras model and writes graph if specified.
4788,on_train_begin,tensorflow/tensorflow/python/keras/callbacks.py,2187,method,
4789,on_train_end,tensorflow/tensorflow/python/keras/callbacks.py,2191,method,
4790,on_train_batch_begin,tensorflow/tensorflow/python/keras/callbacks.py,2209,method,
4791,on_train_batch_end,tensorflow/tensorflow/python/keras/callbacks.py,2217,method,
4792,on_epoch_begin,tensorflow/tensorflow/python/keras/callbacks.py,2227,method,
4793,on_epoch_end,tensorflow/tensorflow/python/keras/callbacks.py,2231,method,Runs metrics and histogram summaries at epoch end.
4794,ReduceLROnPlateau,tensorflow/tensorflow/python/keras/callbacks.py,2324,class,"Reduce learning rate when a metric has stopped improving.
Models often benefit from reducing the learning rate by a factor
of 2-10 once learning stagnates. This callback monitors a
quantity and if no improvement is seen for a 'patience' number
of epochs, the learning rate is reduced.
Example:
```python
reduce_lr = ReduceLROnPlateau(monitor='val_loss', factor=0.2,
patience=5, min_lr=0.001)
model.fit(X_train, Y_train, callbacks=[reduce_lr])
```
Arguments:
monitor: quantity to be monitored.
factor: factor by which the learning rate will be reduced.
`new_lr = lr * factor`.
patience: number of epochs with no improvement after which learning rate
will be reduced.
verbose: int. 0: quiet, 1: update messages.
mode: one of `{'auto', 'min', 'max'}`. In `'min'` mode,
the learning rate will be reduced when the
quantity monitored has stopped decreasing; in `'max'` mode it will be
reduced when the quantity monitored has stopped increasing; in `'auto'`
mode, the direction is automatically inferred from the name of the
monitored quantity.
min_delta: threshold for measuring the new optimum, to only focus on
significant changes.
cooldown: number of epochs to wait before resuming normal operation after
lr has been reduced.
min_lr: lower bound on the learning rate."
4795,on_train_begin,tensorflow/tensorflow/python/keras/callbacks.py,2409,method,
4796,on_epoch_end,tensorflow/tensorflow/python/keras/callbacks.py,2412,method,
4797,in_cooldown,tensorflow/tensorflow/python/keras/callbacks.py,2443,method,
4798,CSVLogger,tensorflow/tensorflow/python/keras/callbacks.py,2448,class,"Callback that streams epoch results to a CSV file.
Supports all values that can be represented as a string,
including 1D iterables such as `np.ndarray`.
Example:
```python
csv_logger = CSVLogger('training.log')
model.fit(X_train, Y_train, callbacks=[csv_logger])
```
Arguments:
filename: Filename of the CSV file, e.g. `'run/log.csv'`.
separator: String used to separate elements in the CSV file.
append: Boolean. True: append if file exists (useful for continuing
training). False: overwrite existing file."
4799,on_train_begin,tensorflow/tensorflow/python/keras/callbacks.py,2483,method,
4800,on_epoch_end,tensorflow/tensorflow/python/keras/callbacks.py,2495,method,
4801,on_train_end,tensorflow/tensorflow/python/keras/callbacks.py,2535,method,
4802,handle_value,tensorflow/tensorflow/python/keras/callbacks.py,2498,method,
4803,LambdaCallback,tensorflow/tensorflow/python/keras/callbacks.py,2541,class,"Callback for creating simple, custom callbacks on-the-fly.
This callback is constructed with anonymous functions that will be called
at the appropriate time. Note that the callbacks expects positional
arguments, as:
- `on_epoch_begin` and `on_epoch_end` expect two positional arguments:
`epoch`, `logs`
- `on_batch_begin` and `on_batch_end` expect two positional arguments:
`batch`, `logs`
- `on_train_begin` and `on_train_end` expect one positional argument:
`logs`
Arguments:
on_epoch_begin: called at the beginning of every epoch.
on_epoch_end: called at the end of every epoch.
on_batch_begin: called at the beginning of every batch.
on_batch_end: called at the end of every batch.
on_train_begin: called at the beginning of model training.
on_train_end: called at the end of model training.
Example:
```python
# Print the batch number at the beginning of every batch.
batch_print_callback = LambdaCallback(
on_batch_begin=lambda batch,logs: print(batch))
# Stream the epoch loss to a file in JSON format. The file content
# is not well-formed JSON but rather has a JSON object per line.
import json
json_log = open('loss_log.json', mode='wt', buffering=1)
json_logging_callback = LambdaCallback(
on_epoch_end=lambda epoch, logs: json_log.write(
json.dumps({'epoch': epoch, 'loss': logs['loss']}) + '\n'),
on_train_end=lambda logs: json_log.close()
)
# Terminate some processes after having finished model training.
processes = ...
cleanup_callback = LambdaCallback(
on_train_end=lambda logs: [
p.terminate() for p in processes if p.is_alive()])
model.fit(...,
callbacks=[batch_print_callback,
json_logging_callback,
cleanup_callback])
```"
4804,Counter,tensorflow/tensorflow/python/keras/callbacks_test.py,77,class,"Counts the number of times each callback method was run.
Attributes:
method_counts: dict. Contains the counts of time each callback method was
run."
4805,wrap_with_counts,tensorflow/tensorflow/python/keras/callbacks_test.py,98,method,
4806,list_summaries,tensorflow/tensorflow/python/keras/callbacks_test.py,1773,function,"Read all summaries under the logdir into a `_SummaryFile`.
Args:
logdir: A path to a directory that contains zero or more event
files, either as direct children or in transitive subdirectories.
Summaries in these events must only contain old-style scalars,
images, and histograms. Non-summary events, like `graph_def`s, are
ignored.
Returns:
A `_SummaryFile` object reflecting all summaries written to any
event files in the logdir or any of its descendant directories.
Raises:
ValueError: If an event file contains an summary of unexpected kind."
4807,TensorBoard,tensorflow/tensorflow/python/keras/callbacks_v1.py,42,class,"Enable visualizations for TensorBoard.
TensorBoard is a visualization tool provided with TensorFlow.
This callback logs events for TensorBoard, including:
* Metrics summary plots
* Training graph visualization
* Activation histograms
* Sampled profiling
If you have installed TensorFlow with pip, you should be able
to launch TensorBoard from the command line:
```sh
tensorboard --logdir=path_to_your_logs
```
You can find more information about TensorBoard
[here](https://www.tensorflow.org/get_started/summaries_and_tensorboard).
Arguments:
log_dir: the path of the directory where to save the log files to be
parsed by TensorBoard.
histogram_freq: frequency (in epochs) at which to compute activation and
weight histograms for the layers of the model. If set to 0, histograms
won't be computed. Validation data (or split) must be specified for
histogram visualizations.
write_graph: whether to visualize the graph in TensorBoard. The log file
can become quite large when write_graph is set to True.
write_grads: whether to visualize gradient histograms in TensorBoard.
`histogram_freq` must be greater than 0.
batch_size: size of batch of inputs to feed to the network for histograms
computation.
write_images: whether to write model weights to visualize as image in
TensorBoard.
embeddings_freq: frequency (in epochs) at which selected embedding layers
will be saved. If set to 0, embeddings won't be computed. Data to be
visualized in TensorBoard's Embedding tab must be passed as
`embeddings_data`.
embeddings_layer_names: a list of names of layers to keep eye on. If None
or empty list all the embedding layer will be watched.
embeddings_metadata: a dictionary which maps layer name to a file name in
which metadata for this embedding layer is saved.
[Here are details](
https://www.tensorflow.org/how_tos/embedding_viz/#metadata_optional)
about metadata files format. In case if the same metadata file is
used for all embedding layers, string can be passed.
embeddings_data: data to be embedded at layers specified in
`embeddings_layer_names`. Numpy array (if the model has a single input)
or list of Numpy arrays (if the model has multiple inputs). Learn more
about embeddings [in this guide](
https://www.tensorflow.org/programmers_guide/embedding).
update_freq: `'batch'` or `'epoch'` or integer. When using `'batch'`,
writes the losses and metrics to TensorBoard after each batch. The same
applies for `'epoch'`. If using an integer, let's say `1000`, the
callback will write the metrics and losses to TensorBoard every 1000
samples. Note that writing too frequently to TensorBoard can slow down
your training.
profile_batch: Profile the batch to sample compute characteristics. By
default, it will profile the second batch. Set profile_batch=0 to
disable profiling.
Raises:
ValueError: If histogram_freq is set and no validation data is provided.
@compatibility(eager)
Using the `TensorBoard` callback will work when eager execution is enabled,
with the restriction that outputting histogram summaries of weights and
gradients is not supported. Consequently, `histogram_freq` will be ignored.
@end_compatibility"
4808,set_model,tensorflow/tensorflow/python/keras/callbacks_v1.py,234,method,Sets Keras model and creates summary ops.
4809,on_train_batch_begin,tensorflow/tensorflow/python/keras/callbacks_v1.py,347,method,
4810,on_train_batch_end,tensorflow/tensorflow/python/keras/callbacks_v1.py,353,method,
4811,on_batch_end,tensorflow/tensorflow/python/keras/callbacks_v1.py,362,method,"Writes scalar summaries for metrics on every training batch.
Performs profiling if current batch is in profiler_batches."
4812,on_train_begin,tensorflow/tensorflow/python/keras/callbacks_v1.py,383,method,
4813,on_epoch_begin,tensorflow/tensorflow/python/keras/callbacks_v1.py,386,method,"Add histogram op to Model eval_function callbacks, reset batch count."
4814,on_epoch_end,tensorflow/tensorflow/python/keras/callbacks_v1.py,401,method,"Checks if summary ops should run next epoch, logs scalar summaries."
4815,on_train_end,tensorflow/tensorflow/python/keras/callbacks_v1.py,465,method,
4816,is_indexed_slices,tensorflow/tensorflow/python/keras/callbacks_v1.py,218,method,
4817,keras_mode_combinations,tensorflow/tensorflow/python/keras/combinations.py,31,function,"Returns the default test combinations for tf.keras tests.
Note that if tf2 is enabled, then v1 session test will be skipped.
Args:
mode: List of modes to run the tests. The valid options are 'graph' and
'eager'. Default to ['graph', 'eager'] if not specified. If a empty list
is provide, then the test will run under the context based on tf's
version, eg graph for v1 and eager for v2.
run_eagerly: List of `run_eagerly` value to be run with the tests.
Default to [True, False] if not specified. Note that for `graph` mode,
run_eagerly value will only be False.
Returns:
A list contains all the combinations to be used to generate test cases."
4818,keras_model_type_combinations,tensorflow/tensorflow/python/keras/combinations.py,60,function,
4819,keras_tensor_combinations,tensorflow/tensorflow/python/keras/combinations.py,64,function,
4820,KerasModeCombination,tensorflow/tensorflow/python/keras/combinations.py,68,class,"Combination for Keras test mode.
It by default includes v1_session, v2_eager and v2_tf_function."
4821,context_managers,tensorflow/tensorflow/python/keras/combinations.py,74,method,
4822,parameter_modifiers,tensorflow/tensorflow/python/keras/combinations.py,82,method,
4823,KerasModelTypeCombination,tensorflow/tensorflow/python/keras/combinations.py,86,class,"Combination for Keras model types when doing model test.
It by default includes 'functional', 'subclass', 'sequential'.
Various methods in `testing_utils` to get models will auto-generate a model
of the currently active Keras model type. This allows unittests to confirm
the equivalence between different Keras models."
4824,context_managers,tensorflow/tensorflow/python/keras/combinations.py,96,method,
4825,parameter_modifiers,tensorflow/tensorflow/python/keras/combinations.py,103,method,
4826,KerasTensorCombination,tensorflow/tensorflow/python/keras/combinations.py,107,class,"Combination for whether KerasTensors are being used or not.
It by default includes `True` and `False`:
running Keras's functional API with KerasTensors
as the inputs, and without."
4827,context_managers,tensorflow/tensorflow/python/keras/combinations.py,115,method,
4828,parameter_modifiers,tensorflow/tensorflow/python/keras/combinations.py,123,method,
4829,Constraint,tensorflow/tensorflow/python/keras/constraints.py,36,class,
4830,get_config,tensorflow/tensorflow/python/keras/constraints.py,41,method,
4831,MaxNorm,tensorflow/tensorflow/python/keras/constraints.py,46,class,"MaxNorm weight constraint.
Constrains the weights incident to each hidden unit
to have a norm less than or equal to a desired value.
Also available via the shortcut function `tf.keras.constraints.max_norm`.
Arguments:
max_value: the maximum norm value for the incoming weights.
axis: integer, axis along which to calculate weight norms.
For instance, in a `Dense` layer the weight matrix
has shape `(input_dim, output_dim)`,
set `axis` to `0` to constrain each weight vector
of length `(input_dim,)`.
In a `Conv2D` layer with `data_format=""channels_last""`,
the weight tensor has shape
`(rows, cols, input_depth, output_depth)`,
set `axis` to `[0, 1, 2]`
to constrain the weights of each filter tensor of size
`(rows, cols, input_depth)`."
4832,get_config,tensorflow/tensorflow/python/keras/constraints.py,82,method,
4833,NonNeg,tensorflow/tensorflow/python/keras/constraints.py,87,class,"Constrains the weights to be non-negative.
Also available via the shortcut function `tf.keras.constraints.non_neg`."
4834,UnitNorm,tensorflow/tensorflow/python/keras/constraints.py,98,class,"Constrains the weights incident to each hidden unit to have unit norm.
Also available via the shortcut function `tf.keras.constraints.unit_norm`.
Arguments:
axis: integer, axis along which to calculate weight norms.
For instance, in a `Dense` layer the weight matrix
has shape `(input_dim, output_dim)`,
set `axis` to `0` to constrain each weight vector
of length `(input_dim,)`.
In a `Conv2D` layer with `data_format=""channels_last""`,
the weight tensor has shape
`(rows, cols, input_depth, output_depth)`,
set `axis` to `[0, 1, 2]`
to constrain the weights of each filter tensor of size
`(rows, cols, input_depth)`."
4835,get_config,tensorflow/tensorflow/python/keras/constraints.py,128,method,
4836,MinMaxNorm,tensorflow/tensorflow/python/keras/constraints.py,133,class,"MinMaxNorm weight constraint.
Constrains the weights incident to each hidden unit
to have the norm between a lower bound and an upper bound.
Also available via the shortcut function `tf.keras.constraints.min_max_norm`.
Arguments:
min_value: the minimum norm for the incoming weights.
max_value: the maximum norm for the incoming weights.
rate: rate for enforcing the constraint: weights will be
rescaled to yield
`(1 - rate) * norm + rate * norm.clip(min_value, max_value)`.
Effectively, this means that rate=1.0 stands for strict
enforcement of the constraint, while rate<1.0 means that
weights will be rescaled at each step to slowly move
towards a value inside the desired interval.
axis: integer, axis along which to calculate weight norms.
For instance, in a `Dense` layer the weight matrix
has shape `(input_dim, output_dim)`,
set `axis` to `0` to constrain each weight vector
of length `(input_dim,)`.
In a `Conv2D` layer with `data_format=""channels_last""`,
the weight tensor has shape
`(rows, cols, input_depth, output_depth)`,
set `axis` to `[0, 1, 2]`
to constrain the weights of each filter tensor of size
`(rows, cols, input_depth)`."
4837,get_config,tensorflow/tensorflow/python/keras/constraints.py,180,method,
4838,RadialConstraint,tensorflow/tensorflow/python/keras/constraints.py,191,class,"Constrains `Conv2D` kernel weights to be the same for each radius.
Also available via the shortcut function
`tf.keras.constraints.radial_constraint`.
For example, the desired output for the following 4-by-4 kernel:
```
kernel = [[v_00, v_01, v_02, v_03],
[v_10, v_11, v_12, v_13],
[v_20, v_21, v_22, v_23],
[v_30, v_31, v_32, v_33]]
```
is this::
```
kernel = [[v_11, v_11, v_11, v_11],
[v_11, v_33, v_33, v_11],
[v_11, v_33, v_33, v_11],
[v_11, v_11, v_11, v_11]]
```
This constraint can be applied to any `Conv2D` layer version, including
`Conv2DTranspose` and `SeparableConv2D`, and with either `""channels_last""` or
`""channels_first""` data format. The method assumes the weight tensor is of
shape `(rows, cols, input_depth, output_depth)`."
4839,body_fn,tensorflow/tensorflow/python/keras/constraints.py,256,method,
4840,serialize,tensorflow/tensorflow/python/keras/constraints.py,286,function,
4841,deserialize,tensorflow/tensorflow/python/keras/constraints.py,291,function,
4842,get,tensorflow/tensorflow/python/keras/constraints.py,300,function,
4843,get_example_array,tensorflow/tensorflow/python/keras/constraints_test.py,35,function,
4844,get_example_kernel,tensorflow/tensorflow/python/keras/constraints_test.py,42,function,
4845,run_with_all_saved_model_formats,tensorflow/tensorflow/python/keras/keras_parameterized.py,49,function,"Execute the decorated test with all Keras saved model formats).
This decorator is intended to be applied either to individual test methods in
a `keras_parameterized.TestCase` class, or directly to a test class that
extends it. Doing so will cause the contents of the individual test
method (or all test methods in the class) to be executed multiple times - once
for each Keras saved model format.
The Keras saved model formats include:
1. HDF5: 'h5'
2. SavedModel: 'tf'
Note: if stacking this decorator with absl.testing's parameterized decorators,
those should be at the bottom of the stack.
Various methods in `testing_utils` to get file path for saved models will
auto-generate a string of the two saved model formats. This allows unittests
to confirm the equivalence between the two Keras saved model formats.
For example, consider the following unittest:
```python
class MyTests(testing_utils.KerasTestCase):
@testing_utils.run_with_all_saved_model_formats
def test_foo(self):
save_format = testing_utils.get_save_format()
saved_model_dir = '/tmp/saved_model/'
model = keras.models.Sequential()
model.add(keras.layers.Dense(2, input_shape=(3,)))
model.add(keras.layers.Dense(3))
model.compile(loss='mse', optimizer='sgd', metrics=['acc'])
keras.models.save_model(model, saved_model_dir, save_format=save_format)
model = keras.models.load_model(saved_model_dir)
if __name__ == ""__main__"":
tf.test.main()
```
This test tries to save the model into the formats of 'hdf5', 'h5', 'keras',
'tensorflow', and 'tf'.
We can also annotate the whole class if we want this to apply to all tests in
the class:
```python
@testing_utils.run_with_all_saved_model_formats
class MyTests(testing_utils.KerasTestCase):
def test_foo(self):
save_format = testing_utils.get_save_format()
saved_model_dir = '/tmp/saved_model/'
model = keras.models.Sequential()
model.add(keras.layers.Dense(2, input_shape=(3,)))
model.add(keras.layers.Dense(3))
model.compile(loss='mse', optimizer='sgd', metrics=['acc'])
keras.models.save_model(model, saved_model_dir, save_format=save_format)
model = tf.keras.models.load_model(saved_model_dir)
if __name__ == ""__main__"":
tf.test.main()
```
Args:
test_or_class: test method or class to be annotated. If None,
this method returns a decorator that can be applied to a test method or
test class. If it is not None this returns the decorator applied to the
test or class.
exclude_formats: A collection of Keras saved model formats to not run.
(May also be a single format not wrapped in a collection).
Defaults to None.
Returns:
Returns a decorator that will run the decorated test method multiple times:
once for each desired Keras saved model format.
Raises:
ImportError: If abseil parameterized is not installed or not included as
a target dependency."
4846,run_with_all_model_types,tensorflow/tensorflow/python/keras/keras_parameterized.py,172,function,"Execute the decorated test with all Keras model types.
This decorator is intended to be applied either to individual test methods in
a `keras_parameterized.TestCase` class, or directly to a test class that
extends it. Doing so will cause the contents of the individual test
method (or all test methods in the class) to be executed multiple times - once
for each Keras model type.
The Keras model types are: ['functional', 'subclass', 'sequential']
Note: if stacking this decorator with absl.testing's parameterized decorators,
those should be at the bottom of the stack.
Various methods in `testing_utils` to get models will auto-generate a model
of the currently active Keras model type. This allows unittests to confirm
the equivalence between different Keras models.
For example, consider the following unittest:
```python
class MyTests(testing_utils.KerasTestCase):
@testing_utils.run_with_all_model_types(
exclude_models = ['sequential'])
def test_foo(self):
model = testing_utils.get_small_mlp(1, 4, input_dim=3)
optimizer = RMSPropOptimizer(learning_rate=0.001)
loss = 'mse'
metrics = ['mae']
model.compile(optimizer, loss, metrics=metrics)
inputs = np.zeros((10, 3))
targets = np.zeros((10, 4))
dataset = dataset_ops.Dataset.from_tensor_slices((inputs, targets))
dataset = dataset.repeat(100)
dataset = dataset.batch(10)
model.fit(dataset, epochs=1, steps_per_epoch=2, verbose=1)
if __name__ == ""__main__"":
tf.test.main()
```
This test tries building a small mlp as both a functional model and as a
subclass model.
We can also annotate the whole class if we want this to apply to all tests in
the class:
```python
@testing_utils.run_with_all_model_types(exclude_models = ['sequential'])
class MyTests(testing_utils.KerasTestCase):
def test_foo(self):
model = testing_utils.get_small_mlp(1, 4, input_dim=3)
optimizer = RMSPropOptimizer(learning_rate=0.001)
loss = 'mse'
metrics = ['mae']
model.compile(optimizer, loss, metrics=metrics)
inputs = np.zeros((10, 3))
targets = np.zeros((10, 4))
dataset = dataset_ops.Dataset.from_tensor_slices((inputs, targets))
dataset = dataset.repeat(100)
dataset = dataset.batch(10)
model.fit(dataset, epochs=1, steps_per_epoch=2, verbose=1)
if __name__ == ""__main__"":
tf.test.main()
```
Args:
test_or_class: test method or class to be annotated. If None,
this method returns a decorator that can be applied to a test method or
test class. If it is not None this returns the decorator applied to the
test or class.
exclude_models: A collection of Keras model types to not run.
(May also be a single model type not wrapped in a collection).
Defaults to None.
Returns:
Returns a decorator that will run the decorated test method multiple times:
once for each desired Keras model type.
Raises:
ImportError: If abseil parameterized is not installed or not included as
a target dependency."
4847,run_all_keras_modes,tensorflow/tensorflow/python/keras/keras_parameterized.py,303,function,"Execute the decorated test with all keras execution modes.
This decorator is intended to be applied either to individual test methods in
a `keras_parameterized.TestCase` class, or directly to a test class that
extends it. Doing so will cause the contents of the individual test
method (or all test methods in the class) to be executed multiple times -
once executing in legacy graph mode, once running eagerly and with
`should_run_eagerly` returning True, and once running eagerly with
`should_run_eagerly` returning False.
If Tensorflow v2 behavior is enabled, legacy graph mode will be skipped, and
the test will only run twice.
Note: if stacking this decorator with absl.testing's parameterized decorators,
those should be at the bottom of the stack.
For example, consider the following unittest:
```python
class MyTests(testing_utils.KerasTestCase):
@testing_utils.run_all_keras_modes
def test_foo(self):
model = testing_utils.get_small_functional_mlp(1, 4, input_dim=3)
optimizer = RMSPropOptimizer(learning_rate=0.001)
loss = 'mse'
metrics = ['mae']
model.compile(
optimizer, loss, metrics=metrics,
run_eagerly=testing_utils.should_run_eagerly())
inputs = np.zeros((10, 3))
targets = np.zeros((10, 4))
dataset = dataset_ops.Dataset.from_tensor_slices((inputs, targets))
dataset = dataset.repeat(100)
dataset = dataset.batch(10)
model.fit(dataset, epochs=1, steps_per_epoch=2, verbose=1)
if __name__ == ""__main__"":
tf.test.main()
```
This test will try compiling & fitting the small functional mlp using all
three Keras execution modes.
Args:
test_or_class: test method or class to be annotated. If None,
this method returns a decorator that can be applied to a test method or
test class. If it is not None this returns the decorator applied to the
test or class.
config: An optional config_pb2.ConfigProto to use to configure the
session when executing graphs.
always_skip_v1: If True, does not try running the legacy graph mode even
when Tensorflow v2 behavior is not enabled.
always_skip_eager: If True, does not execute the decorated test
with eager execution modes.
**kwargs: Additional kwargs for configuring tests for
in-progress Keras behaviors/ refactorings that we haven't fully
rolled out yet
Returns:
Returns a decorator that will run the decorated test method multiple times.
Raises:
ImportError: If abseil parameterized is not installed or not included as
a target dependency."
4848,Loss,tensorflow/tensorflow/python/keras/losses.py,46,class,"Loss base class.
To be implemented by subclasses:
* `call()`: Contains the logic for loss calculation using `y_true`, `y_pred`.
Example subclass implementation:
```python
class MeanSquaredError(Loss):
def call(self, y_true, y_pred):
y_pred = tf.convert_to_tensor_v2(y_pred)
y_true = tf.cast(y_true, y_pred.dtype)
return tf.reduce_mean(math_ops.square(y_pred - y_true), axis=-1)
```
When used with `tf.distribute.Strategy`, outside of built-in training loops
such as `tf.keras` `compile` and `fit`, please use 'SUM' or 'NONE' reduction
types, and reduce losses explicitly in your training loop. Using 'AUTO' or
'SUM_OVER_BATCH_SIZE' will raise an error.
Please see this custom training [tutorial](
https://www.tensorflow.org/tutorials/distribute/custom_training) for more
details on this.
You can implement 'SUM_OVER_BATCH_SIZE' using global batch size like:
```python
with strategy.scope():
loss_obj = tf.keras.losses.CategoricalCrossentropy(
reduction=tf.keras.losses.Reduction.NONE)
....
loss = (tf.reduce_sum(loss_obj(labels, predictions)) *
(1. / global_batch_size))
```"
4849,from_config,tensorflow/tensorflow/python/keras/losses.py,153,method,"Instantiates a `Loss` from its config (output of `get_config()`).
Args:
config: Output of `get_config()`.
Returns:
A `Loss` instance."
4850,get_config,tensorflow/tensorflow/python/keras/losses.py,164,method,Returns the config dictionary for a `Loss` instance.
4851,call,tensorflow/tensorflow/python/keras/losses.py,170,method,"Invokes the `Loss` instance.
Args:
y_true: Ground truth values. shape = `[batch_size, d0, .. dN]`, except
sparse loss functions such as sparse categorical crossentropy where
shape = `[batch_size, d0, .. dN-1]`
y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`
Returns:
Loss values with the shape `[batch_size, d0, .. dN-1]`."
4852,LossFunctionWrapper,tensorflow/tensorflow/python/keras/losses.py,209,class,Wraps a loss function in the `Loss` class.
4853,call,tensorflow/tensorflow/python/keras/losses.py,238,method,"Invokes the `LossFunctionWrapper` instance.
Args:
y_true: Ground truth values.
y_pred: The predicted values.
Returns:
Loss values per sample."
4854,get_config,tensorflow/tensorflow/python/keras/losses.py,254,method,
4855,MeanSquaredError,tensorflow/tensorflow/python/keras/losses.py,263,class,"Computes the mean of squares of errors between labels and predictions.
`loss = square(y_true - y_pred)`
Standalone usage:
>>> y_true = [[0., 1.], [0., 0.]]
>>> y_pred = [[1., 1.], [1., 0.]]
>>> # Using 'auto'/'sum_over_batch_size' reduction type.
>>> mse = tf.keras.losses.MeanSquaredError()
>>> mse(y_true, y_pred).numpy()
0.5
>>> # Calling with 'sample_weight'.
>>> mse(y_true, y_pred, sample_weight=[0.7, 0.3]).numpy()
0.25
>>> # Using 'sum' reduction type.
>>> mse = tf.keras.losses.MeanSquaredError(
... reduction=tf.keras.losses.Reduction.SUM)
>>> mse(y_true, y_pred).numpy()
1.0
>>> # Using 'none' reduction type.
>>> mse = tf.keras.losses.MeanSquaredError(
... reduction=tf.keras.losses.Reduction.NONE)
>>> mse(y_true, y_pred).numpy()
array([0.5, 0.5], dtype=float32)
Usage with the `compile()` API:
```python
model.compile(optimizer='sgd', loss=tf.keras.losses.MeanSquaredError())
```"
4856,MeanAbsoluteError,tensorflow/tensorflow/python/keras/losses.py,322,class,"Computes the mean of absolute difference between labels and predictions.
`loss = abs(y_true - y_pred)`
Standalone usage:
>>> y_true = [[0., 1.], [0., 0.]]
>>> y_pred = [[1., 1.], [1., 0.]]
>>> # Using 'auto'/'sum_over_batch_size' reduction type.
>>> mae = tf.keras.losses.MeanAbsoluteError()
>>> mae(y_true, y_pred).numpy()
0.5
>>> # Calling with 'sample_weight'.
>>> mae(y_true, y_pred, sample_weight=[0.7, 0.3]).numpy()
0.25
>>> # Using 'sum' reduction type.
>>> mae = tf.keras.losses.MeanAbsoluteError(
... reduction=tf.keras.losses.Reduction.SUM)
>>> mae(y_true, y_pred).numpy()
1.0
>>> # Using 'none' reduction type.
>>> mae = tf.keras.losses.MeanAbsoluteError(
... reduction=tf.keras.losses.Reduction.NONE)
>>> mae(y_true, y_pred).numpy()
array([0.5, 0.5], dtype=float32)
Usage with the `compile()` API:
```python
model.compile(optimizer='sgd', loss=tf.keras.losses.MeanAbsoluteError())
```"
4857,MeanAbsolutePercentageError,tensorflow/tensorflow/python/keras/losses.py,381,class,"Computes the mean absolute percentage error between `y_true` and `y_pred`.
`loss = 100 * abs(y_true - y_pred) / y_true`
Standalone usage:
>>> y_true = [[2., 1.], [2., 3.]]
>>> y_pred = [[1., 1.], [1., 0.]]
>>> # Using 'auto'/'sum_over_batch_size' reduction type.
>>> mape = tf.keras.losses.MeanAbsolutePercentageError()
>>> mape(y_true, y_pred).numpy()
50.
>>> # Calling with 'sample_weight'.
>>> mape(y_true, y_pred, sample_weight=[0.7, 0.3]).numpy()
20.
>>> # Using 'sum' reduction type.
>>> mape = tf.keras.losses.MeanAbsolutePercentageError(
... reduction=tf.keras.losses.Reduction.SUM)
>>> mape(y_true, y_pred).numpy()
100.
>>> # Using 'none' reduction type.
>>> mape = tf.keras.losses.MeanAbsolutePercentageError(
... reduction=tf.keras.losses.Reduction.NONE)
>>> mape(y_true, y_pred).numpy()
array([25., 75.], dtype=float32)
Usage with the `compile()` API:
```python
model.compile(optimizer='sgd',
loss=tf.keras.losses.MeanAbsolutePercentageError())
```"
4858,MeanSquaredLogarithmicError,tensorflow/tensorflow/python/keras/losses.py,442,class,"Computes the mean squared logarithmic error between `y_true` and `y_pred`.
`loss = square(log(y_true + 1.) - log(y_pred + 1.))`
Standalone usage:
>>> y_true = [[0., 1.], [0., 0.]]
>>> y_pred = [[1., 1.], [1., 0.]]
>>> # Using 'auto'/'sum_over_batch_size' reduction type.
>>> msle = tf.keras.losses.MeanSquaredLogarithmicError()
>>> msle(y_true, y_pred).numpy()
0.240
>>> # Calling with 'sample_weight'.
>>> msle(y_true, y_pred, sample_weight=[0.7, 0.3]).numpy()
0.120
>>> # Using 'sum' reduction type.
>>> msle = tf.keras.losses.MeanSquaredLogarithmicError(
... reduction=tf.keras.losses.Reduction.SUM)
>>> msle(y_true, y_pred).numpy()
0.480
>>> # Using 'none' reduction type.
>>> msle = tf.keras.losses.MeanSquaredLogarithmicError(
... reduction=tf.keras.losses.Reduction.NONE)
>>> msle(y_true, y_pred).numpy()
array([0.240, 0.240], dtype=float32)
Usage with the `compile()` API:
```python
model.compile(optimizer='sgd',
loss=tf.keras.losses.MeanSquaredLogarithmicError())
```"
4859,BinaryCrossentropy,tensorflow/tensorflow/python/keras/losses.py,503,class,"Computes the cross-entropy loss between true labels and predicted labels.
Use this cross-entropy loss when there are only two label classes (assumed to
be 0 and 1). For each example, there should be a single floating-point value
per prediction.
In the snippet below, each of the four examples has only a single
floating-pointing value, and both `y_pred` and `y_true` have the shape
`[batch_size]`.
Standalone usage:
>>> y_true = [[0., 1.], [0., 0.]]
>>> y_pred = [[0.6, 0.4], [0.4, 0.6]]
>>> # Using 'auto'/'sum_over_batch_size' reduction type.
>>> bce = tf.keras.losses.BinaryCrossentropy()
>>> bce(y_true, y_pred).numpy()
0.815
>>> # Calling with 'sample_weight'.
>>> bce(y_true, y_pred, sample_weight=[1, 0]).numpy()
0.458
>>> # Using 'sum' reduction type.
>>> bce = tf.keras.losses.BinaryCrossentropy(
... reduction=tf.keras.losses.Reduction.SUM)
>>> bce(y_true, y_pred).numpy()
1.630
>>> # Using 'none' reduction type.
>>> bce = tf.keras.losses.BinaryCrossentropy(
... reduction=tf.keras.losses.Reduction.NONE)
>>> bce(y_true, y_pred).numpy()
array([0.916 , 0.714], dtype=float32)
Usage with the `tf.keras` API:
```python
model.compile(optimizer='sgd', loss=tf.keras.losses.BinaryCrossentropy())
```"
4860,CategoricalCrossentropy,tensorflow/tensorflow/python/keras/losses.py,583,class,"Computes the crossentropy loss between the labels and predictions.
Use this crossentropy loss function when there are two or more label classes.
We expect labels to be provided in a `one_hot` representation. If you want to
provide labels as integers, please use `SparseCategoricalCrossentropy` loss.
There should be `# classes` floating point values per feature.
In the snippet below, there is `# classes` floating pointing values per
example. The shape of both `y_pred` and `y_true` are
`[batch_size, num_classes]`.
Standalone usage:
>>> y_true = [[0, 1, 0], [0, 0, 1]]
>>> y_pred = [[0.05, 0.95, 0], [0.1, 0.8, 0.1]]
>>> # Using 'auto'/'sum_over_batch_size' reduction type.
>>> cce = tf.keras.losses.CategoricalCrossentropy()
>>> cce(y_true, y_pred).numpy()
1.177
>>> # Calling with 'sample_weight'.
>>> cce(y_true, y_pred, sample_weight=tf.constant([0.3, 0.7])).numpy()
0.814
>>> # Using 'sum' reduction type.
>>> cce = tf.keras.losses.CategoricalCrossentropy(
... reduction=tf.keras.losses.Reduction.SUM)
>>> cce(y_true, y_pred).numpy()
2.354
>>> # Using 'none' reduction type.
>>> cce = tf.keras.losses.CategoricalCrossentropy(
... reduction=tf.keras.losses.Reduction.NONE)
>>> cce(y_true, y_pred).numpy()
array([0.0513, 2.303], dtype=float32)
Usage with the `compile()` API:
```python
model.compile(optimizer='sgd', loss=tf.keras.losses.CategoricalCrossentropy())
```"
4861,SparseCategoricalCrossentropy,tensorflow/tensorflow/python/keras/losses.py,662,class,"Computes the crossentropy loss between the labels and predictions.
Use this crossentropy loss function when there are two or more label classes.
We expect labels to be provided as integers. If you want to provide labels
using `one-hot` representation, please use `CategoricalCrossentropy` loss.
There should be `# classes` floating point values per feature for `y_pred`
and a single floating point value per feature for `y_true`.
In the snippet below, there is a single floating point value per example for
`y_true` and `# classes` floating pointing values per example for `y_pred`.
The shape of `y_true` is `[batch_size]` and the shape of `y_pred` is
`[batch_size, num_classes]`.
Standalone usage:
>>> y_true = [1, 2]
>>> y_pred = [[0.05, 0.95, 0], [0.1, 0.8, 0.1]]
>>> # Using 'auto'/'sum_over_batch_size' reduction type.
>>> scce = tf.keras.losses.SparseCategoricalCrossentropy()
>>> scce(y_true, y_pred).numpy()
1.177
>>> # Calling with 'sample_weight'.
>>> scce(y_true, y_pred, sample_weight=tf.constant([0.3, 0.7])).numpy()
0.814
>>> # Using 'sum' reduction type.
>>> scce = tf.keras.losses.SparseCategoricalCrossentropy(
... reduction=tf.keras.losses.Reduction.SUM)
>>> scce(y_true, y_pred).numpy()
2.354
>>> # Using 'none' reduction type.
>>> scce = tf.keras.losses.SparseCategoricalCrossentropy(
... reduction=tf.keras.losses.Reduction.NONE)
>>> scce(y_true, y_pred).numpy()
array([0.0513, 2.303], dtype=float32)
Usage with the `compile()` API:
```python
model.compile(optimizer='sgd',
loss=tf.keras.losses.SparseCategoricalCrossentropy())
```"
4862,Hinge,tensorflow/tensorflow/python/keras/losses.py,739,class,"Computes the hinge loss between `y_true` and `y_pred`.
`loss = maximum(1 - y_true * y_pred, 0)`
`y_true` values are expected to be -1 or 1. If binary (0 or 1) labels are
provided we will convert them to -1 or 1.
Standalone usage:
>>> y_true = [[0., 1.], [0., 0.]]
>>> y_pred = [[0.6, 0.4], [0.4, 0.6]]
>>> # Using 'auto'/'sum_over_batch_size' reduction type.
>>> h = tf.keras.losses.Hinge()
>>> h(y_true, y_pred).numpy()
1.3
>>> # Calling with 'sample_weight'.
>>> h(y_true, y_pred, sample_weight=[1, 0]).numpy()
0.55
>>> # Using 'sum' reduction type.
>>> h = tf.keras.losses.Hinge(
... reduction=tf.keras.losses.Reduction.SUM)
>>> h(y_true, y_pred).numpy()
2.6
>>> # Using 'none' reduction type.
>>> h = tf.keras.losses.Hinge(
... reduction=tf.keras.losses.Reduction.NONE)
>>> h(y_true, y_pred).numpy()
array([1.1, 1.5], dtype=float32)
Usage with the `compile()` API:
```python
model.compile(optimizer='sgd', loss=tf.keras.losses.Hinge())
```"
4863,SquaredHinge,tensorflow/tensorflow/python/keras/losses.py,798,class,"Computes the squared hinge loss between `y_true` and `y_pred`.
`loss = square(maximum(1 - y_true * y_pred, 0))`
`y_true` values are expected to be -1 or 1. If binary (0 or 1) labels are
provided we will convert them to -1 or 1.
Standalone usage:
>>> y_true = [[0., 1.], [0., 0.]]
>>> y_pred = [[0.6, 0.4], [0.4, 0.6]]
>>> # Using 'auto'/'sum_over_batch_size' reduction type.
>>> h = tf.keras.losses.SquaredHinge()
>>> h(y_true, y_pred).numpy()
1.86
>>> # Calling with 'sample_weight'.
>>> h(y_true, y_pred, sample_weight=[1, 0]).numpy()
0.73
>>> # Using 'sum' reduction type.
>>> h = tf.keras.losses.SquaredHinge(
... reduction=tf.keras.losses.Reduction.SUM)
>>> h(y_true, y_pred).numpy()
3.72
>>> # Using 'none' reduction type.
>>> h = tf.keras.losses.SquaredHinge(
... reduction=tf.keras.losses.Reduction.NONE)
>>> h(y_true, y_pred).numpy()
array([1.46, 2.26], dtype=float32)
Usage with the `compile()` API:
```python
model.compile(optimizer='sgd', loss=tf.keras.losses.SquaredHinge())
```"
4864,CategoricalHinge,tensorflow/tensorflow/python/keras/losses.py,860,class,"Computes the categorical hinge loss between `y_true` and `y_pred`.
`loss = maximum(neg - pos + 1, 0)`
where `neg=maximum((1-y_true)*y_pred) and pos=sum(y_true*y_pred)`
Standalone usage:
>>> y_true = [[0, 1], [0, 0]]
>>> y_pred = [[0.6, 0.4], [0.4, 0.6]]
>>> # Using 'auto'/'sum_over_batch_size' reduction type.
>>> h = tf.keras.losses.CategoricalHinge()
>>> h(y_true, y_pred).numpy()
1.4
>>> # Calling with 'sample_weight'.
>>> h(y_true, y_pred, sample_weight=[1, 0]).numpy()
0.6
>>> # Using 'sum' reduction type.
>>> h = tf.keras.losses.CategoricalHinge(
... reduction=tf.keras.losses.Reduction.SUM)
>>> h(y_true, y_pred).numpy()
2.8
>>> # Using 'none' reduction type.
>>> h = tf.keras.losses.CategoricalHinge(
... reduction=tf.keras.losses.Reduction.NONE)
>>> h(y_true, y_pred).numpy()
array([1.2, 1.6], dtype=float32)
Usage with the `compile()` API:
```python
model.compile(optimizer='sgd', loss=tf.keras.losses.CategoricalHinge())
```"
4865,Poisson,tensorflow/tensorflow/python/keras/losses.py,920,class,"Computes the Poisson loss between `y_true` and `y_pred`.
`loss = y_pred - y_true * log(y_pred)`
Standalone usage:
>>> y_true = [[0., 1.], [0., 0.]]
>>> y_pred = [[1., 1.], [0., 0.]]
>>> # Using 'auto'/'sum_over_batch_size' reduction type.
>>> p = tf.keras.losses.Poisson()
>>> p(y_true, y_pred).numpy()
0.5
>>> # Calling with 'sample_weight'.
>>> p(y_true, y_pred, sample_weight=[0.8, 0.2]).numpy()
0.4
>>> # Using 'sum' reduction type.
>>> p = tf.keras.losses.Poisson(
... reduction=tf.keras.losses.Reduction.SUM)
>>> p(y_true, y_pred).numpy()
0.999
>>> # Using 'none' reduction type.
>>> p = tf.keras.losses.Poisson(
... reduction=tf.keras.losses.Reduction.NONE)
>>> p(y_true, y_pred).numpy()
array([0.999, 0.], dtype=float32)
Usage with the `compile()` API:
```python
model.compile(optimizer='sgd', loss=tf.keras.losses.Poisson())
```"
4866,LogCosh,tensorflow/tensorflow/python/keras/losses.py,976,class,"Computes the logarithm of the hyperbolic cosine of the prediction error.
`logcosh = log((exp(x) + exp(-x))/2)`,
where x is the error `y_pred - y_true`.
Standalone usage:
>>> y_true = [[0., 1.], [0., 0.]]
>>> y_pred = [[1., 1.], [0., 0.]]
>>> # Using 'auto'/'sum_over_batch_size' reduction type.
>>> l = tf.keras.losses.LogCosh()
>>> l(y_true, y_pred).numpy()
0.108
>>> # Calling with 'sample_weight'.
>>> l(y_true, y_pred, sample_weight=[0.8, 0.2]).numpy()
0.087
>>> # Using 'sum' reduction type.
>>> l = tf.keras.losses.LogCosh(
... reduction=tf.keras.losses.Reduction.SUM)
>>> l(y_true, y_pred).numpy()
0.217
>>> # Using 'none' reduction type.
>>> l = tf.keras.losses.LogCosh(
... reduction=tf.keras.losses.Reduction.NONE)
>>> l(y_true, y_pred).numpy()
array([0.217, 0.], dtype=float32)
Usage with the `compile()` API:
```python
model.compile(optimizer='sgd', loss=tf.keras.losses.LogCosh())
```"
4867,KLDivergence,tensorflow/tensorflow/python/keras/losses.py,1033,class,"Computes Kullback-Leibler divergence loss between `y_true` and `y_pred`.
`loss = y_true * log(y_true / y_pred)`
See: https://en.wikipedia.org/wiki/Kullback%E2%80%93Leibler_divergence
Standalone usage:
>>> y_true = [[0, 1], [0, 0]]
>>> y_pred = [[0.6, 0.4], [0.4, 0.6]]
>>> # Using 'auto'/'sum_over_batch_size' reduction type.
>>> kl = tf.keras.losses.KLDivergence()
>>> kl(y_true, y_pred).numpy()
0.458
>>> # Calling with 'sample_weight'.
>>> kl(y_true, y_pred, sample_weight=[0.8, 0.2]).numpy()
0.366
>>> # Using 'sum' reduction type.
>>> kl = tf.keras.losses.KLDivergence(
... reduction=tf.keras.losses.Reduction.SUM)
>>> kl(y_true, y_pred).numpy()
0.916
>>> # Using 'none' reduction type.
>>> kl = tf.keras.losses.KLDivergence(
... reduction=tf.keras.losses.Reduction.NONE)
>>> kl(y_true, y_pred).numpy()
array([0.916, -3.08e-06], dtype=float32)
Usage with the `compile()` API:
```python
model.compile(optimizer='sgd', loss=tf.keras.losses.KLDivergence())
```"
4868,Huber,tensorflow/tensorflow/python/keras/losses.py,1094,class,"Computes the Huber loss between `y_true` and `y_pred`.
For each value x in `error = y_true - y_pred`:
```
loss = 0.5 * x^2 if |x| <= d
loss = 0.5 * d^2 + d * (|x| - d) if |x| > d
```
where d is `delta`. See: https://en.wikipedia.org/wiki/Huber_loss
Standalone usage:
>>> y_true = [[0, 1], [0, 0]]
>>> y_pred = [[0.6, 0.4], [0.4, 0.6]]
>>> # Using 'auto'/'sum_over_batch_size' reduction type.
>>> h = tf.keras.losses.Huber()
>>> h(y_true, y_pred).numpy()
0.155
>>> # Calling with 'sample_weight'.
>>> h(y_true, y_pred, sample_weight=[1, 0]).numpy()
0.09
>>> # Using 'sum' reduction type.
>>> h = tf.keras.losses.Huber(
... reduction=tf.keras.losses.Reduction.SUM)
>>> h(y_true, y_pred).numpy()
0.31
>>> # Using 'none' reduction type.
>>> h = tf.keras.losses.Huber(
... reduction=tf.keras.losses.Reduction.NONE)
>>> h(y_true, y_pred).numpy()
array([0.18, 0.13], dtype=float32)
Usage with the `compile()` API:
```python
model.compile(optimizer='sgd', loss=tf.keras.losses.Huber())
```"
4869,mean_squared_error,tensorflow/tensorflow/python/keras/losses.py,1168,function,"Computes the mean squared error between labels and predictions.
After computing the squared distance between the inputs, the mean value over
the last dimension is returned.
`loss = mean(square(y_true - y_pred), axis=-1)`
Standalone usage:
>>> y_true = np.random.randint(0, 2, size=(2, 3))
>>> y_pred = np.random.random(size=(2, 3))
>>> loss = tf.keras.losses.mean_squared_error(y_true, y_pred)
>>> assert loss.shape == (2,)
>>> assert np.array_equal(
... loss.numpy(), np.mean(np.square(y_true - y_pred), axis=-1))
Args:
y_true: Ground truth values. shape = `[batch_size, d0, .. dN]`.
y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`.
Returns:
Mean squared error values. shape = `[batch_size, d0, .. dN-1]`."
4870,mean_absolute_error,tensorflow/tensorflow/python/keras/losses.py,1204,function,"Computes the mean absolute error between labels and predictions.
`loss = mean(abs(y_true - y_pred), axis=-1)`
Standalone usage:
>>> y_true = np.random.randint(0, 2, size=(2, 3))
>>> y_pred = np.random.random(size=(2, 3))
>>> loss = tf.keras.losses.mean_absolute_error(y_true, y_pred)
>>> assert loss.shape == (2,)
>>> assert np.array_equal(
... loss.numpy(), np.mean(np.abs(y_true - y_pred), axis=-1))
Args:
y_true: Ground truth values. shape = `[batch_size, d0, .. dN]`.
y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`.
Returns:
Mean absolute error values. shape = `[batch_size, d0, .. dN-1]`."
4871,mean_absolute_percentage_error,tensorflow/tensorflow/python/keras/losses.py,1237,function,"Computes the mean absolute percentage error between `y_true` and `y_pred`.
`loss = 100 * mean(abs((y_true - y_pred) / y_true), axis=-1)`
Standalone usage:
>>> y_true = np.random.random(size=(2, 3))
>>> y_true = np.maximum(y_true, 1e-7) # Prevent division by zero
>>> y_pred = np.random.random(size=(2, 3))
>>> loss = tf.keras.losses.mean_absolute_percentage_error(y_true, y_pred)
>>> assert loss.shape == (2,)
>>> assert np.array_equal(
... loss.numpy(),
... 100. * np.mean(np.abs((y_true - y_pred) / y_true), axis=-1))
Args:
y_true: Ground truth values. shape = `[batch_size, d0, .. dN]`.
y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`.
Returns:
Mean absolute percentage error values. shape = `[batch_size, d0, .. dN-1]`."
4872,mean_squared_logarithmic_error,tensorflow/tensorflow/python/keras/losses.py,1274,function,"Computes the mean squared logarithmic error between `y_true` and `y_pred`.
`loss = mean(square(log(y_true + 1) - log(y_pred + 1)), axis=-1)`
Standalone usage:
>>> y_true = np.random.randint(0, 2, size=(2, 3))
>>> y_pred = np.random.random(size=(2, 3))
>>> loss = tf.keras.losses.mean_squared_logarithmic_error(y_true, y_pred)
>>> assert loss.shape == (2,)
>>> y_true = np.maximum(y_true, 1e-7)
>>> y_pred = np.maximum(y_pred, 1e-7)
>>> assert np.array_equal(
... loss.numpy(),
... np.mean(
... np.square(np.log(y_true + 1.) - np.log(y_pred + 1.)), axis=-1))
Args:
y_true: Ground truth values. shape = `[batch_size, d0, .. dN]`.
y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`.
Returns:
Mean squared logarithmic error values. shape = `[batch_size, d0, .. dN-1]`."
4873,squared_hinge,tensorflow/tensorflow/python/keras/losses.py,1323,function,"Computes the squared hinge loss between `y_true` and `y_pred`.
`loss = mean(square(maximum(1 - y_true * y_pred, 0)), axis=-1)`
Standalone usage:
>>> y_true = np.random.choice([-1, 1], size=(2, 3))
>>> y_pred = np.random.random(size=(2, 3))
>>> loss = tf.keras.losses.squared_hinge(y_true, y_pred)
>>> assert loss.shape == (2,)
>>> assert np.array_equal(
... loss.numpy(),
... np.mean(np.square(np.maximum(1. - y_true * y_pred, 0.)), axis=-1))
Args:
y_true: The ground truth values. `y_true` values are expected to be -1 or 1.
If binary (0 or 1) labels are provided we will convert them to -1 or 1.
shape = `[batch_size, d0, .. dN]`.
y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`.
Returns:
Squared hinge loss values. shape = `[batch_size, d0, .. dN-1]`."
4874,hinge,tensorflow/tensorflow/python/keras/losses.py,1356,function,"Computes the hinge loss between `y_true` and `y_pred`.
`loss = mean(maximum(1 - y_true * y_pred, 0), axis=-1)`
Standalone usage:
>>> y_true = np.random.choice([-1, 1], size=(2, 3))
>>> y_pred = np.random.random(size=(2, 3))
>>> loss = tf.keras.losses.hinge(y_true, y_pred)
>>> assert loss.shape == (2,)
>>> assert np.array_equal(
... loss.numpy(),
... np.mean(np.maximum(1. - y_true * y_pred, 0.), axis=-1))
Args:
y_true: The ground truth values. `y_true` values are expected to be -1 or 1.
If binary (0 or 1) labels are provided they will be converted to -1 or 1.
shape = `[batch_size, d0, .. dN]`.
y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`.
Returns:
Hinge loss values. shape = `[batch_size, d0, .. dN-1]`."
4875,categorical_hinge,tensorflow/tensorflow/python/keras/losses.py,1388,function,"Computes the categorical hinge loss between `y_true` and `y_pred`.
`loss = maximum(neg - pos + 1, 0)`
where `neg=maximum((1-y_true)*y_pred) and pos=sum(y_true*y_pred)`
Standalone usage:
>>> y_true = np.random.randint(0, 3, size=(2,))
>>> y_true = tf.keras.utils.to_categorical(y_true, num_classes=3)
>>> y_pred = np.random.random(size=(2, 3))
>>> loss = tf.keras.losses.categorical_hinge(y_true, y_pred)
>>> assert loss.shape == (2,)
>>> pos = np.sum(y_true * y_pred, axis=-1)
>>> neg = np.amax((1. - y_true) * y_pred, axis=-1)
>>> assert np.array_equal(loss.numpy(), np.maximum(0., neg - pos + 1.))
Args:
y_true: The ground truth values. `y_true` values are expected to be 0 or 1.
y_pred: The predicted values.
Returns:
Categorical hinge loss values."
4876,huber,tensorflow/tensorflow/python/keras/losses.py,1422,function,"Computes Huber loss value.
For each value x in `error = y_true - y_pred`:
```
loss = 0.5 * x^2 if |x| <= d
loss = 0.5 * d^2 + d * (|x| - d) if |x| > d
```
where d is `delta`. See: https://en.wikipedia.org/wiki/Huber_loss
Args:
y_true: tensor of true targets.
y_pred: tensor of predicted targets.
delta: A float, the point where the Huber loss function changes from a
quadratic to linear.
Returns:
Tensor with one scalar loss entry per sample."
4877,log_cosh,tensorflow/tensorflow/python/keras/losses.py,1457,function,"Logarithm of the hyperbolic cosine of the prediction error.
`log(cosh(x))` is approximately equal to `(x ** 2) / 2` for small `x` and
to `abs(x) - log(2)` for large `x`. This means that 'logcosh' works mostly
like the mean squared error, but will not be so strongly affected by the
occasional wildly incorrect prediction.
Standalone usage:
>>> y_true = np.random.random(size=(2, 3))
>>> y_pred = np.random.random(size=(2, 3))
>>> loss = tf.keras.losses.logcosh(y_true, y_pred)
>>> assert loss.shape == (2,)
>>> x = y_pred - y_true
>>> assert np.allclose(
... loss.numpy(),
... np.mean(x + np.log(np.exp(-2. * x) + 1.) - math_ops.log(2.), axis=-1),
... atol=1e-5)
Args:
y_true: Ground truth values. shape = `[batch_size, d0, .. dN]`.
y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`.
Returns:
Logcosh error values. shape = `[batch_size, d0, .. dN-1]`."
4878,categorical_crossentropy,tensorflow/tensorflow/python/keras/losses.py,1496,function,"Computes the categorical crossentropy loss.
Standalone usage:
>>> y_true = [[0, 1, 0], [0, 0, 1]]
>>> y_pred = [[0.05, 0.95, 0], [0.1, 0.8, 0.1]]
>>> loss = tf.keras.losses.categorical_crossentropy(y_true, y_pred)
>>> assert loss.shape == (2,)
>>> loss.numpy()
array([0.0513, 2.303], dtype=float32)
Args:
y_true: Tensor of one-hot true targets.
y_pred: Tensor of predicted targets.
from_logits: Whether `y_pred` is expected to be a logits tensor. By default,
we assume that `y_pred` encodes a probability distribution.
label_smoothing: Float in [0, 1]. If > `0` then smooth the labels.
Returns:
Categorical crossentropy loss value."
4879,sparse_categorical_crossentropy,tensorflow/tensorflow/python/keras/losses.py,1537,function,"Computes the sparse categorical crossentropy loss.
Standalone usage:
>>> y_true = [1, 2]
>>> y_pred = [[0.05, 0.95, 0], [0.1, 0.8, 0.1]]
>>> loss = tf.keras.losses.sparse_categorical_crossentropy(y_true, y_pred)
>>> assert loss.shape == (2,)
>>> loss.numpy()
array([0.0513, 2.303], dtype=float32)
Args:
y_true: Ground truth values.
y_pred: The predicted values.
from_logits: Whether `y_pred` is expected to be a logits tensor. By default,
we assume that `y_pred` encodes a probability distribution.
axis: (Optional) Defaults to -1. The dimension along which the entropy is
computed.
Returns:
Sparse categorical crossentropy loss value."
4880,binary_crossentropy,tensorflow/tensorflow/python/keras/losses.py,1569,function,"Computes the binary crossentropy loss.
Standalone usage:
>>> y_true = [[0, 1], [0, 0]]
>>> y_pred = [[0.6, 0.4], [0.4, 0.6]]
>>> loss = tf.keras.losses.binary_crossentropy(y_true, y_pred)
>>> assert loss.shape == (2,)
>>> loss.numpy()
array([0.916 , 0.714], dtype=float32)
Args:
y_true: Ground truth values. shape = `[batch_size, d0, .. dN]`.
y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`.
from_logits: Whether `y_pred` is expected to be a logits tensor. By default,
we assume that `y_pred` encodes a probability distribution.
label_smoothing: Float in [0, 1]. If > `0` then smooth the labels.
Returns:
Binary crossentropy loss value. shape = `[batch_size, d0, .. dN-1]`."
4881,kl_divergence,tensorflow/tensorflow/python/keras/losses.py,1613,function,"Computes Kullback-Leibler divergence loss between `y_true` and `y_pred`.
`loss = y_true * log(y_true / y_pred)`
See: https://en.wikipedia.org/wiki/Kullback%E2%80%93Leibler_divergence
Standalone usage:
>>> y_true = np.random.randint(0, 2, size=(2, 3)).astype(np.float64)
>>> y_pred = np.random.random(size=(2, 3))
>>> loss = tf.keras.losses.kullback_leibler_divergence(y_true, y_pred)
>>> assert loss.shape == (2,)
>>> y_true = tf.keras.backend.clip(y_true, 1e-7, 1)
>>> y_pred = tf.keras.backend.clip(y_pred, 1e-7, 1)
>>> assert np.array_equal(
... loss.numpy(), np.sum(y_true * np.log(y_true / y_pred), axis=-1))
Args:
y_true: Tensor of true targets.
y_pred: Tensor of predicted targets.
Returns:
A `Tensor` with loss.
Raises:
TypeError: If `y_true` cannot be cast to the `y_pred.dtype`."
4882,poisson,tensorflow/tensorflow/python/keras/losses.py,1650,function,"Computes the Poisson loss between y_true and y_pred.
The Poisson loss is the mean of the elements of the `Tensor`
`y_pred - y_true * log(y_pred)`.
Standalone usage:
>>> y_true = np.random.randint(0, 2, size=(2, 3))
>>> y_pred = np.random.random(size=(2, 3))
>>> loss = tf.keras.losses.poisson(y_true, y_pred)
>>> assert loss.shape == (2,)
>>> y_pred = y_pred + 1e-7
>>> assert np.allclose(
... loss.numpy(), np.mean(y_pred - y_true * np.log(y_pred), axis=-1),
... atol=1e-5)
Args:
y_true: Ground truth values. shape = `[batch_size, d0, .. dN]`.
y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`.
Returns:
Poisson loss value. shape = `[batch_size, d0, .. dN-1]`.
Raises:
InvalidArgumentError: If `y_true` and `y_pred` have incompatible shapes."
4883,cosine_similarity,tensorflow/tensorflow/python/keras/losses.py,1692,function,"Computes the cosine similarity between labels and predictions.
Note that it is a number between -1 and 1. When it is a negative number
between -1 and 0, 0 indicates orthogonality and values closer to -1
indicate greater similarity. The values closer to 1 indicate greater
dissimilarity. This makes it usable as a loss function in a setting
where you try to maximize the proximity between predictions and
targets. If either `y_true` or `y_pred` is a zero vector, cosine
similarity will be 0 regardless of the proximity between predictions
and targets.
`loss = -sum(l2_norm(y_true) * l2_norm(y_pred))`
Standalone usage:
>>> y_true = [[0., 1.], [1., 1.], [1., 1.]]
>>> y_pred = [[1., 0.], [1., 1.], [-1., -1.]]
>>> loss = tf.keras.losses.cosine_similarity(y_true, y_pred, axis=1)
>>> loss.numpy()
array([-0., -0.999, 0.999], dtype=float32)
Args:
y_true: Tensor of true targets.
y_pred: Tensor of predicted targets.
axis: Axis along which to determine similarity.
Returns:
Cosine similarity tensor."
4884,CosineSimilarity,tensorflow/tensorflow/python/keras/losses.py,1728,class,"Computes the cosine similarity between labels and predictions.
Note that it is a negative quantity between -1 and 0, where 0 indicates
orthogonality and values closer to -1 indicate greater similarity. This makes
it usable as a loss function in a setting where you try to maximize the
proximity between predictions and targets. If either `y_true` or `y_pred`
is a zero vector, cosine similarity will be 0 regardless of the proximity
between predictions and targets.
`loss = -sum(l2_norm(y_true) * l2_norm(y_pred))`
Standalone usage:
>>> y_true = [[0., 1.], [1., 1.]]
>>> y_pred = [[1., 0.], [1., 1.]]
>>> # Using 'auto'/'sum_over_batch_size' reduction type.
>>> cosine_loss = tf.keras.losses.CosineSimilarity(axis=1)
>>> # l2_norm(y_true) = [[0., 1.], [1./1.414], 1./1.414]]]
>>> # l2_norm(y_pred) = [[1., 0.], [1./1.414], 1./1.414]]]
>>> # l2_norm(y_true) . l2_norm(y_pred) = [[0., 0.], [0.5, 0.5]]
>>> # loss = mean(sum(l2_norm(y_true) . l2_norm(y_pred), axis=1))
>>> # = -((0. + 0.) + (0.5 + 0.5)) / 2
>>> cosine_loss(y_true, y_pred).numpy()
-0.5
>>> # Calling with 'sample_weight'.
>>> cosine_loss(y_true, y_pred, sample_weight=[0.8, 0.2]).numpy()
-0.0999
>>> # Using 'sum' reduction type.
>>> cosine_loss = tf.keras.losses.CosineSimilarity(axis=1,
... reduction=tf.keras.losses.Reduction.SUM)
>>> cosine_loss(y_true, y_pred).numpy()
-0.999
>>> # Using 'none' reduction type.
>>> cosine_loss = tf.keras.losses.CosineSimilarity(axis=1,
... reduction=tf.keras.losses.Reduction.NONE)
>>> cosine_loss(y_true, y_pred).numpy()
array([-0., -0.999], dtype=float32)
Usage with the `compile()` API:
```python
model.compile(optimizer='sgd', loss=tf.keras.losses.CosineSimilarity(axis=1))
```
Args:
axis: (Optional) Defaults to -1. The dimension along which the cosine
similarity is computed.
reduction: (Optional) Type of `tf.keras.losses.Reduction` to apply to loss.
Default value is `AUTO`. `AUTO` indicates that the reduction option will
be determined by the usage context. For almost all cases this defaults to
`SUM_OVER_BATCH_SIZE`. When used with `tf.distribute.Strategy`, outside of
built-in training loops such as `tf.keras` `compile` and `fit`, using
`AUTO` or `SUM_OVER_BATCH_SIZE` will raise an error. Please see this
custom training [tutorial]
(https://www.tensorflow.org/tutorials/distribute/custom_training) for more
details.
name: Optional name for the op."
4885,is_categorical_crossentropy,tensorflow/tensorflow/python/keras/losses.py,1811,function,
4886,serialize,tensorflow/tensorflow/python/keras/losses.py,1822,function,"Serializes loss function or `Loss` instance.
Arguments:
loss: A Keras `Loss` instance or a loss function.
Returns:
Loss configuration dictionary."
4887,deserialize,tensorflow/tensorflow/python/keras/losses.py,1835,function,"Deserializes a serialized loss class/function instance.
Arguments:
name: Loss configuration.
custom_objects: Optional dictionary mapping names (strings) to custom
objects (classes and functions) to be considered during deserialization.
Returns:
A Keras `Loss` instance or a loss function."
4888,get,tensorflow/tensorflow/python/keras/losses.py,1854,function,"Retrieves a Keras loss as a `function`/`Loss` class instance.
The `identifier` may be the string name of a loss function or `Loss` class.
>>> loss = tf.keras.losses.get(""categorical_crossentropy"")
>>> type(loss)
<class 'function'>
>>> loss = tf.keras.losses.get(""CategoricalCrossentropy"")
>>> type(loss)
<class '...tensorflow.python.keras.losses.CategoricalCrossentropy'>
You can also specify `config` of the loss to this function by passing dict
containing `class_name` and `config` as an identifier. Also note that the
`class_name` must map to a `Loss` class
>>> identifier = {""class_name"": ""CategoricalCrossentropy"",
... ""config"": {""from_logits"": True}}
>>> loss = tf.keras.losses.get(identifier)
>>> type(loss)
<class '...tensorflow.python.keras.losses.CategoricalCrossentropy'>
Arguments:
identifier: A loss identifier. One of None or string name of a loss
function/class or loss configuration dictionary or a loss function or a
loss class instance
Returns:
A Keras loss as a `function`/ `Loss` class instance.
Raises:
ValueError: If `identifier` cannot be interpreted."
4889,BinaryTruePositivesViaControlFlow,tensorflow/tensorflow/python/keras/losses_test.py,1631,class,
4890,call,tensorflow/tensorflow/python/keras/losses_test.py,1636,method,
4891,Metric,tensorflow/tensorflow/python/keras/metrics.py,82,class,"Encapsulates metric logic and state.
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
**kwargs: Additional layer keywords arguments.
Standalone usage:
```python
m = SomeMetric(...)
for input in ...:
m.update_state(input)
print('Final result: ', m.result().numpy())
```
Usage with `compile()` API:
```python
model = tf.keras.Sequential()
model.add(tf.keras.layers.Dense(64, activation='relu'))
model.add(tf.keras.layers.Dense(64, activation='relu'))
model.add(tf.keras.layers.Dense(10, activation='softmax'))
model.compile(optimizer=tf.keras.optimizers.RMSprop(0.01),
loss=tf.keras.losses.CategoricalCrossentropy(),
metrics=[tf.keras.metrics.CategoricalAccuracy()])
data = np.random.random((1000, 32))
labels = np.random.random((1000, 10))
dataset = tf.data.Dataset.from_tensor_slices((data, labels))
dataset = dataset.batch(32)
model.fit(dataset, epochs=10)
```
To be implemented by subclasses:
* `__init__()`: All state variables should be created in this method by
calling `self.add_weight()` like: `self.var = self.add_weight(...)`
* `update_state()`: Has all updates to the state variables like:
self.var.assign_add(...).
* `result()`: Computes and returns a value for the metric
from the state variables.
Example subclass implementation:
```python
class BinaryTruePositives(tf.keras.metrics.Metric):
def __init__(self, name='binary_true_positives', **kwargs):
super(BinaryTruePositives, self).__init__(name=name, **kwargs)
self.true_positives = self.add_weight(name='tp', initializer='zeros')
def update_state(self, y_true, y_pred, sample_weight=None):
y_true = tf.cast(y_true, tf.bool)
y_pred = tf.cast(y_pred, tf.bool)
values = tf.logical_and(tf.equal(y_true, True), tf.equal(y_pred, True))
values = tf.cast(values, self.dtype)
if sample_weight is not None:
sample_weight = tf.cast(sample_weight, self.dtype)
sample_weight = tf.broadcast_to(sample_weight, values.shape)
values = tf.multiply(values, sample_weight)
self.true_positives.assign_add(tf.reduce_sum(values))
def result(self):
return self.true_positives
```"
4892,dtype,tensorflow/tensorflow/python/keras/metrics.py,240,method,
4893,get_config,tensorflow/tensorflow/python/keras/metrics.py,243,method,Returns the serializable config of the metric.
4894,reset_states,tensorflow/tensorflow/python/keras/metrics.py,247,method,"Resets all of the metric state variables.
This function is called between epochs/steps,
when a metric is evaluated during training."
4895,update_state,tensorflow/tensorflow/python/keras/metrics.py,256,method,"Accumulates statistics for the metric.
Note: This function is executed as a graph function in graph mode.
This means:
a) Operations on the same resource are executed in textual order.
This should make it easier to do things like add the updated
value of a variable to another, for example.
b) You don't need to worry about collecting the update ops to execute.
All update ops added to the graph by this function will be executed.
As a result, code should generally work the same way with graph or
eager execution.
Args:
*args:
**kwargs: A mini-batch of inputs to the Metric."
4896,result,tensorflow/tensorflow/python/keras/metrics.py,276,method,"Computes and returns the metric value tensor.
Result computation is an idempotent operation that simply calculates the
metric value using the state variables."
4897,add_weight,tensorflow/tensorflow/python/keras/metrics.py,286,method,Adds state variable. Only for use by subclasses.
4898,result_fn,tensorflow/tensorflow/python/keras/metrics.py,189,method,
4899,replica_local_fn,tensorflow/tensorflow/python/keras/metrics.py,210,method,Updates the state of the metric in a replica-local context.
4900,update_state_fn,tensorflow/tensorflow/python/keras/metrics.py,174,method,
4901,Reduce,tensorflow/tensorflow/python/keras/metrics.py,323,class,"Encapsulates metrics that perform a reduce operation on the values.
Args:
reduction: a `tf.keras.metrics.Reduction` enum value.
name: string name of the metric instance.
dtype: (Optional) data type of the metric result."
4902,update_state,tensorflow/tensorflow/python/keras/metrics.py,342,method,"Accumulates statistics for computing the metric.
Args:
values: Per-example value.
sample_weight: Optional weighting of each example. Defaults to 1.
Returns:
Update op."
4903,result,tensorflow/tensorflow/python/keras/metrics.py,400,method,
4904,Sum,tensorflow/tensorflow/python/keras/metrics.py,414,class,"Computes the (weighted) sum of the given values.
For example, if values is [1, 3, 5, 7] then the sum is 16.
If the weights were specified as [1, 1, 0, 0] then the sum would be 4.
This metric creates one variable, `total`, that is used to compute the sum of
`values`. This is ultimately returned as `sum`.
If `sample_weight` is `None`, weights default to 1. Use `sample_weight` of 0
to mask values.
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.Sum()
>>> m.update_state([1, 3, 5, 7])
>>> m.result().numpy()
16.0
Usage with `compile()` API:
```python
model.add_metric(tf.keras.metrics.Sum(name='sum_1')(outputs))
model.compile(optimizer='sgd', loss='mse')
```"
4905,Mean,tensorflow/tensorflow/python/keras/metrics.py,451,class,"Computes the (weighted) mean of the given values.
For example, if values is [1, 3, 5, 7] then the mean is 4.
If the weights were specified as [1, 1, 0, 0] then the mean would be 2.
This metric creates two variables, `total` and `count` that are used to
compute the average of `values`. This average is ultimately returned as `mean`
which is an idempotent operation that simply divides `total` by `count`.
If `sample_weight` is `None`, weights default to 1.
Use `sample_weight` of 0 to mask values.
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.Mean()
>>> m.update_state([1, 3, 5, 7])
>>> m.result().numpy()
4.0
>>> m.reset_states()
>>> m.update_state([1, 3, 5, 7], sample_weight=[1, 1, 0, 0])
>>> m.result().numpy()
2.0
Usage with `compile()` API:
```python
model.add_metric(tf.keras.metrics.Mean(name='mean_1')(outputs))
model.compile(optimizer='sgd', loss='mse')
```"
4906,MeanRelativeError,tensorflow/tensorflow/python/keras/metrics.py,493,class,"Computes the mean relative error by normalizing with the given values.
This metric creates two local variables, `total` and `count` that are used to
compute the mean relative error. This is weighted by `sample_weight`, and
it is ultimately returned as `mean_relative_error`:
an idempotent operation that simply divides `total` by `count`.
If `sample_weight` is `None`, weights default to 1.
Use `sample_weight` of 0 to mask values.
Args:
normalizer: The normalizer values with same shape as predictions.
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.MeanRelativeError(normalizer=[1, 3, 2, 3])
>>> m.update_state([1, 3, 2, 3], [2, 4, 6, 8])
>>> # metric = mean(|y_pred - y_true| / normalizer)
>>> # = mean([1, 1, 4, 5] / [1, 3, 2, 3]) = mean([1, 1/3, 2, 5/3])
>>> # = 5/4 = 1.25
>>> m.result().numpy()
1.25
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.MeanRelativeError(normalizer=[1, 3])])
```"
4907,update_state,tensorflow/tensorflow/python/keras/metrics.py,535,method,"Accumulates metric statistics.
Args:
y_true: The ground truth values.
y_pred: The predicted values.
sample_weight: Optional weighting of each example. Defaults to 1. Can be a
`Tensor` whose rank is either 0, or the same rank as `y_true`, and must
be broadcastable to `y_true`.
Returns:
Update op."
4908,get_config,tensorflow/tensorflow/python/keras/metrics.py,565,method,
4909,MeanMetricWrapper,tensorflow/tensorflow/python/keras/metrics.py,572,class,"Wraps a stateless metric function with the Mean metric.
Args:
fn: The metric function to wrap, with signature `fn(y_true, y_pred,
**kwargs)`.
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
**kwargs: The keyword arguments that are passed on to `fn`."
4910,update_state,tensorflow/tensorflow/python/keras/metrics.py,588,method,"Accumulates metric statistics.
`y_true` and `y_pred` should have the same shape.
Args:
y_true: Ground truth values. shape = `[batch_size, d0, .. dN]`.
y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`.
sample_weight: Optional `sample_weight` acts as a
coefficient for the metric. If a scalar is provided, then the metric is
simply scaled by the given value. If `sample_weight` is a tensor of size
`[batch_size]`, then the metric for each sample of the batch is rescaled
by the corresponding element in the `sample_weight` vector. If the shape
of `sample_weight` is `[batch_size, d0, .. dN-1]` (or can be broadcasted
to this shape), then each metric element of `y_pred` is scaled by the
corresponding value of `sample_weight`. (Note on `dN-1`: all metric
functions reduce by 1 dimension, usually the last axis (-1)).
Returns:
Update op."
4911,get_config,tensorflow/tensorflow/python/keras/metrics.py,622,method,
4912,from_config,tensorflow/tensorflow/python/keras/metrics.py,636,method,
4913,Accuracy,tensorflow/tensorflow/python/keras/metrics.py,646,class,"Calculates how often predictions equals labels.
This metric creates two local variables, `total` and `count` that are used to
compute the frequency with which `y_pred` matches `y_true`. This frequency is
ultimately returned as `binary accuracy`: an idempotent operation that simply
divides `total` by `count`.
If `sample_weight` is `None`, weights default to 1.
Use `sample_weight` of 0 to mask values.
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.Accuracy()
>>> m.update_state([[1], [2], [3], [4]], [[0], [2], [3], [4]])
>>> m.result().numpy()
0.75
>>> m.reset_states()
>>> m.update_state([[1], [2], [3], [4]], [[0], [2], [3], [4]],
... sample_weight=[1, 1, 0, 0])
>>> m.result().numpy()
0.5
Usage with `compile()` API:
```python
model.compile(optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.Accuracy()])
```"
4914,BinaryAccuracy,tensorflow/tensorflow/python/keras/metrics.py,688,class,"Calculates how often predictions matches binary labels.
This metric creates two local variables, `total` and `count` that are used to
compute the frequency with which `y_pred` matches `y_true`. This frequency is
ultimately returned as `binary accuracy`: an idempotent operation that simply
divides `total` by `count`.
If `sample_weight` is `None`, weights default to 1.
Use `sample_weight` of 0 to mask values.
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
threshold: (Optional) Float representing the threshold for deciding
whether prediction values are 1 or 0.
Standalone usage:
>>> m = tf.keras.metrics.BinaryAccuracy()
>>> m.update_state([[1], [1], [0], [0]], [[0.98], [1], [0], [0.6]])
>>> m.result().numpy()
0.75
>>> m.reset_states()
>>> m.update_state([[1], [1], [0], [0]], [[0.98], [1], [0], [0.6]],
... sample_weight=[1, 0, 0, 1])
>>> m.result().numpy()
0.5
Usage with `compile()` API:
```python
model.compile(optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.BinaryAccuracy()])
```"
4915,CategoricalAccuracy,tensorflow/tensorflow/python/keras/metrics.py,733,class,"Calculates how often predictions matches one-hot labels.
You can provide logits of classes as `y_pred`, since argmax of
logits and probabilities are same.
This metric creates two local variables, `total` and `count` that are used to
compute the frequency with which `y_pred` matches `y_true`. This frequency is
ultimately returned as `categorical accuracy`: an idempotent operation that
simply divides `total` by `count`.
`y_pred` and `y_true` should be passed in as vectors of probabilities, rather
than as labels. If necessary, use `tf.one_hot` to expand `y_true` as a vector.
If `sample_weight` is `None`, weights default to 1.
Use `sample_weight` of 0 to mask values.
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.CategoricalAccuracy()
>>> m.update_state([[0, 0, 1], [0, 1, 0]], [[0.1, 0.9, 0.8],
... [0.05, 0.95, 0]])
>>> m.result().numpy()
0.5
>>> m.reset_states()
>>> m.update_state([[0, 0, 1], [0, 1, 0]], [[0.1, 0.9, 0.8],
... [0.05, 0.95, 0]],
... sample_weight=[0.7, 0.3])
>>> m.result().numpy()
0.3
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.CategoricalAccuracy()])
```"
4916,SparseCategoricalAccuracy,tensorflow/tensorflow/python/keras/metrics.py,785,class,"Calculates how often predictions matches integer labels.
```python
acc = np.dot(sample_weight, np.equal(y_true, np.argmax(y_pred, axis=1))
```
You can provide logits of classes as `y_pred`, since argmax of
logits and probabilities are same.
This metric creates two local variables, `total` and `count` that are used to
compute the frequency with which `y_pred` matches `y_true`. This frequency is
ultimately returned as `sparse categorical accuracy`: an idempotent operation
that simply divides `total` by `count`.
If `sample_weight` is `None`, weights default to 1.
Use `sample_weight` of 0 to mask values.
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.SparseCategoricalAccuracy()
>>> m.update_state([[2], [1]], [[0.1, 0.6, 0.3], [0.05, 0.95, 0]])
>>> m.result().numpy()
0.5
>>> m.reset_states()
>>> m.update_state([[2], [1]], [[0.1, 0.6, 0.3], [0.05, 0.95, 0]],
... sample_weight=[0.7, 0.3])
>>> m.result().numpy()
0.3
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.SparseCategoricalAccuracy()])
```"
4917,TopKCategoricalAccuracy,tensorflow/tensorflow/python/keras/metrics.py,836,class,"Computes how often targets are in the top `K` predictions.
Args:
k: (Optional) Number of top elements to look at for computing accuracy.
Defaults to 5.
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.TopKCategoricalAccuracy(k=1)
>>> m.update_state([[0, 0, 1], [0, 1, 0]],
... [[0.1, 0.9, 0.8], [0.05, 0.95, 0]])
>>> m.result().numpy()
0.5
>>> m.reset_states()
>>> m.update_state([[0, 0, 1], [0, 1, 0]],
... [[0.1, 0.9, 0.8], [0.05, 0.95, 0]],
... sample_weight=[0.7, 0.3])
>>> m.result().numpy()
0.3
Usage with `compile()` API:
```python
model.compile(optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.TopKCategoricalAccuracy()])
```"
4918,SparseTopKCategoricalAccuracy,tensorflow/tensorflow/python/keras/metrics.py,875,class,"Computes how often integer targets are in the top `K` predictions.
Args:
k: (Optional) Number of top elements to look at for computing accuracy.
Defaults to 5.
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.SparseTopKCategoricalAccuracy(k=1)
>>> m.update_state([2, 1], [[0.1, 0.9, 0.8], [0.05, 0.95, 0]])
>>> m.result().numpy()
0.5
>>> m.reset_states()
>>> m.update_state([2, 1], [[0.1, 0.9, 0.8], [0.05, 0.95, 0]],
... sample_weight=[0.7, 0.3])
>>> m.result().numpy()
0.3
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.SparseTopKCategoricalAccuracy()])
```"
4919,FalsePositives,tensorflow/tensorflow/python/keras/metrics.py,980,class,"Calculates the number of false positives.
If `sample_weight` is given, calculates the sum of the weights of
false positives. This metric creates one local variable, `accumulator`
that is used to keep track of the number of false positives.
If `sample_weight` is `None`, weights default to 1.
Use `sample_weight` of 0 to mask values.
Args:
thresholds: (Optional) Defaults to 0.5. A float value or a python
list/tuple of float threshold values in [0, 1]. A threshold is compared
with prediction values to determine the truth value of predictions
(i.e., above the threshold is `true`, below is `false`). One metric
value is generated for each threshold value.
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.FalsePositives()
>>> m.update_state([0, 1, 0, 0], [0, 0, 1, 1])
>>> m.result().numpy()
2.0
>>> m.reset_states()
>>> m.update_state([0, 1, 0, 0], [0, 0, 1, 1], sample_weight=[0, 0, 1, 0])
>>> m.result().numpy()
1.0
Usage with `compile()` API:
```python
model.compile(optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.FalsePositives()])
```"
4920,FalseNegatives,tensorflow/tensorflow/python/keras/metrics.py,1029,class,"Calculates the number of false negatives.
If `sample_weight` is given, calculates the sum of the weights of
false negatives. This metric creates one local variable, `accumulator`
that is used to keep track of the number of false negatives.
If `sample_weight` is `None`, weights default to 1.
Use `sample_weight` of 0 to mask values.
Args:
thresholds: (Optional) Defaults to 0.5. A float value or a python
list/tuple of float threshold values in [0, 1]. A threshold is compared
with prediction values to determine the truth value of predictions
(i.e., above the threshold is `true`, below is `false`). One metric
value is generated for each threshold value.
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.FalseNegatives()
>>> m.update_state([0, 1, 1, 1], [0, 1, 0, 0])
>>> m.result().numpy()
2.0
>>> m.reset_states()
>>> m.update_state([0, 1, 1, 1], [0, 1, 0, 0], sample_weight=[0, 0, 1, 0])
>>> m.result().numpy()
1.0
Usage with `compile()` API:
```python
model.compile(optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.FalseNegatives()])
```"
4921,TrueNegatives,tensorflow/tensorflow/python/keras/metrics.py,1078,class,"Calculates the number of true negatives.
If `sample_weight` is given, calculates the sum of the weights of
true negatives. This metric creates one local variable, `accumulator`
that is used to keep track of the number of true negatives.
If `sample_weight` is `None`, weights default to 1.
Use `sample_weight` of 0 to mask values.
Args:
thresholds: (Optional) Defaults to 0.5. A float value or a python
list/tuple of float threshold values in [0, 1]. A threshold is compared
with prediction values to determine the truth value of predictions
(i.e., above the threshold is `true`, below is `false`). One metric
value is generated for each threshold value.
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.TrueNegatives()
>>> m.update_state([0, 1, 0, 0], [1, 1, 0, 0])
>>> m.result().numpy()
2.0
>>> m.reset_states()
>>> m.update_state([0, 1, 0, 0], [1, 1, 0, 0], sample_weight=[0, 0, 1, 0])
>>> m.result().numpy()
1.0
Usage with `compile()` API:
```python
model.compile(optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.TrueNegatives()])
```"
4922,TruePositives,tensorflow/tensorflow/python/keras/metrics.py,1127,class,"Calculates the number of true positives.
If `sample_weight` is given, calculates the sum of the weights of
true positives. This metric creates one local variable, `true_positives`
that is used to keep track of the number of true positives.
If `sample_weight` is `None`, weights default to 1.
Use `sample_weight` of 0 to mask values.
Args:
thresholds: (Optional) Defaults to 0.5. A float value or a python
list/tuple of float threshold values in [0, 1]. A threshold is compared
with prediction values to determine the truth value of predictions
(i.e., above the threshold is `true`, below is `false`). One metric
value is generated for each threshold value.
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.TruePositives()
>>> m.update_state([0, 1, 1, 1], [1, 0, 1, 1])
>>> m.result().numpy()
2.0
>>> m.reset_states()
>>> m.update_state([0, 1, 1, 1], [1, 0, 1, 1], sample_weight=[0, 0, 1, 0])
>>> m.result().numpy()
1.0
Usage with `compile()` API:
```python
model.compile(optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.TruePositives()])
```"
4923,Precision,tensorflow/tensorflow/python/keras/metrics.py,1176,class,"Computes the precision of the predictions with respect to the labels.
The metric creates two local variables, `true_positives` and `false_positives`
that are used to compute the precision. This value is ultimately returned as
`precision`, an idempotent operation that simply divides `true_positives`
by the sum of `true_positives` and `false_positives`.
If `sample_weight` is `None`, weights default to 1.
Use `sample_weight` of 0 to mask values.
If `top_k` is set, we'll calculate precision as how often on average a class
among the top-k classes with the highest predicted values of a batch entry is
correct and can be found in the label for that entry.
If `class_id` is specified, we calculate precision by considering only the
entries in the batch for which `class_id` is above the threshold and/or in the
top-k highest predictions, and computing the fraction of them for which
`class_id` is indeed a correct label.
Args:
thresholds: (Optional) A float value or a python list/tuple of float
threshold values in [0, 1]. A threshold is compared with prediction
values to determine the truth value of predictions (i.e., above the
threshold is `true`, below is `false`). One metric value is generated
for each threshold value. If neither thresholds nor top_k are set, the
default is to calculate precision with `thresholds=0.5`.
top_k: (Optional) Unset by default. An int value specifying the top-k
predictions to consider when calculating precision.
class_id: (Optional) Integer class ID for which we want binary metrics.
This must be in the half-open interval `[0, num_classes)`, where
`num_classes` is the last dimension of predictions.
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.Precision()
>>> m.update_state([0, 1, 1, 1], [1, 0, 1, 1])
>>> m.result().numpy()
0.6666667
>>> m.reset_states()
>>> m.update_state([0, 1, 1, 1], [1, 0, 1, 1], sample_weight=[0, 0, 1, 0])
>>> m.result().numpy()
1.0
>>> # With top_k=2, it will calculate precision over y_true[:2] and y_pred[:2]
>>> m = tf.keras.metrics.Precision(top_k=2)
>>> m.update_state([0, 0, 1, 1], [1, 1, 1, 1])
>>> m.result().numpy()
0.0
>>> # With top_k=4, it will calculate precision over y_true[:4] and y_pred[:4]
>>> m = tf.keras.metrics.Precision(top_k=4)
>>> m.update_state([0, 0, 1, 1], [1, 1, 1, 1])
>>> m.result().numpy()
0.5
Usage with `compile()` API:
```python
model.compile(optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.Precision()])
```"
4924,update_state,tensorflow/tensorflow/python/keras/metrics.py,1267,method,"Accumulates true positive and false positive statistics.
Args:
y_true: The ground truth values, with the same dimensions as `y_pred`.
Will be cast to `bool`.
y_pred: The predicted values. Each element must be in the range `[0, 1]`.
sample_weight: Optional weighting of each example. Defaults to 1. Can be a
`Tensor` whose rank is either 0, or the same rank as `y_true`, and must
be broadcastable to `y_true`.
Returns:
Update op."
4925,result,tensorflow/tensorflow/python/keras/metrics.py,1293,method,
4926,reset_states,tensorflow/tensorflow/python/keras/metrics.py,1298,method,
4927,get_config,tensorflow/tensorflow/python/keras/metrics.py,1303,method,
4928,Recall,tensorflow/tensorflow/python/keras/metrics.py,1314,class,"Computes the recall of the predictions with respect to the labels.
This metric creates two local variables, `true_positives` and
`false_negatives`, that are used to compute the recall. This value is
ultimately returned as `recall`, an idempotent operation that simply divides
`true_positives` by the sum of `true_positives` and `false_negatives`.
If `sample_weight` is `None`, weights default to 1.
Use `sample_weight` of 0 to mask values.
If `top_k` is set, recall will be computed as how often on average a class
among the labels of a batch entry is in the top-k predictions.
If `class_id` is specified, we calculate recall by considering only the
entries in the batch for which `class_id` is in the label, and computing the
fraction of them for which `class_id` is above the threshold and/or in the
top-k predictions.
Args:
thresholds: (Optional) A float value or a python list/tuple of float
threshold values in [0, 1]. A threshold is compared with prediction
values to determine the truth value of predictions (i.e., above the
threshold is `true`, below is `false`). One metric value is generated
for each threshold value. If neither thresholds nor top_k are set, the
default is to calculate recall with `thresholds=0.5`.
top_k: (Optional) Unset by default. An int value specifying the top-k
predictions to consider when calculating recall.
class_id: (Optional) Integer class ID for which we want binary metrics.
This must be in the half-open interval `[0, num_classes)`, where
`num_classes` is the last dimension of predictions.
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.Recall()
>>> m.update_state([0, 1, 1, 1], [1, 0, 1, 1])
>>> m.result().numpy()
0.6666667
>>> m.reset_states()
>>> m.update_state([0, 1, 1, 1], [1, 0, 1, 1], sample_weight=[0, 0, 1, 0])
>>> m.result().numpy()
1.0
Usage with `compile()` API:
```python
model.compile(optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.Recall()])
```"
4929,update_state,tensorflow/tensorflow/python/keras/metrics.py,1392,method,"Accumulates true positive and false negative statistics.
Args:
y_true: The ground truth values, with the same dimensions as `y_pred`.
Will be cast to `bool`.
y_pred: The predicted values. Each element must be in the range `[0, 1]`.
sample_weight: Optional weighting of each example. Defaults to 1. Can be a
`Tensor` whose rank is either 0, or the same rank as `y_true`, and must
be broadcastable to `y_true`.
Returns:
Update op."
4930,result,tensorflow/tensorflow/python/keras/metrics.py,1418,method,
4931,reset_states,tensorflow/tensorflow/python/keras/metrics.py,1423,method,
4932,get_config,tensorflow/tensorflow/python/keras/metrics.py,1428,method,
4933,SensitivitySpecificityBase,tensorflow/tensorflow/python/keras/metrics.py,1439,class,"Abstract base class for computing sensitivity and specificity.
For additional information about specificity and sensitivity, see
[the following](https://en.wikipedia.org/wiki/Sensitivity_and_specificity)."
4934,update_state,tensorflow/tensorflow/python/keras/metrics.py,1476,method,"Accumulates confusion matrix statistics.
Args:
y_true: The ground truth values.
y_pred: The predicted values.
sample_weight: Optional weighting of each example. Defaults to 1. Can be a
`Tensor` whose rank is either 0, or the same rank as `y_true`, and must
be broadcastable to `y_true`.
Returns:
Update op."
4935,reset_states,tensorflow/tensorflow/python/keras/metrics.py,1501,method,
4936,get_max,tensorflow/tensorflow/python/keras/metrics.py,1524,method,
4937,SensitivityAtSpecificity,tensorflow/tensorflow/python/keras/metrics.py,1531,class,"Computes best sensitivity where specificity is >= specified value.
the sensitivity at a given specificity.
`Sensitivity` measures the proportion of actual positives that are correctly
identified as such (tp / (tp + fn)).
`Specificity` measures the proportion of actual negatives that are correctly
identified as such (tn / (tn + fp)).
This metric creates four local variables, `true_positives`, `true_negatives`,
`false_positives` and `false_negatives` that are used to compute the
sensitivity at the given specificity. The threshold for the given specificity
value is computed and used to evaluate the corresponding sensitivity.
If `sample_weight` is `None`, weights default to 1.
Use `sample_weight` of 0 to mask values.
For additional information about specificity and sensitivity, see
[the following](https://en.wikipedia.org/wiki/Sensitivity_and_specificity).
Args:
specificity: A scalar value in range `[0, 1]`.
num_thresholds: (Optional) Defaults to 200. The number of thresholds to
use for matching the given specificity.
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.SensitivityAtSpecificity(0.5)
>>> m.update_state([0, 0, 0, 1, 1], [0, 0.3, 0.8, 0.3, 0.8])
>>> m.result().numpy()
0.5
>>> m.reset_states()
>>> m.update_state([0, 0, 0, 1, 1], [0, 0.3, 0.8, 0.3, 0.8],
... sample_weight=[1, 1, 2, 2, 1])
>>> m.result().numpy()
0.333333
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.SensitivityAtSpecificity()])
```"
4938,result,tensorflow/tensorflow/python/keras/metrics.py,1590,method,
4939,get_config,tensorflow/tensorflow/python/keras/metrics.py,1598,method,
4940,SpecificityAtSensitivity,tensorflow/tensorflow/python/keras/metrics.py,1608,class,"Computes best specificity where sensitivity is >= specified value.
`Sensitivity` measures the proportion of actual positives that are correctly
identified as such (tp / (tp + fn)).
`Specificity` measures the proportion of actual negatives that are correctly
identified as such (tn / (tn + fp)).
This metric creates four local variables, `true_positives`, `true_negatives`,
`false_positives` and `false_negatives` that are used to compute the
specificity at the given sensitivity. The threshold for the given sensitivity
value is computed and used to evaluate the corresponding specificity.
If `sample_weight` is `None`, weights default to 1.
Use `sample_weight` of 0 to mask values.
For additional information about specificity and sensitivity, see
[the following](https://en.wikipedia.org/wiki/Sensitivity_and_specificity).
Args:
sensitivity: A scalar value in range `[0, 1]`.
num_thresholds: (Optional) Defaults to 200. The number of thresholds to
use for matching the given sensitivity.
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.SpecificityAtSensitivity(0.5)
>>> m.update_state([0, 0, 0, 1, 1], [0, 0.3, 0.8, 0.3, 0.8])
>>> m.result().numpy()
0.66666667
>>> m.reset_states()
>>> m.update_state([0, 0, 0, 1, 1], [0, 0.3, 0.8, 0.3, 0.8],
... sample_weight=[1, 1, 2, 2, 2])
>>> m.result().numpy()
0.5
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.SpecificityAtSensitivity()])
```"
4941,result,tensorflow/tensorflow/python/keras/metrics.py,1665,method,
4942,get_config,tensorflow/tensorflow/python/keras/metrics.py,1673,method,
4943,PrecisionAtRecall,tensorflow/tensorflow/python/keras/metrics.py,1683,class,"Computes best precision where recall is >= specified value.
This metric creates four local variables, `true_positives`, `true_negatives`,
`false_positives` and `false_negatives` that are used to compute the
precision at the given recall. The threshold for the given recall
value is computed and used to evaluate the corresponding precision.
If `sample_weight` is `None`, weights default to 1.
Use `sample_weight` of 0 to mask values.
Args:
recall: A scalar value in range `[0, 1]`.
num_thresholds: (Optional) Defaults to 200. The number of thresholds to
use for matching the given recall.
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.PrecisionAtRecall(0.5)
>>> m.update_state([0, 0, 0, 1, 1], [0, 0.3, 0.8, 0.3, 0.8])
>>> m.result().numpy()
0.5
>>> m.reset_states()
>>> m.update_state([0, 0, 0, 1, 1], [0, 0.3, 0.8, 0.3, 0.8],
... sample_weight=[2, 2, 2, 1, 1])
>>> m.result().numpy()
0.33333333
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.PrecisionAtRecall(recall=0.8)])
```"
4944,result,tensorflow/tensorflow/python/keras/metrics.py,1735,method,
4945,get_config,tensorflow/tensorflow/python/keras/metrics.py,1743,method,
4946,RecallAtPrecision,tensorflow/tensorflow/python/keras/metrics.py,1750,class,"Computes best recall where precision is >= specified value.
For a given score-label-distribution the required precision might not
be achievable, in this case 0.0 is returned as recall.
This metric creates four local variables, `true_positives`, `true_negatives`,
`false_positives` and `false_negatives` that are used to compute the
recall at the given precision. The threshold for the given precision
value is computed and used to evaluate the corresponding recall.
If `sample_weight` is `None`, weights default to 1.
Use `sample_weight` of 0 to mask values.
Args:
precision: A scalar value in range `[0, 1]`.
num_thresholds: (Optional) Defaults to 200. The number of thresholds to
use for matching the given precision.
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.RecallAtPrecision(0.8)
>>> m.update_state([0, 0, 1, 1], [0, 0.5, 0.3, 0.9])
>>> m.result().numpy()
0.5
>>> m.reset_states()
>>> m.update_state([0, 0, 1, 1], [0, 0.5, 0.3, 0.9],
... sample_weight=[1, 0, 0, 1])
>>> m.result().numpy()
1.0
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.RecallAtPrecision(precision=0.8)])
```"
4947,result,tensorflow/tensorflow/python/keras/metrics.py,1805,method,
4948,get_config,tensorflow/tensorflow/python/keras/metrics.py,1813,method,
4949,AUC,tensorflow/tensorflow/python/keras/metrics.py,1821,class,"Computes the approximate AUC (Area under the curve) via a Riemann sum.
This metric creates four local variables, `true_positives`, `true_negatives`,
`false_positives` and `false_negatives` that are used to compute the AUC.
To discretize the AUC curve, a linearly spaced set of thresholds is used to
compute pairs of recall and precision values. The area under the ROC-curve is
therefore computed using the height of the recall values by the false positive
rate, while the area under the PR-curve is the computed using the height of
the precision values by the recall.
This value is ultimately returned as `auc`, an idempotent operation that
computes the area under a discretized curve of precision versus recall values
(computed using the aforementioned variables). The `num_thresholds` variable
controls the degree of discretization with larger numbers of thresholds more
closely approximating the true AUC. The quality of the approximation may vary
dramatically depending on `num_thresholds`. The `thresholds` parameter can be
used to manually specify thresholds which split the predictions more evenly.
For best results, `predictions` should be distributed approximately uniformly
in the range [0, 1] and not peaked around 0 or 1. The quality of the AUC
approximation may be poor if this is not the case. Setting `summation_method`
to 'minoring' or 'majoring' can help quantify the error in the approximation
by providing lower or upper bound estimate of the AUC.
If `sample_weight` is `None`, weights default to 1.
Use `sample_weight` of 0 to mask values.
Args:
num_thresholds: (Optional) Defaults to 200. The number of thresholds to
use when discretizing the roc curve. Values must be > 1.
curve: (Optional) Specifies the name of the curve to be computed, 'ROC'
[default] or 'PR' for the Precision-Recall-curve.
summation_method: (Optional) Specifies the [Riemann summation method](
https://en.wikipedia.org/wiki/Riemann_sum) used.
'interpolation' (default) applies mid-point summation scheme for `ROC`.
For PR-AUC, interpolates (true/false) positives but not the ratio that
is precision (see Davis & Goadrich 2006 for details);
'minoring' applies left summation
for increasing intervals and right summation for decreasing intervals;
'majoring' does the opposite.
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
thresholds: (Optional) A list of floating point values to use as the
thresholds for discretizing the curve. If set, the `num_thresholds`
parameter is ignored. Values should be in [0, 1]. Endpoint thresholds
equal to {-epsilon, 1+epsilon} for a small positive epsilon value will
be automatically included with these to correctly handle predictions
equal to exactly 0 or 1.
multi_label: boolean indicating whether multilabel data should be
treated as such, wherein AUC is computed separately for each label and
then averaged across labels, or (when False) if the data should be
flattened into a single label before AUC computation. In the latter
case, when multilabel data is passed to AUC, each label-prediction pair
is treated as an individual data point. Should be set to False for
multi-class data.
label_weights: (optional) list, array, or tensor of non-negative weights
used to compute AUCs for multilabel data. When `multi_label` is True,
the weights are applied to the individual label AUCs when they are
averaged to produce the multi-label AUC. When it's False, they are used
to weight the individual label predictions in computing the confusion
matrix on the flattened data. Note that this is unlike class_weights in
that class_weights weights the example depending on the value of its
label, whereas label_weights depends only on the index of that label
before flattening; therefore `label_weights` should not be used for
multi-class data.
Standalone usage:
>>> m = tf.keras.metrics.AUC(num_thresholds=3)
>>> m.update_state([0, 0, 1, 1], [0, 0.5, 0.3, 0.9])
>>> # threshold values are [0 - 1e-7, 0.5, 1 + 1e-7]
>>> # tp = [2, 1, 0], fp = [2, 0, 0], fn = [0, 1, 2], tn = [0, 2, 2]
>>> # recall = [1, 0.5, 0], fp_rate = [1, 0, 0]
>>> # auc = ((((1+0.5)/2)*(1-0))+ (((0.5+0)/2)*(0-0))) = 0.75
>>> m.result().numpy()
0.75
>>> m.reset_states()
>>> m.update_state([0, 0, 1, 1], [0, 0.5, 0.3, 0.9],
... sample_weight=[1, 0, 0, 1])
>>> m.result().numpy()
1.0
Usage with `compile()` API:
```python
model.compile(optimizer='sgd', loss='mse', metrics=[tf.keras.metrics.AUC()])
```"
4950,thresholds,tensorflow/tensorflow/python/keras/metrics.py,1987,method,The thresholds used for evaluating AUC.
4951,update_state,tensorflow/tensorflow/python/keras/metrics.py,2033,method,"Accumulates confusion matrix statistics.
Args:
y_true: The ground truth values.
y_pred: The predicted values.
sample_weight: Optional weighting of each example. Defaults to 1. Can be a
`Tensor` whose rank is either 0, or the same rank as `y_true`, and must
be broadcastable to `y_true`.
Returns:
Update op."
4952,interpolate_pr_auc,tensorflow/tensorflow/python/keras/metrics.py,2093,method,"Interpolation formula inspired by section 4 of Davis & Goadrich 2006.
https://www.biostat.wisc.edu/~page/rocpr.pdf
Note here we derive & use a closed formula not present in the paper
as follows:
Precision = TP / (TP + FP) = TP / P
Modeling all of TP (true positive), FP (false positive) and their sum
P = TP + FP (predicted positive) as varying linearly within each interval
[A, B] between successive thresholds, we get
Precision slope = dTP / dP
= (TP_B - TP_A) / (P_B - P_A)
= (TP - TP_A) / (P - P_A)
Precision = (TP_A + slope * (P - P_A)) / P
The area within the interval is (slope / total_pos_weight) times
int_A^B{Precision.dP} = int_A^B{(TP_A + slope * (P - P_A)) * dP / P}
int_A^B{Precision.dP} = int_A^B{slope * dP + intercept * dP / P}
where intercept = TP_A - slope * P_A = TP_B - slope * P_B, resulting in
int_A^B{Precision.dP} = TP_B - TP_A + intercept * log(P_B / P_A)
Bringing back the factor (slope / total_pos_weight) we'd put aside, we get
slope * [dTP + intercept * log(P_B / P_A)] / total_pos_weight
where dTP == TP_B - TP_A.
Note that when P_A == 0 the above calculation simplifies into
int_A^B{Precision.dTP} = int_A^B{slope * dTP} = slope * (TP_B - TP_A)
which is really equivalent to imputing constant precision throughout the
first bucket having >0 true positives.
Returns:
pr_auc: an approximation of the area under the P-R curve."
4953,result,tensorflow/tensorflow/python/keras/metrics.py,2174,method,
4954,reset_states,tensorflow/tensorflow/python/keras/metrics.py,2226,method,
4955,get_config,tensorflow/tensorflow/python/keras/metrics.py,2235,method,
4956,CosineSimilarity,tensorflow/tensorflow/python/keras/metrics.py,2256,class,"Computes the cosine similarity between the labels and predictions.
`cosine similarity = (a . b) / ||a|| ||b||`
See: [Cosine Similarity](https://en.wikipedia.org/wiki/Cosine_similarity).
This metric keeps the average cosine similarity between `predictions` and
`labels` over a stream of data.
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
axis: (Optional) Defaults to -1. The dimension along which the cosine
similarity is computed.
Standalone usage:
>>> # l2_norm(y_true) = [[0., 1.], [1./1.414], 1./1.414]]]
>>> # l2_norm(y_pred) = [[1., 0.], [1./1.414], 1./1.414]]]
>>> # l2_norm(y_true) . l2_norm(y_pred) = [[0., 0.], [0.5, 0.5]]
>>> # result = mean(sum(l2_norm(y_true) . l2_norm(y_pred), axis=1))
>>> # = ((0. + 0.) + (0.5 + 0.5)) / 2
>>> m = tf.keras.metrics.CosineSimilarity(axis=1)
>>> m.update_state([[0., 1.], [1., 1.]], [[1., 0.], [1., 1.]])
>>> m.result().numpy()
0.49999997
>>> m.reset_states()
>>> m.update_state([[0., 1.], [1., 1.]], [[1., 0.], [1., 1.]],
... sample_weight=[0.3, 0.7])
>>> m.result().numpy()
0.6999999
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.CosineSimilarity(axis=1)])
```"
4957,MeanAbsoluteError,tensorflow/tensorflow/python/keras/metrics.py,2306,class,"Computes the mean absolute error between the labels and predictions.
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.MeanAbsoluteError()
>>> m.update_state([[0, 1], [0, 0]], [[1, 1], [0, 0]])
>>> m.result().numpy()
0.25
>>> m.reset_states()
>>> m.update_state([[0, 1], [0, 0]], [[1, 1], [0, 0]],
... sample_weight=[1, 0])
>>> m.result().numpy()
0.5
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.MeanAbsoluteError()])
```"
4958,MeanAbsolutePercentageError,tensorflow/tensorflow/python/keras/metrics.py,2342,class,"Computes the mean absolute percentage error between `y_true` and `y_pred`.
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.MeanAbsolutePercentageError()
>>> m.update_state([[0, 1], [0, 0]], [[1, 1], [0, 0]])
>>> m.result().numpy()
250000000.0
>>> m.reset_states()
>>> m.update_state([[0, 1], [0, 0]], [[1, 1], [0, 0]],
... sample_weight=[1, 0])
>>> m.result().numpy()
500000000.0
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.MeanAbsolutePercentageError()])
```"
4959,MeanSquaredError,tensorflow/tensorflow/python/keras/metrics.py,2378,class,"Computes the mean squared error between `y_true` and `y_pred`.
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.MeanSquaredError()
>>> m.update_state([[0, 1], [0, 0]], [[1, 1], [0, 0]])
>>> m.result().numpy()
0.25
>>> m.reset_states()
>>> m.update_state([[0, 1], [0, 0]], [[1, 1], [0, 0]],
... sample_weight=[1, 0])
>>> m.result().numpy()
0.5
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.MeanSquaredError()])
```"
4960,MeanSquaredLogarithmicError,tensorflow/tensorflow/python/keras/metrics.py,2414,class,"Computes the mean squared logarithmic error between `y_true` and `y_pred`.
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.MeanSquaredLogarithmicError()
>>> m.update_state([[0, 1], [0, 0]], [[1, 1], [0, 0]])
>>> m.result().numpy()
0.12011322
>>> m.reset_states()
>>> m.update_state([[0, 1], [0, 0]], [[1, 1], [0, 0]],
... sample_weight=[1, 0])
>>> m.result().numpy()
0.24022643
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.MeanSquaredLogarithmicError()])
```"
4961,Hinge,tensorflow/tensorflow/python/keras/metrics.py,2450,class,"Computes the hinge metric between `y_true` and `y_pred`.
`y_true` values are expected to be -1 or 1. If binary (0 or 1) labels are
provided we will convert them to -1 or 1.
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.Hinge()
>>> m.update_state([[0, 1], [0, 0]], [[0.6, 0.4], [0.4, 0.6]])
>>> m.result().numpy()
1.3
>>> m.reset_states()
>>> m.update_state([[0, 1], [0, 0]], [[0.6, 0.4], [0.4, 0.6]],
... sample_weight=[1, 0])
>>> m.result().numpy()
1.1
Usage with `compile()` API:
```python
model.compile(optimizer='sgd', loss='mse', metrics=[tf.keras.metrics.Hinge()])
```"
4962,SquaredHinge,tensorflow/tensorflow/python/keras/metrics.py,2485,class,"Computes the squared hinge metric between `y_true` and `y_pred`.
`y_true` values are expected to be -1 or 1. If binary (0 or 1) labels are
provided we will convert them to -1 or 1.
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.SquaredHinge()
>>> m.update_state([[0, 1], [0, 0]], [[0.6, 0.4], [0.4, 0.6]])
>>> m.result().numpy()
1.86
>>> m.reset_states()
>>> m.update_state([[0, 1], [0, 0]], [[0.6, 0.4], [0.4, 0.6]],
... sample_weight=[1, 0])
>>> m.result().numpy()
1.46
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.SquaredHinge()])
```"
4963,CategoricalHinge,tensorflow/tensorflow/python/keras/metrics.py,2523,class,"Computes the categorical hinge metric between `y_true` and `y_pred`.
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.CategoricalHinge()
>>> m.update_state([[0, 1], [0, 0]], [[0.6, 0.4], [0.4, 0.6]])
>>> m.result().numpy()
1.4000001
>>> m.reset_states()
>>> m.update_state([[0, 1], [0, 0]], [[0.6, 0.4], [0.4, 0.6]],
... sample_weight=[1, 0])
>>> m.result().numpy()
1.2
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.CategoricalHinge()])
```"
4964,RootMeanSquaredError,tensorflow/tensorflow/python/keras/metrics.py,2558,class,"Computes root mean squared error metric between `y_true` and `y_pred`.
Standalone usage:
>>> m = tf.keras.metrics.RootMeanSquaredError()
>>> m.update_state([[0, 1], [0, 0]], [[1, 1], [0, 0]])
>>> m.result().numpy()
0.5
>>> m.reset_states()
>>> m.update_state([[0, 1], [0, 0]], [[1, 1], [0, 0]],
... sample_weight=[1, 0])
>>> m.result().numpy()
0.70710677
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.RootMeanSquaredError()])
```"
4965,update_state,tensorflow/tensorflow/python/keras/metrics.py,2587,method,"Accumulates root mean squared error statistics.
Args:
y_true: The ground truth values.
y_pred: The predicted values.
sample_weight: Optional weighting of each example. Defaults to 1. Can be a
`Tensor` whose rank is either 0, or the same rank as `y_true`, and must
be broadcastable to `y_true`.
Returns:
Update op."
4966,result,tensorflow/tensorflow/python/keras/metrics.py,2608,method,
4967,LogCoshError,tensorflow/tensorflow/python/keras/metrics.py,2613,class,"Computes the logarithm of the hyperbolic cosine of the prediction error.
`logcosh = log((exp(x) + exp(-x))/2)`, where x is the error (y_pred - y_true)
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.LogCoshError()
>>> m.update_state([[0, 1], [0, 0]], [[1, 1], [0, 0]])
>>> m.result().numpy()
0.10844523
>>> m.reset_states()
>>> m.update_state([[0, 1], [0, 0]], [[1, 1], [0, 0]],
... sample_weight=[1, 0])
>>> m.result().numpy()
0.21689045
Usage with `compile()` API:
```python
model.compile(optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.LogCoshError()])
```"
4968,Poisson,tensorflow/tensorflow/python/keras/metrics.py,2649,class,"Computes the Poisson metric between `y_true` and `y_pred`.
`metric = y_pred - y_true * log(y_pred)`
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.Poisson()
>>> m.update_state([[0, 1], [0, 0]], [[1, 1], [0, 0]])
>>> m.result().numpy()
0.49999997
>>> m.reset_states()
>>> m.update_state([[0, 1], [0, 0]], [[1, 1], [0, 0]],
... sample_weight=[1, 0])
>>> m.result().numpy()
0.99999994
Usage with `compile()` API:
```python
model.compile(optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.Poisson()])
```"
4969,KLDivergence,tensorflow/tensorflow/python/keras/metrics.py,2685,class,"Computes Kullback-Leibler divergence metric between `y_true` and `y_pred`.
`metric = y_true * log(y_true / y_pred)`
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.KLDivergence()
>>> m.update_state([[0, 1], [0, 0]], [[0.6, 0.4], [0.4, 0.6]])
>>> m.result().numpy()
0.45814306
>>> m.reset_states()
>>> m.update_state([[0, 1], [0, 0]], [[0.6, 0.4], [0.4, 0.6]],
... sample_weight=[1, 0])
>>> m.result().numpy()
0.9162892
Usage with `compile()` API:
```python
model.compile(optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.KLDivergence()])
```"
4970,MeanIoU,tensorflow/tensorflow/python/keras/metrics.py,2722,class,"Computes the mean Intersection-Over-Union metric.
Mean Intersection-Over-Union is a common evaluation metric for semantic image
segmentation, which first computes the IOU for each semantic class and then
computes the average over classes. IOU is defined as follows:
IOU = true_positive / (true_positive + false_positive + false_negative).
The predictions are accumulated in a confusion matrix, weighted by
`sample_weight` and the metric is then calculated from it.
If `sample_weight` is `None`, weights default to 1.
Use `sample_weight` of 0 to mask values.
Args:
num_classes: The possible number of labels the prediction task can have.
This value must be provided, since a confusion matrix of dimension =
[num_classes, num_classes] will be allocated.
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> # cm = [[1, 1],
>>> # [1, 1]]
>>> # sum_row = [2, 2], sum_col = [2, 2], true_positives = [1, 1]
>>> # iou = true_positives / (sum_row + sum_col - true_positives))
>>> # result = (1 / (2 + 2 - 1) + 1 / (2 + 2 - 1)) / 2 = 0.33
>>> m = tf.keras.metrics.MeanIoU(num_classes=2)
>>> m.update_state([0, 0, 1, 1], [0, 1, 0, 1])
>>> m.result().numpy()
0.33333334
>>> m.reset_states()
>>> m.update_state([0, 0, 1, 1], [0, 1, 0, 1],
... sample_weight=[0.3, 0.3, 0.3, 0.1])
>>> m.result().numpy()
0.23809525
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.MeanIoU(num_classes=2)])
```"
4971,update_state,tensorflow/tensorflow/python/keras/metrics.py,2782,method,"Accumulates the confusion matrix statistics.
Args:
y_true: The ground truth values.
y_pred: The predicted values.
sample_weight: Optional weighting of each example. Defaults to 1. Can be a
`Tensor` whose rank is either 0, or the same rank as `y_true`, and must
be broadcastable to `y_true`.
Returns:
Update op."
4972,result,tensorflow/tensorflow/python/keras/metrics.py,2820,method,Compute the mean intersection-over-union via the confusion matrix.
4973,reset_states,tensorflow/tensorflow/python/keras/metrics.py,2844,method,
4974,get_config,tensorflow/tensorflow/python/keras/metrics.py,2847,method,
4975,MeanTensor,tensorflow/tensorflow/python/keras/metrics.py,2854,class,"Computes the element-wise (weighted) mean of the given tensors.
`MeanTensor` returns a tensor with the same shape of the input tensors. The
mean value is updated by keeping local variables `total` and `count`. The
`total` tracks the sum of the weighted values, and `count` stores the sum of
the weighted counts.
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
Standalone usage:
>>> m = tf.keras.metrics.MeanTensor()
>>> m.update_state([0, 1, 2, 3])
>>> m.update_state([4, 5, 6, 7])
>>> m.result().numpy()
array([2., 3., 4., 5.], dtype=float32)
>>> m.update_state([12, 10, 8, 6], sample_weight= [0, 0.2, 0.5, 1])
>>> m.result().numpy()
array([2. , 3.6363635, 4.8 , 5.3333335], dtype=float32)"
4976,total,tensorflow/tensorflow/python/keras/metrics.py,2900,method,
4977,count,tensorflow/tensorflow/python/keras/metrics.py,2904,method,
4978,update_state,tensorflow/tensorflow/python/keras/metrics.py,2907,method,"Accumulates statistics for computing the element-wise mean.
Args:
values: Per-example value.
sample_weight: Optional weighting of each example. Defaults to 1.
Returns:
Update op."
4979,result,tensorflow/tensorflow/python/keras/metrics.py,2950,method,
4980,reset_states,tensorflow/tensorflow/python/keras/metrics.py,2958,method,
4981,BinaryCrossentropy,tensorflow/tensorflow/python/keras/metrics.py,2965,class,"Computes the crossentropy metric between the labels and predictions.
This is the crossentropy metric class to be used when there are only two
label classes (0 and 1).
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
from_logits: (Optional )Whether output is expected to be a logits tensor.
By default, we consider that output encodes a probability distribution.
label_smoothing: (Optional) Float in [0, 1]. When > 0, label values are
smoothed, meaning the confidence on label values are relaxed.
e.g. `label_smoothing=0.2` means that we will use a value of `0.1` for
label `0` and `0.9` for label `1`"".
Standalone usage:
>>> m = tf.keras.metrics.BinaryCrossentropy()
>>> m.update_state([[0, 1], [0, 0]], [[0.6, 0.4], [0.4, 0.6]])
>>> m.result().numpy()
0.81492424
>>> m.reset_states()
>>> m.update_state([[0, 1], [0, 0]], [[0.6, 0.4], [0.4, 0.6]],
... sample_weight=[1, 0])
>>> m.result().numpy()
0.9162905
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.BinaryCrossentropy()])
```"
4982,CategoricalCrossentropy,tensorflow/tensorflow/python/keras/metrics.py,3018,class,"Computes the crossentropy metric between the labels and predictions.
This is the crossentropy metric class to be used when there are multiple
label classes (2 or more). Here we assume that labels are given as a `one_hot`
representation. eg., When labels values are [2, 0, 1],
`y_true` = [[0, 0, 1], [1, 0, 0], [0, 1, 0]].
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
from_logits: (Optional) Whether output is expected to be a logits tensor.
By default, we consider that output encodes a probability distribution.
label_smoothing: (Optional) Float in [0, 1]. When > 0, label values are
smoothed, meaning the confidence on label values are relaxed. e.g.
`label_smoothing=0.2` means that we will use a value of `0.1` for label
`0` and `0.9` for label `1`""
Standalone usage:
>>> # EPSILON = 1e-7, y = y_true, y` = y_pred
>>> # y` = clip_ops.clip_by_value(output, EPSILON, 1. - EPSILON)
>>> # y` = [[0.05, 0.95, EPSILON], [0.1, 0.8, 0.1]]
>>> # xent = -sum(y * log(y'), axis = -1)
>>> # = -((log 0.95), (log 0.1))
>>> # = [0.051, 2.302]
>>> # Reduced xent = (0.051 + 2.302) / 2
>>> m = tf.keras.metrics.CategoricalCrossentropy()
>>> m.update_state([[0, 1, 0], [0, 0, 1]],
... [[0.05, 0.95, 0], [0.1, 0.8, 0.1]])
>>> m.result().numpy()
1.1769392
>>> m.reset_states()
>>> m.update_state([[0, 1, 0], [0, 0, 1]],
... [[0.05, 0.95, 0], [0.1, 0.8, 0.1]],
... sample_weight=tf.constant([0.3, 0.7]))
>>> m.result().numpy()
1.6271976
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.CategoricalCrossentropy()])
```"
4983,SparseCategoricalCrossentropy,tensorflow/tensorflow/python/keras/metrics.py,3082,class,"Computes the crossentropy metric between the labels and predictions.
Use this crossentropy metric when there are two or more label classes.
We expect labels to be provided as integers. If you want to provide labels
using `one-hot` representation, please use `CategoricalCrossentropy` metric.
There should be `# classes` floating point values per feature for `y_pred`
and a single floating point value per feature for `y_true`.
In the snippet below, there is a single floating point value per example for
`y_true` and `# classes` floating pointing values per example for `y_pred`.
The shape of `y_true` is `[batch_size]` and the shape of `y_pred` is
`[batch_size, num_classes]`.
Args:
name: (Optional) string name of the metric instance.
dtype: (Optional) data type of the metric result.
from_logits: (Optional) Whether output is expected to be a logits tensor.
By default, we consider that output encodes a probability distribution.
axis: (Optional) Defaults to -1. The dimension along which the metric is
computed.
Standalone usage:
>>> # y_true = one_hot(y_true) = [[0, 1, 0], [0, 0, 1]]
>>> # logits = log(y_pred)
>>> # softmax = exp(logits) / sum(exp(logits), axis=-1)
>>> # softmax = [[0.05, 0.95, EPSILON], [0.1, 0.8, 0.1]]
>>> # xent = -sum(y * log(softmax), 1)
>>> # log(softmax) = [[-2.9957, -0.0513, -16.1181],
>>> # [-2.3026, -0.2231, -2.3026]]
>>> # y_true * log(softmax) = [[0, -0.0513, 0], [0, 0, -2.3026]]
>>> # xent = [0.0513, 2.3026]
>>> # Reduced xent = (0.0513 + 2.3026) / 2
>>> m = tf.keras.metrics.SparseCategoricalCrossentropy()
>>> m.update_state([1, 2],
... [[0.05, 0.95, 0], [0.1, 0.8, 0.1]])
>>> m.result().numpy()
1.1769392
>>> m.reset_states()
>>> m.update_state([1, 2],
... [[0.05, 0.95, 0], [0.1, 0.8, 0.1]],
... sample_weight=tf.constant([0.3, 0.7]))
>>> m.result().numpy()
1.6271976
Usage with `compile()` API:
```python
model.compile(
optimizer='sgd',
loss='mse',
metrics=[tf.keras.metrics.SparseCategoricalCrossentropy()])
```"
4984,SumOverBatchSize,tensorflow/tensorflow/python/keras/metrics.py,3152,class,"Computes the weighted sum over batch size of the given values.
For example, if values is [1, 3, 5, 7] then the metric value is 4.
If the weights were specified as [1, 1, 0, 0] then the value would be 1.
This metric creates two variables, `total` and `count` that are used to
compute the average of `values`. This average is ultimately returned as sum
over batch size which is an idempotent operation that simply divides `total`
by `count`.
If `sample_weight` is `None`, weights default to 1. Use `sample_weight` of 0
to mask values."
4985,SumOverBatchSizeMetricWrapper,tensorflow/tensorflow/python/keras/metrics.py,3174,class,Wraps a function with the `SumOverBatchSizeMetricWrapper` metric.
4986,update_state,tensorflow/tensorflow/python/keras/metrics.py,3191,method,
4987,get_config,tensorflow/tensorflow/python/keras/metrics.py,3202,method,
4988,accuracy,tensorflow/tensorflow/python/keras/metrics.py,3210,function,
4989,binary_accuracy,tensorflow/tensorflow/python/keras/metrics.py,3222,function,"Calculates how often predictions matches binary labels.
Standalone usage:
>>> y_true = [[1], [1], [0], [0]]
>>> y_pred = [[1], [1], [0], [0]]
>>> m = tf.keras.metrics.binary_accuracy(y_true, y_pred)
>>> assert m.shape == (4,)
>>> m.numpy()
array([1., 1., 1., 1.], dtype=float32)
Args:
y_true: Ground truth values. shape = `[batch_size, d0, .. dN]`.
y_pred: The predicted values. shape = `[batch_size, d0, .. dN]`.
threshold: (Optional) Float representing the threshold for deciding whether
prediction values are 1 or 0.
Returns:
Binary accuracy values. shape = `[batch_size, d0, .. dN-1]`"
4990,categorical_accuracy,tensorflow/tensorflow/python/keras/metrics.py,3250,function,"Calculates how often predictions matches one-hot labels.
Standalone usage:
>>> y_true = [[0, 0, 1], [0, 1, 0]]
>>> y_pred = [[0.1, 0.9, 0.8], [0.05, 0.95, 0]]
>>> m = tf.keras.metrics.categorical_accuracy(y_true, y_pred)
>>> assert m.shape == (2,)
>>> m.numpy()
array([0., 1.], dtype=float32)
You can provide logits of classes as `y_pred`, since argmax of
logits and probabilities are same.
Args:
y_true: One-hot ground truth values.
y_pred: The prediction values.
Returns:
Categorical accuracy values."
4991,sparse_categorical_accuracy,tensorflow/tensorflow/python/keras/metrics.py,3279,function,"Calculates how often predictions matches integer labels.
Standalone usage:
>>> y_true = [2, 1]
>>> y_pred = [[0.1, 0.9, 0.8], [0.05, 0.95, 0]]
>>> m = tf.keras.metrics.sparse_categorical_accuracy(y_true, y_pred)
>>> assert m.shape == (2,)
>>> m.numpy()
array([0., 1.], dtype=float32)
You can provide logits of classes as `y_pred`, since argmax of
logits and probabilities are same.
Args:
y_true: Integer ground truth values.
y_pred: The prediction values.
Returns:
Sparse categorical accuracy values."
4992,top_k_categorical_accuracy,tensorflow/tensorflow/python/keras/metrics.py,3320,function,"Computes how often targets are in the top `K` predictions.
Standalone usage:
>>> y_true = [[0, 0, 1], [0, 1, 0]]
>>> y_pred = [[0.1, 0.9, 0.8], [0.05, 0.95, 0]]
>>> m = tf.keras.metrics.top_k_categorical_accuracy(y_true, y_pred, k=3)
>>> assert m.shape == (2,)
>>> m.numpy()
array([1., 1.], dtype=float32)
Args:
y_true: The ground truth values.
y_pred: The prediction values.
k: (Optional) Number of top elements to look at for computing accuracy.
Defaults to 5.
Returns:
Top K categorical accuracy value."
4993,sparse_top_k_categorical_accuracy,tensorflow/tensorflow/python/keras/metrics.py,3346,function,"Computes how often integer targets are in the top `K` predictions.
Standalone usage:
>>> y_true = [2, 1]
>>> y_pred = [[0.1, 0.9, 0.8], [0.05, 0.95, 0]]
>>> m = tf.keras.metrics.sparse_top_k_categorical_accuracy(
... y_true, y_pred, k=3)
>>> assert m.shape == (2,)
>>> m.numpy()
array([1., 1.], dtype=float32)
Args:
y_true: tensor of true targets.
y_pred: tensor of predicted targets.
k: (Optional) Number of top elements to look at for computing accuracy.
Defaults to 5.
Returns:
Sparse top K categorical accuracy value."
4994,cosine_proximity,tensorflow/tensorflow/python/keras/metrics.py,3380,function,"Computes the cosine similarity between labels and predictions.
Args:
y_true: The ground truth values.
y_pred: The prediction values.
axis: (Optional) Defaults to -1. The dimension along which the cosine
similarity is computed.
Returns:
Cosine similarity value."
4995,clone_metric,tensorflow/tensorflow/python/keras/metrics.py,3407,function,"Returns a clone of the metric if stateful, otherwise returns it as is."
4996,clone_metrics,tensorflow/tensorflow/python/keras/metrics.py,3415,function,Clones the given metric list/dict.
4997,serialize,tensorflow/tensorflow/python/keras/metrics.py,3421,function,"Serializes metric function or `Metric` instance.
Arguments:
metric: A Keras `Metric` instance or a metric function.
Returns:
Metric configuration dictionary."
4998,deserialize,tensorflow/tensorflow/python/keras/metrics.py,3434,function,"Deserializes a serialized metric class/function instance.
Arguments:
config: Metric configuration.
custom_objects: Optional dictionary mapping names (strings) to custom
objects (classes and functions) to be considered during deserialization.
Returns:
A Keras `Metric` instance or a metric function."
4999,get,tensorflow/tensorflow/python/keras/metrics.py,3453,function,"Retrieves a Keras metric as a `function`/`Metric` class instance.
The `identifier` may be the string name of a metric function or class.
>>> metric = tf.keras.metrics.get(""categorical_crossentropy"")
>>> type(metric)
<class 'function'>
>>> metric = tf.keras.metrics.get(""CategoricalCrossentropy"")
>>> type(metric)
<class '...tensorflow.python.keras.metrics.CategoricalCrossentropy'>
You can also specify `config` of the metric to this function by passing dict
containing `class_name` and `config` as an identifier. Also note that the
`class_name` must map to a `Metric` class
>>> identifier = {""class_name"": ""CategoricalCrossentropy"",
... ""config"": {""from_logits"": True}}
>>> metric = tf.keras.metrics.get(identifier)
>>> type(metric)
<class '...tensorflow.python.keras.metrics.CategoricalCrossentropy'>
Arguments:
identifier: A metric identifier. One of None or string name of a metric
function/class or metric configuration dictionary or a metric function or
a metric class instance
Returns:
A Keras metric as a `function`/ `Metric` class instance.
Raises:
ValueError: If `identifier` cannot be interpreted."
5000,is_built_in,tensorflow/tensorflow/python/keras/metrics.py,3497,function,
5001,get_multi_io_model,tensorflow/tensorflow/python/keras/metrics_correctness_test.py,34,function,
5002,custom_generator_multi_io,tensorflow/tensorflow/python/keras/metrics_correctness_test.py,48,function,
5003,BinaryTruePositives,tensorflow/tensorflow/python/keras/metrics_test.py,1931,class,
5004,update_state,tensorflow/tensorflow/python/keras/metrics_test.py,1937,method,
5005,result,tensorflow/tensorflow/python/keras/metrics_test.py,1951,method,
5006,BinaryTruePositivesViaControlFlow,tensorflow/tensorflow/python/keras/metrics_test.py,1955,class,
5007,update_state,tensorflow/tensorflow/python/keras/metrics_test.py,1961,method,
5008,result,tensorflow/tensorflow/python/keras/metrics_test.py,1973,method,
5009,share_weights,tensorflow/tensorflow/python/keras/models.py,56,function,
5010,clone_model,tensorflow/tensorflow/python/keras/models.py,387,function,"Clone any `Model` instance.
Model cloning is similar to calling a model on new inputs,
except that it creates new layers (and thus new weights) instead
of sharing the weights of the existing layers.
Arguments:
model: Instance of `Model`
(could be a functional model or a Sequential model).
input_tensors: optional list of input tensors or InputLayer objects
to build the model upon. If not provided,
placeholders will be created.
clone_function: Callable to be used to clone each layer in the target
model (except `InputLayer` instances). It takes as argument the layer
instance to be cloned, and returns the corresponding layer instance to
be used in the model copy. If unspecified, this callable defaults to
the following serialization/deserialization function:
`lambda layer: layer.__class__.from_config(layer.get_config())`.
By passing a custom callable, you can customize your copy of the
model, e.g. by wrapping certain layers of interest (you might want to
replace all `LSTM` instances with equivalent
`Bidirectional(LSTM(...))` instances, for example).
Returns:
An instance of `Model` reproducing the behavior
of the original model, on top of new inputs tensors,
using newly instantiated weights. The cloned model might behave
differently from the original model if a custom clone_function
modifies the layer.
Raises:
ValueError: in case of invalid `model` argument value."
5011,in_place_subclassed_model_state_restoration,tensorflow/tensorflow/python/keras/models.py,557,function,"Restores the original state of a model after it was ""reset"".
This undoes this action of `_in_place_subclassed_model_reset`, which is called
in `clone_and_build_model` if `in_place_reset` is set to True.
Args:
model: Instance of a Keras model created via subclassing, on which
`_in_place_subclassed_model_reset` was previously called."
5012,clone_and_build_model,tensorflow/tensorflow/python/keras/models.py,589,function,"Clone a `Model` and build/compile it with the same settings used before.
This function can be be run in the same graph or in a separate graph from the
model. When using a separate graph, `in_place_reset` must be `False`.
Note that, currently, the clone produced from this function may not work with
TPU DistributionStrategy. Try at your own risk.
Args:
model: `tf.keras.Model` object. Can be Functional, Sequential, or
sub-classed.
input_tensors: Optional list or dictionary of input tensors to build the
model upon. If not provided, placeholders will be created.
target_tensors: Optional list of target tensors for compiling the model. If
not provided, placeholders will be created.
custom_objects: Optional dictionary mapping string names to custom classes
or functions.
compile_clone: Boolean, whether to compile model clone (default `True`).
in_place_reset: Boolean, whether to reset the model in place. Only used if
the model is a subclassed model. In the case of a subclassed model,
this argument must be set to `True` (default `False`). To restore the
original model, use the function
`in_place_subclassed_model_state_restoration(model)`.
optimizer_iterations: An iterations variable that will be incremented by the
optimizer if the clone is compiled. This argument is used when a Keras
model is cloned into an Estimator model function, because Estimators
create their own global step variable.
optimizer_config: Optimizer config dictionary or list of dictionary
returned from `get_config()`. This argument should be defined if
`clone_and_build_model` is called in a different graph or session from
the original model, and the optimizer is an instance of `OptimizerV2`.
Returns:
Clone of the model.
Raises:
ValueError: Cloning fails in the following cases
- cloning a subclassed model with `in_place_reset` set to False.
- compiling the clone when the original model has not been compiled."
5013,Optimizer,tensorflow/tensorflow/python/keras/optimizers.py,47,class,"Abstract optimizer base class.
Note: this is the parent class of all optimizers, not an actual optimizer
that can be used for training models.
All Keras optimizers support the following keyword arguments:
clipnorm: float >= 0. Gradients will be clipped
when their L2 norm exceeds this value.
clipvalue: float >= 0. Gradients will be clipped
when their absolute value exceeds this value."
5014,get_updates,tensorflow/tensorflow/python/keras/optimizers.py,90,method,
5015,get_gradients,tensorflow/tensorflow/python/keras/optimizers.py,93,method,"Returns gradients of `loss` with respect to `params`.
Arguments:
loss: Loss tensor.
params: List of variables.
Returns:
List of gradient tensors.
Raises:
ValueError: In case any gradient cannot be computed (e.g. if gradient
function not implemented)."
5016,set_weights,tensorflow/tensorflow/python/keras/optimizers.py,123,method,"Sets the weights of the optimizer, from Numpy arrays.
Should only be called after computing the gradients
(otherwise the optimizer has no weights).
Arguments:
weights: a list of Numpy arrays. The number of arrays and their shape
must match number of the dimensions of the weights of the optimizer
(i.e. it should match the output of `get_weights`).
Raises:
ValueError: in case of incompatible weight shapes."
5017,get_weights,tensorflow/tensorflow/python/keras/optimizers.py,153,method,"Returns the current value of the weights of the optimizer.
Returns:
A list of numpy arrays."
5018,get_config,tensorflow/tensorflow/python/keras/optimizers.py,161,method,
5019,from_config,tensorflow/tensorflow/python/keras/optimizers.py,170,method,
5020,SGD,tensorflow/tensorflow/python/keras/optimizers.py,174,class,"Stochastic gradient descent optimizer.
Includes support for momentum,
learning rate decay, and Nesterov momentum.
Arguments:
lr: float >= 0. Learning rate.
momentum: float >= 0. Parameter that accelerates SGD in the relevant
direction and dampens oscillations.
decay: float >= 0. Learning rate decay over each update.
nesterov: boolean. Whether to apply Nesterov momentum."
5021,get_updates,tensorflow/tensorflow/python/keras/optimizers.py,204,method,
5022,get_config,tensorflow/tensorflow/python/keras/optimizers.py,232,method,
5023,RMSprop,tensorflow/tensorflow/python/keras/optimizers.py,243,class,"RMSProp optimizer.
It is recommended to leave the parameters of this optimizer
at their default values
(except the learning rate, which can be freely tuned).
Arguments:
lr: float >= 0. Learning rate.
rho: float >= 0.
epsilon: float >= 0. Fuzz factor. If `None`, defaults to `K.epsilon()`.
decay: float >= 0. Learning rate decay over each update."
5024,get_updates,tensorflow/tensorflow/python/keras/optimizers.py,274,method,
5025,get_config,tensorflow/tensorflow/python/keras/optimizers.py,299,method,
5026,Adagrad,tensorflow/tensorflow/python/keras/optimizers.py,310,class,"Adagrad optimizer.
Adagrad is an optimizer with parameter-specific learning rates,
which are adapted relative to how frequently a parameter gets
updated during training. The more updates a parameter receives,
the smaller the updates.
It is recommended to leave the parameters of this optimizer
at their default values.
# Arguments
lr: float >= 0. Initial learning rate.
epsilon: float >= 0. If `None`, defaults to `K.epsilon()`.
decay: float >= 0. Learning rate decay over each update.
# References
- [Adaptive Subgradient Methods for Online Learning and Stochastic
Optimization](http://www.jmlr.org/papers/volume12/duchi11a/duchi11a.pdf)"
5027,get_updates,tensorflow/tensorflow/python/keras/optimizers.py,348,method,
5028,get_config,tensorflow/tensorflow/python/keras/optimizers.py,373,method,
5029,Adadelta,tensorflow/tensorflow/python/keras/optimizers.py,383,class,"Adadelta optimizer.
Adadelta is a more robust extension of Adagrad
that adapts learning rates based on a moving window of gradient updates,
instead of accumulating all past gradients. This way, Adadelta continues
learning even when many updates have been done. Compared to Adagrad, in the
original version of Adadelta you don't have to set an initial learning
rate. In this version, initial learning rate and decay factor can
be set, as in most other Keras optimizers.
It is recommended to leave the parameters of this optimizer
at their default values.
# Arguments
lr: float >= 0. Initial learning rate, defaults to 1.
It is recommended to leave it at the default value.
rho: float >= 0. Adadelta decay factor, corresponding to fraction of
gradient to keep at each time step.
epsilon: float >= 0. Fuzz factor. If `None`, defaults to `K.epsilon()`.
decay: float >= 0. Initial learning rate decay.
# References
- [Adadelta - an adaptive learning rate
method](http://arxiv.org/abs/1212.5701)"
5030,get_updates,tensorflow/tensorflow/python/keras/optimizers.py,429,method,
5031,get_config,tensorflow/tensorflow/python/keras/optimizers.py,461,method,
5032,Adam,tensorflow/tensorflow/python/keras/optimizers.py,472,class,"Adam optimizer.
Default parameters follow those provided in the original paper.
Arguments:
lr: float >= 0. Learning rate.
beta_1: float, 0 < beta < 1. Generally close to 1.
beta_2: float, 0 < beta < 1. Generally close to 1.
epsilon: float >= 0. Fuzz factor. If `None`, defaults to `K.epsilon()`.
decay: float >= 0. Learning rate decay over each update.
amsgrad: boolean. Whether to apply the AMSGrad variant of this algorithm
from the paper ""On the Convergence of Adam and Beyond""."
5033,get_updates,tensorflow/tensorflow/python/keras/optimizers.py,518,method,
5034,get_config,tensorflow/tensorflow/python/keras/optimizers.py,557,method,
5035,Adamax,tensorflow/tensorflow/python/keras/optimizers.py,570,class,"Adamax optimizer from Adam paper's Section 7.
It is a variant of Adam based on the infinity norm.
Default parameters follow those provided in the paper.
Arguments:
lr: float >= 0. Learning rate.
beta_1/beta_2: floats, 0 < beta < 1. Generally close to 1.
epsilon: float >= 0. Fuzz factor. If `None`, defaults to `K.epsilon()`.
decay: float >= 0. Learning rate decay over each update."
5036,get_updates,tensorflow/tensorflow/python/keras/optimizers.py,612,method,
5037,get_config,tensorflow/tensorflow/python/keras/optimizers.py,646,method,
5038,Nadam,tensorflow/tensorflow/python/keras/optimizers.py,658,class,"Nesterov Adam optimizer.
Much like Adam is essentially RMSprop with momentum,
Nadam is Adam RMSprop with Nesterov momentum.
Default parameters follow those provided in the paper.
It is recommended to leave the parameters of this optimizer
at their default values.
Arguments:
lr: float >= 0. Learning rate.
beta_1/beta_2: floats, 0 < beta < 1. Generally close to 1.
epsilon: float >= 0. Fuzz factor. If `None`, defaults to `K.epsilon()`."
5039,get_updates,tensorflow/tensorflow/python/keras/optimizers.py,701,method,
5040,get_config,tensorflow/tensorflow/python/keras/optimizers.py,744,method,
5041,TFOptimizer,tensorflow/tensorflow/python/keras/optimizers.py,756,class,Wrapper class for native TensorFlow optimizers.
5042,apply_gradients,tensorflow/tensorflow/python/keras/optimizers.py,774,method,
5043,get_grads,tensorflow/tensorflow/python/keras/optimizers.py,777,method,
5044,get_updates,tensorflow/tensorflow/python/keras/optimizers.py,780,method,
5045,weights,tensorflow/tensorflow/python/keras/optimizers.py,809,method,
5046,get_config,tensorflow/tensorflow/python/keras/optimizers.py,812,method,
5047,from_config,tensorflow/tensorflow/python/keras/optimizers.py,815,method,
5048,serialize,tensorflow/tensorflow/python/keras/optimizers.py,831,function,
5049,deserialize,tensorflow/tensorflow/python/keras/optimizers.py,836,function,"Inverse of the `serialize` function.
Arguments:
config: Optimizer configuration dictionary.
custom_objects: Optional dictionary mapping names (strings) to custom
objects (classes and functions) to be considered during deserialization.
Returns:
A Keras Optimizer instance."
5050,get,tensorflow/tensorflow/python/keras/optimizers.py,873,function,"Retrieves a Keras Optimizer instance.
Arguments:
identifier: Optimizer identifier, one of
- String: name of an optimizer
- Dictionary: configuration dictionary. - Keras Optimizer instance (it
will be returned unchanged). - TensorFlow Optimizer instance (it
will be wrapped as a Keras Optimizer).
Returns:
A Keras Optimizer instance.
Raises:
ValueError: If `identifier` cannot be interpreted."
5051,Regularizer,tensorflow/tensorflow/python/keras/regularizers.py,51,class,"Regularizer base class.
Regularizers allow you to apply penalties on layer parameters or layer
activity during optimization. These penalties are summed into the loss
function that the network optimizes.
Regularization penalties are applied on a per-layer basis. The exact API will
depend on the layer, but many layers (e.g. `Dense`, `Conv1D`, `Conv2D` and
`Conv3D`) have a unified API.
These layers expose 3 keyword arguments:
- `kernel_regularizer`: Regularizer to apply a penalty on the layer's kernel
- `bias_regularizer`: Regularizer to apply a penalty on the layer's bias
- `activity_regularizer`: Regularizer to apply a penalty on the layer's output
All layers (including custom layers) expose `activity_regularizer` as a
settable property, whether or not it is in the constructor arguments.
The value returned by the `activity_regularizer` is divided by the input
batch size so that the relative weighting between the weight regularizers and
the activity regularizers does not change with the batch size.
You can access a layer's regularization penalties by calling `layer.losses`
after calling the layer on inputs.
## Example
>>> layer = tf.keras.layers.Dense(
... 5, input_dim=5,
... kernel_initializer='ones',
... kernel_regularizer=tf.keras.regularizers.L1(0.01),
... activity_regularizer=tf.keras.regularizers.L2(0.01))
>>> tensor = tf.ones(shape=(5, 5)) * 2.0
>>> out = layer(tensor)
>>> # The kernel regularization term is 0.25
>>> # The activity regularization term (after dividing by the batch size) is 5
>>> tf.math.reduce_sum(layer.losses)
<tf.Tensor: shape=(), dtype=float32, numpy=5.25>
## Available penalties
```python
tf.keras.regularizers.L1(0.3) # L1 Regularization Penalty
tf.keras.regularizers.L2(0.1) # L2 Regularization Penalty
tf.keras.regularizers.L1L2(l1=0.01, l2=0.01) # L1 + L2 penalties
```
## Directly calling a regularizer
Compute a regularization loss on a tensor by directly calling a regularizer
as if it is a one-argument function.
E.g.
>>> regularizer = tf.keras.regularizers.L2(2.)
>>> tensor = tf.ones(shape=(5, 5))
>>> regularizer(tensor)
<tf.Tensor: shape=(), dtype=float32, numpy=50.0>
## Developing new regularizers
Any function that takes in a weight matrix and returns a scalar
tensor can be used as a regularizer, e.g.:
>>> @tf.keras.utils.register_keras_serializable(package='Custom', name='l1')
... def l1_reg(weight_matrix):
... return 0.01 * tf.math.reduce_sum(tf.math.abs(weight_matrix))
...
>>> layer = tf.keras.layers.Dense(5, input_dim=5,
... kernel_initializer='ones', kernel_regularizer=l1_reg)
>>> tensor = tf.ones(shape=(5, 5))
>>> out = layer(tensor)
>>> layer.losses
[<tf.Tensor: shape=(), dtype=float32, numpy=0.25>]
Alternatively, you can write your custom regularizers in an
object-oriented way by extending this regularizer base class, e.g.:
>>> @tf.keras.utils.register_keras_serializable(package='Custom', name='l2')
... class L2Regularizer(tf.keras.regularizers.Regularizer):
... def __init__(self, l2=0.): # pylint: disable=redefined-outer-name
... self.l2 = l2
...
... def __call__(self, x):
... return self.l2 * tf.math.reduce_sum(tf.math.square(x))
...
... def get_config(self):
... return {'l2': float(self.l2)}
...
>>> layer = tf.keras.layers.Dense(
... 5, input_dim=5, kernel_initializer='ones',
... kernel_regularizer=L2Regularizer(l2=0.5))
>>> tensor = tf.ones(shape=(5, 5))
>>> out = layer(tensor)
>>> layer.losses
[<tf.Tensor: shape=(), dtype=float32, numpy=12.5>]
### A note on serialization and deserialization:
Registering the regularizers as serializable is optional if you are just
training and executing models, exporting to and from SavedModels, or saving
and loading weight checkpoints.
Registration is required for Keras `model_to_estimator`, saving and
loading models to HDF5 formats, Keras model cloning, some visualization
utilities, and exporting models to and from JSON. If using this functionality,
you must make sure any python process running your model has also defined
and registered your custom regularizer.
`tf.keras.utils.register_keras_serializable` is only available in TF 2.1 and
beyond. In earlier versions of TensorFlow you must pass your custom
regularizer to the `custom_objects` argument of methods that expect custom
regularizers to be registered as serializable."
5052,from_config,tensorflow/tensorflow/python/keras/regularizers.py,175,method,"Creates a regularizer from its config.
This method is the reverse of `get_config`,
capable of instantiating the same regularizer from the config
dictionary.
This method is used by Keras `model_to_estimator`, saving and
loading models to HDF5 formats, Keras model cloning, some visualization
utilities, and exporting models to and from JSON.
Arguments:
config: A Python dictionary, typically the output of get_config.
Returns:
A regularizer instance."
5053,get_config,tensorflow/tensorflow/python/keras/regularizers.py,194,method,"Returns the config of the regularizer.
An regularizer config is a Python dictionary (serializable)
containing all configuration parameters of the regularizer.
The same regularizer can be reinstantiated later
(without any saved state) from this configuration.
This method is optional if you are just training and executing models,
exporting to and from SavedModels, or using weight checkpoints.
This method is required for Keras `model_to_estimator`, saving and
loading models to HDF5 formats, Keras model cloning, some visualization
utilities, and exporting models to and from JSON.
Returns:
Python dictionary."
5054,L1L2,tensorflow/tensorflow/python/keras/regularizers.py,216,class,"A regularizer that applies both L1 and L2 regularization penalties.
The L1 regularization penalty is computed as:
`loss = l1 * reduce_sum(abs(x))`
The L2 regularization penalty is computed as
`loss = l2 * reduce_sum(square(x))`
L1L2 may be passed to a layer as a string identifier:
>>> dense = tf.keras.layers.Dense(3, kernel_regularizer='l1_l2')
In this case, the default values used are `l1=0.01` and `l2=0.01`.
Attributes:
l1: Float; L1 regularization factor.
l2: Float; L2 regularization factor."
5055,get_config,tensorflow/tensorflow/python/keras/regularizers.py,256,method,
5056,L1,tensorflow/tensorflow/python/keras/regularizers.py,261,class,"A regularizer that applies a L1 regularization penalty.
The L1 regularization penalty is computed as:
`loss = l1 * reduce_sum(abs(x))`
L1 may be passed to a layer as a string identifier:
>>> dense = tf.keras.layers.Dense(3, kernel_regularizer='l1')
In this case, the default value used is `l1=0.01`.
Attributes:
l1: Float; L1 regularization factor."
5057,get_config,tensorflow/tensorflow/python/keras/regularizers.py,290,method,
5058,L2,tensorflow/tensorflow/python/keras/regularizers.py,295,class,"A regularizer that applies a L2 regularization penalty.
The L2 regularization penalty is computed as:
`loss = l2 * reduce_sum(square(x))`
L2 may be passed to a layer as a string identifier:
>>> dense = tf.keras.layers.Dense(3, kernel_regularizer='l2')
In this case, the default value used is `l2=0.01`.
Attributes:
l2: Float; L2 regularization factor."
5059,get_config,tensorflow/tensorflow/python/keras/regularizers.py,324,method,
5060,l1_l2,tensorflow/tensorflow/python/keras/regularizers.py,329,function,"Create a regularizer that applies both L1 and L2 penalties.
The L1 regularization penalty is computed as:
`loss = l1 * reduce_sum(abs(x))`
The L2 regularization penalty is computed as:
`loss = l2 * reduce_sum(square(x))`
Arguments:
l1: Float; L1 regularization factor.
l2: Float; L2 regularization factor.
Returns:
An L1L2 Regularizer with the given regularization factors."
5061,serialize,tensorflow/tensorflow/python/keras/regularizers.py,354,function,
5062,deserialize,tensorflow/tensorflow/python/keras/regularizers.py,359,function,
5063,get,tensorflow/tensorflow/python/keras/regularizers.py,372,function,Retrieve a regularizer instance from a config or identifier.
5064,model_type_scope,tensorflow/tensorflow/python/keras/testing_utils.py,307,function,"Provides a scope within which the model type to test is equal to `value`.
The model type gets restored to its original value upon exiting the scope.
Arguments:
value: model type value
Yields:
The provided value."
5065,run_eagerly_scope,tensorflow/tensorflow/python/keras/testing_utils.py,328,function,"Provides a scope within which we compile models to run eagerly or not.
The boolean gets restored to its original value upon exiting the scope.
Arguments:
value: Bool specifying if we should run models eagerly in the active test.
Should be True or False.
Yields:
The provided value."
5066,use_keras_tensors_scope,tensorflow/tensorflow/python/keras/testing_utils.py,350,function,"Provides a scope within which we use KerasTensors in the func. API or not.
The boolean gets restored to its original value upon exiting the scope.
Arguments:
value: Bool specifying if we should build functional models
using KerasTensors in the active test.
Should be True or False.
Yields:
The provided value."
5067,should_run_eagerly,tensorflow/tensorflow/python/keras/testing_utils.py,372,function,Returns whether the models we are testing should be run eagerly.
5068,saved_model_format_scope,tensorflow/tensorflow/python/keras/testing_utils.py,383,function,"Provides a scope within which the savde model format to test is `value`.
The saved model format gets restored to its original value upon exiting the
scope.
Arguments:
value: saved model format value
Yields:
The provided value."
5069,get_save_format,tensorflow/tensorflow/python/keras/testing_utils.py,404,function,
5070,get_model_type,tensorflow/tensorflow/python/keras/testing_utils.py,413,function,Gets the model type that should be tested.
5071,get_small_sequential_mlp,tensorflow/tensorflow/python/keras/testing_utils.py,423,function,
5072,get_small_functional_mlp,tensorflow/tensorflow/python/keras/testing_utils.py,434,function,
5073,SmallSubclassMLP,tensorflow/tensorflow/python/keras/testing_utils.py,442,class,A subclass model based small MLP.
5074,call,tensorflow/tensorflow/python/keras/testing_utils.py,458,method,
5075,get_small_subclass_mlp,tensorflow/tensorflow/python/keras/testing_utils.py,487,function,
5076,get_small_subclass_mlp_with_custom_build,tensorflow/tensorflow/python/keras/testing_utils.py,491,function,
5077,get_small_mlp,tensorflow/tensorflow/python/keras/testing_utils.py,495,function,Get a small mlp of the model type specified by `get_model_type`.
5078,get_model_from_layers,tensorflow/tensorflow/python/keras/testing_utils.py,566,function,"Builds a model from a sequence of layers.
Args:
model_layers: The layers used to build the network.
input_shape: Shape tuple of the input or 'TensorShape' instance.
input_dtype: Datatype of the input.
name: Name for the model.
input_ragged: Boolean, whether the input data is a ragged tensor.
input_sparse: Boolean, whether the input data is a sparse tensor.
Returns:
A Keras model."
5079,Bias,tensorflow/tensorflow/python/keras/testing_utils.py,631,class,
5080,build,tensorflow/tensorflow/python/keras/testing_utils.py,633,method,
5081,call,tensorflow/tensorflow/python/keras/testing_utils.py,636,method,
5082,get_multi_io_model,tensorflow/tensorflow/python/keras/testing_utils.py,724,function,"Builds a multi-io model that contains two branches.
The produced model will be of the type specified by `get_model_type`.
To build a two-input, two-output model:
Specify a list of layers for branch a and branch b, but do not specify any
shared input branch or shared output branch. The resulting model will apply
each branch to a different input, to produce two outputs.
The first value in branch_a must be the Keras 'Input' layer for branch a,
and the first value in branch_b must be the Keras 'Input' layer for
branch b.
example usage:
```
branch_a = [Input(shape=(2,), name='a'), Dense(), Dense()]
branch_b = [Input(shape=(3,), name='b'), Dense(), Dense()]
model = get_multi_io_model(branch_a, branch_b)
```
To build a two-input, one-output model:
Specify a list of layers for branch a and branch b, and specify a
shared output branch. The resulting model will apply
each branch to a different input. It will then apply the shared output
branch to a tuple containing the intermediate outputs of each branch,
to produce a single output. The first layer in the shared_output_branch
must be able to merge a tuple of two tensors.
The first value in branch_a must be the Keras 'Input' layer for branch a,
and the first value in branch_b must be the Keras 'Input' layer for
branch b.
example usage:
```
input_branch_a = [Input(shape=(2,), name='a'), Dense(), Dense()]
input_branch_b = [Input(shape=(3,), name='b'), Dense(), Dense()]
shared_output_branch = [Concatenate(), Dense(), Dense()]
model = get_multi_io_model(input_branch_a, input_branch_b,
shared_output_branch=shared_output_branch)
```
To build a one-input, two-output model:
Specify a list of layers for branch a and branch b, and specify a
shared input branch. The resulting model will take one input, and apply
the shared input branch to it. It will then respectively apply each branch
to that intermediate result in parallel, to produce two outputs.
The first value in the shared_input_branch must be the Keras 'Input' layer
for the whole model. Branch a and branch b should not contain any Input
layers.
example usage:
```
shared_input_branch = [Input(shape=(2,), name='in'), Dense(), Dense()]
output_branch_a = [Dense(), Dense()]
output_branch_b = [Dense(), Dense()]
model = get_multi_io_model(output__branch_a, output_branch_b,
shared_input_branch=shared_input_branch)
```
Args:
branch_a: A sequence of layers for branch a of the model.
branch_b: A sequence of layers for branch b of the model.
shared_input_branch: An optional sequence of layers to apply to a single
input, before applying both branches to that intermediate result. If set,
the model will take only one input instead of two. Defaults to None.
shared_output_branch: An optional sequence of layers to merge the
intermediate results produced by branch a and branch b. If set,
the model will produce only one output instead of two. Defaults to None.
Returns:
A multi-io model of the type specified by `get_model_type`, specified
by the different branches."
5083,get_v2_optimizer,tensorflow/tensorflow/python/keras/testing_utils.py,866,function,"Get the v2 optimizer requested.
This is only necessary until v2 are the default, as we are testing in Eager,
and Eager + v1 optimizers fail tests. When we are in v2, the strings alone
should be sufficient, and this mapping can theoretically be removed.
Args:
name: string name of Keras v2 optimizer.
**kwargs: any kwargs to pass to the optimizer constructor.
Returns:
Initialized Keras v2 optimizer.
Raises:
ValueError: if an unknown name was passed."
5084,get_expected_metric_variable_names,tensorflow/tensorflow/python/keras/testing_utils.py,891,function,Returns expected metric variable names given names and prefix/suffix.
5085,enable_v2_dtype_behavior,tensorflow/tensorflow/python/keras/testing_utils.py,900,function,Decorator for enabling the layer V2 dtype behavior on a test.
5086,disable_v2_dtype_behavior,tensorflow/tensorflow/python/keras/testing_utils.py,905,function,Decorator for disabling the layer V2 dtype behavior on a test.
5087,device,tensorflow/tensorflow/python/keras/testing_utils.py,925,function,Uses gpu when requested and available.
5088,use_gpu,tensorflow/tensorflow/python/keras/testing_utils.py,936,function,Uses gpu when requested and available.
5089,dense_block,tensorflow/tensorflow/python/keras/applications/densenet.py,57,function,"A dense block.
Arguments:
x: input tensor.
blocks: integer, the number of building blocks.
name: string, block label.
Returns:
Output tensor for the block."
5090,transition_block,tensorflow/tensorflow/python/keras/applications/densenet.py,73,function,"A transition block.
Arguments:
x: input tensor.
reduction: float, compression rate at transition layers.
name: string, block label.
Returns:
output tensor for the block."
5091,conv_block,tensorflow/tensorflow/python/keras/applications/densenet.py,99,function,"A building block for a dense block.
Arguments:
x: input tensor.
growth_rate: float, growth rate at dense layers.
name: string, block label.
Returns:
Output tensor for the block."
5092,DenseNet,tensorflow/tensorflow/python/keras/applications/densenet.py,129,function,"Instantiates the DenseNet architecture.
Reference:
- [Densely Connected Convolutional Networks](
https://arxiv.org/abs/1608.06993) (CVPR 2017)
Optionally loads weights pre-trained on ImageNet.
Note that the data format convention used by the model is
the one specified in your Keras config at `~/.keras/keras.json`.
Caution: Be sure to properly pre-process your inputs to the application.
Please see `applications.densenet.preprocess_input` for an example.
Arguments:
blocks: numbers of building blocks for the four dense layers.
include_top: whether to include the fully-connected
layer at the top of the network.
weights: one of `None` (random initialization),
'imagenet' (pre-training on ImageNet),
or the path to the weights file to be loaded.
input_tensor: optional Keras tensor
(i.e. output of `layers.Input()`)
to use as image input for the model.
input_shape: optional shape tuple, only to be specified
if `include_top` is False (otherwise the input shape
has to be `(224, 224, 3)` (with `'channels_last'` data format)
or `(3, 224, 224)` (with `'channels_first'` data format).
It should have exactly 3 inputs channels,
and width and height should be no smaller than 32.
E.g. `(200, 200, 3)` would be one valid value.
pooling: optional pooling mode for feature extraction
when `include_top` is `False`.
- `None` means that the output of the model will be
the 4D tensor output of the
last convolutional block.
- `avg` means that global average pooling
will be applied to the output of the
last convolutional block, and thus
the output of the model will be a 2D tensor.
- `max` means that global max pooling will
be applied.
classes: optional number of classes to classify images
into, only to be specified if `include_top` is True, and
if no `weights` argument is specified.
classifier_activation: A `str` or callable. The activation function to use
on the ""top"" layer. Ignored unless `include_top=True`. Set
`classifier_activation=None` to return the logits of the ""top"" layer.
Returns:
A `keras.Model` instance.
Raises:
ValueError: in case of invalid argument for `weights`,
or invalid input shape.
ValueError: if `classifier_activation` is not `softmax` or `None` when
using a pretrained top layer."
5093,DenseNet121,tensorflow/tensorflow/python/keras/applications/densenet.py,322,function,Instantiates the Densenet121 architecture.
5094,DenseNet169,tensorflow/tensorflow/python/keras/applications/densenet.py,335,function,Instantiates the Densenet169 architecture.
5095,DenseNet201,tensorflow/tensorflow/python/keras/applications/densenet.py,348,function,Instantiates the Densenet201 architecture.
5096,preprocess_input,tensorflow/tensorflow/python/keras/applications/densenet.py,360,function,
5097,decode_predictions,tensorflow/tensorflow/python/keras/applications/densenet.py,366,function,
5098,EfficientNet,tensorflow/tensorflow/python/keras/applications/efficientnet.py,194,function,"Instantiates the EfficientNet architecture using given scaling coefficients.
Reference:
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](
https://arxiv.org/abs/1905.11946) (ICML 2019)
Optionally loads weights pre-trained on ImageNet.
Note that the data format convention used by the model is
the one specified in your Keras config at `~/.keras/keras.json`.
Arguments:
width_coefficient: float, scaling coefficient for network width.
depth_coefficient: float, scaling coefficient for network depth.
default_size: integer, default input image size.
dropout_rate: float, dropout rate before final classifier layer.
drop_connect_rate: float, dropout rate at skip connections.
depth_divisor: integer, a unit of network width.
activation: activation function.
blocks_args: list of dicts, parameters to construct block modules.
model_name: string, model name.
include_top: whether to include the fully-connected
layer at the top of the network.
weights: one of `None` (random initialization),
'imagenet' (pre-training on ImageNet),
or the path to the weights file to be loaded.
input_tensor: optional Keras tensor
(i.e. output of `layers.Input()`)
to use as image input for the model.
input_shape: optional shape tuple, only to be specified
if `include_top` is False.
It should have exactly 3 inputs channels.
pooling: optional pooling mode for feature extraction
when `include_top` is `False`.
- `None` means that the output of the model will be
the 4D tensor output of the
last convolutional layer.
- `avg` means that global average pooling
will be applied to the output of the
last convolutional layer, and thus
the output of the model will be a 2D tensor.
- `max` means that global max pooling will
be applied.
classes: optional number of classes to classify images
into, only to be specified if `include_top` is True, and
if no `weights` argument is specified.
classifier_activation: A `str` or callable. The activation function to use
on the ""top"" layer. Ignored unless `include_top=True`. Set
`classifier_activation=None` to return the logits of the ""top"" layer.
Returns:
A `keras.Model` instance.
Raises:
ValueError: in case of invalid argument for `weights`,
or invalid input shape.
ValueError: if `classifier_activation` is not `softmax` or `None` when
using a pretrained top layer."
5099,block,tensorflow/tensorflow/python/keras/applications/efficientnet.py,413,function,"An inverted residual block.
Arguments:
inputs: input tensor.
activation: activation function.
drop_rate: float between 0 and 1, fraction of the input units to drop.
name: string, block label.
filters_in: integer, the number of input filters.
filters_out: integer, the number of output filters.
kernel_size: integer, the dimension of the convolution window.
strides: integer, the stride of the convolution.
expand_ratio: integer, scaling coefficient for the input filters.
se_ratio: float between 0 and 1, fraction to squeeze the input filters.
id_skip: boolean.
Returns:
output tensor for the block."
5100,EfficientNetB0,tensorflow/tensorflow/python/keras/applications/efficientnet.py,519,function,
5101,EfficientNetB1,tensorflow/tensorflow/python/keras/applications/efficientnet.py,545,function,
5102,EfficientNetB2,tensorflow/tensorflow/python/keras/applications/efficientnet.py,571,function,
5103,EfficientNetB3,tensorflow/tensorflow/python/keras/applications/efficientnet.py,597,function,
5104,EfficientNetB4,tensorflow/tensorflow/python/keras/applications/efficientnet.py,623,function,
5105,EfficientNetB5,tensorflow/tensorflow/python/keras/applications/efficientnet.py,649,function,
5106,EfficientNetB6,tensorflow/tensorflow/python/keras/applications/efficientnet.py,675,function,
5107,EfficientNetB7,tensorflow/tensorflow/python/keras/applications/efficientnet.py,701,function,
5108,preprocess_input,tensorflow/tensorflow/python/keras/applications/efficientnet.py,736,function,
5109,decode_predictions,tensorflow/tensorflow/python/keras/applications/efficientnet.py,741,function,
5110,write_ckpt_to_h5,tensorflow/tensorflow/python/keras/applications/efficientnet_weight_update_util.py,51,function,"Map the weights in checkpoint file (tf) to h5 file (keras).
Args:
path_h5: str, path to output hdf5 file to write weights loaded from ckpt
files.
path_ckpt: str, path to the ckpt files (e.g. 'efficientnet-b0/model.ckpt')
that records efficientnet weights from original repo
https://github.com/tensorflow/tpu/tree/master/models/official/efficientnet
keras_model: keras model, built from keras.applications efficientnet
functions (e.g. EfficientNetB0)
use_ema: Bool, whether to use ExponentialMovingAverage result or not"
5111,get_variable_names_from_ckpt,tensorflow/tensorflow/python/keras/applications/efficientnet_weight_update_util.py,121,function,"Get list of tensor names from checkpoint.
Args:
path_ckpt: str, path to the ckpt files
use_ema: Bool, whether to use ExponentialMovingAverage result or not.
Returns:
List of variable names from checkpoint."
5112,get_tf_blocks,tensorflow/tensorflow/python/keras/applications/efficientnet_weight_update_util.py,145,function,Extract the block names from list of full weight names.
5113,get_keras_blocks,tensorflow/tensorflow/python/keras/applications/efficientnet_weight_update_util.py,154,function,Extract the block names from list of full weight names.
5114,keras_name_to_tf_name_stem_top,tensorflow/tensorflow/python/keras/applications/efficientnet_weight_update_util.py,161,function,"Mapping name in h5 to ckpt that is in stem or top (head).
we map name keras_name that points to a weight in h5 file
to a name of weight in ckpt file.
Args:
keras_name: str, the name of weight in the h5 file of keras implementation
use_ema: Bool, use the ExponentialMovingAverage resuolt in ckpt or not
model_name_tf: str, the name of model in ckpt.
Returns:
String for the name of weight as in ckpt file.
Raises:
KeyError: if we cannot parse the keras_name."
5115,keras_name_to_tf_name_block,tensorflow/tensorflow/python/keras/applications/efficientnet_weight_update_util.py,213,function,"Mapping name in h5 to ckpt that belongs to a block.
we map name keras_name that points to a weight in h5 file
to a name of weight in ckpt file.
Args:
keras_name: str, the name of weight in the h5 file of keras implementation
keras_block: str, the block name for keras implementation (e.g. 'block1a')
tf_block: str, the block name for tf implementation (e.g. 'blocks_0')
use_ema: Bool, use the ExponentialMovingAverage resuolt in ckpt or not
model_name_tf: str, the name of model in ckpt.
Returns:
String for the name of weight as in ckpt file.
Raises:
ValueError if keras_block does not show up in keras_name"
5116,check_match,tensorflow/tensorflow/python/keras/applications/efficientnet_weight_update_util.py,300,function,"Check if the weights in h5 and ckpt match.
we match each name from keras_weight_names that is in keras_block
and check if there is 1-1 correspondence to names from tf_weight_names
that is in tf_block
Args:
keras_block: str, the block name for keras implementation (e.g. 'block1a')
tf_block: str, the block name for tf implementation (e.g. 'blocks_0')
keras_weight_names: list of str, weight names in keras implementation
tf_weight_names: list of str, weight names in tf implementation
model_name_tf: str, the name of model in ckpt."
5117,preprocess_input,tensorflow/tensorflow/python/keras/applications/imagenet_utils.py,104,function,Preprocesses a tensor or Numpy array encoding a batch of images.
5118,decode_predictions,tensorflow/tensorflow/python/keras/applications/imagenet_utils.py,129,function,"Decodes the prediction of an ImageNet model.
Arguments:
preds: Numpy array encoding a batch of predictions.
top: Integer, how many top-guesses to return. Defaults to 5.
Returns:
A list of lists of top class prediction tuples
`(class_name, class_description, score)`.
One list of tuples per sample in batch input.
Raises:
ValueError: In case of invalid shape of the `pred` array
(must be 2D)."
5119,obtain_input_shape,tensorflow/tensorflow/python/keras/applications/imagenet_utils.py,296,function,"Internal utility to compute/validate a model's input shape.
Arguments:
input_shape: Either None (will return the default network input shape),
or a user-provided shape to be validated.
default_size: Default input width/height for the model.
min_size: Minimum input width/height accepted by the model.
data_format: Image data format to use.
require_flatten: Whether the model is expected to
be linked to a classifier via a Flatten layer.
weights: One of `None` (random initialization)
or 'imagenet' (pre-training on ImageNet).
If weights='imagenet' input channels must be equal to 3.
Returns:
An integer shape tuple (may include None entries).
Raises:
ValueError: In case of invalid argument values."
5120,correct_pad,tensorflow/tensorflow/python/keras/applications/imagenet_utils.py,388,function,"Returns a tuple for zero-padding for 2D convolution with downsampling.
Arguments:
inputs: Input tensor.
kernel_size: An integer or tuple/list of 2 integers.
Returns:
A tuple."
5121,validate_activation,tensorflow/tensorflow/python/keras/applications/imagenet_utils.py,411,function,"validates that the classifer_activation is compatible with the weights.
Args:
classifier_activation: str or callable activation function
weights: The pretrained weights to load.
Raises:
ValueError: if an activation other than `None` or `softmax` are used with
pretrained weights."
5122,InceptionResNetV2,tensorflow/tensorflow/python/keras/applications/inception_resnet_v2.py,45,function,"Instantiates the Inception-ResNet v2 architecture.
Reference:
- [Inception-v4, Inception-ResNet and the Impact of
Residual Connections on Learning](https://arxiv.org/abs/1602.07261)
(AAAI 2017)
Optionally loads weights pre-trained on ImageNet.
Note that the data format convention used by the model is
the one specified in your Keras config at `~/.keras/keras.json`.
Caution: Be sure to properly pre-process your inputs to the application.
Please see `applications.inception_resnet_v2.preprocess_input` for an example.
Arguments:
include_top: whether to include the fully-connected
layer at the top of the network.
weights: one of `None` (random initialization),
'imagenet' (pre-training on ImageNet),
or the path to the weights file to be loaded.
input_tensor: optional Keras tensor (i.e. output of `layers.Input()`)
to use as image input for the model.
input_shape: optional shape tuple, only to be specified
if `include_top` is `False` (otherwise the input shape
has to be `(299, 299, 3)` (with `'channels_last'` data format)
or `(3, 299, 299)` (with `'channels_first'` data format).
It should have exactly 3 inputs channels,
and width and height should be no smaller than 75.
E.g. `(150, 150, 3)` would be one valid value.
pooling: Optional pooling mode for feature extraction
when `include_top` is `False`.
- `None` means that the output of the model will be
the 4D tensor output of the last convolutional block.
- `'avg'` means that global average pooling
will be applied to the output of the
last convolutional block, and thus
the output of the model will be a 2D tensor.
- `'max'` means that global max pooling will be applied.
classes: optional number of classes to classify images
into, only to be specified if `include_top` is `True`, and
if no `weights` argument is specified.
classifier_activation: A `str` or callable. The activation function to use
on the ""top"" layer. Ignored unless `include_top=True`. Set
`classifier_activation=None` to return the logits of the ""top"" layer.
**kwargs: For backwards compatibility only.
Returns:
A `keras.Model` instance.
Raises:
ValueError: in case of invalid argument for `weights`,
or invalid input shape.
ValueError: if `classifier_activation` is not `softmax` or `None` when
using a pretrained top layer."
5123,conv2d_bn,tensorflow/tensorflow/python/keras/applications/inception_resnet_v2.py,251,function,"Utility function to apply conv + BN.
Arguments:
x: input tensor.
filters: filters in `Conv2D`.
kernel_size: kernel size as in `Conv2D`.
strides: strides in `Conv2D`.
padding: padding mode in `Conv2D`.
activation: activation in `Conv2D`.
use_bias: whether to use a bias in `Conv2D`.
name: name of the ops; will become `name + '_ac'` for the activation
and `name + '_bn'` for the batch norm layer.
Returns:
Output tensor after applying `Conv2D` and `BatchNormalization`."
5124,inception_resnet_block,tensorflow/tensorflow/python/keras/applications/inception_resnet_v2.py,293,function,"Adds an Inception-ResNet block.
This function builds 3 types of Inception-ResNet blocks mentioned
in the paper, controlled by the `block_type` argument (which is the
block name used in the official TF-slim implementation):
- Inception-ResNet-A: `block_type='block35'`
- Inception-ResNet-B: `block_type='block17'`
- Inception-ResNet-C: `block_type='block8'`
Arguments:
x: input tensor.
scale: scaling factor to scale the residuals (i.e., the output of passing
`x` through an inception module) before adding them to the shortcut
branch. Let `r` be the output from the residual branch, the output of this
block will be `x + scale * r`.
block_type: `'block35'`, `'block17'` or `'block8'`, determines the network
structure in the residual branch.
block_idx: an `int` used for generating layer names. The Inception-ResNet
blocks are repeated many times in this network. We use `block_idx` to
identify each of the repetitions. For example, the first
Inception-ResNet-A block will have `block_type='block35', block_idx=0`,
and the layer names will have a common prefix `'block35_0'`.
activation: activation function to use at the end of the block (see
[activations](../activations.md)). When `activation=None`, no activation
is applied
(i.e., ""linear"" activation: `a(x) = x`).
Returns:
Output tensor for the block.
Raises:
ValueError: if `block_type` is not one of `'block35'`,
`'block17'` or `'block8'`."
5125,preprocess_input,tensorflow/tensorflow/python/keras/applications/inception_resnet_v2.py,377,function,
5126,decode_predictions,tensorflow/tensorflow/python/keras/applications/inception_resnet_v2.py,382,function,
5127,InceptionV3,tensorflow/tensorflow/python/keras/applications/inception_v3.py,48,function,"Instantiates the Inception v3 architecture.
Reference:
- [Rethinking the Inception Architecture for Computer Vision](
http://arxiv.org/abs/1512.00567) (CVPR 2016)
Optionally loads weights pre-trained on ImageNet.
Note that the data format convention used by the model is
the one specified in the `tf.keras.backend.image_data_format()`.
Caution: Be sure to properly pre-process your inputs to the application.
Please see `applications.inception_v3.preprocess_input` for an example.
Arguments:
include_top: Boolean, whether to include the fully-connected
layer at the top, as the last layer of the network. Default to `True`.
weights: One of `None` (random initialization),
`imagenet` (pre-training on ImageNet),
or the path to the weights file to be loaded. Default to `imagenet`.
input_tensor: Optional Keras tensor (i.e. output of `layers.Input()`)
to use as image input for the model. `input_tensor` is useful for sharing
inputs between multiple different networks. Default to None.
input_shape: Optional shape tuple, only to be specified
if `include_top` is False (otherwise the input shape
has to be `(299, 299, 3)` (with `channels_last` data format)
or `(3, 299, 299)` (with `channels_first` data format).
It should have exactly 3 inputs channels,
and width and height should be no smaller than 75.
E.g. `(150, 150, 3)` would be one valid value.
`input_shape` will be ignored if the `input_tensor` is provided.
pooling: Optional pooling mode for feature extraction
when `include_top` is `False`.
- `None` (default) means that the output of the model will be
the 4D tensor output of the last convolutional block.
- `avg` means that global average pooling
will be applied to the output of the
last convolutional block, and thus
the output of the model will be a 2D tensor.
- `max` means that global max pooling will be applied.
classes: optional number of classes to classify images
into, only to be specified if `include_top` is True, and
if no `weights` argument is specified. Default to 1000.
classifier_activation: A `str` or callable. The activation function to use
on the ""top"" layer. Ignored unless `include_top=True`. Set
`classifier_activation=None` to return the logits of the ""top"" layer.
Returns:
A `keras.Model` instance.
Raises:
ValueError: in case of invalid argument for `weights`,
or invalid input shape.
ValueError: if `classifier_activation` is not `softmax` or `None` when
using a pretrained top layer."
5128,conv2d_bn,tensorflow/tensorflow/python/keras/applications/inception_v3.py,362,function,"Utility function to apply conv + BN.
Arguments:
x: input tensor.
filters: filters in `Conv2D`.
num_row: height of the convolution kernel.
num_col: width of the convolution kernel.
padding: padding mode in `Conv2D`.
strides: strides in `Conv2D`.
name: name of the ops; will become `name + '_conv'`
for the convolution and `name + '_bn'` for the
batch norm layer.
Returns:
Output tensor after applying `Conv2D` and `BatchNormalization`."
5129,preprocess_input,tensorflow/tensorflow/python/keras/applications/inception_v3.py,408,function,
5130,decode_predictions,tensorflow/tensorflow/python/keras/applications/inception_v3.py,413,function,
5131,MobileNet,tensorflow/tensorflow/python/keras/applications/mobilenet.py,85,function,"Instantiates the MobileNet architecture.
Reference:
- [MobileNets: Efficient Convolutional Neural Networks
for Mobile Vision Applications](
https://arxiv.org/abs/1704.04861)
Optionally loads weights pre-trained on ImageNet.
Note that the data format convention used by the model is
the one specified in the `tf.keras.backend.image_data_format()`.
Caution: Be sure to properly pre-process your inputs to the application.
Please see `applications.mobilenet.preprocess_input` for an example.
Arguments:
input_shape: Optional shape tuple, only to be specified if `include_top`
is False (otherwise the input shape has to be `(224, 224, 3)` (with
`channels_last` data format) or (3, 224, 224) (with `channels_first`
data format). It should have exactly 3 inputs channels, and width and
height should be no smaller than 32. E.g. `(200, 200, 3)` would be one
valid value. Default to `None`.
`input_shape` will be ignored if the `input_tensor` is provided.
alpha: Controls the width of the network. This is known as the width
multiplier in the MobileNet paper. - If `alpha` < 1.0, proportionally
decreases the number of filters in each layer. - If `alpha` > 1.0,
proportionally increases the number of filters in each layer. - If
`alpha` = 1, default number of filters from the paper are used at each
layer. Default to 1.0.
depth_multiplier: Depth multiplier for depthwise convolution. This is
called the resolution multiplier in the MobileNet paper. Default to 1.0.
dropout: Dropout rate. Default to 0.001.
include_top: Boolean, whether to include the fully-connected layer at the
top of the network. Default to `True`.
weights: One of `None` (random initialization), 'imagenet' (pre-training
on ImageNet), or the path to the weights file to be loaded. Default to
`imagenet`.
input_tensor: Optional Keras tensor (i.e. output of `layers.Input()`) to
use as image input for the model. `input_tensor` is useful for sharing
inputs between multiple different networks. Default to None.
pooling: Optional pooling mode for feature extraction when `include_top`
is `False`.
- `None` (default) means that the output of the model will be
the 4D tensor output of the last convolutional block.
- `avg` means that global average pooling
will be applied to the output of the
last convolutional block, and thus
the output of the model will be a 2D tensor.
- `max` means that global max pooling will be applied.
classes: Optional number of classes to classify images into, only to be
specified if `include_top` is True, and if no `weights` argument is
specified. Defaults to 1000.
classifier_activation: A `str` or callable. The activation function to use
on the ""top"" layer. Ignored unless `include_top=True`. Set
`classifier_activation=None` to return the logits of the ""top"" layer.
**kwargs: For backwards compatibility only.
Returns:
A `keras.Model` instance.
Raises:
ValueError: in case of invalid argument for `weights`,
or invalid input shape.
ValueError: if `classifier_activation` is not `softmax` or `None` when
using a pretrained top layer."
5132,preprocess_input,tensorflow/tensorflow/python/keras/applications/mobilenet.py,445,function,
5133,decode_predictions,tensorflow/tensorflow/python/keras/applications/mobilenet.py,450,function,
5134,MobileNetV2,tensorflow/tensorflow/python/keras/applications/mobilenet_v2.py,97,function,"Instantiates the MobileNetV2 architecture.
Reference:
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](
https://arxiv.org/abs/1801.04381) (CVPR 2018)
Optionally loads weights pre-trained on ImageNet.
Caution: Be sure to properly pre-process your inputs to the application.
Please see `applications.mobilenet_v2.preprocess_input` for an example.
Arguments:
input_shape: Optional shape tuple, to be specified if you would
like to use a model with an input image resolution that is not
(224, 224, 3).
It should have exactly 3 inputs channels (224, 224, 3).
You can also omit this option if you would like
to infer input_shape from an input_tensor.
If you choose to include both input_tensor and input_shape then
input_shape will be used if they match, if the shapes
do not match then we will throw an error.
E.g. `(160, 160, 3)` would be one valid value.
alpha: Float between 0 and 1. controls the width of the network.
This is known as the width multiplier in the MobileNetV2 paper,
but the name is kept for consistency with `applications.MobileNetV1`
model in Keras.
- If `alpha` < 1.0, proportionally decreases the number
of filters in each layer.
- If `alpha` > 1.0, proportionally increases the number
of filters in each layer.
- If `alpha` = 1, default number of filters from the paper
are used at each layer.
include_top: Boolean, whether to include the fully-connected
layer at the top of the network. Defaults to `True`.
weights: String, one of `None` (random initialization),
'imagenet' (pre-training on ImageNet),
or the path to the weights file to be loaded.
input_tensor: Optional Keras tensor (i.e. output of
`layers.Input()`)
to use as image input for the model.
pooling: String, optional pooling mode for feature extraction
when `include_top` is `False`.
- `None` means that the output of the model
will be the 4D tensor output of the
last convolutional block.
- `avg` means that global average pooling
will be applied to the output of the
last convolutional block, and thus
the output of the model will be a
2D tensor.
- `max` means that global max pooling will
be applied.
classes: Integer, optional number of classes to classify images
into, only to be specified if `include_top` is True, and
if no `weights` argument is specified.
classifier_activation: A `str` or callable. The activation function to use
on the ""top"" layer. Ignored unless `include_top=True`. Set
`classifier_activation=None` to return the logits of the ""top"" layer.
**kwargs: For backwards compatibility only.
Returns:
A `keras.Model` instance.
Raises:
ValueError: in case of invalid argument for `weights`,
or invalid input shape or invalid alpha, rows when
weights='imagenet'
ValueError: if `classifier_activation` is not `softmax` or `None` when
using a pretrained top layer."
5135,preprocess_input,tensorflow/tensorflow/python/keras/applications/mobilenet_v2.py,501,function,
5136,decode_predictions,tensorflow/tensorflow/python/keras/applications/mobilenet_v2.py,506,function,
5137,NASNet,tensorflow/tensorflow/python/keras/applications/nasnet.py,65,function,"Instantiates a NASNet model.
Reference:
- [Learning Transferable Architectures for Scalable Image Recognition](
https://arxiv.org/abs/1707.07012) (CVPR 2018)
Optionally loads weights pre-trained on ImageNet.
Note that the data format convention used by the model is
the one specified in your Keras config at `~/.keras/keras.json`.
Caution: Be sure to properly pre-process your inputs to the application.
Please see `applications.nasnet.preprocess_input` for an example.
Arguments:
input_shape: Optional shape tuple, the input shape
is by default `(331, 331, 3)` for NASNetLarge and
`(224, 224, 3)` for NASNetMobile.
It should have exactly 3 input channels,
and width and height should be no smaller than 32.
E.g. `(224, 224, 3)` would be one valid value.
penultimate_filters: Number of filters in the penultimate layer.
NASNet models use the notation `NASNet (N @ P)`, where:
- N is the number of blocks
- P is the number of penultimate filters
num_blocks: Number of repeated blocks of the NASNet model.
NASNet models use the notation `NASNet (N @ P)`, where:
- N is the number of blocks
- P is the number of penultimate filters
stem_block_filters: Number of filters in the initial stem block
skip_reduction: Whether to skip the reduction step at the tail
end of the network.
filter_multiplier: Controls the width of the network.
- If `filter_multiplier` < 1.0, proportionally decreases the number
of filters in each layer.
- If `filter_multiplier` > 1.0, proportionally increases the number
of filters in each layer.
- If `filter_multiplier` = 1, default number of filters from the
paper are used at each layer.
include_top: Whether to include the fully-connected
layer at the top of the network.
weights: `None` (random initialization) or
`imagenet` (ImageNet weights)
input_tensor: Optional Keras tensor (i.e. output of
`layers.Input()`)
to use as image input for the model.
pooling: Optional pooling mode for feature extraction
when `include_top` is `False`.
- `None` means that the output of the model
will be the 4D tensor output of the
last convolutional block.
- `avg` means that global average pooling
will be applied to the output of the
last convolutional block, and thus
the output of the model will be a
2D tensor.
- `max` means that global max pooling will
be applied.
classes: Optional number of classes to classify images
into, only to be specified if `include_top` is True, and
if no `weights` argument is specified.
default_size: Specifies the default image size of the model
classifier_activation: A `str` or callable. The activation function to use
on the ""top"" layer. Ignored unless `include_top=True`. Set
`classifier_activation=None` to return the logits of the ""top"" layer.
Returns:
A `keras.Model` instance.
Raises:
ValueError: In case of invalid argument for `weights`,
invalid input shape or invalid `penultimate_filters` value.
ValueError: if `classifier_activation` is not `softmax` or `None` when
using a pretrained top layer."
5138,NASNetMobile,tensorflow/tensorflow/python/keras/applications/nasnet.py,327,function,"Instantiates a Mobile NASNet model in ImageNet mode.
Reference:
- [Learning Transferable Architectures for Scalable Image Recognition](
https://arxiv.org/abs/1707.07012) (CVPR 2018)
Optionally loads weights pre-trained on ImageNet.
Note that the data format convention used by the model is
the one specified in your Keras config at `~/.keras/keras.json`.
Caution: Be sure to properly pre-process your inputs to the application.
Please see `applications.nasnet.preprocess_input` for an example.
Arguments:
input_shape: Optional shape tuple, only to be specified
if `include_top` is False (otherwise the input shape
has to be `(224, 224, 3)` for NASNetMobile
It should have exactly 3 inputs channels,
and width and height should be no smaller than 32.
E.g. `(224, 224, 3)` would be one valid value.
include_top: Whether to include the fully-connected
layer at the top of the network.
weights: `None` (random initialization) or
`imagenet` (ImageNet weights)
For loading `imagenet` weights, `input_shape` should be (224, 224, 3)
input_tensor: Optional Keras tensor (i.e. output of
`layers.Input()`)
to use as image input for the model.
pooling: Optional pooling mode for feature extraction
when `include_top` is `False`.
- `None` means that the output of the model
will be the 4D tensor output of the
last convolutional layer.
- `avg` means that global average pooling
will be applied to the output of the
last convolutional layer, and thus
the output of the model will be a
2D tensor.
- `max` means that global max pooling will
be applied.
classes: Optional number of classes to classify images
into, only to be specified if `include_top` is True, and
if no `weights` argument is specified.
Returns:
A Keras model instance.
Raises:
ValueError: In case of invalid argument for `weights`,
or invalid input shape.
RuntimeError: If attempting to run this model with a
backend that does not support separable convolutions."
5139,NASNetLarge,tensorflow/tensorflow/python/keras/applications/nasnet.py,403,function,"Instantiates a NASNet model in ImageNet mode.
Reference:
- [Learning Transferable Architectures for Scalable Image Recognition](
https://arxiv.org/abs/1707.07012) (CVPR 2018)
Optionally loads weights pre-trained on ImageNet.
Note that the data format convention used by the model is
the one specified in your Keras config at `~/.keras/keras.json`.
Caution: Be sure to properly pre-process your inputs to the application.
Please see `applications.nasnet.preprocess_input` for an example.
Arguments:
input_shape: Optional shape tuple, only to be specified
if `include_top` is False (otherwise the input shape
has to be `(331, 331, 3)` for NASNetLarge.
It should have exactly 3 inputs channels,
and width and height should be no smaller than 32.
E.g. `(224, 224, 3)` would be one valid value.
include_top: Whether to include the fully-connected
layer at the top of the network.
weights: `None` (random initialization) or
`imagenet` (ImageNet weights)
For loading `imagenet` weights, `input_shape` should be (331, 331, 3)
input_tensor: Optional Keras tensor (i.e. output of
`layers.Input()`)
to use as image input for the model.
pooling: Optional pooling mode for feature extraction
when `include_top` is `False`.
- `None` means that the output of the model
will be the 4D tensor output of the
last convolutional layer.
- `avg` means that global average pooling
will be applied to the output of the
last convolutional layer, and thus
the output of the model will be a
2D tensor.
- `max` means that global max pooling will
be applied.
classes: Optional number of classes to classify images
into, only to be specified if `include_top` is True, and
if no `weights` argument is specified.
Returns:
A Keras model instance.
Raises:
ValueError: in case of invalid argument for `weights`,
or invalid input shape.
RuntimeError: If attempting to run this model with a
backend that does not support separable convolutions."
5140,preprocess_input,tensorflow/tensorflow/python/keras/applications/nasnet.py,803,function,
5141,decode_predictions,tensorflow/tensorflow/python/keras/applications/nasnet.py,808,function,
5142,ResNet,tensorflow/tensorflow/python/keras/applications/resnet.py,60,function,"Instantiates the ResNet, ResNetV2, and ResNeXt architecture.
Reference:
- [Deep Residual Learning for Image Recognition](
https://arxiv.org/abs/1512.03385) (CVPR 2015)
Optionally loads weights pre-trained on ImageNet.
Note that the data format convention used by the model is
the one specified in your Keras config at `~/.keras/keras.json`.
Caution: Be sure to properly pre-process your inputs to the application.
Please see `applications.resnet.preprocess_input` for an example.
Arguments:
stack_fn: a function that returns output tensor for the
stacked residual blocks.
preact: whether to use pre-activation or not
(True for ResNetV2, False for ResNet and ResNeXt).
use_bias: whether to use biases for convolutional layers or not
(True for ResNet and ResNetV2, False for ResNeXt).
model_name: string, model name.
include_top: whether to include the fully-connected
layer at the top of the network.
weights: one of `None` (random initialization),
'imagenet' (pre-training on ImageNet),
or the path to the weights file to be loaded.
input_tensor: optional Keras tensor
(i.e. output of `layers.Input()`)
to use as image input for the model.
input_shape: optional shape tuple, only to be specified
if `include_top` is False (otherwise the input shape
has to be `(224, 224, 3)` (with `channels_last` data format)
or `(3, 224, 224)` (with `channels_first` data format).
It should have exactly 3 inputs channels.
pooling: optional pooling mode for feature extraction
when `include_top` is `False`.
- `None` means that the output of the model will be
the 4D tensor output of the
last convolutional layer.
- `avg` means that global average pooling
will be applied to the output of the
last convolutional layer, and thus
the output of the model will be a 2D tensor.
- `max` means that global max pooling will
be applied.
classes: optional number of classes to classify images
into, only to be specified if `include_top` is True, and
if no `weights` argument is specified.
classifier_activation: A `str` or callable. The activation function to use
on the ""top"" layer. Ignored unless `include_top=True`. Set
`classifier_activation=None` to return the logits of the ""top"" layer.
**kwargs: For backwards compatibility only.
Returns:
A `keras.Model` instance.
Raises:
ValueError: in case of invalid argument for `weights`,
or invalid input shape.
ValueError: if `classifier_activation` is not `softmax` or `None` when
using a pretrained top layer."
5143,block1,tensorflow/tensorflow/python/keras/applications/resnet.py,229,function,"A residual block.
Arguments:
x: input tensor.
filters: integer, filters of the bottleneck layer.
kernel_size: default 3, kernel size of the bottleneck layer.
stride: default 1, stride of the first layer.
conv_shortcut: default True, use convolution shortcut if True,
otherwise identity shortcut.
name: string, block label.
Returns:
Output tensor for the residual block."
5144,stack1,tensorflow/tensorflow/python/keras/applications/resnet.py,274,function,"A set of stacked residual blocks.
Arguments:
x: input tensor.
filters: integer, filters of the bottleneck layer in a block.
blocks: integer, blocks in the stacked blocks.
stride1: default 2, stride of the first layer in the first block.
name: string, stack label.
Returns:
Output tensor for the stacked blocks."
5145,block2,tensorflow/tensorflow/python/keras/applications/resnet.py,293,function,"A residual block.
Arguments:
x: input tensor.
filters: integer, filters of the bottleneck layer.
kernel_size: default 3, kernel size of the bottleneck layer.
stride: default 1, stride of the first layer.
conv_shortcut: default False, use convolution shortcut if True,
otherwise identity shortcut.
name: string, block label.
Returns:
Output tensor for the residual block."
5146,stack2,tensorflow/tensorflow/python/keras/applications/resnet.py,342,function,"A set of stacked residual blocks.
Arguments:
x: input tensor.
filters: integer, filters of the bottleneck layer in a block.
blocks: integer, blocks in the stacked blocks.
stride1: default 2, stride of the first layer in the first block.
name: string, stack label.
Returns:
Output tensor for the stacked blocks."
5147,block3,tensorflow/tensorflow/python/keras/applications/resnet.py,362,function,"A residual block.
Arguments:
x: input tensor.
filters: integer, filters of the bottleneck layer.
kernel_size: default 3, kernel size of the bottleneck layer.
stride: default 1, stride of the first layer.
groups: default 32, group size for grouped convolution.
conv_shortcut: default True, use convolution shortcut if True,
otherwise identity shortcut.
name: string, block label.
Returns:
Output tensor for the residual block."
5148,stack3,tensorflow/tensorflow/python/keras/applications/resnet.py,431,function,"A set of stacked residual blocks.
Arguments:
x: input tensor.
filters: integer, filters of the bottleneck layer in a block.
blocks: integer, blocks in the stacked blocks.
stride1: default 2, stride of the first layer in the first block.
groups: default 32, group size for grouped convolution.
name: string, stack label.
Returns:
Output tensor for the stacked blocks."
5149,ResNet50,tensorflow/tensorflow/python/keras/applications/resnet.py,459,function,Instantiates the ResNet50 architecture.
5150,ResNet101,tensorflow/tensorflow/python/keras/applications/resnet.py,480,function,Instantiates the ResNet101 architecture.
5151,ResNet152,tensorflow/tensorflow/python/keras/applications/resnet.py,501,function,Instantiates the ResNet152 architecture.
5152,preprocess_input,tensorflow/tensorflow/python/keras/applications/resnet.py,522,function,
5153,decode_predictions,tensorflow/tensorflow/python/keras/applications/resnet.py,529,function,
5154,ResNet50V2,tensorflow/tensorflow/python/keras/applications/resnet_v2.py,33,function,Instantiates the ResNet50V2 architecture.
5155,ResNet101V2,tensorflow/tensorflow/python/keras/applications/resnet_v2.py,64,function,Instantiates the ResNet101V2 architecture.
5156,ResNet152V2,tensorflow/tensorflow/python/keras/applications/resnet_v2.py,95,function,Instantiates the ResNet152V2 architecture.
5157,preprocess_input,tensorflow/tensorflow/python/keras/applications/resnet_v2.py,125,function,
5158,decode_predictions,tensorflow/tensorflow/python/keras/applications/resnet_v2.py,131,function,
5159,VGG16,tensorflow/tensorflow/python/keras/applications/vgg16.py,46,function,"Instantiates the VGG16 model.
Reference:
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](
https://arxiv.org/abs/1409.1556) (ICLR 2015)
By default, it loads weights pre-trained on ImageNet. Check 'weights' for
other options.
This model can be built both with 'channels_first' data format
(channels, height, width) or 'channels_last' data format
(height, width, channels).
The default input size for this model is 224x224.
Caution: Be sure to properly pre-process your inputs to the application.
Please see `applications.vgg16.preprocess_input` for an example.
Arguments:
include_top: whether to include the 3 fully-connected
layers at the top of the network.
weights: one of `None` (random initialization),
'imagenet' (pre-training on ImageNet),
or the path to the weights file to be loaded.
input_tensor: optional Keras tensor
(i.e. output of `layers.Input()`)
to use as image input for the model.
input_shape: optional shape tuple, only to be specified
if `include_top` is False (otherwise the input shape
has to be `(224, 224, 3)`
(with `channels_last` data format)
or `(3, 224, 224)` (with `channels_first` data format).
It should have exactly 3 input channels,
and width and height should be no smaller than 32.
E.g. `(200, 200, 3)` would be one valid value.
pooling: Optional pooling mode for feature extraction
when `include_top` is `False`.
- `None` means that the output of the model will be
the 4D tensor output of the
last convolutional block.
- `avg` means that global average pooling
will be applied to the output of the
last convolutional block, and thus
the output of the model will be a 2D tensor.
- `max` means that global max pooling will
be applied.
classes: optional number of classes to classify images
into, only to be specified if `include_top` is True, and
if no `weights` argument is specified.
classifier_activation: A `str` or callable. The activation function to use
on the ""top"" layer. Ignored unless `include_top=True`. Set
`classifier_activation=None` to return the logits of the ""top"" layer.
Returns:
A `keras.Model` instance.
Raises:
ValueError: in case of invalid argument for `weights`,
or invalid input shape.
ValueError: if `classifier_activation` is not `softmax` or `None` when
using a pretrained top layer."
5160,preprocess_input,tensorflow/tensorflow/python/keras/applications/vgg16.py,229,function,
5161,decode_predictions,tensorflow/tensorflow/python/keras/applications/vgg16.py,235,function,
5162,VGG19,tensorflow/tensorflow/python/keras/applications/vgg19.py,46,function,"Instantiates the VGG19 architecture.
Reference:
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](
https://arxiv.org/abs/1409.1556) (ICLR 2015)
By default, it loads weights pre-trained on ImageNet. Check 'weights' for
other options.
This model can be built both with 'channels_first' data format
(channels, height, width) or 'channels_last' data format
(height, width, channels).
The default input size for this model is 224x224.
Caution: Be sure to properly pre-process your inputs to the application.
Please see `applications.vgg19.preprocess_input` for an example.
Arguments:
include_top: whether to include the 3 fully-connected
layers at the top of the network.
weights: one of `None` (random initialization),
'imagenet' (pre-training on ImageNet),
or the path to the weights file to be loaded.
input_tensor: optional Keras tensor
(i.e. output of `layers.Input()`)
to use as image input for the model.
input_shape: optional shape tuple, only to be specified
if `include_top` is False (otherwise the input shape
has to be `(224, 224, 3)`
(with `channels_last` data format)
or `(3, 224, 224)` (with `channels_first` data format).
It should have exactly 3 inputs channels,
and width and height should be no smaller than 32.
E.g. `(200, 200, 3)` would be one valid value.
pooling: Optional pooling mode for feature extraction
when `include_top` is `False`.
- `None` means that the output of the model will be
the 4D tensor output of the
last convolutional block.
- `avg` means that global average pooling
will be applied to the output of the
last convolutional block, and thus
the output of the model will be a 2D tensor.
- `max` means that global max pooling will
be applied.
classes: optional number of classes to classify images
into, only to be specified if `include_top` is True, and
if no `weights` argument is specified.
classifier_activation: A `str` or callable. The activation function to use
on the ""top"" layer. Ignored unless `include_top=True`. Set
`classifier_activation=None` to return the logits of the ""top"" layer.
Returns:
A `keras.Model` instance.
Raises:
ValueError: in case of invalid argument for `weights`,
or invalid input shape.
ValueError: if `classifier_activation` is not `softmax` or `None` when
using a pretrained top layer."
5163,preprocess_input,tensorflow/tensorflow/python/keras/applications/vgg19.py,234,function,
5164,decode_predictions,tensorflow/tensorflow/python/keras/applications/vgg19.py,240,function,
5165,Xception,tensorflow/tensorflow/python/keras/applications/xception.py,52,function,"Instantiates the Xception architecture.
Reference:
- [Xception: Deep Learning with Depthwise Separable Convolutions](
https://arxiv.org/abs/1610.02357) (CVPR 2017)
Optionally loads weights pre-trained on ImageNet.
Note that the data format convention used by the model is
the one specified in your Keras config at `~/.keras/keras.json`.
Note that the default input image size for this model is 299x299.
Caution: Be sure to properly pre-process your inputs to the application.
Please see `applications.xception.preprocess_input` for an example.
Arguments:
include_top: whether to include the fully-connected
layer at the top of the network.
weights: one of `None` (random initialization),
'imagenet' (pre-training on ImageNet),
or the path to the weights file to be loaded.
input_tensor: optional Keras tensor
(i.e. output of `layers.Input()`)
to use as image input for the model.
input_shape: optional shape tuple, only to be specified
if `include_top` is False (otherwise the input shape
has to be `(299, 299, 3)`.
It should have exactly 3 inputs channels,
and width and height should be no smaller than 71.
E.g. `(150, 150, 3)` would be one valid value.
pooling: Optional pooling mode for feature extraction
when `include_top` is `False`.
- `None` means that the output of the model will be
the 4D tensor output of the
last convolutional block.
- `avg` means that global average pooling
will be applied to the output of the
last convolutional block, and thus
the output of the model will be a 2D tensor.
- `max` means that global max pooling will
be applied.
classes: optional number of classes to classify images
into, only to be specified if `include_top` is True,
and if no `weights` argument is specified.
classifier_activation: A `str` or callable. The activation function to use
on the ""top"" layer. Ignored unless `include_top=True`. Set
`classifier_activation=None` to return the logits of the ""top"" layer.
Returns:
A `keras.Model` instance.
Raises:
ValueError: in case of invalid argument for `weights`,
or invalid input shape.
ValueError: if `classifier_activation` is not `softmax` or `None` when
using a pretrained top layer."
5166,preprocess_input,tensorflow/tensorflow/python/keras/applications/xception.py,318,function,
5167,decode_predictions,tensorflow/tensorflow/python/keras/applications/xception.py,323,function,
5168,TimerCallBack,tensorflow/tensorflow/python/keras/benchmarks/benchmark_util.py,28,class,Callback for logging time in each epoch or batch.
5169,on_epoch_begin,tensorflow/tensorflow/python/keras/benchmarks/benchmark_util.py,37,method,
5170,on_epoch_end,tensorflow/tensorflow/python/keras/benchmarks/benchmark_util.py,40,method,
5171,on_batch_end,tensorflow/tensorflow/python/keras/benchmarks/benchmark_util.py,43,method,
5172,measure_performance,tensorflow/tensorflow/python/keras/benchmarks/benchmark_util.py,49,function,"Run models and measure the performance.
Arguments:
model_fn: Model function to be benchmarked.
x: Input data. See `x` in the `fit()` method of `keras.Model`.
y: Target data. See `y` in the `fit()` method of `keras.Model`.
epochs: Integer. Number of epochs to train the model.
If unspecified, `epochs` will default to 2.
batch_size: Integer. Number of samples per gradient update. If unspecified,
`batch_size` will default to 32.
run_iters: Integer. Number of iterations to run the performance measurement.
If unspecified, `run_iters` will default to 4.
optimizer: String (name of optimizer) or optimizer instance. See
`tf.keras.optimizers`.
loss: String (name of objective function), objective function or
`tf.keras.losses.Loss` instance. See `tf.keras.losses`.
metrics: Lists of metrics to be evaluated by the model during training. See
`metrics` in the `compile()` method of `keras.Model`.
verbose: 0, 1, 2. Verbosity mode. See `verbose` in the `fit()` method of
`keras.Model`. If unspecified, `verbose` will default to 0.
num_gpus: Number of GPUs to run the model.
distribution_strategy: Distribution strategies. It could be
`multi_worker_mirrored`, `one_device`, `mirrored`. If unspecified,
`distribution_strategy` will default to 'off'. Note that, `TPU`
and `parameter_server` are not supported yet.
Returns:
Performance summary, which contains build_time, compile_time,
startup_time, avg_epoch_time, wall_time, exp_per_sec, epochs,
distribution_strategy.
Raise:
ValueError: If `x` is none or if `optimizer` is not provided or
if `loss` is not provided or if `num_gpus` is negative."
5173,get_distribution_strategy,tensorflow/tensorflow/python/keras/benchmarks/distribution_util.py,86,function,"Return a DistributionStrategy for running the model.
Args:
distribution_strategy: a string specifying which distribution strategy to
use. Accepted values are ""off"", ""one_device"", ""mirrored"", and
""multi_worker_mirrored"" -- case insensitive. ""off"" means not to use
Distribution Strategy.
num_gpus: Number of GPUs to run this model.
Returns:
tf.distribute.DistibutionStrategy object.
Raises:
ValueError: if `distribution_strategy` is ""off"" or ""one_device"" and
`num_gpus` is larger than 1; or `num_gpus` is negative."
5174,configure_cluster,tensorflow/tensorflow/python/keras/benchmarks/distribution_util.py,141,function,"Set multi-worker cluster spec in TF_CONFIG environment variable.
Args:
worker_hosts: comma-separated list of worker ip:port pairs.
Returns:
Number of workers in the cluster."
5175,get_strategy_scope,tensorflow/tensorflow/python/keras/benchmarks/distribution_util.py,175,function,
5176,DummyContextManager,tensorflow/tensorflow/python/keras/benchmarks/distribution_util.py,184,class,
5177,MicroBenchmarksBase,tensorflow/tensorflow/python/keras/benchmarks/eager_microbenchmarks_test.py,47,class,Run and report benchmark results.
5178,run_report,tensorflow/tensorflow/python/keras/benchmarks/eager_microbenchmarks_test.py,50,method,Run and report benchmark results.
5179,benchmark_layers_call_overhead,tensorflow/tensorflow/python/keras/benchmarks/eager_microbenchmarks_test.py,88,method,
5180,benchmark_op_layer_call_overhead,tensorflow/tensorflow/python/keras/benchmarks/eager_microbenchmarks_test.py,103,method,
5181,benchmark_model_predict_tensorlike_overhead,tensorflow/tensorflow/python/keras/benchmarks/eager_microbenchmarks_test.py,118,method,
5182,benchmark_layers_embeddings_embedding_overhead,tensorflow/tensorflow/python/keras/benchmarks/eager_microbenchmarks_test.py,133,method,
5183,fn,tensorflow/tensorflow/python/keras/benchmarks/eager_microbenchmarks_test.py,98,method,
5184,fn,tensorflow/tensorflow/python/keras/benchmarks/eager_microbenchmarks_test.py,112,method,
5185,fn,tensorflow/tensorflow/python/keras/benchmarks/eager_microbenchmarks_test.py,128,method,
5186,fn,tensorflow/tensorflow/python/keras/benchmarks/eager_microbenchmarks_test.py,138,method,
5187,call,tensorflow/tensorflow/python/keras/benchmarks/eager_microbenchmarks_test.py,92,method,
5188,call,tensorflow/tensorflow/python/keras/benchmarks/eager_microbenchmarks_test.py,122,method,
5189,KerasLayerCallOverheadBenchmarks,tensorflow/tensorflow/python/keras/benchmarks/eager_microbenchmarks_test.py,144,class,
5190,benchmark_layer,tensorflow/tensorflow/python/keras/benchmarks/eager_microbenchmarks_test.py,190,method,
5191,fn,tensorflow/tensorflow/python/keras/benchmarks/eager_microbenchmarks_test.py,194,method,
5192,KerasModelCPUBenchmark,tensorflow/tensorflow/python/keras/benchmarks/keras_cpu_benchmark_test.py,33,class,"Required Arguments for measure_performance.
x: Input data, it could be Numpy or load from tfds.
y: Target data. If `x` is a dataset, generator instance,
`y` should not be specified.
loss: Loss function for model.
optimizer: Optimizer for model.
Other details can see in `measure_performance()` method of
benchmark_util."
5193,benchmark_mnist_mlp,tensorflow/tensorflow/python/keras/benchmarks/keras_cpu_benchmark_test.py,94,method,Benchmark for MLP model on synthetic mnist data.
5194,benchmark_mnist_convnet,tensorflow/tensorflow/python/keras/benchmarks/keras_cpu_benchmark_test.py,109,method,Benchmark for Convnet model on synthetic mnist data.
5195,benchmark_imdb_lstm,tensorflow/tensorflow/python/keras/benchmarks/keras_cpu_benchmark_test.py,124,method,Benchmark for LSTM model on synthetic imdb review dataset.
5196,SubclassedKerasModel,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,32,class,
5197,call,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,47,method,
5198,make_keras_model,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,55,function,
5199,make_sequential_keras_model,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,70,function,
5200,run_benchmark,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,86,function,
5201,KerasComponentsBenchmarks,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,103,class,
5202,benchmark_keras_model_subclassed,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,118,method,
5203,benchmark_keras_model_functional,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,133,method,
5204,benchmark_keras_model_sequential,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,142,method,
5205,benchmark_keras_model_subclassed_fit,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,189,method,
5206,benchmark_keras_model_subclassed_fit_graph_mode,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,193,method,
5207,benchmark_keras_model_subclassed_fit_run_model_eagerly,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,198,method,
5208,benchmark_keras_model_functional_fit,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,202,method,
5209,benchmark_keras_model_functional_fit_graph_mode,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,206,method,
5210,benchmark_keras_model_functional_fit_graph_mode_with_profiler,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,211,method,
5211,benchmark_keras_model_functional_fit_run_model_eagerly,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,219,method,
5212,benchmark_keras_model_functional_fit_run_model_eagerly_with_profiler,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,223,method,
5213,benchmark_keras_model_sequential_fit,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,231,method,
5214,benchmark_keras_model_sequential_fit_graph_mode,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,235,method,
5215,benchmark_keras_model_sequential_fit_run_model_eagerly,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,240,method,
5216,benchmark_keras_model_subclassed_evaluate,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,244,method,
5217,benchmark_keras_model_subclassed_evaluate_run_model_eagerly,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,248,method,
5218,benchmark_keras_model_functional_evaluate,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,252,method,
5219,benchmark_keras_model_functional_evaluate_run_model_eagerly,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,256,method,
5220,benchmark_keras_model_sequential_evaluate,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,260,method,
5221,benchmark_keras_model_sequential_evaluate_run_model_eagerly,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,264,method,
5222,benchmark_keras_model_subclassed_predict,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,268,method,
5223,benchmark_keras_model_subclassed_predict_run_model_eagerly,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,272,method,
5224,benchmark_keras_model_functional_predict,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,276,method,
5225,benchmark_keras_model_functional_predict_run_model_eagerly,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,280,method,
5226,benchmark_keras_model_sequential_predict,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,284,method,
5227,benchmark_keras_model_sequential_predict_run_model_eagerly,tensorflow/tensorflow/python/keras/benchmarks/model_components_benchmarks_test.py,288,method,
5228,AntirectifierBenchmark,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/antirectifier_benchmark_test.py,25,class,Benchmarks for Antirectifier using `tf.test.Benchmark`.
5229,benchmark_pixel_cnn_bs_128,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/antirectifier_benchmark_test.py,56,method,Measure performance with batch_size=128 and run_iters=2.
5230,benchmark_pixel_cnn_bs_256,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/antirectifier_benchmark_test.py,73,method,Measure performance with batch_size=256 and run_iters=3.
5231,benchmark_pixel_cnn_bs_512,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/antirectifier_benchmark_test.py,90,method,Measure performance with batch_size=512 and run_iters=4.
5232,Antirectifier,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/antirectifier_benchmark_test.py,108,class,Build simple custome layer.
5233,build,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/antirectifier_benchmark_test.py,115,method,
5234,call,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/antirectifier_benchmark_test.py,124,method,
5235,get_config,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/antirectifier_benchmark_test.py,132,method,
5236,BidirectionalLSTMBenchmark,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/bidirectional_lstm_benchmark_test.py,25,class,Benchmarks for Bidirectional LSTM using `tf.test.Benchmark`.
5237,benchmark_bidirect_lstm_imdb_bs_128,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/bidirectional_lstm_benchmark_test.py,58,method,Measure performance with batch_size=128 and run_iters=3.
5238,benchmark_bidirect_lstm_imdb_bs_256,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/bidirectional_lstm_benchmark_test.py,75,method,Measure performance with batch_size=256 and run_iters=2.
5239,benchmark_bidirect_lstm_imdb_bs_512,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/bidirectional_lstm_benchmark_test.py,92,method,Measure performance with batch_size=512 and run_iters=4.
5240,Cifar10CNNBenchmark,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/cifar10_cnn_benchmark_test.py,25,class,Benchmarks for CNN using `tf.test.Benchmark`.
5241,benchmark_cnn_cifar10_bs_256,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/cifar10_cnn_benchmark_test.py,72,method,Measure performance with batch_size=256 and run_iters=3.
5242,benchmark_cnn_cifar10_bs_512,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/cifar10_cnn_benchmark_test.py,90,method,Measure performance with batch_size=512 and run_iters=3.
5243,benchmark_cnn_cifar10_bs_1024,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/cifar10_cnn_benchmark_test.py,108,method,Measure performance with batch_size=1024 and run_iters=2.
5244,ConvMnistBenchmark,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/mnist_conv_benchmark_test.py,27,class,Benchmarks for Convnet using `tf.test.Benchmark`.
5245,benchmark_conv_mnist_bs_128,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/mnist_conv_benchmark_test.py,63,method,Measure performance with batch_size=128 and run_iters=2.
5246,benchmark_conv_mnist_bs_256,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/mnist_conv_benchmark_test.py,81,method,Measure performance with batch_size=256 and run_iters=3.
5247,benchmark_conv_mnist_bs_512,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/mnist_conv_benchmark_test.py,99,method,Measure performance with batch_size=512 and run_iters=3.
5248,HierarchicalRNNBenchmark,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/mnist_hierarchical_rnn_benchmark_test.py,25,class,Benchmarks for Hierarchical RNN using `tf.test.Benchmark`.
5249,benchmark_hrnn_mnist_bs_256,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/mnist_hierarchical_rnn_benchmark_test.py,64,method,Measure performance with batch_size=256 and run_iters=4.
5250,benchmark_hrnn_mnist_bs_512,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/mnist_hierarchical_rnn_benchmark_test.py,81,method,Measure performance with batch_size=512 and run_iters=5.
5251,benchmark_hrnn_mnist_bs_1024,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/mnist_hierarchical_rnn_benchmark_test.py,98,method,Measure performance with batch_size=1024 and run_iters=3.
5252,IRNNMnistBenchmark,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/mnist_irnn_benchmark_test.py,25,class,Benchmarks for IRNN using `tf.test.Benchmark`.
5253,benchmark_irnn_mnist_bs_256,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/mnist_irnn_benchmark_test.py,64,method,Measure performance with batch_size=256 and run_iters=4.
5254,benchmark_irnn_mnist_bs_512,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/mnist_irnn_benchmark_test.py,81,method,Measure performance with batch_size=512 and run_iters=3.
5255,benchmark_irnn_mnist_bs_1024,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/mnist_irnn_benchmark_test.py,98,method,Measure performance with batch_size=1024 and run_iters=3.
5256,MLPReutersBenchmark,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/reuters_mlp_benchmark_test.py,27,class,Benchmarks for MLP using `tf.test.Benchmark`.
5257,benchmark_mlp_reuters_bs_128,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/reuters_mlp_benchmark_test.py,63,method,Measure performance with batch_size=128 and run_iters=2.
5258,benchmark_mlp_reuters_bs_256,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/reuters_mlp_benchmark_test.py,81,method,Measure performance with batch_size=256 and run_iters=3.
5259,benchmark_mlp_reuters_bs_512,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/reuters_mlp_benchmark_test.py,99,method,Measure performance with batch_size=512 and run_iters=4.
5260,TextWithTransformerBenchmark,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/text_classification_transformer_benchmark_test.py,25,class,"Benchmarks for Text classification with Transformer
using `tf.test.Benchmark`."
5261,benchmark_text_classification_bs_128,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/text_classification_transformer_benchmark_test.py,68,method,Measure performance with batch_size=128 and run_iters=3.
5262,benchmark_text_classification_bs_512,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/text_classification_transformer_benchmark_test.py,85,method,Measure performance with batch_size=512 and run_iters=4.
5263,benchmark_text_classification_bs_256,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/text_classification_transformer_benchmark_test.py,102,method,Measure performance with batch_size=256 and run_iters=3.
5264,MultiHeadSelfAttention,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/text_classification_transformer_benchmark_test.py,120,class,Implement multi head self attention as a Keras layer.
5265,attention,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/text_classification_transformer_benchmark_test.py,136,method,
5266,separate_heads,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/text_classification_transformer_benchmark_test.py,144,method,
5267,call,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/text_classification_transformer_benchmark_test.py,148,method,
5268,TransformerBlock,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/text_classification_transformer_benchmark_test.py,172,class,Implement a Transformer block as a layer.
5269,call,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/text_classification_transformer_benchmark_test.py,187,method,
5270,TokenAndPositionEmbedding,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/text_classification_transformer_benchmark_test.py,196,class,Implement embedding layer.
5271,call,tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/text_classification_transformer_benchmark_test.py,206,method,
5272,BenchmarkSaveApplications,tensorflow/tensorflow/python/keras/benchmarks/saved_model_benchmarks/densenet_benchmark_test.py,25,class,
5273,benchmark_save_and_load_densenet_201,tensorflow/tensorflow/python/keras/benchmarks/saved_model_benchmarks/densenet_benchmark_test.py,27,method,
5274,BenchmarkSaveApplications,tensorflow/tensorflow/python/keras/benchmarks/saved_model_benchmarks/efficientnet_benchmark_test.py,25,class,
5275,benchmark_save_and_load_efficient_net_b7,tensorflow/tensorflow/python/keras/benchmarks/saved_model_benchmarks/efficientnet_benchmark_test.py,27,method,
5276,BenchmarkSaveApplications,tensorflow/tensorflow/python/keras/benchmarks/saved_model_benchmarks/inception_resnet_v2_benchmark_test.py,25,class,
5277,benchmark_save_and_load_inception_resnet_v2,tensorflow/tensorflow/python/keras/benchmarks/saved_model_benchmarks/inception_resnet_v2_benchmark_test.py,27,method,
5278,BenchmarkSaveApplications,tensorflow/tensorflow/python/keras/benchmarks/saved_model_benchmarks/mobilenet_benchmark_test.py,25,class,
5279,benchmark_save_and_load_mobilenet_v2,tensorflow/tensorflow/python/keras/benchmarks/saved_model_benchmarks/mobilenet_benchmark_test.py,27,method,
5280,BenchmarkSaveApplications,tensorflow/tensorflow/python/keras/benchmarks/saved_model_benchmarks/nasnet_large_benchmark_test.py,25,class,
5281,benchmark_save_and_load_nasnet_large,tensorflow/tensorflow/python/keras/benchmarks/saved_model_benchmarks/nasnet_large_benchmark_test.py,27,method,
5282,BenchmarkSaveApplications,tensorflow/tensorflow/python/keras/benchmarks/saved_model_benchmarks/resnet152_v2_benchmark_test.py,25,class,
5283,benchmark_save_and_load_resnet152_v2,tensorflow/tensorflow/python/keras/benchmarks/saved_model_benchmarks/resnet152_v2_benchmark_test.py,27,method,
5284,save_and_load_benchmark,tensorflow/tensorflow/python/keras/benchmarks/saved_model_benchmarks/saved_model_benchmark_util.py,30,function,Util for saved model benchmarks.
5285,BenchmarkSaveApplications,tensorflow/tensorflow/python/keras/benchmarks/saved_model_benchmarks/vgg_benchmark_test.py,25,class,
5286,benchmark_save_and_load_vgg19,tensorflow/tensorflow/python/keras/benchmarks/saved_model_benchmarks/vgg_benchmark_test.py,27,method,
5287,BenchmarkSaveApplications,tensorflow/tensorflow/python/keras/benchmarks/saved_model_benchmarks/xception_benchmark_test.py,25,class,
5288,benchmark_save_and_load_xception,tensorflow/tensorflow/python/keras/benchmarks/saved_model_benchmarks/xception_benchmark_test.py,27,method,
5289,load_data,tensorflow/tensorflow/python/keras/datasets/boston_housing.py,28,function,"Loads the Boston Housing dataset.
This is a dataset taken from the StatLib library which is maintained at
Carnegie Mellon University.
Samples contain 13 attributes of houses at different locations around the
Boston suburbs in the late 1970s. Targets are the median values of
the houses at a location (in k$).
The attributes themselves are defined in the
[StatLib website](http://lib.stat.cmu.edu/datasets/boston).
Arguments:
path: path where to cache the dataset locally
(relative to `~/.keras/datasets`).
test_split: fraction of the data to reserve as test set.
seed: Random seed for shuffling the data
before computing the test split.
Returns:
Tuple of Numpy arrays: `(x_train, y_train), (x_test, y_test)`.
**x_train, x_test**: numpy arrays with shape `(num_samples, 13)`
containing either the training samples (for x_train),
or test samples (for y_train).
**y_train, y_test**: numpy arrays of shape `(num_samples,)` containing the
target scalars. The targets are float scalars typically between 10 and
50 that represent the home prices in k$."
5290,load_batch,tensorflow/tensorflow/python/keras/datasets/cifar.py,26,function,"Internal utility for parsing CIFAR data.
Arguments:
fpath: path the file to parse.
label_key: key for label data in the retrieve
dictionary.
Returns:
A tuple `(data, labels)`."
5291,load_data,tensorflow/tensorflow/python/keras/datasets/cifar10.py,32,function,"Loads [CIFAR10 dataset](https://www.cs.toronto.edu/~kriz/cifar.html).
This is a dataset of 50,000 32x32 color training images and 10,000 test
images, labeled over 10 categories. See more info at the
[CIFAR homepage](https://www.cs.toronto.edu/~kriz/cifar.html).
Returns:
Tuple of Numpy arrays: `(x_train, y_train), (x_test, y_test)`.
**x_train, x_test**: uint8 arrays of RGB image data with shape
`(num_samples, 3, 32, 32)` if `tf.keras.backend.image_data_format()` is
`'channels_first'`, or `(num_samples, 32, 32, 3)` if the data format
is `'channels_last'`.
**y_train, y_test**: uint8 arrays of category labels
(integers in range 0-9) each with shape (num_samples, 1)."
5292,load_data,tensorflow/tensorflow/python/keras/datasets/cifar100.py,32,function,"Loads [CIFAR100 dataset](https://www.cs.toronto.edu/~kriz/cifar.html).
This is a dataset of 50,000 32x32 color training images and
10,000 test images, labeled over 100 fine-grained classes that are
grouped into 20 coarse-grained classes. See more info at the
[CIFAR homepage](https://www.cs.toronto.edu/~kriz/cifar.html).
Arguments:
label_mode: one of ""fine"", ""coarse"". If it is ""fine"" the category labels
are the fine-grained labels, if it is ""coarse"" the output labels are the
coarse-grained superclasses.
Returns:
Tuple of Numpy arrays: `(x_train, y_train), (x_test, y_test)`.
**x_train, x_test**: uint8 arrays of RGB image data with shape
`(num_samples, 3, 32, 32)` if `tf.keras.backend.image_data_format()` is
`'channels_first'`, or `(num_samples, 32, 32, 3)` if the data format
is `'channels_last'`.
**y_train, y_test**: uint8 arrays of category labels with shape
(num_samples, 1).
Raises:
ValueError: in case of invalid `label_mode`."
5293,load_data,tensorflow/tensorflow/python/keras/datasets/fashion_mnist.py,31,function,"Loads the Fashion-MNIST dataset.
This is a dataset of 60,000 28x28 grayscale images of 10 fashion categories,
along with a test set of 10,000 images. This dataset can be used as
a drop-in replacement for MNIST. The class labels are:
| Label | Description |
|:-----:|-------------|
| 0 | T-shirt/top |
| 1 | Trouser |
| 2 | Pullover |
| 3 | Dress |
| 4 | Coat |
| 5 | Sandal |
| 6 | Shirt |
| 7 | Sneaker |
| 8 | Bag |
| 9 | Ankle boot |
Returns:
Tuple of Numpy arrays: `(x_train, y_train), (x_test, y_test)`.
**x_train, x_test**: uint8 arrays of grayscale image data with shape
(num_samples, 28, 28).
**y_train, y_test**: uint8 arrays of labels (integers in range 0-9)
with shape (num_samples,).
License:
The copyright for Fashion-MNIST is held by Zalando SE.
Fashion-MNIST is licensed under the [MIT license](
https://github.com/zalandoresearch/fashion-mnist/blob/master/LICENSE)."
5294,load_data,tensorflow/tensorflow/python/keras/datasets/imdb.py,32,function,"Loads the [IMDB dataset](https://ai.stanford.edu/~amaas/data/sentiment/).
This is a dataset of 25,000 movies reviews from IMDB, labeled by sentiment
(positive/negative). Reviews have been preprocessed, and each review is
encoded as a list of word indexes (integers).
For convenience, words are indexed by overall frequency in the dataset,
so that for instance the integer ""3"" encodes the 3rd most frequent word in
the data. This allows for quick filtering operations such as:
""only consider the top 10,000 most
common words, but eliminate the top 20 most common words"".
As a convention, ""0"" does not stand for a specific word, but instead is used
to encode any unknown word.
Arguments:
path: where to cache the data (relative to `~/.keras/dataset`).
num_words: integer or None. Words are
ranked by how often they occur (in the training set) and only
the `num_words` most frequent words are kept. Any less frequent word
will appear as `oov_char` value in the sequence data. If None,
all words are kept. Defaults to None, so all words are kept.
skip_top: skip the top N most frequently occurring words
(which may not be informative). These words will appear as
`oov_char` value in the dataset. Defaults to 0, so no words are
skipped.
maxlen: int or None. Maximum sequence length.
Any longer sequence will be truncated. Defaults to None, which
means no truncation.
seed: int. Seed for reproducible data shuffling.
start_char: int. The start of a sequence will be marked with this
character. Defaults to 1 because 0 is usually the padding character.
oov_char: int. The out-of-vocabulary character.
Words that were cut out because of the `num_words` or
`skip_top` limits will be replaced with this character.
index_from: int. Index actual words with this index and higher.
**kwargs: Used for backwards compatibility.
Returns:
Tuple of Numpy arrays: `(x_train, y_train), (x_test, y_test)`.
**x_train, x_test**: lists of sequences, which are lists of indexes
(integers). If the num_words argument was specific, the maximum
possible index value is `num_words - 1`. If the `maxlen` argument was
specified, the largest possible sequence length is `maxlen`.
**y_train, y_test**: lists of integer labels (1 or 0).
Raises:
ValueError: in case `maxlen` is so low
that no input sequence could be kept.
Note that the 'out of vocabulary' character is only used for
words that were present in the training set but are not included
because they're not making the `num_words` cut here.
Words that were not seen in the training set but are in the test set
have simply been skipped."
5295,get_word_index,tensorflow/tensorflow/python/keras/datasets/imdb.py,166,function,"Retrieves a dict mapping words to their index in the IMDB dataset.
Arguments:
path: where to cache the data (relative to `~/.keras/dataset`).
Returns:
The word index dictionary. Keys are word strings, values are their index."
5296,load_data,tensorflow/tensorflow/python/keras/datasets/mnist.py,28,function,"Loads the [MNIST dataset](http://yann.lecun.com/exdb/mnist/).
This is a dataset of 60,000 28x28 grayscale images of the 10 digits,
along with a test set of 10,000 images.
More info can be found at the
[MNIST homepage](http://yann.lecun.com/exdb/mnist/).
Arguments:
path: path where to cache the dataset locally
(relative to `~/.keras/datasets`).
Returns:
Tuple of Numpy arrays: `(x_train, y_train), (x_test, y_test)`.
**x_train, x_test**: uint8 arrays of grayscale image data with shapes
(num_samples, 28, 28).
**y_train, y_test**: uint8 arrays of digit labels (integers in range 0-9)
with shapes (num_samples,).
License:
Yann LeCun and Corinna Cortes hold the copyright of MNIST dataset,
which is a derivative work from original NIST datasets.
MNIST dataset is made available under the terms of the
[Creative Commons Attribution-Share Alike 3.0 license.](
https://creativecommons.org/licenses/by-sa/3.0/)"
5297,load_data,tensorflow/tensorflow/python/keras/datasets/reuters.py,32,function,"Loads the Reuters newswire classification dataset.
This is a dataset of 11,228 newswires from Reuters, labeled over 46 topics.
This was originally generated by parsing and preprocessing the classic
Reuters-21578 dataset, but the preprocessing code is no longer packaged
with Keras. See this
[github discussion](https://github.com/keras-team/keras/issues/12072)
for more info.
Each newswire is encoded as a list of word indexes (integers).
For convenience, words are indexed by overall frequency in the dataset,
so that for instance the integer ""3"" encodes the 3rd most frequent word in
the data. This allows for quick filtering operations such as:
""only consider the top 10,000 most
common words, but eliminate the top 20 most common words"".
As a convention, ""0"" does not stand for a specific word, but instead is used
to encode any unknown word.
Arguments:
path: where to cache the data (relative to `~/.keras/dataset`).
num_words: integer or None. Words are
ranked by how often they occur (in the training set) and only
the `num_words` most frequent words are kept. Any less frequent word
will appear as `oov_char` value in the sequence data. If None,
all words are kept. Defaults to None, so all words are kept.
skip_top: skip the top N most frequently occurring words
(which may not be informative). These words will appear as
`oov_char` value in the dataset. Defaults to 0, so no words are
skipped.
maxlen: int or None. Maximum sequence length.
Any longer sequence will be truncated. Defaults to None, which
means no truncation.
test_split: Float between 0 and 1. Fraction of the dataset to be used
as test data. Defaults to 0.2, meaning 20% of the dataset is used as
test data.
seed: int. Seed for reproducible data shuffling.
start_char: int. The start of a sequence will be marked with this
character. Defaults to 1 because 0 is usually the padding character.
oov_char: int. The out-of-vocabulary character.
Words that were cut out because of the `num_words` or
`skip_top` limits will be replaced with this character.
index_from: int. Index actual words with this index and higher.
**kwargs: Used for backwards compatibility.
Returns:
Tuple of Numpy arrays: `(x_train, y_train), (x_test, y_test)`.
**x_train, x_test**: lists of sequences, which are lists of indexes
(integers). If the num_words argument was specific, the maximum
possible index value is `num_words - 1`. If the `maxlen` argument was
specified, the largest possible sequence length is `maxlen`.
**y_train, y_test**: lists of integer labels (1 or 0).
Note: The 'out of vocabulary' character is only used for
words that were present in the training set but are not included
because they're not making the `num_words` cut here.
Words that were not seen in the training set but are in the test set
have simply been skipped."
5298,get_word_index,tensorflow/tensorflow/python/keras/datasets/reuters.py,155,function,"Retrieves a dict mapping words to their index in the Reuters dataset.
Arguments:
path: where to cache the data (relative to `~/.keras/dataset`).
Returns:
The word index dictionary. Keys are word strings, values are their index."
5299,LocalCollectiveAllReduceStrategy,tensorflow/tensorflow/python/keras/distribute/collective_all_reduce_strategy_test.py,300,class,
5300,MaybeStrategyScope,tensorflow/tensorflow/python/keras/distribute/ctl_correctness_test.py,48,class,Provides a context allowing no distribution strategy.
5301,get_model,tensorflow/tensorflow/python/keras/distribute/ctl_correctness_test.py,66,function,
5302,get_data,tensorflow/tensorflow/python/keras/distribute/ctl_correctness_test.py,81,function,
5303,compute_loss,tensorflow/tensorflow/python/keras/distribute/ctl_correctness_test.py,91,function,
5304,iteration_inside_func,tensorflow/tensorflow/python/keras/distribute/ctl_correctness_test.py,99,function,Helper function to test iterating over data inside a tf.function.
5305,iteration_outside_func,tensorflow/tensorflow/python/keras/distribute/ctl_correctness_test.py,163,function,Helper function to test iterating over data outside a tf.function.
5306,CustomModel,tensorflow/tensorflow/python/keras/distribute/custom_training_loop_models_test.py,39,class,
5307,simple_sequential_model,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,70,function,
5308,simple_subclassed_model,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,78,function,
5309,simple_multi_inputs_multi_outputs_model,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,92,function,
5310,get_multi_inputs_multi_outputs_data,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,104,function,
5311,batch_wrapper,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,147,function,
5312,get_model,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,159,function,
5313,get_sample_weights_model,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,166,function,
5314,get_dataset,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,175,function,
5315,get_predict_dataset,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,184,function,
5316,convert_numpy_to_dataset_with_unknown_cardinality,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,192,function,
5317,multi_input_output_model,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,206,function,
5318,strategy_minus_tpu_combinations,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,245,function,
5319,tpu_strategy_combinations,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,250,function,
5320,tpu_strategy_combinations_graph_only,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,255,function,
5321,all_strategy_combinations,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,259,function,
5322,all_strategy_minus_default_and_tpu_combinations,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,263,function,
5323,all_strategy_combinations_minus_default,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,274,function,
5324,strategy_and_optimizer_combinations,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,279,function,
5325,BatchCountingCB,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,322,class,
5326,on_train_batch_begin,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,333,method,
5327,on_train_batch_end,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,336,method,
5328,on_predict_batch_begin,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,345,method,
5329,on_predict_batch_end,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,348,method,
5330,DeterministicModel,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,2416,class,"Deterministic Model that always outputs the same initial result.
It verifies the `call` method is run inside the same distribution
strategy that the model was initially passed."
5331,build,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,2428,method,
5332,call,tensorflow/tensorflow/python/keras/distribute/distribute_strategy_test.py,2431,method,
5333,set_weights,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,55,function,"Sets the weights of the replicated models.
The weights of the replicated models are set to the weights of the original
model. The weights of the replicated model are Mirrored variables and hence
we need to use the `update` call within a DistributionStrategy scope.
Args:
distribution_strategy: DistributionStrategy used to distribute training
and validation.
dist_model: The replicated models on the different devices.
weights: The weights of the original model."
5334,unwrap_values,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,83,function,"Unwrap the list of values contained in the PerReplica parameters.
This function calls `flatten_per_replica_values` to parse each of the input
parameters into a list of values on the different devices. If we set
`with_loss_tensor` to be True, we also call `reduce` on the list of losses on
the different devices to give us one loss tensor.
Args:
distribution_strategy: DistributionStrategy used to distribute training and
validation.
grouped_inputs: PerReplica inputs returned from the train or test function
that we ran on each device.
grouped_outputs: PerReplica outputs returned from the train or test function
that we ran on each device.
grouped_updates: PerReplica updates returned from the train or test function
that we ran on each device.
grouped_session_args: PerReplica session args returned from the train or
test function that we ran on each device.
with_loss_tensor: Boolean that indicates if we need to add the reduced loss
tensor as one of the outputs.
Returns:
Values of each of the PerReplica parameters."
5335,unwrap_output_dict,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,140,function,Unwrap the list of outputs contained in the PerReplica parameters.
5336,unwrap_outputs,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,172,function,"Unwrap the list of outputs contained in the PerReplica parameters.
This function calls `flatten_per_replica_values` to parse each of the input
parameters into a list of outputs on the different devices. If we set
`with_loss_tensor` to be True, we also call `reduce` on the list of losses on
the different devices to give us one loss tensor.
Args:
distribution_strategy: DistributionStrategy used to distribute training and
validation.
grouped_outputs: PerReplica outputs returned from the train or test function
that we ran on each device.
with_loss_tensor: Boolean that indicates if we need to add the reduced loss
tensor as one of the outputs.
Returns:
Values of each of the PerReplica outputs."
5337,flatten_per_replica_values,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,216,function,"Unwraps and flattens a nest of PerReplica parameters.
PerReplica values have one value associated with each device. Each entry in
the PerReplica dict has a device `key` and the corresponding value on the
device as the `value`. In this function we take a PerReplica value or a list
of PerReplica values and return all the values in the PerReplica dict.
Args:
distribution_strategy: DistributionStrategy used to distribute training and
validation.
per_replica_values: List of PerReplica object or a single PerReplica object.
Returns:
List of values of all the PerReplica objects."
5338,validate_callbacks,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,240,function,"Validate whether given callbacks are supported by DistributionStrategy.
Args:
input_callbacks: List of callbacks passed by the user to fit.
optimizer: Optimizer instance used to train the model.
Raises:
ValueError: If `LearningRateScheduler` or `ReduceLROnPlateau` is one of the
callbacks passed.
ValueError: If `write_grads` is one of the parameters passed as part of the
TensorBoard callback."
5339,validate_distributed_dataset_inputs,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,275,function,"Validate all the components of a DistributedValue Dataset input.
Args:
distribution_strategy: The current DistributionStrategy used to call
`fit`/`evaluate`.
x: Input Dataset DistributedValue object. For example, when we use
`MirroredStrategy` this is a PerReplica object with a tensor for each
device set in the dict. x can also be a tuple or dict. The keys of the
dict should match the names of the input layers of the model.
y: Target Dataset DistributedValue object. For example, when we use
`MirroredStrategy` this is a PerReplica object with a tensor for each
device set in the dict. y can also be a tuple or dict. The keys of the
dict should match the names of the output layers of the model.
sample_weights: Sample weights Dataset DistributedValue object. For example,
when we use `MirroredStrategy` this is a PerReplica object with a tensor
for each device set in the dict.
Returns:
The unwrapped values list of the x and y DistributedValues inputs.
Raises:
ValueError: If x and y do not have support for being evaluated as tensors.
or if x and y contain elements that are not tensors or if x and y
contain elements that have a shape or dtype mismatch."
5340,validate_per_replica_inputs,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,325,function,"Validates PerReplica dataset input list.
Args:
distribution_strategy: The current DistributionStrategy used to call
`fit`, `evaluate` and `predict`.
x: A list of PerReplica objects that represent the input or
target values.
Returns:
List containing the first element of each of the PerReplica objects in
the input list.
Raises:
ValueError: If any of the objects in the `per_replica_list` is not a tensor."
5341,validate_all_tensor_types,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,362,function,
5342,validate_all_tensor_shapes,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,370,function,
5343,init_restore_or_wait_for_variables,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,402,function,Initialize or restore variables or wait for variables to be initialized.
5344,validate_inputs,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,413,function,"Validate inputs when using DistributionStrategy.
Args:
x: Model Inputs.
y: Model Targets.
Raises:
ValueError: if input is not a Dataset or a numpy array(when we use
MirroredStrategy)."
5345,global_batch_size_supported,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,434,function,
5346,is_tpu_strategy,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,439,function,We're executing TPU Strategy.
5347,is_dataset_shape_fully_defined,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,445,function,Returns whether a dataset contains a final partial batch.
5348,process_batch_and_step_size,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,452,function,Process the batch size and step size based on input and dist strategy.
5349,get_input_params,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,472,function,"Calculate the number of batches and steps/steps_per_epoch.
Args:
distribution_strategy: The DistributionStrategy used to compile the model.
num_samples: The number of samples from which we determine the batch size
and steps.
steps: The specified number of steps.
batch_size: The specified batch_size.
mode: ModeKey representing whether input will be used for training,
evaluation, or prediction. This is used to relax the constraints on
consuming all the training samples to keep compatibility till we support
partial batches. If none, then partial batches are not allowed.
Returns:
steps: The steps or steps_per_epoch argument depending on if a user is
calling `fit`, `evaluate` or `predict`. If the is_training flag is set
we don't require the number of samples to be used completely.
batch_size: The batch size to be used in model iterations.
Raises:
ValueError: If the number of batches or steps evaluates to 0."
5350,get_batch_dimension,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,576,function,
5351,get_iterator,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,584,function,
5352,initialize_iterator,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,591,function,
5353,is_distributing_by_cloning,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,672,function,"Decide whether this model is going to be distributed via cloning.
We are going to distribute the model by cloning in graph mode.
Args:
model: Keras model to distribute.
Returns:
True if the `model` is going to be distributed using cloning and False
otherwise."
5354,clone_model_on_replicas,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,828,function,Create a cloned model on each replica.
5355,get_distributed_model,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,1074,function,
5356,set_distributed_model,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,1079,function,
5357,get_distributed_function,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,1084,function,
5358,set_distributed_function,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,1089,function,
5359,distributed_scope,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,1100,function,
5360,call_replica_local_fn,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,1105,function,"Call a function that uses replica-local variables.
This function correctly handles calling `fn` in a cross-replica
context.
Arguments:
fn: The function to call.
*args: Positional arguments to the `fn`.
**kwargs: Keyword argument to `fn`.
Returns:
The result of calling `fn`."
5361,is_current_worker_chief,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,1136,function,
5362,filter_distributed_callbacks,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,1140,function,"Filter Callbacks based on the worker context when running multi-worker.
Arguments:
callbacks_list: A list of `Callback` instances.
model: Keras model instance.
Returns:
The list of `Callback` instances that should be run on this worker."
5363,concat_along_batch_dimension,tensorflow/tensorflow/python/keras/distribute/distributed_training_utils.py,1195,function,Concats prediction outputs along the batch dimension.
5364,all_strategy_and_input_config_combinations,tensorflow/tensorflow/python/keras/distribute/keras_correctness_test_base.py,67,function,
5365,strategy_minus_tpu_and_input_config_combinations_eager,tensorflow/tensorflow/python/keras/distribute/keras_correctness_test_base.py,74,function,
5366,strategies_for_embedding_models,tensorflow/tensorflow/python/keras/distribute/keras_correctness_test_base.py,81,function,"Returns distribution strategies to test for embedding models.
Since embedding models take longer to train, we disregard DefaultStrategy
in order to prevent testing timeouts."
5367,MaybeDistributionScope,tensorflow/tensorflow/python/keras/distribute/keras_correctness_test_base.py,120,class,Provides a context allowing no distribution strategy.
5368,batch_wrapper,tensorflow/tensorflow/python/keras/distribute/keras_correctness_test_base.py,138,function,
5369,get_batch_size,tensorflow/tensorflow/python/keras/distribute/keras_correctness_test_base.py,144,function,
5370,get_data_size,tensorflow/tensorflow/python/keras/distribute/keras_correctness_test_base.py,155,function,"Gets the size of data in list, tuple, dict, or a numpy array."
5371,get_shapes,tensorflow/tensorflow/python/keras/distribute/keras_correctness_test_base.py,168,function,
5372,fit_eval_and_predict,tensorflow/tensorflow/python/keras/distribute/keras_correctness_test_base.py,246,function,Generates results for fit/predict/evaluate for given model.
5373,compare_results,tensorflow/tensorflow/python/keras/distribute/keras_correctness_test_base.py,289,function,Compares results of model compiled with/without distribution strategy.
5374,should_skip_tpu_with_eager,tensorflow/tensorflow/python/keras/distribute/keras_correctness_test_base.py,356,function,
5375,LearningRateBatchScheduler,tensorflow/tensorflow/python/keras/distribute/keras_correctness_test_base.py,362,class,Scheduler that dynamically sets the learning rate of model.
5376,on_batch_begin,tensorflow/tensorflow/python/keras/distribute/keras_correctness_test_base.py,368,method,
5377,all_strategy_combinations_with_eager_and_graph_modes,tensorflow/tensorflow/python/keras/distribute/keras_dnn_correctness_test.py,33,function,
5378,all_strategy_combinations_with_graph_mode,tensorflow/tensorflow/python/keras/distribute/keras_dnn_correctness_test.py,39,function,
5379,is_default_strategy,tensorflow/tensorflow/python/keras/distribute/keras_dnn_correctness_test.py,45,function,
5380,SubclassedModel,tensorflow/tensorflow/python/keras/distribute/keras_dnn_correctness_test.py,219,class,
5381,call,tensorflow/tensorflow/python/keras/distribute/keras_dnn_correctness_test.py,236,method,
5382,all_combinations,tensorflow/tensorflow/python/keras/distribute/keras_metrics_test.py,73,function,
5383,tpu_combinations,tensorflow/tensorflow/python/keras/distribute/keras_metrics_test.py,84,function,
5384,get_model,tensorflow/tensorflow/python/keras/distribute/keras_optimizer_v2_test.py,39,function,
5385,strategy_combinations_eager_data_fn,tensorflow/tensorflow/python/keras/distribute/keras_premade_models_test.py,34,function,
5386,get_numpy,tensorflow/tensorflow/python/keras/distribute/keras_premade_models_test.py,47,function,
5387,get_dataset,tensorflow/tensorflow/python/keras/distribute/keras_premade_models_test.py,53,function,
5388,strategies_for_stateful_embedding_model,tensorflow/tensorflow/python/keras/distribute/keras_stateful_lstm_model_correctness_test.py,29,function,Returns TPUStrategy with single core device assignment.
5389,Counter,tensorflow/tensorflow/python/keras/distribute/keras_utils_test.py,43,class,"Counts the number of times each callback method was run.
Attributes:
method_counts: dict. Contains the counts of time each callback method was
run."
5390,wrap_with_counts,tensorflow/tensorflow/python/keras/distribute/keras_utils_test.py,64,method,
5391,MiniModel,tensorflow/tensorflow/python/keras/distribute/mirrored_strategy_test.py,37,class,"Minimal model for mnist.
Useful for testing and debugging on slow TPU simulators."
5392,call,tensorflow/tensorflow/python/keras/distribute/mirrored_strategy_test.py,48,method,
5393,ModelAndInput,tensorflow/tensorflow/python/keras/distribute/model_collection_base.py,21,class,Base class to provide model and its corresponding inputs.
5394,get_model,tensorflow/tensorflow/python/keras/distribute/model_collection_base.py,24,method,"Returns a compiled keras model object, together with output name.
Returns:
model: a keras model object
output_name: a string for the name of the output layer"
5395,get_data,tensorflow/tensorflow/python/keras/distribute/model_collection_base.py,33,method,"Returns data for training and predicting.
Returns:
x_train: data used for training
y_train: label used for training
x_predict: data used for predicting"
5396,get_batch_size,tensorflow/tensorflow/python/keras/distribute/model_collection_base.py,43,method,Returns the batch_size used by the model.
5397,checkpoint_exists,tensorflow/tensorflow/python/keras/distribute/multi_worker_callback_tf2_test.py,37,function,Returns whether the checkpoint `filepath` refers to exists.
5398,ParameterServerStrategy,tensorflow/tensorflow/python/keras/distribute/multi_worker_test.py,52,class,Temporarily mock the original strategy to bypass cluster_spec check.
5399,MultiWorkerVerificationCallback,tensorflow/tensorflow/python/keras/distribute/multi_worker_test.py,96,class,"MultiWorkerVerificationCallback verifies the callbacks in multi-worker scheme.
This Callback is intended to be used for verifying the callback is indeed
called the correct number of times in various task types.
Attributes:
_task_dict: A nested dictionary storing the number of times a callback has
been called in specific task type, task index, and method name.
Look up structure is
task_name -> task_id -> tracking_method_name -> invoke_count
For example, a _task_dict of
{
'ps': {
0: {
'on_epoch_begin': 2
},
1: {
'on_epoch_begin': 2
}
},
'worker': {
0: {
'on_epoch_begin': 2
},
1: {
'on_epoch_begin': 2
}
}
}
indicates the ps task has 'on_epoch_begin' called twice on each
of the two indices, and likewise for worker task."
5400,is_between_graph,tensorflow/tensorflow/python/keras/distribute/multi_worker_test.py,152,method,
5401,is_between_graph,tensorflow/tensorflow/python/keras/distribute/multi_worker_test.py,156,method,
5402,wrap_methods,tensorflow/tensorflow/python/keras/distribute/multi_worker_test.py,159,method,"Wrap methods so that the counts of calls are tracked.
Args:
method_names: A list of names of methods to track calls."
5403,verify,tensorflow/tensorflow/python/keras/distribute/multi_worker_test.py,178,method,
5404,wrapped_method,tensorflow/tensorflow/python/keras/distribute/multi_worker_test.py,168,method,
5405,mnist_synthetic_dataset,tensorflow/tensorflow/python/keras/distribute/multi_worker_testing_utils.py,28,function,Generate synthetic MNIST dataset for testing.
5406,get_mnist_model,tensorflow/tensorflow/python/keras/distribute/multi_worker_testing_utils.py,52,function,Define a deterministically-initialized CNN model for MNIST testing.
5407,distributions_and_v1_optimizers,tensorflow/tensorflow/python/keras/distribute/optimizer_combinations.py,81,function,A common set of combination with DistributionStrategies and Optimizers.
5408,distributions_and_v2_optimizers,tensorflow/tensorflow/python/keras/distribute/optimizer_combinations.py,92,function,A common set of combination with DistributionStrategies and Optimizers.
5409,distributions_and_v1_and_v2_optimizers,tensorflow/tensorflow/python/keras/distribute/optimizer_combinations.py,103,function,A common set of combination with DistributionStrategies and Optimizers.
5410,is_tpu_strategy,tensorflow/tensorflow/python/keras/distribute/saved_model_test_base.py,66,function,
5411,get_tolerance,tensorflow/tensorflow/python/keras/distribute/saved_model_test_base.py,71,function,
5412,simple_models_with_strategies,tensorflow/tensorflow/python/keras/distribute/saved_model_test_base.py,78,function,
5413,simple_models_with_strategy_pairs,tensorflow/tensorflow/python/keras/distribute/saved_model_test_base.py,85,function,
5414,tfmodule_models_with_strategies,tensorflow/tensorflow/python/keras/distribute/saved_model_test_base.py,93,function,
5415,tfmodule_models_with_strategy_pairs,tensorflow/tensorflow/python/keras/distribute/saved_model_test_base.py,100,function,
5416,load_and_run_with_saved_model_api,tensorflow/tensorflow/python/keras/distribute/saved_model_test_base.py,108,function,"Loads a saved_model using tf.saved_model API, and runs it."
5417,SimpleFunctionalModel,tensorflow/tensorflow/python/keras/distribute/simple_models.py,44,class,A simple functional model and its inputs.
5418,get_model,tensorflow/tensorflow/python/keras/distribute/simple_models.py,47,method,
5419,get_data,tensorflow/tensorflow/python/keras/distribute/simple_models.py,62,method,
5420,get_batch_size,tensorflow/tensorflow/python/keras/distribute/simple_models.py,65,method,
5421,SimpleSequentialModel,tensorflow/tensorflow/python/keras/distribute/simple_models.py,69,class,A simple sequential model and its inputs.
5422,get_model,tensorflow/tensorflow/python/keras/distribute/simple_models.py,72,method,
5423,get_data,tensorflow/tensorflow/python/keras/distribute/simple_models.py,87,method,
5424,get_batch_size,tensorflow/tensorflow/python/keras/distribute/simple_models.py,90,method,
5425,SimpleSubclassModel,tensorflow/tensorflow/python/keras/distribute/simple_models.py,104,class,A simple subclass model and its data.
5426,get_model,tensorflow/tensorflow/python/keras/distribute/simple_models.py,107,method,
5427,get_data,tensorflow/tensorflow/python/keras/distribute/simple_models.py,118,method,
5428,get_batch_size,tensorflow/tensorflow/python/keras/distribute/simple_models.py,121,method,
5429,SimpleTFModuleModel,tensorflow/tensorflow/python/keras/distribute/simple_models.py,135,class,A simple model based on tf.Module and its data.
5430,get_model,tensorflow/tensorflow/python/keras/distribute/simple_models.py,138,method,
5431,get_data,tensorflow/tensorflow/python/keras/distribute/simple_models.py,142,method,
5432,get_batch_size,tensorflow/tensorflow/python/keras/distribute/simple_models.py,145,method,
5433,get_tpu_cluster_resolver,tensorflow/tensorflow/python/keras/distribute/tpu_strategy_test_utils.py,34,function,
5434,get_tpu_strategy,tensorflow/tensorflow/python/keras/distribute/tpu_strategy_test_utils.py,43,function,
5435,WorkerTrainingState,tensorflow/tensorflow/python/keras/distribute/worker_training_state.py,40,class,"Training state management class.
This class provides apis for backing up and restoring the training state.
This allows model and epoch information to be saved periodically and restore
for fault-tolerance, also known as preemption-recovery purpose."
5436,back_up,tensorflow/tensorflow/python/keras/distribute/worker_training_state.py,87,method,"Back up the current state of training into a checkpoint file.
Arguments:
epoch: The current epoch information to be saved."
5437,restore,tensorflow/tensorflow/python/keras/distribute/worker_training_state.py,100,method,"Restore the training state from the backed up checkpoint file.
Returns:
True if the training state is successfully restored. False if the training
state doesn't need to be restored, or error occurred so it can't."
5438,delete_backup,tensorflow/tensorflow/python/keras/distribute/worker_training_state.py,115,method,"Delete the backup directories.
Delete the backup directories which should not exist after `fit()`
successfully finishes."
5439,maybe_load_initial_epoch_from_ckpt,tensorflow/tensorflow/python/keras/distribute/worker_training_state.py,129,method,"Maybe load initial epoch from ckpt considering possible worker recovery.
When `_ckpt_saved_epoch` attribute exists and is not
`CKPT_SAVED_EPOCH_UNUSED_VALUE`, this is under multi-worker training setting
and indicates the worker is recovering from previous failure. In this case,
infer `initial_epoch` from `self._ckpt_saved_epoch` to continue previous
unfinished training from certain epoch.
Arguments:
initial_epoch: The original initial_epoch user passes in in `fit()`.
mode: The mode for running `model.fit()`.
Returns:
If the training is recovering from previous failure under multi-worker
training setting, return the epoch the training is supposed to continue
at. Otherwise, return the `initial_epoch` the user passes in."
5440,Layer,tensorflow/tensorflow/python/keras/engine/base_layer.py,103,class,"This is the class from which all layers inherit.
A layer is a callable object that takes as input one or more tensors and
that outputs one or more tensors. It involves *computation*, defined
in the `call()` method, and a *state* (weight variables), defined
either in the constructor `__init__()` or in the `build()` method.
Users will just instantiate a layer and then treat it as a callable.
Arguments:
trainable: Boolean, whether the layer's variables should be trainable.
name: String name of the layer.
dtype: The dtype of the layer's computations and weights (default of
`None` means use `tf.keras.backend.floatx` in TensorFlow 2, or the type
of the first input in TensorFlow 1).
dynamic: Set this to `True` if your layer should only be run eagerly, and
should not be used to generate a static computation graph.
This would be the case for a Tree-RNN or a recursive network,
for example, or generally for any layer that manipulates tensors
using Python control flow. If `False`, we assume that the layer can
safely be used to generate a static computation graph.
Attributes:
name: The name of the layer (string).
dtype: The dtype of the layer's computations and weights. If mixed
precision is used with a `tf.keras.mixed_precision.experimental.Policy`,
this is instead just the dtype of the layer's weights, as the computations
are done in a different dtype.
trainable_weights: List of variables to be included in backprop.
non_trainable_weights: List of variables that should not be
included in backprop.
weights: The concatenation of the lists trainable_weights and
non_trainable_weights (in this order).
trainable: Whether the layer should be trained (boolean), i.e. whether
its potentially-trainable weights should be returned as part of
`layer.trainable_weights`.
input_spec: Optional (list of) `InputSpec` object(s) specifying the
constraints on inputs that can be accepted by the layer.
We recommend that descendants of `Layer` implement the following methods:
* `__init__()`: Defines custom layer attributes, and creates layer state
variables that do not depend on input shapes, using `add_weight()`.
* `build(self, input_shape)`: This method can be used to create weights that
depend on the shape(s) of the input(s), using `add_weight()`. `__call__()`
will automatically build the layer (if it has not been built yet) by
calling `build()`.
* `call(self, *args, **kwargs)`: Called in `__call__` after making sure
`build()` has been called. `call()` performs the logic of applying the
layer to the input tensors (which should be passed in as argument).
Two reserved keyword arguments you can optionally use in `call()` are:
- `training` (boolean, whether the call is in
inference mode or training mode)
- `mask` (boolean tensor encoding masked timesteps in the input, used
in RNN layers)
* `get_config(self)`: Returns a dictionary containing the configuration used
to initialize this layer. If the keys differ from the arguments
in `__init__`, then override `from_config(self)` as well.
This method is used when saving
the layer or a model that contains this layer.
Examples:
Here's a basic example: a layer with two variables, `w` and `b`,
that returns `y = w . x + b`.
It shows how to implement `build()` and `call()`.
Variables set as attributes of a layer are tracked as weights
of the layers (in `layer.weights`).
```python
class SimpleDense(Layer):
def __init__(self, units=32):
super(SimpleDense, self).__init__()
self.units = units
def build(self, input_shape): # Create the state of the layer (weights)
w_init = tf.random_normal_initializer()
self.w = tf.Variable(
initial_value=w_init(shape=(input_shape[-1], self.units),
dtype='float32'),
trainable=True)
b_init = tf.zeros_initializer()
self.b = tf.Variable(
initial_value=b_init(shape=(self.units,), dtype='float32'),
trainable=True)
def call(self, inputs): # Defines the computation from inputs to outputs
return tf.matmul(inputs, self.w) + self.b
# Instantiates the layer.
linear_layer = SimpleDense(4)
# This will also call `build(input_shape)` and create the weights.
y = linear_layer(tf.ones((2, 2)))
assert len(linear_layer.weights) == 2
# These weights are trainable, so they're listed in `trainable_weights`:
assert len(linear_layer.trainable_weights) == 2
```
Note that the method `add_weight()` offers a shortcut to create weights:
```python
class SimpleDense(Layer):
def __init__(self, units=32):
super(SimpleDense, self).__init__()
self.units = units
def build(self, input_shape):
self.w = self.add_weight(shape=(input_shape[-1], self.units),
initializer='random_normal',
trainable=True)
self.b = self.add_weight(shape=(self.units,),
initializer='random_normal',
trainable=True)
def call(self, inputs):
return tf.matmul(inputs, self.w) + self.b
```
Besides trainable weights, updated via backpropagation during training,
layers can also have non-trainable weights. These weights are meant to
be updated manually during `call()`. Here's a example layer that computes
the running sum of its inputs:
```python
class ComputeSum(Layer):
def __init__(self, input_dim):
super(ComputeSum, self).__init__()
# Create a non-trainable weight.
self.total = tf.Variable(initial_value=tf.zeros((input_dim,)),
trainable=False)
def call(self, inputs):
self.total.assign_add(tf.reduce_sum(inputs, axis=0))
return self.total
my_sum = ComputeSum(2)
x = tf.ones((2, 2))
y = my_sum(x)
print(y.numpy()) # [2. 2.]
y = my_sum(x)
print(y.numpy()) # [4. 4.]
assert my_sum.weights == [my_sum.total]
assert my_sum.non_trainable_weights == [my_sum.total]
assert my_sum.trainable_weights == []
```
For more information about creating layers, see the guide
[Writing custom layers and models with Keras](
https://www.tensorflow.org/guide/keras/custom_layers_and_models)
About the layer's `dtype` attribute:
Each layer has a dtype, which is typically the dtype of the layer's
computations and variables. A layer's dtype can be queried via the
`Layer.dtype` property. The dtype is specified with the `dtype` constructor
argument. In TensorFlow 2, the dtype defaults to `tf.keras.backend.floatx()`
if no dtype is passed. `floatx()` itself defaults to ""float32"". Additionally,
layers will cast their inputs to the layer's dtype in TensorFlow 2. When mixed
precision is used, layers may have different computation and variable dtypes.
See `tf.keras.mixed_precision.experimental.Policy` for details on layer
dtypes."
5441,build,tensorflow/tensorflow/python/keras/engine/base_layer.py,425,method,"Creates the variables of the layer (optional, for subclass implementers).
This is a method that implementers of subclasses of `Layer` or `Model`
can override if they need a state-creation step in-between
layer instantiation and layer call.
This is typically used to create the weights of `Layer` subclasses.
Arguments:
input_shape: Instance of `TensorShape`, or list of instances of
`TensorShape` if the layer expects a list of inputs
(one instance per input)."
5442,call,tensorflow/tensorflow/python/keras/engine/base_layer.py,445,method,"This is where the layer's logic lives.
Note here that `call()` method in `tf.keras` is little bit different
from `keras` API. In `keras` API, you can pass support masking for
layers as additional arguments. Whereas `tf.keras` has `compute_mask()`
method to support masking.
Arguments:
inputs: Input tensor, or list/tuple of input tensors.
**kwargs: Additional keyword arguments. Currently unused.
Returns:
A tensor or list/tuple of tensors."
5443,add_weight,tensorflow/tensorflow/python/keras/engine/base_layer.py,483,method,"Adds a new variable to the layer.
Arguments:
name: Variable name.
shape: Variable shape. Defaults to scalar if unspecified.
dtype: The type of the variable. Defaults to `self.dtype` or `float32`.
initializer: Initializer instance (callable).
regularizer: Regularizer instance (callable).
trainable: Boolean, whether the variable should be part of the layer's
""trainable_variables"" (e.g. variables, biases)
or ""non_trainable_variables"" (e.g. BatchNorm mean and variance).
Note that `trainable` cannot be `True` if `synchronization`
is set to `ON_READ`.
constraint: Constraint instance (callable).
use_resource: Whether to use `ResourceVariable`.
synchronization: Indicates when a distributed a variable will be
aggregated. Accepted values are constants defined in the class
`tf.VariableSynchronization`. By default the synchronization is set to
`AUTO` and the current `DistributionStrategy` chooses
when to synchronize. If `synchronization` is set to `ON_READ`,
`trainable` must not be set to `True`.
aggregation: Indicates how a distributed variable will be aggregated.
Accepted values are constants defined in the class
`tf.VariableAggregation`.
**kwargs: Additional keyword arguments. Accepted values are `getter`,
`collections`, `experimental_autocast` and `caching_device`.
Returns:
The variable created.
Raises:
ValueError: When giving unsupported dtype and no initializer or when
trainable has been set to True with synchronization set as `ON_READ`."
5444,get_config,tensorflow/tensorflow/python/keras/engine/base_layer.py,638,method,"Returns the config of the layer.
A layer config is a Python dictionary (serializable)
containing the configuration of a layer.
The same layer can be reinstantiated later
(without its trained weights) from this configuration.
The config of a layer does not include connectivity
information, nor the layer class name. These are handled
by `Network` (one layer of abstraction above).
Returns:
Python dictionary."
5445,from_config,tensorflow/tensorflow/python/keras/engine/base_layer.py,679,method,"Creates a layer from its config.
This method is the reverse of `get_config`,
capable of instantiating the same layer from the config
dictionary. It does not handle layer connectivity
(handled by Network), nor weights (handled by `set_weights`).
Arguments:
config: A Python dictionary, typically the
output of get_config.
Returns:
A layer instance."
5446,compute_output_shape,tensorflow/tensorflow/python/keras/engine/base_layer.py,696,method,"Computes the output shape of the layer.
If the layer has not been built, this method will call `build` on the
layer. This assumes that the layer will later be used with inputs that
match the input shape provided here.
Arguments:
input_shape: Shape tuple (tuple of integers)
or list of shape tuples (one per output tensor of the layer).
Shape tuples can include None for free dimensions,
instead of an integer.
Returns:
An input shape tuple."
5447,compute_output_signature,tensorflow/tensorflow/python/keras/engine/base_layer.py,743,method,"Compute the output tensor signature of the layer based on the inputs.
Unlike a TensorShape object, a TensorSpec object contains both shape
and dtype information for a tensor. This method allows layers to provide
output dtype information if it is different from the input dtype.
For any layer that doesn't implement this function,
the framework will fall back to use `compute_output_shape`, and will
assume that the output dtype matches the input dtype.
Args:
input_signature: Single TensorSpec or nested structure of TensorSpec
objects, describing a candidate input for the layer.
Returns:
Single TensorSpec or nested structure of TensorSpec objects, describing
how the layer would transform the provided input.
Raises:
TypeError: If input_signature contains a non-TensorSpec object."
5448,compute_mask,tensorflow/tensorflow/python/keras/engine/base_layer.py,854,method,"Computes an output mask tensor.
Arguments:
inputs: Tensor or list of tensors.
mask: Tensor or list of tensors.
Returns:
None or a tensor (or list of tensors,
one per output tensor of the layer)."
5449,dtype,tensorflow/tensorflow/python/keras/engine/base_layer.py,1214,method,"Dtype used by the weights of the layer, set in the constructor."
5450,name,tensorflow/tensorflow/python/keras/engine/base_layer.py,1219,method,"Name of the layer (string), set in the constructor."
5451,supports_masking,tensorflow/tensorflow/python/keras/engine/base_layer.py,1224,method,Whether this layer supports computing a mask using `compute_mask`.
5452,supports_masking,tensorflow/tensorflow/python/keras/engine/base_layer.py,1229,method,
5453,dynamic,tensorflow/tensorflow/python/keras/engine/base_layer.py,1233,method,Whether the layer is dynamic (eager-only); set in the constructor.
5454,stateful,tensorflow/tensorflow/python/keras/engine/base_layer.py,1239,method,
5455,stateful,tensorflow/tensorflow/python/keras/engine/base_layer.py,1243,method,
5456,trainable,tensorflow/tensorflow/python/keras/engine/base_layer.py,1247,method,
5457,trainable,tensorflow/tensorflow/python/keras/engine/base_layer.py,1251,method,
5458,activity_regularizer,tensorflow/tensorflow/python/keras/engine/base_layer.py,1256,method,Optional regularizer function for the output of this layer.
5459,activity_regularizer,tensorflow/tensorflow/python/keras/engine/base_layer.py,1261,method,Optional regularizer function for the output of this layer.
5460,input_spec,tensorflow/tensorflow/python/keras/engine/base_layer.py,1266,method,"`InputSpec` instance(s) describing the input format for this layer.
When you create a layer subclass, you can set `self.input_spec` to enable
the layer to run input compatibility checks when it is called.
Consider a `Conv2D` layer: it can only be called on a single input tensor
of rank 4. As such, you can set, in `__init__()`:
```python
self.input_spec = tf.keras.layers.InputSpec(ndim=4)
```
Now, if you try to call the layer on an input that isn't rank 4
(for instance, an input of shape `(2,)`, it will raise a nicely-formatted
error:
```
ValueError: Input 0 of layer conv2d is incompatible with the layer:
expected ndim=4, found ndim=1. Full shape received: [2]
```
Input checks that can be specified via `input_spec` include:
- Structure (e.g. a single input, a list of 2 inputs, etc)
- Shape
- Rank (ndim)
- Dtype
For more information, see `tf.keras.layers.InputSpec`.
Returns:
A `tf.keras.layers.InputSpec` instance, or nested structure thereof."
5461,input_spec,tensorflow/tensorflow/python/keras/engine/base_layer.py,1304,method,
5462,trainable_weights,tensorflow/tensorflow/python/keras/engine/base_layer.py,1312,method,"List of all trainable weights tracked by this layer.
Trainable weights are updated via gradient descent during training.
Returns:
A list of trainable variables."
5463,non_trainable_weights,tensorflow/tensorflow/python/keras/engine/base_layer.py,1327,method,"List of all non-trainable weights tracked by this layer.
Non-trainable weights are *not* updated during training. They are expected
to be updated manually in `call()`.
Returns:
A list of non-trainable variables."
5464,weights,tensorflow/tensorflow/python/keras/engine/base_layer.py,1348,method,"Returns the list of all layer variables/weights.
Returns:
A list of variables."
5465,updates,tensorflow/tensorflow/python/keras/engine/base_layer.py,1362,method,
5466,losses,tensorflow/tensorflow/python/keras/engine/base_layer.py,1379,method,"List of losses added using the `add_loss()` API.
Variable regularization tensors are created when this property is accessed,
so it is eager safe: accessing `losses` under a `tf.GradientTape` will
propagate gradients back to the corresponding variables.
Examples:
>>> class MyLayer(tf.keras.layers.Layer):
... def call(self, inputs):
... self.add_loss(tf.abs(tf.reduce_mean(inputs)))
... return inputs
>>> l = MyLayer()
>>> l(np.ones((10, 1)))
>>> l.losses
[1.0]
>>> inputs = tf.keras.Input(shape=(10,))
>>> x = tf.keras.layers.Dense(10)(inputs)
>>> outputs = tf.keras.layers.Dense(1)(x)
>>> model = tf.keras.Model(inputs, outputs)
>>> # Activity regularization.
>>> len(model.losses)
0
>>> model.add_loss(tf.abs(tf.reduce_mean(x)))
>>> len(model.losses)
1
>>> inputs = tf.keras.Input(shape=(10,))
>>> d = tf.keras.layers.Dense(10, kernel_initializer='ones')
>>> x = d(inputs)
>>> outputs = tf.keras.layers.Dense(1)(x)
>>> model = tf.keras.Model(inputs, outputs)
>>> # Weight regularization.
>>> model.add_loss(lambda: tf.reduce_mean(d.kernel))
>>> model.losses
[<tf.Tensor: shape=(), dtype=float32, numpy=1.0>]
Returns:
A list of tensors."
5467,add_loss,tensorflow/tensorflow/python/keras/engine/base_layer.py,1440,method,"Add loss tensor(s), potentially dependent on layer inputs.
Some losses (for instance, activity regularization losses) may be dependent
on the inputs passed when calling a layer. Hence, when reusing the same
layer on different inputs `a` and `b`, some entries in `layer.losses` may
be dependent on `a` and some on `b`. This method automatically keeps track
of dependencies.
This method can be used inside a subclassed layer or model's `call`
function, in which case `losses` should be a Tensor or list of Tensors.
Example:
```python
class MyLayer(tf.keras.layers.Layer):
def call(self, inputs):
self.add_loss(tf.abs(tf.reduce_mean(inputs)))
return inputs
```
This method can also be called directly on a Functional Model during
construction. In this case, any loss Tensors passed to this Model must
be symbolic and be able to be traced back to the model's `Input`s. These
losses become part of the model's topology and are tracked in `get_config`.
Example:
```python
inputs = tf.keras.Input(shape=(10,))
x = tf.keras.layers.Dense(10)(inputs)
outputs = tf.keras.layers.Dense(1)(x)
model = tf.keras.Model(inputs, outputs)
# Activity regularization.
model.add_loss(tf.abs(tf.reduce_mean(x)))
```
If this is not the case for your loss (if, for example, your loss references
a `Variable` of one of the model's layers), you can wrap your loss in a
zero-argument lambda. These losses are not tracked as part of the model's
topology since they can't be serialized.
Example:
```python
inputs = tf.keras.Input(shape=(10,))
d = tf.keras.layers.Dense(10)
x = d(inputs)
outputs = tf.keras.layers.Dense(1)(x)
model = tf.keras.Model(inputs, outputs)
# Weight regularization.
model.add_loss(lambda: tf.reduce_mean(d.kernel))
```
Arguments:
losses: Loss tensor, or list/tuple of tensors. Rather than tensors, losses
may also be zero-argument callables which create a loss tensor.
**kwargs: Additional keyword arguments for backward compatibility.
Accepted values:
inputs - Deprecated, will be automatically inferred."
5468,metrics,tensorflow/tensorflow/python/keras/engine/base_layer.py,1572,method,"List of metrics added using the `add_metric()` API.
Example:
>>> input = tf.keras.layers.Input(shape=(3,))
>>> d = tf.keras.layers.Dense(2)
>>> output = d(input)
>>> d.add_metric(tf.reduce_max(output), name='max')
>>> d.add_metric(tf.reduce_min(output), name='min')
>>> [m.name for m in d.metrics]
['max', 'min']
Returns:
A list of tensors."
5469,add_metric,tensorflow/tensorflow/python/keras/engine/base_layer.py,1594,method,"Adds metric tensor to the layer.
This method can be used inside the `call()` method of a subclassed layer
or model.
```python
class MyMetricLayer(tf.keras.layers.Layer):
def __init__(self):
super(MyMetricLayer, self).__init__(name='my_metric_layer')
self.mean = metrics_module.Mean(name='metric_1')
def call(self, inputs):
self.add_metric(self.mean(x))
self.add_metric(math_ops.reduce_sum(x), name='metric_2')
return inputs
```
This method can also be called directly on a Functional Model during
construction. In this case, any tensor passed to this Model must
be symbolic and be able to be traced back to the model's `Input`s. These
metrics become part of the model's topology and are tracked when you
save the model via `save()`.
```python
inputs = tf.keras.Input(shape=(10,))
x = tf.keras.layers.Dense(10)(inputs)
outputs = tf.keras.layers.Dense(1)(x)
model = tf.keras.Model(inputs, outputs)
model.add_metric(math_ops.reduce_sum(x), name='metric_1')
```
Note: Calling `add_metric()` with the result of a metric object on a
Functional Model, as shown in the example below, is not supported. This is
because we cannot trace the metric result tensor back to the model's inputs.
```python
inputs = tf.keras.Input(shape=(10,))
x = tf.keras.layers.Dense(10)(inputs)
outputs = tf.keras.layers.Dense(1)(x)
model = tf.keras.Model(inputs, outputs)
model.add_metric(tf.keras.metrics.Mean()(x), name='metric_1')
```
Args:
value: Metric tensor.
name: String metric name.
**kwargs: Additional keyword arguments for backward compatibility.
Accepted values:
`aggregation` - When the `value` tensor provided is not the result of
calling a `keras.Metric` instance, it will be aggregated by default
using a `keras.Metric.Mean`."
5470,add_update,tensorflow/tensorflow/python/keras/engine/base_layer.py,1720,method,"Add update op(s), potentially dependent on layer inputs.
Weight updates (for instance, the updates of the moving mean and variance
in a BatchNormalization layer) may be dependent on the inputs passed
when calling a layer. Hence, when reusing the same layer on
different inputs `a` and `b`, some entries in `layer.updates` may be
dependent on `a` and some on `b`. This method automatically keeps track
of dependencies.
This call is ignored when eager execution is enabled (in that case, variable
updates are run on the fly and thus do not need to be tracked for later
execution).
Arguments:
updates: Update op, or list/tuple of update ops, or zero-arg callable
that returns an update op. A zero-arg callable should be passed in
order to disable running the updates by setting `trainable=False`
on this Layer, when executing in Eager mode.
inputs: Deprecated, will be automatically inferred."
5471,set_weights,tensorflow/tensorflow/python/keras/engine/base_layer.py,1756,method,"Sets the weights of the layer, from Numpy arrays.
The weights of a layer represent the state of the layer. This function
sets the weight values from numpy arrays. The weight values should be
passed in the order they are created by the layer. Note that the layer's
weights must be instantiated before calling this function by calling
the layer.
For example, a Dense layer returns a list of two values-- per-output
weights and the bias value. These can be used to set the weights of another
Dense layer:
>>> a = tf.keras.layers.Dense(1,
... kernel_initializer=tf.constant_initializer(1.))
>>> a_out = a(tf.convert_to_tensor([[1., 2., 3.]]))
>>> a.get_weights()
[array([[1.],
[1.],
[1.]], dtype=float32), array([0.], dtype=float32)]
>>> b = tf.keras.layers.Dense(1,
... kernel_initializer=tf.constant_initializer(2.))
>>> b_out = b(tf.convert_to_tensor([[10., 20., 30.]]))
>>> b.get_weights()
[array([[2.],
[2.],
[2.]], dtype=float32), array([0.], dtype=float32)]
>>> b.set_weights(a.get_weights())
>>> b.get_weights()
[array([[1.],
[1.],
[1.]], dtype=float32), array([0.], dtype=float32)]
Arguments:
weights: a list of Numpy arrays. The number
of arrays and their shape must match
number of the dimensions of the weights
of the layer (i.e. it should match the
output of `get_weights`).
Raises:
ValueError: If the provided weights list does not match the
layer's specifications."
5472,get_weights,tensorflow/tensorflow/python/keras/engine/base_layer.py,1836,method,"Returns the current weights of the layer.
The weights of a layer represent the state of the layer. This function
returns both trainable and non-trainable weight values associated with this
layer as a list of Numpy arrays, which can in turn be used to load state
into similarly parameterized layers.
For example, a Dense layer returns a list of two values-- per-output
weights and the bias value. These can be used to set the weights of another
Dense layer:
>>> a = tf.keras.layers.Dense(1,
... kernel_initializer=tf.constant_initializer(1.))
>>> a_out = a(tf.convert_to_tensor([[1., 2., 3.]]))
>>> a.get_weights()
[array([[1.],
[1.],
[1.]], dtype=float32), array([0.], dtype=float32)]
>>> b = tf.keras.layers.Dense(1,
... kernel_initializer=tf.constant_initializer(2.))
>>> b_out = b(tf.convert_to_tensor([[10., 20., 30.]]))
>>> b.get_weights()
[array([[2.],
[2.],
[2.]], dtype=float32), array([0.], dtype=float32)]
>>> b.set_weights(a.get_weights())
>>> b.get_weights()
[array([[1.],
[1.],
[1.]], dtype=float32), array([0.], dtype=float32)]
Returns:
Weights values as a list of numpy arrays."
5473,get_updates_for,tensorflow/tensorflow/python/keras/engine/base_layer.py,1883,method,"Deprecated, do NOT use!
Retrieves updates relevant to a specific set of inputs.
Arguments:
inputs: Input tensor or list/tuple of input tensors.
Returns:
List of update ops of the layer that depend on `inputs`."
5474,get_losses_for,tensorflow/tensorflow/python/keras/engine/base_layer.py,1899,method,"Deprecated, do NOT use!
Retrieves losses relevant to a specific set of inputs.
Arguments:
inputs: Input tensor or list/tuple of input tensors.
Returns:
List of loss tensors of the layer that depend on `inputs`."
5475,get_input_mask_at,tensorflow/tensorflow/python/keras/engine/base_layer.py,1913,method,"Retrieves the input mask tensor(s) of a layer at a given node.
Arguments:
node_index: Integer, index of the node
from which to retrieve the attribute.
E.g. `node_index=0` will correspond to the
first time the layer was called.
Returns:
A mask tensor
(or list of tensors if the layer has multiple inputs)."
5476,get_output_mask_at,tensorflow/tensorflow/python/keras/engine/base_layer.py,1933,method,"Retrieves the output mask tensor(s) of a layer at a given node.
Arguments:
node_index: Integer, index of the node
from which to retrieve the attribute.
E.g. `node_index=0` will correspond to the
first time the layer was called.
Returns:
A mask tensor
(or list of tensors if the layer has multiple outputs)."
5477,input_mask,tensorflow/tensorflow/python/keras/engine/base_layer.py,1954,method,"Retrieves the input mask tensor(s) of a layer.
Only applicable if the layer has exactly one inbound node,
i.e. if it is connected to one incoming layer.
Returns:
Input mask tensor (potentially None) or list of input
mask tensors.
Raises:
AttributeError: if the layer is connected to
more than one incoming layers."
5478,output_mask,tensorflow/tensorflow/python/keras/engine/base_layer.py,1976,method,"Retrieves the output mask tensor(s) of a layer.
Only applicable if the layer has exactly one inbound node,
i.e. if it is connected to one incoming layer.
Returns:
Output mask tensor (potentially None) or list of output
mask tensors.
Raises:
AttributeError: if the layer is connected to
more than one incoming layers."
5479,get_input_shape_at,tensorflow/tensorflow/python/keras/engine/base_layer.py,1997,method,"Retrieves the input shape(s) of a layer at a given node.
Arguments:
node_index: Integer, index of the node
from which to retrieve the attribute.
E.g. `node_index=0` will correspond to the
first time the layer was called.
Returns:
A shape tuple
(or list of shape tuples if the layer has multiple inputs).
Raises:
RuntimeError: If called in Eager mode."
5480,get_output_shape_at,tensorflow/tensorflow/python/keras/engine/base_layer.py,2017,method,"Retrieves the output shape(s) of a layer at a given node.
Arguments:
node_index: Integer, index of the node
from which to retrieve the attribute.
E.g. `node_index=0` will correspond to the
first time the layer was called.
Returns:
A shape tuple
(or list of shape tuples if the layer has multiple outputs).
Raises:
RuntimeError: If called in Eager mode."
5481,get_input_at,tensorflow/tensorflow/python/keras/engine/base_layer.py,2037,method,"Retrieves the input tensor(s) of a layer at a given node.
Arguments:
node_index: Integer, index of the node
from which to retrieve the attribute.
E.g. `node_index=0` will correspond to the
first time the layer was called.
Returns:
A tensor (or list of tensors if the layer has multiple inputs).
Raises:
RuntimeError: If called in Eager mode."
5482,get_output_at,tensorflow/tensorflow/python/keras/engine/base_layer.py,2056,method,"Retrieves the output tensor(s) of a layer at a given node.
Arguments:
node_index: Integer, index of the node
from which to retrieve the attribute.
E.g. `node_index=0` will correspond to the
first time the layer was called.
Returns:
A tensor (or list of tensors if the layer has multiple outputs).
Raises:
RuntimeError: If called in Eager mode."
5483,input,tensorflow/tensorflow/python/keras/engine/base_layer.py,2075,method,"Retrieves the input tensor(s) of a layer.
Only applicable if the layer has exactly one input,
i.e. if it is connected to one incoming layer.
Returns:
Input tensor or list of input tensors.
Raises:
RuntimeError: If called in Eager mode.
AttributeError: If no inbound nodes are found."
5484,output,tensorflow/tensorflow/python/keras/engine/base_layer.py,2094,method,"Retrieves the output tensor(s) of a layer.
Only applicable if the layer has exactly one output,
i.e. if it is connected to one incoming layer.
Returns:
Output tensor or list of output tensors.
Raises:
AttributeError: if the layer is connected to more than one incoming
layers.
RuntimeError: if called in Eager mode."
5485,input_shape,tensorflow/tensorflow/python/keras/engine/base_layer.py,2114,method,"Retrieves the input shape(s) of a layer.
Only applicable if the layer has exactly one input,
i.e. if it is connected to one incoming layer, or if all inputs
have the same shape.
Returns:
Input shape, as an integer shape tuple
(or list of shape tuples, one tuple per input tensor).
Raises:
AttributeError: if the layer has no defined input_shape.
RuntimeError: if called in Eager mode."
5486,count_params,tensorflow/tensorflow/python/keras/engine/base_layer.py,2145,method,"Count the total number of scalars composing the weights.
Returns:
An integer count.
Raises:
ValueError: if the layer isn't yet built
(in which case its weights aren't yet defined)."
5487,output_shape,tensorflow/tensorflow/python/keras/engine/base_layer.py,2168,method,"Retrieves the output shape(s) of a layer.
Only applicable if the layer has one output,
or if all outputs have the same shape.
Returns:
Output shape, as an integer shape tuple
(or list of shape tuples, one tuple per output tensor).
Raises:
AttributeError: if the layer has no defined output shape.
RuntimeError: if called in Eager mode."
5488,inbound_nodes,tensorflow/tensorflow/python/keras/engine/base_layer.py,2200,method,"Deprecated, do NOT use! Only for compatibility with external Keras."
5489,outbound_nodes,tensorflow/tensorflow/python/keras/engine/base_layer.py,2206,method,"Deprecated, do NOT use! Only for compatibility with external Keras."
5490,apply,tensorflow/tensorflow/python/keras/engine/base_layer.py,2217,method,"Deprecated, do NOT use!
This is an alias of `self.__call__`.
Arguments:
inputs: Input tensor(s).
*args: additional positional arguments to be passed to `self.call`.
**kwargs: additional keyword arguments to be passed to `self.call`.
Returns:
Output tensor(s)."
5491,add_variable,tensorflow/tensorflow/python/keras/engine/base_layer.py,2235,method,"Deprecated, do NOT use! Alias for `add_weight`."
5492,variables,tensorflow/tensorflow/python/keras/engine/base_layer.py,2241,method,"Returns the list of all layer variables/weights.
Alias of `self.weights`.
Returns:
A list of variables."
5493,trainable_variables,tensorflow/tensorflow/python/keras/engine/base_layer.py,2253,method,
5494,non_trainable_variables,tensorflow/tensorflow/python/keras/engine/base_layer.py,2258,method,
5495,check_type_return_shape,tensorflow/tensorflow/python/keras/engine/base_layer.py,764,method,
5496,getter,tensorflow/tensorflow/python/keras/engine/base_layer.py,586,method,
5497,TensorFlowOpLayer,tensorflow/tensorflow/python/keras/engine/base_layer.py,3042,class,"Wraps a TensorFlow Operation in a Layer.
This class is used internally by the Functional API. When a user
uses a raw TensorFlow Operation on symbolic tensors originating
from an `Input` Layer, the resultant operation will be wrapped
with this Layer object in order to make the operation compatible
with the Keras API.
This Layer will create a new, identical operation (except for inputs
and outputs) every time it is called. If `run_eagerly` is `True`,
the op creation and calculation will happen inside an Eager function.
Instances of this Layer are created when `autolambda` is called, which
is whenever a Layer's `__call__` encounters symbolic inputs that do
not have Keras metadata, or when a Network's `__init__` encounters
outputs that do not have Keras metadata.
Attributes:
node_def: String, the serialized NodeDef of the Op this layer will wrap.
name: String, the name of the Layer.
constants: Dict of NumPy arrays, the values of any Tensors needed for this
Operation that do not originate from a Keras `Input` Layer. Since all
placeholders must come from Keras `Input` Layers, these Tensors must be
treated as constant in the Functional API.
trainable: Bool, whether this Layer is trainable. Currently Variables are
not supported, and so this parameter has no effect.
dtype: The default dtype of this Layer. Inherited from `Layer` and has no
effect on this class, however is used in `get_config`."
5498,call,tensorflow/tensorflow/python/keras/engine/base_layer.py,3103,method,
5499,get_config,tensorflow/tensorflow/python/keras/engine/base_layer.py,3152,method,
5500,AddLoss,tensorflow/tensorflow/python/keras/engine/base_layer.py,3165,class,"Adds its inputs as a loss.
Attributes:
unconditional: Whether or not the loss should be conditioned on the inputs."
5501,call,tensorflow/tensorflow/python/keras/engine/base_layer.py,3179,method,
5502,get_config,tensorflow/tensorflow/python/keras/engine/base_layer.py,3183,method,
5503,AddMetric,tensorflow/tensorflow/python/keras/engine/base_layer.py,3189,class,"Adds its inputs as a metric.
Attributes:
aggregation: 'mean' or None. How the inputs should be aggregated.
metric_name: The name to use for this metric."
5504,call,tensorflow/tensorflow/python/keras/engine/base_layer.py,3202,method,
5505,get_config,tensorflow/tensorflow/python/keras/engine/base_layer.py,3206,method,
5506,DynamicLayer,tensorflow/tensorflow/python/keras/engine/base_layer_test.py,64,class,
5507,call,tensorflow/tensorflow/python/keras/engine/base_layer_test.py,69,method,
5508,compute_output_shape,tensorflow/tensorflow/python/keras/engine/base_layer_test.py,76,method,
5509,InvalidLayer,tensorflow/tensorflow/python/keras/engine/base_layer_test.py,80,class,
5510,call,tensorflow/tensorflow/python/keras/engine/base_layer_test.py,82,method,
5511,AddLayer,tensorflow/tensorflow/python/keras/engine/base_layer_test.py,1402,class,"A layer which adds its input to a variable.
Useful for testing a layer with a variable"
5512,build,tensorflow/tensorflow/python/keras/engine/base_layer_test.py,1408,method,
5513,call,tensorflow/tensorflow/python/keras/engine/base_layer_test.py,1412,method,
5514,IdentityLayer,tensorflow/tensorflow/python/keras/engine/base_layer_test.py,1416,class,"A layer that returns its input.
Useful for testing a layer without a variable."
5515,call,tensorflow/tensorflow/python/keras/engine/base_layer_test.py,1422,method,
5516,create_mean_metric,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,47,function,
5517,make_variable,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,55,function,"Temporary util to create a variable (relies on `variable_scope.variable`).
Some reuse-related technicalities prevent us from using
`variable_scope.get_variable()` directly, so we use a subcomponent
that has fewer constraints (`variable_scope.variable()`).
In the longer term, it seems like a similar ""default variable creator"" method
should exist in `Trackable` instead. When this happens, we can get
rid of this temporary solution.
TODO(fchollet): remove this method when no longer needed.
Arguments:
name: Variable name.
shape: Variable shape.
dtype: The type of the variable. Defaults to `self.dtype` or `float32`.
initializer: Initializer instance (callable).
trainable: Whether the variable should be part of the layer's
""trainable_variables"" (e.g. variables, biases)
or ""non_trainable_variables"" (e.g. BatchNorm mean, stddev).
Note, if the current variable scope is marked as non-trainable
then this parameter is ignored and any added variables are also
marked as non-trainable. `trainable` defaults to `True` unless
`synchronization` is set to `ON_READ`.
caching_device: Passed to `tf.Variable`.
validate_shape: Passed to `tf.Variable`.
constraint: Constraint instance (callable).
use_resource: Whether to use a `ResourceVariable`.
collections: List of graph collections keys. The new variable is added to
these collections. Defaults to `[GraphKeys.GLOBAL_VARIABLES]`.
synchronization: Indicates when a distributed a variable will be
aggregated. Accepted values are constants defined in the class
`tf.VariableSynchronization`. By default the synchronization is set to
`AUTO` and the current `DistributionStrategy` chooses
when to synchronize. If `synchronization` is set to `ON_READ`,
`trainable` must not be set to `True`.
aggregation: Indicates how a distributed variable will be aggregated.
Accepted values are constants defined in the class
`tf.VariableAggregation`.
partitioner: Not handled at this time.
Returns:
Variable instance."
5518,collect_previous_mask,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,146,function,"Retrieves the output mask(s) of the previous node.
Arguments:
input_tensors: An arbitrary structure of Tensors.
Returns:
A mask tensor or list of mask tensors."
5519,have_all_keras_metadata,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,162,function,
5520,generate_placeholders_from_shape,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,166,function,
5521,create_keras_history,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,170,function,"Wraps TensorFlow Operations for compatibility with the Functional API.
This method checks to see if a Tensor in `tensors` is missing Keras metadata
and has its origin in a Keras `Input` Layer. If so, this method will replace
the raw TensorFlow Operations that created this tensor with
`TensorFlowOpLayer` instances that create identical operations.
Any Tensors not originating from a Keras `Input` Layer will be treated as
constants when constructing `TensorFlowOpLayer` instances.
Arguments:
tensors: A structure of Tensors, some of which come from raw TensorFlow
operations and need to have Keras metadata assigned to them.
Returns:
created_layers: List. The `TensorFlowOpLayer` instances created to wrap
the raw Tensorflow operations."
5522,unnest_if_single_tensor,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,283,function,
5523,needs_keras_history,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,294,function,"Check if any Tensors need to be wrapped in TensorFlowOpLayers.
This will never return True inside a sublayer, because sublayers
do not need to create Keras History. Otherwise, this returns True
if one or more of `tensors` originates from a `keras.Input` and
does not have `_keras_history` set.
Arguments:
tensors: An arbitrary nested structure of Tensors.
ignore_call_context: Whether to ignore the check of if currently
outside of a `call` context. This is `True` when creating
KerasHistory inside `Node`, where we always know that Tensors
are being used with the Functional API.
Returns:
Bool, whether at least one Tensor needs to be wrapped."
5524,is_in_keras_graph,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,323,function,Returns if currently executing inside of a Keras graph.
5525,is_in_eager_or_tf_function,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,328,function,Returns if in eager mode or inside of a tf.function.
5526,is_in_tf_function,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,333,function,Returns if inside of a tf.function.
5527,uses_keras_history,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,351,function,"Check if at least one Tensor originates from a `keras.Input`.
This is `True` if at least one Tensor has its origin in a `keras.Input`.
Any Tensor that originates from a `keras.Input` will have a dependency
Tensor with a `_keras_history` attribute attached. Tensors that have
already been checked to not originate from a `keras.Input`
are marked as `_keras_history_checked`.
Arguments:
tensors: An arbitrary nested structure of Tensors.
Returns:
Bool, whether at least one Tensor originates from a `keras.Input`."
5528,mark_checked,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,396,function,"Marks that these Tensors should not be tracked.
This prevents Layers from attempting to create TensorFlowOpLayers
for these Tensors.
Arguments:
tensors: An arbitrary structure of Tensors."
5529,call_context,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,412,function,Returns currently active `CallContext`.
5530,CallContext,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,424,class,"Keeps track of properties currently inside a Layer/Model's `call`.
Attributes:
in_call: Whether currently inside the `call` of a Layer.
layer: The `Layer` whose `call` is currently active.
inputs: The inputs to the currently active `Layer`.
build_graph: Whether currently inside a Graph or FuncGraph.
training: Whether currently executing in training or inference mode.
saving: Whether currently saving to SavedModel.
frozen: Whether currently executing inside a `Layer` with `trainable` set to
`False`.
in_keras_graph: Whether executing inside the Keras Graph."
5531,enter,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,454,method,"Push a Layer and its inputs and state onto the current call context.
Arguments:
layer: The `Layer` whose `call` is currently active.
inputs: The inputs to the currently active `Layer`.
build_graph: Whether currently inside a Graph or FuncGraph.
training: Whether currently executing in training or inference mode.
saving: Whether currently saving to SavedModel.
Returns:
Context manager."
5532,layer,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,477,method,
5533,inputs,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,481,method,
5534,build_graph,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,485,method,
5535,training,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,489,method,
5536,saving,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,493,method,
5537,frozen,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,497,method,
5538,in_keras_graph,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,504,method,
5539,CallContextManager,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,513,class,Context manager for `CallContext`.
5540,training_arg_passed_to_call,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,546,function,Returns whether a user passed the `training` argument in `__call__`.
5541,is_subclassed,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,554,function,Returns True if the object is a subclassed layer or subclassed model.
5542,from_saved_model,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,560,function,Returns whether the layer is loaded from a SavedModel.
5543,check_graph_consistency,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,565,function,"Checks that tensors passed to `add_*` method match the Keras graph.
When one of the `add_*` method is called inside a V2 conditional branch,
the underlying tensor gets created in a FuncGraph managed by control_flow_v2.
We need to raise clear error messages in such cases.
Arguments:
tensor: Tensor to check, or `False` if it is known that an error
should be raised.
method: Caller method, one of {'add_metric', 'add_loss', 'add_update'}.
force_raise: If an error should be raised regardless of `tensor`.
Raises:
RuntimeError: In case of an out-of-graph tensor."
5544,mark_as_return,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,693,function,Marks `outputs` as the return values for automatic control deps.
5545,enable_v2_dtype_behavior,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,723,function,"Enable the V2 dtype behavior for Keras layers.
By default, the V2 dtype behavior is enabled in TensorFlow 2, so this function
is only useful if `tf.compat.v1.disable_v2_behavior` has been called. Since
mixed precision requires V2 dtype behavior to be enabled, this function allows
you to use mixed precision in Keras layers if `disable_v2_behavior` has been
called.
When enabled, the dtype of Keras layers defaults to floatx (which is typically
float32) instead of None. In addition, layers will automatically cast
floating-point inputs to the layer's dtype.
>>> x = tf.ones((4, 4, 4, 4), dtype='float64')
>>> layer = tf.keras.layers.Conv2D(filters=4, kernel_size=2)
>>> print(layer.dtype) # float32 since V2 dtype behavior is enabled
float32
>>> y = layer(x) # Layer casts inputs since V2 dtype behavior is enabled
>>> print(y.dtype.name)
float32
A layer author can opt-out their layer from the automatic input casting by
passing `autocast=False` to the base Layer's constructor. This disables the
autocasting part of the V2 behavior for that layer, but not the defaulting to
floatx part of the V2 behavior.
When a global `tf.keras.mixed_precision.experimental.Policy` is set, a Keras
layer's dtype will default to the global policy instead of floatx. Layers
will automatically cast inputs to the policy's compute_dtype."
5546,disable_v2_dtype_behavior,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,758,function,"Disables the V2 dtype behavior for Keras layers.
See `tf.compat.v1.keras.layers.enable_v2_dtype_behavior`."
5547,v2_dtype_behavior_enabled,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,767,function,Returns True if the V2 dtype behavior is enabled.
5548,TrackableWeightHandler,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,774,class,"Keras wrapper for handling tracking.Trackable object saving and restoring.
This class handles Trackables in both V1 and V2 modes, ensuring that they can
be saved and restored with the correct data and without adding additional ops
on every save.
Attributes:
trackable: The trackable to wrap.
num_tensors: The number of tensors that this trackable requires for saving."
5549,num_tensors,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,825,method,
5550,set_weights,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,828,method,
5551,get_tensors,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,836,method,
5552,no_ragged_support,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,846,function,
5553,is_split_variable,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,854,function,Returns True if `v` is either a PartionedVariable or a SharedVariable.
5554,has_weights,tensorflow/tensorflow/python/keras/engine/base_layer_utils.py,859,function,
5555,Layer,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,77,class,"Base layer class.
This is the class from which all layers inherit.
A layer is a class implementing common neural networks operations, such
as convolution, batch norm, etc. These operations require managing weights,
losses, updates, and inter-layer connectivity.
Users will just instantiate a layer and then treat it as a callable.
We recommend that descendants of `Layer` implement the following methods:
* `__init__()`: Save configuration in member variables
* `build()`: Called once from `__call__`, when we know the shapes of inputs
and `dtype`. Should have the calls to `add_weight()`, and then
call the super's `build()` (which sets `self.built = True`, which is
nice in case the user wants to call `build()` manually before the
first `__call__`).
* `call()`: Called in `__call__` after making sure `build()` has been called
once. Should actually perform the logic of applying the layer to the
input tensors (which should be passed in as the first argument).
Arguments:
trainable: Boolean, whether the layer's variables should be trainable.
name: String name of the layer.
dtype: The dtype of the layer's computations and weights (default of
`None` means use `tf.keras.backend.floatx` in TensorFlow 2, or the type
of the first input in TensorFlow 1).
dynamic: Set this to `True` if your layer should only be run eagerly, and
should not be used to generate a static computation graph.
This would be the case for a Tree-RNN or a recursive network,
for example, or generally for any layer that manipulates tensors
using Python control flow. If `False`, we assume that the layer can
safely be used to generate a static computation graph.
Attributes:
name: The name of the layer (string).
dtype: The dtype of the layer's computations and weights. If mixed
precision is used with a `tf.keras.mixed_precision.experimental.Policy`,
this is instead just the dtype of the layer's weights, as the computations
are done in a different dtype.
updates: List of update ops of this layer.
losses: List of losses added by this layer.
trainable_weights: List of variables to be included in backprop.
non_trainable_weights: List of variables that should not be
included in backprop.
weights: The concatenation of the lists trainable_weights and
non_trainable_weights (in this order).
trainable: Whether the layer should be trained (boolean).
input_spec: Optional (list of) `InputSpec` object(s) specifying the
constraints on inputs that can be accepted by the layer.
Each layer has a dtype, which is typically the dtype of the layer's
computations and variables. A layer's dtype can be queried via the
`Layer.dtype` property. The dtype is specified with the `dtype` constructor
argument. In TensorFlow 2, the dtype defaults to `tf.keras.backend.floatx()`
if no dtype is passed. `floatx()` itself defaults to ""float32"". Additionally,
layers will cast their inputs to the layer's dtype in TensorFlow 2. When mixed
precision is used, layers may have different computation and variable dtypes.
See `tf.keras.mixed_precision.experimental.Policy` for details on layer
dtypes."
5556,build,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,266,method,"Creates the variables of the layer (optional, for subclass implementers).
This is a method that implementers of subclasses of `Layer` or `Model`
can override if they need a state-creation step in-between
layer instantiation and layer call.
This is typically used to create the weights of `Layer` subclasses.
Arguments:
input_shape: Instance of `TensorShape`, or list of instances of
`TensorShape` if the layer expects a list of inputs
(one instance per input)."
5557,call,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,285,method,"This is where the layer's logic lives.
Arguments:
inputs: Input tensor, or list/tuple of input tensors.
**kwargs: Additional keyword arguments.
Returns:
A tensor or list/tuple of tensors."
5558,add_weight,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,318,method,"Adds a new variable to the layer.
Arguments:
name: Variable name.
shape: Variable shape. Defaults to scalar if unspecified.
dtype: The type of the variable. Defaults to `self.dtype` or `float32`.
initializer: Initializer instance (callable).
regularizer: Regularizer instance (callable).
trainable: Boolean, whether the variable should be part of the layer's
""trainable_variables"" (e.g. variables, biases)
or ""non_trainable_variables"" (e.g. BatchNorm mean and variance).
Note that `trainable` cannot be `True` if `synchronization`
is set to `ON_READ`.
constraint: Constraint instance (callable).
partitioner: Partitioner to be passed to the `Trackable` API.
use_resource: Whether to use `ResourceVariable`.
synchronization: Indicates when a distributed a variable will be
aggregated. Accepted values are constants defined in the class
`tf.VariableSynchronization`. By default the synchronization is set to
`AUTO` and the current `DistributionStrategy` chooses
when to synchronize. If `synchronization` is set to `ON_READ`,
`trainable` must not be set to `True`.
aggregation: Indicates how a distributed variable will be aggregated.
Accepted values are constants defined in the class
`tf.VariableAggregation`.
**kwargs: Additional keyword arguments. Accepted values are `getter`,
`collections`, `experimental_autocast` and `caching_device`.
Returns:
The created variable. Usually either a `Variable` or `ResourceVariable`
instance. If `partitioner` is not `None`, a `PartitionedVariable`
instance is returned.
Raises:
RuntimeError: If called with partitioned variable regularization and
eager execution is enabled.
ValueError: When giving unsupported dtype and no initializer or when
trainable has been set to True with synchronization set as `ON_READ`."
5559,get_config,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,479,method,"Returns the config of the layer.
A layer config is a Python dictionary (serializable)
containing the configuration of a layer.
The same layer can be reinstantiated later
(without its trained weights) from this configuration.
The config of a layer does not include connectivity
information, nor the layer class name. These are handled
by `Network` (one layer of abstraction above).
Returns:
Python dictionary."
5560,from_config,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,516,method,"Creates a layer from its config.
This method is the reverse of `get_config`,
capable of instantiating the same layer from the config
dictionary. It does not handle layer connectivity
(handled by Network), nor weights (handled by `set_weights`).
Arguments:
config: A Python dictionary, typically the
output of get_config.
Returns:
A layer instance."
5561,compute_output_shape,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,533,method,"Computes the output shape of the layer.
If the layer has not been built, this method will call `build` on the
layer. This assumes that the layer will later be used with inputs that
match the input shape provided here.
Arguments:
input_shape: Shape tuple (tuple of integers)
or list of shape tuples (one per output tensor of the layer).
Shape tuples can include None for free dimensions,
instead of an integer.
Returns:
An input shape tuple."
5562,compute_output_signature,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,577,method,"Compute the output tensor signature of the layer based on the inputs.
Unlike a TensorShape object, a TensorSpec object contains both shape
and dtype information for a tensor. This method allows layers to provide
output dtype information if it is different from the input dtype.
For any layer that doesn't implement this function,
the framework will fall back to use `compute_output_shape`, and will
assume that the output dtype matches the input dtype.
Args:
input_signature: Single TensorSpec or nested structure of TensorSpec
objects, describing a candidate input for the layer.
Returns:
Single TensorSpec or nested structure of TensorSpec objects, describing
how the layer would transform the provided input.
Raises:
TypeError: If input_signature contains a non-TensorSpec object."
5563,compute_mask,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,617,method,"Computes an output mask tensor.
Arguments:
inputs: Tensor or list of tensors.
mask: Tensor or list of tensors.
Returns:
None or a tensor (or list of tensors,
one per output tensor of the layer)."
5564,dtype,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,847,method,
5565,name,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,851,method,
5566,dynamic,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,855,method,
5567,stateful,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,860,method,
5568,stateful,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,864,method,
5569,trainable,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,868,method,
5570,trainable,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,872,method,
5571,activity_regularizer,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,878,method,Optional regularizer function for the output of this layer.
5572,activity_regularizer,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,883,method,Optional regularizer function for the output of this layer.
5573,input_spec,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,888,method,
5574,input_spec,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,895,method,
5575,trainable_weights,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,903,method,
5576,non_trainable_weights,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,911,method,
5577,weights,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,924,method,"Returns the list of all layer variables/weights.
Returns:
A list of variables."
5578,updates,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,933,method,
5579,losses,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,953,method,"Losses which are associated with this `Layer`.
Variable regularization tensors are created when this property is accessed,
so it is eager safe: accessing `losses` under a `tf.GradientTape` will
propagate gradients back to the corresponding variables.
Returns:
A list of tensors."
5580,add_loss,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,977,method,"Add loss tensor(s), potentially dependent on layer inputs.
Some losses (for instance, activity regularization losses) may be dependent
on the inputs passed when calling a layer. Hence, when reusing the same
layer on different inputs `a` and `b`, some entries in `layer.losses` may
be dependent on `a` and some on `b`. This method automatically keeps track
of dependencies.
This method can be used inside a subclassed layer or model's `call`
function, in which case `losses` should be a Tensor or list of Tensors.
Example:
```python
class MyLayer(tf.keras.layers.Layer):
def call(inputs, self):
self.add_loss(tf.abs(tf.reduce_mean(inputs)), inputs=True)
return inputs
```
This method can also be called directly on a Functional Model during
construction. In this case, any loss Tensors passed to this Model must
be symbolic and be able to be traced back to the model's `Input`s. These
losses become part of the model's topology and are tracked in `get_config`.
Example:
```python
inputs = tf.keras.Input(shape=(10,))
x = tf.keras.layers.Dense(10)(inputs)
outputs = tf.keras.layers.Dense(1)(x)
model = tf.keras.Model(inputs, outputs)
# Actvity regularization.
model.add_loss(tf.abs(tf.reduce_mean(x)))
```
If this is not the case for your loss (if, for example, your loss references
a `Variable` of one of the model's layers), you can wrap your loss in a
zero-argument lambda. These losses are not tracked as part of the model's
topology since they can't be serialized.
Example:
```python
inputs = tf.keras.Input(shape=(10,))
x = tf.keras.layers.Dense(10)(inputs)
outputs = tf.keras.layers.Dense(1)(x)
model = tf.keras.Model(inputs, outputs)
# Weight regularization.
model.add_loss(lambda: tf.reduce_mean(x.kernel))
```
The `get_losses_for` method allows to retrieve the losses relevant to a
specific set of inputs.
Arguments:
losses: Loss tensor, or list/tuple of tensors. Rather than tensors, losses
may also be zero-argument callables which create a loss tensor.
inputs: Ignored when executing eagerly. If anything other than None is
passed, it signals the losses are conditional on some of the layer's
inputs, and thus they should only be run where these inputs are
available. This is the case for activity regularization losses, for
instance. If `None` is passed, the losses are assumed
to be unconditional, and will apply across all dataflows of the layer
(e.g. weight regularization losses)."
5581,metrics,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1092,method,
5582,add_metric,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1099,method,"Adds metric tensor to the layer.
Args:
value: Metric tensor.
aggregation: Sample-wise metric reduction function. If `aggregation=None`,
it indicates that the metric tensor provided has been aggregated
already. eg, `bin_acc = BinaryAccuracy(name='acc')` followed by
`model.add_metric(bin_acc(y_true, y_pred))`. If aggregation='mean', the
given metric tensor will be sample-wise reduced using `mean` function.
eg, `model.add_metric(tf.reduce_sum(outputs), name='output_mean',
aggregation='mean')`.
name: String metric name.
Raises:
ValueError: If `aggregation` is anything other than None or `mean`."
5583,add_update,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1167,method,"Add update op(s), potentially dependent on layer inputs.
Weight updates (for instance, the updates of the moving mean and variance
in a BatchNormalization layer) may be dependent on the inputs passed
when calling a layer. Hence, when reusing the same layer on
different inputs `a` and `b`, some entries in `layer.updates` may be
dependent on `a` and some on `b`. This method automatically keeps track
of dependencies.
The `get_updates_for` method allows to retrieve the updates relevant to a
specific set of inputs.
This call is ignored when eager execution is enabled (in that case, variable
updates are run on the fly and thus do not need to be tracked for later
execution).
Arguments:
updates: Update op, or list/tuple of update ops, or zero-arg callable
that returns an update op. A zero-arg callable should be passed in
order to disable running the updates by setting `trainable=False`
on this Layer, when executing in Eager mode.
inputs: Deprecated, will be automatically inferred."
5584,set_weights,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1239,method,"Sets the weights of the layer, from Numpy arrays.
The weights of a layer represent the state of the layer. This function
sets the weight values from numpy arrays. The weight values should be
passed in the order they are created by the layer. Note that the layer's
weights must be instantiated before calling this function by calling
the layer.
For example, a Dense layer returns a list of two values-- per-output
weights and the bias value. These can be used to set the weights of another
Dense layer:
>>> a = tf.keras.layers.Dense(1,
... kernel_initializer=tf.constant_initializer(1.))
>>> a_out = a(tf.convert_to_tensor([[1., 2., 3.]]))
>>> a.get_weights()
[array([[1.],
[1.],
[1.]], dtype=float32), array([0.], dtype=float32)]
>>> b = tf.keras.layers.Dense(1,
... kernel_initializer=tf.constant_initializer(2.))
>>> b_out = b(tf.convert_to_tensor([[10., 20., 30.]]))
>>> b.get_weights()
[array([[2.],
[2.],
[2.]], dtype=float32), array([0.], dtype=float32)]
>>> b.set_weights(a.get_weights())
>>> b.get_weights()
[array([[1.],
[1.],
[1.]], dtype=float32), array([0.], dtype=float32)]
Arguments:
weights: a list of Numpy arrays. The number
of arrays and their shape must match
number of the dimensions of the weights
of the layer (i.e. it should match the
output of `get_weights`).
Raises:
ValueError: If the provided weights list does not match the
layer's specifications."
5585,get_weights,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1319,method,"Returns the current weights of the layer.
The weights of a layer represent the state of the layer. This function
returns both trainable and non-trainable weight values associated with this
layer as a list of Numpy arrays, which can in turn be used to load state
into similarly parameterized layers.
For example, a Dense layer returns a list of two values-- per-output
weights and the bias value. These can be used to set the weights of another
Dense layer:
>>> a = tf.keras.layers.Dense(1,
... kernel_initializer=tf.constant_initializer(1.))
>>> a_out = a(tf.convert_to_tensor([[1., 2., 3.]]))
>>> a.get_weights()
[array([[1.],
[1.],
[1.]], dtype=float32), array([0.], dtype=float32)]
>>> b = tf.keras.layers.Dense(1,
... kernel_initializer=tf.constant_initializer(2.))
>>> b_out = b(tf.convert_to_tensor([[10., 20., 30.]]))
>>> b.get_weights()
[array([[2.],
[2.],
[2.]], dtype=float32), array([0.], dtype=float32)]
>>> b.set_weights(a.get_weights())
>>> b.get_weights()
[array([[1.],
[1.],
[1.]], dtype=float32), array([0.], dtype=float32)]
Returns:
Weights values as a list of numpy arrays."
5586,get_updates_for,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1363,method,"Retrieves updates relevant to a specific set of inputs.
Arguments:
inputs: Input tensor or list/tuple of input tensors.
Returns:
List of update ops of the layer that depend on `inputs`."
5587,get_losses_for,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1382,method,"Retrieves losses relevant to a specific set of inputs.
Arguments:
inputs: Input tensor or list/tuple of input tensors.
Returns:
List of loss tensors of the layer that depend on `inputs`."
5588,get_input_mask_at,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1401,method,"Retrieves the input mask tensor(s) of a layer at a given node.
Arguments:
node_index: Integer, index of the node
from which to retrieve the attribute.
E.g. `node_index=0` will correspond to the
first time the layer was called.
Returns:
A mask tensor
(or list of tensors if the layer has multiple inputs)."
5589,get_output_mask_at,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1420,method,"Retrieves the output mask tensor(s) of a layer at a given node.
Arguments:
node_index: Integer, index of the node
from which to retrieve the attribute.
E.g. `node_index=0` will correspond to the
first time the layer was called.
Returns:
A mask tensor
(or list of tensors if the layer has multiple outputs)."
5590,input_mask,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1440,method,"Retrieves the input mask tensor(s) of a layer.
Only applicable if the layer has exactly one inbound node,
i.e. if it is connected to one incoming layer.
Returns:
Input mask tensor (potentially None) or list of input
mask tensors.
Raises:
AttributeError: if the layer is connected to
more than one incoming layers."
5591,output_mask,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1461,method,"Retrieves the output mask tensor(s) of a layer.
Only applicable if the layer has exactly one inbound node,
i.e. if it is connected to one incoming layer.
Returns:
Output mask tensor (potentially None) or list of output
mask tensors.
Raises:
AttributeError: if the layer is connected to
more than one incoming layers."
5592,get_input_shape_at,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1481,method,"Retrieves the input shape(s) of a layer at a given node.
Arguments:
node_index: Integer, index of the node
from which to retrieve the attribute.
E.g. `node_index=0` will correspond to the
first time the layer was called.
Returns:
A shape tuple
(or list of shape tuples if the layer has multiple inputs).
Raises:
RuntimeError: If called in Eager mode."
5593,get_output_shape_at,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1500,method,"Retrieves the output shape(s) of a layer at a given node.
Arguments:
node_index: Integer, index of the node
from which to retrieve the attribute.
E.g. `node_index=0` will correspond to the
first time the layer was called.
Returns:
A shape tuple
(or list of shape tuples if the layer has multiple outputs).
Raises:
RuntimeError: If called in Eager mode."
5594,get_input_at,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1519,method,"Retrieves the input tensor(s) of a layer at a given node.
Arguments:
node_index: Integer, index of the node
from which to retrieve the attribute.
E.g. `node_index=0` will correspond to the
first time the layer was called.
Returns:
A tensor (or list of tensors if the layer has multiple inputs).
Raises:
RuntimeError: If called in Eager mode."
5595,get_output_at,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1537,method,"Retrieves the output tensor(s) of a layer at a given node.
Arguments:
node_index: Integer, index of the node
from which to retrieve the attribute.
E.g. `node_index=0` will correspond to the
first time the layer was called.
Returns:
A tensor (or list of tensors if the layer has multiple outputs).
Raises:
RuntimeError: If called in Eager mode."
5596,input,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1556,method,"Retrieves the input tensor(s) of a layer.
Only applicable if the layer has exactly one input,
i.e. if it is connected to one incoming layer.
Returns:
Input tensor or list of input tensors.
Raises:
RuntimeError: If called in Eager mode.
AttributeError: If no inbound nodes are found."
5597,output,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1575,method,"Retrieves the output tensor(s) of a layer.
Only applicable if the layer has exactly one output,
i.e. if it is connected to one incoming layer.
Returns:
Output tensor or list of output tensors.
Raises:
AttributeError: if the layer is connected to more than one incoming
layers.
RuntimeError: if called in Eager mode."
5598,input_shape,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1594,method,"Retrieves the input shape(s) of a layer.
Only applicable if the layer has exactly one input,
i.e. if it is connected to one incoming layer, or if all inputs
have the same shape.
Returns:
Input shape, as an integer shape tuple
(or list of shape tuples, one tuple per input tensor).
Raises:
AttributeError: if the layer has no defined input_shape.
RuntimeError: if called in Eager mode."
5599,count_params,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1625,method,"Count the total number of scalars composing the weights.
Returns:
An integer count.
Raises:
ValueError: if the layer isn't yet built
(in which case its weights aren't yet defined)."
5600,output_shape,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1647,method,"Retrieves the output shape(s) of a layer.
Only applicable if the layer has one output,
or if all outputs have the same shape.
Returns:
Output shape, as an integer shape tuple
(or list of shape tuples, one tuple per output tensor).
Raises:
AttributeError: if the layer has no defined output shape.
RuntimeError: if called in Eager mode."
5601,inbound_nodes,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1679,method,"Deprecated, do NOT use! Only for compatibility with external Keras."
5602,outbound_nodes,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1685,method,"Deprecated, do NOT use! Only for compatibility with external Keras."
5603,apply,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1696,method,"Deprecated, do NOT use!
This is an alias of `self.__call__`.
Arguments:
inputs: Input tensor(s).
*args: additional positional arguments to be passed to `self.call`.
**kwargs: additional keyword arguments to be passed to `self.call`.
Returns:
Output tensor(s)."
5604,add_variable,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1714,method,"Deprecated, do NOT use! Alias for `add_weight`."
5605,variables,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1719,method,"Returns the list of all layer variables/weights.
Alias of `self.weights`.
Returns:
A list of variables."
5606,trainable_variables,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1730,method,
5607,non_trainable_variables,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1734,method,
5608,check_type_return_shape,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,598,method,
5609,process_update,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1213,method,"Standardize update ops.
Arguments:
x: Tensor, op, or callable.
Returns:
An update op."
5610,getter,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,426,method,
5611,f,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,1806,method,Cast a single Tensor or TensorSpec to the compute dtype.
5612,KerasHistory,tensorflow/tensorflow/python/keras/engine/base_layer_v1.py,2406,class,"Tracks the Layer call that created a Tensor, for Keras Graph Networks.
During construction of Keras Graph Networks, this metadata is added to
each Tensor produced as the output of a Layer, starting with an
`InputLayer`. This allows Keras to track how each Tensor was produced, and
this information is later retraced by the `keras.engine.Network` class to
reconstruct the Keras Graph Network.
Attributes:
layer: The Layer that produced the Tensor.
node_index: The specific call to the Layer that produced this Tensor. Layers
can be called multiple times in order to share weights. A new node is
created every time a Tensor is called.
tensor_index: The output index for this Tensor. Always zero if the Layer
that produced this Tensor only has one output. Nested structures of
Tensors are deterministically assigned an index via `nest.flatten`."
5613,PreprocessingLayer,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer.py,49,class,Base class for PreprocessingLayers.
5614,adapt,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer.py,53,method,"Fits the state of the preprocessing layer to the data being passed.
Arguments:
data: The data to train on. It can be passed either as a tf.data
Dataset, or as a numpy array.
reset_state: Optional argument specifying whether to clear the state of
the layer at the start of the call to `adapt`, or whether to start
from the existing state. This argument may not be relevant to all
preprocessing layers: a subclass of PreprocessingLayer may choose to
throw if 'reset_state' is set to False."
5615,CombinerPreprocessingLayer,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer.py,69,class,"Base class for PreprocessingLayers that do computation using a Combiner.
This class provides several helper methods to make creating a
PreprocessingLayer easier. It assumes that the core of your computation will
be done via a Combiner object. Subclassing this class to create a
PreprocessingLayer allows your layer to be compatible with distributed
computation.
This class is compatible with Tensorflow 2.0+."
5616,adapt,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer.py,136,method,"Fits the state of the preprocessing layer to the data being passed.
Arguments:
data: The data to train on. It can be passed either as a tf.data Dataset,
or as a numpy array.
reset_state: Optional argument specifying whether to clear the state of
the layer at the start of the call to `adapt`, or whether to start from
the existing state. Subclasses may choose to throw if reset_state is set
to 'False'."
5617,convert_to_list,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer.py,246,function,"Convert a TensorLike, CompositeTensor, or ndarray into a Python list."
5618,Combiner,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer.py,285,class,"Functional object that defines a shardable computation.
This object defines functions required to create and manipulate data objects.
These data objects, referred to below as 'accumulators', are computation-
specific and may be implemented alongside concrete subclasses of Combiner
(if necessary - some computations may be simple enough that standard Python
types can be used as accumulators).
The intent for this class is that by describing computations in this way, we
can arbitrarily shard a dataset, perform computations on a subset, and then
merge the computation into a final result. This enables distributed
computation.
The combiner itself does not own any state - all computational state is owned
by the accumulator objects. This is so that we can have an arbitrary number of
Combiners (thus sharding the computation N ways) without risking any change
to the underlying computation. These accumulator objects are uniquely
associated with each Combiner; a Combiner defines what the accumulator object
should be and will only work with accumulators of that type."
5619,compute,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer.py,312,method,"Compute a step in this computation, returning a new accumulator.
This method computes a step of the computation described by this Combiner.
If an accumulator is passed, the data in that accumulator is also used; so
compute(batch_values) results in f(batch_values), while
compute(batch_values, accumulator) results in
merge(f(batch_values), accumulator).
Args:
batch_values: A list of ndarrays representing the values of the inputs for
this step of the computation.
accumulator: the current accumulator. Can be None.
Returns:
An accumulator that includes the passed batch of inputs."
5620,merge,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer.py,332,method,"Merge several accumulators to a single accumulator.
This method takes the partial values in several accumulators and combines
them into a single accumulator. This computation must not be order-specific
(that is, merge([a, b]) must return the same result as merge([b, a]).
Args:
accumulators: the accumulators to merge, as a list.
Returns:
A merged accumulator."
5621,extract,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer.py,348,method,"Convert an accumulator into a dict of output values.
Args:
accumulator: The accumulator to convert.
Returns:
A dict of ndarrays representing the data in this accumulator."
5622,restore,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer.py,360,method,"Create an accumulator based on 'output'.
This method creates a new accumulator with identical internal state to the
one used to create the data in 'output'. This means that if you do
output_data = combiner.extract(accumulator_1)
accumulator_2 = combiner.restore(output_data)
then accumulator_1 and accumulator_2 will have identical internal state, and
computations using either of them will be equivalent.
Args:
output: The data output from a previous computation. Should be in the same
form as provided by 'extract_output'.
Returns:
A new accumulator."
5623,serialize,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer.py,382,method,"Serialize an accumulator for a remote call.
This function serializes an accumulator to be sent to a remote process.
Args:
accumulator: The accumulator to serialize.
Returns:
A byte string representing the passed accumulator."
5624,deserialize,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer.py,396,method,"Deserialize an accumulator received from 'serialize()'.
This function deserializes an accumulator serialized by 'serialize()'.
Args:
encoded_accumulator: A byte string representing an accumulator.
Returns:
The accumulator represented by the passed byte_string."
5625,AddingPreprocessingLayer,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer_test.py,47,class,
5626,build,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer_test.py,55,method,
5627,set_total,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer_test.py,63,method,"This is an example of how a subclass would implement a direct setter.
These methods should generally just create a dict mapping the correct names
to the relevant passed values, and call self._set_state_variables() with the
dict of data.
Args:
sum_value: The total to set."
5628,call,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer_test.py,75,method,
5629,compute,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer_test.py,81,method,"Compute a step in this computation, returning a new accumulator."
5630,merge,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer_test.py,89,method,Merge several accumulators to a single accumulator.
5631,extract,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer_test.py,97,method,Convert an accumulator into a dict of output values.
5632,restore,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer_test.py,103,method,Create an accumulator based on 'output'.
5633,serialize,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer_test.py,110,method,Serialize an accumulator for a remote call.
5634,deserialize,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer_test.py,114,method,Deserialize an accumulator received from 'serialize()'.
5635,AddingPreprocessingLayerV1,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer_test.py,119,class,
5636,get_layer,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer_test.py,125,function,
5637,CombinerPreprocessingLayer,tensorflow/tensorflow/python/keras/engine/base_preprocessing_layer_v1.py,26,class,"V1-compatible CombinerPreprocessingLayer.
This class overrides several methods of the CombinerPreprocessingLayer to
make it compatible with V1 execution. End users should not need to worry about
the implementation details here; Keras will export the appropriate class under
the 'CombinerPreprocessingLayer' symbol. (Users should not directly
instantiate engine.base_preprocessing_layer/_v1.CombinerPreprocessingLayer).
When creating a subclass of PreprocessingLayer, you can create a V1-compatible
subclass as follows:
class MyProcLayer(MyProcLayer,
base_preprocessing_layer_v1.CombinerPreprocessingLayer):
pass
Note that the same classname is required for serialization purposes.
This is only necessary for internal classes, since any class that inherits
from tf.keras.[...].CombinerPreprocessingLayer will get the right symbol."
5638,Container,tensorflow/tensorflow/python/keras/engine/compile_utils.py,33,class,Base Container class.
5639,build,tensorflow/tensorflow/python/keras/engine/compile_utils.py,39,method,
5640,LossesContainer,tensorflow/tensorflow/python/keras/engine/compile_utils.py,106,class,A container class for losses passed to `Model.compile`.
5641,metrics,tensorflow/tensorflow/python/keras/engine/compile_utils.py,123,method,Per-output loss metrics.
5642,build,tensorflow/tensorflow/python/keras/engine/compile_utils.py,133,method,One-time setup of loss objects.
5643,MetricsContainer,tensorflow/tensorflow/python/keras/engine/compile_utils.py,276,class,A container class for metrics passed to `Model.compile`.
5644,metrics,tensorflow/tensorflow/python/keras/engine/compile_utils.py,291,method,Metrics created by this container.
5645,build,tensorflow/tensorflow/python/keras/engine/compile_utils.py,297,method,One-time setup of metric objects.
5646,update_state,tensorflow/tensorflow/python/keras/engine/compile_utils.py,381,method,Updates the state of per-output metrics.
5647,create_pseudo_output_names,tensorflow/tensorflow/python/keras/engine/compile_utils.py,493,function,Create pseudo output names for a subclassed Model.
5648,create_pseudo_input_names,tensorflow/tensorflow/python/keras/engine/compile_utils.py,498,function,Create pseudo input names for a subclassed Model.
5649,map_to_output_names,tensorflow/tensorflow/python/keras/engine/compile_utils.py,548,function,"Maps a dict to a list using `output_names` as keys.
This is a convenience feature only. When a `Model`'s outputs
are a list, you can specify per-output losses and metrics as
a dict, where the keys are the output names. If you specify
per-output losses and metrics via the same structure as the
`Model`'s outputs (recommended), no mapping is performed.
For the Functional API, the output names are the names of the
last layer of each output. For the Subclass API, the output names
are determined by `create_pseudo_output_names` (For example:
`['output_1', 'output_2']` for a list of outputs).
This mapping preserves backwards compatibility for `compile` and
`fit`.
Arguments:
y_pred: Sample outputs of the Model, to determine if this convenience
feature should be applied (`struct` is returned unmodified if `y_pred`
isn't a flat list).
output_names: List. The names of the outputs of the Model.
struct: The structure to map.
Returns:
`struct` mapped to a list in same order as `output_names`."
5650,map_missing_dict_keys,tensorflow/tensorflow/python/keras/engine/compile_utils.py,595,function,Replaces missing dict keys in `struct` with `None` placeholders.
5651,match_dtype_and_rank,tensorflow/tensorflow/python/keras/engine/compile_utils.py,605,function,Match dtype and rank of predictions.
5652,get_mask,tensorflow/tensorflow/python/keras/engine/compile_utils.py,625,function,Returns Keras mask from tensor.
5653,apply_mask,tensorflow/tensorflow/python/keras/engine/compile_utils.py,630,function,Applies any mask on predictions to sample weights.
5654,ControlFlowLayer1,tensorflow/tensorflow/python/keras/engine/control_flow_test.py,37,class,Layer with an `if` condition in call.
5655,call,tensorflow/tensorflow/python/keras/engine/control_flow_test.py,40,method,
5656,ControlFlowLayer2,tensorflow/tensorflow/python/keras/engine/control_flow_test.py,47,class,Layer with a `for` loop in call.
5657,call,tensorflow/tensorflow/python/keras/engine/control_flow_test.py,50,method,
5658,NestedControlFlowLayer,tensorflow/tensorflow/python/keras/engine/control_flow_test.py,60,class,Layer nested with a control flow layer.
5659,call,tensorflow/tensorflow/python/keras/engine/control_flow_test.py,67,method,
5660,ControlFlowModel,tensorflow/tensorflow/python/keras/engine/control_flow_test.py,71,class,Model with an `if` condition in call.
5661,call,tensorflow/tensorflow/python/keras/engine/control_flow_test.py,74,method,
5662,NestedControlFlowModel,tensorflow/tensorflow/python/keras/engine/control_flow_test.py,81,class,Model with an `if` condition in call using a control flow layer.
5663,call,tensorflow/tensorflow/python/keras/engine/control_flow_test.py,88,method,
5664,FunctionControlFlowModel,tensorflow/tensorflow/python/keras/engine/control_flow_test.py,96,class,Model with control flow where `call` is wrapped in function already.
5665,call,tensorflow/tensorflow/python/keras/engine/control_flow_test.py,100,method,
5666,MultiInputSubclassed,tensorflow/tensorflow/python/keras/engine/correctness_test.py,30,class,Subclassed Model that adds its inputs and then adds a bias.
5667,call,tensorflow/tensorflow/python/keras/engine/correctness_test.py,38,method,
5668,multi_input_functional,tensorflow/tensorflow/python/keras/engine/correctness_test.py,43,function,Functional Model that adds its inputs and then adds a bias.
5669,DataAdapter,tensorflow/tensorflow/python/keras/engine/data_adapter.py,73,class,"Base class for input data adapter.
In TF 2.0, tf.data is the preferred API for user to feed in data. In order
to simplify the training code path, all the input data object will be
converted to `tf.data.Dataset` if possible.
Note that since this class is mainly targeted for TF 2.0, it might have a lot
of assumptions under the hood, eg eager context by default, distribution
strategy, etc. In the meantime, some legacy feature support might be dropped,
eg, Iterator from dataset API in v1, etc.
The sample usage of this class is like:
```
x = tf.data.Dataset.range(100)
adapter_cls = [NumpyArrayDataAdapter, ..., DatasetAdapter]
applicable_adapters = [cls for cls in adapter_cls if cls.can_handle(x)]
if len(applicable_adapters) != 1:
raise ValueError(""Expect only one adapter class to handle the input"")
dataset = applicable_adapters[0](x).get_dataset()
for data in dataset:
# training
```"
5670,can_handle,tensorflow/tensorflow/python/keras/engine/data_adapter.py,101,method,"Whether the current DataAdapter could handle the input x and y.
Structure wise, x and y can be single object, or list of objects if there
multiple input/output, or dictionary of objects when the intput/output are
named.
Args:
x: input features.
y: target labels. Note that y could be None in the case of prediction.
Returns:
boolean"
5671,get_dataset,tensorflow/tensorflow/python/keras/engine/data_adapter.py,144,method,"Get a dataset instance for the current DataAdapter.
Note that the dataset returned does not repeat for epoch, so caller might
need to create new iterator for the same dataset at the beginning of the
epoch. This behavior might change in future.
Returns:
An tf.dataset.Dataset. Caller might use the dataset in different
context, eg iter(dataset) in eager to get the value directly, or in graph
mode, provide the iterator tensor to Keras model function."
5672,get_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,159,method,"Return the size (number of batches) for the dataset created.
For certain type of the data input, the number of batches is known, eg for
Numpy data, the size is same as (number_of_element / batch_size). Whereas
for dataset or python generator, the size is unknown since it may or may not
have a end state.
Returns:
int, the number of batches for the dataset, or None if it is unknown. The
caller could use this to control the loop of training, show progress bar,
or handle unexpected StopIteration error."
5673,batch_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,175,method,"Return the batch size of the dataset created.
For certain type of the data input, the batch size is known, and even
required, like numpy array. Where as for dataset, the batch is unknown
unless we take a peek.
Returns:
int, the batch size of the dataset, or None if it is unknown."
5674,representative_batch_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,187,method,"Return a representative size for batches in the dataset.
This is not guaranteed to be the batch size for all batches in the
dataset. It just needs to be a rough approximation for batch sizes in
the dataset.
Returns:
int, a representative size for batches found in the dataset,
or None if it is unknown."
5675,has_partial_batch,tensorflow/tensorflow/python/keras/engine/data_adapter.py,201,method,Whether the dataset has partial batch at the end.
5676,partial_batch_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,206,method,"The size of the final partial batch for dataset.
Will return None if has_partial_batch is False or batch_size is None."
5677,should_recreate_iterator,tensorflow/tensorflow/python/keras/engine/data_adapter.py,214,method,Returns whether a new iterator should be created every epoch.
5678,get_samples,tensorflow/tensorflow/python/keras/engine/data_adapter.py,218,method,"Returns number of samples in the data, or `None`."
5679,on_epoch_end,tensorflow/tensorflow/python/keras/engine/data_adapter.py,227,method,A hook called after each epoch.
5680,TensorLikeDataAdapter,tensorflow/tensorflow/python/keras/engine/data_adapter.py,232,class,"Adapter that handles Tensor-like objects, e.g. EagerTensor and NumPy."
5681,can_handle,tensorflow/tensorflow/python/keras/engine/data_adapter.py,236,method,
5682,slice_inputs,tensorflow/tensorflow/python/keras/engine/data_adapter.py,366,method,"Slice inputs into a Dataset of batches.
Given a Dataset of batch indices and the unsliced inputs,
this step slices the inputs in a parallelized fashion
and produces a dataset of input batches.
Args:
indices_dataset: A Dataset of batched indices
inputs: A python data structure that contains the inputs, targets,
and possibly sample weights.
Returns:
A Dataset of input batches matching the batch indices."
5683,get_dataset,tensorflow/tensorflow/python/keras/engine/data_adapter.py,403,method,
5684,get_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,406,method,
5685,batch_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,409,method,
5686,has_partial_batch,tensorflow/tensorflow/python/keras/engine/data_adapter.py,412,method,
5687,partial_batch_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,415,method,
5688,should_recreate_iterator,tensorflow/tensorflow/python/keras/engine/data_adapter.py,418,method,
5689,permutation,tensorflow/tensorflow/python/keras/engine/data_adapter.py,308,method,
5690,slice_batch_indices,tensorflow/tensorflow/python/keras/engine/data_adapter.py,323,method,"Convert a Tensor of indices into a dataset of batched indices.
This step can be accomplished in several ways. The most natural is to
slice the Tensor in a Dataset map. (With a condition on the upper index to
handle the partial batch.) However it turns out that coercing the Tensor
into a shape which is divisible by the batch size (and handling the last
partial batch separately) allows for a much more favorable memory access
pattern and improved performance.
Args:
indices: Tensor which determines the data order for an entire epoch.
Returns:
A Dataset of batched indices."
5691,grab_batch,tensorflow/tensorflow/python/keras/engine/data_adapter.py,386,method,
5692,shuffle_batch,tensorflow/tensorflow/python/keras/engine/data_adapter.py,360,method,
5693,GenericArrayLikeDataAdapter,tensorflow/tensorflow/python/keras/engine/data_adapter.py,423,class,"Adapter that handles array-like data without forcing it into memory.
As an example, this adapter handles `keras.utils.HDF5Matrix` which holds
datasets that may be too big to fully fit into memory.
Specifically, this adapter handles any Python class which implements:
`__get_item__`, `__len__`, `shape`, and `dtype` with the same meanings
as Numpy, but it ignores any case where all the inputs are Tensors or Numpy
arrays (because that case is handled by the base TensorLikeDataAdapter).
It ignores scipy sparse matrices and Composite Tensors because those are
handled by the CompositeTensorDataAdapter.
It also does not handle lists/tuples of scalars, because those are handled
by the ListsOfScalarsDataAdapter."
5694,can_handle,tensorflow/tensorflow/python/keras/engine/data_adapter.py,442,method,
5695,slice_inputs,tensorflow/tensorflow/python/keras/engine/data_adapter.py,471,method,"Slice inputs into a Dataset of batches.
Given a Dataset of batch indices and the unsliced inputs,
this step slices the inputs in a parallelized fashion
and produces a dataset of input batches.
Args:
indices_dataset: A Dataset of batched indices
inputs: A python data structure that contains the inputs, targets,
and possibly sample weights.
Returns:
A Dataset of input batches matching the batch indices."
5696,dynamic_shape_like,tensorflow/tensorflow/python/keras/engine/data_adapter.py,487,method,
5697,grab_batch,tensorflow/tensorflow/python/keras/engine/data_adapter.py,497,method,Grab a batch of data from the inputs.
5698,py_method,tensorflow/tensorflow/python/keras/engine/data_adapter.py,502,method,
5699,slice_array,tensorflow/tensorflow/python/keras/engine/data_adapter.py,503,method,
5700,CompositeTensorDataAdapter,tensorflow/tensorflow/python/keras/engine/data_adapter.py,519,class,Adapter that handles composite tensor.
5701,can_handle,tensorflow/tensorflow/python/keras/engine/data_adapter.py,523,method,
5702,get_dataset,tensorflow/tensorflow/python/keras/engine/data_adapter.py,588,method,
5703,get_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,591,method,
5704,batch_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,594,method,
5705,has_partial_batch,tensorflow/tensorflow/python/keras/engine/data_adapter.py,597,method,
5706,partial_batch_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,600,method,
5707,should_recreate_iterator,tensorflow/tensorflow/python/keras/engine/data_adapter.py,603,method,
5708,ListsOfScalarsDataAdapter,tensorflow/tensorflow/python/keras/engine/data_adapter.py,607,class,Adapter that handles lists of scalars and lists of lists of scalars.
5709,can_handle,tensorflow/tensorflow/python/keras/engine/data_adapter.py,611,method,
5710,get_dataset,tensorflow/tensorflow/python/keras/engine/data_adapter.py,652,method,
5711,get_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,655,method,
5712,batch_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,658,method,
5713,has_partial_batch,tensorflow/tensorflow/python/keras/engine/data_adapter.py,661,method,
5714,partial_batch_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,664,method,
5715,should_recreate_iterator,tensorflow/tensorflow/python/keras/engine/data_adapter.py,667,method,
5716,DatasetAdapter,tensorflow/tensorflow/python/keras/engine/data_adapter.py,671,class,Adapter that handles `tf.data.Dataset`.
5717,can_handle,tensorflow/tensorflow/python/keras/engine/data_adapter.py,675,method,
5718,get_dataset,tensorflow/tensorflow/python/keras/engine/data_adapter.py,695,method,
5719,get_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,698,method,
5720,batch_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,701,method,
5721,has_partial_batch,tensorflow/tensorflow/python/keras/engine/data_adapter.py,704,method,
5722,partial_batch_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,707,method,
5723,should_recreate_iterator,tensorflow/tensorflow/python/keras/engine/data_adapter.py,710,method,
5724,GeneratorDataAdapter,tensorflow/tensorflow/python/keras/engine/data_adapter.py,746,class,Adapter that handles python generators and iterators.
5725,can_handle,tensorflow/tensorflow/python/keras/engine/data_adapter.py,750,method,
5726,get_dataset,tensorflow/tensorflow/python/keras/engine/data_adapter.py,852,method,
5727,get_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,855,method,
5728,batch_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,858,method,
5729,representative_batch_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,861,method,
5730,has_partial_batch,tensorflow/tensorflow/python/keras/engine/data_adapter.py,864,method,
5731,partial_batch_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,867,method,
5732,should_recreate_iterator,tensorflow/tensorflow/python/keras/engine/data_adapter.py,870,method,
5733,wrapped_generator,tensorflow/tensorflow/python/keras/engine/data_adapter.py,806,method,
5734,generator_fn,tensorflow/tensorflow/python/keras/engine/data_adapter.py,843,method,
5735,KerasSequenceAdapter,tensorflow/tensorflow/python/keras/engine/data_adapter.py,874,class,Adapter that handles `keras.utils.Sequence`.
5736,can_handle,tensorflow/tensorflow/python/keras/engine/data_adapter.py,878,method,
5737,get_size,tensorflow/tensorflow/python/keras/engine/data_adapter.py,937,method,
5738,should_recreate_iterator,tensorflow/tensorflow/python/keras/engine/data_adapter.py,940,method,
5739,on_epoch_end,tensorflow/tensorflow/python/keras/engine/data_adapter.py,943,method,
5740,generator_fn,tensorflow/tensorflow/python/keras/engine/data_adapter.py,918,method,
5741,generator_fn,tensorflow/tensorflow/python/keras/engine/data_adapter.py,925,method,
5742,select_data_adapter,tensorflow/tensorflow/python/keras/engine/data_adapter.py,956,function,Selects a data adapter than can handle a given x and y.
5743,is_none_or_empty,tensorflow/tensorflow/python/keras/engine/data_adapter.py,1018,function,
5744,broadcast_sample_weight_modes,tensorflow/tensorflow/python/keras/engine/data_adapter.py,1027,function,Match sample_weight_modes structure with output structure.
5745,DataHandler,tensorflow/tensorflow/python/keras/engine/data_adapter.py,1063,class,Handles iterating over epoch-level `tf.data.Iterator` objects.
5746,enumerate_epochs,tensorflow/tensorflow/python/keras/engine/data_adapter.py,1128,method,"Yields `(epoch, tf.data.Iterator)`."
5747,catch_stop_iteration,tensorflow/tensorflow/python/keras/engine/data_adapter.py,1158,method,Catches errors when an iterator runs out of data.
5748,steps,tensorflow/tensorflow/python/keras/engine/data_adapter.py,1177,method,Yields steps for the current epoch.
5749,step_increment,tensorflow/tensorflow/python/keras/engine/data_adapter.py,1206,method,The number to increment the step for `on_batch_end` methods.
5750,inferred_steps,tensorflow/tensorflow/python/keras/engine/data_adapter.py,1211,method,"The inferred steps per epoch of the created `Dataset`.
This will be `None` in the case where:
(1) A `Dataset` of unknown cardinality was passed to the `DataHandler`, and
(2) `steps_per_epoch` was not provided, and
(3) The first epoch of iteration has not yet completed.
Returns:
The inferred steps per epoch of the created `Dataset`."
5751,should_sync,tensorflow/tensorflow/python/keras/engine/data_adapter.py,1226,method,
5752,expand_1d,tensorflow/tensorflow/python/keras/engine/data_adapter.py,1330,function,Expands 1-dimensional `Tensor`s into 2-dimensional `Tensor`s.
5753,train_validation_split,tensorflow/tensorflow/python/keras/engine/data_adapter.py,1343,function,"Split arrays into train and validation subsets in deterministic order.
The last part of data will become validation data.
Arguments:
arrays: Tensors to split. Allowed inputs are arbitrarily nested structures
of Tensors and NumPy arrays.
validation_split: Float between 0 and 1. The proportion of the dataset to
include in the validation split. The rest of the dataset will be included
in the training split.
Returns:
`(train_arrays, validation_arrays)`"
5754,unpack_x_y_sample_weight,tensorflow/tensorflow/python/keras/engine/data_adapter.py,1406,function,"Unpacks user-provided data tuple.
This is a convenience utility to be used when overriding
`Model.train_step`, `Model.test_step`, or `Model.predict_step`.
This utility makes it easy to support data of the form `(x,)`,
`(x, y)`, or `(x, y, sample_weight)`.
Standalone usage:
>>> features_batch = tf.ones((10, 5))
>>> labels_batch = tf.zeros((10, 5))
>>> data = (features_batch, labels_batch)
>>> # `y` and `sample_weight` will default to `None` if not provided.
>>> x, y, sample_weight = tf.keras.utils.unpack_x_y_sample_weight(data)
>>> sample_weight is None
True
Example in overridden `Model.train_step`:
```python
class MyModel(tf.keras.Model):
def train_step(self, data):
# If `sample_weight` is not provided, all samples will be weighted
# equally.
x, y, sample_weight = tf.keras.utils.unpack_x_y_sample_weight(data)
with tf.GradientTape() as tape:
y_pred = self(x, training=True)
loss = self.compiled_loss(
y, y_pred, sample_weight, regularization_losses=self.losses)
trainable_variables = self.trainable_variables
gradients = tape.gradient(loss, trainable_variables)
self.optimizer.apply_gradients(zip(gradients, trainable_variables))
self.compiled_metrics.update_state(y, y_pred, sample_weight)
return {m.name: m.result() for m in self.metrics}
```
Arguments:
data: A tuple of the form `(x,)`, `(x, y)`, or `(x, y, sample_weight)`.
Returns:
The unpacked tuple, with `None`s for `y` and `sample_weight` if they are not
provided."
5755,pack_x_y_sample_weight,tensorflow/tensorflow/python/keras/engine/data_adapter.py,1468,function,"Packs user-provided data into a tuple.
This is a convenience utility for packing data into the tuple formats
that `Model.fit` uses.
Standalone usage:
>>> x = tf.ones((10, 1))
>>> data = tf.keras.utils.pack_x_y_sample_weight(x)
>>> isinstance(data, tf.Tensor)
True
>>> y = tf.ones((10, 1))
>>> data = tf.keras.utils.pack_x_y_sample_weight(x, y)
>>> isinstance(data, tuple)
True
>>> x, y = data
Arguments:
x: Features to pass to `Model`.
y: Ground-truth targets to pass to `Model`.
sample_weight: Sample weight for each element.
Returns:
Tuple in the format used in `Model.fit`."
5756,single_batch_iterator,tensorflow/tensorflow/python/keras/engine/data_adapter.py,1509,function,Creates a single-batch dataset.
5757,DummyArrayLike,tensorflow/tensorflow/python/keras/engine/data_adapter_test.py,44,class,Dummy array-like object.
5758,shape,tensorflow/tensorflow/python/keras/engine/data_adapter_test.py,57,method,
5759,dtype,tensorflow/tensorflow/python/keras/engine/data_adapter_test.py,61,method,
5760,fail_on_convert,tensorflow/tensorflow/python/keras/engine/data_adapter_test.py,65,function,
5761,get_model,tensorflow/tensorflow/python/keras/engine/deferred_sequential_test.py,206,function,
5762,Functional,tensorflow/tensorflow/python/keras/engine/functional.py,52,class,"A `Functional` model is a `Model` defined as a directed graph of layers.
Three types of `Model` exist: subclassed `Model`, `Functional` model,
and `Sequential` (a special case of `Functional`).
In general, more Keras features are supported with `Functional`
than with subclassed `Model`s, specifically:
- Model cloning (`keras.models.clone`)
- Serialization (`model.get_config()/from_config`, `model.to_json()/to_yaml()`
- Whole-model saving (`model.save()`)
A `Functional` model can be instantiated by passing two arguments to
`__init__`. The first argument is the `keras.Input` Tensors that represent
the inputs to the model. The second argument specifies the output
tensors that represent the outputs of this model. Both arguments can be a
nested structure of tensors.
Example:
```
inputs = {'x1': keras.Input(shape=(10,)), 'x2': keras.Input(shape=(1,))}
t = keras.layers.Dense(1, activation='relu')(inputs['x1'])
outputs = keras.layers.Add()([t, inputs['x2'])
model = keras.Model(inputs, outputs)
```
A `Functional` model constructed using the Functional API can also include raw
TensorFlow functions, with the exception of functions that create Variables
or assign ops.
Example:
```
inputs = keras.Input(shape=(10,))
x = keras.layers.Dense(1)(inputs)
outputs = tf.nn.relu(x)
model = keras.Model(inputs, outputs)
```
Arguments:
inputs: List of input tensors (must be created via `tf.keras.Input()`).
outputs: List of outputs tensors.
name: String, optional. Name of the model.
trainable: Boolean, whether the model's variables should be trainable."
5763,input,tensorflow/tensorflow/python/keras/engine/functional.py,218,method,"Retrieves the input tensor(s) of a layer.
Only applicable if the layer has exactly one input,
i.e. if it is connected to one incoming layer.
Returns:
Input tensor or list of input tensors.
Raises:
RuntimeError: If called in Eager mode.
AttributeError: If no inbound nodes are found."
5764,input_shape,tensorflow/tensorflow/python/keras/engine/functional.py,234,method,"Retrieves the input shape(s) of a layer.
Only applicable if the layer has exactly one input,
i.e. if it is connected to one incoming layer, or if all inputs
have the same shape.
Returns:
Input shape, as an integer shape tuple
(or list of shape tuples, one tuple per input tensor).
Raises:
AttributeError: if the layer has no defined input_shape.
RuntimeError: if called in Eager mode."
5765,output,tensorflow/tensorflow/python/keras/engine/functional.py,252,method,"Retrieves the output tensor(s) of a layer.
Only applicable if the layer has exactly one output,
i.e. if it is connected to one incoming layer.
Returns:
Output tensor or list of output tensors.
Raises:
AttributeError: if the layer is connected to more than one incoming
layers.
RuntimeError: if called in Eager mode."
5766,output_shape,tensorflow/tensorflow/python/keras/engine/functional.py,269,method,"Retrieves the output shape(s) of a layer.
Only applicable if the layer has one output,
or if all outputs have the same shape.
Returns:
Output shape, as an integer shape tuple
(or list of shape tuples, one tuple per output tensor).
Raises:
AttributeError: if the layer has no defined output shape.
RuntimeError: if called in Eager mode."
5767,compute_mask,tensorflow/tensorflow/python/keras/engine/functional.py,355,method,
5768,call,tensorflow/tensorflow/python/keras/engine/functional.py,363,method,"Calls the model on new inputs.
In this case `call` just reapplies
all ops in the graph to the new inputs
(e.g. build a new computational graph from the provided inputs).
Arguments:
inputs: A tensor or list of tensors.
training: Boolean or boolean scalar tensor, indicating whether to run
the `Network` in training mode or inference mode.
mask: A mask or list of masks. A mask can be
either a tensor or None (no mask).
Returns:
A tensor if there is a single output, or
a list of tensors if there are more than one outputs."
5769,compute_output_shape,tensorflow/tensorflow/python/keras/engine/functional.py,384,method,
5770,get_config,tensorflow/tensorflow/python/keras/engine/functional.py,593,method,
5771,from_config,tensorflow/tensorflow/python/keras/engine/functional.py,597,method,"Instantiates a Model from its config (output of `get_config()`).
Arguments:
config: Model config dictionary.
custom_objects: Optional dictionary mapping names
(strings) to custom classes or functions to be
considered during deserialization.
Returns:
A model instance.
Raises:
ValueError: In case of improperly formatted config dict."
5772,connect_ancillary_layers,tensorflow/tensorflow/python/keras/engine/functional.py,1052,function,Adds layers that are not connected to the outputs to the model.
5773,reconstruct_from_config,tensorflow/tensorflow/python/keras/engine/functional.py,1068,function,"Reconstructs graph from config object.
Args:
config: Dictionary returned from Network.get_config()
custom_objects: Optional dictionary mapping names (strings) to custom
classes or functions to be considered during deserialization.
created_layers: Optional dictionary mapping names to Layer objects. Any
layer not in this dictionary will be be created and added to the dict.
This function will add new nodes to all layers (excluding InputLayers),
instead of re-using pre-existing nodes in the layers.
Returns:
Tuple of (input tensors, output tensors, dictionary of created layers)"
5774,get_network_config,tensorflow/tensorflow/python/keras/engine/functional.py,1242,function,"Builds the config, which consists of the node graph and serialized layers.
Args:
network: A Network object.
serialize_layer_fn: Function used to serialize layers.
Returns:
Config dictionary."
5775,AttrTrackingLayer,tensorflow/tensorflow/python/keras/engine/functional_test.py,2103,class,"Count how many times `dynamic` and `stateful` are called.
These counts are used to test that the attribute cache behaves as expected."
5776,stateful,tensorflow/tensorflow/python/keras/engine/functional_test.py,2114,method,
5777,dynamic,tensorflow/tensorflow/python/keras/engine/functional_test.py,2119,method,
5778,InputLayer,tensorflow/tensorflow/python/keras/engine/input_layer.py,37,class,"Layer to be used as an entry point into a Network (a graph of layers).
It can either wrap an existing tensor (pass an `input_tensor` argument)
or create a placeholder tensor (pass arguments `input_shape`, and
optionally, `dtype`).
It is generally recommend to use the functional layer API via `Input`,
(which creates an `InputLayer`) without directly using `InputLayer`.
When using InputLayer with Keras Sequential model, it can be skipped by
moving the input_shape parameter to the first layer after the InputLayer.
This class can create placeholders for tf.Tensors, tf.SparseTensors, and
tf.RaggedTensors by choosing 'sparse=True' or 'ragged=True'. Note that
'sparse' and 'ragged' can't be configured to True at same time.
Usage:
```python
# With explicit InputLayer.
model = tf.keras.Sequential([
tf.keras.layers.InputLayer(input_shape=(4,)),
tf.keras.layers.Dense(8)])
model.compile(tf.optimizers.RMSprop(0.001), loss='mse')
model.fit(np.zeros((10, 4)),
np.ones((10, 8)))
# Without InputLayer and let the first layer to have the input_shape.
# Keras will add a input for the model behind the scene.
model = tf.keras.Sequential([
tf.keras.layers.Dense(8, input_shape=(4,))])
model.compile(tf.optimizers.RMSprop(0.001), loss='mse')
model.fit(np.zeros((10, 4)),
np.ones((10, 8)))
```
Arguments:
input_shape: Shape tuple (not including the batch axis), or `TensorShape`
instance (not including the batch axis).
batch_size: Optional input batch size (integer or None).
dtype: Optional datatype of the input. When not provided, the Keras
default float type will be used.
input_tensor: Optional tensor to use as layer input
instead of creating a placeholder.
sparse: Boolean, whether the placeholder created is meant to be sparse.
Default to False.
ragged: Boolean, whether the placeholder created is meant to be ragged.
In this case, values of 'None' in the 'shape' argument represent
ragged dimensions. For more information about RaggedTensors, see
[this guide](https://www.tensorflow.org/guide/ragged_tensors).
Default to False.
name: Optional name of the layer (string)."
5779,get_config,tensorflow/tensorflow/python/keras/engine/input_layer.py,195,method,
5780,Input,tensorflow/tensorflow/python/keras/engine/input_layer.py,211,function,"`Input()` is used to instantiate a Keras tensor.
A Keras tensor is a TensorFlow symbolic tensor object,
which we augment with certain attributes that allow us to build a Keras model
just by knowing the inputs and outputs of the model.
For instance, if `a`, `b` and `c` are Keras tensors,
it becomes possible to do:
`model = Model(input=[a, b], output=c)`
Arguments:
shape: A shape tuple (integers), not including the batch size.
For instance, `shape=(32,)` indicates that the expected input
will be batches of 32-dimensional vectors. Elements of this tuple
can be None; 'None' elements represent dimensions where the shape is
not known.
batch_size: optional static batch size (integer).
name: An optional name string for the layer.
Should be unique in a model (do not reuse the same name twice).
It will be autogenerated if it isn't provided.
dtype: The data type expected by the input, as a string
(`float32`, `float64`, `int32`...)
sparse: A boolean specifying whether the placeholder to be created is
sparse. Only one of 'ragged' and 'sparse' can be True. Note that,
if `sparse` is False, sparse tensors can still be passed into the
input - they will be densified with a default value of 0.
tensor: Optional existing tensor to wrap into the `Input` layer.
If set, the layer will not create a placeholder tensor.
ragged: A boolean specifying whether the placeholder to be created is
ragged. Only one of 'ragged' and 'sparse' can be True. In this case,
values of 'None' in the 'shape' argument represent ragged dimensions.
For more information about RaggedTensors, see
[this guide](https://www.tensorflow.org/guide/ragged_tensors).
**kwargs: deprecated arguments support. Supports `batch_shape` and
`batch_input_shape`.
Returns:
A `tensor`.
Example:
```python
# this is a logistic regression in Keras
x = Input(shape=(32,))
y = Dense(16, activation='softmax')(x)
model = Model(x, y)
```
Note that even if eager execution is enabled,
`Input` produces a symbolic tensor (i.e. a placeholder).
This symbolic tensor can be used with other
TensorFlow ops, as such:
```python
x = Input(shape=(32,))
y = tf.square(x)
```
Raises:
ValueError: If both `sparse` and `ragged` are provided.
ValueError: If both `shape` and (`batch_input_shape` or `batch_shape`) are
provided.
ValueError: If both `shape` and `tensor` are None.
ValueError: if any unrecognized parameters are provided."
5781,InputSpec,tensorflow/tensorflow/python/keras/engine/input_spec.py,34,class,"Specifies the rank, dtype and shape of every input to a layer.
Layers can expose (if appropriate) an `input_spec` attribute:
an instance of `InputSpec`, or a nested structure of `InputSpec` instances
(one per input tensor). These objects enable the layer to run input
compatibility checks for input structure, input rank, input shape, and
input dtype.
A None entry in a shape is compatible with any dimension,
a None shape is compatible with any shape.
Arguments:
dtype: Expected DataType of the input.
shape: Shape tuple, expected shape of the input
(may include None for unchecked axes).
ndim: Integer, expected rank of the input.
max_ndim: Integer, maximum rank of the input.
min_ndim: Integer, minimum rank of the input.
axes: Dictionary mapping integer axes to
a specific dimension value."
5782,get_config,tensorflow/tensorflow/python/keras/engine/input_spec.py,95,method,
5783,from_config,tensorflow/tensorflow/python/keras/engine/input_spec.py,105,method,
5784,to_tensor_shape,tensorflow/tensorflow/python/keras/engine/input_spec.py,109,function,"Returns a tf.TensorShape object that matches the shape specifications.
If the InputSpec's shape or ndim is defined, this method will return a fully
or partially-known shape. Otherwise, the returned TensorShape is None.
Args:
spec: an InputSpec object.
Returns:
a tf.TensorShape object"
5785,assert_input_compatibility,tensorflow/tensorflow/python/keras/engine/input_spec.py,132,function,"Checks compatibility between the layer and provided inputs.
This checks that the tensor(s) `inputs` verify the input assumptions
of a layer (if any). If not, a clear and actional exception gets raised.
Arguments:
input_spec: An InputSpec instance, list of InputSpec instances, a nested
structure of InputSpec instances, or None.
inputs: Input tensor, list of input tensors, or a nested structure of
input tensors.
layer_name: String, name of the layer (for error message formatting).
Raises:
ValueError: in case of mismatch between
the provided inputs and the expectations of the layer."
5786,to_tensor_spec,tensorflow/tensorflow/python/keras/engine/input_spec.py,237,function,Converts a Keras InputSpec object to a TensorSpec.
5787,enable_keras_tensors,tensorflow/tensorflow/python/keras/engine/keras_tensor.py,36,function,Enable using KerasTensors in Keras's functional API.
5788,disable_keras_tensors,tensorflow/tensorflow/python/keras/engine/keras_tensor.py,42,function,Disable using KerasTensors in Keras's functional API.
5789,keras_tensors_enabled,tensorflow/tensorflow/python/keras/engine/keras_tensor.py,48,function,Return a bool specifying if KerasTensors are enabled.
5790,KerasTensor,tensorflow/tensorflow/python/keras/engine/keras_tensor.py,53,class,"A representation of a Keras in/output during Functional API construction.
`KerasTensor`s are tensor-like objects that represent the symbolic inputs
and outputs of Keras layers during Functional model construction. They are
comprised of the `tf.TypeSpec` of the (Composite)Tensor that will be
consumed/produced in the corresponding location of the Functional model.
KerasTensors are intended as a private API, so users should never need to
directly instantiate `KerasTensor`s.
**Building Functional Models with KerasTensors**
`tf.keras.Input` produces `KerasTensor`s that represent the symbolic inputs
to your model.
Passing a `KerasTensor` to a `tf.keras.Layer` `__call__` lets the layer know
that you are building a Functional model. The layer __call__ will
infer the output signature and return `KerasTensor`s with `tf.TypeSpec`s
corresponding to the symbolic outputs of that layer call. These output
`KerasTensor`s will have all of the internal KerasHistory metadata attached
to them that Keras needs to construct a Functional Model.
Currently, layers infer the output signature by:
* creating a scratch `FuncGraph`
* making placeholders in the scratch graph that match the input typespecs
* Calling `layer.call` on these placeholders
* extracting the signatures of the outputs before clearing the scratch graph
(Note: names assigned to KerasTensors by this process are not guaranteed to
be unique, and are subject to implementation details).
`tf.nest` methods are used to insure all of the inputs/output data
structures get maintained, with elements swapped between KerasTensors and
placeholders.
In rare cases (such as when directly manipulating shapes using Keras layers),
the layer may be able to partially infer the value of of the output in
addition to just inferring the signature.
When this happens, the returned KerasTensor will also contain the inferred
value information. Follow-on layers can use this information.
during their own output signature inference.
E.g. if one layer produces a symbolic `KerasTensor` that the next layer uses
as the shape of its outputs, partially knowing the value helps infer the
output shape.
**Automatically converting TF APIs to layers**:
If you passing a `KerasTensor` to a TF API that supports dispatching,
Keras will automatically turn that API call into a lambda
layer in the Functional model, and return KerasTensors representing the
symbolic outputs.
Most TF APIs that take only tensors as input and produce output tensors
will support dispatching.
Calling a `tf.function` does not support dispatching, so you cannot pass
`KerasTensor`s as inputs to a `tf.function`.
Higher-order apis that take methods which produce tensors (e.g. `tf.while`,
`tf.map_fn`, `tf.cond`) also do not currently support dispatching. So, you
cannot directly pass KerasTensors as inputs to these APIs either. If you
want to use these APIs inside of a Functional model, you must put them inside
of a custom layer.
Args:
type_spec: The `tf.TypeSpec` for the symbolic input created by
`tf.keras.Input`, or symbolically inferred for the output
during a symbolic layer `__call__`.
inferred_value: (Optional) a non-symbolic static value, possibly partially
specified, that could be symbolically inferred for the outputs during
a symbolic layer `__call__`. This will generally only happen when
grabbing and manipulating `tf.int32` shapes directly as tensors.
Statically inferring values in this way and storing them in the
KerasTensor allows follow-on layers to infer output signatures
more effectively. (e.g. when using a symbolic shape tensor to later
construct a tensor with that shape).
name: (optional) string name for this KerasTensor. Names automatically
generated by symbolic layer `__call__`s are not guaranteed to be unique,
and are subject to implementation details."
5791,type_spec,tensorflow/tensorflow/python/keras/engine/keras_tensor.py,143,method,Returns the `tf.TypeSpec` symbolically inferred for this Keras output.
5792,shape,tensorflow/tensorflow/python/keras/engine/keras_tensor.py,148,method,Returns the `TensorShape` symbolically inferred for this Keras output.
5793,get_shape,tensorflow/tensorflow/python/keras/engine/keras_tensor.py,156,method,
5794,op,tensorflow/tensorflow/python/keras/engine/keras_tensor.py,170,method,
5795,is_tensor_like,tensorflow/tensorflow/python/keras/engine/keras_tensor.py,204,method,
5796,set_shape,tensorflow/tensorflow/python/keras/engine/keras_tensor.py,207,method,Updates the shape of this KerasTensor. Mimics `tf.Tensor.set_shape()`.
5797,dtype,tensorflow/tensorflow/python/keras/engine/keras_tensor.py,269,method,Returns the `dtype` symbolically inferred for this Keras output.
5798,ref,tensorflow/tensorflow/python/keras/engine/keras_tensor.py,277,method,"Returns a hashable reference object to this KerasTensor.
The primary use case for this API is to put KerasTensors in a
set/dictionary. We can't put tensors in a set/dictionary as
`tensor.__hash__()` is not available and tensor equality (`==`) is supposed
to produce a tensor representing if the two inputs are equal.
See the documentation of `tf.Tensor.ref()` for more info."
5799,name,tensorflow/tensorflow/python/keras/engine/keras_tensor.py,304,method,"Returns the (non-unique, optional) name of this symbolic Keras value."
5800,keras_tensor_to_placeholder,tensorflow/tensorflow/python/keras/engine/keras_tensor.py,362,function,Construct a graph placeholder to represent a KerasTensor when tracing.
5801,UserRegisteredSpec,tensorflow/tensorflow/python/keras/engine/keras_tensor.py,413,class,TypeSpec to represent user-registered symbolic objects.
5802,value_type,tensorflow/tensorflow/python/keras/engine/keras_tensor.py,433,method,
5803,keras_tensor_from_tensor,tensorflow/tensorflow/python/keras/engine/keras_tensor.py,437,function,Convert a traced (composite)tensor to a representative KerasTensor.
5804,Node,tensorflow/tensorflow/python/keras/engine/node.py,38,class,"A `Node` describes the connectivity between two layers.
Each time a layer is connected to some new input,
a node is added to `layer._inbound_nodes`.
Each time the output of a layer is used by another layer,
a node is added to `layer._outbound_nodes`.
Arguments:
layer: The Layer for the Layer.__call__ this node represents.
call_args: The positional arguments the Layer was called with.
call_kwargs: The keyword arguments the Layer was called with.
outputs: The outputs of the Layer.__call__"
5805,keras_inputs,tensorflow/tensorflow/python/keras/engine/node.py,118,method,Tensors input to this node that can be traced back to a `keras.Input`.
5806,parent_nodes,tensorflow/tensorflow/python/keras/engine/node.py,123,method,Returns all the `Node`s whose output this node immediately depends on.
5807,iterate_inbound,tensorflow/tensorflow/python/keras/engine/node.py,133,method,"Yields tuples representing the data inbound from other nodes.
Yields:
tuples like: (inbound_layer, node_index, tensor_index, tensor)."
5808,map_arguments,tensorflow/tensorflow/python/keras/engine/node.py,146,method,Maps Keras Tensors to computed Tensors using `tensor_dict`.
5809,serialize,tensorflow/tensorflow/python/keras/engine/node.py,161,method,Serializes `Node` for Functional API's `get_config`.
5810,input_tensors,tensorflow/tensorflow/python/keras/engine/node.py,214,method,
5811,output_tensors,tensorflow/tensorflow/python/keras/engine/node.py,220,method,
5812,input_shapes,tensorflow/tensorflow/python/keras/engine/node.py,226,method,
5813,output_shapes,tensorflow/tensorflow/python/keras/engine/node.py,233,method,
5814,outbound_layer,tensorflow/tensorflow/python/keras/engine/node.py,237,method,
5815,inbound_layers,tensorflow/tensorflow/python/keras/engine/node.py,241,method,
5816,serialize_first_arg_tensor,tensorflow/tensorflow/python/keras/engine/node.py,185,method,
5817,KerasHistory,tensorflow/tensorflow/python/keras/engine/node.py,249,class,"Tracks the Layer call that created a Tensor, for Keras Graph Networks.
During construction of Keras Graph Networks, this metadata is added to
each Tensor produced as the output of a Layer, starting with an
`InputLayer`. This allows Keras to track how each Tensor was produced, and
this information is later retraced by the `keras.engine.Network` class to
reconstruct the Keras Graph Network.
Attributes:
layer: The Layer that produced the Tensor.
node_index: The specific call to the Layer that produced this Tensor. Layers
can be called multiple times in order to share weights. A new node is
created every time a Layer is called.
tensor_index: The output index for this Tensor. Always zero if the Layer
that produced this Tensor only has one output. Nested structures of
Tensors are deterministically assigned an index via `nest.flatten`."
5818,is_keras_tensor,tensorflow/tensorflow/python/keras/engine/node.py,274,function,
5819,DummyTensor,tensorflow/tensorflow/python/keras/engine/node_test.py,27,class,
5820,DummyLayer,tensorflow/tensorflow/python/keras/engine/node_test.py,33,class,
5821,PartialBatchPaddingHandler,tensorflow/tensorflow/python/keras/engine/partial_batch_padding_handler.py,30,class,A container that holds info about partial batches for `predict()`.
5822,get_real_batch_size,tensorflow/tensorflow/python/keras/engine/partial_batch_padding_handler.py,38,method,Returns the number of elements in a potentially partial batch.
5823,update_mask,tensorflow/tensorflow/python/keras/engine/partial_batch_padding_handler.py,56,method,Calculate and cache the amount of padding required for a batch.
5824,pad_batch,tensorflow/tensorflow/python/keras/engine/partial_batch_padding_handler.py,64,method,Pads out the batch dimension of a tensor to the complete batch size.
5825,apply_mask,tensorflow/tensorflow/python/keras/engine/partial_batch_padding_handler.py,89,method,Removes prediction output that corresponds to padded input.
5826,Sequential,tensorflow/tensorflow/python/keras/engine/sequential.py,50,class,"`Sequential` groups a linear stack of layers into a `tf.keras.Model`.
`Sequential` provides training and inference features on this model.
Examples:
>>> # Optionally, the first layer can receive an `input_shape` argument:
>>> model = tf.keras.Sequential()
>>> model.add(tf.keras.layers.Dense(8, input_shape=(16,)))
>>> # Afterwards, we do automatic shape inference:
>>> model.add(tf.keras.layers.Dense(4))
>>> # This is identical to the following:
>>> model = tf.keras.Sequential()
>>> model.add(tf.keras.Input(shape=(16,)))
>>> model.add(tf.keras.layers.Dense(8))
>>> # Note that you can also omit the `input_shape` argument.
>>> # In that case the model doesn't have any weights until the first call
>>> # to a training/evaluation method (since it isn't yet built):
>>> model = tf.keras.Sequential()
>>> model.add(tf.keras.layers.Dense(8))
>>> model.add(tf.keras.layers.Dense(4))
>>> # model.weights not created yet
>>> # Whereas if you specify the input shape, the model gets built
>>> # continuously as you are adding layers:
>>> model = tf.keras.Sequential()
>>> model.add(tf.keras.layers.Dense(8, input_shape=(16,)))
>>> model.add(tf.keras.layers.Dense(4))
>>> len(model.weights)
4
>>> # When using the delayed-build pattern (no input shape specified), you can
>>> # choose to manually build your model by calling
>>> # `build(batch_input_shape)`:
>>> model = tf.keras.Sequential()
>>> model.add(tf.keras.layers.Dense(8))
>>> model.add(tf.keras.layers.Dense(4))
>>> model.build((None, 16))
>>> len(model.weights)
4
```python
# Note that when using the delayed-build pattern (no input shape specified),
# the model gets built the first time you call `fit`, `eval`, or `predict`,
# or the first time you call the model on some input data.
model = tf.keras.Sequential()
model.add(tf.keras.layers.Dense(8))
model.add(tf.keras.layers.Dense(1))
model.compile(optimizer='sgd', loss='mse')
# This builds the model for the first time:
model.fit(x, y, batch_size=32, epochs=10)
```"
5827,layers,tensorflow/tensorflow/python/keras/engine/sequential.py,145,method,
5828,add,tensorflow/tensorflow/python/keras/engine/sequential.py,157,method,"Adds a layer instance on top of the layer stack.
Arguments:
layer: layer instance.
Raises:
TypeError: If `layer` is not a layer instance.
ValueError: In case the `layer` argument does not
know its input shape.
ValueError: In case the `layer` argument has
multiple output tensors, or is already connected
somewhere else (forbidden in `Sequential` models)."
5829,pop,tensorflow/tensorflow/python/keras/engine/sequential.py,237,method,"Removes the last layer in the model.
Raises:
TypeError: if there are no layers in the model."
5830,build,tensorflow/tensorflow/python/keras/engine/sequential.py,339,method,
5831,call,tensorflow/tensorflow/python/keras/engine/sequential.py,352,method,
5832,compute_output_shape,tensorflow/tensorflow/python/keras/engine/sequential.py,395,method,
5833,compute_mask,tensorflow/tensorflow/python/keras/engine/sequential.py,401,method,
5834,predict_proba,tensorflow/tensorflow/python/keras/engine/sequential.py,409,method,"Generates class probability predictions for the input samples.
The input samples are processed batch by batch.
Arguments:
x: input data, as a Numpy array or list of Numpy arrays
(if the model has multiple inputs).
batch_size: integer.
verbose: verbosity mode, 0 or 1.
Returns:
A Numpy array of probability predictions."
5835,predict_classes,tensorflow/tensorflow/python/keras/engine/sequential.py,439,method,"Generate class predictions for the input samples.
The input samples are processed batch by batch.
Arguments:
x: input data, as a Numpy array or list of Numpy arrays
(if the model has multiple inputs).
batch_size: integer.
verbose: verbosity mode, 0 or 1.
Returns:
A numpy array of class predictions."
5836,get_config,tensorflow/tensorflow/python/keras/engine/sequential.py,459,method,
5837,from_config,tensorflow/tensorflow/python/keras/engine/sequential.py,475,method,
5838,input_spec,tensorflow/tensorflow/python/keras/engine/sequential.py,495,method,
5839,relax_input_shape,tensorflow/tensorflow/python/keras/engine/sequential.py,527,function,
5840,clear_previously_created_nodes,tensorflow/tensorflow/python/keras/engine/sequential.py,535,function,Remove nodes from `created_nodes` from the layer's inbound_nodes.
5841,track_nodes_created_by_last_call,tensorflow/tensorflow/python/keras/engine/sequential.py,547,function,Adds to `created_nodes` the nodes created by the last call to `layer`.
5842,enable_multi_worker,tensorflow/tensorflow/python/keras/engine/training.py,103,function,Decorator that handles running `method` with multi-worker strategy.
5843,disable_multi_worker,tensorflow/tensorflow/python/keras/engine/training.py,123,function,Decorator that disallows multi-worker use of `method`.
5844,inject_functional_model_class,tensorflow/tensorflow/python/keras/engine/training.py,136,function,Inject `Functional` into the hierarchy of this class if needed.
5845,is_functional_model_init_params,tensorflow/tensorflow/python/keras/engine/training.py,152,function,
5846,Model,tensorflow/tensorflow/python/keras/engine/training.py,159,class,"`Model` groups layers into an object with training and inference features.
Arguments:
inputs: The input(s) of the model: a `keras.Input` object or list of
`keras.Input` objects.
outputs: The output(s) of the model. See Functional API example below.
name: String, the name of the model.
There are two ways to instantiate a `Model`:
1 - With the ""Functional API"", where you start from `Input`,
you chain layer calls to specify the model's forward pass,
and finally you create your model from inputs and outputs:
```python
import tensorflow as tf
inputs = tf.keras.Input(shape=(3,))
x = tf.keras.layers.Dense(4, activation=tf.nn.relu)(inputs)
outputs = tf.keras.layers.Dense(5, activation=tf.nn.softmax)(x)
model = tf.keras.Model(inputs=inputs, outputs=outputs)
```
2 - By subclassing the `Model` class: in that case, you should define your
layers in `__init__` and you should implement the model's forward pass
in `call`.
```python
import tensorflow as tf
class MyModel(tf.keras.Model):
def __init__(self):
super(MyModel, self).__init__()
self.dense1 = tf.keras.layers.Dense(4, activation=tf.nn.relu)
self.dense2 = tf.keras.layers.Dense(5, activation=tf.nn.softmax)
def call(self, inputs):
x = self.dense1(inputs)
return self.dense2(x)
model = MyModel()
```
If you subclass `Model`, you can optionally have
a `training` argument (boolean) in `call`, which you can use to specify
a different behavior in training and inference:
```python
import tensorflow as tf
class MyModel(tf.keras.Model):
def __init__(self):
super(MyModel, self).__init__()
self.dense1 = tf.keras.layers.Dense(4, activation=tf.nn.relu)
self.dense2 = tf.keras.layers.Dense(5, activation=tf.nn.softmax)
self.dropout = tf.keras.layers.Dropout(0.5)
def call(self, inputs, training=False):
x = self.dense1(inputs)
if training:
x = self.dropout(x, training=training)
return self.dense2(x)
model = MyModel()
```
Once the model is created, you can config the model with losses and metrics
with `model.compile()`, train the model with `model.fit()`, or use the model
to do prediction with `model.predict()`."
5847,build,tensorflow/tensorflow/python/keras/engine/training.py,345,method,"Builds the model based on input shapes received.
This is to be used for subclassed models, which do not know at instantiation
time what their inputs look like.
This method only exists for users who want to call `model.build()` in a
standalone way (as a substitute for calling the model on real data to
build it). It will never be called by the framework (and thus it will
never throw unexpected errors in an unrelated workflow).
Args:
input_shape: Single tuple, TensorShape, or list of shapes, where shapes
are tuples, integers, or TensorShapes.
Raises:
ValueError:
1. In case of invalid user-provided data (not of type tuple,
list, or TensorShape).
2. If the model requires call arguments that are agnostic
to the input shapes (positional or kwarg in call signature).
3. If not all layers were properly built.
4. If float type inputs are not supported within the layers.
In each of these cases, the user should build their model by calling it
on real tensor data."
5848,call,tensorflow/tensorflow/python/keras/engine/training.py,443,method,"Calls the model on new inputs.
In this case `call` just reapplies
all ops in the graph to the new inputs
(e.g. build a new computational graph from the provided inputs).
Arguments:
inputs: A tensor or list of tensors.
training: Boolean or boolean scalar tensor, indicating whether to run
the `Network` in training mode or inference mode.
mask: A mask or list of masks. A mask can be
either a tensor or None (no mask).
Returns:
A tensor if there is a single output, or
a list of tensors if there are more than one outputs."
5849,compile,tensorflow/tensorflow/python/keras/engine/training.py,464,method,"Configures the model for training.
Arguments:
optimizer: String (name of optimizer) or optimizer instance. See
`tf.keras.optimizers`.
loss: String (name of objective function), objective function or
`tf.keras.losses.Loss` instance. See `tf.keras.losses`. An objective
function is any callable with the signature `loss = fn(y_true,
y_pred)`, where y_true = ground truth values with shape =
`[batch_size, d0, .. dN]`, except sparse loss functions such as sparse
categorical crossentropy where shape = `[batch_size, d0, .. dN-1]`.
y_pred = predicted values with shape = `[batch_size, d0, .. dN]`. It
returns a weighted loss float tensor. If a custom `Loss` instance is
used and reduction is set to NONE, return value has the shape
[batch_size, d0, .. dN-1] ie. per-sample or per-timestep loss values;
otherwise, it is a scalar. If the model has multiple outputs, you can
use a different loss on each output by passing a dictionary or a list
of losses. The loss value that will be minimized by the model will
then be the sum of all individual losses.
metrics: List of metrics to be evaluated by the model during training
and testing. Each of this can be a string (name of a built-in
function), function or a `tf.keras.metrics.Metric` instance. See
`tf.keras.metrics`. Typically you will use `metrics=['accuracy']`. A
function is any callable with the signature `result = fn(y_true,
y_pred)`. To specify different metrics for different outputs of a
multi-output model, you could also pass a dictionary, such as
`metrics={'output_a': 'accuracy', 'output_b': ['accuracy', 'mse']}`.
You can also pass a list (len = len(outputs)) of lists of metrics
such as `metrics=[['accuracy'], ['accuracy', 'mse']]` or
`metrics=['accuracy', ['accuracy', 'mse']]`. When you pass the
strings 'accuracy' or 'acc', we convert this to one of
`tf.keras.metrics.BinaryAccuracy`,
`tf.keras.metrics.CategoricalAccuracy`,
`tf.keras.metrics.SparseCategoricalAccuracy` based on the loss
function used and the model output shape. We do a similar
conversion for the strings 'crossentropy' and 'ce' as well.
loss_weights: Optional list or dictionary specifying scalar coefficients
(Python floats) to weight the loss contributions of different model
outputs. The loss value that will be minimized by the model will then
be the *weighted sum* of all individual losses, weighted by the
`loss_weights` coefficients.
If a list, it is expected to have a 1:1 mapping to the model's
outputs. If a dict, it is expected to map output names (strings)
to scalar coefficients.
weighted_metrics: List of metrics to be evaluated and weighted by
sample_weight or class_weight during training and testing.
run_eagerly: Bool. Defaults to `False`. If `True`, this `Model`'s
logic will not be wrapped in a `tf.function`. Recommended to leave
this as `None` unless your `Model` cannot be run inside a
`tf.function`.
**kwargs: Any additional arguments. Supported arguments:
- `experimental_steps_per_execution`: Int. The number of batches to
run during each `tf.function` call. Running multiple batches
inside a single `tf.function` call can greatly improve performance
on TPUs or small models with a large Python overhead. Note that if
this value is set to `N`, `Callback.on_batch` methods will only be
called every `N` batches. This currently defaults to `1`. At most,
one full epoch will be run each execution. If a number larger than
the size of the epoch is passed, the execution will be truncated
to the size of the epoch.
- `sample_weight_mode` for backward compatibility.
Raises:
ValueError: In case of invalid arguments for
`optimizer`, `loss` or `metrics`."
5850,metrics,tensorflow/tensorflow/python/keras/engine/training.py,592,method,"Returns the model's metrics added using `compile`, `add_metric` APIs.
Note: Metrics passed to `compile()` are available only after a `keras.Model`
has been trained/evaluated on actual data.
Examples:
>>> inputs = tf.keras.layers.Input(shape=(3,))
>>> outputs = tf.keras.layers.Dense(2)(inputs)
>>> model = tf.keras.models.Model(inputs=inputs, outputs=outputs)
>>> model.compile(optimizer=""Adam"", loss=""mse"", metrics=[""mae""])
>>> [m.name for m in model.metrics]
[]
>>> x = np.random.random((2, 3))
>>> y = np.random.randint(0, 2, (2, 2))
>>> model.fit(x, y)
>>> [m.name for m in model.metrics]
['loss', 'mae']
>>> inputs = tf.keras.layers.Input(shape=(3,))
>>> d = tf.keras.layers.Dense(2, name='out')
>>> output_1 = d(inputs)
>>> output_2 = d(inputs)
>>> model = tf.keras.models.Model(
... inputs=inputs, outputs=[output_1, output_2])
>>> model.add_metric(
... tf.reduce_sum(output_2), name='mean', aggregation='mean')
>>> model.compile(optimizer=""Adam"", loss=""mse"", metrics=[""mae"", ""acc""])
>>> model.fit(x, (y, y))
>>> [m.name for m in model.metrics]
['loss', 'out_loss', 'out_1_loss', 'out_mae', 'out_acc', 'out_1_mae',
'out_1_acc', 'mean']"
5851,metrics_names,tensorflow/tensorflow/python/keras/engine/training.py,642,method,"Returns the model's display labels for all outputs.
Note: `metrics_names` are available only after a `keras.Model` has been
trained/evaluated on actual data.
Examples:
>>> inputs = tf.keras.layers.Input(shape=(3,))
>>> outputs = tf.keras.layers.Dense(2)(inputs)
>>> model = tf.keras.models.Model(inputs=inputs, outputs=outputs)
>>> model.compile(optimizer=""Adam"", loss=""mse"", metrics=[""mae""])
>>> model.metrics_names
[]
>>> x = np.random.random((2, 3))
>>> y = np.random.randint(0, 2, (2, 2))
>>> model.fit(x, y)
>>> model.metrics_names
['loss', 'mae']
>>> inputs = tf.keras.layers.Input(shape=(3,))
>>> d = tf.keras.layers.Dense(2, name='out')
>>> output_1 = d(inputs)
>>> output_2 = d(inputs)
>>> model = tf.keras.models.Model(
... inputs=inputs, outputs=[output_1, output_2])
>>> model.compile(optimizer=""Adam"", loss=""mse"", metrics=[""mae"", ""acc""])
>>> model.fit(x, (y, y))
>>> model.metrics_names
['loss', 'out_loss', 'out_1_loss', 'out_mae', 'out_acc', 'out_1_mae',
'out_1_acc']"
5852,distribute_strategy,tensorflow/tensorflow/python/keras/engine/training.py,682,method,The `tf.distribute.Strategy` this model was created under.
5853,run_eagerly,tensorflow/tensorflow/python/keras/engine/training.py,687,method,"Settable attribute indicating whether the model should run eagerly.
Running eagerly means that your model will be run step by step,
like Python code. Your model might run slower, but it should become easier
for you to debug it by stepping into individual layer calls.
By default, we will attempt to compile your model to a static graph to
deliver the best execution performance.
Returns:
Boolean, whether the model should run eagerly."
5854,run_eagerly,tensorflow/tensorflow/python/keras/engine/training.py,716,method,
5855,train_step,tensorflow/tensorflow/python/keras/engine/training.py,719,method,"The logic for one training step.
This method can be overridden to support custom training logic.
This method is called by `Model.make_train_function`.
This method should contain the mathemetical logic for one step of training.
This typically includes the forward pass, loss calculation, backpropagation,
and metric updates.
Configuration details for *how* this logic is run (e.g. `tf.function` and
`tf.distribute.Strategy` settings), should be left to
`Model.make_train_function`, which can also be overridden.
Arguments:
data: A nested structure of `Tensor`s.
Returns:
A `dict` containing values that will be passed to
`tf.keras.callbacks.CallbackList.on_train_batch_end`. Typically, the
values of the `Model`'s metrics are returned. Example:
`{'loss': 0.2, 'accuracy': 0.7}`."
5856,make_train_function,tensorflow/tensorflow/python/keras/engine/training.py,765,method,"Creates a function that executes one step of training.
This method can be overridden to support custom training logic.
This method is called by `Model.fit` and `Model.train_on_batch`.
Typically, this method directly controls `tf.function` and
`tf.distribute.Strategy` settings, and delegates the actual training
logic to `Model.train_step`.
This function is cached the first time `Model.fit` or
`Model.train_on_batch` is called. The cache is cleared whenever
`Model.compile` is called.
Returns:
Function. The function created by this method should accept a
`tf.data.Iterator`, and return a `dict` containing values that will
be passed to `tf.keras.Callbacks.on_train_batch_end`, such as
`{'loss': 0.2, 'accuracy': 0.7}`."
5857,fit,tensorflow/tensorflow/python/keras/engine/training.py,828,method,"Trains the model for a fixed number of epochs (iterations on a dataset).
Arguments:
x: Input data. It could be:
- A Numpy array (or array-like), or a list of arrays
(in case the model has multiple inputs).
- A TensorFlow tensor, or a list of tensors
(in case the model has multiple inputs).
- A dict mapping input names to the corresponding array/tensors,
if the model has named inputs.
- A `tf.data` dataset. Should return a tuple
of either `(inputs, targets)` or
`(inputs, targets, sample_weights)`.
- A generator or `keras.utils.Sequence` returning `(inputs, targets)`
or `(inputs, targets, sample_weights)`.
A more detailed description of unpacking behavior for iterator types
(Dataset, generator, Sequence) is given below.
y: Target data. Like the input data `x`,
it could be either Numpy array(s) or TensorFlow tensor(s).
It should be consistent with `x` (you cannot have Numpy inputs and
tensor targets, or inversely). If `x` is a dataset, generator,
or `keras.utils.Sequence` instance, `y` should
not be specified (since targets will be obtained from `x`).
batch_size: Integer or `None`.
Number of samples per gradient update.
If unspecified, `batch_size` will default to 32.
Do not specify the `batch_size` if your data is in the
form of datasets, generators, or `keras.utils.Sequence` instances
(since they generate batches).
epochs: Integer. Number of epochs to train the model.
An epoch is an iteration over the entire `x` and `y`
data provided.
Note that in conjunction with `initial_epoch`,
`epochs` is to be understood as ""final epoch"".
The model is not trained for a number of iterations
given by `epochs`, but merely until the epoch
of index `epochs` is reached.
verbose: 0, 1, or 2. Verbosity mode.
0 = silent, 1 = progress bar, 2 = one line per epoch.
Note that the progress bar is not particularly useful when
logged to a file, so verbose=2 is recommended when not running
interactively (eg, in a production environment).
callbacks: List of `keras.callbacks.Callback` instances.
List of callbacks to apply during training.
See `tf.keras.callbacks`.
validation_split: Float between 0 and 1.
Fraction of the training data to be used as validation data.
The model will set apart this fraction of the training data,
will not train on it, and will evaluate
the loss and any model metrics
on this data at the end of each epoch.
The validation data is selected from the last samples
in the `x` and `y` data provided, before shuffling. This argument is
not supported when `x` is a dataset, generator or
`keras.utils.Sequence` instance.
validation_data: Data on which to evaluate
the loss and any model metrics at the end of each epoch.
The model will not be trained on this data. Thus, note the fact
that the validation loss of data provided using `validation_split`
or `validation_data` is not affected by regularization layers like
noise and dropuout.
`validation_data` will override `validation_split`.
`validation_data` could be:
- tuple `(x_val, y_val)` of Numpy arrays or tensors
- tuple `(x_val, y_val, val_sample_weights)` of Numpy arrays
- dataset
For the first two cases, `batch_size` must be provided.
For the last case, `validation_steps` could be provided.
Note that `validation_data` does not support all the data types that
are supported in `x`, eg, dict, generator or `keras.utils.Sequence`.
shuffle: Boolean (whether to shuffle the training data
before each epoch) or str (for 'batch'). This argument is ignored
when `x` is a generator. 'batch' is a special option for dealing
with the limitations of HDF5 data; it shuffles in batch-sized
chunks. Has no effect when `steps_per_epoch` is not `None`.
class_weight: Optional dictionary mapping class indices (integers)
to a weight (float) value, used for weighting the loss function
(during training only).
This can be useful to tell the model to
""pay more attention"" to samples from
an under-represented class.
sample_weight: Optional Numpy array of weights for
the training samples, used for weighting the loss function
(during training only). You can either pass a flat (1D)
Numpy array with the same length as the input samples
(1:1 mapping between weights and samples),
or in the case of temporal data,
you can pass a 2D array with shape
`(samples, sequence_length)`,
to apply a different weight to every timestep of every sample. This
argument is not supported when `x` is a dataset, generator, or
`keras.utils.Sequence` instance, instead provide the sample_weights
as the third element of `x`.
initial_epoch: Integer.
Epoch at which to start training
(useful for resuming a previous training run).
steps_per_epoch: Integer or `None`.
Total number of steps (batches of samples)
before declaring one epoch finished and starting the
next epoch. When training with input tensors such as
TensorFlow data tensors, the default `None` is equal to
the number of samples in your dataset divided by
the batch size, or 1 if that cannot be determined. If x is a
`tf.data` dataset, and 'steps_per_epoch'
is None, the epoch will run until the input dataset is exhausted.
When passing an infinitely repeating dataset, you must specify the
`steps_per_epoch` argument. This argument is not supported with
array inputs.
validation_steps: Only relevant if `validation_data` is provided and
is a `tf.data` dataset. Total number of steps (batches of
samples) to draw before stopping when performing validation
at the end of every epoch. If 'validation_steps' is None, validation
will run until the `validation_data` dataset is exhausted. In the
case of an infinitely repeated dataset, it will run into an
infinite loop. If 'validation_steps' is specified and only part of
the dataset will be consumed, the evaluation will start from the
beginning of the dataset at each epoch. This ensures that the same
validation samples are used every time.
validation_batch_size: Integer or `None`.
Number of samples per validation batch.
If unspecified, will default to `batch_size`.
Do not specify the `validation_batch_size` if your data is in the
form of datasets, generators, or `keras.utils.Sequence` instances
(since they generate batches).
validation_freq: Only relevant if validation data is provided. Integer
or `collections_abc.Container` instance (e.g. list, tuple, etc.).
If an integer, specifies how many training epochs to run before a
new validation run is performed, e.g. `validation_freq=2` runs
validation every 2 epochs. If a Container, specifies the epochs on
which to run validation, e.g. `validation_freq=[1, 2, 10]` runs
validation at the end of the 1st, 2nd, and 10th epochs.
max_queue_size: Integer. Used for generator or `keras.utils.Sequence`
input only. Maximum size for the generator queue.
If unspecified, `max_queue_size` will default to 10.
workers: Integer. Used for generator or `keras.utils.Sequence` input
only. Maximum number of processes to spin up
when using process-based threading. If unspecified, `workers`
will default to 1. If 0, will execute the generator on the main
thread.
use_multiprocessing: Boolean. Used for generator or
`keras.utils.Sequence` input only. If `True`, use process-based
threading. If unspecified, `use_multiprocessing` will default to
`False`. Note that because this implementation relies on
multiprocessing, you should not pass non-picklable arguments to
the generator as they can't be passed easily to children processes.
Unpacking behavior for iterator-like inputs:
A common pattern is to pass a tf.data.Dataset, generator, or
tf.keras.utils.Sequence to the `x` argument of fit, which will in fact
yield not only features (x) but optionally targets (y) and sample weights.
Keras requires that the output of such iterator-likes be unambiguous. The
iterator should return a tuple of length 1, 2, or 3, where the optional
second and third elements will be used for y and sample_weight
respectively. Any other type provided will be wrapped in a length one
tuple, effectively treating everything as 'x'. When yielding dicts, they
should still adhere to the top-level tuple structure.
e.g. `({""x0"": x0, ""x1"": x1}, y)`. Keras will not attempt to separate
features, targets, and weights from the keys of a single dict.
A notable unsupported data type is the namedtuple. The reason is that
it behaves like both an ordered datatype (tuple) and a mapping
datatype (dict). So given a namedtuple of the form:
`namedtuple(""example_tuple"", [""y"", ""x""])`
it is ambiguous whether to reverse the order of the elements when
interpreting the value. Even worse is a tuple of the form:
`namedtuple(""other_tuple"", [""x"", ""y"", ""z""])`
where it is unclear if the tuple was intended to be unpacked into x, y,
and sample_weight or passed through as a single element to `x`. As a
result the data processing code will simply raise a ValueError if it
encounters a namedtuple. (Along with instructions to remedy the issue.)
Returns:
A `History` object. Its `History.history` attribute is
a record of training loss values and metrics values
at successive epochs, as well as validation loss values
and validation metrics values (if applicable).
Raises:
RuntimeError: 1. If the model was never compiled or,
2. If `model.fit` is wrapped in `tf.function`.
ValueError: In case of mismatch between the provided input data
and what the model expects or when the input data is empty."
5858,evaluate,tensorflow/tensorflow/python/keras/engine/training.py,1250,method,"Returns the loss value & metrics values for the model in test mode.
Computation is done in batches (see the `batch_size` arg.)
Arguments:
x: Input data. It could be:
- A Numpy array (or array-like), or a list of arrays
(in case the model has multiple inputs).
- A TensorFlow tensor, or a list of tensors
(in case the model has multiple inputs).
- A dict mapping input names to the corresponding array/tensors,
if the model has named inputs.
- A `tf.data` dataset. Should return a tuple
of either `(inputs, targets)` or
`(inputs, targets, sample_weights)`.
- A generator or `keras.utils.Sequence` returning `(inputs, targets)`
or `(inputs, targets, sample_weights)`.
A more detailed description of unpacking behavior for iterator types
(Dataset, generator, Sequence) is given in the `Unpacking behavior
for iterator-like inputs` section of `Model.fit`.
y: Target data. Like the input data `x`, it could be either Numpy
array(s) or TensorFlow tensor(s). It should be consistent with `x`
(you cannot have Numpy inputs and tensor targets, or inversely). If
`x` is a dataset, generator or `keras.utils.Sequence` instance, `y`
should not be specified (since targets will be obtained from the
iterator/dataset).
batch_size: Integer or `None`. Number of samples per batch of
computation. If unspecified, `batch_size` will default to 32. Do not
specify the `batch_size` if your data is in the form of a dataset,
generators, or `keras.utils.Sequence` instances (since they generate
batches).
verbose: 0 or 1. Verbosity mode. 0 = silent, 1 = progress bar.
sample_weight: Optional Numpy array of weights for the test samples,
used for weighting the loss function. You can either pass a flat (1D)
Numpy array with the same length as the input samples
(1:1 mapping between weights and samples), or in the case of
temporal data, you can pass a 2D array with shape `(samples,
sequence_length)`, to apply a different weight to every timestep
of every sample. This argument is not supported when `x` is a
dataset, instead pass sample weights as the third element of `x`.
steps: Integer or `None`. Total number of steps (batches of samples)
before declaring the evaluation round finished. Ignored with the
default value of `None`. If x is a `tf.data` dataset and `steps` is
None, 'evaluate' will run until the dataset is exhausted. This
argument is not supported with array inputs.
callbacks: List of `keras.callbacks.Callback` instances. List of
callbacks to apply during evaluation. See
[callbacks](/api_docs/python/tf/keras/callbacks).
max_queue_size: Integer. Used for generator or `keras.utils.Sequence`
input only. Maximum size for the generator queue. If unspecified,
`max_queue_size` will default to 10.
workers: Integer. Used for generator or `keras.utils.Sequence` input
only. Maximum number of processes to spin up when using process-based
threading. If unspecified, `workers` will default to 1. If 0, will
execute the generator on the main thread.
use_multiprocessing: Boolean. Used for generator or
`keras.utils.Sequence` input only. If `True`, use process-based
threading. If unspecified, `use_multiprocessing` will default to
`False`. Note that because this implementation relies on
multiprocessing, you should not pass non-picklable arguments to the
generator as they can't be passed easily to children processes.
return_dict: If `True`, loss and metric results are returned as a dict,
with each key being the name of the metric. If `False`, they are
returned as a list.
See the discussion of `Unpacking behavior for iterator-like inputs` for
`Model.fit`.
Returns:
Scalar test loss (if the model has a single output and no metrics)
or list of scalars (if the model has multiple outputs
and/or metrics). The attribute `model.metrics_names` will give you
the display labels for the scalar outputs.
Raises:
RuntimeError: If `model.evaluate` is wrapped in `tf.function`.
ValueError: in case of invalid arguments."
5859,predict_step,tensorflow/tensorflow/python/keras/engine/training.py,1403,method,"The logic for one inference step.
This method can be overridden to support custom inference logic.
This method is called by `Model.make_predict_function`.
This method should contain the mathemetical logic for one step of inference.
This typically includes the forward pass.
Configuration details for *how* this logic is run (e.g. `tf.function` and
`tf.distribute.Strategy` settings), should be left to
`Model.make_predict_function`, which can also be overridden.
Arguments:
data: A nested structure of `Tensor`s.
Returns:
The result of one inference step, typically the output of calling the
`Model` on data."
5860,make_predict_function,tensorflow/tensorflow/python/keras/engine/training.py,1427,method,"Creates a function that executes one step of inference.
This method can be overridden to support custom inference logic.
This method is called by `Model.predict` and `Model.predict_on_batch`.
Typically, this method directly controls `tf.function` and
`tf.distribute.Strategy` settings, and delegates the actual evaluation
logic to `Model.predict_step`.
This function is cached the first time `Model.predict` or
`Model.predict_on_batch` is called. The cache is cleared whenever
`Model.compile` is called.
Returns:
Function. The function created by this method should accept a
`tf.data.Iterator`, and return the outputs of the `Model`."
5861,predict,tensorflow/tensorflow/python/keras/engine/training.py,1494,method,"Generates output predictions for the input samples.
Computation is done in batches. This method is designed for performance in
large scale inputs. For small amount of inputs that fit in one batch,
directly using `__call__` is recommended for faster execution, e.g.,
`model(x)`, or `model(x, training=False)` if you have layers such as
`tf.keras.layers.BatchNormalization` that behaves differently during
inference. Also, note the fact that test loss is not affected by
regularization layers like noise and dropout.
Arguments:
x: Input samples. It could be:
- A Numpy array (or array-like), or a list of arrays
(in case the model has multiple inputs).
- A TensorFlow tensor, or a list of tensors
(in case the model has multiple inputs).
- A `tf.data` dataset.
- A generator or `keras.utils.Sequence` instance.
A more detailed description of unpacking behavior for iterator types
(Dataset, generator, Sequence) is given in the `Unpacking behavior
for iterator-like inputs` section of `Model.fit`.
batch_size: Integer or `None`.
Number of samples per batch.
If unspecified, `batch_size` will default to 32.
Do not specify the `batch_size` if your data is in the
form of dataset, generators, or `keras.utils.Sequence` instances
(since they generate batches).
verbose: Verbosity mode, 0 or 1.
steps: Total number of steps (batches of samples)
before declaring the prediction round finished.
Ignored with the default value of `None`. If x is a `tf.data`
dataset and `steps` is None, `predict` will
run until the input dataset is exhausted.
callbacks: List of `keras.callbacks.Callback` instances.
List of callbacks to apply during prediction.
See [callbacks](/api_docs/python/tf/keras/callbacks).
max_queue_size: Integer. Used for generator or `keras.utils.Sequence`
input only. Maximum size for the generator queue.
If unspecified, `max_queue_size` will default to 10.
workers: Integer. Used for generator or `keras.utils.Sequence` input
only. Maximum number of processes to spin up when using
process-based threading. If unspecified, `workers` will default
to 1. If 0, will execute the generator on the main thread.
use_multiprocessing: Boolean. Used for generator or
`keras.utils.Sequence` input only. If `True`, use process-based
threading. If unspecified, `use_multiprocessing` will default to
`False`. Note that because this implementation relies on
multiprocessing, you should not pass non-picklable arguments to
the generator as they can't be passed easily to children processes.
See the discussion of `Unpacking behavior for iterator-like inputs` for
`Model.fit`. Note that Model.predict uses the same interpretation rules as
`Model.fit` and `Model.evaluate`, so inputs must be unambiguous for all
three methods.
Returns:
Numpy array(s) of predictions.
Raises:
RuntimeError: If `model.predict` is wrapped in `tf.function`.
ValueError: In case of mismatch between the provided
input data and the model's expectations,
or in case a stateful model receives a number of samples
that is not a multiple of the batch size."
5862,reset_metrics,tensorflow/tensorflow/python/keras/engine/training.py,1627,method,"Resets the state of all the metrics in the model.
Examples:
>>> inputs = tf.keras.layers.Input(shape=(3,))
>>> outputs = tf.keras.layers.Dense(2)(inputs)
>>> model = tf.keras.models.Model(inputs=inputs, outputs=outputs)
>>> model.compile(optimizer=""Adam"", loss=""mse"", metrics=[""mae""])
>>> x = np.random.random((2, 3))
>>> y = np.random.randint(0, 2, (2, 2))
>>> _ = model.fit(x, y, verbose=0)
>>> assert all(float(m.result()) for m in model.metrics)
>>> model.reset_metrics()
>>> assert all(float(m.result()) == 0 for m in model.metrics)"
5863,train_on_batch,tensorflow/tensorflow/python/keras/engine/training.py,1649,method,"Runs a single gradient update on a single batch of data.
Arguments:
x: Input data. It could be:
- A Numpy array (or array-like), or a list of arrays
(in case the model has multiple inputs).
- A TensorFlow tensor, or a list of tensors
(in case the model has multiple inputs).
- A dict mapping input names to the corresponding array/tensors,
if the model has named inputs.
y: Target data. Like the input data `x`, it could be either Numpy
array(s) or TensorFlow tensor(s). It should be consistent with `x`
(you cannot have Numpy inputs and tensor targets, or inversely).
sample_weight: Optional array of the same length as x, containing
weights to apply to the model's loss for each sample. In the case of
temporal data, you can pass a 2D array with shape (samples,
sequence_length), to apply a different weight to every timestep of
every sample.
class_weight: Optional dictionary mapping class indices (integers) to a
weight (float) to apply to the model's loss for the samples from this
class during training. This can be useful to tell the model to ""pay
more attention"" to samples from an under-represented class.
reset_metrics: If `True`, the metrics returned will be only for this
batch. If `False`, the metrics will be statefully accumulated across
batches.
return_dict: If `True`, loss and metric results are returned as a dict,
with each key being the name of the metric. If `False`, they are
returned as a list.
Returns:
Scalar training loss
(if the model has a single output and no metrics)
or list of scalars (if the model has multiple outputs
and/or metrics). The attribute `model.metrics_names` will give you
the display labels for the scalar outputs.
Raises:
RuntimeError: If `model.train_on_batch` is wrapped in `tf.function`.
ValueError: In case of invalid user-provided arguments."
5864,predict_on_batch,tensorflow/tensorflow/python/keras/engine/training.py,1777,method,"Returns predictions for a single batch of samples.
Arguments:
x: Input data. It could be: - A Numpy array (or array-like), or a list
of arrays (in case the model has multiple inputs). - A TensorFlow
tensor, or a list of tensors (in case the model has multiple inputs).
Returns:
Numpy array(s) of predictions.
Raises:
RuntimeError: If `model.predict_on_batch` is wrapped in `tf.function`.
ValueError: In case of mismatch between given number of inputs and
expectations of the model."
5865,fit_generator,tensorflow/tensorflow/python/keras/engine/training.py,1803,method,"Fits the model on data yielded batch-by-batch by a Python generator.
DEPRECATED:
`Model.fit` now supports generators, so there is no longer any need to use
this endpoint."
5866,evaluate_generator,tensorflow/tensorflow/python/keras/engine/training.py,1843,method,"Evaluates the model on a data generator.
DEPRECATED:
`Model.evaluate` now supports generators, so there is no longer any need
to use this endpoint."
5867,predict_generator,tensorflow/tensorflow/python/keras/engine/training.py,1871,method,"Generates predictions for the input samples from a data generator.
DEPRECATED:
`Model.predict` now supports generators, so there is no longer any need
to use this endpoint."
5868,trainable_weights,tensorflow/tensorflow/python/keras/engine/training.py,1901,method,
5869,non_trainable_weights,tensorflow/tensorflow/python/keras/engine/training.py,1910,method,
5870,get_weights,tensorflow/tensorflow/python/keras/engine/training.py,1919,method,"Retrieves the weights of the model.
Returns:
A flat list of Numpy arrays."
5871,save,tensorflow/tensorflow/python/keras/engine/training.py,1928,method,"Saves the model to Tensorflow SavedModel or a single HDF5 file.
The savefile includes:
- The model architecture, allowing to re-instantiate the model.
- The model weights.
- The state of the optimizer, allowing to resume training
exactly where you left off.
This allows you to save the entirety of the state of a model
in a single file.
Saved models can be reinstantiated via `keras.models.load_model`.
The model returned by `load_model` is a compiled model ready to be used
(unless the saved model was never compiled in the first place).
Models built with the Sequential and Functional API can be saved to both the
HDF5 and SavedModel formats. Subclassed models can only be saved with the
SavedModel format.
Note that the model weights may have different scoped names after being
loaded. Scoped names include the model/layer names, such as
`""dense_1/kernel:0""`. It is recommended that you use the layer properties to
access specific variables, e.g. `model.get_layer(""dense_1"").kernel`.
Arguments:
filepath: String, PathLike, path to SavedModel or H5 file to save the
model.
overwrite: Whether to silently overwrite any existing file at the
target location, or provide the user with a manual prompt.
include_optimizer: If True, save optimizer's state together.
save_format: Either `'tf'` or `'h5'`, indicating whether to save the
model to Tensorflow SavedModel or HDF5. Defaults to 'tf' in TF 2.X,
and 'h5' in TF 1.X.
signatures: Signatures to save with the SavedModel. Applicable to the
'tf' format only. Please see the `signatures` argument in
`tf.saved_model.save` for details.
options: Optional `tf.saved_model.SaveOptions` object that specifies
options for saving to SavedModel.
Example:
```python
from keras.models import load_model
model.save('my_model.h5') # creates a HDF5 file 'my_model.h5'
del model # deletes the existing model
# returns a compiled model
# identical to the previous one
model = load_model('my_model.h5')
```"
5872,save_weights,tensorflow/tensorflow/python/keras/engine/training.py,1991,method,"Saves all layer weights.
Either saves in HDF5 or in TensorFlow format based on the `save_format`
argument.
When saving in HDF5 format, the weight file has:
- `layer_names` (attribute), a list of strings
(ordered names of model layers).
- For every layer, a `group` named `layer.name`
- For every such layer group, a group attribute `weight_names`,
a list of strings
(ordered names of weights tensor of the layer).
- For every weight in the layer, a dataset
storing the weight value, named after the weight tensor.
When saving in TensorFlow format, all objects referenced by the network are
saved in the same format as `tf.train.Checkpoint`, including any `Layer`
instances or `Optimizer` instances assigned to object attributes. For
networks constructed from inputs and outputs using `tf.keras.Model(inputs,
outputs)`, `Layer` instances used by the network are tracked/saved
automatically. For user-defined classes which inherit from `tf.keras.Model`,
`Layer` instances must be assigned to object attributes, typically in the
constructor. See the documentation of `tf.train.Checkpoint` and
`tf.keras.Model` for details.
While the formats are the same, do not mix `save_weights` and
`tf.train.Checkpoint`. Checkpoints saved by `Model.save_weights` should be
loaded using `Model.load_weights`. Checkpoints saved using
`tf.train.Checkpoint.save` should be restored using the corresponding
`tf.train.Checkpoint.restore`. Prefer `tf.train.Checkpoint` over
`save_weights` for training checkpoints.
The TensorFlow format matches objects and variables by starting at a root
object, `self` for `save_weights`, and greedily matching attribute
names. For `Model.save` this is the `Model`, and for `Checkpoint.save` this
is the `Checkpoint` even if the `Checkpoint` has a model attached. This
means saving a `tf.keras.Model` using `save_weights` and loading into a
`tf.train.Checkpoint` with a `Model` attached (or vice versa) will not match
the `Model`'s variables. See the [guide to training
checkpoints](https://www.tensorflow.org/guide/checkpoint) for details
on the TensorFlow format.
Arguments:
filepath: String or PathLike, path to the file to save the weights to.
When saving in TensorFlow format, this is the prefix used for
checkpoint files (multiple files are generated). Note that the '.h5'
suffix causes weights to be saved in HDF5 format.
overwrite: Whether to silently overwrite any existing file at the
target location, or provide the user with a manual prompt.
save_format: Either 'tf' or 'h5'. A `filepath` ending in '.h5' or
'.keras' will default to HDF5 if `save_format` is `None`. Otherwise
`None` defaults to 'tf'.
options: Optional `tf.train.CheckpointOptions` object that specifies
options for saving weights.
Raises:
ImportError: If h5py is not available when attempting to save in HDF5
format.
ValueError: For invalid/unknown format arguments."
5873,load_weights,tensorflow/tensorflow/python/keras/engine/training.py,2119,method,"Loads all layer weights, either from a TensorFlow or an HDF5 weight file.
If `by_name` is False weights are loaded based on the network's
topology. This means the architecture should be the same as when the weights
were saved. Note that layers that don't have weights are not taken into
account in the topological ordering, so adding or removing layers is fine as
long as they don't have weights.
If `by_name` is True, weights are loaded into layers only if they share the
same name. This is useful for fine-tuning or transfer-learning models where
some of the layers have changed.
Only topological loading (`by_name=False`) is supported when loading weights
from the TensorFlow format. Note that topological loading differs slightly
between TensorFlow and HDF5 formats for user-defined classes inheriting from
`tf.keras.Model`: HDF5 loads based on a flattened list of weights, while the
TensorFlow format loads based on the object-local names of attributes to
which layers are assigned in the `Model`'s constructor.
Arguments:
filepath: String, path to the weights file to load. For weight files in
TensorFlow format, this is the file prefix (the same as was passed
to `save_weights`).
by_name: Boolean, whether to load weights by name or by topological
order. Only topological loading is supported for weight files in
TensorFlow format.
skip_mismatch: Boolean, whether to skip loading of layers where there is
a mismatch in the number of weights, or a mismatch in the shape of
the weight (only valid when `by_name=True`).
options: Optional `tf.train.CheckpointOptions` object that specifies
options for loading weights.
Returns:
When loading a weight file in TensorFlow format, returns the same status
object as `tf.train.Checkpoint.restore`. When graph building, restore
ops are run automatically as soon as the network is built (on first call
for user-defined classes inheriting from `Model`, immediately if it is
already built).
When loading weights in HDF5 format, returns `None`.
Raises:
ImportError: If h5py is not available and the weight file is in HDF5
format.
ValueError: If `skip_mismatch` is set to `True` when `by_name` is
`False`."
5874,get_config,tensorflow/tensorflow/python/keras/engine/training.py,2240,method,
5875,from_config,tensorflow/tensorflow/python/keras/engine/training.py,2244,method,
5876,to_json,tensorflow/tensorflow/python/keras/engine/training.py,2251,method,"Returns a JSON string containing the network configuration.
To load a network from a JSON save file, use
`keras.models.model_from_json(json_string, custom_objects={})`.
Arguments:
**kwargs: Additional keyword arguments
to be passed to `json.dumps()`.
Returns:
A JSON string."
5877,to_yaml,tensorflow/tensorflow/python/keras/engine/training.py,2268,method,"Returns a yaml string containing the network configuration.
To load a network from a yaml save file, use
`keras.models.model_from_yaml(yaml_string, custom_objects={})`.
`custom_objects` should be a dictionary mapping
the names of custom losses / layers / etc to the corresponding
functions / classes.
Arguments:
**kwargs: Additional keyword arguments
to be passed to `yaml.dump()`.
Returns:
A YAML string.
Raises:
ImportError: if yaml module is not found."
5878,reset_states,tensorflow/tensorflow/python/keras/engine/training.py,2293,method,
5879,state_updates,tensorflow/tensorflow/python/keras/engine/training.py,2304,method,"Deprecated, do NOT use!
Returns the `updates` from all layers that are stateful.
This is useful for separating training updates and
state updates, e.g. when we need to update a layer's internal state
during prediction.
Returns:
A list of update ops."
5880,weights,tensorflow/tensorflow/python/keras/engine/training.py,2324,method,"Returns the list of all layer variables/weights.
Returns:
A list of variables."
5881,summary,tensorflow/tensorflow/python/keras/engine/training.py,2342,method,"Prints a string summary of the network.
Arguments:
line_length: Total length of printed lines
(e.g. set this to adapt the display to different
terminal window sizes).
positions: Relative or absolute positions of log elements
in each line. If not provided,
defaults to `[.33, .55, .67, 1.]`.
print_fn: Print function to use. Defaults to `print`.
It will be called on each line of the summary.
You can set it to a custom function
in order to capture the string summary.
Raises:
ValueError: if `summary()` is called before the model is built."
5882,layers,tensorflow/tensorflow/python/keras/engine/training.py,2372,method,
5883,get_layer,tensorflow/tensorflow/python/keras/engine/training.py,2375,method,"Retrieves a layer based on either its name (unique) or index.
If `name` and `index` are both provided, `index` will take precedence.
Indices are based on order of horizontal graph traversal (bottom-up).
Arguments:
name: String, name of layer.
index: Integer, index of layer.
Returns:
A layer instance.
Raises:
ValueError: In case of invalid layer name or index."
5884,step_function,tensorflow/tensorflow/python/keras/engine/training.py,788,method,Runs a single training step.
5885,step_function,tensorflow/tensorflow/python/keras/engine/training.py,1211,method,Runs a single evaluation step.
5886,step_function,tensorflow/tensorflow/python/keras/engine/training.py,1448,method,Runs a single evaluation step.
5887,run_step,tensorflow/tensorflow/python/keras/engine/training.py,791,method,
5888,train_function,tensorflow/tensorflow/python/keras/engine/training.py,807,method,Runs a training execution with one step.
5889,train_function,tensorflow/tensorflow/python/keras/engine/training.py,813,method,Runs a training execution with multiple steps.
5890,run_step,tensorflow/tensorflow/python/keras/engine/training.py,1214,method,
5891,run_step,tensorflow/tensorflow/python/keras/engine/training.py,1451,method,
5892,predict_function,tensorflow/tensorflow/python/keras/engine/training.py,1467,method,Runs an evaluation execution with one step.
5893,predict_function,tensorflow/tensorflow/python/keras/engine/training.py,1473,method,Runs an evaluation execution with multiple steps.
5894,reduce_per_replica,tensorflow/tensorflow/python/keras/engine/training.py,2647,function,"Reduce PerReplica objects.
Arguments:
values: Structure of `PerReplica` objects or `Tensor`s. `Tensor`s are
returned as-is.
strategy: `tf.distribute.Strategy` object.
reduction: One of 'first', 'concat'.
Returns:
Structure of `Tensor`s."
5895,concat,tensorflow/tensorflow/python/keras/engine/training.py,2677,function,Concats `tensor`s along `axis`.
5896,write_scalar_summaries,tensorflow/tensorflow/python/keras/engine/training.py,2764,function,
5897,model_iteration,tensorflow/tensorflow/python/keras/engine/training_arrays.py,46,function,"Loop function for arrays of data with modes TRAIN/TEST/PREDICT.
Arguments:
model: Keras Model instance.
inputs: Either a list or dictionary of arrays, or a dataset instance.
targets: List/dictionary of input arrays.
sample_weights: Optional list of sample weight arrays.
batch_size: Integer batch size or None if unknown.
epochs: Number of times to iterate over the data
verbose: 0, 1, or 2. Verbosity mode.
0 = silent, 1 = progress bar, 2 = one line per epoch.
Note that the progress bar is not particularly useful when
logged to a file, so verbose=2 is recommended when not running
interactively (eg, in a production environment).
callbacks: List of callbacks to be called during training
val_inputs: Either a list or dictionary of arrays, or a dataset instance.
val_targets: List/dictionary of target arrays.
val_sample_weights: Optional list of sample weight arrays.
shuffle: Whether to shuffle the data at the beginning of each epoch
concatenation of list the display names of the outputs of `f` and the
list of display names of the outputs of `f_val`.
initial_epoch: Epoch at which to start training (useful for resuming a
previous training run)
steps_per_epoch: Total number of steps (batches of samples) before
declaring one epoch finished and starting the next epoch. Ignored with
the default value of `None`.
validation_steps: Number of steps to run validation for (only if doing
validation from data tensors). Ignored with the default value of
`None`.
validation_freq: Only relevant if validation data is provided. Integer or
`collections_abc.Container` instance (e.g. list, tuple, etc.). If an
integer, specifies how many training epochs to run before a new
validation run is performed, e.g. `validation_freq=2` runs
validation every 2 epochs. If a Container, specifies the epochs on
which to run validation, e.g. `validation_freq=[1, 2, 10]` runs
validation at the end of the 1st, 2nd, and 10th epochs.
mode: One of ModeKeys.TRAIN/ModeKeys.TEST/ModeKeys.PREDICT.
validation_in_fit: if true, then this method is invoked from within
training iteration (for validation). In the case where `val_inputs` is
a dataset, this flag indicates that its iterator and feed values are
already created so should properly reuse resources.
prepared_feed_values_from_dataset: if True, `inputs` is a list of feed
tensors returned from `_prepare_feed_values` call on the validation
dataset, so do not call it again on `inputs`. Should only be used for
inline validation (i.e., only if `validation_in_fit` is also True).
steps_name: The string name of the steps argument, either `steps`,
`validation_steps`, or `steps_per_epoch`. Only used for error message
formatting.
**kwargs: Additional arguments for backwards compatibility.
Returns:
- In TRAIN mode: `History` object.
- In TEST mode: Evaluation metrics.
- In PREDICT mode: Outputs of the Model called on inputs.
Raises:
ValueError: in case of invalid arguments."
5898,ArrayLikeTrainingLoop,tensorflow/tensorflow/python/keras/engine/training_arrays.py,594,class,"TrainingLoop that handle inputs like array.
This is the default handler for most of the input data types, includes
symbolic tensors or Numpy array-like, Datasets and iterators in graph mode
(since they generate symbolic tensors). This Function is used to handle model
with `run_eagerly` = False."
5899,fit,tensorflow/tensorflow/python/keras/engine/training_arrays.py,603,method,
5900,evaluate,tensorflow/tensorflow/python/keras/engine/training_arrays.py,668,method,
5901,predict,tensorflow/tensorflow/python/keras/engine/training_arrays.py,697,method,
5902,BatchCounterCallback,tensorflow/tensorflow/python/keras/engine/training_dataset_test.py,40,class,
5903,on_batch_begin,tensorflow/tensorflow/python/keras/engine/training_dataset_test.py,46,method,
5904,on_batch_end,tensorflow/tensorflow/python/keras/engine/training_dataset_test.py,49,method,
5905,experimental_tpu_fit_loop,tensorflow/tensorflow/python/keras/engine/training_distributed.py,123,function,"Fit loop for training with TPU tf.distribute.Strategy.
Arguments:
model: Keras Model instance.
dataset: Dataset that returns inputs and targets
epochs: Number of times to iterate over the data
verbose: Integer, Verbosity mode, 0, 1 or 2
callbacks: List of callbacks to be called during training
initial_epoch: Epoch at which to start training
(useful for resuming a previous training run)
steps_per_epoch: Total number of steps (batches of samples)
before declaring one epoch finished and starting the
next epoch. Ignored with the default value of `None`.
val_dataset: Dataset for validation data.
validation_steps: Number of steps to run validation for
(only if doing validation from data tensors).
Ignored with the default value of `None`.
validation_freq: Only relevant if validation data is provided. Integer or
`collections.abc.Container` instance (e.g. list, tuple, etc.). If an
integer, specifies how many training epochs to run before a new
validation run is performed, e.g. `validation_freq=2` runs
validation every 2 epochs. If a Container, specifies the epochs on
which to run validation, e.g. `validation_freq=[1, 2, 10]` runs
validation at the end of the 1st, 2nd, and 10th epochs.
Returns:
Returns `None`.
Raises:
ValueError: in case of invalid arguments."
5906,experimental_tpu_predict_loop,tensorflow/tensorflow/python/keras/engine/training_distributed.py,423,function,"Predict loop for predicting with TPU tf.distribute.Strategy.
Arguments:
model: Keras Model instance.
dataset: Dataset for input data.
verbose: Integer, Verbosity mode 0 or 1.
steps: Total number of steps (batches of samples)
before declaring `_predict_loop` finished.
Ignored with the default value of `None`.
callbacks: List of callbacks to be called during training
Returns:
Array of predictions (if the model has a single output)
or list of arrays of predictions
(if the model has multiple outputs)."
5907,DistributionSingleWorkerTrainingLoop,tensorflow/tensorflow/python/keras/engine/training_distributed.py,577,class,Training loop for distribution strategy with single worker.
5908,fit,tensorflow/tensorflow/python/keras/engine/training_distributed.py,580,method,Fit loop for Distribution Strategies.
5909,evaluate,tensorflow/tensorflow/python/keras/engine/training_distributed.py,687,method,Evaluate loop for Distribution Strategies.
5910,predict,tensorflow/tensorflow/python/keras/engine/training_distributed.py,728,method,Predict loop for Distribution Strategies.
5911,DistributionMultiWorkerTrainingLoop,tensorflow/tensorflow/python/keras/engine/training_distributed.py,782,class,Training loop for distribution strategy with multiple worker.
5912,fit,tensorflow/tensorflow/python/keras/engine/training_distributed.py,788,method,
5913,evaluate,tensorflow/tensorflow/python/keras/engine/training_distributed.py,792,method,
5914,predict,tensorflow/tensorflow/python/keras/engine/training_distributed.py,796,method,
5915,train_on_batch,tensorflow/tensorflow/python/keras/engine/training_eager.py,285,function,"Calculates the loss and gradient updates for one input batch.
Arguments:
model: Model whose loss has to be calculated.
inputs: Input batch data.
targets: Target batch data.
sample_weights: Sample weight batch data.
output_loss_metrics: List of metrics that are used to aggregated output
loss values.
Returns:
Dict with three items:
'total_loss': list with a single tensor for overall loss,
'output_losses': list of tensors for loss corresponding to each of the
model output. Could be a empty list when model has only one output.
'metrics': list of tensors for metric specified."
5916,model_iteration,tensorflow/tensorflow/python/keras/engine/training_generator.py,41,function,"Loop function for arrays of data with modes TRAIN/TEST/PREDICT.
Arguments:
model: Keras Model instance.
data: Either a tuple of NumPy/Tensor inputs (i.e. `(x,)` or `(x, y)` or
`(x, y, sample_weights)`) or a generator or
`keras.utils.data_utils.Sequence` object or Eager Iterator or Dataset.
steps_per_epoch: Total number of steps (batches of samples) before
declaring one epoch finished and starting the next epoch. Ignored with
the default value of `None`.
epochs: Number of times to iterate over the data.
verbose: 0, 1, or 2. Verbosity mode.
0 = silent, 1 = progress bar, 2 = one line per epoch.
Note that the progress bar is not particularly useful when
logged to a file, so verbose=2 is recommended when not running
interactively (eg, in a production environment).
callbacks: List of callbacks to be called during training.
validation_data: Either a tuple of NumPy/Tensor inputs (i.e. `(x,)` or
`(x, y)` or `(x, y, sample_weights)`) or a generator or
`keras.utils.data_utils.Sequence` object or Eager Iterator or Dataset.
validation_steps: Total number of steps (batches of samples) before
declaring validation finished.
validation_freq: Only relevant if validation data is provided. Integer or
`collections.abc.Container` instance (e.g. list, tuple, etc.). If an
integer, specifies how many training epochs to run before a new
validation run is performed, e.g. `validation_freq=2` runs
validation every 2 epochs. If a Container, specifies the epochs on
which to run validation, e.g. `validation_freq=[1, 2, 10]` runs
validation at the end of the 1st, 2nd, and 10th epochs.
class_weight: Dictionary mapping class indices to a weight for the class.
max_queue_size: Integer. Maximum size for the generator queue. If
unspecified, `max_queue_size` will default to 10.
workers: Integer. Maximum number of processes to spin up when using
process-based threading. If unspecified, `workers` will default to 1. If
0, will execute the generator on the main thread.
use_multiprocessing: Boolean. If `True`, use process-based threading. If
unspecified, `use_multiprocessing` will default to `False`. Note that
because this implementation relies on multiprocessing, you should not
pass non-picklable arguments to the generator as they can't be passed
easily to children processes.
shuffle: Boolean. Whether to shuffle the order of the batches at the
beginning of each epoch. Only used with instances of `Sequence`
(`keras.utils.Sequence`). Has no effect when `steps_per_epoch` is not
`None`.
initial_epoch: Epoch at which to start training (useful for resuming a
previous training run).
mode: One of ModeKeys.TRAIN/ModeKeys.TEST/ModeKeys.PREDICT.
batch_size: Integer batch size or None if unknown. Will only be used if
`data` is in NumPy/Tensor format.
steps_name: The string name of the steps argument, either `steps`,
`validation_steps`, or `steps_per_epoch`. Only used for error message
formatting.
**kwargs: Additional arguments for backwards compatibility. `steps` is
accepted as an alias for `steps_per_epoch`.
Returns:
- In TRAIN mode: `History` object.
- In TEST mode: Evaluation metrics.
- In PREDICT mode: Outputs of the Model called on inputs.
Raises:
ValueError: in case of invalid arguments."
5917,convert_to_generator_like,tensorflow/tensorflow/python/keras/engine/training_generator.py,422,function,"Make a generator out of NumPy or EagerTensor inputs.
Arguments:
data: Either a generator or `keras.utils.data_utils.Sequence` object or
`Dataset`, `Iterator`, or a {1,2,3}-tuple of NumPy arrays or EagerTensors.
If a tuple, the elements represent `(x, y, sample_weights)` and may be
`None` or `[None]`.
batch_size: Used when creating a generator out of tuples of NumPy arrays or
EagerTensors.
steps_per_epoch: Steps of the generator to run each epoch. If `None` the
number of steps will be read from the data (for
`keras.utils.data_utils.Sequence` types).
epochs: Total number of epochs to run.
shuffle: Whether the data should be shuffled.
Returns:
- Generator, `keras.utils.data_utils.Sequence`, or `Iterator`.
Raises:
- ValueError: If `batch_size` is not provided for NumPy or EagerTensor
inputs."
5918,GeneratorOrSequenceTrainingLoop,tensorflow/tensorflow/python/keras/engine/training_generator.py,541,class,"Generator-like.
Input is Python generator, or Sequence object.
The difference between this class and `GeneratorLikeTrainingFunction` is that
this class only handles inputs that with x, y and sample_weight fused into one
param."
5919,fit,tensorflow/tensorflow/python/keras/engine/training_generator.py,551,method,
5920,evaluate,tensorflow/tensorflow/python/keras/engine/training_generator.py,592,method,
5921,predict,tensorflow/tensorflow/python/keras/engine/training_generator.py,616,method,
5922,EagerDatasetOrIteratorTrainingLoop,tensorflow/tensorflow/python/keras/engine/training_generator.py,638,class,A non-distributed Dataset or iterator in eager execution.
5923,fit,tensorflow/tensorflow/python/keras/engine/training_generator.py,641,method,
5924,evaluate,tensorflow/tensorflow/python/keras/engine/training_generator.py,682,method,
5925,predict,tensorflow/tensorflow/python/keras/engine/training_generator.py,698,method,
5926,GeneratorLikeTrainingLoop,tensorflow/tensorflow/python/keras/engine/training_generator.py,711,class,"TrainingLoop that handle inputs like python generator.
This is the default handler for most of the input data types, includes
symbolic tensors or Numpy array-like, Datasets and iterators in graph mode
(since they generate symbolic tensors). This Function is used to handle model
with `run_eagerly` = True."
5927,fit,tensorflow/tensorflow/python/keras/engine/training_generator.py,720,method,
5928,evaluate,tensorflow/tensorflow/python/keras/engine/training_generator.py,781,method,
5929,predict,tensorflow/tensorflow/python/keras/engine/training_generator.py,808,method,
5930,custom_generator,tensorflow/tensorflow/python/keras/engine/training_generator_test.py,44,function,
5931,custom_generator_changing_batch_size,tensorflow/tensorflow/python/keras/engine/training_generator_test.py,67,function,
5932,BareUpdateLayer,tensorflow/tensorflow/python/keras/engine/training_test.py,3431,class,
5933,build,tensorflow/tensorflow/python/keras/engine/training_test.py,3433,method,
5934,call,tensorflow/tensorflow/python/keras/engine/training_test.py,3441,method,
5935,LambdaUpdateLayer,tensorflow/tensorflow/python/keras/engine/training_test.py,3446,class,
5936,build,tensorflow/tensorflow/python/keras/engine/training_test.py,3448,method,
5937,call,tensorflow/tensorflow/python/keras/engine/training_test.py,3456,method,
5938,NestedUpdateLayer,tensorflow/tensorflow/python/keras/engine/training_test.py,3462,class,
5939,build,tensorflow/tensorflow/python/keras/engine/training_test.py,3464,method,
5940,counter,tensorflow/tensorflow/python/keras/engine/training_test.py,3469,method,
5941,call,tensorflow/tensorflow/python/keras/engine/training_test.py,3472,method,
5942,SubgraphUpdateLayer,tensorflow/tensorflow/python/keras/engine/training_test.py,3476,class,
5943,build,tensorflow/tensorflow/python/keras/engine/training_test.py,3478,method,
5944,call,tensorflow/tensorflow/python/keras/engine/training_test.py,3486,method,
5945,Aggregator,tensorflow/tensorflow/python/keras/engine/training_utils.py,65,class,"Abstract base class used to aggregate batch-level outputs of a loop.
Attributes:
use_steps: Whether the loop is using `step` or `batch_size`.
num_samples: Total number of samples: `batch_size * num_batches`.
steps: Total number of steps.
batch_size: Batch size. It is used for validation checks between inputs and
outputs.
results: What to return at the end of the aggregation loop."
5946,create,tensorflow/tensorflow/python/keras/engine/training_utils.py,85,method,"Creates the initial results from the first batch outputs.
Arguments:
batch_outs: A list of batch-level outputs."
5947,aggregate,tensorflow/tensorflow/python/keras/engine/training_utils.py,94,method,"Aggregates batch-level results into total results.
Arguments:
batch_outs: A list of batch-level outputs.
batch_start: The start index of this batch. Always `None` if `use_steps`
is `True`.
batch_end: The end index of this batch. Always `None` if `use_steps` is
`True`."
5948,finalize,tensorflow/tensorflow/python/keras/engine/training_utils.py,107,method,Prepares the total results to be returned.
5949,MetricsAggregator,tensorflow/tensorflow/python/keras/engine/training_utils.py,112,class,"Aggregator that calculates loss and metrics info.
Attributes:
use_steps: Whether the loop is using `step` or `batch_size`.
num_samples: Total number of samples: `batch_size*num_batches`.
steps: Total number of steps, ie number of times to iterate over a dataset
to cover all samples."
5950,create,tensorflow/tensorflow/python/keras/engine/training_utils.py,129,method,
5951,aggregate,tensorflow/tensorflow/python/keras/engine/training_utils.py,132,method,
5952,finalize,tensorflow/tensorflow/python/keras/engine/training_utils.py,141,method,
5953,ConcatAggregator,tensorflow/tensorflow/python/keras/engine/training_utils.py,147,class,"Combine tensor-likes which cannot be merged on the fly.
This class expects to aggregate a single tensor-like rather than a nested
structure of tensor-likes."
5954,create,tensorflow/tensorflow/python/keras/engine/training_utils.py,159,method,
5955,aggregate,tensorflow/tensorflow/python/keras/engine/training_utils.py,163,method,
5956,finalize,tensorflow/tensorflow/python/keras/engine/training_utils.py,175,method,
5957,get_copy_pool,tensorflow/tensorflow/python/keras/engine/training_utils.py,198,function,"Shared threadpool for copying arrays.
Pool instantiation takes ~ 2ms, so a singleton pool is used rather than
creating a pool per SliceAggregator.
Returns:
The global copy threadpool."
5958,SliceAggregator,tensorflow/tensorflow/python/keras/engine/training_utils.py,214,class,"Combine arrays where the final size is known.
This class expects to aggregate a single tensor-like rather than a nested
structure of tensor-likes.
NumPy copies are an operation that threads handle quite well because all of
the heavy lifting is in c and does not need the GIL. Moreover, we can perform
lock-free writes to the same buffer in multiple threads because the nature of
result aggregation guarantees that either the indices are disjoint or the
aggregator will throw an exception in finalize. Moreover, because aggregation
is performed on the slowest varying dimension, assignments for a given batch
will write to contiguous blocks of memory, further minimizing contention.
There is, however, some scheduling and context switching overhead which will
offset the gains from pipelining the slice assignment. Below a given threshold
it is faster to simply assign in the main thread rather than enqueue the
assignment in a side thread. The exact threshold will vary from system to
system, but the time is not very sensitive to the exact transition so a value
of 2 ** 14 was chosen which should be reasonable on most systems."
5959,create,tensorflow/tensorflow/python/keras/engine/training_utils.py,249,method,
5960,aggregate,tensorflow/tensorflow/python/keras/engine/training_utils.py,259,method,
5961,finalize,tensorflow/tensorflow/python/keras/engine/training_utils.py,301,method,
5962,OutputsAggregator,tensorflow/tensorflow/python/keras/engine/training_utils.py,312,class,Aggregator that concatenates outputs.
5963,create,tensorflow/tensorflow/python/keras/engine/training_utils.py,317,method,
5964,aggregate,tensorflow/tensorflow/python/keras/engine/training_utils.py,343,method,
5965,finalize,tensorflow/tensorflow/python/keras/engine/training_utils.py,348,method,
5966,get_progbar,tensorflow/tensorflow/python/keras/engine/training_utils.py,355,function,Get Progbar.
5967,slice_arrays,tensorflow/tensorflow/python/keras/engine/training_utils.py,366,function,"Slices batches out of provided arrays (workaround for eager tensors).
Unfortunately eager tensors don't have the same slicing behavior as
Numpy arrays (they follow the same slicing behavior as symbolic TF tensors),
hence we cannot use `generic_utils.slice_arrays` directly
and we have to implement this workaround based on `concat`. This has a
performance cost.
Arguments:
arrays: Single array or list of arrays.
indices: List of indices in the array that should be included in the output
batch.
contiguous: Boolean flag indicating whether the indices are contiguous.
Returns:
Slice of data (either single array or list of arrays)."
5968,check_num_samples,tensorflow/tensorflow/python/keras/engine/training_utils.py,402,function,"Determine the number of samples provided for training and evaluation.
The number of samples is not defined when running with `steps`,
in which case the number of samples is set to `None`.
Arguments:
ins: List of tensors to be fed to the Keras function.
batch_size: Integer batch size or `None` if not defined.
steps: Total number of steps (batches of samples) before declaring
`_predict_loop` finished. Ignored with the default value of `None`.
steps_name: The public API's parameter name for `steps`.
Raises:
ValueError: when `steps` is `None` and the attribute `ins.shape`
does not exist. Also raises ValueError when `steps` is not `None`
and `batch_size` is not `None` because they are mutually
exclusive.
Returns:
When steps is `None`, returns the number of samples to be
processed based on the size of the first dimension of the
first input numpy array. When steps is not `None` and
`batch_size` is `None`, returns `None`."
5969,standardize_single_array,tensorflow/tensorflow/python/keras/engine/training_utils.py,438,function,"Expand data of shape (x,) to (x, 1), unless len(expected_shape)==1."
5970,standardize_input_data,tensorflow/tensorflow/python/keras/engine/training_utils.py,459,function,"Normalizes inputs and targets provided by users.
Users may pass data as a list of arrays, dictionary of arrays,
or as a single array. We normalize this to an ordered list of
arrays (same order as `names`), while checking that the provided
arrays have shapes that match the network's expectations.
Arguments:
data: User-provided input data (polymorphic).
names: List of expected array names.
shapes: Optional list of expected array shapes.
check_batch_axis: Boolean; whether to check that the batch axis of the
arrays matches the expected value found in `shapes`.
exception_prefix: String prefix used for exception formatting.
Returns:
List of standardized input arrays (one array per model input).
Raises:
ValueError: in case of improperly formatted user-provided data."
5971,standardize_sample_or_class_weights,tensorflow/tensorflow/python/keras/engine/training_utils.py,585,function,"Maps `sample_weight` or `class_weight` to model outputs.
Arguments:
x_weight: User-provided `sample_weight` or `class_weight` argument.
output_names: List of output names (strings) in the model.
weight_type: A string used purely for exception printing.
Returns:
A list of `sample_weight` or `class_weight` where there are exactly
one element per model output.
Raises:
ValueError: In case of invalid user-provided argument."
5972,standardize_class_weights,tensorflow/tensorflow/python/keras/engine/training_utils.py,631,function,
5973,standardize_sample_weights,tensorflow/tensorflow/python/keras/engine/training_utils.py,636,function,
5974,handle_partial_sample_weights,tensorflow/tensorflow/python/keras/engine/training_utils.py,641,function,"Adds 1.0 as sample weights for the outputs for which there is no weight.
Args:
outputs: List of model outputs.
sample_weights: List of sample weight inputs.
sample_weight_modes: List of sample weight modes or None.
check_all_flat: Ensure that inputs are not nested structures. This is not
a free check, so we may not want to run it eagerly every iteration.
Returns:
Tuple of sample weights, one sample weight for every output, and booleans
describing the raw sample weights."
5975,check_array_lengths,tensorflow/tensorflow/python/keras/engine/training_utils.py,700,function,"Does user input validation for numpy arrays.
Arguments:
inputs: list of Numpy arrays of inputs.
targets: list of Numpy arrays of targets.
weights: list of Numpy arrays of sample weights.
Raises:
ValueError: in case of incorrectly formatted data."
5976,check_loss_and_target_compatibility,tensorflow/tensorflow/python/keras/engine/training_utils.py,755,function,"Does validation on the compatibility of targets and loss functions.
This helps prevent users from using loss functions incorrectly. This check
is purely for UX purposes.
Arguments:
targets: list of Numpy arrays of targets.
loss_fns: list of loss functions.
output_shapes: list of shapes of model outputs.
Raises:
ValueError: if a loss function or target array
is incompatible with an output."
5977,collect_per_output_metric_info,tensorflow/tensorflow/python/keras/engine/training_utils.py,814,function,"Maps metric names and functions to model outputs.
Arguments:
metrics: a list or a list of lists or a dict of metric functions.
output_names: a list of the names (strings) of model outputs.
output_shapes: a list of the shapes (strings) of model outputs.
loss_fns: a list of the loss functions corresponding to the model outputs.
is_weighted: Boolean indicating whether the given metrics are weighted.
Returns:
A list (one entry per model output) of dicts.
For instance, if the model has 2 outputs, and for the first output
we want to compute ""binary_accuracy"" and ""binary_crossentropy"",
and just ""binary_accuracy"" for the second output,
the list would look like: `[{
'acc': binary_accuracy(),
'ce': binary_crossentropy(),
}, {
'acc': binary_accuracy(),
}]`
Raises:
TypeError: if an incorrect type is passed for the `metrics` argument."
5978,batch_shuffle,tensorflow/tensorflow/python/keras/engine/training_utils.py,893,function,"Shuffles an array in a batch-wise fashion.
Useful for shuffling HDF5 arrays
(where one cannot access arbitrary indices).
Arguments:
index_array: array of indices to be shuffled.
batch_size: integer.
Returns:
The `index_array` array, shuffled in a batch-wise fashion."
5979,standardize_weights,tensorflow/tensorflow/python/keras/engine/training_utils.py,917,function,"Performs sample weight validation and standardization.
Everything gets normalized to a single sample-wise (or timestep-wise)
weight array. If both `sample_weight` and `class_weight` are provided,
the weights are multiplied.
Arguments:
y: Numpy array or Tensor of model targets to be weighted.
sample_weight: User-provided `sample_weight` argument.
class_weight: User-provided `class_weight` argument.
sample_weight_mode: One of `None` or `""temporal""`. `""temporal""` indicated
that we expect 2D weight data that will be applied to the last 2
dimensions of the targets (i.e. we are weighting timesteps, not
samples).
Returns:
A numpy array of target weights, one entry per sample to weight.
Raises:
ValueError: In case of invalid user-provided arguments."
5980,has_symbolic_tensors,tensorflow/tensorflow/python/keras/engine/training_utils.py,1044,function,
5981,has_tensors,tensorflow/tensorflow/python/keras/engine/training_utils.py,1050,function,Returns true if `ls` contains tensors.
5982,get_metric_name,tensorflow/tensorflow/python/keras/engine/training_utils.py,1068,function,"Returns the name corresponding to the given metric input.
Arguments:
metric: Metric function name or reference.
weighted: Boolean indicating if the given metric is weighted.
Returns:
The metric name."
5983,get_metric_function,tensorflow/tensorflow/python/keras/engine/training_utils.py,1103,function,"Returns the metric function corresponding to the given metric input.
Arguments:
metric: Metric function name or reference.
output_shape: The shape of the output that this metric will be calculated
for.
loss_fn: The loss function used.
Returns:
The metric function."
5984,call_metric_function,tensorflow/tensorflow/python/keras/engine/training_utils.py,1145,function,Invokes metric function and returns the metric result tensor.
5985,get_loss_function,tensorflow/tensorflow/python/keras/engine/training_utils.py,1169,function,Returns the loss corresponding to the loss input in `compile` API.
5986,RespectCompiledTrainableState,tensorflow/tensorflow/python/keras/engine/training_utils.py,1201,class,"Set and restore trainable state if it has changed since compile.
The keras API guarantees that the value of each Layer's `trainable` property
at `Model.compile` time will be used when training that model. In order to
respect this requirement, it may be necessary to set the trainable value of
layers to their compile time values before beginning a training endpoint and
restore the values before returing from said endpoint. This scope checks if
any layer's trainable state has changed since Model compile, and performs this
set and un-set bookkeeping.
However, the trainable state of a layer changes quite infrequently, if ever,
for many kinds of workflows. Moreover, updating every layer in a model is an
expensive operation. As a result, we will only explicitly set and unset the
trainable state of a model if a trainable value has changed since compile."
5987,validate_dataset_input,tensorflow/tensorflow/python/keras/engine/training_utils.py,1247,function,"Validates user input arguments when a dataset iterator is passed.
Arguments:
x: Input data. A `tf.data` dataset or iterator.
y: Target data. It could be either Numpy array(s) or TensorFlow tensor(s).
Expected to be `None` when `x` is a dataset iterator.
sample_weight: An optional sample-weight array passed by the user to weight
the importance of each sample in `x`. Expected to be `None` when `x` is a
dataset iterator
validation_split: Float between 0 and 1. Fraction of the training data to be
used as validation data. Expected to be `None` when `x` is a dataset
iterator.
Raises:
ValueError: if argument `y` or `sample_weight` or `validation_split` are
provided by user."
5988,validate_input_types,tensorflow/tensorflow/python/keras/engine/training_utils.py,1285,function,Helper function to validate either inputs or targets.
5989,check_generator_arguments,tensorflow/tensorflow/python/keras/engine/training_utils.py,1303,function,Validates arguments passed when using a generator.
5990,check_steps_argument,tensorflow/tensorflow/python/keras/engine/training_utils.py,1319,function,"Validates `steps` argument based on input data's type.
The cases when `steps` value must be provided are when
1. input data passed is an iterator.
2. model was built on top of symbolic tensors, input data is not
required and is `None`.
3. input data passed is a symbolic tensor.
Arguments:
input_data: Input data. Can be Numpy array(s) or TensorFlow tensor(s) or
tf.data.Dataset iterator or `None`.
steps: Integer or `None`. Total number of steps (batches of samples) to
execute.
steps_name: The public API's parameter name for `steps`.
Returns:
boolean, True if `steps` argument is required, else False.
Raises:
ValueError: if `steps` argument is required for given input data type
but not provided."
5991,cast_single_tensor,tensorflow/tensorflow/python/keras/engine/training_utils.py,1367,function,
5992,cast_if_floating_dtype_and_mismatch,tensorflow/tensorflow/python/keras/engine/training_utils.py,1376,function,"Returns target data tensors using correct datatype.
Checks that each target and output pair are the same datatype. If not, casts
the target to the output's datatype.
Args:
targets: tensor or list of targets.
outputs: tensor or list of outputs.
Returns:
Targets in appropriate datatype."
5993,cast_if_floating_dtype,tensorflow/tensorflow/python/keras/engine/training_utils.py,1403,function,"Casts the given data tensors to the default floating point type.
Casts only if the input is already a floating point type.
Args:
x: tensor or list/tuple of tensors.
dtype: The dtype to which Tensors should be cast.
Returns:
Converted input."
5994,cast_to_model_input_dtypes,tensorflow/tensorflow/python/keras/engine/training_utils.py,1418,function,"Casts the given data tensors to the dtypes of the model inputs.
Args:
x: tensor or list/tuple of tensors.
model: The model.
Returns:
Converted input. Each tensor is casted to the corresponding input in
`model.inputs`."
5995,prepare_sample_weight_modes,tensorflow/tensorflow/python/keras/engine/training_utils.py,1433,function,"Prepares sample weight modes for the model.
Args:
training_endpoints: List of model _TrainingEndpoints.
sample_weight_mode: sample weight mode user input passed from compile API.
Raises:
ValueError: In case of invalid `sample_weight_mode` input."
5996,prepare_loss_functions,tensorflow/tensorflow/python/keras/engine/training_utils.py,1473,function,"Converts loss to a list of loss functions.
Arguments:
loss: String (name of objective function), objective function or
`tf.losses.Loss` instance. See `tf.losses`. If the model has multiple
outputs, you can use a different loss on each output by passing a
dictionary or a list of losses. The loss value that will be minimized by
the model will then be the sum of all individual losses.
output_names: List of model output names.
Returns:
A list of loss objective functions.
Raises:
ValueError: If loss is a dict with keys not in model output names,
or if loss is a list with len not equal to model outputs."
5997,prepare_loss_weights,tensorflow/tensorflow/python/keras/engine/training_utils.py,1515,function,"Converts loss weights to a list of loss weights.
The result loss weights will be populated on the training endpoint.
Arguments:
training_endpoints: List of model training endpoints.
loss_weights: Optional list or dictionary specifying scalar coefficients
(Python floats) to weight the loss contributions of different model
outputs. The loss value that will be minimized by the model will then be
the *weighted sum* of all individual losses, weighted by the
`loss_weights` coefficients. If a list, it is expected to have a 1:1
mapping to the model's outputs. If a dict, it is expected to map
output names (strings) to scalar coefficients.
Raises:
ValueError: If loss weight is a dict with key not in model output names,
or if loss is a list with len not equal to model outputs."
5998,is_feature_layer,tensorflow/tensorflow/python/keras/engine/training_utils.py,1559,function,Returns whether `layer` is a FeatureLayer or not.
5999,is_eager_dataset_or_iterator,tensorflow/tensorflow/python/keras/engine/training_utils.py,1564,function,
6000,assert_not_batched,tensorflow/tensorflow/python/keras/engine/training_utils.py,1571,function,"Asserts that `dataset` is not batched.
The algorithm used by this method is sound but not complete. In other words,
if the method fails to establish the assertion, it does not mean the dataset
is batched.
Example usage:
```python
try:
assert_not_batched(dataset)
# safe to assume `dataset` it not batched here
expect ValueError:
# make no assumptions about `dataset`
```
Args:
dataset: The dataset to analyze.
Raises:
ValueError: If the method cannot establish the assertion."
6001,assert_not_shuffled,tensorflow/tensorflow/python/keras/engine/training_utils.py,1626,function,"Asserts that `dataset` is not shuffled.
The algorithm used by this method is sound but not complete. In other words,
if the method fails to establish the assertion, it does not mean the dataset
is shuffled.
Example usage:
```python
try:
assert_not_shuffled(dataset)
# safe to assume `dataset` it not shuffled here
expect ValueError:
# make no assumptions about `dataset`
```
Args:
dataset: The dataset to analyze.
Raises:
ValueError: If the method cannot establish the assertion."
6002,verify_dataset_shuffled,tensorflow/tensorflow/python/keras/engine/training_utils.py,1682,function,"Verifies that the dataset is shuffled.
Args:
x: Dataset passed as an input to the model.
Raises:
ValueError: if the dataset is not already shuffled."
6003,is_dataset_or_iterator,tensorflow/tensorflow/python/keras/engine/training_utils.py,1702,function,
6004,get_iterator,tensorflow/tensorflow/python/keras/engine/training_utils.py,1707,function,Create and initialize an iterator from a dataset.
6005,initialize_iterator,tensorflow/tensorflow/python/keras/engine/training_utils.py,1717,function,
6006,extract_tensors_from_dataset,tensorflow/tensorflow/python/keras/engine/training_utils.py,1723,function,"Extract a tuple of tensors `inputs, targets, sample_weight` from a dataset.
Arguments:
dataset: Dataset instance.
Returns:
Tuple of tensors `x, y, weights`. `y` and `weights` entry may be None."
6007,unpack_iterator_input,tensorflow/tensorflow/python/keras/engine/training_utils.py,1737,function,"Convert a dataset iterator to a tuple of tensors `x, y, sample_weights`.
Arguments:
iterator: Instance of a dataset iterator.
Returns:
Tuple of tensors `x, y, weights`. `y` and `weights` entry may be None."
6008,infer_steps_for_dataset,tensorflow/tensorflow/python/keras/engine/training_utils.py,1771,function,"Infers steps_per_epoch needed to loop through a dataset.
Arguments:
model: Keras model instance.
dataset: Input data of type tf.data.Dataset.
steps: Number of steps to draw from the dataset (may be None if unknown).
epochs: Number of times to iterate over the dataset.
steps_name: The string name of the steps argument, either `steps`,
`validation_steps`, or `steps_per_epoch`. Only used for error message
formatting.
Returns:
Integer or `None`. Inferred number of steps to loop through the dataset.
`None` is returned if 1) the size of the dataset is unknown and `steps` was
not specified, or 2) this is multi-worker training and auto sharding is
enabled.
Raises:
ValueError: In case of invalid argument values."
6009,ModelInputs,tensorflow/tensorflow/python/keras/engine/training_utils.py,1829,class,"Encapsulates model inputs.
Allows for transforming model inputs while keeping the same structure."
6010,get_input_names,tensorflow/tensorflow/python/keras/engine/training_utils.py,1853,method,"Returns keys to name inputs by.
In case inputs provided were a list, tuple or single entry, we make up a
key 'input_%d'. For dictionary case, we return a sorted list of keys."
6011,get_symbolic_inputs,tensorflow/tensorflow/python/keras/engine/training_utils.py,1861,method,Returns inputs to be set as self.inputs for a model.
6012,as_dict,tensorflow/tensorflow/python/keras/engine/training_utils.py,1898,method,An iterable over a dictionary version of inputs.
6013,as_list,tensorflow/tensorflow/python/keras/engine/training_utils.py,1903,method,Returning the inputs as a list.
6014,get_input_shape_and_dtype,tensorflow/tensorflow/python/keras/engine/training_utils.py,1910,function,"Retrieves input shape and input dtype of layer if applicable.
Args:
layer: Layer (or model) instance.
Returns:
Tuple (input_shape, input_dtype). Both could be None if the layer
does not have a defined input shape.
Raises:
ValueError: in case an empty Sequential or Functional model is passed."
6015,get_static_batch_size,tensorflow/tensorflow/python/keras/engine/training_utils.py,1944,function,"Gets the static batch size of a Layer.
Arguments:
layer: a `Layer` instance.
Returns:
The static batch size of a Layer."
6016,generic_output_names,tensorflow/tensorflow/python/keras/engine/training_utils.py,1959,function,
6017,convert_eager_tensors_to_numpy,tensorflow/tensorflow/python/keras/engine/training_utils.py,1963,function,"Convert every EagerTensor in `structure` to NumPy.
Arguments:
structure: An arbitrary structure of elements to be converted to NumPy
arrays.
Returns:
An identical structure with EagerTensors converted to NumPy arrays."
6018,list_to_tuple,tensorflow/tensorflow/python/keras/engine/training_utils.py,1982,function,"Datasets will stack the list of tensor, so switch them to tuples."
6019,should_run_validation,tensorflow/tensorflow/python/keras/engine/training_utils.py,1989,function,"Checks if validation should be run this epoch.
Arguments:
validation_freq: Integer or list. If an integer, specifies how many training
epochs to run before a new validation run is performed. If a list,
specifies the epochs on which to run validation.
epoch: Integer, the number of the training epoch just completed.
Returns:
Bool, True if validation should be run.
Raises:
ValueError: if `validation_freq` is an Integer and less than 1, or if
it is neither an Integer nor a Sequence."
6020,split_training_and_validation_data,tensorflow/tensorflow/python/keras/engine/training_utils.py,2019,function,Split input data into train/eval section based on validation_split.
6021,unpack_validation_data,tensorflow/tensorflow/python/keras/engine/training_utils.py,2042,function,"Unpack validation data based input type.
The validation data is not touched if its dataset or dataset iterator.
For other type of input (Numpy or tensor), it will be unpacked into tuple of
3 which is x, y and sample weights.
Args:
validation_data: dataset, dataset iterator, or numpy, tensor tuple.
raise_if_ambiguous: boolean on whether to fail if validation_data cannot be
parsed. Otherwise simply return validation_data, None, None and defer the
decision to the caller.
Returns:
tuple of 3, (x, y, sample_weights) for numpy and tensor input."
6022,TrainingLoop,tensorflow/tensorflow/python/keras/engine/training_utils.py,2090,class,"TrainingLoop is a wrapper class around the training logic.
This class is trying to encapsulate the different logic of fit/eval/predict
with regard to different data input and model condition.
Note that TrainingLoop is stateless, which means it doesn't contain any
internal field and can be reused with different model and inputs."
6023,fit,tensorflow/tensorflow/python/keras/engine/training_utils.py,2100,method,Train the model with the inputs and targets.
6024,evaluate,tensorflow/tensorflow/python/keras/engine/training_utils.py,2121,method,Returns the loss value & metrics values for the model in test mode.
6025,predict,tensorflow/tensorflow/python/keras/engine/training_utils.py,2134,method,
6026,MonitoredPool,tensorflow/tensorflow/python/keras/engine/training_utils_test.py,287,class,
6027,apply_async,tensorflow/tensorflow/python/keras/engine/training_utils_test.py,294,method,
6028,add_sleep,tensorflow/tensorflow/python/keras/engine/training_utils_test.py,301,function,
6029,cause_error,tensorflow/tensorflow/python/keras/engine/training_utils_test.py,309,function,
6030,Model,tensorflow/tensorflow/python/keras/engine/training_v1.py,79,class,"`Model` groups layers into an object with training and inference features.
There are two ways to instantiate a `Model`:
1 - With the ""functional API"", where you start from `Input`,
you chain layer calls to specify the model's forward pass,
and finally you create your model from inputs and outputs:
```python
import tensorflow as tf
inputs = tf.keras.Input(shape=(3,))
x = tf.keras.layers.Dense(4, activation=tf.nn.relu)(inputs)
outputs = tf.keras.layers.Dense(5, activation=tf.nn.softmax)(x)
model = tf.keras.Model(inputs=inputs, outputs=outputs)
```
2 - By subclassing the `Model` class: in that case, you should define your
layers in `__init__` and you should implement the model's forward pass
in `call`.
```python
import tensorflow as tf
class MyModel(tf.keras.Model):
def __init__(self):
super(MyModel, self).__init__()
self.dense1 = tf.keras.layers.Dense(4, activation=tf.nn.relu)
self.dense2 = tf.keras.layers.Dense(5, activation=tf.nn.softmax)
def call(self, inputs):
x = self.dense1(inputs)
return self.dense2(x)
model = MyModel()
```
If you subclass `Model`, you can optionally have
a `training` argument (boolean) in `call`, which you can use to specify
a different behavior in training and inference:
```python
import tensorflow as tf
class MyModel(tf.keras.Model):
def __init__(self):
super(MyModel, self).__init__()
self.dense1 = tf.keras.layers.Dense(4, activation=tf.nn.relu)
self.dense2 = tf.keras.layers.Dense(5, activation=tf.nn.softmax)
self.dropout = tf.keras.layers.Dropout(0.5)
def call(self, inputs, training=False):
x = self.dense1(inputs)
if training:
x = self.dropout(x, training=training)
return self.dense2(x)
model = MyModel()
```"
6031,get_weights,tensorflow/tensorflow/python/keras/engine/training_v1.py,173,method,"Retrieves the weights of the model.
Returns:
A flat list of Numpy arrays."
6032,load_weights,tensorflow/tensorflow/python/keras/engine/training_v1.py,186,method,"Loads all layer weights, either from a TensorFlow or an HDF5 weight file.
If `by_name` is False weights are loaded based on the network's
topology. This means the architecture should be the same as when the weights
were saved. Note that layers that don't have weights are not taken into
account in the topological ordering, so adding or removing layers is fine as
long as they don't have weights.
If `by_name` is True, weights are loaded into layers only if they share the
same name. This is useful for fine-tuning or transfer-learning models where
some of the layers have changed.
Only topological loading (`by_name=False`) is supported when loading weights
from the TensorFlow format. Note that topological loading differs slightly
between TensorFlow and HDF5 formats for user-defined classes inheriting from
`tf.keras.Model`: HDF5 loads based on a flattened list of weights, while the
TensorFlow format loads based on the object-local names of attributes to
which layers are assigned in the `Model`'s constructor.
Arguments:
filepath: String, path to the weights file to load. For weight files in
TensorFlow format, this is the file prefix (the same as was passed
to `save_weights`).
by_name: Boolean, whether to load weights by name or by topological
order. Only topological loading is supported for weight files in
TensorFlow format.
skip_mismatch: Boolean, whether to skip loading of layers where there is
a mismatch in the number of weights, or a mismatch in the shape of
the weight (only valid when `by_name=True`).
Returns:
When loading a weight file in TensorFlow format, returns the same status
object as `tf.train.Checkpoint.restore`. When graph building, restore
ops are run automatically as soon as the network is built (on first call
for user-defined classes inheriting from `Model`, immediately if it is
already built).
When loading weights in HDF5 format, returns `None`.
Raises:
ImportError: If h5py is not available and the weight file is in HDF5
format.
ValueError: If `skip_mismatch` is set to `True` when `by_name` is
`False`."
6033,compile,tensorflow/tensorflow/python/keras/engine/training_v1.py,240,method,"Configures the model for training.
Arguments:
optimizer: String (name of optimizer) or optimizer instance.
See `tf.keras.optimizers`.
loss: String (name of objective function), objective function or
`tf.keras.losses.Loss` instance. See `tf.keras.losses`. An objective
function is any callable with the signature
`scalar_loss = fn(y_true, y_pred)`. If the model has multiple
outputs, you can use a different loss on each output by passing a
dictionary or a list of losses. The loss value that will be
minimized by the model will then be the sum of all individual
losses.
metrics: List of metrics to be evaluated by the model during training
and testing. Typically you will use `metrics=['accuracy']`.
To specify different metrics for different outputs of a
multi-output model, you could also pass a dictionary, such as
`metrics={'output_a': 'accuracy', 'output_b': ['accuracy', 'mse']}`.
You can also pass a list (len = len(outputs)) of lists of metrics
such as `metrics=[['accuracy'], ['accuracy', 'mse']]` or
`metrics=['accuracy', ['accuracy', 'mse']]`.
loss_weights: Optional list or dictionary specifying scalar
coefficients (Python floats) to weight the loss contributions
of different model outputs.
The loss value that will be minimized by the model
will then be the *weighted sum* of all individual losses,
weighted by the `loss_weights` coefficients.
If a list, it is expected to have a 1:1 mapping
to the model's outputs. If a tensor, it is expected to map
output names (strings) to scalar coefficients.
sample_weight_mode: If you need to do timestep-wise
sample weighting (2D weights), set this to `""temporal""`.
`None` defaults to sample-wise weights (1D).
If the model has multiple outputs, you can use a different
`sample_weight_mode` on each output by passing a
dictionary or a list of modes.
weighted_metrics: List of metrics to be evaluated and weighted
by sample_weight or class_weight during training and testing.
target_tensors: By default, Keras will create placeholders for the
model's target, which will be fed with the target data during
training. If instead you would like to use your own
target tensors (in turn, Keras will not expect external
Numpy data for these targets at training time), you
can specify them via the `target_tensors` argument. It can be
a single tensor (for a single-output model), a list of tensors,
or a dict mapping output names to target tensors.
distribute: NOT SUPPORTED IN TF 2.0, please create and compile the
model under distribution strategy scope instead of passing it to
compile.
**kwargs: Any additional arguments.
Raises:
ValueError: In case of invalid arguments for
`optimizer`, `loss`, `metrics` or `sample_weight_mode`."
6034,metrics,tensorflow/tensorflow/python/keras/engine/training_v1.py,490,method,"Returns the model's metrics added using `compile`, `add_metric` APIs."
6035,metrics_names,tensorflow/tensorflow/python/keras/engine/training_v1.py,505,method,Returns the model's display labels for all outputs.
6036,run_eagerly,tensorflow/tensorflow/python/keras/engine/training_v1.py,531,method,"Settable attribute indicating whether the model should run eagerly.
Running eagerly means that your model will be run step by step,
like Python code. Your model might run slower, but it should become easier
for you to debug it by stepping into individual layer calls.
By default, we will attempt to compile your model to a static graph to
deliver the best execution performance.
Returns:
Boolean, whether the model should run eagerly."
6037,run_eagerly,tensorflow/tensorflow/python/keras/engine/training_v1.py,570,method,
6038,fit,tensorflow/tensorflow/python/keras/engine/training_v1.py,610,method,"Trains the model for a fixed number of epochs (iterations on a dataset).
Arguments:
x: Input data. It could be:
- A Numpy array (or array-like), or a list of arrays
(in case the model has multiple inputs).
- A TensorFlow tensor, or a list of tensors
(in case the model has multiple inputs).
- A dict mapping input names to the corresponding array/tensors,
if the model has named inputs.
- A `tf.data` dataset. Should return a tuple
of either `(inputs, targets)` or
`(inputs, targets, sample_weights)`.
- A generator or `keras.utils.Sequence` returning `(inputs, targets)`
or `(inputs, targets, sample weights)`.
y: Target data. Like the input data `x`,
it could be either Numpy array(s) or TensorFlow tensor(s).
It should be consistent with `x` (you cannot have Numpy inputs and
tensor targets, or inversely). If `x` is a dataset, generator,
or `keras.utils.Sequence` instance, `y` should
not be specified (since targets will be obtained from `x`).
batch_size: Integer or `None`.
Number of samples per gradient update.
If unspecified, `batch_size` will default to 32.
Do not specify the `batch_size` if your data is in the
form of symbolic tensors, datasets,
generators, or `keras.utils.Sequence` instances (since they generate
batches).
epochs: Integer. Number of epochs to train the model.
An epoch is an iteration over the entire `x` and `y`
data provided.
Note that in conjunction with `initial_epoch`,
`epochs` is to be understood as ""final epoch"".
The model is not trained for a number of iterations
given by `epochs`, but merely until the epoch
of index `epochs` is reached.
verbose: 0, 1, or 2. Verbosity mode.
0 = silent, 1 = progress bar, 2 = one line per epoch.
Note that the progress bar is not particularly useful when
logged to a file, so verbose=2 is recommended when not running
interactively (eg, in a production environment).
callbacks: List of `keras.callbacks.Callback` instances.
List of callbacks to apply during training.
See `tf.keras.callbacks`.
validation_split: Float between 0 and 1.
Fraction of the training data to be used as validation data.
The model will set apart this fraction of the training data,
will not train on it, and will evaluate
the loss and any model metrics
on this data at the end of each epoch.
The validation data is selected from the last samples
in the `x` and `y` data provided, before shuffling. This argument is
not supported when `x` is a dataset, generator or
`keras.utils.Sequence` instance.
validation_data: Data on which to evaluate
the loss and any model metrics at the end of each epoch.
The model will not be trained on this data.
`validation_data` will override `validation_split`.
`validation_data` could be:
- tuple `(x_val, y_val)` of Numpy arrays or tensors
- tuple `(x_val, y_val, val_sample_weights)` of Numpy arrays
- dataset
For the first two cases, `batch_size` must be provided.
For the last case, `validation_steps` could be provided.
shuffle: Boolean (whether to shuffle the training data
before each epoch) or str (for 'batch').
'batch' is a special option for dealing with the
limitations of HDF5 data; it shuffles in batch-sized chunks.
Has no effect when `steps_per_epoch` is not `None`.
class_weight: Optional dictionary mapping class indices (integers)
to a weight (float) value, used for weighting the loss function
(during training only).
This can be useful to tell the model to
""pay more attention"" to samples from
an under-represented class.
sample_weight: Optional Numpy array of weights for
the training samples, used for weighting the loss function
(during training only). You can either pass a flat (1D)
Numpy array with the same length as the input samples
(1:1 mapping between weights and samples),
or in the case of temporal data,
you can pass a 2D array with shape
`(samples, sequence_length)`,
to apply a different weight to every timestep of every sample.
In this case you should make sure to specify
`sample_weight_mode=""temporal""` in `compile()`. This argument is not
supported when `x` is a dataset, generator, or
`keras.utils.Sequence` instance, instead provide the sample_weights
as the third element of `x`.
initial_epoch: Integer.
Epoch at which to start training
(useful for resuming a previous training run).
steps_per_epoch: Integer or `None`.
Total number of steps (batches of samples)
before declaring one epoch finished and starting the
next epoch. When training with input tensors such as
TensorFlow data tensors, the default `None` is equal to
the number of samples in your dataset divided by
the batch size, or 1 if that cannot be determined. If x is a
`tf.data` dataset, and 'steps_per_epoch'
is None, the epoch will run until the input dataset is exhausted.
This argument is not supported with array inputs.
validation_steps: Only relevant if `validation_data` is provided and
is a `tf.data` dataset. Total number of steps (batches of
samples) to draw before stopping when performing validation
at the end of every epoch. If 'validation_steps' is None, validation
will run until the `validation_data` dataset is exhausted. In the
case of a infinite dataset, it will run into a infinite loop.
If 'validation_steps' is specified and only part of the dataset
will be consumed, the evaluation will start from the beginning of
the dataset at each epoch. This ensures that the same validation
samples are used every time.
validation_freq: Only relevant if validation data is provided. Integer
or `collections_abc.Container` instance (e.g. list, tuple, etc.).
If an integer, specifies how many training epochs to run before a
new validation run is performed, e.g. `validation_freq=2` runs
validation every 2 epochs. If a Container, specifies the epochs on
which to run validation, e.g. `validation_freq=[1, 2, 10]` runs
validation at the end of the 1st, 2nd, and 10th epochs.
max_queue_size: Integer. Used for generator or `keras.utils.Sequence`
input only. Maximum size for the generator queue.
If unspecified, `max_queue_size` will default to 10.
workers: Integer. Used for generator or `keras.utils.Sequence` input
only. Maximum number of processes to spin up
when using process-based threading. If unspecified, `workers`
will default to 1. If 0, will execute the generator on the main
thread.
use_multiprocessing: Boolean. Used for generator or
`keras.utils.Sequence` input only. If `True`, use process-based
threading. If unspecified, `use_multiprocessing` will default to
`False`. Note that because this implementation relies on
multiprocessing, you should not pass non-picklable arguments to
the generator as they can't be passed easily to children processes.
**kwargs: Used for backwards compatibility.
Returns:
A `History` object. Its `History.history` attribute is
a record of training loss values and metrics values
at successive epochs, as well as validation loss values
and validation metrics values (if applicable).
Raises:
RuntimeError: If the model was never compiled.
ValueError: In case of mismatch between the provided input data
and what the model expects."
6039,evaluate,tensorflow/tensorflow/python/keras/engine/training_v1.py,810,method,"Returns the loss value & metrics values for the model in test mode.
Computation is done in batches (see the `batch_size` arg.)
Arguments:
x: Input data. It could be:
- A Numpy array (or array-like), or a list of arrays
(in case the model has multiple inputs).
- A TensorFlow tensor, or a list of tensors
(in case the model has multiple inputs).
- A dict mapping input names to the corresponding array/tensors,
if the model has named inputs.
- A `tf.data` dataset.
- A generator or `keras.utils.Sequence` instance.
y: Target data. Like the input data `x`,
it could be either Numpy array(s) or TensorFlow tensor(s).
It should be consistent with `x` (you cannot have Numpy inputs and
tensor targets, or inversely).
If `x` is a dataset, generator or
`keras.utils.Sequence` instance, `y` should not be specified (since
targets will be obtained from the iterator/dataset).
batch_size: Integer or `None`.
Number of samples per batch of computation.
If unspecified, `batch_size` will default to 32.
Do not specify the `batch_size` if your data is in the
form of symbolic tensors, dataset,
generators, or `keras.utils.Sequence` instances (since they generate
batches).
verbose: 0 or 1. Verbosity mode.
0 = silent, 1 = progress bar.
sample_weight: Optional Numpy array of weights for
the test samples, used for weighting the loss function.
You can either pass a flat (1D)
Numpy array with the same length as the input samples
(1:1 mapping between weights and samples),
or in the case of temporal data,
you can pass a 2D array with shape
`(samples, sequence_length)`,
to apply a different weight to every timestep of every sample.
In this case you should make sure to specify
`sample_weight_mode=""temporal""` in `compile()`. This argument is not
supported when `x` is a dataset, instead pass
sample weights as the third element of `x`.
steps: Integer or `None`.
Total number of steps (batches of samples)
before declaring the evaluation round finished.
Ignored with the default value of `None`.
If x is a `tf.data` dataset and `steps` is
None, 'evaluate' will run until the dataset is exhausted.
This argument is not supported with array inputs.
callbacks: List of `keras.callbacks.Callback` instances.
List of callbacks to apply during evaluation.
See [callbacks](/api_docs/python/tf/keras/callbacks).
max_queue_size: Integer. Used for generator or `keras.utils.Sequence`
input only. Maximum size for the generator queue.
If unspecified, `max_queue_size` will default to 10.
workers: Integer. Used for generator or `keras.utils.Sequence` input
only. Maximum number of processes to spin up when using
process-based threading. If unspecified, `workers` will default
to 1. If 0, will execute the generator on the main thread.
use_multiprocessing: Boolean. Used for generator or
`keras.utils.Sequence` input only. If `True`, use process-based
threading. If unspecified, `use_multiprocessing` will default to
`False`. Note that because this implementation relies on
multiprocessing, you should not pass non-picklable arguments to
the generator as they can't be passed easily to children processes.
Returns:
Scalar test loss (if the model has a single output and no metrics)
or list of scalars (if the model has multiple outputs
and/or metrics). The attribute `model.metrics_names` will give you
the display labels for the scalar outputs.
Raises:
ValueError: in case of invalid arguments."
6040,predict,tensorflow/tensorflow/python/keras/engine/training_v1.py,916,method,"Generates output predictions for the input samples.
Computation is done in batches (see the `batch_size` arg.)
Arguments:
x: Input samples. It could be:
- A Numpy array (or array-like), or a list of arrays
(in case the model has multiple inputs).
- A TensorFlow tensor, or a list of tensors
(in case the model has multiple inputs).
- A `tf.data` dataset.
- A generator or `keras.utils.Sequence` instance.
batch_size: Integer or `None`.
Number of samples per batch of computation.
If unspecified, `batch_size` will default to 32.
Do not specify the `batch_size` if your data is in the
form of symbolic tensors, dataset,
generators, or `keras.utils.Sequence` instances (since they generate
batches).
verbose: Verbosity mode, 0 or 1.
steps: Total number of steps (batches of samples)
before declaring the prediction round finished.
Ignored with the default value of `None`. If x is a `tf.data`
dataset and `steps` is None, `predict` will
run until the input dataset is exhausted.
callbacks: List of `keras.callbacks.Callback` instances.
List of callbacks to apply during prediction.
See [callbacks](/api_docs/python/tf/keras/callbacks).
max_queue_size: Integer. Used for generator or `keras.utils.Sequence`
input only. Maximum size for the generator queue.
If unspecified, `max_queue_size` will default to 10.
workers: Integer. Used for generator or `keras.utils.Sequence` input
only. Maximum number of processes to spin up when using
process-based threading. If unspecified, `workers` will default
to 1. If 0, will execute the generator on the main thread.
use_multiprocessing: Boolean. Used for generator or
`keras.utils.Sequence` input only. If `True`, use process-based
threading. If unspecified, `use_multiprocessing` will default to
`False`. Note that because this implementation relies on
multiprocessing, you should not pass non-picklable arguments to
the generator as they can't be passed easily to children processes.
Returns:
Numpy array(s) of predictions.
Raises:
ValueError: In case of mismatch between the provided
input data and the model's expectations,
or in case a stateful model receives a number of samples
that is not a multiple of the batch size."
6041,reset_metrics,tensorflow/tensorflow/python/keras/engine/training_v1.py,993,method,Resets the state of metrics.
6042,train_on_batch,tensorflow/tensorflow/python/keras/engine/training_v1.py,1003,method,"Runs a single gradient update on a single batch of data.
Arguments:
x: Input data. It could be:
- A Numpy array (or array-like), or a list of arrays
(in case the model has multiple inputs).
- A TensorFlow tensor, or a list of tensors
(in case the model has multiple inputs).
- A dict mapping input names to the corresponding array/tensors,
if the model has named inputs.
- A `tf.data` dataset.
y: Target data. Like the input data `x`, it could be either Numpy
array(s) or TensorFlow tensor(s). It should be consistent with `x`
(you cannot have Numpy inputs and tensor targets, or inversely). If
`x` is a dataset, `y` should not be specified
(since targets will be obtained from the iterator).
sample_weight: Optional array of the same length as x, containing
weights to apply to the model's loss for each sample. In the case of
temporal data, you can pass a 2D array with shape (samples,
sequence_length), to apply a different weight to every timestep of
every sample. In this case you should make sure to specify
sample_weight_mode=""temporal"" in compile(). This argument is not
supported when `x` is a dataset.
class_weight: Optional dictionary mapping class indices (integers) to a
weight (float) to apply to the model's loss for the samples from this
class during training. This can be useful to tell the model to ""pay
more attention"" to samples from an under-represented class.
reset_metrics: If `True`, the metrics returned will be only for this
batch. If `False`, the metrics will be statefully accumulated across
batches.
Returns:
Scalar training loss
(if the model has a single output and no metrics)
or list of scalars (if the model has multiple outputs
and/or metrics). The attribute `model.metrics_names` will give you
the display labels for the scalar outputs.
Raises:
ValueError: In case of invalid user-provided arguments."
6043,predict_on_batch,tensorflow/tensorflow/python/keras/engine/training_v1.py,1173,method,"Returns predictions for a single batch of samples.
Arguments:
x: Input data. It could be:
- A Numpy array (or array-like), or a list of arrays
(in case the model has multiple inputs).
- A TensorFlow tensor, or a list of tensors
(in case the model has multiple inputs).
- A `tf.data` dataset.
Returns:
Numpy array(s) of predictions.
Raises:
ValueError: In case of mismatch between given number of inputs and
expectations of the model."
6044,fit_generator,tensorflow/tensorflow/python/keras/engine/training_v1.py,1221,method,"Fits the model on data yielded batch-by-batch by a Python generator.
DEPRECATED:
`Model.fit` now supports generators, so there is no longer any need to use
this endpoint."
6045,evaluate_generator,tensorflow/tensorflow/python/keras/engine/training_v1.py,1260,method,"Evaluates the model on a data generator.
DEPRECATED:
`Model.evaluate` now supports generators, so there is no longer any need
to use this endpoint."
6046,predict_generator,tensorflow/tensorflow/python/keras/engine/training_v1.py,1287,method,"Generates predictions for the input samples from a data generator.
DEPRECATED:
`Model.predict` now supports generators, so there is no longer any need
to use this endpoint."
6047,sample_weights,tensorflow/tensorflow/python/keras/engine/training_v1.py,2730,method,
6048,create_tensor_spec,tensorflow/tensorflow/python/keras/engine/training_v1.py,2514,method,
6049,DistributedCallbackModel,tensorflow/tensorflow/python/keras/engine/training_v1.py,2834,class,Model that is used for callbacks with tf.distribute.Strategy.
6050,set_original_model,tensorflow/tensorflow/python/keras/engine/training_v1.py,2841,method,
6051,save_weights,tensorflow/tensorflow/python/keras/engine/training_v1.py,2844,method,
6052,save,tensorflow/tensorflow/python/keras/engine/training_v1.py,2848,method,
6053,load_weights,tensorflow/tensorflow/python/keras/engine/training_v1.py,2856,method,
6054,model_to_estimator,tensorflow/tensorflow/python/keras/estimator/__init__.py,35,function,"Constructs an `Estimator` instance from given keras model.
If you use infrastructure or other tooling that relies on Estimators, you can
still build a Keras model and use model_to_estimator to convert the Keras
model to an Estimator for use with downstream systems.
For usage example, please see:
[Creating estimators from Keras Models](
https://www.tensorflow.org/guide/estimators#creating_estimators_from_keras_models).
Sample Weights:
Estimators returned by `model_to_estimator` are configured so that they can
handle sample weights (similar to `keras_model.fit(x, y, sample_weights)`).
To pass sample weights when training or evaluating the Estimator, the first
item returned by the input function should be a dictionary with keys
`features` and `sample_weights`. Example below:
```python
keras_model = tf.keras.Model(...)
keras_model.compile(...)
estimator = tf.keras.estimator.model_to_estimator(keras_model)
def input_fn():
return dataset_ops.Dataset.from_tensors(
({'features': features, 'sample_weights': sample_weights},
targets))
estimator.train(input_fn, steps=1)
```
Args:
keras_model: A compiled Keras model object. This argument is mutually
exclusive with `keras_model_path`. Estimator's `model_fn` uses the
structure of the model to clone the model. Defaults to `None`.
keras_model_path: Path to a compiled Keras model saved on disk, in HDF5
format, which can be generated with the `save()` method of a Keras model.
This argument is mutually exclusive with `keras_model`.
Defaults to `None`.
custom_objects: Dictionary for cloning customized objects. This is
used with classes that is not part of this pip package. For example, if
user maintains a `relu6` class that inherits from `tf.keras.layers.Layer`,
then pass `custom_objects={'relu6': relu6}`. Defaults to `None`.
model_dir: Directory to save `Estimator` model parameters, graph, summary
files for TensorBoard, etc. If unset a directory will be created with
`tempfile.mkdtemp`
config: `RunConfig` to config `Estimator`. Allows setting up things in
`model_fn` based on configuration such as `num_ps_replicas`, or
`model_dir`. Defaults to `None`. If both `config.model_dir` and the
`model_dir` argument (above) are specified the `model_dir` **argument**
takes precedence.
checkpoint_format: Sets the format of the checkpoint saved by the estimator
when training. May be `saver` or `checkpoint`, depending on whether to
save checkpoints from `tf.train.Saver` or `tf.train.Checkpoint`. This
argument currently defaults to `saver`. When 2.0 is released, the default
will be `checkpoint`. Estimators use name-based `tf.train.Saver`
checkpoints, while Keras models use object-based checkpoints from
`tf.train.Checkpoint`. Currently, saving object-based checkpoints from
`model_to_estimator` is only supported by Functional and Sequential
models. Defaults to 'saver'.
Returns:
An Estimator from given keras model.
Raises:
ValueError: If neither keras_model nor keras_model_path was given.
ValueError: If both keras_model and keras_model_path was given.
ValueError: If the keras_model_path is a GCS URI.
ValueError: If keras_model has not been compiled.
ValueError: If an invalid checkpoint_format was given."
6055,model_to_estimator_v2,tensorflow/tensorflow/python/keras/estimator/__init__.py,133,function,"Constructs an `Estimator` instance from given keras model.
If you use infrastructure or other tooling that relies on Estimators, you can
still build a Keras model and use model_to_estimator to convert the Keras
model to an Estimator for use with downstream systems.
For usage example, please see:
[Creating estimators from Keras Models](
https://www.tensorflow.org/guide/estimators#creating_estimators_from_keras_models).
Sample Weights:
Estimators returned by `model_to_estimator` are configured so that they can
handle sample weights (similar to `keras_model.fit(x, y, sample_weights)`).
To pass sample weights when training or evaluating the Estimator, the first
item returned by the input function should be a dictionary with keys
`features` and `sample_weights`. Example below:
```python
keras_model = tf.keras.Model(...)
keras_model.compile(...)
estimator = tf.keras.estimator.model_to_estimator(keras_model)
def input_fn():
return dataset_ops.Dataset.from_tensors(
({'features': features, 'sample_weights': sample_weights},
targets))
estimator.train(input_fn, steps=1)
```
Note: We do not support creating weighted metrics in Keras and converting them
to weighted metrics in the Estimator API using `model_to_estimator`.
You will have to create these metrics directly on the estimator spec using the
`add_metrics` function.
To customize the estimator `eval_metric_ops` names, you can pass in the
`metric_names_map` dictionary mapping the keras model output metric names
to the custom names as follows:
```python
input_a = tf.keras.layers.Input(shape=(16,), name='input_a')
input_b = tf.keras.layers.Input(shape=(16,), name='input_b')
dense = tf.keras.layers.Dense(8, name='dense_1')
interm_a = dense(input_a)
interm_b = dense(input_b)
merged = tf.keras.layers.concatenate([interm_a, interm_b], name='merge')
output_a = tf.keras.layers.Dense(3, activation='softmax', name='dense_2')(
merged)
output_b = tf.keras.layers.Dense(2, activation='softmax', name='dense_3')(
merged)
keras_model = tf.keras.models.Model(
inputs=[input_a, input_b], outputs=[output_a, output_b])
keras_model.compile(
loss='categorical_crossentropy',
optimizer='rmsprop',
metrics={
'dense_2': 'categorical_accuracy',
'dense_3': 'categorical_accuracy'
})
metric_names_map = {
'dense_2_categorical_accuracy': 'acc_1',
'dense_3_categorical_accuracy': 'acc_2',
}
keras_est = tf.keras.estimator.model_to_estimator(
keras_model=keras_model,
config=config,
metric_names_map=metric_names_map)
```
Args:
keras_model: A compiled Keras model object. This argument is mutually
exclusive with `keras_model_path`. Estimator's `model_fn` uses the
structure of the model to clone the model. Defaults to `None`.
keras_model_path: Path to a compiled Keras model saved on disk, in HDF5
format, which can be generated with the `save()` method of a Keras model.
This argument is mutually exclusive with `keras_model`.
Defaults to `None`.
custom_objects: Dictionary for cloning customized objects. This is
used with classes that is not part of this pip package. For example, if
user maintains a `relu6` class that inherits from `tf.keras.layers.Layer`,
then pass `custom_objects={'relu6': relu6}`. Defaults to `None`.
model_dir: Directory to save `Estimator` model parameters, graph, summary
files for TensorBoard, etc. If unset a directory will be created with
`tempfile.mkdtemp`
config: `RunConfig` to config `Estimator`. Allows setting up things in
`model_fn` based on configuration such as `num_ps_replicas`, or
`model_dir`. Defaults to `None`. If both `config.model_dir` and the
`model_dir` argument (above) are specified the `model_dir` **argument**
takes precedence.
checkpoint_format: Sets the format of the checkpoint saved by the estimator
when training. May be `saver` or `checkpoint`, depending on whether to
save checkpoints from `tf.compat.v1.train.Saver` or `tf.train.Checkpoint`.
The default is `checkpoint`. Estimators use name-based `tf.train.Saver`
checkpoints, while Keras models use object-based checkpoints from
`tf.train.Checkpoint`. Currently, saving object-based checkpoints from
`model_to_estimator` is only supported by Functional and Sequential
models. Defaults to 'checkpoint'.
metric_names_map: Optional dictionary mapping Keras model output metric
names to custom names. This can be used to override the default Keras
model output metrics names in a multi IO model use case and provide custom
names for the `eval_metric_ops` in Estimator.
The Keras model metric names can be obtained using `model.metrics_names`
excluding any loss metrics such as total loss and output losses.
For example, if your Keras model has two outputs `out_1` and `out_2`,
with `mse` loss and `acc` metric, then `model.metrics_names` will be
`['loss', 'out_1_loss', 'out_2_loss', 'out_1_acc', 'out_2_acc']`.
The model metric names excluding the loss metrics will be
`['out_1_acc', 'out_2_acc']`.
Returns:
An Estimator from given keras model.
Raises:
ValueError: If neither keras_model nor keras_model_path was given.
ValueError: If both keras_model and keras_model_path was given.
ValueError: If the keras_model_path is a GCS URI.
ValueError: If keras_model has not been compiled.
ValueError: If an invalid checkpoint_format was given."
6056,DenseFeatures,tensorflow/tensorflow/python/keras/feature_column/dense_features.py,31,class,"A layer that produces a dense `Tensor` based on given `feature_columns`.
Generally a single example in training data is described with FeatureColumns.
At the first layer of the model, this column-oriented data should be converted
to a single `Tensor`.
This layer can be called multiple times with different features.
This is the V1 version of this layer that uses variable_scope's or partitioner
to create variables which works well with PartitionedVariables. Variable
scopes are deprecated in V2, so the V2 version uses name_scopes instead. But
currently that lacks support for partitioned variables. Use this if you need
partitioned variables. Use the partitioner argument if you have a Keras model
and uses `tf.compat.v1.keras.estimator.model_to_estimator` for training.
Example:
```python
price = tf.feature_column.numeric_column('price')
keywords_embedded = tf.feature_column.embedding_column(
tf.feature_column.categorical_column_with_hash_bucket(""keywords"", 10K),
dimension=16)
columns = [price, keywords_embedded, ...]
partitioner = tf.compat.v1.fixed_size_partitioner(num_shards=4)
feature_layer = tf.compat.v1.keras.layers.DenseFeatures(
feature_columns=columns, partitioner=partitioner)
features = tf.io.parse_example(
..., features=tf.feature_column.make_parse_example_spec(columns))
dense_tensor = feature_layer(features)
for units in [128, 64, 32]:
dense_tensor = tf.compat.v1.keras.layers.Dense(
units, activation='relu')(dense_tensor)
prediction = tf.compat.v1.keras.layers.Dense(1)(dense_tensor)
```"
6057,call,tensorflow/tensorflow/python/keras/feature_column/dense_features.py,119,method,"Returns a dense tensor corresponding to the `feature_columns`.
Example usage:
>>> t1 = tf.feature_column.embedding_column(
... tf.feature_column.categorical_column_with_hash_bucket(""t1"", 2),
... dimension=8)
>>> t2 = tf.feature_column.numeric_column('t2')
>>> feature_layer = tf.compat.v1.keras.layers.DenseFeatures([t1, t2])
>>> features = {""t1"": tf.constant([""a"", ""b""]), ""t2"": tf.constant([1, 2])}
>>> dense_tensor = feature_layer(features, training=True)
Args:
features: A mapping from key to tensors. `FeatureColumn`s look up via
these keys. For example `numeric_column('price')` will look at 'price'
key in this dict. Values can be a `SparseTensor` or a `Tensor` depends
on corresponding `FeatureColumn`.
cols_to_output_tensors: If not `None`, this will be filled with a dict
mapping feature columns to output tensors created.
training: Python boolean or None, indicating whether to the layer is being
run in training mode. This argument is passed to the call method of any
`FeatureColumn` that takes a `training` argument. For example, if a
`FeatureColumn` performed dropout, the column could expose a `training`
argument to control whether the dropout should be applied. If `None`,
defaults to `tf.keras.backend.learning_phase()`.
Returns:
A `Tensor` which represents input layer of a model. Its shape
is (batch_size, first_layer_dimension) and its dtype is `float32`.
first_layer_dimension is determined based on given `feature_columns`.
Raises:
ValueError: If features are not a dictionary."
6058,DenseFeatures,tensorflow/tensorflow/python/keras/feature_column/dense_features_v2.py,29,class,"A layer that produces a dense `Tensor` based on given `feature_columns`.
Generally a single example in training data is described with FeatureColumns.
At the first layer of the model, this column oriented data should be converted
to a single `Tensor`.
This layer can be called multiple times with different features.
This is the V2 version of this layer that uses name_scopes to create
variables instead of variable_scopes. But this approach currently lacks
support for partitioned variables. In that case, use the V1 version instead.
Example:
```python
price = tf.feature_column.numeric_column('price')
keywords_embedded = tf.feature_column.embedding_column(
tf.feature_column.categorical_column_with_hash_bucket(""keywords"", 10K),
dimensions=16)
columns = [price, keywords_embedded, ...]
feature_layer = tf.keras.layers.DenseFeatures(columns)
features = tf.io.parse_example(
..., features=tf.feature_column.make_parse_example_spec(columns))
dense_tensor = feature_layer(features)
for units in [128, 64, 32]:
dense_tensor = tf.keras.layers.Dense(units, activation='relu')(dense_tensor)
prediction = tf.keras.layers.Dense(1)(dense_tensor)
```"
6059,build,tensorflow/tensorflow/python/keras/feature_column/dense_features_v2.py,90,method,
6060,SequenceFeatures,tensorflow/tensorflow/python/keras/feature_column/sequence_feature_column.py,36,class,"A layer for sequence input.
All `feature_columns` must be sequence dense columns with the same
`sequence_length`. The output of this method can be fed into sequence
networks, such as RNN.
The output of this method is a 3D `Tensor` of shape `[batch_size, T, D]`.
`T` is the maximum sequence length for this batch, which could differ from
batch to batch.
If multiple `feature_columns` are given with `Di` `num_elements` each, their
outputs are concatenated. So, the final `Tensor` has shape
`[batch_size, T, D0 + D1 + ... + Dn]`.
Example:
```python
# Behavior of some cells or feature columns may depend on whether we are in
# training or inference mode, e.g. applying dropout.
training = True
rating = sequence_numeric_column('rating')
watches = sequence_categorical_column_with_identity(
'watches', num_buckets=1000)
watches_embedding = embedding_column(watches, dimension=10)
columns = [rating, watches_embedding]
sequence_input_layer = SequenceFeatures(columns)
features = tf.io.parse_example(...,
features=make_parse_example_spec(columns))
sequence_input, sequence_length = sequence_input_layer(
features, training=training)
sequence_length_mask = tf.sequence_mask(sequence_length)
rnn_cell = tf.keras.layers.SimpleRNNCell(hidden_size, training=training)
rnn_layer = tf.keras.layers.RNN(rnn_cell, training=training)
outputs, state = rnn_layer(sequence_input, mask=sequence_length_mask)
```"
6061,call,tensorflow/tensorflow/python/keras/feature_column/sequence_feature_column.py,111,method,"Returns sequence input corresponding to the `feature_columns`.
Args:
features: A dict mapping keys to tensors.
training: Python boolean or None, indicating whether to the layer is being
run in training mode. This argument is passed to the call method of any
`FeatureColumn` that takes a `training` argument. For example, if a
`FeatureColumn` performed dropout, the column could expose a `training`
argument to control whether the dropout should be applied. If `None`,
defaults to `tf.keras.backend.learning_phase()`.
Returns:
An `(input_layer, sequence_length)` tuple where:
- input_layer: A float `Tensor` of shape `[batch_size, T, D]`.
`T` is the maximum sequence length for this batch, which could differ
from batch to batch. `D` is the sum of `num_elements` for all
`feature_columns`.
- sequence_length: An int `Tensor` of shape `[batch_size]`. The sequence
length for each example.
Raises:
ValueError: If features are not a dictionary."
6062,populate_deserializable_objects,tensorflow/tensorflow/python/keras/initializers/__init__.py,38,function,"Populates dict ALL_OBJECTS with every built-in initializer.
"
6063,serialize,tensorflow/tensorflow/python/keras/initializers/__init__.py,134,function,
6064,deserialize,tensorflow/tensorflow/python/keras/initializers/__init__.py,139,function,Return an `Initializer` object from its config.
6065,get,tensorflow/tensorflow/python/keras/initializers/__init__.py,150,function,
6066,RandomNormal,tensorflow/tensorflow/python/keras/initializers/initializers_v1.py,44,class,
6067,RandomUniform,tensorflow/tensorflow/python/keras/initializers/initializers_v1.py,54,class,
6068,TruncatedNormal,tensorflow/tensorflow/python/keras/initializers/initializers_v1.py,64,class,
6069,LecunNormal,tensorflow/tensorflow/python/keras/initializers/initializers_v1.py,72,class,
6070,get_config,tensorflow/tensorflow/python/keras/initializers/initializers_v1.py,78,method,
6071,LecunUniform,tensorflow/tensorflow/python/keras/initializers/initializers_v1.py,83,class,
6072,get_config,tensorflow/tensorflow/python/keras/initializers/initializers_v1.py,89,method,
6073,HeNormal,tensorflow/tensorflow/python/keras/initializers/initializers_v1.py,94,class,
6074,get_config,tensorflow/tensorflow/python/keras/initializers/initializers_v1.py,100,method,
6075,HeUniform,tensorflow/tensorflow/python/keras/initializers/initializers_v1.py,105,class,
6076,get_config,tensorflow/tensorflow/python/keras/initializers/initializers_v1.py,111,method,
6077,Initializer,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,30,class,"Initializer base class: all Keras initializers inherit from this class.
Initializers should implement a `__call__` method with the following
signature:
```python
def __call__(self, shape, dtype=None)`:
# returns a tensor of shape `shape` and dtype `dtype`
# containing values drawn from a distribution of your choice.
```
Optionally, you an also implement the method `get_config` and the class
method `from_config` in order to support serialization -- just like with
any Keras object.
Here's a simple example: a random normal initializer.
```python
import tensorflow as tf
class ExampleRandomNormal(tf.keras.initializers.Initializer):
def __init__(self, mean, stddev):
self.mean = mean
self.stddev = stddev
def __call__(self, shape, dtype=None)`:
return tf.random.normal(
shape, mean=self.mean, stddev=self.stddev, dtype=dtype)
def get_config(self): # To support serialization
return {""mean"": self.mean, ""stddev"": self.stddev}
```
Note that we don't have to implement `from_config` in the example above since
the constructor arguments of the class the keys in the config returned by
`get_config` are the same. In this case, the default `from_config`
works fine."
6078,get_config,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,80,method,"Returns the configuration of the initializer as a JSON-serializable dict.
Returns:
A JSON-serializable Python dict."
6079,from_config,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,89,method,"Instantiates an initializer from a configuration dictionary.
Example:
```python
initializer = RandomUniform(-1, 1)
config = initializer.get_config()
initializer = RandomUniform.from_config(config)
```
Args:
config: A Python dictionary, the output of `get_config`.
Returns:
A `tf.keras.initializers.Initializer` instance."
6080,Zeros,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,111,class,"Initializer that generates tensors initialized to 0.
Also available via the shortcut function `tf.keras.initializers.zeros`.
Examples:
>>> # Standalone usage:
>>> initializer = tf.keras.initializers.Zeros()
>>> values = initializer(shape=(2, 2))
>>> # Usage in a Keras layer:
>>> initializer = tf.keras.initializers.Zeros()
>>> layer = tf.keras.layers.Dense(3, kernel_initializer=initializer)"
6081,Ones,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,141,class,"Initializer that generates tensors initialized to 1.
Also available via the shortcut function `tf.keras.initializers.ones`.
Examples:
>>> # Standalone usage:
>>> initializer = tf.keras.initializers.Ones()
>>> values = initializer(shape=(2, 2))
>>> # Usage in a Keras layer:
>>> initializer = tf.keras.initializers.Ones()
>>> layer = tf.keras.layers.Dense(3, kernel_initializer=initializer)"
6082,Constant,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,173,class,"Initializer that generates tensors with constant values.
Also available via the shortcut function `tf.keras.initializers.constant`.
Only scalar values are allowed.
The constant value provided must be convertible to the dtype requested
when calling the initializer.
Examples:
>>> # Standalone usage:
>>> initializer = tf.keras.initializers.Constant(3.)
>>> values = initializer(shape=(2, 2))
>>> # Usage in a Keras layer:
>>> initializer = tf.keras.initializers.Constant(3.)
>>> layer = tf.keras.layers.Dense(3, kernel_initializer=initializer)
Args:
value: A Python scalar."
6083,get_config,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,212,method,
6084,RandomUniform,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,219,class,"Initializer that generates tensors with a uniform distribution.
Also available via the shortcut function
`tf.keras.initializers.random_uniform`.
Examples:
>>> # Standalone usage:
>>> initializer = tf.keras.initializers.RandomUniform(minval=0., maxval=1.)
>>> values = initializer(shape=(2, 2))
>>> # Usage in a Keras layer:
>>> initializer = tf.keras.initializers.RandomUniform(minval=0., maxval=1.)
>>> layer = tf.keras.layers.Dense(3, kernel_initializer=initializer)
Args:
minval: A python scalar or a scalar tensor. Lower bound of the range of
random values to generate (inclusive).
maxval: A python scalar or a scalar tensor. Upper bound of the range of
random values to generate (exclusive).
seed: A Python integer. An initializer created with a given seed will
always produce the same random tensor for a given shape and dtype."
6085,RandomNormal,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,261,class,"Initializer that generates tensors with a normal distribution.
Also available via the shortcut function
`tf.keras.initializers.random_normal`.
Examples:
>>> # Standalone usage:
>>> initializer = tf.keras.initializers.RandomNormal(mean=0., stddev=1.)
>>> values = initializer(shape=(2, 2))
>>> # Usage in a Keras layer:
>>> initializer = tf.keras.initializers.RandomNormal(mean=0., stddev=1.)
>>> layer = tf.keras.layers.Dense(3, kernel_initializer=initializer)
Args:
mean: a python scalar or a scalar tensor. Mean of the random values to
generate.
stddev: a python scalar or a scalar tensor. Standard deviation of the random
values to generate.
seed: A Python integer. An initializer created with a given seed will
always produce the same random tensor for a given shape and dtype."
6086,TruncatedNormal,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,302,class,"Initializer that generates a truncated normal distribution.
Also available via the shortcut function
`tf.keras.initializers.truncated_normal`.
The values generated are similar to values from a
`tf.keras.initializers.RandomNormal` initializer except that values more
than two standard deviations from the mean are
discarded and re-drawn.
Examples:
>>> # Standalone usage:
>>> initializer = tf.keras.initializers.TruncatedNormal(mean=0., stddev=1.)
>>> values = initializer(shape=(2, 2))
>>> # Usage in a Keras layer:
>>> initializer = tf.keras.initializers.TruncatedNormal(mean=0., stddev=1.)
>>> layer = tf.keras.layers.Dense(3, kernel_initializer=initializer)
Args:
mean: a python scalar or a scalar tensor. Mean of the random values
to generate.
stddev: a python scalar or a scalar tensor. Standard deviation of the
random values to generate.
seed: A Python integer. An initializer created with a given seed will
always produce the same random tensor for a given shape and dtype."
6087,VarianceScaling,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,348,class,"Initializer capable of adapting its scale to the shape of weights tensors.
Also available via the shortcut function
`tf.keras.initializers.variance_scaling`.
With `distribution=""truncated_normal"" or ""untruncated_normal""`, samples are
drawn from a truncated/untruncated normal distribution with a mean of zero and
a standard deviation (after truncation, if used) `stddev = sqrt(scale / n)`,
where `n` is:
- number of input units in the weight tensor, if `mode=""fan_in""`
- number of output units, if `mode=""fan_out""`
- average of the numbers of input and output units, if `mode=""fan_avg""`
With `distribution=""uniform""`, samples are drawn from a uniform distribution
within `[-limit, limit]`, where `limit = sqrt(3 * scale / n)`.
Examples:
>>> # Standalone usage:
>>> initializer = tf.keras.initializers.VarianceScaling(
... scale=0.1, mode='fan_in', distribution='uniform')
>>> values = initializer(shape=(2, 2))
>>> # Usage in a Keras layer:
>>> initializer = tf.keras.initializers.VarianceScaling(
... scale=0.1, mode='fan_in', distribution='uniform')
>>> layer = tf.keras.layers.Dense(3, kernel_initializer=initializer)
Args:
scale: Scaling factor (positive float).
mode: One of ""fan_in"", ""fan_out"", ""fan_avg"".
distribution: Random distribution to use. One of ""truncated_normal"",
""untruncated_normal"" and ""uniform"".
seed: A Python integer. An initializer created with a given seed will
always produce the same random tensor for a given shape and dtype."
6088,Orthogonal,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,403,class,"Initializer that generates an orthogonal matrix.
Also available via the shortcut function `tf.keras.initializers.orthogonal`.
If the shape of the tensor to initialize is two-dimensional, it is initialized
with an orthogonal matrix obtained from the QR decomposition of a matrix of
random numbers drawn from a normal distribution.
If the matrix has fewer rows than columns then the output will have orthogonal
rows. Otherwise, the output will have orthogonal columns.
If the shape of the tensor to initialize is more than two-dimensional,
a matrix of shape `(shape[0] * ... * shape[n - 2], shape[n - 1])`
is initialized, where `n` is the length of the shape vector.
The matrix is subsequently reshaped to give a tensor of the desired shape.
Examples:
>>> # Standalone usage:
>>> initializer = tf.keras.initializers.Orthogonal()
>>> values = initializer(shape=(2, 2))
>>> # Usage in a Keras layer:
>>> initializer = tf.keras.initializers.Orthogonal()
>>> layer = tf.keras.layers.Dense(3, kernel_initializer=initializer)
Args:
gain: multiplicative factor to apply to the orthogonal matrix
seed: A Python integer. An initializer created with a given seed will
always produce the same random tensor for a given shape and dtype.
References:
[Saxe et al., 2014](https://openreview.net/forum?id=_wzZwKpTDF_9C)
([pdf](https://arxiv.org/pdf/1312.6120.pdf))"
6089,Identity,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,455,class,"Initializer that generates the identity matrix.
Also available via the shortcut function `tf.keras.initializers.identity`.
Only usable for generating 2D matrices.
Examples:
>>> # Standalone usage:
>>> initializer = tf.keras.initializers.Identity()
>>> values = initializer(shape=(2, 2))
>>> # Usage in a Keras layer:
>>> initializer = tf.keras.initializers.Identity()
>>> layer = tf.keras.layers.Dense(3, kernel_initializer=initializer)
Args:
gain: Multiplicative factor to apply to the identity matrix."
6090,GlorotUniform,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,492,class,"The Glorot uniform initializer, also called Xavier uniform initializer.
Also available via the shortcut function
`tf.keras.initializers.glorot_uniform`.
Draws samples from a uniform distribution within `[-limit, limit]`, where
`limit = sqrt(6 / (fan_in + fan_out))` (`fan_in` is the number of input units
in the weight tensor and `fan_out` is the number of output units).
Examples:
>>> # Standalone usage:
>>> initializer = tf.keras.initializers.GlorotUniform()
>>> values = initializer(shape=(2, 2))
>>> # Usage in a Keras layer:
>>> initializer = tf.keras.initializers.GlorotUniform()
>>> layer = tf.keras.layers.Dense(3, kernel_initializer=initializer)
Args:
seed: A Python integer. An initializer created with a given seed will
always produce the same random tensor for a given shape and dtype.
References:
[Glorot et al., 2010](http://proceedings.mlr.press/v9/glorot10a.html)
([pdf](http://jmlr.org/proceedings/papers/v9/glorot10a/glorot10a.pdf))"
6091,get_config,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,528,method,
6092,GlorotNormal,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,535,class,"The Glorot normal initializer, also called Xavier normal initializer.
Also available via the shortcut function
`tf.keras.initializers.glorot_normal`.
Draws samples from a truncated normal distribution centered on 0 with `stddev
= sqrt(2 / (fan_in + fan_out))` where `fan_in` is the number of input units in
the weight tensor and `fan_out` is the number of output units in the weight
tensor.
Examples:
>>> # Standalone usage:
>>> initializer = tf.keras.initializers.GlorotNormal()
>>> values = initializer(shape=(2, 2))
>>> # Usage in a Keras layer:
>>> initializer = tf.keras.initializers.GlorotNormal()
>>> layer = tf.keras.layers.Dense(3, kernel_initializer=initializer)
Args:
seed: A Python integer. An initializer created with a given seed will
always produce the same random tensor for a given shape and dtype.
References:
[Glorot et al., 2010](http://proceedings.mlr.press/v9/glorot10a.html)
([pdf](http://jmlr.org/proceedings/papers/v9/glorot10a/glorot10a.pdf))"
6093,get_config,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,572,method,
6094,LecunNormal,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,579,class,"Lecun normal initializer.
Also available via the shortcut function
`tf.keras.initializers.lecun_normal`.
Initializers allow you to pre-specify an initialization strategy, encoded in
the Initializer object, without knowing the shape and dtype of the variable
being initialized.
Draws samples from a truncated normal distribution centered on 0 with `stddev
= sqrt(1 / fan_in)` where `fan_in` is the number of input units in the weight
tensor.
Examples:
>>> # Standalone usage:
>>> initializer = tf.keras.initializers.LecunNormal()
>>> values = initializer(shape=(2, 2))
>>> # Usage in a Keras layer:
>>> initializer = tf.keras.initializers.LecunNormal()
>>> layer = tf.keras.layers.Dense(3, kernel_initializer=initializer)
Arguments:
seed: A Python integer. Used to seed the random generator.
References:
- Self-Normalizing Neural Networks,
[Klambauer et al., 2017]
(https://papers.nips.cc/paper/6698-self-normalizing-neural-networks)
([pdf]
(https://papers.nips.cc/paper/6698-self-normalizing-neural-networks.pdf))
- Efficient Backprop,
[Lecun et al., 1998](http://yann.lecun.com/exdb/publis/pdf/lecun-98b.pdf)"
6095,get_config,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,620,method,
6096,LecunUniform,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,627,class,"Lecun uniform initializer.
Also available via the shortcut function
`tf.keras.initializers.lecun_uniform`.
Draws samples from a uniform distribution within `[-limit, limit]`,
where `limit = sqrt(3 / fan_in)` (`fan_in` is the number of input units in the
weight tensor).
Examples:
>>> # Standalone usage:
>>> initializer = tf.keras.initializers.LecunUniform()
>>> values = initializer(shape=(2, 2))
>>> # Usage in a Keras layer:
>>> initializer = tf.keras.initializers.LecunUniform()
>>> layer = tf.keras.layers.Dense(3, kernel_initializer=initializer)
Arguments:
seed: A Python integer. An initializer created with a given seed will
always produce the same random tensor for a given shape and dtype.
References:
- Self-Normalizing Neural Networks,
[Klambauer et al., 2017](https://papers.nips.cc/paper/6698-self-normalizing-neural-networks) # pylint: disable=line-too-long
([pdf](https://papers.nips.cc/paper/6698-self-normalizing-neural-networks.pdf))
- Efficient Backprop,
[Lecun et al., 1998](http://yann.lecun.com/exdb/publis/pdf/lecun-98b.pdf)"
6097,get_config,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,663,method,
6098,HeNormal,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,670,class,"He normal initializer.
Also available via the shortcut function
`tf.keras.initializers.he_normal`.
It draws samples from a truncated normal distribution centered on 0 with
`stddev = sqrt(2 / fan_in)` where `fan_in` is the number of input units in the
weight tensor.
Examples:
>>> # Standalone usage:
>>> initializer = tf.keras.initializers.HeNormal()
>>> values = initializer(shape=(2, 2))
>>> # Usage in a Keras layer:
>>> initializer = tf.keras.initializers.HeNormal()
>>> layer = tf.keras.layers.Dense(3, kernel_initializer=initializer)
Arguments:
seed: A Python integer. An initializer created with a given seed will
always produce the same random tensor for a given shape and dtype.
References:
[He et al., 2015](https://www.cv-foundation.org/openaccess/content_iccv_2015/html/He_Delving_Deep_into_ICCV_2015_paper.html) # pylint: disable=line-too-long
([pdf](https://www.cv-foundation.org/openaccess/content_iccv_2015/papers/He_Delving_Deep_into_ICCV_2015_paper.pdf))"
6099,get_config,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,703,method,
6100,HeUniform,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,710,class,"He uniform variance scaling initializer.
Also available via the shortcut function
`tf.keras.initializers.he_uniform`.
Draws samples from a uniform distribution within `[-limit, limit]`, where
`limit = sqrt(6 / fan_in)` (`fan_in` is the number of input units in the
weight tensor).
Examples:
>>> # Standalone usage:
>>> initializer = tf.keras.initializers.HeUniform()
>>> values = initializer(shape=(2, 2))
>>> # Usage in a Keras layer:
>>> initializer = tf.keras.initializers.HeUniform()
>>> layer = tf.keras.layers.Dense(3, kernel_initializer=initializer)
Arguments:
seed: A Python integer. An initializer created with a given seed will
always produce the same random tensor for a given shape and dtype.
References:
[He et al., 2015](https://www.cv-foundation.org/openaccess/content_iccv_2015/html/He_Delving_Deep_into_ICCV_2015_paper.html) # pylint: disable=line-too-long
([pdf](https://www.cv-foundation.org/openaccess/content_iccv_2015/papers/He_Delving_Deep_into_ICCV_2015_paper.pdf))"
6101,get_config,tensorflow/tensorflow/python/keras/initializers/initializers_v2.py,743,method,
6102,MiniModel,tensorflow/tensorflow/python/keras/integration_test/function_test.py,24,class,"Minimal model for mnist.
Useful for testing and debugging on slow TPU simulators."
6103,call,tensorflow/tensorflow/python/keras/integration_test/function_test.py,35,method,
6104,DefunnedMiniModel,tensorflow/tensorflow/python/keras/integration_test/function_test.py,39,class,
6105,call,tensorflow/tensorflow/python/keras/integration_test/function_test.py,42,method,
6106,ModelWithOptimizer,tensorflow/tensorflow/python/keras/integration_test/function_test.py,46,class,
6107,call,tensorflow/tensorflow/python/keras/integration_test/function_test.py,56,method,
6108,KerasNetworkTFRNNs,tensorflow/tensorflow/python/keras/integration_test/legacy_rnn_test.py,25,class,
6109,call,tensorflow/tensorflow/python/keras/integration_test/legacy_rnn_test.py,32,method,
6110,KerasNetworkKerasRNNs,tensorflow/tensorflow/python/keras/integration_test/legacy_rnn_test.py,36,class,
6111,call,tensorflow/tensorflow/python/keras/integration_test/legacy_rnn_test.py,43,method,
6112,cycle,tensorflow/tensorflow/python/keras/integration_test/saved_model_test.py,27,function,
6113,get_tpu_cluster_resolver,tensorflow/tensorflow/python/keras/integration_test/tpu_strategy_test.py,32,function,
6114,get_tpu_strategy,tensorflow/tensorflow/python/keras/integration_test/tpu_strategy_test.py,41,function,
6115,VersionAwareLayers,tensorflow/tensorflow/python/keras/layers/__init__.py,275,class,"Utility to be used internally to access layers in a V1/V2-aware fashion.
When using layers within the Keras codebase, under the constraint that
e.g. `layers.BatchNormalization` should be the `BatchNormalization` version
corresponding to the current runtime (TF1 or TF2), do not simply access
`layers.BatchNormalization` since it would ignore e.g. an early
`compat.v2.disable_v2_behavior()` call. Instead, use an instance
of `VersionAwareLayers` (which you can use just like the `layers` module)."
6116,LeakyReLU,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,33,class,"Leaky version of a Rectified Linear Unit.
It allows a small gradient when the unit is not active:
```
f(x) = alpha * x if x < 0
f(x) = x if x >= 0
```
Usage:
>>> layer = tf.keras.layers.LeakyReLU()
>>> output = layer([-3.0, -1.0, 0.0, 2.0])
>>> list(output.numpy())
[-0.9, -0.3, 0.0, 2.0]
>>> layer = tf.keras.layers.LeakyReLU(alpha=0.1)
>>> output = layer([-3.0, -1.0, 0.0, 2.0])
>>> list(output.numpy())
[-0.3, -0.1, 0.0, 2.0]
Input shape:
Arbitrary. Use the keyword argument `input_shape`
(tuple of integers, does not include the batch axis)
when using this layer as the first layer in a model.
Output shape:
Same shape as the input.
Arguments:
alpha: Float >= 0. Negative slope coefficient. Default to 0.3."
6117,call,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,72,method,
6118,get_config,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,75,method,
6119,compute_output_shape,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,81,method,
6120,PReLU,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,86,class,"Parametric Rectified Linear Unit.
It follows:
```
f(x) = alpha * x for x < 0
f(x) = x for x >= 0
```
where `alpha` is a learned array with the same shape as x.
Input shape:
Arbitrary. Use the keyword argument `input_shape`
(tuple of integers, does not include the samples axis)
when using this layer as the first layer in a model.
Output shape:
Same shape as the input.
Arguments:
alpha_initializer: Initializer function for the weights.
alpha_regularizer: Regularizer for the weights.
alpha_constraint: Constraint for the weights.
shared_axes: The axes along which to share learnable
parameters for the activation function.
For example, if the incoming feature maps
are from a 2D convolution
with output shape `(batch, height, width, channels)`,
and you wish to share parameters across space
so that each filter only has one set of parameters,
set `shared_axes=[1, 2]`."
6121,build,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,139,method,
6122,call,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,159,method,
6123,get_config,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,164,method,
6124,compute_output_shape,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,175,method,
6125,ELU,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,180,class,"Exponential Linear Unit.
It follows:
```
f(x) = alpha * (exp(x) - 1.) for x < 0
f(x) = x for x >= 0
```
Input shape:
Arbitrary. Use the keyword argument `input_shape`
(tuple of integers, does not include the samples axis)
when using this layer as the first layer in a model.
Output shape:
Same shape as the input.
Arguments:
alpha: Scale for the negative factor."
6126,call,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,207,method,
6127,get_config,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,210,method,
6128,compute_output_shape,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,216,method,
6129,ThresholdedReLU,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,221,class,"Thresholded Rectified Linear Unit.
It follows:
```
f(x) = x for x > theta
f(x) = 0 otherwise`
```
Input shape:
Arbitrary. Use the keyword argument `input_shape`
(tuple of integers, does not include the samples axis)
when using this layer as the first layer in a model.
Output shape:
Same shape as the input.
Arguments:
theta: Float >= 0. Threshold location of activation."
6130,call,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,248,method,
6131,get_config,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,252,method,
6132,compute_output_shape,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,258,method,
6133,Softmax,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,263,class,"Softmax activation function.
Input shape:
Arbitrary. Use the keyword argument `input_shape`
(tuple of integers, does not include the samples axis)
when using this layer as the first layer in a model.
Output shape:
Same shape as the input.
Arguments:
axis: Integer, axis along which the softmax normalization is applied."
6134,call,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,283,method,
6135,get_config,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,286,method,
6136,compute_output_shape,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,292,method,
6137,ReLU,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,297,class,"Rectified Linear Unit activation function.
With default values, it returns element-wise `max(x, 0)`.
Otherwise, it follows:
```
f(x) = max_value if x >= max_value
f(x) = x if threshold <= x < max_value
f(x) = negative_slope * (x - threshold) otherwise
```
Usage:
>>> layer = tf.keras.layers.ReLU()
>>> output = layer([-3.0, -1.0, 0.0, 2.0])
>>> list(output.numpy())
[0.0, 0.0, 0.0, 2.0]
>>> layer = tf.keras.layers.ReLU(max_value=1.0)
>>> output = layer([-3.0, -1.0, 0.0, 2.0])
>>> list(output.numpy())
[0.0, 0.0, 0.0, 1.0]
>>> layer = tf.keras.layers.ReLU(negative_slope=1.0)
>>> output = layer([-3.0, -1.0, 0.0, 2.0])
>>> list(output.numpy())
[-3.0, -1.0, 0.0, 2.0]
>>> layer = tf.keras.layers.ReLU(threshold=1.5)
>>> output = layer([-3.0, -1.0, 1.0, 2.0])
>>> list(output.numpy())
[0.0, 0.0, 0.0, 2.0]
Input shape:
Arbitrary. Use the keyword argument `input_shape`
(tuple of integers, does not include the batch axis)
when using this layer as the first layer in a model.
Output shape:
Same shape as the input.
Arguments:
max_value: Float >= 0. Maximum activation value. Default to None, which
means unlimited.
negative_slope: Float >= 0. Negative slope coefficient. Default to 0.
threshold: Float. Threshold value for thresholded activation. Default to 0."
6138,call,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,363,method,
6139,get_config,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,371,method,
6140,compute_output_shape,tensorflow/tensorflow/python/keras/layers/advanced_activations.py,381,method,
6141,Conv,tensorflow/tensorflow/python/keras/layers/convolutional.py,52,class,"Abstract N-D convolution layer (private, used as implementation base).
This layer creates a convolution kernel that is convolved
(actually cross-correlated) with the layer input to produce a tensor of
outputs. If `use_bias` is True (and a `bias_initializer` is provided),
a bias vector is created and added to the outputs. Finally, if
`activation` is not `None`, it is applied to the outputs as well.
Note: layer attributes cannot be modified after the layer has been called
once (except the `trainable` attribute).
Arguments:
rank: An integer, the rank of the convolution, e.g. ""2"" for 2D convolution.
filters: Integer, the dimensionality of the output space (i.e. the number
of filters in the convolution).
kernel_size: An integer or tuple/list of n integers, specifying the
length of the convolution window.
strides: An integer or tuple/list of n integers,
specifying the stride length of the convolution.
Specifying any stride value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: One of `""valid""`, `""same""`, or `""causal""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input. `""causal""` results in causal
(dilated) convolutions, e.g. `output[t]` does not depend on `input[t+1:]`.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch_size, ..., channels)` while `channels_first` corresponds to
inputs with shape `(batch_size, channels, ...)`.
dilation_rate: An integer or tuple/list of n integers, specifying
the dilation rate to use for dilated convolution.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any `strides` value != 1.
groups: A positive integer specifying the number of groups in which the
input is split along the channel axis. Each group is convolved
separately with `filters / groups` filters. The output is the
concatenation of all the `groups` results along the channel axis.
Input channels and `filters` must both be divisible by `groups`.
activation: Activation function to use.
If you don't specify anything, no activation is applied.
use_bias: Boolean, whether the layer uses a bias.
kernel_initializer: An initializer for the convolution kernel.
bias_initializer: An initializer for the bias vector. If None, the default
initializer will be used.
kernel_regularizer: Optional regularizer for the convolution kernel.
bias_regularizer: Optional regularizer for the bias vector.
activity_regularizer: Optional regularizer function for the output.
kernel_constraint: Optional projection function to be applied to the
kernel after being updated by an `Optimizer` (e.g. used to implement
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training.
bias_constraint: Optional projection function to be applied to the
bias after being updated by an `Optimizer`.
trainable: Boolean, if `True` the weights of this layer will be marked as
trainable (and listed in `layer.trainable_weights`).
name: A string, the name of the layer."
6142,build,tensorflow/tensorflow/python/keras/layers/convolutional.py,189,method,
6143,call,tensorflow/tensorflow/python/keras/layers/convolutional.py,247,method,
6144,compute_output_shape,tensorflow/tensorflow/python/keras/layers/convolutional.py,287,method,
6145,get_config,tensorflow/tensorflow/python/keras/layers/convolutional.py,303,method,
6146,Conv1D,tensorflow/tensorflow/python/keras/layers/convolutional.py,378,class,"1D convolution layer (e.g. temporal convolution).
This layer creates a convolution kernel that is convolved
with the layer input over a single spatial (or temporal) dimension
to produce a tensor of outputs.
If `use_bias` is True, a bias vector is created and added to the outputs.
Finally, if `activation` is not `None`,
it is applied to the outputs as well.
When using this layer as the first layer in a model,
provide an `input_shape` argument
(tuple of integers or `None`, e.g.
`(10, 128)` for sequences of 10 vectors of 128-dimensional vectors,
or `(None, 128)` for variable-length sequences of 128-dimensional vectors.
Examples:
>>> # The inputs are 128-length vectors with 10 timesteps, and the batch size
>>> # is 4.
>>> input_shape = (4, 10, 128)
>>> x = tf.random.normal(input_shape)
>>> y = tf.keras.layers.Conv1D(
... 32, 3, activation='relu',input_shape=input_shape[1:])(x)
>>> print(y.shape)
(4, 8, 32)
>>> # With extended batch shape [4, 7] (e.g. weather data where batch
>>> # dimensions correspond to spatial location and the third dimension
>>> # corresponds to time.)
>>> input_shape = (4, 7, 10, 128)
>>> x = tf.random.normal(input_shape)
>>> y = tf.keras.layers.Conv1D(
... 32, 3, activation='relu', input_shape=input_shape[2:])(x)
>>> print(y.shape)
(4, 7, 8, 32)
Arguments:
filters: Integer, the dimensionality of the output space
(i.e. the number of output filters in the convolution).
kernel_size: An integer or tuple/list of a single integer,
specifying the length of the 1D convolution window.
strides: An integer or tuple/list of a single integer,
specifying the stride length of the convolution.
Specifying any stride value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: One of `""valid""`, `""same""` or `""causal""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
`""causal""` results in causal (dilated) convolutions, e.g. `output[t]`
does not depend on `input[t+1:]`. Useful when modeling temporal data
where the model should not violate the temporal order.
See [WaveNet: A Generative Model for Raw Audio, section
2.1](https://arxiv.org/abs/1609.03499).
data_format: A string,
one of `channels_last` (default) or `channels_first`.
dilation_rate: an integer or tuple/list of a single integer, specifying
the dilation rate to use for dilated convolution.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any `strides` value != 1.
groups: A positive integer specifying the number of groups in which the
input is split along the channel axis. Each group is convolved
separately with `filters / groups` filters. The output is the
concatenation of all the `groups` results along the channel axis.
Input channels and `filters` must both be divisible by `groups`.
activation: Activation function to use.
If you don't specify anything, no activation is applied (
see `keras.activations`).
use_bias: Boolean, whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix (
see `keras.initializers`).
bias_initializer: Initializer for the bias vector (
see `keras.initializers`).
kernel_regularizer: Regularizer function applied to
the `kernel` weights matrix (see `keras.regularizers`).
bias_regularizer: Regularizer function applied to the bias vector (
see `keras.regularizers`).
activity_regularizer: Regularizer function applied to
the output of the layer (its ""activation"") (
see `keras.regularizers`).
kernel_constraint: Constraint function applied to the kernel matrix (
see `keras.constraints`).
bias_constraint: Constraint function applied to the bias vector (
see `keras.constraints`).
Input shape:
3+D tensor with shape: `batch_shape + (steps, input_dim)`
Output shape:
3+D tensor with shape: `batch_shape + (new_steps, filters)`
`steps` value might have changed due to padding or strides.
Returns:
A tensor of rank 3 representing
`activation(conv1d(inputs, kernel) + bias)`.
Raises:
ValueError: when both `strides > 1` and `dilation_rate > 1`."
6147,Conv2D,tensorflow/tensorflow/python/keras/layers/convolutional.py,519,class,"2D convolution layer (e.g. spatial convolution over images).
This layer creates a convolution kernel that is convolved
with the layer input to produce a tensor of
outputs. If `use_bias` is True,
a bias vector is created and added to the outputs. Finally, if
`activation` is not `None`, it is applied to the outputs as well.
When using this layer as the first layer in a model,
provide the keyword argument `input_shape`
(tuple of integers, does not include the sample axis),
e.g. `input_shape=(128, 128, 3)` for 128x128 RGB pictures
in `data_format=""channels_last""`.
Examples:
>>> # The inputs are 28x28 RGB images with `channels_last` and the batch
>>> # size is 4.
>>> input_shape = (4, 28, 28, 3)
>>> x = tf.random.normal(input_shape)
>>> y = tf.keras.layers.Conv2D(
... 2, 3, activation='relu', input_shape=input_shape[1:])(x)
>>> print(y.shape)
(4, 26, 26, 2)
>>> # With `dilation_rate` as 2.
>>> input_shape = (4, 28, 28, 3)
>>> x = tf.random.normal(input_shape)
>>> y = tf.keras.layers.Conv2D(
... 2, 3, activation='relu', dilation_rate=2, input_shape=input_shape[1:])(x)
>>> print(y.shape)
(4, 24, 24, 2)
>>> # With `padding` as ""same"".
>>> input_shape = (4, 28, 28, 3)
>>> x = tf.random.normal(input_shape)
>>> y = tf.keras.layers.Conv2D(
... 2, 3, activation='relu', padding=""same"", input_shape=input_shape[1:])(x)
>>> print(y.shape)
(4, 28, 28, 2)
>>> # With extended batch shape [4, 7]:
>>> input_shape = (4, 7, 28, 28, 3)
>>> x = tf.random.normal(input_shape)
>>> y = tf.keras.layers.Conv2D(
... 2, 3, activation='relu', input_shape=input_shape[2:])(x)
>>> print(y.shape)
(4, 7, 26, 26, 2)
Arguments:
filters: Integer, the dimensionality of the output space (i.e. the number of
output filters in the convolution).
kernel_size: An integer or tuple/list of 2 integers, specifying the height
and width of the 2D convolution window. Can be a single integer to specify
the same value for all spatial dimensions.
strides: An integer or tuple/list of 2 integers, specifying the strides of
the convolution along the height and width. Can be a single integer to
specify the same value for all spatial dimensions. Specifying any stride
value != 1 is incompatible with specifying any `dilation_rate` value != 1.
padding: one of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs. `channels_last` corresponds
to inputs with shape `(batch_size, height, width, channels)` while
`channels_first` corresponds to inputs with shape `(batch_size, channels,
height, width)`. It defaults to the `image_data_format` value found in
your Keras config file at `~/.keras/keras.json`. If you never set it, then
it will be `channels_last`.
dilation_rate: an integer or tuple/list of 2 integers, specifying the
dilation rate to use for dilated convolution. Can be a single integer to
specify the same value for all spatial dimensions. Currently, specifying
any `dilation_rate` value != 1 is incompatible with specifying any stride
value != 1.
groups: A positive integer specifying the number of groups in which the
input is split along the channel axis. Each group is convolved separately
with `filters / groups` filters. The output is the concatenation of all
the `groups` results along the channel axis. Input channels and `filters`
must both be divisible by `groups`.
activation: Activation function to use. If you don't specify anything, no
activation is applied (see `keras.activations`).
use_bias: Boolean, whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix (see
`keras.initializers`).
bias_initializer: Initializer for the bias vector (see
`keras.initializers`).
kernel_regularizer: Regularizer function applied to the `kernel` weights
matrix (see `keras.regularizers`).
bias_regularizer: Regularizer function applied to the bias vector (see
`keras.regularizers`).
activity_regularizer: Regularizer function applied to the output of the
layer (its ""activation"") (see `keras.regularizers`).
kernel_constraint: Constraint function applied to the kernel matrix (see
`keras.constraints`).
bias_constraint: Constraint function applied to the bias vector (see
`keras.constraints`).
Input shape:
4+D tensor with shape: `batch_shape + (channels, rows, cols)` if
`data_format='channels_first'`
or 4+D tensor with shape: `batch_shape + (rows, cols, channels)` if
`data_format='channels_last'`.
Output shape:
4+D tensor with shape: `batch_shape + (filters, new_rows, new_cols)` if
`data_format='channels_first'` or 4+D tensor with shape: `batch_shape +
(new_rows, new_cols, filters)` if `data_format='channels_last'`. `rows`
and `cols` values might have changed due to padding.
Returns:
A tensor of rank 4+ representing
`activation(conv2d(inputs, kernel) + bias)`.
Raises:
ValueError: if `padding` is `""causal""`.
ValueError: when both `strides > 1` and `dilation_rate > 1`."
6148,Conv3D,tensorflow/tensorflow/python/keras/layers/convolutional.py,678,class,"3D convolution layer (e.g. spatial convolution over volumes).
This layer creates a convolution kernel that is convolved
with the layer input to produce a tensor of
outputs. If `use_bias` is True,
a bias vector is created and added to the outputs. Finally, if
`activation` is not `None`, it is applied to the outputs as well.
When using this layer as the first layer in a model,
provide the keyword argument `input_shape`
(tuple of integers, does not include the sample axis),
e.g. `input_shape=(128, 128, 128, 1)` for 128x128x128 volumes
with a single channel,
in `data_format=""channels_last""`.
Examples:
>>> # The inputs are 28x28x28 volumes with a single channel, and the
>>> # batch size is 4
>>> input_shape =(4, 28, 28, 28, 1)
>>> x = tf.random.normal(input_shape)
>>> y = tf.keras.layers.Conv3D(
... 2, 3, activation='relu', input_shape=input_shape[1:])(x)
>>> print(y.shape)
(4, 26, 26, 26, 2)
>>> # With extended batch shape [4, 7], e.g. a batch of 4 videos of 3D frames,
>>> # with 7 frames per video.
>>> input_shape = (4, 7, 28, 28, 28, 1)
>>> x = tf.random.normal(input_shape)
>>> y = tf.keras.layers.Conv3D(
... 2, 3, activation='relu', input_shape=input_shape[2:])(x)
>>> print(y.shape)
(4, 7, 26, 26, 26, 2)
Arguments:
filters: Integer, the dimensionality of the output space (i.e. the number of
output filters in the convolution).
kernel_size: An integer or tuple/list of 3 integers, specifying the depth,
height and width of the 3D convolution window. Can be a single integer to
specify the same value for all spatial dimensions.
strides: An integer or tuple/list of 3 integers, specifying the strides of
the convolution along each spatial dimension. Can be a single integer to
specify the same value for all spatial dimensions. Specifying any stride
value != 1 is incompatible with specifying any `dilation_rate` value != 1.
padding: one of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs. `channels_last` corresponds
to inputs with shape `batch_shape + (spatial_dim1, spatial_dim2,
spatial_dim3, channels)` while `channels_first` corresponds to inputs with
shape `batch_shape + (channels, spatial_dim1, spatial_dim2,
spatial_dim3)`. It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`. If you never set it, then it
will be ""channels_last"".
dilation_rate: an integer or tuple/list of 3 integers, specifying the
dilation rate to use for dilated convolution. Can be a single integer to
specify the same value for all spatial dimensions. Currently, specifying
any `dilation_rate` value != 1 is incompatible with specifying any stride
value != 1.
groups: A positive integer specifying the number of groups in which the
input is split along the channel axis. Each group is convolved separately
with `filters / groups` filters. The output is the concatenation of all
the `groups` results along the channel axis. Input channels and `filters`
must both be divisible by `groups`.
activation: Activation function to use. If you don't specify anything, no
activation is applied (see `keras.activations`).
use_bias: Boolean, whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix (see
`keras.initializers`).
bias_initializer: Initializer for the bias vector (see
`keras.initializers`).
kernel_regularizer: Regularizer function applied to the `kernel` weights
matrix (see `keras.regularizers`).
bias_regularizer: Regularizer function applied to the bias vector (see
`keras.regularizers`).
activity_regularizer: Regularizer function applied to the output of the
layer (its ""activation"") (see `keras.regularizers`).
kernel_constraint: Constraint function applied to the kernel matrix (see
`keras.constraints`).
bias_constraint: Constraint function applied to the bias vector (see
`keras.constraints`).
Input shape:
5+D tensor with shape: `batch_shape + (channels, conv_dim1, conv_dim2,
conv_dim3)` if data_format='channels_first'
or 5+D tensor with shape: `batch_shape + (conv_dim1, conv_dim2, conv_dim3,
channels)` if data_format='channels_last'.
Output shape:
5+D tensor with shape: `batch_shape + (filters, new_conv_dim1,
new_conv_dim2, new_conv_dim3)` if data_format='channels_first'
or 5+D tensor with shape: `batch_shape + (new_conv_dim1, new_conv_dim2,
new_conv_dim3, filters)` if data_format='channels_last'. `new_conv_dim1`,
`new_conv_dim2` and `new_conv_dim3` values might have changed due to
padding.
Returns:
A tensor of rank 5+ representing
`activation(conv3d(inputs, kernel) + bias)`.
Raises:
ValueError: if `padding` is ""causal"".
ValueError: when both `strides > 1` and `dilation_rate > 1`."
6149,Conv1DTranspose,tensorflow/tensorflow/python/keras/layers/convolutional.py,826,class,"Transposed convolution layer (sometimes called Deconvolution).
The need for transposed convolutions generally arises
from the desire to use a transformation going in the opposite direction
of a normal convolution, i.e., from something that has the shape of the
output of some convolution to something that has the shape of its input
while maintaining a connectivity pattern that is compatible with
said convolution.
When using this layer as the first layer in a model,
provide the keyword argument `input_shape`
(tuple of integers, does not include the sample axis),
e.g. `input_shape=(128, 3)` for data with 128 time steps and 3 channels.
Arguments:
filters: Integer, the dimensionality of the output space
(i.e. the number of output filters in the convolution).
kernel_size: An integer length of the 1D convolution window.
strides: An integer specifying the stride of the convolution along the
time dimension. Specifying a stride value != 1 is incompatible with
specifying a `dilation_rate` value != 1. Defaults to 1.
padding: one of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
output_padding: An integer specifying the amount of padding along
the time dimension of the output tensor.
The amount of output padding must be lower than the stride.
If set to `None` (default), the output shape is inferred.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch_size, length, channels)` while `channels_first` corresponds to
inputs with shape `(batch_size, channels, length)`.
dilation_rate: an integer, specifying
the dilation rate to use for dilated convolution.
Currently, specifying a `dilation_rate` value != 1 is
incompatible with specifying a stride value != 1.
activation: Activation function to use.
If you don't specify anything, no activation is applied (
see `keras.activations`).
use_bias: Boolean, whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix (
see `keras.initializers`).
bias_initializer: Initializer for the bias vector (
see `keras.initializers`).
kernel_regularizer: Regularizer function applied to
the `kernel` weights matrix (see `keras.regularizers`).
bias_regularizer: Regularizer function applied to the bias vector (
see `keras.regularizers`).
activity_regularizer: Regularizer function applied to
the output of the layer (its ""activation"") (see `keras.regularizers`).
kernel_constraint: Constraint function applied to the kernel matrix (
see `keras.constraints`).
bias_constraint: Constraint function applied to the bias vector (
see `keras.constraints`).
Input shape:
3D tensor with shape:
`(batch_size, steps, channels)`
Output shape:
3D tensor with shape:
`(batch_size, new_steps, filters)`
If `output_padding` is specified:
```
new_timesteps = ((timesteps - 1) * strides + kernel_size -
2 * padding + output_padding)
```
Returns:
A tensor of rank 3 representing
`activation(conv1dtranspose(inputs, kernel) + bias)`.
Raises:
ValueError: if `padding` is ""causal"".
ValueError: when both `strides` > 1 and `dilation_rate` > 1.
References:
- [A guide to convolution arithmetic for deep learning](
https://arxiv.org/abs/1603.07285v1)
- [Deconvolutional Networks](
https://www.matthewzeiler.com/mattzeiler/deconvolutionalnetworks.pdf)"
6150,build,tensorflow/tensorflow/python/keras/layers/convolutional.py,958,method,
6151,call,tensorflow/tensorflow/python/keras/layers/convolutional.py,992,method,
6152,compute_output_shape,tensorflow/tensorflow/python/keras/layers/convolutional.py,1042,method,
6153,get_config,tensorflow/tensorflow/python/keras/layers/convolutional.py,1064,method,
6154,Conv2DTranspose,tensorflow/tensorflow/python/keras/layers/convolutional.py,1072,class,"Transposed convolution layer (sometimes called Deconvolution).
The need for transposed convolutions generally arises
from the desire to use a transformation going in the opposite direction
of a normal convolution, i.e., from something that has the shape of the
output of some convolution to something that has the shape of its input
while maintaining a connectivity pattern that is compatible with
said convolution.
When using this layer as the first layer in a model,
provide the keyword argument `input_shape`
(tuple of integers, does not include the sample axis),
e.g. `input_shape=(128, 128, 3)` for 128x128 RGB pictures
in `data_format=""channels_last""`.
Arguments:
filters: Integer, the dimensionality of the output space
(i.e. the number of output filters in the convolution).
kernel_size: An integer or tuple/list of 2 integers, specifying the
height and width of the 2D convolution window.
Can be a single integer to specify the same value for
all spatial dimensions.
strides: An integer or tuple/list of 2 integers,
specifying the strides of the convolution along the height and width.
Can be a single integer to specify the same value for
all spatial dimensions.
Specifying any stride value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: one of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
output_padding: An integer or tuple/list of 2 integers,
specifying the amount of padding along the height and width
of the output tensor.
Can be a single integer to specify the same value for all
spatial dimensions.
The amount of output padding along a given dimension must be
lower than the stride along that same dimension.
If set to `None` (default), the output shape is inferred.
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch_size, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch_size, channels, height, width)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
dilation_rate: an integer or tuple/list of 2 integers, specifying
the dilation rate to use for dilated convolution.
Can be a single integer to specify the same value for
all spatial dimensions.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any stride value != 1.
activation: Activation function to use.
If you don't specify anything, no activation is applied (
see `keras.activations`).
use_bias: Boolean, whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix (
see `keras.initializers`).
bias_initializer: Initializer for the bias vector (
see `keras.initializers`).
kernel_regularizer: Regularizer function applied to
the `kernel` weights matrix (see `keras.regularizers`).
bias_regularizer: Regularizer function applied to the bias vector (
see `keras.regularizers`).
activity_regularizer: Regularizer function applied to
the output of the layer (its ""activation"") (see `keras.regularizers`).
kernel_constraint: Constraint function applied to the kernel matrix (
see `keras.constraints`).
bias_constraint: Constraint function applied to the bias vector (
see `keras.constraints`).
Input shape:
4D tensor with shape:
`(batch_size, channels, rows, cols)` if data_format='channels_first'
or 4D tensor with shape:
`(batch_size, rows, cols, channels)` if data_format='channels_last'.
Output shape:
4D tensor with shape:
`(batch_size, filters, new_rows, new_cols)` if data_format='channels_first'
or 4D tensor with shape:
`(batch_size, new_rows, new_cols, filters)` if data_format='channels_last'.
`rows` and `cols` values might have changed due to padding.
If `output_padding` is specified:
```
new_rows = ((rows - 1) * strides[0] + kernel_size[0] - 2 * padding[0] +
output_padding[0])
new_cols = ((cols - 1) * strides[1] + kernel_size[1] - 2 * padding[1] +
output_padding[1])
```
Returns:
A tensor of rank 4 representing
`activation(conv2dtranspose(inputs, kernel) + bias)`.
Raises:
ValueError: if `padding` is ""causal"".
ValueError: when both `strides` > 1 and `dilation_rate` > 1.
References:
- [A guide to convolution arithmetic for deep
learning](https://arxiv.org/abs/1603.07285v1)
- [Deconvolutional
Networks](https://www.matthewzeiler.com/mattzeiler/deconvolutionalnetworks.pdf)"
6155,build,tensorflow/tensorflow/python/keras/layers/convolutional.py,1229,method,
6156,call,tensorflow/tensorflow/python/keras/layers/convolutional.py,1263,method,
6157,compute_output_shape,tensorflow/tensorflow/python/keras/layers/convolutional.py,1334,method,
6158,get_config,tensorflow/tensorflow/python/keras/layers/convolutional.py,1367,method,
6159,Conv3DTranspose,tensorflow/tensorflow/python/keras/layers/convolutional.py,1375,class,"Transposed convolution layer (sometimes called Deconvolution).
The need for transposed convolutions generally arises
from the desire to use a transformation going in the opposite direction
of a normal convolution, i.e., from something that has the shape of the
output of some convolution to something that has the shape of its input
while maintaining a connectivity pattern that is compatible with
said convolution.
When using this layer as the first layer in a model,
provide the keyword argument `input_shape`
(tuple of integers, does not include the sample axis),
e.g. `input_shape=(128, 128, 128, 3)` for a 128x128x128 volume with 3 channels
if `data_format=""channels_last""`.
Arguments:
filters: Integer, the dimensionality of the output space
(i.e. the number of output filters in the convolution).
kernel_size: An integer or tuple/list of 3 integers, specifying the
depth, height and width of the 3D convolution window.
Can be a single integer to specify the same value for
all spatial dimensions.
strides: An integer or tuple/list of 3 integers,
specifying the strides of the convolution along the depth, height
and width.
Can be a single integer to specify the same value for
all spatial dimensions.
Specifying any stride value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: one of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
output_padding: An integer or tuple/list of 3 integers,
specifying the amount of padding along the depth, height, and
width.
Can be a single integer to specify the same value for all
spatial dimensions.
The amount of output padding along a given dimension must be
lower than the stride along that same dimension.
If set to `None` (default), the output shape is inferred.
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch_size, depth, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch_size, channels, depth, height, width)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
dilation_rate: an integer or tuple/list of 3 integers, specifying
the dilation rate to use for dilated convolution.
Can be a single integer to specify the same value for
all spatial dimensions.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any stride value != 1.
activation: Activation function to use.
If you don't specify anything, no activation is applied (
see `keras.activations`).
use_bias: Boolean, whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix.
bias_initializer: Initializer for the bias vector.
kernel_regularizer: Regularizer function applied to
the `kernel` weights matrix (
see `keras.regularizers`).
bias_regularizer: Regularizer function applied to the bias vector (
see `keras.regularizers`).
activity_regularizer: Regularizer function applied to
the output of the layer (its ""activation"") (
see `keras.regularizers`).
kernel_constraint: Constraint function applied to the kernel matrix (
see `keras.constraints`).
bias_constraint: Constraint function applied to the bias vector (
see `keras.constraints`).
Input shape:
5D tensor with shape:
`(batch_size, channels, depth, rows, cols)` if data_format='channels_first'
or 5D tensor with shape:
`(batch_size, depth, rows, cols, channels)` if data_format='channels_last'.
Output shape:
5D tensor with shape:
`(batch_size, filters, new_depth, new_rows, new_cols)` if
data_format='channels_first'
or 5D tensor with shape:
`(batch_size, new_depth, new_rows, new_cols, filters)` if
data_format='channels_last'.
`depth` and `rows` and `cols` values might have changed due to padding.
If `output_padding` is specified::
```
new_depth = ((depth - 1) * strides[0] + kernel_size[0] - 2 * padding[0] +
output_padding[0])
new_rows = ((rows - 1) * strides[1] + kernel_size[1] - 2 * padding[1] +
output_padding[1])
new_cols = ((cols - 1) * strides[2] + kernel_size[2] - 2 * padding[2] +
output_padding[2])
```
Returns:
A tensor of rank 5 representing
`activation(conv3dtranspose(inputs, kernel) + bias)`.
Raises:
ValueError: if `padding` is ""causal"".
ValueError: when both `strides` > 1 and `dilation_rate` > 1.
References:
- [A guide to convolution arithmetic for deep
learning](https://arxiv.org/abs/1603.07285v1)
- [Deconvolutional
Networks](https://www.matthewzeiler.com/mattzeiler/deconvolutionalnetworks.pdf)"
6160,build,tensorflow/tensorflow/python/keras/layers/convolutional.py,1537,method,
6161,call,tensorflow/tensorflow/python/keras/layers/convolutional.py,1571,method,
6162,compute_output_shape,tensorflow/tensorflow/python/keras/layers/convolutional.py,1640,method,
6163,get_config,tensorflow/tensorflow/python/keras/layers/convolutional.py,1677,method,
6164,SeparableConv,tensorflow/tensorflow/python/keras/layers/convolutional.py,1684,class,"Abstract base layer for separable nD convolution.
This layer performs a depthwise convolution that acts separately on
channels, followed by a pointwise convolution that mixes channels.
If `use_bias` is True and a bias initializer is provided,
it adds a bias vector to the output.
It then optionally applies an activation function to produce the final output.
Arguments:
rank: An integer, the rank of the convolution, e.g. ""2"" for 2D convolution.
filters: Integer, the dimensionality of the output space (i.e. the number
of filters in the convolution).
kernel_size: A tuple or list of integers specifying the spatial
dimensions of the filters. Can be a single integer to specify the same
value for all spatial dimensions.
strides: A tuple or list of integers specifying the strides
of the convolution. Can be a single integer to specify the same value for
all spatial dimensions.
Specifying any `stride` value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch_size, ..., channels)` while `channels_first` corresponds to
inputs with shape `(batch_size, channels, ...)`.
dilation_rate: An integer or tuple/list of 2 integers, specifying
the dilation rate to use for dilated convolution.
Can be a single integer to specify the same value for
all spatial dimensions.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any stride value != 1.
depth_multiplier: The number of depthwise convolution output channels for
each input channel. The total number of depthwise convolution output
channels will be equal to `num_filters_in * depth_multiplier`.
activation: Activation function to use.
If you don't specify anything, no activation is applied (
see `keras.activations`).
use_bias: Boolean, whether the layer uses a bias.
depthwise_initializer: An initializer for the depthwise convolution kernel.
pointwise_initializer: An initializer for the pointwise convolution kernel.
bias_initializer: An initializer for the bias vector. If None, the default
initializer will be used.
depthwise_regularizer: Optional regularizer for the depthwise
convolution kernel.
pointwise_regularizer: Optional regularizer for the pointwise
convolution kernel.
bias_regularizer: Optional regularizer for the bias vector.
activity_regularizer: Optional regularizer function for the output.
depthwise_constraint: Optional projection function to be applied to the
depthwise kernel after being updated by an `Optimizer` (e.g. used for
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training.
pointwise_constraint: Optional projection function to be applied to the
pointwise kernel after being updated by an `Optimizer`.
bias_constraint: Optional projection function to be applied to the
bias after being updated by an `Optimizer`.
trainable: Boolean, if `True` the weights of this layer will be marked as
trainable (and listed in `layer.trainable_weights`).
name: A string, the name of the layer."
6165,build,tensorflow/tensorflow/python/keras/layers/convolutional.py,1801,method,
6166,call,tensorflow/tensorflow/python/keras/layers/convolutional.py,1844,method,
6167,get_config,tensorflow/tensorflow/python/keras/layers/convolutional.py,1847,method,
6168,SeparableConv1D,tensorflow/tensorflow/python/keras/layers/convolutional.py,1894,class,"Depthwise separable 1D convolution.
This layer performs a depthwise convolution that acts separately on
channels, followed by a pointwise convolution that mixes channels.
If `use_bias` is True and a bias initializer is provided,
it adds a bias vector to the output.
It then optionally applies an activation function to produce the final output.
Arguments:
filters: Integer, the dimensionality of the output space (i.e. the number
of filters in the convolution).
kernel_size: A single integer specifying the spatial
dimensions of the filters.
strides: A single integer specifying the strides
of the convolution.
Specifying any `stride` value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: One of `""valid""`, `""same""`, or `""causal""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input. `""causal""` results in causal
(dilated) convolutions, e.g. `output[t]` does not depend on `input[t+1:]`.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch_size, length, channels)` while `channels_first` corresponds to
inputs with shape `(batch_size, channels, length)`.
dilation_rate: A single integer, specifying
the dilation rate to use for dilated convolution.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any stride value != 1.
depth_multiplier: The number of depthwise convolution output channels for
each input channel. The total number of depthwise convolution output
channels will be equal to `num_filters_in * depth_multiplier`.
activation: Activation function to use.
If you don't specify anything, no activation is applied (
see `keras.activations`).
use_bias: Boolean, whether the layer uses a bias.
depthwise_initializer: An initializer for the depthwise convolution kernel (
see `keras.initializers`).
pointwise_initializer: An initializer for the pointwise convolution kernel (
see `keras.initializers`).
bias_initializer: An initializer for the bias vector. If None, the default
initializer will be used (see `keras.initializers`).
depthwise_regularizer: Optional regularizer for the depthwise
convolution kernel (see `keras.regularizers`).
pointwise_regularizer: Optional regularizer for the pointwise
convolution kernel (see `keras.regularizers`).
bias_regularizer: Optional regularizer for the bias vector (
see `keras.regularizers`).
activity_regularizer: Optional regularizer function for the output (
see `keras.regularizers`).
depthwise_constraint: Optional projection function to be applied to the
depthwise kernel after being updated by an `Optimizer` (e.g. used for
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training (
see `keras.constraints`).
pointwise_constraint: Optional projection function to be applied to the
pointwise kernel after being updated by an `Optimizer` (
see `keras.constraints`).
bias_constraint: Optional projection function to be applied to the
bias after being updated by an `Optimizer` (
see `keras.constraints`).
trainable: Boolean, if `True` the weights of this layer will be marked as
trainable (and listed in `layer.trainable_weights`).
name: A string, the name of the layer.
Input shape:
3D tensor with shape:
`(batch_size, channels, steps)` if data_format='channels_first'
or 5D tensor with shape:
`(batch_size, steps, channels)` if data_format='channels_last'.
Output shape:
3D tensor with shape:
`(batch_size, filters, new_steps)` if data_format='channels_first'
or 3D tensor with shape:
`(batch_size, new_steps, filters)` if data_format='channels_last'.
`new_steps` value might have changed due to padding or strides.
Returns:
A tensor of rank 3 representing
`activation(separableconv1d(inputs, kernel) + bias)`.
Raises:
ValueError: when both `strides` > 1 and `dilation_rate` > 1."
6169,call,tensorflow/tensorflow/python/keras/layers/convolutional.py,2029,method,
6170,SeparableConv2D,tensorflow/tensorflow/python/keras/layers/convolutional.py,2074,class,"Depthwise separable 2D convolution.
Separable convolutions consist of first performing
a depthwise spatial convolution
(which acts on each input channel separately)
followed by a pointwise convolution which mixes the resulting
output channels. The `depth_multiplier` argument controls how many
output channels are generated per input channel in the depthwise step.
Intuitively, separable convolutions can be understood as
a way to factorize a convolution kernel into two smaller kernels,
or as an extreme version of an Inception block.
Arguments:
filters: Integer, the dimensionality of the output space
(i.e. the number of output filters in the convolution).
kernel_size: An integer or tuple/list of 2 integers, specifying the
height and width of the 2D convolution window.
Can be a single integer to specify the same value for
all spatial dimensions.
strides: An integer or tuple/list of 2 integers,
specifying the strides of the convolution along the height and width.
Can be a single integer to specify the same value for
all spatial dimensions.
Specifying any stride value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: one of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch_size, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch_size, channels, height, width)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
dilation_rate: An integer or tuple/list of 2 integers, specifying
the dilation rate to use for dilated convolution.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any `strides` value != 1.
depth_multiplier: The number of depthwise convolution output channels
for each input channel.
The total number of depthwise convolution output
channels will be equal to `filters_in * depth_multiplier`.
activation: Activation function to use.
If you don't specify anything, no activation is applied (
see `keras.activations`).
use_bias: Boolean, whether the layer uses a bias vector.
depthwise_initializer: Initializer for the depthwise kernel matrix (
see `keras.initializers`).
pointwise_initializer: Initializer for the pointwise kernel matrix (
see `keras.initializers`).
bias_initializer: Initializer for the bias vector (
see `keras.initializers`).
depthwise_regularizer: Regularizer function applied to
the depthwise kernel matrix (see `keras.regularizers`).
pointwise_regularizer: Regularizer function applied to
the pointwise kernel matrix (see `keras.regularizers`).
bias_regularizer: Regularizer function applied to the bias vector (
see `keras.regularizers`).
activity_regularizer: Regularizer function applied to
the output of the layer (its ""activation"") (
see `keras.regularizers`).
depthwise_constraint: Constraint function applied to
the depthwise kernel matrix (
see `keras.constraints`).
pointwise_constraint: Constraint function applied to
the pointwise kernel matrix (
see `keras.constraints`).
bias_constraint: Constraint function applied to the bias vector (
see `keras.constraints`).
Input shape:
4D tensor with shape:
`(batch_size, channels, rows, cols)` if data_format='channels_first'
or 4D tensor with shape:
`(batch_size, rows, cols, channels)` if data_format='channels_last'.
Output shape:
4D tensor with shape:
`(batch_size, filters, new_rows, new_cols)` if data_format='channels_first'
or 4D tensor with shape:
`(batch_size, new_rows, new_cols, filters)` if data_format='channels_last'.
`rows` and `cols` values might have changed due to padding.
Returns:
A tensor of rank 4 representing
`activation(separableconv2d(inputs, kernel) + bias)`.
Raises:
ValueError: if `padding` is ""causal"".
ValueError: when both `strides` > 1 and `dilation_rate` > 1."
6171,call,tensorflow/tensorflow/python/keras/layers/convolutional.py,2217,method,
6172,DepthwiseConv2D,tensorflow/tensorflow/python/keras/layers/convolutional.py,2244,class,"Depthwise separable 2D convolution.
Depthwise Separable convolutions consist of performing
just the first step in a depthwise spatial convolution
(which acts on each input channel separately).
The `depth_multiplier` argument controls how many
output channels are generated per input channel in the depthwise step.
Arguments:
kernel_size: An integer or tuple/list of 2 integers, specifying the
height and width of the 2D convolution window.
Can be a single integer to specify the same value for
all spatial dimensions.
strides: An integer or tuple/list of 2 integers,
specifying the strides of the convolution along the height and width.
Can be a single integer to specify the same value for
all spatial dimensions.
Specifying any stride value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: one of `'valid'` or `'same'` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
depth_multiplier: The number of depthwise convolution output channels
for each input channel.
The total number of depthwise convolution output
channels will be equal to `filters_in * depth_multiplier`.
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch_size, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch_size, channels, height, width)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be 'channels_last'.
dilation_rate: An integer or tuple/list of 2 integers, specifying
the dilation rate to use for dilated convolution.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any `strides` value != 1.
activation: Activation function to use.
If you don't specify anything, no activation is applied (
see `keras.activations`).
use_bias: Boolean, whether the layer uses a bias vector.
depthwise_initializer: Initializer for the depthwise kernel matrix (
see `keras.initializers`).
bias_initializer: Initializer for the bias vector (
see `keras.initializers`).
depthwise_regularizer: Regularizer function applied to
the depthwise kernel matrix (see `keras.regularizers`).
bias_regularizer: Regularizer function applied to the bias vector (
see `keras.regularizers`).
activity_regularizer: Regularizer function applied to
the output of the layer (its 'activation') (
see `keras.regularizers`).
depthwise_constraint: Constraint function applied to
the depthwise kernel matrix (
see `keras.constraints`).
bias_constraint: Constraint function applied to the bias vector (
see `keras.constraints`).
Input shape:
4D tensor with shape:
`[batch_size, channels, rows, cols]` if data_format='channels_first'
or 4D tensor with shape:
`[batch_size, rows, cols, channels]` if data_format='channels_last'.
Output shape:
4D tensor with shape:
`[batch_size, filters, new_rows, new_cols]` if data_format='channels_first'
or 4D tensor with shape:
`[batch_size, new_rows, new_cols, filters]` if data_format='channels_last'.
`rows` and `cols` values might have changed due to padding.
Returns:
A tensor of rank 4 representing
`activation(depthwiseconv2d(inputs, kernel) + bias)`.
Raises:
ValueError: if `padding` is ""causal"".
ValueError: when both `strides` > 1 and `dilation_rate` > 1."
6173,build,tensorflow/tensorflow/python/keras/layers/convolutional.py,2365,method,
6174,call,tensorflow/tensorflow/python/keras/layers/convolutional.py,2400,method,
6175,compute_output_shape,tensorflow/tensorflow/python/keras/layers/convolutional.py,2421,method,
6176,get_config,tensorflow/tensorflow/python/keras/layers/convolutional.py,2444,method,
6177,UpSampling1D,tensorflow/tensorflow/python/keras/layers/convolutional.py,2461,class,"Upsampling layer for 1D inputs.
Repeats each temporal step `size` times along the time axis.
Examples:
>>> input_shape = (2, 2, 3)
>>> x = np.arange(np.prod(input_shape)).reshape(input_shape)
>>> print(x)
[[[ 0 1 2]
[ 3 4 5]]
[[ 6 7 8]
[ 9 10 11]]]
>>> y = tf.keras.layers.UpSampling1D(size=2)(x)
>>> print(y)
tf.Tensor(
[[[ 0 1 2]
[ 0 1 2]
[ 3 4 5]
[ 3 4 5]]
[[ 6 7 8]
[ 6 7 8]
[ 9 10 11]
[ 9 10 11]]], shape=(2, 4, 3), dtype=int64)
Arguments:
size: Integer. Upsampling factor.
Input shape:
3D tensor with shape: `(batch_size, steps, features)`.
Output shape:
3D tensor with shape: `(batch_size, upsampled_steps, features)`."
6178,compute_output_shape,tensorflow/tensorflow/python/keras/layers/convolutional.py,2502,method,
6179,call,tensorflow/tensorflow/python/keras/layers/convolutional.py,2507,method,
6180,get_config,tensorflow/tensorflow/python/keras/layers/convolutional.py,2511,method,
6181,UpSampling2D,tensorflow/tensorflow/python/keras/layers/convolutional.py,2518,class,"Upsampling layer for 2D inputs.
Repeats the rows and columns of the data
by `size[0]` and `size[1]` respectively.
Examples:
>>> input_shape = (2, 2, 1, 3)
>>> x = np.arange(np.prod(input_shape)).reshape(input_shape)
>>> print(x)
[[[[ 0 1 2]]
[[ 3 4 5]]]
[[[ 6 7 8]]
[[ 9 10 11]]]]
>>> y = tf.keras.layers.UpSampling2D(size=(1, 2))(x)
>>> print(y)
tf.Tensor(
[[[[ 0 1 2]
[ 0 1 2]]
[[ 3 4 5]
[ 3 4 5]]]
[[[ 6 7 8]
[ 6 7 8]]
[[ 9 10 11]
[ 9 10 11]]]], shape=(2, 2, 2, 3), dtype=int64)
Arguments:
size: Int, or tuple of 2 integers.
The upsampling factors for rows and columns.
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch_size, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch_size, channels, height, width)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
interpolation: A string, one of `nearest` or `bilinear`.
Input shape:
4D tensor with shape:
- If `data_format` is `""channels_last""`:
`(batch_size, rows, cols, channels)`
- If `data_format` is `""channels_first""`:
`(batch_size, channels, rows, cols)`
Output shape:
4D tensor with shape:
- If `data_format` is `""channels_last""`:
`(batch_size, upsampled_rows, upsampled_cols, channels)`
- If `data_format` is `""channels_first""`:
`(batch_size, channels, upsampled_rows, upsampled_cols)`"
6182,compute_output_shape,tensorflow/tensorflow/python/keras/layers/convolutional.py,2589,method,
6183,call,tensorflow/tensorflow/python/keras/layers/convolutional.py,2606,method,
6184,get_config,tensorflow/tensorflow/python/keras/layers/convolutional.py,2611,method,
6185,UpSampling3D,tensorflow/tensorflow/python/keras/layers/convolutional.py,2622,class,"Upsampling layer for 3D inputs.
Repeats the 1st, 2nd and 3rd dimensions
of the data by `size[0]`, `size[1]` and `size[2]` respectively.
Examples:
>>> input_shape = (2, 1, 2, 1, 3)
>>> x = tf.constant(1, shape=input_shape)
>>> y = tf.keras.layers.UpSampling3D(size=2)(x)
>>> print(y.shape)
(2, 2, 4, 2, 3)
Arguments:
size: Int, or tuple of 3 integers.
The upsampling factors for dim1, dim2 and dim3.
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch_size, spatial_dim1, spatial_dim2, spatial_dim3, channels)`
while `channels_first` corresponds to inputs with shape
`(batch_size, channels, spatial_dim1, spatial_dim2, spatial_dim3)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
Input shape:
5D tensor with shape:
- If `data_format` is `""channels_last""`:
`(batch_size, dim1, dim2, dim3, channels)`
- If `data_format` is `""channels_first""`:
`(batch_size, channels, dim1, dim2, dim3)`
Output shape:
5D tensor with shape:
- If `data_format` is `""channels_last""`:
`(batch_size, upsampled_dim1, upsampled_dim2, upsampled_dim3, channels)`
- If `data_format` is `""channels_first""`:
`(batch_size, channels, upsampled_dim1, upsampled_dim2, upsampled_dim3)`"
6186,compute_output_shape,tensorflow/tensorflow/python/keras/layers/convolutional.py,2671,method,
6187,call,tensorflow/tensorflow/python/keras/layers/convolutional.py,2692,method,
6188,get_config,tensorflow/tensorflow/python/keras/layers/convolutional.py,2696,method,
6189,ZeroPadding1D,tensorflow/tensorflow/python/keras/layers/convolutional.py,2703,class,"Zero-padding layer for 1D input (e.g. temporal sequence).
Examples:
>>> input_shape = (2, 2, 3)
>>> x = np.arange(np.prod(input_shape)).reshape(input_shape)
>>> print(x)
[[[ 0 1 2]
[ 3 4 5]]
[[ 6 7 8]
[ 9 10 11]]]
>>> y = tf.keras.layers.ZeroPadding1D(padding=2)(x)
>>> print(y)
tf.Tensor(
[[[ 0 0 0]
[ 0 0 0]
[ 0 1 2]
[ 3 4 5]
[ 0 0 0]
[ 0 0 0]]
[[ 0 0 0]
[ 0 0 0]
[ 6 7 8]
[ 9 10 11]
[ 0 0 0]
[ 0 0 0]]], shape=(2, 6, 3), dtype=int64)
Arguments:
padding: Int, or tuple of int (length 2), or dictionary.
- If int:
How many zeros to add at the beginning and end of
the padding dimension (axis 1).
- If tuple of int (length 2):
How many zeros to add at the beginning and the end of
the padding dimension (`(left_pad, right_pad)`).
Input shape:
3D tensor with shape `(batch_size, axis_to_pad, features)`
Output shape:
3D tensor with shape `(batch_size, padded_axis, features)`"
6190,compute_output_shape,tensorflow/tensorflow/python/keras/layers/convolutional.py,2752,method,
6191,call,tensorflow/tensorflow/python/keras/layers/convolutional.py,2759,method,
6192,get_config,tensorflow/tensorflow/python/keras/layers/convolutional.py,2762,method,
6193,ZeroPadding2D,tensorflow/tensorflow/python/keras/layers/convolutional.py,2769,class,"Zero-padding layer for 2D input (e.g. picture).
This layer can add rows and columns of zeros
at the top, bottom, left and right side of an image tensor.
Examples:
>>> input_shape = (1, 1, 2, 2)
>>> x = np.arange(np.prod(input_shape)).reshape(input_shape)
>>> print(x)
[[[[0 1]
[2 3]]]]
>>> y = tf.keras.layers.ZeroPadding2D(padding=1)(x)
>>> print(y)
tf.Tensor(
[[[[0 0]
[0 0]
[0 0]
[0 0]]
[[0 0]
[0 1]
[2 3]
[0 0]]
[[0 0]
[0 0]
[0 0]
[0 0]]]], shape=(1, 3, 4, 2), dtype=int64)
Arguments:
padding: Int, or tuple of 2 ints, or tuple of 2 tuples of 2 ints.
- If int: the same symmetric padding
is applied to height and width.
- If tuple of 2 ints:
interpreted as two different
symmetric padding values for height and width:
`(symmetric_height_pad, symmetric_width_pad)`.
- If tuple of 2 tuples of 2 ints:
interpreted as
`((top_pad, bottom_pad), (left_pad, right_pad))`
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch_size, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch_size, channels, height, width)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
Input shape:
4D tensor with shape:
- If `data_format` is `""channels_last""`:
`(batch_size, rows, cols, channels)`
- If `data_format` is `""channels_first""`:
`(batch_size, channels, rows, cols)`
Output shape:
4D tensor with shape:
- If `data_format` is `""channels_last""`:
`(batch_size, padded_rows, padded_cols, channels)`
- If `data_format` is `""channels_first""`:
`(batch_size, channels, padded_rows, padded_cols)`"
6194,compute_output_shape,tensorflow/tensorflow/python/keras/layers/convolutional.py,2858,method,
6195,call,tensorflow/tensorflow/python/keras/layers/convolutional.py,2883,method,
6196,get_config,tensorflow/tensorflow/python/keras/layers/convolutional.py,2887,method,
6197,ZeroPadding3D,tensorflow/tensorflow/python/keras/layers/convolutional.py,2894,class,"Zero-padding layer for 3D data (spatial or spatio-temporal).
Examples:
>>> input_shape = (1, 1, 2, 2, 3)
>>> x = np.arange(np.prod(input_shape)).reshape(input_shape)
>>> y = tf.keras.layers.ZeroPadding3D(padding=2)(x)
>>> print(y.shape)
(1, 5, 6, 6, 3)
Arguments:
padding: Int, or tuple of 3 ints, or tuple of 3 tuples of 2 ints.
- If int: the same symmetric padding
is applied to height and width.
- If tuple of 3 ints:
interpreted as two different
symmetric padding values for height and width:
`(symmetric_dim1_pad, symmetric_dim2_pad, symmetric_dim3_pad)`.
- If tuple of 3 tuples of 2 ints:
interpreted as
`((left_dim1_pad, right_dim1_pad), (left_dim2_pad,
right_dim2_pad), (left_dim3_pad, right_dim3_pad))`
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch_size, spatial_dim1, spatial_dim2, spatial_dim3, channels)`
while `channels_first` corresponds to inputs with shape
`(batch_size, channels, spatial_dim1, spatial_dim2, spatial_dim3)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
Input shape:
5D tensor with shape:
- If `data_format` is `""channels_last""`:
`(batch_size, first_axis_to_pad, second_axis_to_pad, third_axis_to_pad,
depth)`
- If `data_format` is `""channels_first""`:
`(batch_size, depth, first_axis_to_pad, second_axis_to_pad,
third_axis_to_pad)`
Output shape:
5D tensor with shape:
- If `data_format` is `""channels_last""`:
`(batch_size, first_padded_axis, second_padded_axis, third_axis_to_pad,
depth)`
- If `data_format` is `""channels_first""`:
`(batch_size, depth, first_padded_axis, second_padded_axis,
third_axis_to_pad)`"
6198,compute_output_shape,tensorflow/tensorflow/python/keras/layers/convolutional.py,2976,method,
6199,call,tensorflow/tensorflow/python/keras/layers/convolutional.py,3009,method,
6200,get_config,tensorflow/tensorflow/python/keras/layers/convolutional.py,3013,method,
6201,Cropping1D,tensorflow/tensorflow/python/keras/layers/convolutional.py,3020,class,"Cropping layer for 1D input (e.g. temporal sequence).
It crops along the time dimension (axis 1).
Examples:
>>> input_shape = (2, 3, 2)
>>> x = np.arange(np.prod(input_shape)).reshape(input_shape)
>>> print(x)
[[[ 0 1]
[ 2 3]
[ 4 5]]
[[ 6 7]
[ 8 9]
[10 11]]]
>>> y = tf.keras.layers.Cropping1D(cropping=1)(x)
>>> print(y)
tf.Tensor(
[[[2 3]]
[[8 9]]], shape=(2, 1, 2), dtype=int64)
Arguments:
cropping: Int or tuple of int (length 2)
How many units should be trimmed off at the beginning and end of
the cropping dimension (axis 1).
If a single int is provided, the same value will be used for both.
Input shape:
3D tensor with shape `(batch_size, axis_to_crop, features)`
Output shape:
3D tensor with shape `(batch_size, cropped_axis, features)`"
6202,compute_output_shape,tensorflow/tensorflow/python/keras/layers/convolutional.py,3060,method,
6203,call,tensorflow/tensorflow/python/keras/layers/convolutional.py,3068,method,
6204,get_config,tensorflow/tensorflow/python/keras/layers/convolutional.py,3074,method,
6205,Cropping2D,tensorflow/tensorflow/python/keras/layers/convolutional.py,3081,class,"Cropping layer for 2D input (e.g. picture).
It crops along spatial dimensions, i.e. height and width.
Examples:
>>> input_shape = (2, 28, 28, 3)
>>> x = np.arange(np.prod(input_shape)).reshape(input_shape)
>>> y = tf.keras.layers.Cropping2D(cropping=((2, 2), (4, 4)))(x)
>>> print(y.shape)
(2, 24, 20, 3)
Arguments:
cropping: Int, or tuple of 2 ints, or tuple of 2 tuples of 2 ints.
- If int: the same symmetric cropping
is applied to height and width.
- If tuple of 2 ints:
interpreted as two different
symmetric cropping values for height and width:
`(symmetric_height_crop, symmetric_width_crop)`.
- If tuple of 2 tuples of 2 ints:
interpreted as
`((top_crop, bottom_crop), (left_crop, right_crop))`
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch_size, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch_size, channels, height, width)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
Input shape:
4D tensor with shape:
- If `data_format` is `""channels_last""`:
`(batch_size, rows, cols, channels)`
- If `data_format` is `""channels_first""`:
`(batch_size, channels, rows, cols)`
Output shape:
4D tensor with shape:
- If `data_format` is `""channels_last""`:
`(batch_size, cropped_rows, cropped_cols, channels)`
- If `data_format` is `""channels_first""`:
`(batch_size, channels, cropped_rows, cropped_cols)`"
6206,compute_output_shape,tensorflow/tensorflow/python/keras/layers/convolutional.py,3154,method,
6207,call,tensorflow/tensorflow/python/keras/layers/convolutional.py,3175,method,
6208,get_config,tensorflow/tensorflow/python/keras/layers/convolutional.py,3201,method,
6209,Cropping3D,tensorflow/tensorflow/python/keras/layers/convolutional.py,3208,class,"Cropping layer for 3D data (e.g. spatial or spatio-temporal).
Examples:
>>> input_shape = (2, 28, 28, 10, 3)
>>> x = np.arange(np.prod(input_shape)).reshape(input_shape)
>>> y = tf.keras.layers.Cropping3D(cropping=(2, 4, 2))(x)
>>> print(y.shape)
(2, 24, 20, 6, 3)
Arguments:
cropping: Int, or tuple of 3 ints, or tuple of 3 tuples of 2 ints.
- If int: the same symmetric cropping
is applied to depth, height, and width.
- If tuple of 3 ints: interpreted as two different
symmetric cropping values for depth, height, and width:
`(symmetric_dim1_crop, symmetric_dim2_crop, symmetric_dim3_crop)`.
- If tuple of 3 tuples of 2 ints: interpreted as
`((left_dim1_crop, right_dim1_crop), (left_dim2_crop,
right_dim2_crop), (left_dim3_crop, right_dim3_crop))`
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch_size, spatial_dim1, spatial_dim2, spatial_dim3, channels)`
while `channels_first` corresponds to inputs with shape
`(batch_size, channels, spatial_dim1, spatial_dim2, spatial_dim3)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
Input shape:
5D tensor with shape:
- If `data_format` is `""channels_last""`:
`(batch_size, first_axis_to_crop, second_axis_to_crop, third_axis_to_crop,
depth)`
- If `data_format` is `""channels_first""`:
`(batch_size, depth, first_axis_to_crop, second_axis_to_crop,
third_axis_to_crop)`
Output shape:
5D tensor with shape:
- If `data_format` is `""channels_last""`:
`(batch_size, first_cropped_axis, second_cropped_axis, third_cropped_axis,
depth)`
- If `data_format` is `""channels_first""`:
`(batch_size, depth, first_cropped_axis, second_cropped_axis,
third_cropped_axis)`"
6210,compute_output_shape,tensorflow/tensorflow/python/keras/layers/convolutional.py,3291,method,
6211,call,tensorflow/tensorflow/python/keras/layers/convolutional.py,3326,method,
6212,get_config,tensorflow/tensorflow/python/keras/layers/convolutional.py,3383,method,
6213,ConvRNN2D,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,40,class,"Base class for convolutional-recurrent layers.
Arguments:
cell: A RNN cell instance. A RNN cell is a class that has:
- a `call(input_at_t, states_at_t)` method, returning
`(output_at_t, states_at_t_plus_1)`. The call method of the
cell can also take the optional argument `constants`, see
section ""Note on passing external constants"" below.
- a `state_size` attribute. This can be a single integer
(single state) in which case it is
the number of channels of the recurrent state
(which should be the same as the number of channels of the cell
output). This can also be a list/tuple of integers
(one size per state). In this case, the first entry
(`state_size[0]`) should be the same as
the size of the cell output.
return_sequences: Boolean. Whether to return the last output.
in the output sequence, or the full sequence.
return_state: Boolean. Whether to return the last state
in addition to the output.
go_backwards: Boolean (default False).
If True, process the input sequence backwards and return the
reversed sequence.
stateful: Boolean (default False). If True, the last state
for each sample at index i in a batch will be used as initial
state for the sample of index i in the following batch.
input_shape: Use this argument to specify the shape of the
input when this layer is the first one in a model.
Call arguments:
inputs: A 5D tensor.
mask: Binary tensor of shape `(samples, timesteps)` indicating whether
a given timestep should be masked.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. This argument is passed to the cell
when calling it. This is for use with cells that use dropout.
initial_state: List of initial state tensors to be passed to the first
call of the cell.
constants: List of constant tensors to be passed to the cell at each
timestep.
Input shape:
5D tensor with shape:
`(samples, timesteps, channels, rows, cols)`
if data_format='channels_first' or 5D tensor with shape:
`(samples, timesteps, rows, cols, channels)`
if data_format='channels_last'.
Output shape:
- If `return_state`: a list of tensors. The first tensor is
the output. The remaining tensors are the last states,
each 4D tensor with shape:
`(samples, filters, new_rows, new_cols)`
if data_format='channels_first'
or 4D tensor with shape:
`(samples, new_rows, new_cols, filters)`
if data_format='channels_last'.
`rows` and `cols` values might have changed due to padding.
- If `return_sequences`: 5D tensor with shape:
`(samples, timesteps, filters, new_rows, new_cols)`
if data_format='channels_first'
or 5D tensor with shape:
`(samples, timesteps, new_rows, new_cols, filters)`
if data_format='channels_last'.
- Else, 4D tensor with shape:
`(samples, filters, new_rows, new_cols)`
if data_format='channels_first'
or 4D tensor with shape:
`(samples, new_rows, new_cols, filters)`
if data_format='channels_last'.
Masking:
This layer supports masking for input data with a variable number
of timesteps.
Note on using statefulness in RNNs:
You can set RNN layers to be 'stateful', which means that the states
computed for the samples in one batch will be reused as initial states
for the samples in the next batch. This assumes a one-to-one mapping
between samples in different successive batches.
To enable statefulness:
- Specify `stateful=True` in the layer constructor.
- Specify a fixed batch size for your model, by passing
- If sequential model:
`batch_input_shape=(...)` to the first layer in your model.
- If functional model with 1 or more Input layers:
`batch_shape=(...)` to all the first layers in your model.
This is the expected shape of your inputs
*including the batch size*.
It should be a tuple of integers,
e.g. `(32, 10, 100, 100, 32)`.
Note that the number of rows and columns should be specified
too.
- Specify `shuffle=False` when calling fit().
To reset the states of your model, call `.reset_states()` on either
a specific layer, or on your entire model.
Note on specifying the initial state of RNNs:
You can specify the initial state of RNN layers symbolically by
calling them with the keyword argument `initial_state`. The value of
`initial_state` should be a tensor or list of tensors representing
the initial state of the RNN layer.
You can specify the initial state of RNN layers numerically by
calling `reset_states` with the keyword argument `states`. The value of
`states` should be a numpy array or list of numpy arrays representing
the initial state of the RNN layer.
Note on passing external constants to RNNs:
You can pass ""external"" constants to the cell using the `constants`
keyword argument of `RNN.__call__` (as well as `RNN.call`) method. This
requires that the `cell.call` method accepts the same keyword argument
`constants`. Such constants can be used to condition the cell
transformation on additional static inputs (not changing over time),
a.k.a. an attention mechanism."
6214,compute_output_shape,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,184,method,
6215,build,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,225,method,
6216,get_initial_state,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,277,method,
6217,call,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,294,method,
6218,reset_states,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,353,method,
6219,get_tuple_shape,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,377,method,
6220,step,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,317,method,
6221,step,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,322,method,
6222,ConvLSTM2DCell,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,424,class,"Cell class for the ConvLSTM2D layer.
Arguments:
filters: Integer, the dimensionality of the output space
(i.e. the number of output filters in the convolution).
kernel_size: An integer or tuple/list of n integers, specifying the
dimensions of the convolution window.
strides: An integer or tuple/list of n integers,
specifying the strides of the convolution.
Specifying any stride value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string,
one of `channels_last` (default) or `channels_first`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
dilation_rate: An integer or tuple/list of n integers, specifying
the dilation rate to use for dilated convolution.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any `strides` value != 1.
activation: Activation function to use.
If you don't specify anything, no activation is applied
(ie. ""linear"" activation: `a(x) = x`).
recurrent_activation: Activation function to use
for the recurrent step.
use_bias: Boolean, whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix,
used for the linear transformation of the inputs.
recurrent_initializer: Initializer for the `recurrent_kernel`
weights matrix,
used for the linear transformation of the recurrent state.
bias_initializer: Initializer for the bias vector.
unit_forget_bias: Boolean.
If True, add 1 to the bias of the forget gate at initialization.
Use in combination with `bias_initializer=""zeros""`.
This is recommended in [Jozefowicz et al., 2015](
http://www.jmlr.org/proceedings/papers/v37/jozefowicz15.pdf)
kernel_regularizer: Regularizer function applied to
the `kernel` weights matrix.
recurrent_regularizer: Regularizer function applied to
the `recurrent_kernel` weights matrix.
bias_regularizer: Regularizer function applied to the bias vector.
kernel_constraint: Constraint function applied to
the `kernel` weights matrix.
recurrent_constraint: Constraint function applied to
the `recurrent_kernel` weights matrix.
bias_constraint: Constraint function applied to the bias vector.
dropout: Float between 0 and 1.
Fraction of the units to drop for
the linear transformation of the inputs.
recurrent_dropout: Float between 0 and 1.
Fraction of the units to drop for
the linear transformation of the recurrent state.
Call arguments:
inputs: A 4D tensor.
states: List of state tensors corresponding to the previous timestep.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. Only relevant when `dropout` or
`recurrent_dropout` is used."
6223,build,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,543,method,
6224,call,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,590,method,
6225,input_conv,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,650,method,
6226,recurrent_conv,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,660,method,
6227,get_config,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,666,method,
6228,bias_initializer,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,572,method,
6229,ConvLSTM2D,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,700,class,"Convolutional LSTM.
It is similar to an LSTM layer, but the input transformations
and recurrent transformations are both convolutional.
Arguments:
filters: Integer, the dimensionality of the output space
(i.e. the number of output filters in the convolution).
kernel_size: An integer or tuple/list of n integers, specifying the
dimensions of the convolution window.
strides: An integer or tuple/list of n integers,
specifying the strides of the convolution.
Specifying any stride value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, time, ..., channels)`
while `channels_first` corresponds to
inputs with shape `(batch, time, channels, ...)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
dilation_rate: An integer or tuple/list of n integers, specifying
the dilation rate to use for dilated convolution.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any `strides` value != 1.
activation: Activation function to use.
By default hyperbolic tangent activation function is applied
(`tanh(x)`).
recurrent_activation: Activation function to use
for the recurrent step.
use_bias: Boolean, whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix,
used for the linear transformation of the inputs.
recurrent_initializer: Initializer for the `recurrent_kernel`
weights matrix,
used for the linear transformation of the recurrent state.
bias_initializer: Initializer for the bias vector.
unit_forget_bias: Boolean.
If True, add 1 to the bias of the forget gate at initialization.
Use in combination with `bias_initializer=""zeros""`.
This is recommended in [Jozefowicz et al., 2015](
http://www.jmlr.org/proceedings/papers/v37/jozefowicz15.pdf)
kernel_regularizer: Regularizer function applied to
the `kernel` weights matrix.
recurrent_regularizer: Regularizer function applied to
the `recurrent_kernel` weights matrix.
bias_regularizer: Regularizer function applied to the bias vector.
activity_regularizer: Regularizer function applied to.
kernel_constraint: Constraint function applied to
the `kernel` weights matrix.
recurrent_constraint: Constraint function applied to
the `recurrent_kernel` weights matrix.
bias_constraint: Constraint function applied to the bias vector.
return_sequences: Boolean. Whether to return the last output
in the output sequence, or the full sequence. (default False)
return_state: Boolean Whether to return the last state
in addition to the output. (default False)
go_backwards: Boolean (default False).
If True, process the input sequence backwards.
stateful: Boolean (default False). If True, the last state
for each sample at index i in a batch will be used as initial
state for the sample of index i in the following batch.
dropout: Float between 0 and 1.
Fraction of the units to drop for
the linear transformation of the inputs.
recurrent_dropout: Float between 0 and 1.
Fraction of the units to drop for
the linear transformation of the recurrent state.
Call arguments:
inputs: A 5D tensor.
mask: Binary tensor of shape `(samples, timesteps)` indicating whether
a given timestep should be masked.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. This argument is passed to the cell
when calling it. This is only relevant if `dropout` or `recurrent_dropout`
are set.
initial_state: List of initial state tensors to be passed to the first
call of the cell.
Input shape:
- If data_format='channels_first'
5D tensor with shape:
`(samples, time, channels, rows, cols)`
- If data_format='channels_last'
5D tensor with shape:
`(samples, time, rows, cols, channels)`
Output shape:
- If `return_state`: a list of tensors. The first tensor is
the output. The remaining tensors are the last states,
each 4D tensor with shape:
`(samples, filters, new_rows, new_cols)`
if data_format='channels_first'
or 4D tensor with shape:
`(samples, new_rows, new_cols, filters)`
if data_format='channels_last'.
`rows` and `cols` values might have changed due to padding.
- If `return_sequences`: 5D tensor with shape:
`(samples, timesteps, filters, new_rows, new_cols)`
if data_format='channels_first'
or 5D tensor with shape:
`(samples, timesteps, new_rows, new_cols, filters)`
if data_format='channels_last'.
- Else, 4D tensor with shape:
`(samples, filters, new_rows, new_cols)`
if data_format='channels_first'
or 4D tensor with shape:
`(samples, new_rows, new_cols, filters)`
if data_format='channels_last'.
Raises:
ValueError: in case of invalid constructor arguments.
References:
- [Shi et al., 2015](http://arxiv.org/abs/1506.04214v1)
(the current implementation does not include the feedback loop on the
cells output)."
6230,call,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,886,method,
6231,filters,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,894,method,
6232,kernel_size,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,898,method,
6233,strides,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,902,method,
6234,padding,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,906,method,
6235,data_format,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,910,method,
6236,dilation_rate,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,914,method,
6237,activation,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,918,method,
6238,recurrent_activation,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,922,method,
6239,use_bias,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,926,method,
6240,kernel_initializer,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,930,method,
6241,recurrent_initializer,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,934,method,
6242,bias_initializer,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,938,method,
6243,unit_forget_bias,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,942,method,
6244,kernel_regularizer,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,946,method,
6245,recurrent_regularizer,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,950,method,
6246,bias_regularizer,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,954,method,
6247,kernel_constraint,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,958,method,
6248,recurrent_constraint,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,962,method,
6249,bias_constraint,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,966,method,
6250,dropout,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,970,method,
6251,recurrent_dropout,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,974,method,
6252,get_config,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,977,method,
6253,from_config,tensorflow/tensorflow/python/keras/layers/convolutional_recurrent.py,1013,method,
6254,Masking,tensorflow/tensorflow/python/keras/layers/core.py,67,class,"Masks a sequence by using a mask value to skip timesteps.
For each timestep in the input tensor (dimension #1 in the tensor),
if all values in the input tensor at that timestep
are equal to `mask_value`, then the timestep will be masked (skipped)
in all downstream layers (as long as they support masking).
If any downstream layer does not support masking yet receives such
an input mask, an exception will be raised.
Example:
Consider a Numpy data array `x` of shape `(samples, timesteps, features)`,
to be fed to an LSTM layer. You want to mask timestep #3 and #5 because you
lack data for these timesteps. You can:
- Set `x[:, 3, :] = 0.` and `x[:, 5, :] = 0.`
- Insert a `Masking` layer with `mask_value=0.` before the LSTM layer:
```python
samples, timesteps, features = 32, 10, 8
inputs = np.random.random([samples, timesteps, features]).astype(np.float32)
inputs[:, 3, :] = 0.
inputs[:, 5, :] = 0.
model = tf.keras.models.Sequential()
model.add(tf.keras.layers.Masking(mask_value=0.,
input_shape=(timesteps, features)))
model.add(tf.keras.layers.LSTM(32))
output = model(inputs)
# The time step 3 and 5 will be skipped from LSTM calculation.
```
See [the masking and padding guide](
https://www.tensorflow.org/guide/keras/masking_and_padding)
for more details."
6255,compute_mask,tensorflow/tensorflow/python/keras/layers/core.py,113,method,
6256,call,tensorflow/tensorflow/python/keras/layers/core.py,116,method,
6257,compute_output_shape,tensorflow/tensorflow/python/keras/layers/core.py,124,method,
6258,get_config,tensorflow/tensorflow/python/keras/layers/core.py,127,method,
6259,Dropout,tensorflow/tensorflow/python/keras/layers/core.py,134,class,"Applies Dropout to the input.
The Dropout layer randomly sets input units to 0 with a frequency of `rate`
at each step during training time, which helps prevent overfitting.
Inputs not set to 0 are scaled up by 1/(1 - rate) such that the sum over
all inputs is unchanged.
Note that the Dropout layer only applies when `training` is set to True
such that no values are dropped during inference. When using `model.fit`,
`training` will be appropriately set to True automatically, and in other
contexts, you can set the kwarg explicitly to True when calling the layer.
(This is in contrast to setting `trainable=False` for a Dropout layer.
`trainable` does not affect the layer's behavior, as Dropout does
not have any variables/weights that can be frozen during training.)
>>> tf.random.set_seed(0)
>>> layer = tf.keras.layers.Dropout(.2, input_shape=(2,))
>>> data = np.arange(10).reshape(5, 2).astype(np.float32)
>>> print(data)
[[0. 1.]
[2. 3.]
[4. 5.]
[6. 7.]
[8. 9.]]
>>> outputs = layer(data, training=True)
>>> print(outputs)
tf.Tensor(
[[ 0. 1.25]
[ 2.5 3.75]
[ 5. 6.25]
[ 7.5 8.75]
[10. 0. ]], shape=(5, 2), dtype=float32)
Arguments:
rate: Float between 0 and 1. Fraction of the input units to drop.
noise_shape: 1D integer tensor representing the shape of the
binary dropout mask that will be multiplied with the input.
For instance, if your inputs have shape
`(batch_size, timesteps, features)` and
you want the dropout mask to be the same for all timesteps,
you can use `noise_shape=(batch_size, 1, features)`.
seed: A Python integer to use as random seed.
Call arguments:
inputs: Input tensor (of any rank).
training: Python boolean indicating whether the layer should behave in
training mode (adding dropout) or in inference mode (doing nothing)."
6260,call,tensorflow/tensorflow/python/keras/layers/core.py,205,method,
6261,compute_output_shape,tensorflow/tensorflow/python/keras/layers/core.py,221,method,
6262,get_config,tensorflow/tensorflow/python/keras/layers/core.py,224,method,
6263,dropped_inputs,tensorflow/tensorflow/python/keras/layers/core.py,209,method,
6264,SpatialDropout1D,tensorflow/tensorflow/python/keras/layers/core.py,235,class,"Spatial 1D version of Dropout.
This version performs the same function as Dropout, however, it drops
entire 1D feature maps instead of individual elements. If adjacent frames
within feature maps are strongly correlated (as is normally the case in
early convolution layers) then regular dropout will not regularize the
activations and will otherwise just result in an effective learning rate
decrease. In this case, SpatialDropout1D will help promote independence
between feature maps and should be used instead.
Arguments:
rate: Float between 0 and 1. Fraction of the input units to drop.
Call arguments:
inputs: A 3D tensor.
training: Python boolean indicating whether the layer should behave in
training mode (adding dropout) or in inference mode (doing nothing).
Input shape:
3D tensor with shape:
`(samples, timesteps, channels)`
Output shape:
Same as input.
References:
- [Efficient Object Localization Using Convolutional
Networks](https://arxiv.org/abs/1411.4280)"
6265,SpatialDropout2D,tensorflow/tensorflow/python/keras/layers/core.py,277,class,"Spatial 2D version of Dropout.
This version performs the same function as Dropout, however, it drops
entire 2D feature maps instead of individual elements. If adjacent pixels
within feature maps are strongly correlated (as is normally the case in
early convolution layers) then regular dropout will not regularize the
activations and will otherwise just result in an effective learning rate
decrease. In this case, SpatialDropout2D will help promote independence
between feature maps and should be used instead.
Arguments:
rate: Float between 0 and 1. Fraction of the input units to drop.
data_format: 'channels_first' or 'channels_last'.
In 'channels_first' mode, the channels dimension
(the depth) is at index 1,
in 'channels_last' mode is it at index 3.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
Call arguments:
inputs: A 4D tensor.
training: Python boolean indicating whether the layer should behave in
training mode (adding dropout) or in inference mode (doing nothing).
Input shape:
4D tensor with shape:
`(samples, channels, rows, cols)` if data_format='channels_first'
or 4D tensor with shape:
`(samples, rows, cols, channels)` if data_format='channels_last'.
Output shape:
Same as input.
References:
- [Efficient Object Localization Using Convolutional
Networks](https://arxiv.org/abs/1411.4280)"
6266,SpatialDropout3D,tensorflow/tensorflow/python/keras/layers/core.py,336,class,"Spatial 3D version of Dropout.
This version performs the same function as Dropout, however, it drops
entire 3D feature maps instead of individual elements. If adjacent voxels
within feature maps are strongly correlated (as is normally the case in
early convolution layers) then regular dropout will not regularize the
activations and will otherwise just result in an effective learning rate
decrease. In this case, SpatialDropout3D will help promote independence
between feature maps and should be used instead.
Arguments:
rate: Float between 0 and 1. Fraction of the input units to drop.
data_format: 'channels_first' or 'channels_last'.
In 'channels_first' mode, the channels dimension (the depth)
is at index 1, in 'channels_last' mode is it at index 4.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
Call arguments:
inputs: A 5D tensor.
training: Python boolean indicating whether the layer should behave in
training mode (adding dropout) or in inference mode (doing nothing).
Input shape:
5D tensor with shape:
`(samples, channels, dim1, dim2, dim3)` if data_format='channels_first'
or 5D tensor with shape:
`(samples, dim1, dim2, dim3, channels)` if data_format='channels_last'.
Output shape:
Same as input.
References:
- [Efficient Object Localization Using Convolutional
Networks](https://arxiv.org/abs/1411.4280)"
6267,Activation,tensorflow/tensorflow/python/keras/layers/core.py,394,class,"Applies an activation function to an output.
Arguments:
activation: Activation function, such as `tf.nn.relu`, or string name of
built-in activation function, such as ""relu"".
Usage:
>>> layer = tf.keras.layers.Activation('relu')
>>> output = layer([-3.0, -1.0, 0.0, 2.0])
>>> list(output.numpy())
[0.0, 0.0, 0.0, 2.0]
>>> layer = tf.keras.layers.Activation(tf.nn.relu)
>>> output = layer([-3.0, -1.0, 0.0, 2.0])
>>> list(output.numpy())
[0.0, 0.0, 0.0, 2.0]
Input shape:
Arbitrary. Use the keyword argument `input_shape`
(tuple of integers, does not include the batch axis)
when using this layer as the first layer in a model.
Output shape:
Same shape as input."
6268,call,tensorflow/tensorflow/python/keras/layers/core.py,426,method,
6269,compute_output_shape,tensorflow/tensorflow/python/keras/layers/core.py,429,method,
6270,get_config,tensorflow/tensorflow/python/keras/layers/core.py,432,method,
6271,Reshape,tensorflow/tensorflow/python/keras/layers/core.py,439,class,"Layer that reshapes inputs into the given shape.
Input shape:
Arbitrary, although all dimensions in the input shape must be known/fixed.
Use the keyword argument `input_shape` (tuple of integers, does not include
the samples/batch size axis) when using this layer as the first layer
in a model.
Output shape:
`(batch_size,) + target_shape`
Example:
>>> # as first layer in a Sequential model
>>> model = tf.keras.Sequential()
>>> model.add(tf.keras.layers.Reshape((3, 4), input_shape=(12,)))
>>> # model.output_shape == (None, 3, 4), `None` is the batch size.
>>> model.output_shape
(None, 3, 4)
>>> # as intermediate layer in a Sequential model
>>> model.add(tf.keras.layers.Reshape((6, 2)))
>>> model.output_shape
(None, 6, 2)
>>> # also supports shape inference using `-1` as dimension
>>> model.add(tf.keras.layers.Reshape((-1, 2, 2)))
>>> model.output_shape
(None, 3, 2, 2)"
6272,compute_output_shape,tensorflow/tensorflow/python/keras/layers/core.py,526,method,
6273,call,tensorflow/tensorflow/python/keras/layers/core.py,538,method,
6274,get_config,tensorflow/tensorflow/python/keras/layers/core.py,547,method,
6275,Permute,tensorflow/tensorflow/python/keras/layers/core.py,554,class,"Permutes the dimensions of the input according to a given pattern.
Useful e.g. connecting RNNs and convnets.
Example:
```python
model = Sequential()
model.add(Permute((2, 1), input_shape=(10, 64)))
# now: model.output_shape == (None, 64, 10)
# note: `None` is the batch dimension
```
Arguments:
dims: Tuple of integers. Permutation pattern does not include the
samples dimension. Indexing starts at 1.
For instance, `(2, 1)` permutes the first and second dimensions
of the input.
Input shape:
Arbitrary. Use the keyword argument `input_shape`
(tuple of integers, does not include the samples axis)
when using this layer as the first layer in a model.
Output shape:
Same as the input shape, but with the dimensions re-ordered according
to the specified pattern."
6276,compute_output_shape,tensorflow/tensorflow/python/keras/layers/core.py,594,method,
6277,call,tensorflow/tensorflow/python/keras/layers/core.py,602,method,
6278,get_config,tensorflow/tensorflow/python/keras/layers/core.py,605,method,
6279,Flatten,tensorflow/tensorflow/python/keras/layers/core.py,612,class,"Flattens the input. Does not affect the batch size.
Note: If inputs are shaped `(batch,)` without a feature axis, then
flattening adds an extra channel dimension and output shape is `(batch, 1)`.
Arguments:
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, ..., channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, ...)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
Example:
>>> model = tf.keras.Sequential()
>>> model.add(tf.keras.layers.Conv2D(64, 3, 3, input_shape=(3, 32, 32)))
>>> model.output_shape
(None, 1, 10, 64)
>>> model.add(Flatten())
>>> model.output_shape
(None, 640)"
6280,call,tensorflow/tensorflow/python/keras/layers/core.py,648,method,
6281,compute_output_shape,tensorflow/tensorflow/python/keras/layers/core.py,681,method,
6282,get_config,tensorflow/tensorflow/python/keras/layers/core.py,693,method,
6283,RepeatVector,tensorflow/tensorflow/python/keras/layers/core.py,700,class,"Repeats the input n times.
Example:
```python
model = Sequential()
model.add(Dense(32, input_dim=32))
# now: model.output_shape == (None, 32)
# note: `None` is the batch dimension
model.add(RepeatVector(3))
# now: model.output_shape == (None, 3, 32)
```
Arguments:
n: Integer, repetition factor.
Input shape:
2D tensor of shape `(num_samples, features)`.
Output shape:
3D tensor of shape `(num_samples, n, features)`."
6284,compute_output_shape,tensorflow/tensorflow/python/keras/layers/core.py,730,method,
6285,call,tensorflow/tensorflow/python/keras/layers/core.py,734,method,
6286,get_config,tensorflow/tensorflow/python/keras/layers/core.py,737,method,
6287,Lambda,tensorflow/tensorflow/python/keras/layers/core.py,744,class,"Wraps arbitrary expressions as a `Layer` object.
The `Lambda` layer exists so that arbitrary TensorFlow functions
can be used when constructing `Sequential` and Functional API
models. `Lambda` layers are best suited for simple operations or
quick experimentation. For more advanced use cases, follow
[this guide](https://www.tensorflow.org/guide/keras/custom_layers_and_models)
for subclassing `tf.keras.layers.Layer`.
The main reason to subclass `tf.keras.layers.Layer` instead of using a
`Lambda` layer is saving and inspecting a Model. `Lambda` layers
are saved by serializing the Python bytecode, whereas subclassed
Layers can be saved via overriding their `get_config` method. Overriding
`get_config` improves the portability of Models. Models that rely on
subclassed Layers are also often easier to visualize and reason about.
Examples:
```python
# add a x -> x^2 layer
model.add(Lambda(lambda x: x ** 2))
```
```python
# add a layer that returns the concatenation
# of the positive part of the input and
# the opposite of the negative part
def antirectifier(x):
x -= K.mean(x, axis=1, keepdims=True)
x = K.l2_normalize(x, axis=1)
pos = K.relu(x)
neg = K.relu(-x)
return K.concatenate([pos, neg], axis=1)
model.add(Lambda(antirectifier))
```
Variables:
While it is possible to use Variables with Lambda layers, this practice is
discouraged as it can easily lead to bugs. For instance, consider the
following layer:
```python
scale = tf.Variable(1.)
scale_layer = tf.keras.layers.Lambda(lambda x: x * scale)
```
Because scale_layer does not directly track the `scale` variable, it will
not appear in `scale_layer.trainable_weights` and will therefore not be
trained if `scale_layer` is used in a Model.
A better pattern is to write a subclassed Layer:
```python
class ScaleLayer(tf.keras.layers.Layer):
def __init__(self):
super(ScaleLayer, self).__init__()
self.scale = tf.Variable(1.)
def call(self, inputs):
return inputs * self.scale
```
In general, Lambda layers can be convenient for simple stateless
computation, but anything more complex should use a subclass Layer instead.
Arguments:
function: The function to be evaluated. Takes input tensor as first
argument.
output_shape: Expected output shape from function. This argument can be
inferred if not explicitly provided. Can be a tuple or function. If a
tuple, it only specifies the first dimension onward;
sample dimension is assumed either the same as the input: `output_shape =
(input_shape[0], ) + output_shape` or, the input is `None` and
the sample dimension is also `None`: `output_shape = (None, ) +
output_shape` If a function, it specifies the entire shape as a function
of the
input shape: `output_shape = f(input_shape)`
mask: Either None (indicating no masking) or a callable with the same
signature as the `compute_mask` layer method, or a tensor that will be
returned as output mask regardless of what the input is.
arguments: Optional dictionary of keyword arguments to be passed to the
function.
Input shape:
Arbitrary. Use the keyword argument input_shape (tuple of
integers, does not include the samples axis) when using this layer as the
first layer in a model.
Output shape:
Specified by `output_shape` argument"
6288,compute_output_shape,tensorflow/tensorflow/python/keras/layers/core.py,859,method,
6289,call,tensorflow/tensorflow/python/keras/layers/core.py,887,method,
6290,compute_mask,tensorflow/tensorflow/python/keras/layers/core.py,954,method,
6291,get_config,tensorflow/tensorflow/python/keras/layers/core.py,959,method,
6292,from_config,tensorflow/tensorflow/python/keras/layers/core.py,1003,method,
6293,Dense,tensorflow/tensorflow/python/keras/layers/core.py,1067,class,"Just your regular densely-connected NN layer.
`Dense` implements the operation:
`output = activation(dot(input, kernel) + bias)`
where `activation` is the element-wise activation function
passed as the `activation` argument, `kernel` is a weights matrix
created by the layer, and `bias` is a bias vector created by the layer
(only applicable if `use_bias` is `True`).
Note: If the input to the layer has a rank greater than 2, then `Dense`
computes the dot product between the `inputs` and the `kernel` along the
last axis of the `inputs` and axis 1 of the `kernel` (using `tf.tensordot`).
For example, if input has dimensions `(batch_size, d0, d1)`,
then we create a `kernel` with shape `(d1, units)`, and the `kernel` operates
along axis 2 of the `input`, on every sub-tensor of shape `(1, 1, d1)`
(there are `batch_size * d0` such sub-tensors).
The output in this case will have shape `(batch_size, d0, units)`.
Besides, layer attributes cannot be modified after the layer has been called
once (except the `trainable` attribute).
Example:
>>> # Create a `Sequential` model and add a Dense layer as the first layer.
>>> model = tf.keras.models.Sequential()
>>> model.add(tf.keras.Input(shape=(16,)))
>>> model.add(tf.keras.layers.Dense(32, activation='relu'))
>>> # Now the model will take as input arrays of shape (None, 16)
>>> # and output arrays of shape (None, 32).
>>> # Note that after the first layer, you don't need to specify
>>> # the size of the input anymore:
>>> model.add(tf.keras.layers.Dense(32))
>>> model.output_shape
(None, 32)
Arguments:
units: Positive integer, dimensionality of the output space.
activation: Activation function to use.
If you don't specify anything, no activation is applied
(ie. ""linear"" activation: `a(x) = x`).
use_bias: Boolean, whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix.
bias_initializer: Initializer for the bias vector.
kernel_regularizer: Regularizer function applied to
the `kernel` weights matrix.
bias_regularizer: Regularizer function applied to the bias vector.
activity_regularizer: Regularizer function applied to
the output of the layer (its ""activation"").
kernel_constraint: Constraint function applied to
the `kernel` weights matrix.
bias_constraint: Constraint function applied to the bias vector.
Input shape:
N-D tensor with shape: `(batch_size, ..., input_dim)`.
The most common situation would be
a 2D input with shape `(batch_size, input_dim)`.
Output shape:
N-D tensor with shape: `(batch_size, ..., units)`.
For instance, for a 2D input with shape `(batch_size, input_dim)`,
the output would have shape `(batch_size, units)`."
6294,build,tensorflow/tensorflow/python/keras/layers/core.py,1159,method,
6295,call,tensorflow/tensorflow/python/keras/layers/core.py,1192,method,
6296,compute_output_shape,tensorflow/tensorflow/python/keras/layers/core.py,1200,method,
6297,get_config,tensorflow/tensorflow/python/keras/layers/core.py,1209,method,
6298,ActivityRegularization,tensorflow/tensorflow/python/keras/layers/core.py,1237,class,"Layer that applies an update to the cost function based input activity.
Arguments:
l1: L1 regularization factor (positive float).
l2: L2 regularization factor (positive float).
Input shape:
Arbitrary. Use the keyword argument `input_shape`
(tuple of integers, does not include the samples axis)
when using this layer as the first layer in a model.
Output shape:
Same shape as input."
6299,compute_output_shape,tensorflow/tensorflow/python/keras/layers/core.py,1260,method,
6300,get_config,tensorflow/tensorflow/python/keras/layers/core.py,1263,method,
6301,TFOpLambda,tensorflow/tensorflow/python/keras/layers/core.py,1269,class,"Wraps TF API symbols in a `Layer` object.
It is inserted by the Functional API construction whenever users call
a supported TF symbol on KerasTensors.
Like Lambda layers, this layer tries to raise warnings when it detects users
explicitly use variables in the call. (To let them know
that the layer will not capture the variables).
This is useful in the case where users do something like:
x = keras.Input(...)
y = tf.Variable(...)
out = x * tf_variable"
6302,get_config,tensorflow/tensorflow/python/keras/layers/core.py,1384,method,
6303,from_config,tensorflow/tensorflow/python/keras/layers/core.py,1403,method,
6304,KerasOpDispatcher,tensorflow/tensorflow/python/keras/layers/core.py,1416,class,A global dispatcher that allows building a functional model with TF Ops.
6305,handle,tensorflow/tensorflow/python/keras/layers/core.py,1419,method,Handle the specified operation with the specified arguments.
6306,SlicingOpLambda,tensorflow/tensorflow/python/keras/layers/core.py,1443,class,"Wraps TF API symbols in a `Layer` object.
It is inserted by the Functional API construction whenever users call
a supported TF symbol on KerasTensors.
Like Lambda layers, this layer tries to raise warnings when it detects users
explicitly use variables in the call. (To let them know
that the layer will not capture the variables).
This is useful in the case where users do something like:
x = keras.Input(...)
y = tf.Variable(...)
out = x * tf_variable"
6307,TFSlicingOpDispatcher,tensorflow/tensorflow/python/keras/layers/core.py,1499,class,A global dispatcher that allows building a functional model with TF Ops.
6308,handle,tensorflow/tensorflow/python/keras/layers/core.py,1505,method,Handle the specified operation with the specified arguments.
6309,CuDNNGRU,tensorflow/tensorflow/python/keras/layers/cudnn_recurrent.py,162,class,"Fast GRU implementation backed by cuDNN.
More information about cuDNN can be found on the [NVIDIA
developer website](https://developer.nvidia.com/cudnn).
Can only be run on GPU.
Arguments:
units: Positive integer, dimensionality of the output space.
kernel_initializer: Initializer for the `kernel` weights matrix, used for
the linear transformation of the inputs.
recurrent_initializer: Initializer for the `recurrent_kernel` weights
matrix, used for the linear transformation of the recurrent state.
bias_initializer: Initializer for the bias vector.
kernel_regularizer: Regularizer function applied to the `kernel` weights
matrix.
recurrent_regularizer: Regularizer function applied to the
`recurrent_kernel` weights matrix.
bias_regularizer: Regularizer function applied to the bias vector.
activity_regularizer: Regularizer function applied to the output of the
layer (its ""activation"").
kernel_constraint: Constraint function applied to the `kernel` weights
matrix.
recurrent_constraint: Constraint function applied to the
`recurrent_kernel` weights matrix.
bias_constraint: Constraint function applied to the bias vector.
return_sequences: Boolean. Whether to return the last output in the output
sequence, or the full sequence.
return_state: Boolean. Whether to return the last state in addition to the
output.
go_backwards: Boolean (default False). If True, process the input sequence
backwards and return the reversed sequence.
stateful: Boolean (default False). If True, the last state for each sample
at index i in a batch will be used as initial state for the sample of
index i in the following batch."
6310,cell,tensorflow/tensorflow/python/keras/layers/cudnn_recurrent.py,240,method,
6311,build,tensorflow/tensorflow/python/keras/layers/cudnn_recurrent.py,243,method,
6312,get_config,tensorflow/tensorflow/python/keras/layers/cudnn_recurrent.py,319,method,
6313,CuDNNLSTM,tensorflow/tensorflow/python/keras/layers/cudnn_recurrent.py,342,class,"Fast LSTM implementation backed by cuDNN.
More information about cuDNN can be found on the [NVIDIA
developer website](https://developer.nvidia.com/cudnn).
Can only be run on GPU.
Arguments:
units: Positive integer, dimensionality of the output space.
kernel_initializer: Initializer for the `kernel` weights matrix, used for
the linear transformation of the inputs.
unit_forget_bias: Boolean. If True, add 1 to the bias of the forget gate
at initialization. Setting it to true will also force
`bias_initializer=""zeros""`. This is recommended in [Jozefowicz et
al.](http://www.jmlr.org/proceedings/papers/v37/jozefowicz15.pdf)
recurrent_initializer: Initializer for the `recurrent_kernel` weights
matrix, used for the linear transformation of the recurrent state.
bias_initializer: Initializer for the bias vector.
kernel_regularizer: Regularizer function applied to the `kernel` weights
matrix.
recurrent_regularizer: Regularizer function applied to the
`recurrent_kernel` weights matrix.
bias_regularizer: Regularizer function applied to the bias vector.
activity_regularizer: Regularizer function applied to the output of the
layer (its ""activation"").
kernel_constraint: Constraint function applied to the `kernel` weights
matrix.
recurrent_constraint: Constraint function applied to the
`recurrent_kernel` weights matrix.
bias_constraint: Constraint function applied to the bias vector.
return_sequences: Boolean. Whether to return the last output. in the
output sequence, or the full sequence.
return_state: Boolean. Whether to return the last state in addition to the
output.
go_backwards: Boolean (default False). If True, process the input sequence
backwards and return the reversed sequence.
stateful: Boolean (default False). If True, the last state for each sample
at index i in a batch will be used as initial state for the sample of
index i in the following batch."
6314,cell,tensorflow/tensorflow/python/keras/layers/cudnn_recurrent.py,426,method,
6315,build,tensorflow/tensorflow/python/keras/layers/cudnn_recurrent.py,429,method,
6316,get_config,tensorflow/tensorflow/python/keras/layers/cudnn_recurrent.py,521,method,
6317,bias_initializer,tensorflow/tensorflow/python/keras/layers/cudnn_recurrent.py,451,method,
6318,BaseDenseAttention,tensorflow/tensorflow/python/keras/layers/dense_attention.py,38,class,"Base Attention class for Dense networks.
This class is suitable for Dense or CNN networks, and not for RNN networks.
Implementations of attention mechanisms should inherit from this class, and
reuse the `apply_attention_scores()` method.
Args:
causal: Boolean. Set to `True` for decoder self-attention. Adds a mask such
that position `i` cannot attend to positions `j > i`. This prevents the
flow of information from the future towards the past.
dropout: Float between 0 and 1. Fraction of the units to drop for the
attention scores.
Call Arguments:
inputs: List of the following tensors:
* query: Query `Tensor` of shape `[batch_size, Tq, dim]`.
* value: Value `Tensor` of shape `[batch_size, Tv, dim]`.
* key: Optional key `Tensor` of shape `[batch_size, Tv, dim]`. If not
given, will use `value` for both `key` and `value`, which is the
most common case.
mask: List of the following tensors:
* query_mask: A boolean mask `Tensor` of shape `[batch_size, Tq]`.
If given, the output will be zero at the positions where
`mask==False`.
* value_mask: A boolean mask `Tensor` of shape `[batch_size, Tv]`.
If given, will apply the mask such that values at positions where
`mask==False` do not contribute to the result.
training: Python boolean indicating whether the layer should behave in
training mode (adding dropout) or in inference mode (no dropout).
Output shape:
Attention outputs of shape `[batch_size, Tq, dim]`."
6319,call,tensorflow/tensorflow/python/keras/layers/dense_attention.py,137,method,
6320,compute_mask,tensorflow/tensorflow/python/keras/layers/dense_attention.py,169,method,
6321,get_config,tensorflow/tensorflow/python/keras/layers/dense_attention.py,200,method,
6322,dropped_weights,tensorflow/tensorflow/python/keras/layers/dense_attention.py,127,method,
6323,Attention,tensorflow/tensorflow/python/keras/layers/dense_attention.py,210,class,"Dot-product attention layer, a.k.a. Luong-style attention.
Inputs are `query` tensor of shape `[batch_size, Tq, dim]`, `value` tensor of
shape `[batch_size, Tv, dim]` and `key` tensor of shape
`[batch_size, Tv, dim]`. The calculation follows the steps:
1. Calculate scores with shape `[batch_size, Tq, Tv]` as a `query`-`key` dot
product: `scores = tf.matmul(query, key, transpose_b=True)`.
2. Use scores to calculate a distribution with shape
`[batch_size, Tq, Tv]`: `distribution = tf.nn.softmax(scores)`.
3. Use `distribution` to create a linear combination of `value` with
shape `[batch_size, Tq, dim]`:
`return tf.matmul(distribution, value)`.
Args:
use_scale: If `True`, will create a scalar variable to scale the attention
scores.
causal: Boolean. Set to `True` for decoder self-attention. Adds a mask such
that position `i` cannot attend to positions `j > i`. This prevents the
flow of information from the future towards the past.
dropout: Float between 0 and 1. Fraction of the units to drop for the
attention scores.
Call Arguments:
inputs: List of the following tensors:
* query: Query `Tensor` of shape `[batch_size, Tq, dim]`.
* value: Value `Tensor` of shape `[batch_size, Tv, dim]`.
* key: Optional key `Tensor` of shape `[batch_size, Tv, dim]`. If not
given, will use `value` for both `key` and `value`, which is the
most common case.
mask: List of the following tensors:
* query_mask: A boolean mask `Tensor` of shape `[batch_size, Tq]`.
If given, the output will be zero at the positions where
`mask==False`.
* value_mask: A boolean mask `Tensor` of shape `[batch_size, Tv]`.
If given, will apply the mask such that values at positions where
`mask==False` do not contribute to the result.
training: Python boolean indicating whether the layer should behave in
training mode (adding dropout) or in inference mode (no dropout).
Output shape:
Attention outputs of shape `[batch_size, Tq, dim]`.
The meaning of `query`, `value` and `key` depend on the application. In the
case of text similarity, for example, `query` is the sequence embeddings of
the first piece of text and `value` is the sequence embeddings of the second
piece of text. `key` is usually the same tensor as `value`.
Here is a code example for using `Attention` in a CNN+Attention network:
```python
# Variable-length int sequences.
query_input = tf.keras.Input(shape=(None,), dtype='int32')
value_input = tf.keras.Input(shape=(None,), dtype='int32')
# Embedding lookup.
token_embedding = tf.keras.layers.Embedding(input_dim=1000, output_dim=64)
# Query embeddings of shape [batch_size, Tq, dimension].
query_embeddings = token_embedding(query_input)
# Value embeddings of shape [batch_size, Tv, dimension].
value_embeddings = token_embedding(value_input)
# CNN layer.
cnn_layer = tf.keras.layers.Conv1D(
filters=100,
kernel_size=4,
# Use 'same' padding so outputs have the same shape as inputs.
padding='same')
# Query encoding of shape [batch_size, Tq, filters].
query_seq_encoding = cnn_layer(query_embeddings)
# Value encoding of shape [batch_size, Tv, filters].
value_seq_encoding = cnn_layer(value_embeddings)
# Query-value attention of shape [batch_size, Tq, filters].
query_value_attention_seq = tf.keras.layers.Attention()(
[query_seq_encoding, value_seq_encoding])
# Reduce over the sequence axis to produce encodings of shape
# [batch_size, filters].
query_encoding = tf.keras.layers.GlobalAveragePooling1D()(
query_seq_encoding)
query_value_attention = tf.keras.layers.GlobalAveragePooling1D()(
query_value_attention_seq)
# Concatenate query and document encodings to produce a DNN input layer.
input_layer = tf.keras.layers.Concatenate()(
[query_encoding, query_value_attention])
# Add DNN layers, and create Model.
# ...
```"
6324,build,tensorflow/tensorflow/python/keras/layers/dense_attention.py,310,method,Creates scale variable if use_scale==True.
6325,get_config,tensorflow/tensorflow/python/keras/layers/dense_attention.py,337,method,
6326,AdditiveAttention,tensorflow/tensorflow/python/keras/layers/dense_attention.py,344,class,"Additive attention layer, a.k.a. Bahdanau-style attention.
Inputs are `query` tensor of shape `[batch_size, Tq, dim]`, `value` tensor of
shape `[batch_size, Tv, dim]` and `key` tensor of shape
`[batch_size, Tv, dim]`. The calculation follows the steps:
1. Reshape `query` and `value` into shapes `[batch_size, Tq, 1, dim]`
and `[batch_size, 1, Tv, dim]` respectively.
2. Calculate scores with shape `[batch_size, Tq, Tv]` as a non-linear
sum: `scores = tf.reduce_sum(tf.tanh(query + value), axis=-1)`
3. Use scores to calculate a distribution with shape
`[batch_size, Tq, Tv]`: `distribution = tf.nn.softmax(scores)`.
4. Use `distribution` to create a linear combination of `value` with
shape `batch_size, Tq, dim]`:
`return tf.matmul(distribution, value)`.
Args:
use_scale: If `True`, will create a variable to scale the attention scores.
causal: Boolean. Set to `True` for decoder self-attention. Adds a mask such
that position `i` cannot attend to positions `j > i`. This prevents the
flow of information from the future towards the past.
dropout: Float between 0 and 1. Fraction of the units to drop for the
attention scores.
Call Arguments:
inputs: List of the following tensors:
* query: Query `Tensor` of shape `[batch_size, Tq, dim]`.
* value: Value `Tensor` of shape `[batch_size, Tv, dim]`.
* key: Optional key `Tensor` of shape `[batch_size, Tv, dim]`. If not
given, will use `value` for both `key` and `value`, which is the
most common case.
mask: List of the following tensors:
* query_mask: A boolean mask `Tensor` of shape `[batch_size, Tq]`.
If given, the output will be zero at the positions where
`mask==False`.
* value_mask: A boolean mask `Tensor` of shape `[batch_size, Tv]`.
If given, will apply the mask such that values at positions where
`mask==False` do not contribute to the result.
training: Python boolean indicating whether the layer should behave in
training mode (adding dropout) or in inference mode (no dropout).
Output shape:
Attention outputs of shape `[batch_size, Tq, dim]`.
The meaning of `query`, `value` and `key` depend on the application. In the
case of text similarity, for example, `query` is the sequence embeddings of
the first piece of text and `value` is the sequence embeddings of the second
piece of text. `key` is usually the same tensor as `value`.
Here is a code example for using `AdditiveAttention` in a CNN+Attention
network:
```python
# Variable-length int sequences.
query_input = tf.keras.Input(shape=(None,), dtype='int32')
value_input = tf.keras.Input(shape=(None,), dtype='int32')
# Embedding lookup.
token_embedding = tf.keras.layers.Embedding(max_tokens, dimension)
# Query embeddings of shape [batch_size, Tq, dimension].
query_embeddings = token_embedding(query_input)
# Value embeddings of shape [batch_size, Tv, dimension].
value_embeddings = token_embedding(value_input)
# CNN layer.
cnn_layer = tf.keras.layers.Conv1D(
filters=100,
kernel_size=4,
# Use 'same' padding so outputs have the same shape as inputs.
padding='same')
# Query encoding of shape [batch_size, Tq, filters].
query_seq_encoding = cnn_layer(query_embeddings)
# Value encoding of shape [batch_size, Tv, filters].
value_seq_encoding = cnn_layer(value_embeddings)
# Query-value attention of shape [batch_size, Tq, filters].
query_value_attention_seq = tf.keras.layers.AdditiveAttention()(
[query_seq_encoding, value_seq_encoding])
# Reduce over the sequence axis to produce encodings of shape
# [batch_size, filters].
query_encoding = tf.keras.layers.GlobalAveragePooling1D()(
query_seq_encoding)
query_value_attention = tf.keras.layers.GlobalAveragePooling1D()(
query_value_attention_seq)
# Concatenate query and document encodings to produce a DNN input layer.
input_layer = tf.keras.layers.Concatenate()(
[query_encoding, query_value_attention])
# Add DNN layers, and create Model.
# ...
```"
6327,build,tensorflow/tensorflow/python/keras/layers/dense_attention.py,446,method,
6328,get_config,tensorflow/tensorflow/python/keras/layers/dense_attention.py,483,method,
6329,EinsumDense,tensorflow/tensorflow/python/keras/layers/einsum_dense.py,34,class,"A layer that uses tf.einsum as the backing computation.
This layer can perform einsum calculations of arbitrary dimensionality.
Arguments:
equation: An equation describing the einsum to perform. This equation must
be a valid einsum string of the form `ab,bc->ac`, `...ab,bc->...ac`, or
`ab...,bc->ac...` where 'ab', 'bc', and 'ac' can be any valid einsum axis
expression sequence.
output_shape: The expected shape of the output tensor (excluding the batch
dimension and any dimensions represented by ellipses). You can specify
None for any dimension that is unknown or can be inferred from the input
shape.
activation: Activation function to use. If you don't specify anything, no
activation is applied (that is, a ""linear"" activation: `a(x) = x`).
bias_axes: A string containing the output dimension(s) to apply a bias to.
Each character in the `bias_axes` string should correspond to a character
in the output portion of the `equation` string.
kernel_initializer: Initializer for the `kernel` weights matrix.
bias_initializer: Initializer for the bias vector.
kernel_regularizer: Regularizer function applied to the `kernel` weights
matrix.
bias_regularizer: Regularizer function applied to the bias vector.
activity_regularizer: Regularizer function applied to the output of the
layer (its ""activation"")..
kernel_constraint: Constraint function applied to the `kernel` weights
matrix.
bias_constraint: Constraint function applied to the bias vector.
Examples:
**Biased dense layer with einsums**
This example shows how to instantiate a standard Keras dense layer using
einsum operations. This example is equivalent to
`tf.keras.layers.Dense(64, use_bias=True)`.
>>> layer = EinsumDense(""ab,bc->ac"", output_shape=64, bias_axes=""c"")
>>> input_tensor = tf.keras.Input(shape=[32])
>>> output_tensor = layer(input_tensor)
>>> output_tensor
<... shape=(None, 64) dtype=...>
**Applying a dense layer to a sequence**
This example shows how to instantiate a layer that applies the same dense
operation to every element in a sequence. Here, the 'output_shape' has two
values (since there are two non-batch dimensions in the output); the first
dimension in the output_shape is `None`, because the sequence dimension `b`
has an unknown shape.
>>> layer = EinsumDense(""abc,cd->abd"",
... output_shape=(None, 64),
... bias_axes=""d"")
>>> input_tensor = tf.keras.Input(shape=[32, 128])
>>> output_tensor = layer(input_tensor)
>>> output_tensor
<... shape=(None, 32, 64) dtype=...>
**Applying a dense layer to a sequence using ellipses**
This example shows how to instantiate a layer that applies the same dense
operation to every element in a sequence, but uses the ellipsis notation
instead of specifying the batch and sequence dimensions.
Because we are using ellipsis notation and have specified only one axis, the
output_shape arg is a single value. When instantiated in this way, the layer
can handle any number of sequence dimensions - including the case where no
sequence dimension exists.
>>> layer = EinsumDense(""...x,xy->...y"", output_shape=64, bias_axes=""y"")
>>> input_tensor = tf.keras.Input(shape=[32, 128])
>>> output_tensor = layer(input_tensor)
>>> output_tensor
<... shape=(None, 32, 64) dtype=...>"
6330,build,tensorflow/tensorflow/python/keras/layers/einsum_dense.py,140,method,
6331,compute_output_shape,tensorflow/tensorflow/python/keras/layers/einsum_dense.py,169,method,
6332,get_config,tensorflow/tensorflow/python/keras/layers/einsum_dense.py,172,method,
6333,call,tensorflow/tensorflow/python/keras/layers/einsum_dense.py,200,method,
6334,Embedding,tensorflow/tensorflow/python/keras/layers/embeddings.py,36,class,"Turns positive integers (indexes) into dense vectors of fixed size.
e.g. `[[4], [20]] -> [[0.25, 0.1], [0.6, -0.2]]`
This layer can only be used as the first layer in a model.
Example:
>>> model = tf.keras.Sequential()
>>> model.add(tf.keras.layers.Embedding(1000, 64, input_length=10))
>>> # The model will take as input an integer matrix of size (batch,
>>> # input_length), and the largest integer (i.e. word index) in the input
>>> # should be no larger than 999 (vocabulary size).
>>> # Now model.output_shape is (None, 10, 64), where `None` is the batch
>>> # dimension.
>>> input_array = np.random.randint(1000, size=(32, 10))
>>> model.compile('rmsprop', 'mse')
>>> output_array = model.predict(input_array)
>>> print(output_array.shape)
(32, 10, 64)
Arguments:
input_dim: Integer. Size of the vocabulary,
i.e. maximum integer index + 1.
output_dim: Integer. Dimension of the dense embedding.
embeddings_initializer: Initializer for the `embeddings`
matrix (see `keras.initializers`).
embeddings_regularizer: Regularizer function applied to
the `embeddings` matrix (see `keras.regularizers`).
embeddings_constraint: Constraint function applied to
the `embeddings` matrix (see `keras.constraints`).
mask_zero: Boolean, whether or not the input value 0 is a special ""padding""
value that should be masked out.
This is useful when using recurrent layers
which may take variable length input.
If this is `True`, then all subsequent layers
in the model need to support masking or an exception will be raised.
If mask_zero is set to True, as a consequence, index 0 cannot be
used in the vocabulary (input_dim should equal size of
vocabulary + 1).
input_length: Length of input sequences, when it is constant.
This argument is required if you are going to connect
`Flatten` then `Dense` layers upstream
(without it, the shape of the dense outputs cannot be computed).
Input shape:
2D tensor with shape: `(batch_size, input_length)`.
Output shape:
3D tensor with shape: `(batch_size, input_length, output_dim)`."
6335,build,tensorflow/tensorflow/python/keras/layers/embeddings.py,127,method,
6336,compute_mask,tensorflow/tensorflow/python/keras/layers/embeddings.py,152,method,
6337,compute_output_shape,tensorflow/tensorflow/python/keras/layers/embeddings.py,159,method,
6338,call,tensorflow/tensorflow/python/keras/layers/embeddings.py,182,method,
6339,get_config,tensorflow/tensorflow/python/keras/layers/embeddings.py,192,method,
6340,RandomFourierFeatures,tensorflow/tensorflow/python/keras/layers/kernelized.py,40,class,"Layer that projects its inputs into a random feature space.
This layer implements a mapping from input space to a space with `output_dim`
dimensions, which approximates shift-invariant kernels. A kernel function
`K(x, y)` is shift-invariant if `K(x, y) == k(x - y)` for some function `k`.
Many popular Radial Basis Functions (RBF), including Gaussian and
Laplacian kernels, are shift-invariant.
The implementation of this layer is based on the following paper:
[""Random Features for Large-Scale Kernel Machines""](
https://people.eecs.berkeley.edu/~brecht/papers/07.rah.rec.nips.pdf)
by Ali Rahimi and Ben Recht.
The distribution from which the parameters of the random features map (layer)
are sampled determines which shift-invariant kernel the layer approximates
(see paper for more details). You can use the distribution of your
choice. The layer supports out-of-the-box
approximation sof the following two RBF kernels:
- Gaussian: `K(x, y) == exp(- square(x - y) / (2 * square(scale)))`
- Laplacian: `K(x, y) = exp(-abs(x - y) / scale))`
**Note:** Unlike what is described in the paper and unlike what is used in
the Scikit-Learn implementation, the output of this layer does not apply
the `sqrt(2 / D)` normalization factor.
**Usage:** Typically, this layer is used to ""kernelize"" linear models by
applying a non-linear transformation (this layer) to the input features and
then training a linear model on top of the transformed features. Depending on
the loss function of the linear model, the composition of this layer and the
linear model results to models that are equivalent (up to approximation) to
kernel SVMs (for hinge loss), kernel logistic regression (for logistic loss),
kernel linear regression (for squared loss), etc.
Examples:
A kernel multinomial logistic regression model with Gaussian kernel for MNIST:
```python
model = keras.Sequential([
keras.Input(shape=(784,)),
RandomFourierFeatures(
output_dim=4096,
scale=10.,
kernel_initializer='gaussian'),
layers.Dense(units=10, activation='softmax'),
])
model.compile(
optimizer='adam',
loss='categorical_crossentropy',
metrics=['categorical_accuracy']
)
```
A quasi-SVM classifier for MNIST:
```python
model = keras.Sequential([
keras.Input(shape=(784,)),
RandomFourierFeatures(
output_dim=4096,
scale=10.,
kernel_initializer='gaussian'),
layers.Dense(units=10),
])
model.compile(
optimizer='adam',
loss='hinge',
metrics=['categorical_accuracy']
)
```
To use another kernel, just replace the layer creation line with:
```python
random_features_layer = RandomFourierFeatures(
output_dim=500,
kernel_initializer=<my_initializer>,
scale=...,
...)
```
Arguments:
output_dim: Positive integer, the dimension of the layer's output, i.e., the
number of random features used to approximate the kernel.
kernel_initializer: Determines the distribution of the parameters of the
random features map (and therefore the kernel approximated by the layer).
It can be either a string identifier or a Keras `Initializer` instance.
Currently only 'gaussian' and 'laplacian' are supported string
identifiers (case insensitive). Note that the kernel matrix is not
trainable.
scale: For Gaussian and Laplacian kernels, this corresponds to a scaling
factor of the corresponding kernel approximated by the layer (see concrete
definitions above). When provided, it should be a positive float. If None,
a default value is used: if the kernel initializer is set to ""gaussian"",
`scale` defaults to `sqrt(input_dim / 2)`, otherwise, it defaults to 1.0.
Both the approximation error of the kernel and the classification quality
are sensitive to this parameter. If `trainable` is set to `True`, this
parameter is learned end-to-end during training and the provided value
serves as the initial value.
**Note:** When features from this layer are fed to a linear model,
by making `scale` trainable, the resulting optimization problem is
no longer convex (even if the loss function used by the linear model
is convex).
trainable: Whether the scaling parameter of the layer should be trainable.
Defaults to `False`.
name: String, name to use for this layer."
6341,build,tensorflow/tensorflow/python/keras/layers/kernelized.py,175,method,
6342,call,tensorflow/tensorflow/python/keras/layers/kernelized.py,220,method,
6343,compute_output_shape,tensorflow/tensorflow/python/keras/layers/kernelized.py,228,method,
6344,get_config,tensorflow/tensorflow/python/keras/layers/kernelized.py,237,method,
6345,LocallyConnected1D,tensorflow/tensorflow/python/keras/layers/local.py,36,class,"Locally-connected layer for 1D inputs.
The `LocallyConnected1D` layer works similarly to
the `Conv1D` layer, except that weights are unshared,
that is, a different set of filters is applied at each different patch
of the input.
Note: layer attributes cannot be modified after the layer has been called
once (except the `trainable` attribute).
Example:
```python
# apply a unshared weight convolution 1d of length 3 to a sequence with
# 10 timesteps, with 64 output filters
model = Sequential()
model.add(LocallyConnected1D(64, 3, input_shape=(10, 32)))
# now model.output_shape == (None, 8, 64)
# add a new conv1d on top
model.add(LocallyConnected1D(32, 3))
# now model.output_shape == (None, 6, 32)
```
Arguments:
filters: Integer, the dimensionality of the output space
(i.e. the number of output filters in the convolution).
kernel_size: An integer or tuple/list of a single integer,
specifying the length of the 1D convolution window.
strides: An integer or tuple/list of a single integer,
specifying the stride length of the convolution.
Specifying any stride value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: Currently only supports `""valid""` (case-insensitive).
`""same""` may be supported in the future.
`""valid""` means no padding.
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, length, channels)` while `channels_first`
corresponds to inputs with shape
`(batch, channels, length)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
activation: Activation function to use.
If you don't specify anything, no activation is applied
(ie. ""linear"" activation: `a(x) = x`).
use_bias: Boolean, whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix.
bias_initializer: Initializer for the bias vector.
kernel_regularizer: Regularizer function applied to
the `kernel` weights matrix.
bias_regularizer: Regularizer function applied to the bias vector.
activity_regularizer: Regularizer function applied to
the output of the layer (its ""activation"")..
kernel_constraint: Constraint function applied to the kernel matrix.
bias_constraint: Constraint function applied to the bias vector.
implementation: implementation mode, either `1`, `2`, or `3`.
`1` loops over input spatial locations to perform the forward pass.
It is memory-efficient but performs a lot of (small) ops.
`2` stores layer weights in a dense but sparsely-populated 2D matrix
and implements the forward pass as a single matrix-multiply. It uses
a lot of RAM but performs few (large) ops.
`3` stores layer weights in a sparse tensor and implements the forward
pass as a single sparse matrix-multiply.
How to choose:
`1`: large, dense models,
`2`: small models,
`3`: large, sparse models,
where ""large"" stands for large input/output activations
(i.e. many `filters`, `input_filters`, large `input_size`,
`output_size`), and ""sparse"" stands for few connections between inputs
and outputs, i.e. small ratio
`filters * input_filters * kernel_size / (input_size * strides)`,
where inputs to and outputs of the layer are assumed to have shapes
`(input_size, input_filters)`, `(output_size, filters)`
respectively.
It is recommended to benchmark each in the setting of interest to pick
the most efficient one (in terms of speed and memory usage). Correct
choice of implementation can lead to dramatic speed improvements (e.g.
50X), potentially at the expense of RAM.
Also, only `padding=""valid""` is supported by `implementation=1`.
Input shape:
3D tensor with shape: `(batch_size, steps, input_dim)`
Output shape:
3D tensor with shape: `(batch_size, new_steps, filters)`
`steps` value might have changed due to padding or strides."
6346,build,tensorflow/tensorflow/python/keras/layers/local.py,175,method,
6347,compute_output_shape,tensorflow/tensorflow/python/keras/layers/local.py,263,method,
6348,call,tensorflow/tensorflow/python/keras/layers/local.py,277,method,
6349,get_config,tensorflow/tensorflow/python/keras/layers/local.py,301,method,
6350,LocallyConnected2D,tensorflow/tensorflow/python/keras/layers/local.py,339,class,"Locally-connected layer for 2D inputs.
The `LocallyConnected2D` layer works similarly
to the `Conv2D` layer, except that weights are unshared,
that is, a different set of filters is applied at each
different patch of the input.
Note: layer attributes cannot be modified after the layer has been called
once (except the `trainable` attribute).
Examples:
```python
# apply a 3x3 unshared weights convolution with 64 output filters on a
32x32 image
# with `data_format=""channels_last""`:
model = Sequential()
model.add(LocallyConnected2D(64, (3, 3), input_shape=(32, 32, 3)))
# now model.output_shape == (None, 30, 30, 64)
# notice that this layer will consume (30*30)*(3*3*3*64) + (30*30)*64
parameters
# add a 3x3 unshared weights convolution on top, with 32 output filters:
model.add(LocallyConnected2D(32, (3, 3)))
# now model.output_shape == (None, 28, 28, 32)
```
Arguments:
filters: Integer, the dimensionality of the output space
(i.e. the number of output filters in the convolution).
kernel_size: An integer or tuple/list of 2 integers, specifying the
width and height of the 2D convolution window.
Can be a single integer to specify the same value for
all spatial dimensions.
strides: An integer or tuple/list of 2 integers,
specifying the strides of the convolution along the width and height.
Can be a single integer to specify the same value for
all spatial dimensions.
padding: Currently only support `""valid""` (case-insensitive).
`""same""` will be supported in future.
`""valid""` means no padding.
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch, channels, height, width)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
activation: Activation function to use.
If you don't specify anything, no activation is applied
(ie. ""linear"" activation: `a(x) = x`).
use_bias: Boolean, whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix.
bias_initializer: Initializer for the bias vector.
kernel_regularizer: Regularizer function applied to
the `kernel` weights matrix.
bias_regularizer: Regularizer function applied to the bias vector.
activity_regularizer: Regularizer function applied to
the output of the layer (its ""activation"").
kernel_constraint: Constraint function applied to the kernel matrix.
bias_constraint: Constraint function applied to the bias vector.
implementation: implementation mode, either `1`, `2`, or `3`.
`1` loops over input spatial locations to perform the forward pass.
It is memory-efficient but performs a lot of (small) ops.
`2` stores layer weights in a dense but sparsely-populated 2D matrix
and implements the forward pass as a single matrix-multiply. It uses
a lot of RAM but performs few (large) ops.
`3` stores layer weights in a sparse tensor and implements the forward
pass as a single sparse matrix-multiply.
How to choose:
`1`: large, dense models,
`2`: small models,
`3`: large, sparse models,
where ""large"" stands for large input/output activations
(i.e. many `filters`, `input_filters`, large `np.prod(input_size)`,
`np.prod(output_size)`), and ""sparse"" stands for few connections
between inputs and outputs, i.e. small ratio
`filters * input_filters * np.prod(kernel_size) / (np.prod(input_size)
* np.prod(strides))`, where inputs to and outputs of the layer are
assumed to have shapes `input_size + (input_filters,)`,
`output_size + (filters,)` respectively.
It is recommended to benchmark each in the setting of interest to pick
the most efficient one (in terms of speed and memory usage). Correct
choice of implementation can lead to dramatic speed improvements (e.g.
50X), potentially at the expense of RAM.
Also, only `padding=""valid""` is supported by `implementation=1`.
Input shape:
4D tensor with shape:
`(samples, channels, rows, cols)` if data_format='channels_first'
or 4D tensor with shape:
`(samples, rows, cols, channels)` if data_format='channels_last'.
Output shape:
4D tensor with shape:
`(samples, filters, new_rows, new_cols)` if data_format='channels_first'
or 4D tensor with shape:
`(samples, new_rows, new_cols, filters)` if data_format='channels_last'.
`rows` and `cols` values might have changed due to padding."
6351,build,tensorflow/tensorflow/python/keras/layers/local.py,490,method,
6352,compute_output_shape,tensorflow/tensorflow/python/keras/layers/local.py,586,method,
6353,call,tensorflow/tensorflow/python/keras/layers/local.py,604,method,
6354,get_config,tensorflow/tensorflow/python/keras/layers/local.py,629,method,
6355,get_locallyconnected_mask,tensorflow/tensorflow/python/keras/layers/local.py,666,function,"Return a mask representing connectivity of a locally-connected operation.
This method returns a masking numpy array of 0s and 1s (of type `np.float32`)
that, when element-wise multiplied with a fully-connected weight tensor, masks
out the weights between disconnected input-output pairs and thus implements
local connectivity through a sparse fully-connected weight tensor.
Assume an unshared convolution with given parameters is applied to an input
having N spatial dimensions with `input_shape = (d_in1, ..., d_inN)`
to produce an output with spatial shape `(d_out1, ..., d_outN)` (determined
by layer parameters such as `strides`).
This method returns a mask which can be broadcast-multiplied (element-wise)
with a 2*(N+1)-D weight matrix (equivalent to a fully-connected layer between
(N+1)-D activations (N spatial + 1 channel dimensions for input and output)
to make it perform an unshared convolution with given `kernel_shape`,
`strides`, `padding` and `data_format`.
Arguments:
input_shape: tuple of size N: `(d_in1, ..., d_inN)`
spatial shape of the input.
kernel_shape: tuple of size N, spatial shape of the convolutional kernel
/ receptive field.
strides: tuple of size N, strides along each spatial dimension.
padding: type of padding, string `""same""` or `""valid""`.
data_format: a string, `""channels_first""` or `""channels_last""`.
Returns:
a `np.float32`-type `np.ndarray` of shape
`(1, d_in1, ..., d_inN, 1, d_out1, ..., d_outN)`
if `data_format == `""channels_first""`, or
`(d_in1, ..., d_inN, 1, d_out1, ..., d_outN, 1)`
if `data_format == ""channels_last""`.
Raises:
ValueError: if `data_format` is neither `""channels_first""` nor
`""channels_last""`."
6356,local_conv_matmul,tensorflow/tensorflow/python/keras/layers/local.py,729,function,"Apply N-D convolution with un-shared weights using a single matmul call.
This method outputs `inputs . (kernel * kernel_mask)`
(with `.` standing for matrix-multiply and `*` for element-wise multiply)
and requires a precomputed `kernel_mask` to zero-out weights in `kernel` and
hence perform the same operation as a convolution with un-shared
(the remaining entries in `kernel`) weights. It also does the necessary
reshapes to make `inputs` and `kernel` 2-D and `output` (N+2)-D.
Arguments:
inputs: (N+2)-D tensor with shape
`(batch_size, channels_in, d_in1, ..., d_inN)`
or
`(batch_size, d_in1, ..., d_inN, channels_in)`.
kernel: the unshared weights for N-D convolution,
an (N+2)-D tensor of shape:
`(d_in1, ..., d_inN, channels_in, d_out2, ..., d_outN, channels_out)`
or
`(channels_in, d_in1, ..., d_inN, channels_out, d_out2, ..., d_outN)`,
with the ordering of channels and spatial dimensions matching
that of the input.
Each entry is the weight between a particular input and
output location, similarly to a fully-connected weight matrix.
kernel_mask: a float 0/1 mask tensor of shape:
`(d_in1, ..., d_inN, 1, d_out2, ..., d_outN, 1)`
or
`(1, d_in1, ..., d_inN, 1, d_out2, ..., d_outN)`,
with the ordering of singleton and spatial dimensions
matching that of the input.
Mask represents the connectivity pattern of the layer and is
precomputed elsewhere based on layer parameters: stride,
padding, and the receptive field shape.
output_shape: a tuple of (N+2) elements representing the output shape:
`(batch_size, channels_out, d_out1, ..., d_outN)`
or
`(batch_size, d_out1, ..., d_outN, channels_out)`,
with the ordering of channels and spatial dimensions matching that of
the input.
Returns:
Output (N+2)-D tensor with shape `output_shape`."
6357,local_conv_sparse_matmul,tensorflow/tensorflow/python/keras/layers/local.py,783,function,"Apply N-D convolution with un-shared weights using a single sparse matmul.
This method outputs `inputs . tf.sparse.SparseTensor(indices=kernel_idxs,
values=kernel, dense_shape=kernel_shape)`, with `.` standing for
matrix-multiply. It also reshapes `inputs` to 2-D and `output` to (N+2)-D.
Arguments:
inputs: (N+2)-D tensor with shape `(batch_size, channels_in, d_in1, ...,
d_inN)` or `(batch_size, d_in1, ..., d_inN, channels_in)`.
kernel: a 1-D tensor with shape `(len(kernel_idxs),)` containing all the
weights of the layer.
kernel_idxs: a list of integer tuples representing indices in a sparse
matrix performing the un-shared convolution as a matrix-multiply.
kernel_shape: a tuple `(input_size, output_size)`, where `input_size =
channels_in * d_in1 * ... * d_inN` and `output_size = channels_out *
d_out1 * ... * d_outN`.
output_shape: a tuple of (N+2) elements representing the output shape:
`(batch_size, channels_out, d_out1, ..., d_outN)` or `(batch_size,
d_out1, ..., d_outN, channels_out)`, with the ordering of channels and
spatial dimensions matching that of the input.
Returns:
Output (N+2)-D dense tensor with shape `output_shape`."
6358,make_2d,tensorflow/tensorflow/python/keras/layers/local.py,821,function,"Reshapes an N-dimensional tensor into a 2D tensor.
Dimensions before (excluding) and after (including) `split_dim` are grouped
together.
Arguments:
tensor: a tensor of shape `(d0, ..., d(N-1))`.
split_dim: an integer from 1 to N-1, index of the dimension to group
dimensions before (excluding) and after (including).
Returns:
Tensor of shape
`(d0 * ... * d(split_dim-1), d(split_dim) * ... * d(N-1))`."
6359,get_inputs,tensorflow/tensorflow/python/keras/layers/local_test.py,402,function,
6360,xent,tensorflow/tensorflow/python/keras/layers/local_test.py,423,function,
6361,get_model,tensorflow/tensorflow/python/keras/layers/local_test.py,433,function,
6362,copy_lc_weights_2_to_1,tensorflow/tensorflow/python/keras/layers/local_test.py,471,function,
6363,copy_lc_weights_2_to_3,tensorflow/tensorflow/python/keras/layers/local_test.py,512,function,
6364,copy_model_weights,tensorflow/tensorflow/python/keras/layers/local_test.py,529,function,
6365,Add,tensorflow/tensorflow/python/keras/layers/merge.py,221,class,"Layer that adds a list of inputs.
It takes as input a list of tensors,
all of the same shape, and returns
a single tensor (also of the same shape).
Examples:
>>> input_shape = (2, 3, 4)
>>> x1 = tf.random.normal(input_shape)
>>> x2 = tf.random.normal(input_shape)
>>> y = tf.keras.layers.Add()([x1, x2])
>>> print(y.shape)
(2, 3, 4)
Used in a functional model:
>>> input1 = tf.keras.layers.Input(shape=(16,))
>>> x1 = tf.keras.layers.Dense(8, activation='relu')(input1)
>>> input2 = tf.keras.layers.Input(shape=(32,))
>>> x2 = tf.keras.layers.Dense(8, activation='relu')(input2)
>>> # equivalent to `added = tf.keras.layers.add([x1, x2])`
>>> added = tf.keras.layers.Add()([x1, x2])
>>> out = tf.keras.layers.Dense(4)(added)
>>> model = tf.keras.models.Model(inputs=[input1, input2], outputs=out)"
6366,Subtract,tensorflow/tensorflow/python/keras/layers/merge.py,258,class,"Layer that subtracts two inputs.
It takes as input a list of tensors of size 2,
both of the same shape, and returns a single tensor, (inputs[0] - inputs[1]),
also of the same shape.
Examples:
```python
import keras
input1 = keras.layers.Input(shape=(16,))
x1 = keras.layers.Dense(8, activation='relu')(input1)
input2 = keras.layers.Input(shape=(32,))
x2 = keras.layers.Dense(8, activation='relu')(input2)
# Equivalent to subtracted = keras.layers.subtract([x1, x2])
subtracted = keras.layers.Subtract()([x1, x2])
out = keras.layers.Dense(4)(subtracted)
model = keras.models.Model(inputs=[input1, input2], outputs=out)
```"
6367,build,tensorflow/tensorflow/python/keras/layers/merge.py,283,method,
6368,Multiply,tensorflow/tensorflow/python/keras/layers/merge.py,297,class,"Layer that multiplies (element-wise) a list of inputs.
It takes as input a list of tensors, all of the same shape, and returns
a single tensor (also of the same shape).
>>> tf.keras.layers.Multiply()([np.arange(5).reshape(5, 1),
... np.arange(5, 10).reshape(5, 1)])
<tf.Tensor: shape=(5, 1), dtype=int64, numpy=
array([[ 0],
[ 6],
[14],
[24],
[36]])>
>>> x1 = tf.keras.layers.Dense(8)(np.arange(10).reshape(5, 2))
>>> x2 = tf.keras.layers.Dense(8)(np.arange(10, 20).reshape(5, 2))
>>> multiplied = tf.keras.layers.Multiply()([x1, x2])
>>> multiplied.shape
TensorShape([5, 8])"
6369,Average,tensorflow/tensorflow/python/keras/layers/merge.py,327,class,"Layer that averages a list of inputs element-wise.
It takes as input a list of tensors, all of the same shape, and returns
a single tensor (also of the same shape).
Example:
>>> x1 = np.ones((2, 2))
>>> x2 = np.zeros((2, 2))
>>> y = tf.keras.layers.Average()([x1, x2])
>>> y.numpy().tolist()
[[0.5, 0.5], [0.5, 0.5]]
Usage in a functional model:
>>> input1 = tf.keras.layers.Input(shape=(16,))
>>> x1 = tf.keras.layers.Dense(8, activation='relu')(input1)
>>> input2 = tf.keras.layers.Input(shape=(32,))
>>> x2 = tf.keras.layers.Dense(8, activation='relu')(input2)
>>> avg = tf.keras.layers.Average()([x1, x2])
>>> out = tf.keras.layers.Dense(4)(avg)
>>> model = tf.keras.models.Model(inputs=[input1, input2], outputs=out)
Raises:
ValueError: If there is a shape mismatch between the inputs and the shapes
cannot be broadcasted to match."
6370,Maximum,tensorflow/tensorflow/python/keras/layers/merge.py,364,class,"Layer that computes the maximum (element-wise) a list of inputs.
It takes as input a list of tensors, all of the same shape, and returns
a single tensor (also of the same shape).
>>> tf.keras.layers.Maximum()([np.arange(5).reshape(5, 1),
... np.arange(5, 10).reshape(5, 1)])
<tf.Tensor: shape=(5, 1), dtype=int64, numpy=
array([[5],
[6],
[7],
[8],
[9]])>
>>> x1 = tf.keras.layers.Dense(8)(np.arange(10).reshape(5, 2))
>>> x2 = tf.keras.layers.Dense(8)(np.arange(10, 20).reshape(5, 2))
>>> maxed = tf.keras.layers.Maximum()([x1, x2])
>>> maxed.shape
TensorShape([5, 8])"
6371,Minimum,tensorflow/tensorflow/python/keras/layers/merge.py,394,class,"Layer that computes the minimum (element-wise) a list of inputs.
It takes as input a list of tensors, all of the same shape, and returns
a single tensor (also of the same shape).
>>> tf.keras.layers.Minimum()([np.arange(5).reshape(5, 1),
... np.arange(5, 10).reshape(5, 1)])
<tf.Tensor: shape=(5, 1), dtype=int64, numpy=
array([[0],
[1],
[2],
[3],
[4]])>
>>> x1 = tf.keras.layers.Dense(8)(np.arange(10).reshape(5, 2))
>>> x2 = tf.keras.layers.Dense(8)(np.arange(10, 20).reshape(5, 2))
>>> minned = tf.keras.layers.Minimum()([x1, x2])
>>> minned.shape
TensorShape([5, 8])"
6372,Concatenate,tensorflow/tensorflow/python/keras/layers/merge.py,424,class,"Layer that concatenates a list of inputs.
It takes as input a list of tensors, all of the same shape except
for the concatenation axis, and returns a single tensor that is the
concatenation of all inputs.
>>> x = np.arange(20).reshape(2, 2, 5)
>>> print(x)
[[[ 0 1 2 3 4]
[ 5 6 7 8 9]]
[[10 11 12 13 14]
[15 16 17 18 19]]]
>>> y = np.arange(20, 30).reshape(2, 1, 5)
>>> print(y)
[[[20 21 22 23 24]]
[[25 26 27 28 29]]]
>>> tf.keras.layers.Concatenate(axis=1)([x, y])
<tf.Tensor: shape=(2, 3, 5), dtype=int64, numpy=
array([[[ 0, 1, 2, 3, 4],
[ 5, 6, 7, 8, 9],
[20, 21, 22, 23, 24]],
[[10, 11, 12, 13, 14],
[15, 16, 17, 18, 19],
[25, 26, 27, 28, 29]]])>
>>> x1 = tf.keras.layers.Dense(8)(np.arange(10).reshape(5, 2))
>>> x2 = tf.keras.layers.Dense(8)(np.arange(10, 20).reshape(5, 2))
>>> concatted = tf.keras.layers.Concatenate()([x1, x2])
>>> concatted.shape
TensorShape([5, 16])"
6373,build,tensorflow/tensorflow/python/keras/layers/merge.py,490,method,
6374,compute_output_shape,tensorflow/tensorflow/python/keras/layers/merge.py,525,method,
6375,compute_mask,tensorflow/tensorflow/python/keras/layers/merge.py,542,method,
6376,get_config,tensorflow/tensorflow/python/keras/layers/merge.py,570,method,
6377,Dot,tensorflow/tensorflow/python/keras/layers/merge.py,579,class,"Layer that computes a dot product between samples in two tensors.
E.g. if applied to a list of two tensors `a` and `b` of shape
`(batch_size, n)`, the output will be a tensor of shape `(batch_size, 1)`
where each entry `i` will be the dot product between
`a[i]` and `b[i]`.
>>> x = np.arange(10).reshape(1, 5, 2)
>>> print(x)
[[[0 1]
[2 3]
[4 5]
[6 7]
[8 9]]]
>>> y = np.arange(10, 20).reshape(1, 2, 5)
>>> print(y)
[[[10 11 12 13 14]
[15 16 17 18 19]]]
>>> tf.keras.layers.Dot(axes=(1, 2))([x, y])
<tf.Tensor: shape=(1, 2, 2), dtype=int64, numpy=
array([[[260, 360],
[320, 445]]])>
>>> x1 = tf.keras.layers.Dense(8)(np.arange(10).reshape(5, 2))
>>> x2 = tf.keras.layers.Dense(8)(np.arange(10, 20).reshape(5, 2))
>>> dotted = tf.keras.layers.Dot(axes=1)([x1, x2])
>>> dotted.shape
TensorShape([5, 1])"
6378,build,tensorflow/tensorflow/python/keras/layers/merge.py,660,method,
6379,compute_output_shape,tensorflow/tensorflow/python/keras/layers/merge.py,707,method,
6380,compute_mask,tensorflow/tensorflow/python/keras/layers/merge.py,728,method,
6381,get_config,tensorflow/tensorflow/python/keras/layers/merge.py,731,method,
6382,add,tensorflow/tensorflow/python/keras/layers/merge.py,741,function,"Functional interface to the `tf.keras.layers.Add` layer.
Arguments:
inputs: A list of input tensors (at least 2) with the same shape.
**kwargs: Standard layer keyword arguments.
Returns:
A tensor as the sum of the inputs. It has the same shape as the inputs.
Examples:
>>> input_shape = (2, 3, 4)
>>> x1 = tf.random.normal(input_shape)
>>> x2 = tf.random.normal(input_shape)
>>> y = tf.keras.layers.add([x1, x2])
>>> print(y.shape)
(2, 3, 4)
Used in a functiona model:
>>> input1 = tf.keras.layers.Input(shape=(16,))
>>> x1 = tf.keras.layers.Dense(8, activation='relu')(input1)
>>> input2 = tf.keras.layers.Input(shape=(32,))
>>> x2 = tf.keras.layers.Dense(8, activation='relu')(input2)
>>> added = tf.keras.layers.add([x1, x2])
>>> out = tf.keras.layers.Dense(4)(added)
>>> model = tf.keras.models.Model(inputs=[input1, input2], outputs=out)"
6383,subtract,tensorflow/tensorflow/python/keras/layers/merge.py,775,function,"Functional interface to the `Subtract` layer.
Arguments:
inputs: A list of input tensors (exactly 2).
**kwargs: Standard layer keyword arguments.
Returns:
A tensor, the difference of the inputs.
Examples:
```python
import keras
input1 = keras.layers.Input(shape=(16,))
x1 = keras.layers.Dense(8, activation='relu')(input1)
input2 = keras.layers.Input(shape=(32,))
x2 = keras.layers.Dense(8, activation='relu')(input2)
subtracted = keras.layers.subtract([x1, x2])
out = keras.layers.Dense(4)(subtracted)
model = keras.models.Model(inputs=[input1, input2], outputs=out)
```"
6384,multiply,tensorflow/tensorflow/python/keras/layers/merge.py,804,function,"Functional interface to the `Multiply` layer.
Arguments:
inputs: A list of input tensors (at least 2).
**kwargs: Standard layer keyword arguments.
Returns:
A tensor, the element-wise product of the inputs."
6385,average,tensorflow/tensorflow/python/keras/layers/merge.py,818,function,"Functional interface to the `tf.keras.layers.Average` layer.
Example:
>>> x1 = np.ones((2, 2))
>>> x2 = np.zeros((2, 2))
>>> y = tf.keras.layers.Average()([x1, x2])
>>> y.numpy().tolist()
[[0.5, 0.5], [0.5, 0.5]]
Usage in a functional model:
>>> input1 = tf.keras.layers.Input(shape=(16,))
>>> x1 = tf.keras.layers.Dense(8, activation='relu')(input1)
>>> input2 = tf.keras.layers.Input(shape=(32,))
>>> x2 = tf.keras.layers.Dense(8, activation='relu')(input2)
>>> avg = tf.keras.layers.Average()([x1, x2])
>>> out = tf.keras.layers.Dense(4)(avg)
>>> model = tf.keras.models.Model(inputs=[input1, input2], outputs=out)
Arguments:
inputs: A list of input tensors (at least 2).
**kwargs: Standard layer keyword arguments.
Returns:
A tensor, the average of the inputs.
Raises:
ValueError: If there is a shape mismatch between the inputs and the shapes
cannot be broadcasted to match."
6386,maximum,tensorflow/tensorflow/python/keras/layers/merge.py,854,function,"Functional interface to compute maximum (element-wise) list of `inputs`.
This is equivalent to the `tf.keras.layers.Maximum` layer.
For example:
```python
input1 = tf.keras.layers.Input(shape=(16,))
x1 = tf.keras.layers.Dense(8, activation='relu')(input1) #shape=(None, 8)
input2 = tf.keras.layers.Input(shape=(32,))
x2 = tf.keras.layers.Dense(8, activation='relu')(input2) #shape=(None, 8)
max_inp=tf.keras.layers.maximum([x1,x2]) #shape=(None, 8)
out = tf.keras.layers.Dense(4)(max_inp)
model = tf.keras.models.Model(inputs=[input1, input2], outputs=out)
```
Arguments:
inputs: A list of input tensors (at least 2) of same shape.
**kwargs: Standard layer keyword arguments.
Returns:
A tensor (of same shape as input tensor) with the element-wise
maximum of the inputs.
Raises:
ValueError: If input tensors are of different shape."
6387,minimum,tensorflow/tensorflow/python/keras/layers/merge.py,886,function,"Functional interface to the `Minimum` layer.
Arguments:
inputs: A list of input tensors (at least 2).
**kwargs: Standard layer keyword arguments.
Returns:
A tensor, the element-wise minimum of the inputs."
6388,concatenate,tensorflow/tensorflow/python/keras/layers/merge.py,900,function,"Functional interface to the `Concatenate` layer.
>>> x = np.arange(20).reshape(2, 2, 5)
>>> print(x)
[[[ 0 1 2 3 4]
[ 5 6 7 8 9]]
[[10 11 12 13 14]
[15 16 17 18 19]]]
>>> y = np.arange(20, 30).reshape(2, 1, 5)
>>> print(y)
[[[20 21 22 23 24]]
[[25 26 27 28 29]]]
>>> tf.keras.layers.concatenate([x, y],
... axis=1)
<tf.Tensor: shape=(2, 3, 5), dtype=int64, numpy=
array([[[ 0, 1, 2, 3, 4],
[ 5, 6, 7, 8, 9],
[20, 21, 22, 23, 24]],
[[10, 11, 12, 13, 14],
[15, 16, 17, 18, 19],
[25, 26, 27, 28, 29]]])>
Arguments:
inputs: A list of input tensors (at least 2).
axis: Concatenation axis.
**kwargs: Standard layer keyword arguments.
Returns:
A tensor, the concatenation of the inputs alongside axis `axis`."
6389,dot,tensorflow/tensorflow/python/keras/layers/merge.py,935,function,"Functional interface to the `Dot` layer.
Arguments:
inputs: A list of input tensors (at least 2).
axes: Integer or tuple of integers,
axis or axes along which to take the dot product.
normalize: Whether to L2-normalize samples along the
dot product axis before taking the dot product.
If set to True, then the output of the dot product
is the cosine proximity between the two samples.
**kwargs: Standard layer keyword arguments.
Returns:
A tensor, the dot product of the samples from the inputs."
6390,GaussianNoise,tensorflow/tensorflow/python/keras/layers/noise.py,32,class,"Apply additive zero-centered Gaussian noise.
This is useful to mitigate overfitting
(you could see it as a form of random data augmentation).
Gaussian Noise (GS) is a natural choice as corruption process
for real valued inputs.
As it is a regularization layer, it is only active at training time.
Arguments:
stddev: Float, standard deviation of the noise distribution.
Call arguments:
inputs: Input tensor (of any rank).
training: Python boolean indicating whether the layer should behave in
training mode (adding noise) or in inference mode (doing nothing).
Input shape:
Arbitrary. Use the keyword argument `input_shape`
(tuple of integers, does not include the samples axis)
when using this layer as the first layer in a model.
Output shape:
Same shape as input."
6391,call,tensorflow/tensorflow/python/keras/layers/noise.py,64,method,
6392,get_config,tensorflow/tensorflow/python/keras/layers/noise.py,75,method,
6393,compute_output_shape,tensorflow/tensorflow/python/keras/layers/noise.py,81,method,
6394,noised,tensorflow/tensorflow/python/keras/layers/noise.py,66,method,
6395,GaussianDropout,tensorflow/tensorflow/python/keras/layers/noise.py,86,class,"Apply multiplicative 1-centered Gaussian noise.
As it is a regularization layer, it is only active at training time.
Arguments:
rate: Float, drop probability (as with `Dropout`).
The multiplicative noise will have
standard deviation `sqrt(rate / (1 - rate))`.
Call arguments:
inputs: Input tensor (of any rank).
training: Python boolean indicating whether the layer should behave in
training mode (adding dropout) or in inference mode (doing nothing).
Input shape:
Arbitrary. Use the keyword argument `input_shape`
(tuple of integers, does not include the samples axis)
when using this layer as the first layer in a model.
Output shape:
Same shape as input."
6396,call,tensorflow/tensorflow/python/keras/layers/noise.py,115,method,
6397,get_config,tensorflow/tensorflow/python/keras/layers/noise.py,129,method,
6398,compute_output_shape,tensorflow/tensorflow/python/keras/layers/noise.py,135,method,
6399,noised,tensorflow/tensorflow/python/keras/layers/noise.py,118,method,
6400,AlphaDropout,tensorflow/tensorflow/python/keras/layers/noise.py,140,class,"Applies Alpha Dropout to the input.
Alpha Dropout is a `Dropout` that keeps mean and variance of inputs
to their original values, in order to ensure the self-normalizing property
even after this dropout.
Alpha Dropout fits well to Scaled Exponential Linear Units
by randomly setting activations to the negative saturation value.
Arguments:
rate: float, drop probability (as with `Dropout`).
The multiplicative noise will have
standard deviation `sqrt(rate / (1 - rate))`.
seed: A Python integer to use as random seed.
Call arguments:
inputs: Input tensor (of any rank).
training: Python boolean indicating whether the layer should behave in
training mode (adding dropout) or in inference mode (doing nothing).
Input shape:
Arbitrary. Use the keyword argument `input_shape`
(tuple of integers, does not include the samples axis)
when using this layer as the first layer in a model.
Output shape:
Same shape as input."
6401,call,tensorflow/tensorflow/python/keras/layers/noise.py,179,method,
6402,get_config,tensorflow/tensorflow/python/keras/layers/noise.py,205,method,
6403,compute_output_shape,tensorflow/tensorflow/python/keras/layers/noise.py,211,method,
6404,dropped_inputs,tensorflow/tensorflow/python/keras/layers/noise.py,183,method,
6405,BatchNormalizationBase,tensorflow/tensorflow/python/keras/layers/normalization.py,43,class,"Normalize and scale inputs or activations.
Normalize the activations of the previous layer at each batch,
i.e. applies a transformation that maintains the mean activation
close to 0 and the activation standard deviation close to 1.
Batch normalization differs from other layers in several key aspects:
1) Adding BatchNormalization with `training=True` to a model causes the
result of one example to depend on the contents of all other examples in a
minibatch. Be careful when padding batches or masking examples, as these can
change the minibatch statistics and affect other examples.
2) Updates to the weights (moving statistics) are based on the forward pass
of a model rather than the result of gradient computations.
3) When performing inference using a model containing batch normalization, it
is generally (though not always) desirable to use accumulated statistics
rather than mini-batch statistics. This is accomplished by passing
`training=False` when calling the model, or using `model.predict`.
Arguments:
axis: Integer, the axis that should be normalized (typically the features
axis). For instance, after a `Conv2D` layer with
`data_format=""channels_first""`, set `axis=1` in `BatchNormalization`.
momentum: Momentum for the moving average.
epsilon: Small float added to variance to avoid dividing by zero.
center: If True, add offset of `beta` to normalized tensor. If False, `beta`
is ignored.
scale: If True, multiply by `gamma`. If False, `gamma` is not used. When the
next layer is linear (also e.g. `nn.relu`), this can be disabled since the
scaling will be done by the next layer.
beta_initializer: Initializer for the beta weight.
gamma_initializer: Initializer for the gamma weight.
moving_mean_initializer: Initializer for the moving mean.
moving_variance_initializer: Initializer for the moving variance.
beta_regularizer: Optional regularizer for the beta weight.
gamma_regularizer: Optional regularizer for the gamma weight.
beta_constraint: Optional constraint for the beta weight.
gamma_constraint: Optional constraint for the gamma weight.
renorm: Whether to use [Batch Renormalization](
https://arxiv.org/abs/1702.03275). This adds extra variables during
training. The inference is the same for either value of this parameter.
renorm_clipping: A dictionary that may map keys 'rmax', 'rmin', 'dmax' to
scalar `Tensors` used to clip the renorm correction. The correction `(r,
d)` is used as `corrected_value = normalized_value * r + d`, with `r`
clipped to [rmin, rmax], and `d` to [-dmax, dmax]. Missing rmax, rmin,
dmax are set to inf, 0, inf, respectively.
renorm_momentum: Momentum used to update the moving means and standard
deviations with renorm. Unlike `momentum`, this affects training and
should be neither too small (which would add noise) nor too large (which
would give stale estimates). Note that `momentum` is still applied to get
the means and variances for inference.
fused: if `True`, use a faster, fused implementation, or raise a ValueError
if the fused implementation cannot be used. If `None`, use the faster
implementation if possible. If False, do not used the fused
implementation.
trainable: Boolean, if `True` the variables will be marked as trainable.
virtual_batch_size: An `int`. By default, `virtual_batch_size` is `None`,
which means batch normalization is performed across the whole batch. When
`virtual_batch_size` is not `None`, instead perform ""Ghost Batch
Normalization"", which creates virtual sub-batches which are each
normalized separately (with shared gamma, beta, and moving statistics).
Must divide the actual batch size during execution.
adjustment: A function taking the `Tensor` containing the (dynamic) shape of
the input tensor and returning a pair (scale, bias) to apply to the
normalized values (before gamma and beta), only during training. For
example, if axis==-1,
`adjustment = lambda shape: (
tf.random.uniform(shape[-1:], 0.93, 1.07),
tf.random.uniform(shape[-1:], -0.1, 0.1))` will scale the normalized
value by up to 7% up or down, then shift the result by up to 0.1
(with independent scaling and bias for each feature but shared
across all examples), and finally apply gamma and/or beta. If
`None`, no adjustment is applied. Cannot be specified if
virtual_batch_size is specified.
Call arguments:
inputs: Input tensor (of any rank).
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode.
- `training=True`: The layer will normalize its inputs using the mean and
variance of the current batch of inputs.
- `training=False`: The layer will normalize its inputs using the mean and
variance of its moving statistics, learned during training.
Input shape: Arbitrary. Use the keyword argument `input_shape` (tuple of
integers, does not include the samples axis) when using this layer as the
first layer in a model.
Output shape: Same shape as input. {{TRAINABLE_ATTRIBUTE_NOTE}}
Normalization equations: Consider the intermediate activations \(x\) of a
mini-batch of size
\\(m\\): We can compute the mean and variance of the batch \\({\mu_B} =
\frac{1}{m} \sum_{i=1}^{m} {x_i}\\) \\({\sigma_B^2} = \frac{1}{m}
\sum_{i=1}^{m} ({x_i} - {\mu_B})^2\\) and then compute a normalized
\\(x\\), including a small factor \\({\epsilon}\\) for numerical
stability. \\(\hat{x_i} = \frac{x_i - \mu_B}{\sqrt{\sigma_B^2 +
\epsilon}}\\) And finally \\(\hat{x}\) is linearly transformed by
\({\gamma}\\)
and \\({\beta}\\), which are learned parameters: \\({y_i} = {\gamma *
\hat{x_i} + \beta}\\)
Reference:
- [Ioffe and Szegedy, 2015](https://arxiv.org/abs/1502.03167)."
6406,trainable,tensorflow/tensorflow/python/keras/layers/normalization.py,264,method,
6407,trainable,tensorflow/tensorflow/python/keras/layers/normalization.py,268,method,
6408,build,tensorflow/tensorflow/python/keras/layers/normalization.py,284,method,
6409,call,tensorflow/tensorflow/python/keras/layers/normalization.py,697,method,
6410,compute_output_shape,tensorflow/tensorflow/python/keras/layers/normalization.py,875,method,
6411,get_config,tensorflow/tensorflow/python/keras/layers/normalization.py,878,method,
6412,mean_update,tensorflow/tensorflow/python/keras/layers/normalization.py,589,method,Update self.moving_mean with the most recent data point.
6413,variance_update,tensorflow/tensorflow/python/keras/layers/normalization.py,597,method,Update self.moving_variance with the most recent data point.
6414,undo_virtual_batching,tensorflow/tensorflow/python/keras/layers/normalization.py,715,method,
6415,mean_update,tensorflow/tensorflow/python/keras/layers/normalization.py,826,method,
6416,variance_update,tensorflow/tensorflow/python/keras/layers/normalization.py,831,method,Update the moving variance.
6417,moving_stddev_initializer,tensorflow/tensorflow/python/keras/layers/normalization.py,429,method,
6418,true_branch_renorm,tensorflow/tensorflow/python/keras/layers/normalization.py,834,method,
6419,replace_in_base_docstring,tensorflow/tensorflow/python/keras/layers/normalization.py,925,function,
6420,BatchNormalization,tensorflow/tensorflow/python/keras/layers/normalization.py,934,class,
6421,LayerNormalization,tensorflow/tensorflow/python/keras/layers/normalization.py,949,class,"Layer normalization layer (Ba et al., 2016).
Normalize the activations of the previous layer for each given example in a
batch independently, rather than across a batch like Batch Normalization.
i.e. applies a transformation that maintains the mean activation within each
example close to 0 and the activation standard deviation close to 1.
Given a tensor `inputs`, moments are calculated and normalization
is performed across the axes specified in `axis`.
Example:
>>> data = tf.constant(np.arange(10).reshape(5, 2) * 10, dtype=tf.float32)
>>> print(data)
tf.Tensor(
[[ 0. 10.]
[20. 30.]
[40. 50.]
[60. 70.]
[80. 90.]], shape=(5, 2), dtype=float32)
>>> layer = tf.keras.layers.LayerNormalization(axis=1)
>>> output = layer(data)
>>> print(output)
tf.Tensor(
[[-1. 1.]
[-1. 1.]
[-1. 1.]
[-1. 1.]
[-1. 1.]], shape=(5, 2), dtype=float32)
Notice that with Layer Normalization the normalization happens across the
axes *within* each example, rather than across different examples in the
batch.
If `scale` or `center` are enabled, the layer will scale the normalized
outputs by broadcasting them with a trainable variable `gamma`, and center
the outputs by broadcasting with a trainable variable `beta`. `gamma` will
default to a ones tensor and `beta` will default to a zeros tensor, so that
centering and scaling are no-ops before training has begun.
So, with scaling and centering enabled the normalization equations
are as follows:
Let the intermediate activations for a mini-batch to be the `inputs`.
For each sample `x_i` in `inputs` with `k` features, we compute the mean and
variance of the sample:
```python
mean_i = sum(x_i[j] for j in range(k)) / k
var_i = sum((x_i[j] - mean_i) ** 2 for j in range(k)) / k
```
and then compute a normalized `x_i_normalized`, including a small factor
`epsilon` for numerical stability.
```python
x_i_normalized = (x_i - mean_i) / sqrt(var_i + epsilon)
```
And finally `x_i_normalized ` is linearly transformed by `gamma` and `beta`,
which are learned parameters:
```python
output_i = x_i_normalized * gamma + beta
```
`gamma` and `beta` will span the axes of `inputs` specified in `axis`, and
this part of the inputs' shape must be fully defined.
For example:
>>> layer = tf.keras.layers.LayerNormalization(axis=[1, 2, 3])
>>> layer.build([5, 20, 30, 40])
>>> print(layer.beta.shape)
(20, 30, 40)
>>> print(layer.gamma.shape)
(20, 30, 40)
Note that other implementations of layer normalization may choose to define
`gamma` and `beta` over a separate set of axes from the axes being
normalized across. For example, Group Normalization
([Wu et al. 2018](https://arxiv.org/abs/1803.08494)) with group size of 1
corresponds to a Layer Normalization that normalizes across height, width,
and channel and has `gamma` and `beta` span only the channel dimension.
So, this Layer Normalization implementation will not match a Group
Normalization layer with group size set to 1.
Arguments:
axis: Integer or List/Tuple. The axis or axes to normalize across. Typically
this is the features axis/axes. The left-out axes are typically the batch
axis/axes. This argument defaults to `-1`, the last dimension in the
input.
epsilon: Small float added to variance to avoid dividing by zero. Defaults
to 1e-3
center: If True, add offset of `beta` to normalized tensor. If False, `beta`
is ignored. Defaults to True.
scale: If True, multiply by `gamma`. If False, `gamma` is not used. Defaults
to True. When the next layer is linear (also e.g. `nn.relu`), this can be
disabled since the scaling will be done by the next layer.
beta_initializer: Initializer for the beta weight. Defaults to zeros.
gamma_initializer: Initializer for the gamma weight. Defaults to ones.
beta_regularizer: Optional regularizer for the beta weight. None by default.
gamma_regularizer: Optional regularizer for the gamma weight. None by
default.
beta_constraint: Optional constraint for the beta weight. None by default.
gamma_constraint: Optional constraint for the gamma weight. None by default.
trainable: Boolean, if `True` the variables will be marked as trainable.
Defaults to True.
Input shape: Arbitrary. Use the keyword argument `input_shape` (tuple of
integers, does not include the samples axis) when using this layer as the
first layer in a model.
Output shape: Same shape as input.
Reference:
- [Lei Ba et al., 2016](https://arxiv.org/abs/1607.06450)."
6422,build,tensorflow/tensorflow/python/keras/layers/normalization.py,1129,method,
6423,call,tensorflow/tensorflow/python/keras/layers/normalization.py,1179,method,
6424,compute_output_shape,tensorflow/tensorflow/python/keras/layers/normalization.py,1267,method,
6425,get_config,tensorflow/tensorflow/python/keras/layers/normalization.py,1270,method,
6426,SyncBatchNormalization,tensorflow/tensorflow/python/keras/layers/normalization_v2.py,32,class,"Normalize and scale inputs or activations synchronously across replicas.
Applies batch normalization to activations of the previous layer at each batch
by synchronizing the global batch statistics across all devices that are
training the model. For specific details about batch normalization please
refer to the `tf.keras.layers.BatchNormalization` layer docs.
If this layer is used when using tf.distribute strategy to train models
across devices/workers, there will be an allreduce call to aggregate batch
statistics across all replicas at every training step. Without tf.distribute
strategy, this layer behaves as a regular `tf.keras.layers.BatchNormalization`
layer.
Example usage:
```
strategy = tf.distribute.MirroredStrategy()
with strategy.scope():
model = tf.keras.Sequential()
model.add(tf.keras.layers.Dense(16))
model.add(tf.keras.layers.experimental.SyncBatchNormalization())
```
Arguments:
axis: Integer, the axis that should be normalized
(typically the features axis).
For instance, after a `Conv2D` layer with
`data_format=""channels_first""`,
set `axis=1` in `BatchNormalization`.
momentum: Momentum for the moving average.
epsilon: Small float added to variance to avoid dividing by zero.
center: If True, add offset of `beta` to normalized tensor.
If False, `beta` is ignored.
scale: If True, multiply by `gamma`.
If False, `gamma` is not used.
When the next layer is linear (also e.g. `nn.relu`),
this can be disabled since the scaling
will be done by the next layer.
beta_initializer: Initializer for the beta weight.
gamma_initializer: Initializer for the gamma weight.
moving_mean_initializer: Initializer for the moving mean.
moving_variance_initializer: Initializer for the moving variance.
beta_regularizer: Optional regularizer for the beta weight.
gamma_regularizer: Optional regularizer for the gamma weight.
beta_constraint: Optional constraint for the beta weight.
gamma_constraint: Optional constraint for the gamma weight.
renorm: Whether to use [Batch Renormalization](
https://arxiv.org/abs/1702.03275). This adds extra variables during
training. The inference is the same for either value of this parameter.
renorm_clipping: A dictionary that may map keys 'rmax', 'rmin', 'dmax' to
scalar `Tensors` used to clip the renorm correction. The correction
`(r, d)` is used as `corrected_value = normalized_value * r + d`, with
`r` clipped to [rmin, rmax], and `d` to [-dmax, dmax]. Missing rmax, rmin,
dmax are set to inf, 0, inf, respectively.
renorm_momentum: Momentum used to update the moving means and standard
deviations with renorm. Unlike `momentum`, this affects training
and should be neither too small (which would add noise) nor too large
(which would give stale estimates). Note that `momentum` is still applied
to get the means and variances for inference.
trainable: Boolean, if `True` the variables will be marked as trainable.
Call arguments:
inputs: Input tensor (of any rank).
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode.
- `training=True`: The layer will normalize its inputs using the
mean and variance of the current batch of inputs.
- `training=False`: The layer will normalize its inputs using the
mean and variance of its moving statistics, learned during training.
Input shape:
Arbitrary. Use the keyword argument `input_shape`
(tuple of integers, does not include the samples axis)
when using this layer as the first layer in a model.
Output shape:
Same shape as input."
6427,BatchNormalization,tensorflow/tensorflow/python/keras/layers/normalization_v2.py,208,class,
6428,Pooling1D,tensorflow/tensorflow/python/keras/layers/pooling.py,34,class,"Pooling layer for arbitrary pooling functions, for 1D inputs.
This class only exists for code reuse. It will never be an exposed API.
Arguments:
pool_function: The pooling function to apply, e.g. `tf.nn.max_pool2d`.
pool_size: An integer or tuple/list of a single integer,
representing the size of the pooling window.
strides: An integer or tuple/list of a single integer, specifying the
strides of the pooling operation.
padding: A string. The padding method, either 'valid' or 'same'.
Case-insensitive.
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, steps, features)` while `channels_first`
corresponds to inputs with shape
`(batch, features, steps)`.
name: A string, the name of the layer."
6429,call,tensorflow/tensorflow/python/keras/layers/pooling.py,72,method,
6430,compute_output_shape,tensorflow/tensorflow/python/keras/layers/pooling.py,83,method,
6431,get_config,tensorflow/tensorflow/python/keras/layers/pooling.py,100,method,
6432,MaxPooling1D,tensorflow/tensorflow/python/keras/layers/pooling.py,112,class,"Max pooling operation for 1D temporal data.
Downsamples the input representation by taking the maximum value over the
window defined by `pool_size`. The window is shifted by `strides`. The
resulting output when using ""valid"" padding option has a shape of:
`output_shape = (input_shape - pool_size + 1) / strides)`
The resulting output shape when using the ""same"" padding option is:
`output_shape = input_shape / strides`
For example, for strides=1 and padding=""valid"":
>>> x = tf.constant([1., 2., 3., 4., 5.])
>>> x = tf.reshape(x, [1, 5, 1])
>>> max_pool_1d = tf.keras.layers.MaxPooling1D(pool_size=2,
... strides=1, padding='valid')
>>> max_pool_1d(x)
<tf.Tensor: shape=(1, 4, 1), dtype=float32, numpy=
array([[[2.],
[3.],
[4.],
[5.]]], dtype=float32)>
For example, for strides=2 and padding=""valid"":
>>> x = tf.constant([1., 2., 3., 4., 5.])
>>> x = tf.reshape(x, [1, 5, 1])
>>> max_pool_1d = tf.keras.layers.MaxPooling1D(pool_size=2,
... strides=2, padding='valid')
>>> max_pool_1d(x)
<tf.Tensor: shape=(1, 2, 1), dtype=float32, numpy=
array([[[2.],
[4.]]], dtype=float32)>
For example, for strides=1 and padding=""same"":
>>> x = tf.constant([1., 2., 3., 4., 5.])
>>> x = tf.reshape(x, [1, 5, 1])
>>> max_pool_1d = tf.keras.layers.MaxPooling1D(pool_size=2,
... strides=1, padding='same')
>>> max_pool_1d(x)
<tf.Tensor: shape=(1, 5, 1), dtype=float32, numpy=
array([[[2.],
[3.],
[4.],
[5.],
[5.]]], dtype=float32)>
Arguments:
pool_size: Integer, size of the max pooling window.
strides: Integer, or None. Specifies how much the pooling window moves
for each pooling step.
If None, it will default to `pool_size`.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, steps, features)` while `channels_first`
corresponds to inputs with shape
`(batch, features, steps)`.
Input shape:
- If `data_format='channels_last'`:
3D tensor with shape `(batch_size, steps, features)`.
- If `data_format='channels_first'`:
3D tensor with shape `(batch_size, features, steps)`.
Output shape:
- If `data_format='channels_last'`:
3D tensor with shape `(batch_size, downsampled_steps, features)`.
- If `data_format='channels_first'`:
3D tensor with shape `(batch_size, features, downsampled_steps)`."
6433,AveragePooling1D,tensorflow/tensorflow/python/keras/layers/pooling.py,204,class,"Average pooling for temporal data.
Arguments:
pool_size: Integer, size of the average pooling windows.
strides: Integer, or None. Factor by which to downscale.
E.g. 2 will halve the input.
If None, it will default to `pool_size`.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, steps, features)` while `channels_first`
corresponds to inputs with shape
`(batch, features, steps)`.
Input shape:
- If `data_format='channels_last'`:
3D tensor with shape `(batch_size, steps, features)`.
- If `data_format='channels_first'`:
3D tensor with shape `(batch_size, features, steps)`.
Output shape:
- If `data_format='channels_last'`:
3D tensor with shape `(batch_size, downsampled_steps, features)`.
- If `data_format='channels_first'`:
3D tensor with shape `(batch_size, features, downsampled_steps)`."
6434,Pooling2D,tensorflow/tensorflow/python/keras/layers/pooling.py,248,class,"Pooling layer for arbitrary pooling functions, for 2D inputs (e.g. images).
This class only exists for code reuse. It will never be an exposed API.
Arguments:
pool_function: The pooling function to apply, e.g. `tf.nn.max_pool2d`.
pool_size: An integer or tuple/list of 2 integers: (pool_height, pool_width)
specifying the size of the pooling window.
Can be a single integer to specify the same value for
all spatial dimensions.
strides: An integer or tuple/list of 2 integers,
specifying the strides of the pooling operation.
Can be a single integer to specify the same value for
all spatial dimensions.
padding: A string. The padding method, either 'valid' or 'same'.
Case-insensitive.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, height, width, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, height, width)`.
name: A string, the name of the layer."
6435,call,tensorflow/tensorflow/python/keras/layers/pooling.py,288,method,
6436,compute_output_shape,tensorflow/tensorflow/python/keras/layers/pooling.py,303,method,
6437,get_config,tensorflow/tensorflow/python/keras/layers/pooling.py,322,method,
6438,MaxPooling2D,tensorflow/tensorflow/python/keras/layers/pooling.py,334,class,"Max pooling operation for 2D spatial data.
Downsamples the input representation by taking the maximum value over the
window defined by `pool_size` for each dimension along the features axis.
The window is shifted by `strides` in each dimension. The resulting output
when using ""valid"" padding option has a shape(number of rows or columns) of:
`output_shape = (input_shape - pool_size + 1) / strides)`
The resulting output shape when using the ""same"" padding option is:
`output_shape = input_shape / strides`
For example, for stride=(1,1) and padding=""valid"":
>>> x = tf.constant([[1., 2., 3.],
... [4., 5., 6.],
... [7., 8., 9.]])
>>> x = tf.reshape(x, [1, 3, 3, 1])
>>> max_pool_2d = tf.keras.layers.MaxPooling2D(pool_size=(2, 2),
... strides=(1, 1), padding='valid')
>>> max_pool_2d(x)
<tf.Tensor: shape=(1, 2, 2, 1), dtype=float32, numpy=
array([[[[5.],
[6.]],
[[8.],
[9.]]]], dtype=float32)>
For example, for stride=(2,2) and padding=""valid"":
>>> x = tf.constant([[1., 2., 3., 4.],
... [5., 6., 7., 8.],
... [9., 10., 11., 12.]])
>>> x = tf.reshape(x, [1, 3, 4, 1])
>>> max_pool_2d = tf.keras.layers.MaxPooling2D(pool_size=(2, 2),
... strides=(1, 1), padding='valid')
>>> max_pool_2d(x)
<tf.Tensor: shape=(1, 2, 3, 1), dtype=float32, numpy=
array([[[[ 6.],
[ 7.],
[ 8.]],
[[10.],
[11.],
[12.]]]], dtype=float32)>
Usage Example:
>>> input_image = tf.constant([[[[1.], [1.], [2.], [4.]],
... [[2.], [2.], [3.], [2.]],
... [[4.], [1.], [1.], [1.]],
... [[2.], [2.], [1.], [4.]]]])
>>> output = tf.constant([[[[1], [0]],
... [[0], [1]]]])
>>> model = tf.keras.models.Sequential()
>>> model.add(tf.keras.layers.MaxPooling2D(pool_size=(2, 2),
... input_shape=(4,4,1)))
>>> model.compile('adam', 'mean_squared_error')
>>> model.predict(input_image, steps=1)
array([[[[2.],
[4.]],
[[4.],
[4.]]]], dtype=float32)
For example, for stride=(1,1) and padding=""same"":
>>> x = tf.constant([[1., 2., 3.],
... [4., 5., 6.],
... [7., 8., 9.]])
>>> x = tf.reshape(x, [1, 3, 3, 1])
>>> max_pool_2d = tf.keras.layers.MaxPooling2D(pool_size=(2, 2),
... strides=(1, 1), padding='same')
>>> max_pool_2d(x)
<tf.Tensor: shape=(1, 3, 3, 1), dtype=float32, numpy=
array([[[[5.],
[6.],
[6.]],
[[8.],
[9.],
[9.]],
[[8.],
[9.],
[9.]]]], dtype=float32)>
Arguments:
pool_size: integer or tuple of 2 integers,
window size over which to take the maximum.
`(2, 2)` will take the max value over a 2x2 pooling window.
If only one integer is specified, the same window length
will be used for both dimensions.
strides: Integer, tuple of 2 integers, or None.
Strides values. Specifies how far the pooling window moves
for each pooling step. If None, it will default to `pool_size`.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch, channels, height, width)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
Input shape:
- If `data_format='channels_last'`:
4D tensor with shape `(batch_size, rows, cols, channels)`.
- If `data_format='channels_first'`:
4D tensor with shape `(batch_size, channels, rows, cols)`.
Output shape:
- If `data_format='channels_last'`:
4D tensor with shape `(batch_size, pooled_rows, pooled_cols, channels)`.
- If `data_format='channels_first'`:
4D tensor with shape `(batch_size, channels, pooled_rows, pooled_cols)`.
Returns:
A tensor of rank 4 representing the maximum pooled values. See above for
output shape."
6439,AveragePooling2D,tensorflow/tensorflow/python/keras/layers/pooling.py,470,class,"Average pooling operation for spatial data.
Arguments:
pool_size: integer or tuple of 2 integers,
factors by which to downscale (vertical, horizontal).
`(2, 2)` will halve the input in both spatial dimension.
If only one integer is specified, the same window length
will be used for both dimensions.
strides: Integer, tuple of 2 integers, or None.
Strides values.
If None, it will default to `pool_size`.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch, channels, height, width)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
Input shape:
- If `data_format='channels_last'`:
4D tensor with shape `(batch_size, rows, cols, channels)`.
- If `data_format='channels_first'`:
4D tensor with shape `(batch_size, channels, rows, cols)`.
Output shape:
- If `data_format='channels_last'`:
4D tensor with shape `(batch_size, pooled_rows, pooled_cols, channels)`.
- If `data_format='channels_first'`:
4D tensor with shape `(batch_size, channels, pooled_rows, pooled_cols)`."
6440,Pooling3D,tensorflow/tensorflow/python/keras/layers/pooling.py,522,class,"Pooling layer for arbitrary pooling functions, for 3D inputs.
This class only exists for code reuse. It will never be an exposed API.
Arguments:
pool_function: The pooling function to apply, e.g. `tf.nn.max_pool2d`.
pool_size: An integer or tuple/list of 3 integers:
(pool_depth, pool_height, pool_width)
specifying the size of the pooling window.
Can be a single integer to specify the same value for
all spatial dimensions.
strides: An integer or tuple/list of 3 integers,
specifying the strides of the pooling operation.
Can be a single integer to specify the same value for
all spatial dimensions.
padding: A string. The padding method, either 'valid' or 'same'.
Case-insensitive.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, depth, height, width, channels)`
while `channels_first` corresponds to
inputs with shape `(batch, channels, depth, height, width)`.
name: A string, the name of the layer."
6441,call,tensorflow/tensorflow/python/keras/layers/pooling.py,564,method,
6442,compute_output_shape,tensorflow/tensorflow/python/keras/layers/pooling.py,584,method,
6443,get_config,tensorflow/tensorflow/python/keras/layers/pooling.py,607,method,
6444,MaxPooling3D,tensorflow/tensorflow/python/keras/layers/pooling.py,619,class,"Max pooling operation for 3D data (spatial or spatio-temporal).
Arguments:
pool_size: Tuple of 3 integers,
factors by which to downscale (dim1, dim2, dim3).
`(2, 2, 2)` will halve the size of the 3D input in each dimension.
strides: tuple of 3 integers, or None. Strides values.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, spatial_dim1, spatial_dim2, spatial_dim3, channels)`
while `channels_first` corresponds to inputs with shape
`(batch, channels, spatial_dim1, spatial_dim2, spatial_dim3)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
Input shape:
- If `data_format='channels_last'`:
5D tensor with shape:
`(batch_size, spatial_dim1, spatial_dim2, spatial_dim3, channels)`
- If `data_format='channels_first'`:
5D tensor with shape:
`(batch_size, channels, spatial_dim1, spatial_dim2, spatial_dim3)`
Output shape:
- If `data_format='channels_last'`:
5D tensor with shape:
`(batch_size, pooled_dim1, pooled_dim2, pooled_dim3, channels)`
- If `data_format='channels_first'`:
5D tensor with shape:
`(batch_size, channels, pooled_dim1, pooled_dim2, pooled_dim3)`"
6445,AveragePooling3D,tensorflow/tensorflow/python/keras/layers/pooling.py,672,class,"Average pooling operation for 3D data (spatial or spatio-temporal).
Arguments:
pool_size: tuple of 3 integers,
factors by which to downscale (dim1, dim2, dim3).
`(2, 2, 2)` will halve the size of the 3D input in each dimension.
strides: tuple of 3 integers, or None. Strides values.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, spatial_dim1, spatial_dim2, spatial_dim3, channels)`
while `channels_first` corresponds to inputs with shape
`(batch, channels, spatial_dim1, spatial_dim2, spatial_dim3)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
Input shape:
- If `data_format='channels_last'`:
5D tensor with shape:
`(batch_size, spatial_dim1, spatial_dim2, spatial_dim3, channels)`
- If `data_format='channels_first'`:
5D tensor with shape:
`(batch_size, channels, spatial_dim1, spatial_dim2, spatial_dim3)`
Output shape:
- If `data_format='channels_last'`:
5D tensor with shape:
`(batch_size, pooled_dim1, pooled_dim2, pooled_dim3, channels)`
- If `data_format='channels_first'`:
5D tensor with shape:
`(batch_size, channels, pooled_dim1, pooled_dim2, pooled_dim3)`"
6446,GlobalPooling1D,tensorflow/tensorflow/python/keras/layers/pooling.py,724,class,Abstract class for different global pooling 1D layers.
6447,compute_output_shape,tensorflow/tensorflow/python/keras/layers/pooling.py,732,method,
6448,call,tensorflow/tensorflow/python/keras/layers/pooling.py,739,method,
6449,get_config,tensorflow/tensorflow/python/keras/layers/pooling.py,742,method,
6450,GlobalAveragePooling1D,tensorflow/tensorflow/python/keras/layers/pooling.py,750,class,"Global average pooling operation for temporal data.
Examples:
>>> input_shape = (2, 3, 4)
>>> x = tf.random.normal(input_shape)
>>> y = tf.keras.layers.GlobalAveragePooling1D()(x)
>>> print(y.shape)
(2, 4)
Arguments:
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, steps, features)` while `channels_first`
corresponds to inputs with shape
`(batch, features, steps)`.
Call arguments:
inputs: A 3D tensor.
mask: Binary tensor of shape `(batch_size, steps)` indicating whether
a given step should be masked (excluded from the average).
Input shape:
- If `data_format='channels_last'`:
3D tensor with shape:
`(batch_size, steps, features)`
- If `data_format='channels_first'`:
3D tensor with shape:
`(batch_size, features, steps)`
Output shape:
2D tensor with shape `(batch_size, features)`."
6451,call,tensorflow/tensorflow/python/keras/layers/pooling.py,792,method,
6452,compute_mask,tensorflow/tensorflow/python/keras/layers/pooling.py,804,method,
6453,GlobalMaxPooling1D,tensorflow/tensorflow/python/keras/layers/pooling.py,809,class,"Global max pooling operation for 1D temporal data.
Downsamples the input representation by taking the maximum value over
the time dimension.
For example:
>>> x = tf.constant([[1., 2., 3.], [4., 5., 6.], [7., 8., 9.]])
>>> x = tf.reshape(x, [3, 3, 1])
>>> x
<tf.Tensor: shape=(3, 3, 1), dtype=float32, numpy=
array([[[1.], [2.], [3.]],
[[4.], [5.], [6.]],
[[7.], [8.], [9.]]], dtype=float32)>
>>> max_pool_1d = tf.keras.layers.GlobalMaxPooling1D()
>>> max_pool_1d(x)
<tf.Tensor: shape=(3, 1), dtype=float32, numpy=
array([[3.],
[6.],
[9.], dtype=float32)>
Arguments:
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, steps, features)` while `channels_first`
corresponds to inputs with shape
`(batch, features, steps)`.
Input shape:
- If `data_format='channels_last'`:
3D tensor with shape:
`(batch_size, steps, features)`
- If `data_format='channels_first'`:
3D tensor with shape:
`(batch_size, features, steps)`
Output shape:
2D tensor with shape `(batch_size, features)`."
6454,call,tensorflow/tensorflow/python/keras/layers/pooling.py,852,method,
6455,GlobalPooling2D,tensorflow/tensorflow/python/keras/layers/pooling.py,857,class,"Abstract class for different global pooling 2D layers.
"
6456,compute_output_shape,tensorflow/tensorflow/python/keras/layers/pooling.py,866,method,
6457,call,tensorflow/tensorflow/python/keras/layers/pooling.py,873,method,
6458,get_config,tensorflow/tensorflow/python/keras/layers/pooling.py,876,method,
6459,GlobalAveragePooling2D,tensorflow/tensorflow/python/keras/layers/pooling.py,884,class,"Global average pooling operation for spatial data.
Examples:
>>> input_shape = (2, 4, 5, 3)
>>> x = tf.random.normal(input_shape)
>>> y = tf.keras.layers.GlobalAveragePooling2D()(x)
>>> print(y.shape)
(2, 3)
Arguments:
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch, channels, height, width)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
Input shape:
- If `data_format='channels_last'`:
4D tensor with shape `(batch_size, rows, cols, channels)`.
- If `data_format='channels_first'`:
4D tensor with shape `(batch_size, channels, rows, cols)`.
Output shape:
2D tensor with shape `(batch_size, channels)`."
6460,call,tensorflow/tensorflow/python/keras/layers/pooling.py,917,method,
6461,GlobalMaxPooling2D,tensorflow/tensorflow/python/keras/layers/pooling.py,925,class,"Global max pooling operation for spatial data.
Examples:
>>> input_shape = (2, 4, 5, 3)
>>> x = tf.random.normal(input_shape)
>>> y = tf.keras.layers.GlobalMaxPool2D()(x)
>>> print(y.shape)
(2, 3)
Arguments:
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch, channels, height, width)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
Input shape:
- If `data_format='channels_last'`:
4D tensor with shape `(batch_size, rows, cols, channels)`.
- If `data_format='channels_first'`:
4D tensor with shape `(batch_size, channels, rows, cols)`.
Output shape:
2D tensor with shape `(batch_size, channels)`."
6462,call,tensorflow/tensorflow/python/keras/layers/pooling.py,958,method,
6463,GlobalPooling3D,tensorflow/tensorflow/python/keras/layers/pooling.py,965,class,Abstract class for different global pooling 3D layers.
6464,compute_output_shape,tensorflow/tensorflow/python/keras/layers/pooling.py,973,method,
6465,call,tensorflow/tensorflow/python/keras/layers/pooling.py,980,method,
6466,get_config,tensorflow/tensorflow/python/keras/layers/pooling.py,983,method,
6467,GlobalAveragePooling3D,tensorflow/tensorflow/python/keras/layers/pooling.py,991,class,"Global Average pooling operation for 3D data.
Arguments:
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, spatial_dim1, spatial_dim2, spatial_dim3, channels)`
while `channels_first` corresponds to inputs with shape
`(batch, channels, spatial_dim1, spatial_dim2, spatial_dim3)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
Input shape:
- If `data_format='channels_last'`:
5D tensor with shape:
`(batch_size, spatial_dim1, spatial_dim2, spatial_dim3, channels)`
- If `data_format='channels_first'`:
5D tensor with shape:
`(batch_size, channels, spatial_dim1, spatial_dim2, spatial_dim3)`
Output shape:
2D tensor with shape `(batch_size, channels)`."
6468,call,tensorflow/tensorflow/python/keras/layers/pooling.py,1018,method,
6469,GlobalMaxPooling3D,tensorflow/tensorflow/python/keras/layers/pooling.py,1026,class,"Global Max pooling operation for 3D data.
Arguments:
data_format: A string,
one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, spatial_dim1, spatial_dim2, spatial_dim3, channels)`
while `channels_first` corresponds to inputs with shape
`(batch, channels, spatial_dim1, spatial_dim2, spatial_dim3)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
Input shape:
- If `data_format='channels_last'`:
5D tensor with shape:
`(batch_size, spatial_dim1, spatial_dim2, spatial_dim3, channels)`
- If `data_format='channels_first'`:
5D tensor with shape:
`(batch_size, channels, spatial_dim1, spatial_dim2, spatial_dim3)`
Output shape:
2D tensor with shape `(batch_size, channels)`."
6470,call,tensorflow/tensorflow/python/keras/layers/pooling.py,1053,method,
6471,StackedRNNCells,tensorflow/tensorflow/python/keras/layers/recurrent.py,59,class,"Wrapper allowing a stack of RNN cells to behave as a single cell.
Used to implement efficient stacked RNNs.
Arguments:
cells: List of RNN cell instances.
Examples:
```python
batch_size = 3
sentence_max_length = 5
n_features = 2
new_shape = (batch_size, sentence_max_length, n_features)
x = tf.constant(np.reshape(np.arange(30), new_shape), dtype = tf.float32)
rnn_cells = [tf.keras.layers.LSTMCell(128) for _ in range(2)]
stacked_lstm = tf.keras.layers.StackedRNNCells(rnn_cells)
lstm_layer = tf.keras.layers.RNN(stacked_lstm)
result = lstm_layer(x)
```"
6472,state_size,tensorflow/tensorflow/python/keras/layers/recurrent.py,107,method,
6473,output_size,tensorflow/tensorflow/python/keras/layers/recurrent.py,112,method,
6474,get_initial_state,tensorflow/tensorflow/python/keras/layers/recurrent.py,120,method,
6475,call,tensorflow/tensorflow/python/keras/layers/recurrent.py,133,method,
6476,build,tensorflow/tensorflow/python/keras/layers/recurrent.py,164,method,
6477,get_config,tensorflow/tensorflow/python/keras/layers/recurrent.py,182,method,
6478,from_config,tensorflow/tensorflow/python/keras/layers/recurrent.py,194,method,
6479,RNN,tensorflow/tensorflow/python/keras/layers/recurrent.py,204,class,"Base class for recurrent layers.
See [the Keras RNN API guide](https://www.tensorflow.org/guide/keras/rnn)
for details about the usage of RNN API.
Arguments:
cell: A RNN cell instance or a list of RNN cell instances.
A RNN cell is a class that has:
- A `call(input_at_t, states_at_t)` method, returning
`(output_at_t, states_at_t_plus_1)`. The call method of the
cell can also take the optional argument `constants`, see
section ""Note on passing external constants"" below.
- A `state_size` attribute. This can be a single integer
(single state) in which case it is the size of the recurrent
state. This can also be a list/tuple of integers (one size per state).
The `state_size` can also be TensorShape or tuple/list of
TensorShape, to represent high dimension state.
- A `output_size` attribute. This can be a single integer or a
TensorShape, which represent the shape of the output. For backward
compatible reason, if this attribute is not available for the
cell, the value will be inferred by the first element of the
`state_size`.
- A `get_initial_state(inputs=None, batch_size=None, dtype=None)`
method that creates a tensor meant to be fed to `call()` as the
initial state, if the user didn't specify any initial state via other
means. The returned initial state should have a shape of
[batch_size, cell.state_size]. The cell might choose to create a
tensor full of zeros, or full of other values based on the cell's
implementation.
`inputs` is the input tensor to the RNN layer, which should
contain the batch size as its shape[0], and also dtype. Note that
the shape[0] might be `None` during the graph construction. Either
the `inputs` or the pair of `batch_size` and `dtype` are provided.
`batch_size` is a scalar tensor that represents the batch size
of the inputs. `dtype` is `tf.DType` that represents the dtype of
the inputs.
For backward compatible reason, if this method is not implemented
by the cell, the RNN layer will create a zero filled tensor with the
size of [batch_size, cell.state_size].
In the case that `cell` is a list of RNN cell instances, the cells
will be stacked on top of each other in the RNN, resulting in an
efficient stacked RNN.
return_sequences: Boolean (default `False`). Whether to return the last
output in the output sequence, or the full sequence.
return_state: Boolean (default `False`). Whether to return the last state
in addition to the output.
go_backwards: Boolean (default `False`).
If True, process the input sequence backwards and return the
reversed sequence.
stateful: Boolean (default `False`). If True, the last state
for each sample at index i in a batch will be used as initial
state for the sample of index i in the following batch.
unroll: Boolean (default `False`).
If True, the network will be unrolled, else a symbolic loop will be used.
Unrolling can speed-up a RNN, although it tends to be more
memory-intensive. Unrolling is only suitable for short sequences.
time_major: The shape format of the `inputs` and `outputs` tensors.
If True, the inputs and outputs will be in shape
`(timesteps, batch, ...)`, whereas in the False case, it will be
`(batch, timesteps, ...)`. Using `time_major = True` is a bit more
efficient because it avoids transposes at the beginning and end of the
RNN calculation. However, most TensorFlow data is batch-major, so by
default this function accepts input and emits output in batch-major
form.
zero_output_for_mask: Boolean (default `False`).
Whether the output should use zeros for the masked timesteps. Note that
this field is only used when `return_sequences` is True and mask is
provided. It can useful if you want to reuse the raw output sequence of
the RNN without interference from the masked timesteps, eg, merging
bidirectional RNNs.
Call arguments:
inputs: Input tensor.
mask: Binary tensor of shape `[batch_size, timesteps]` indicating whether
a given timestep should be masked.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. This argument is passed to the cell
when calling it. This is for use with cells that use dropout.
initial_state: List of initial state tensors to be passed to the first
call of the cell.
constants: List of constant tensors to be passed to the cell at each
timestep.
Input shape:
N-D tensor with shape `[batch_size, timesteps, ...]` or
`[timesteps, batch_size, ...]` when time_major is True.
Output shape:
- If `return_state`: a list of tensors. The first tensor is
the output. The remaining tensors are the last states,
each with shape `[batch_size, state_size]`, where `state_size` could
be a high dimension tensor shape.
- If `return_sequences`: N-D tensor with shape
`[batch_size, timesteps, output_size]`, where `output_size` could
be a high dimension tensor shape, or
`[timesteps, batch_size, output_size]` when `time_major` is True.
- Else, N-D tensor with shape `[batch_size, output_size]`, where
`output_size` could be a high dimension tensor shape.
Masking:
This layer supports masking for input data with a variable number
of timesteps. To introduce masks to your data,
use an [tf.keras.layers.Embedding] layer with the `mask_zero` parameter
set to `True`.
Note on using statefulness in RNNs:
You can set RNN layers to be 'stateful', which means that the states
computed for the samples in one batch will be reused as initial states
for the samples in the next batch. This assumes a one-to-one mapping
between samples in different successive batches.
To enable statefulness:
- Specify `stateful=True` in the layer constructor.
- Specify a fixed batch size for your model, by passing
If sequential model:
`batch_input_shape=(...)` to the first layer in your model.
Else for functional model with 1 or more Input layers:
`batch_shape=(...)` to all the first layers in your model.
This is the expected shape of your inputs
*including the batch size*.
It should be a tuple of integers, e.g. `(32, 10, 100)`.
- Specify `shuffle=False` when calling `fit()`.
To reset the states of your model, call `.reset_states()` on either
a specific layer, or on your entire model.
Note on specifying the initial state of RNNs:
You can specify the initial state of RNN layers symbolically by
calling them with the keyword argument `initial_state`. The value of
`initial_state` should be a tensor or list of tensors representing
the initial state of the RNN layer.
You can specify the initial state of RNN layers numerically by
calling `reset_states` with the keyword argument `states`. The value of
`states` should be a numpy array or list of numpy arrays representing
the initial state of the RNN layer.
Note on passing external constants to RNNs:
You can pass ""external"" constants to the cell using the `constants`
keyword argument of `RNN.__call__` (as well as `RNN.call`) method. This
requires that the `cell.call` method accepts the same keyword argument
`constants`. Such constants can be used to condition the cell
transformation on additional static inputs (not changing over time),
a.k.a. an attention mechanism.
Examples:
```python
# First, let's define a RNN Cell, as a layer subclass.
class MinimalRNNCell(keras.layers.Layer):
def __init__(self, units, **kwargs):
self.units = units
self.state_size = units
super(MinimalRNNCell, self).__init__(**kwargs)
def build(self, input_shape):
self.kernel = self.add_weight(shape=(input_shape[-1], self.units),
initializer='uniform',
name='kernel')
self.recurrent_kernel = self.add_weight(
shape=(self.units, self.units),
initializer='uniform',
name='recurrent_kernel')
self.built = True
def call(self, inputs, states):
prev_output = states[0]
h = K.dot(inputs, self.kernel)
output = h + K.dot(prev_output, self.recurrent_kernel)
return output, [output]
# Let's use this cell in a RNN layer:
cell = MinimalRNNCell(32)
x = keras.Input((None, 5))
layer = RNN(cell)
y = layer(x)
# Here's how to use the cell to build a stacked RNN:
cells = [MinimalRNNCell(32), MinimalRNNCell(64)]
x = keras.Input((None, 5))
layer = RNN(cells)
y = layer(x)
```"
6480,states,tensorflow/tensorflow/python/keras/layers/recurrent.py,448,method,
6481,states,tensorflow/tensorflow/python/keras/layers/recurrent.py,458,method,
6482,compute_output_shape,tensorflow/tensorflow/python/keras/layers/recurrent.py,461,method,
6483,compute_mask,tensorflow/tensorflow/python/keras/layers/recurrent.py,514,method,
6484,build,tensorflow/tensorflow/python/keras/layers/recurrent.py,528,method,
6485,get_initial_state,tensorflow/tensorflow/python/keras/layers/recurrent.py,632,method,
6486,call,tensorflow/tensorflow/python/keras/layers/recurrent.py,721,method,
6487,reset_states,tensorflow/tensorflow/python/keras/layers/recurrent.py,888,method,"Reset the recorded states for the stateful RNN layer.
Can only be used when RNN layer is constructed with `stateful` = `True`.
Args:
states: Numpy arrays that contains the value for the initial state, which
will be feed to cell at the first time step. When the value is None,
zero filled numpy array will be created based on the cell state size.
Raises:
AttributeError: When the RNN layer is not stateful.
ValueError: When the batch size of the RNN layer is unknown.
ValueError: When the input numpy array is not compatible with the RNN
layer state, either size wise or dtype wise."
6488,get_config,tensorflow/tensorflow/python/keras/layers/recurrent.py,958,method,
6489,from_config,tensorflow/tensorflow/python/keras/layers/recurrent.py,981,method,
6490,get_input_spec,tensorflow/tensorflow/python/keras/layers/recurrent.py,535,method,Convert input shape to InputSpec.
6491,get_step_input_shape,tensorflow/tensorflow/python/keras/layers/recurrent.py,547,method,
6492,step,tensorflow/tensorflow/python/keras/layers/recurrent.py,778,method,
6493,step,tensorflow/tensorflow/python/keras/layers/recurrent.py,790,method,
6494,create_state_variable,tensorflow/tensorflow/python/keras/layers/recurrent.py,928,method,
6495,AbstractRNNCell,tensorflow/tensorflow/python/keras/layers/recurrent.py,995,class,"Abstract object representing an RNN cell.
See [the Keras RNN API guide](https://www.tensorflow.org/guide/keras/rnn)
for details about the usage of RNN API.
This is the base class for implementing RNN cells with custom behavior.
Every `RNNCell` must have the properties below and implement `call` with
the signature `(output, next_state) = call(input, state)`.
Examples:
```python
class MinimalRNNCell(AbstractRNNCell):
def __init__(self, units, **kwargs):
self.units = units
super(MinimalRNNCell, self).__init__(**kwargs)
@property
def state_size(self):
return self.units
def build(self, input_shape):
self.kernel = self.add_weight(shape=(input_shape[-1], self.units),
initializer='uniform',
name='kernel')
self.recurrent_kernel = self.add_weight(
shape=(self.units, self.units),
initializer='uniform',
name='recurrent_kernel')
self.built = True
def call(self, inputs, states):
prev_output = states[0]
h = K.dot(inputs, self.kernel)
output = h + K.dot(prev_output, self.recurrent_kernel)
return output, output
```
This definition of cell differs from the definition used in the literature.
In the literature, 'cell' refers to an object with a single scalar output.
This definition refers to a horizontal array of such units.
An RNN cell, in the most abstract setting, is anything that has
a state and performs some operation that takes a matrix of inputs.
This operation results in an output matrix with `self.output_size` columns.
If `self.state_size` is an integer, this operation also results in a new
state matrix with `self.state_size` columns. If `self.state_size` is a
(possibly nested tuple of) TensorShape object(s), then it should return a
matching structure of Tensors having shape `[batch_size].concatenate(s)`
for each `s` in `self.batch_size`."
6496,call,tensorflow/tensorflow/python/keras/layers/recurrent.py,1050,method,"The function that contains the logic for one RNN step calculation.
Args:
inputs: the input tensor, which is a slide from the overall RNN input by
the time dimension (usually the second dimension).
states: the state tensor from previous step, which has the same shape
as `(batch, state_size)`. In the case of timestep 0, it will be the
initial state user specified, or zero filled tensor otherwise.
Returns:
A tuple of two tensors:
1. output tensor for the current timestep, with size `output_size`.
2. state tensor for next step, which has the shape of `state_size`."
6497,state_size,tensorflow/tensorflow/python/keras/layers/recurrent.py,1068,method,"size(s) of state(s) used by this cell.
It can be represented by an Integer, a TensorShape or a tuple of Integers
or TensorShapes."
6498,output_size,tensorflow/tensorflow/python/keras/layers/recurrent.py,1077,method,Integer or TensorShape: size of outputs produced by this cell.
6499,get_initial_state,tensorflow/tensorflow/python/keras/layers/recurrent.py,1081,method,
6500,DropoutRNNCellMixin,tensorflow/tensorflow/python/keras/layers/recurrent.py,1086,class,"Object that hold dropout related fields for RNN Cell.
This class is not a standalone RNN cell. It suppose to be used with a RNN cell
by multiple inheritance. Any cell that mix with class should have following
fields:
dropout: a float number within range [0, 1). The ratio that the input
tensor need to dropout.
recurrent_dropout: a float number within range [0, 1). The ratio that the
recurrent state weights need to dropout.
This object will create and cache created dropout masks, and reuse them for
the incoming data, so that the same mask is used for every batch input."
6501,reset_dropout_mask,tensorflow/tensorflow/python/keras/layers/recurrent.py,1126,method,"Reset the cached dropout masks if any.
This is important for the RNN layer to invoke this in it `call()` method so
that the cached mask is cleared before calling the `cell.call()`. The mask
should be cached across the timestep within the same batch, but shouldn't
be cached between batches. Otherwise it will introduce unreasonable bias
against certain index of data within the batch."
6502,reset_recurrent_dropout_mask,tensorflow/tensorflow/python/keras/layers/recurrent.py,1137,method,"Reset the cached recurrent dropout masks if any.
This is important for the RNN layer to invoke this in it call() method so
that the cached mask is cleared before calling the cell.call(). The mask
should be cached across the timestep within the same batch, but shouldn't
be cached between batches. Otherwise it will introduce unreasonable bias
against certain index of data within the batch."
6503,get_dropout_mask_for_cell,tensorflow/tensorflow/python/keras/layers/recurrent.py,1162,method,"Get the dropout mask for RNN cell's input.
It will create mask based on context if there isn't any existing cached
mask. If a new mask is generated, it will update the cache in the cell.
Args:
inputs: The input tensor whose shape will be used to generate dropout
mask.
training: Boolean tensor, whether its in training mode, dropout will be
ignored in non-training mode.
count: Int, how many dropout mask will be generated. It is useful for cell
that has internal weights fused together.
Returns:
List of mask tensor, generated or cached mask based on context."
6504,get_recurrent_dropout_mask_for_cell,tensorflow/tensorflow/python/keras/layers/recurrent.py,1183,method,"Get the recurrent dropout mask for RNN cell.
It will create mask based on context if there isn't any existing cached
mask. If a new mask is generated, it will update the cache in the cell.
Args:
inputs: The input tensor whose shape will be used to generate dropout
mask.
training: Boolean tensor, whether its in training mode, dropout will be
ignored in non-training mode.
count: Int, how many dropout mask will be generated. It is useful for cell
that has internal weights fused together.
Returns:
List of mask tensor, generated or cached mask based on context."
6505,SimpleRNNCell,tensorflow/tensorflow/python/keras/layers/recurrent.py,1221,class,"Cell class for SimpleRNN.
See [the Keras RNN API guide](https://www.tensorflow.org/guide/keras/rnn)
for details about the usage of RNN API.
This class processes one step within the whole time sequence input, whereas
`tf.keras.layer.SimpleRNN` processes the whole sequence.
Arguments:
units: Positive integer, dimensionality of the output space.
activation: Activation function to use.
Default: hyperbolic tangent (`tanh`).
If you pass `None`, no activation is applied
(ie. ""linear"" activation: `a(x) = x`).
use_bias: Boolean, (default `True`), whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix,
used for the linear transformation of the inputs. Default:
`glorot_uniform`.
recurrent_initializer: Initializer for the `recurrent_kernel`
weights matrix, used for the linear transformation of the recurrent state.
Default: `orthogonal`.
bias_initializer: Initializer for the bias vector. Default: `zeros`.
kernel_regularizer: Regularizer function applied to the `kernel` weights
matrix. Default: `None`.
recurrent_regularizer: Regularizer function applied to the
`recurrent_kernel` weights matrix. Default: `None`.
bias_regularizer: Regularizer function applied to the bias vector. Default:
`None`.
kernel_constraint: Constraint function applied to the `kernel` weights
matrix. Default: `None`.
recurrent_constraint: Constraint function applied to the `recurrent_kernel`
weights matrix. Default: `None`.
bias_constraint: Constraint function applied to the bias vector. Default:
`None`.
dropout: Float between 0 and 1. Fraction of the units to drop for the linear
transformation of the inputs. Default: 0.
recurrent_dropout: Float between 0 and 1. Fraction of the units to drop for
the linear transformation of the recurrent state. Default: 0.
Call arguments:
inputs: A 2D tensor, with shape of `[batch, feature]`.
states: A 2D tensor with shape of `[batch, units]`, which is the state from
the previous time step. For timestep 0, the initial state provided by user
will be feed to cell.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. Only relevant when `dropout` or
`recurrent_dropout` is used.
Examples:
```python
inputs = np.random.random([32, 10, 8]).astype(np.float32)
rnn = tf.keras.layers.RNN(tf.keras.layers.SimpleRNNCell(4))
output = rnn(inputs) # The output has shape `[32, 4]`.
rnn = tf.keras.layers.RNN(
tf.keras.layers.SimpleRNNCell(4),
return_sequences=True,
return_state=True)
# whole_sequence_output has shape `[32, 10, 4]`.
# final_state has shape `[32, 4]`.
whole_sequence_output, final_state = rnn(inputs)
```"
6506,build,tensorflow/tensorflow/python/keras/layers/recurrent.py,1333,method,
6507,call,tensorflow/tensorflow/python/keras/layers/recurrent.py,1361,method,
6508,get_initial_state,tensorflow/tensorflow/python/keras/layers/recurrent.py,1383,method,
6509,get_config,tensorflow/tensorflow/python/keras/layers/recurrent.py,1386,method,
6510,SimpleRNN,tensorflow/tensorflow/python/keras/layers/recurrent.py,1423,class,"Fully-connected RNN where the output is to be fed back to input.
See [the Keras RNN API guide](https://www.tensorflow.org/guide/keras/rnn)
for details about the usage of RNN API.
Arguments:
units: Positive integer, dimensionality of the output space.
activation: Activation function to use.
Default: hyperbolic tangent (`tanh`).
If you pass None, no activation is applied
(ie. ""linear"" activation: `a(x) = x`).
use_bias: Boolean, (default `True`), whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix,
used for the linear transformation of the inputs. Default:
`glorot_uniform`.
recurrent_initializer: Initializer for the `recurrent_kernel`
weights matrix, used for the linear transformation of the recurrent state.
Default: `orthogonal`.
bias_initializer: Initializer for the bias vector. Default: `zeros`.
kernel_regularizer: Regularizer function applied to the `kernel` weights
matrix. Default: `None`.
recurrent_regularizer: Regularizer function applied to the
`recurrent_kernel` weights matrix. Default: `None`.
bias_regularizer: Regularizer function applied to the bias vector. Default:
`None`.
activity_regularizer: Regularizer function applied to the output of the
layer (its ""activation""). Default: `None`.
kernel_constraint: Constraint function applied to the `kernel` weights
matrix. Default: `None`.
recurrent_constraint: Constraint function applied to the `recurrent_kernel`
weights matrix. Default: `None`.
bias_constraint: Constraint function applied to the bias vector. Default:
`None`.
dropout: Float between 0 and 1.
Fraction of the units to drop for the linear transformation of the inputs.
Default: 0.
recurrent_dropout: Float between 0 and 1.
Fraction of the units to drop for the linear transformation of the
recurrent state. Default: 0.
return_sequences: Boolean. Whether to return the last output
in the output sequence, or the full sequence. Default: `False`.
return_state: Boolean. Whether to return the last state
in addition to the output. Default: `False`
go_backwards: Boolean (default False).
If True, process the input sequence backwards and return the
reversed sequence.
stateful: Boolean (default False). If True, the last state
for each sample at index i in a batch will be used as initial
state for the sample of index i in the following batch.
unroll: Boolean (default False).
If True, the network will be unrolled,
else a symbolic loop will be used.
Unrolling can speed-up a RNN,
although it tends to be more memory-intensive.
Unrolling is only suitable for short sequences.
Call arguments:
inputs: A 3D tensor, with shape `[batch, timesteps, feature]`.
mask: Binary tensor of shape `[batch, timesteps]` indicating whether
a given timestep should be masked.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. This argument is passed to the cell
when calling it. This is only relevant if `dropout` or
`recurrent_dropout` is used.
initial_state: List of initial state tensors to be passed to the first
call of the cell.
Examples:
```python
inputs = np.random.random([32, 10, 8]).astype(np.float32)
simple_rnn = tf.keras.layers.SimpleRNN(4)
output = simple_rnn(inputs) # The output has shape `[32, 4]`.
simple_rnn = tf.keras.layers.SimpleRNN(
4, return_sequences=True, return_state=True)
# whole_sequence_output has shape `[32, 10, 4]`.
# final_state has shape `[32, 4]`.
whole_sequence_output, final_state = simple_rnn(inputs)
```"
6511,call,tensorflow/tensorflow/python/keras/layers/recurrent.py,1569,method,
6512,units,tensorflow/tensorflow/python/keras/layers/recurrent.py,1575,method,
6513,activation,tensorflow/tensorflow/python/keras/layers/recurrent.py,1579,method,
6514,use_bias,tensorflow/tensorflow/python/keras/layers/recurrent.py,1583,method,
6515,kernel_initializer,tensorflow/tensorflow/python/keras/layers/recurrent.py,1587,method,
6516,recurrent_initializer,tensorflow/tensorflow/python/keras/layers/recurrent.py,1591,method,
6517,bias_initializer,tensorflow/tensorflow/python/keras/layers/recurrent.py,1595,method,
6518,kernel_regularizer,tensorflow/tensorflow/python/keras/layers/recurrent.py,1599,method,
6519,recurrent_regularizer,tensorflow/tensorflow/python/keras/layers/recurrent.py,1603,method,
6520,bias_regularizer,tensorflow/tensorflow/python/keras/layers/recurrent.py,1607,method,
6521,kernel_constraint,tensorflow/tensorflow/python/keras/layers/recurrent.py,1611,method,
6522,recurrent_constraint,tensorflow/tensorflow/python/keras/layers/recurrent.py,1615,method,
6523,bias_constraint,tensorflow/tensorflow/python/keras/layers/recurrent.py,1619,method,
6524,dropout,tensorflow/tensorflow/python/keras/layers/recurrent.py,1623,method,
6525,recurrent_dropout,tensorflow/tensorflow/python/keras/layers/recurrent.py,1627,method,
6526,get_config,tensorflow/tensorflow/python/keras/layers/recurrent.py,1630,method,
6527,from_config,tensorflow/tensorflow/python/keras/layers/recurrent.py,1669,method,
6528,GRUCell,tensorflow/tensorflow/python/keras/layers/recurrent.py,1676,class,"Cell class for the GRU layer.
Arguments:
units: Positive integer, dimensionality of the output space.
activation: Activation function to use.
Default: hyperbolic tangent (`tanh`).
If you pass None, no activation is applied
(ie. ""linear"" activation: `a(x) = x`).
recurrent_activation: Activation function to use
for the recurrent step.
Default: hard sigmoid (`hard_sigmoid`).
If you pass `None`, no activation is applied
(ie. ""linear"" activation: `a(x) = x`).
use_bias: Boolean, whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix,
used for the linear transformation of the inputs.
recurrent_initializer: Initializer for the `recurrent_kernel`
weights matrix,
used for the linear transformation of the recurrent state.
bias_initializer: Initializer for the bias vector.
kernel_regularizer: Regularizer function applied to
the `kernel` weights matrix.
recurrent_regularizer: Regularizer function applied to
the `recurrent_kernel` weights matrix.
bias_regularizer: Regularizer function applied to the bias vector.
kernel_constraint: Constraint function applied to
the `kernel` weights matrix.
recurrent_constraint: Constraint function applied to
the `recurrent_kernel` weights matrix.
bias_constraint: Constraint function applied to the bias vector.
dropout: Float between 0 and 1.
Fraction of the units to drop for the linear transformation of the inputs.
recurrent_dropout: Float between 0 and 1.
Fraction of the units to drop for
the linear transformation of the recurrent state.
implementation: Implementation mode, either 1 or 2.
Mode 1 will structure its operations as a larger number of
smaller dot products and additions, whereas mode 2 will
batch them into fewer, larger operations. These modes will
have different performance profiles on different hardware and
for different applications.
reset_after: GRU convention (whether to apply reset gate after or
before matrix multiplication). False = ""before"" (default),
True = ""after"" (CuDNN compatible).
Call arguments:
inputs: A 2D tensor.
states: List of state tensors corresponding to the previous timestep.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. Only relevant when `dropout` or
`recurrent_dropout` is used."
6529,build,tensorflow/tensorflow/python/keras/layers/recurrent.py,1784,method,
6530,call,tensorflow/tensorflow/python/keras/layers/recurrent.py,1821,method,
6531,get_config,tensorflow/tensorflow/python/keras/layers/recurrent.py,1923,method,
6532,get_initial_state,tensorflow/tensorflow/python/keras/layers/recurrent.py,1951,method,
6533,GRU,tensorflow/tensorflow/python/keras/layers/recurrent.py,1956,class,"Gated Recurrent Unit - Cho et al. 2014.
There are two variants. The default one is based on 1406.1078v3 and
has reset gate applied to hidden state before matrix multiplication. The
other one is based on original 1406.1078v1 and has the order reversed.
The second variant is compatible with CuDNNGRU (GPU-only) and allows
inference on CPU. Thus it has separate biases for `kernel` and
`recurrent_kernel`. Use `'reset_after'=True` and
`recurrent_activation='sigmoid'`.
Arguments:
units: Positive integer, dimensionality of the output space.
activation: Activation function to use.
Default: hyperbolic tangent (`tanh`).
If you pass `None`, no activation is applied
(ie. ""linear"" activation: `a(x) = x`).
recurrent_activation: Activation function to use
for the recurrent step.
Default: hard sigmoid (`hard_sigmoid`).
If you pass `None`, no activation is applied
(ie. ""linear"" activation: `a(x) = x`).
use_bias: Boolean, whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix,
used for the linear transformation of the inputs.
recurrent_initializer: Initializer for the `recurrent_kernel`
weights matrix, used for the linear transformation of the recurrent state.
bias_initializer: Initializer for the bias vector.
kernel_regularizer: Regularizer function applied to
the `kernel` weights matrix.
recurrent_regularizer: Regularizer function applied to
the `recurrent_kernel` weights matrix.
bias_regularizer: Regularizer function applied to the bias vector.
activity_regularizer: Regularizer function applied to
the output of the layer (its ""activation"")..
kernel_constraint: Constraint function applied to
the `kernel` weights matrix.
recurrent_constraint: Constraint function applied to
the `recurrent_kernel` weights matrix.
bias_constraint: Constraint function applied to the bias vector.
dropout: Float between 0 and 1.
Fraction of the units to drop for
the linear transformation of the inputs.
recurrent_dropout: Float between 0 and 1.
Fraction of the units to drop for
the linear transformation of the recurrent state.
implementation: Implementation mode, either 1 or 2.
Mode 1 will structure its operations as a larger number of
smaller dot products and additions, whereas mode 2 will
batch them into fewer, larger operations. These modes will
have different performance profiles on different hardware and
for different applications.
return_sequences: Boolean. Whether to return the last output
in the output sequence, or the full sequence.
return_state: Boolean. Whether to return the last state
in addition to the output.
go_backwards: Boolean (default False).
If True, process the input sequence backwards and return the
reversed sequence.
stateful: Boolean (default False). If True, the last state
for each sample at index i in a batch will be used as initial
state for the sample of index i in the following batch.
unroll: Boolean (default False).
If True, the network will be unrolled,
else a symbolic loop will be used.
Unrolling can speed-up a RNN,
although it tends to be more memory-intensive.
Unrolling is only suitable for short sequences.
time_major: The shape format of the `inputs` and `outputs` tensors.
If True, the inputs and outputs will be in shape
`(timesteps, batch, ...)`, whereas in the False case, it will be
`(batch, timesteps, ...)`. Using `time_major = True` is a bit more
efficient because it avoids transposes at the beginning and end of the
RNN calculation. However, most TensorFlow data is batch-major, so by
default this function accepts input and emits output in batch-major
form.
reset_after: GRU convention (whether to apply reset gate after or
before matrix multiplication). False = ""before"" (default),
True = ""after"" (CuDNN compatible).
Call arguments:
inputs: A 3D tensor.
mask: Binary tensor of shape `(samples, timesteps)` indicating whether
a given timestep should be masked.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. This argument is passed to the cell
when calling it. This is only relevant if `dropout` or
`recurrent_dropout` is used.
initial_state: List of initial state tensors to be passed to the first
call of the cell."
6534,call,tensorflow/tensorflow/python/keras/layers/recurrent.py,2115,method,
6535,units,tensorflow/tensorflow/python/keras/layers/recurrent.py,2121,method,
6536,activation,tensorflow/tensorflow/python/keras/layers/recurrent.py,2125,method,
6537,recurrent_activation,tensorflow/tensorflow/python/keras/layers/recurrent.py,2129,method,
6538,use_bias,tensorflow/tensorflow/python/keras/layers/recurrent.py,2133,method,
6539,kernel_initializer,tensorflow/tensorflow/python/keras/layers/recurrent.py,2137,method,
6540,recurrent_initializer,tensorflow/tensorflow/python/keras/layers/recurrent.py,2141,method,
6541,bias_initializer,tensorflow/tensorflow/python/keras/layers/recurrent.py,2145,method,
6542,kernel_regularizer,tensorflow/tensorflow/python/keras/layers/recurrent.py,2149,method,
6543,recurrent_regularizer,tensorflow/tensorflow/python/keras/layers/recurrent.py,2153,method,
6544,bias_regularizer,tensorflow/tensorflow/python/keras/layers/recurrent.py,2157,method,
6545,kernel_constraint,tensorflow/tensorflow/python/keras/layers/recurrent.py,2161,method,
6546,recurrent_constraint,tensorflow/tensorflow/python/keras/layers/recurrent.py,2165,method,
6547,bias_constraint,tensorflow/tensorflow/python/keras/layers/recurrent.py,2169,method,
6548,dropout,tensorflow/tensorflow/python/keras/layers/recurrent.py,2173,method,
6549,recurrent_dropout,tensorflow/tensorflow/python/keras/layers/recurrent.py,2177,method,
6550,implementation,tensorflow/tensorflow/python/keras/layers/recurrent.py,2181,method,
6551,reset_after,tensorflow/tensorflow/python/keras/layers/recurrent.py,2185,method,
6552,get_config,tensorflow/tensorflow/python/keras/layers/recurrent.py,2188,method,
6553,from_config,tensorflow/tensorflow/python/keras/layers/recurrent.py,2233,method,
6554,LSTMCell,tensorflow/tensorflow/python/keras/layers/recurrent.py,2240,class,"Cell class for the LSTM layer.
Arguments:
units: Positive integer, dimensionality of the output space.
activation: Activation function to use.
Default: hyperbolic tangent (`tanh`).
If you pass `None`, no activation is applied
(ie. ""linear"" activation: `a(x) = x`).
recurrent_activation: Activation function to use
for the recurrent step.
Default: hard sigmoid (`hard_sigmoid`).
If you pass `None`, no activation is applied
(ie. ""linear"" activation: `a(x) = x`).
use_bias: Boolean, whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix,
used for the linear transformation of the inputs.
recurrent_initializer: Initializer for the `recurrent_kernel`
weights matrix,
used for the linear transformation of the recurrent state.
bias_initializer: Initializer for the bias vector.
unit_forget_bias: Boolean.
If True, add 1 to the bias of the forget gate at initialization.
Setting it to true will also force `bias_initializer=""zeros""`.
This is recommended in [Jozefowicz et al., 2015](
http://www.jmlr.org/proceedings/papers/v37/jozefowicz15.pdf)
kernel_regularizer: Regularizer function applied to
the `kernel` weights matrix.
recurrent_regularizer: Regularizer function applied to
the `recurrent_kernel` weights matrix.
bias_regularizer: Regularizer function applied to the bias vector.
kernel_constraint: Constraint function applied to
the `kernel` weights matrix.
recurrent_constraint: Constraint function applied to
the `recurrent_kernel` weights matrix.
bias_constraint: Constraint function applied to the bias vector.
dropout: Float between 0 and 1.
Fraction of the units to drop for
the linear transformation of the inputs.
recurrent_dropout: Float between 0 and 1.
Fraction of the units to drop for
the linear transformation of the recurrent state.
implementation: Implementation mode, either 1 or 2.
Mode 1 will structure its operations as a larger number of
smaller dot products and additions, whereas mode 2 will
batch them into fewer, larger operations. These modes will
have different performance profiles on different hardware and
for different applications.
Call arguments:
inputs: A 2D tensor.
states: List of state tensors corresponding to the previous timestep.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. Only relevant when `dropout` or
`recurrent_dropout` is used."
6555,build,tensorflow/tensorflow/python/keras/layers/recurrent.py,2357,method,
6556,call,tensorflow/tensorflow/python/keras/layers/recurrent.py,2420,method,
6557,get_config,tensorflow/tensorflow/python/keras/layers/recurrent.py,2480,method,
6558,get_initial_state,tensorflow/tensorflow/python/keras/layers/recurrent.py,2521,method,
6559,bias_initializer,tensorflow/tensorflow/python/keras/layers/recurrent.py,2378,method,
6560,PeepholeLSTMCell,tensorflow/tensorflow/python/keras/layers/recurrent.py,2527,class,"Equivalent to LSTMCell class but adds peephole connections.
Peephole connections allow the gates to utilize the previous internal state as
well as the previous hidden state (which is what LSTMCell is limited to).
This allows PeepholeLSTMCell to better learn precise timings over LSTMCell.
From [Gers et al., 2002](
http://www.jmlr.org/papers/volume3/gers02a/gers02a.pdf):
""We find that LSTM augmented by 'peephole connections' from its internal
cells to its multiplicative gates can learn the fine distinction between
sequences of spikes spaced either 50 or 49 time steps apart without the help
of any short training exemplars.""
The peephole implementation is based on:
[Sak et al., 2014](https://research.google.com/pubs/archive/43905.pdf)
Example:
```python
# Create 2 PeepholeLSTMCells
peephole_lstm_cells = [PeepholeLSTMCell(size) for size in [128, 256]]
# Create a layer composed sequentially of the peephole LSTM cells.
layer = RNN(peephole_lstm_cells)
input = keras.Input((timesteps, input_dim))
output = layer(input)
```"
6561,build,tensorflow/tensorflow/python/keras/layers/recurrent.py,2599,method,
6562,LSTM,tensorflow/tensorflow/python/keras/layers/recurrent.py,2645,class,"Long Short-Term Memory layer - Hochreiter 1997.
Note that this cell is not optimized for performance on GPU. Please use
`tf.compat.v1.keras.layers.CuDNNLSTM` for better performance on GPU.
Arguments:
units: Positive integer, dimensionality of the output space.
activation: Activation function to use.
Default: hyperbolic tangent (`tanh`).
If you pass `None`, no activation is applied
(ie. ""linear"" activation: `a(x) = x`).
recurrent_activation: Activation function to use
for the recurrent step.
Default: hard sigmoid (`hard_sigmoid`).
If you pass `None`, no activation is applied
(ie. ""linear"" activation: `a(x) = x`).
use_bias: Boolean, whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix,
used for the linear transformation of the inputs..
recurrent_initializer: Initializer for the `recurrent_kernel`
weights matrix,
used for the linear transformation of the recurrent state.
bias_initializer: Initializer for the bias vector.
unit_forget_bias: Boolean.
If True, add 1 to the bias of the forget gate at initialization.
Setting it to true will also force `bias_initializer=""zeros""`.
This is recommended in [Jozefowicz et al., 2015](
http://www.jmlr.org/proceedings/papers/v37/jozefowicz15.pdf).
kernel_regularizer: Regularizer function applied to
the `kernel` weights matrix.
recurrent_regularizer: Regularizer function applied to
the `recurrent_kernel` weights matrix.
bias_regularizer: Regularizer function applied to the bias vector.
activity_regularizer: Regularizer function applied to
the output of the layer (its ""activation"").
kernel_constraint: Constraint function applied to
the `kernel` weights matrix.
recurrent_constraint: Constraint function applied to
the `recurrent_kernel` weights matrix.
bias_constraint: Constraint function applied to the bias vector.
dropout: Float between 0 and 1.
Fraction of the units to drop for
the linear transformation of the inputs.
recurrent_dropout: Float between 0 and 1.
Fraction of the units to drop for
the linear transformation of the recurrent state.
implementation: Implementation mode, either 1 or 2.
Mode 1 will structure its operations as a larger number of
smaller dot products and additions, whereas mode 2 will
batch them into fewer, larger operations. These modes will
have different performance profiles on different hardware and
for different applications.
return_sequences: Boolean. Whether to return the last output.
in the output sequence, or the full sequence.
return_state: Boolean. Whether to return the last state
in addition to the output.
go_backwards: Boolean (default False).
If True, process the input sequence backwards and return the
reversed sequence.
stateful: Boolean (default False). If True, the last state
for each sample at index i in a batch will be used as initial
state for the sample of index i in the following batch.
unroll: Boolean (default False).
If True, the network will be unrolled,
else a symbolic loop will be used.
Unrolling can speed-up a RNN,
although it tends to be more memory-intensive.
Unrolling is only suitable for short sequences.
time_major: The shape format of the `inputs` and `outputs` tensors.
If True, the inputs and outputs will be in shape
`(timesteps, batch, ...)`, whereas in the False case, it will be
`(batch, timesteps, ...)`. Using `time_major = True` is a bit more
efficient because it avoids transposes at the beginning and end of the
RNN calculation. However, most TensorFlow data is batch-major, so by
default this function accepts input and emits output in batch-major
form.
Call arguments:
inputs: A 3D tensor.
mask: Binary tensor of shape `(samples, timesteps)` indicating whether
a given timestep should be masked.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. This argument is passed to the cell
when calling it. This is only relevant if `dropout` or
`recurrent_dropout` is used.
initial_state: List of initial state tensors to be passed to the first
call of the cell."
6563,call,tensorflow/tensorflow/python/keras/layers/recurrent.py,2801,method,
6564,units,tensorflow/tensorflow/python/keras/layers/recurrent.py,2807,method,
6565,activation,tensorflow/tensorflow/python/keras/layers/recurrent.py,2811,method,
6566,recurrent_activation,tensorflow/tensorflow/python/keras/layers/recurrent.py,2815,method,
6567,use_bias,tensorflow/tensorflow/python/keras/layers/recurrent.py,2819,method,
6568,kernel_initializer,tensorflow/tensorflow/python/keras/layers/recurrent.py,2823,method,
6569,recurrent_initializer,tensorflow/tensorflow/python/keras/layers/recurrent.py,2827,method,
6570,bias_initializer,tensorflow/tensorflow/python/keras/layers/recurrent.py,2831,method,
6571,unit_forget_bias,tensorflow/tensorflow/python/keras/layers/recurrent.py,2835,method,
6572,kernel_regularizer,tensorflow/tensorflow/python/keras/layers/recurrent.py,2839,method,
6573,recurrent_regularizer,tensorflow/tensorflow/python/keras/layers/recurrent.py,2843,method,
6574,bias_regularizer,tensorflow/tensorflow/python/keras/layers/recurrent.py,2847,method,
6575,kernel_constraint,tensorflow/tensorflow/python/keras/layers/recurrent.py,2851,method,
6576,recurrent_constraint,tensorflow/tensorflow/python/keras/layers/recurrent.py,2855,method,
6577,bias_constraint,tensorflow/tensorflow/python/keras/layers/recurrent.py,2859,method,
6578,dropout,tensorflow/tensorflow/python/keras/layers/recurrent.py,2863,method,
6579,recurrent_dropout,tensorflow/tensorflow/python/keras/layers/recurrent.py,2867,method,
6580,implementation,tensorflow/tensorflow/python/keras/layers/recurrent.py,2871,method,
6581,get_config,tensorflow/tensorflow/python/keras/layers/recurrent.py,2874,method,
6582,from_config,tensorflow/tensorflow/python/keras/layers/recurrent.py,2919,method,
6583,RNNCellWithConstants,tensorflow/tensorflow/python/keras/layers/recurrent_test.py,1700,class,
6584,build,tensorflow/tensorflow/python/keras/layers/recurrent_test.py,1708,method,
6585,call,tensorflow/tensorflow/python/keras/layers/recurrent_test.py,1723,method,
6586,get_config,tensorflow/tensorflow/python/keras/layers/recurrent_test.py,1732,method,
6587,Minimal2DRNNCell,tensorflow/tensorflow/python/keras/layers/recurrent_test.py,1738,class,"The minimal 2D RNN cell is a simple combination of 2 1-D RNN cell.
Both internal state and output have 2 dimensions and are orthogonal
between each other."
6588,build,tensorflow/tensorflow/python/keras/layers/recurrent_test.py,1752,method,
6589,call,tensorflow/tensorflow/python/keras/layers/recurrent_test.py,1767,method,
6590,PlusOneRNNCell,tensorflow/tensorflow/python/keras/layers/recurrent_test.py,1776,class,"Add one to the input and state.
This cell is used for testing state_size and output_size."
6591,build,tensorflow/tensorflow/python/keras/layers/recurrent_test.py,1785,method,
6592,call,tensorflow/tensorflow/python/keras/layers/recurrent_test.py,1788,method,
6593,NestedCell,tensorflow/tensorflow/python/keras/layers/recurrent_test.py,1792,class,
6594,build,tensorflow/tensorflow/python/keras/layers/recurrent_test.py,1808,method,
6595,call,tensorflow/tensorflow/python/keras/layers/recurrent_test.py,1824,method,
6596,GRUCell,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,69,class,"Cell class for the GRU layer.
See [the Keras RNN API guide](https://www.tensorflow.org/guide/keras/rnn)
for details about the usage of RNN API.
This class processes one step within the whole time sequence input, whereas
`tf.keras.layer.GRU` processes the whole sequence.
For example:
>>> inputs = tf.random.normal([32, 10, 8])
>>> rnn = tf.keras.layers.RNN(tf.keras.layers.GRUCell(4))
>>> output = rnn(inputs)
>>> print(output.shape)
(32, 4)
>>> rnn = tf.keras.layers.RNN(
... tf.keras.layers.GRUCell(4),
... return_sequences=True,
... return_state=True)
>>> whole_sequence_output, final_state = rnn(inputs)
>>> print(whole_sequence_output.shape)
(32, 10, 4)
>>> print(final_state.shape)
(32, 4)
Arguments:
units: Positive integer, dimensionality of the output space.
activation: Activation function to use. Default: hyperbolic tangent
(`tanh`). If you pass None, no activation is applied
(ie. ""linear"" activation: `a(x) = x`).
recurrent_activation: Activation function to use for the recurrent step.
Default: sigmoid (`sigmoid`). If you pass `None`, no activation is
applied (ie. ""linear"" activation: `a(x) = x`).
use_bias: Boolean, (default `True`), whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix,
used for the linear transformation of the inputs. Default:
`glorot_uniform`.
recurrent_initializer: Initializer for the `recurrent_kernel`
weights matrix, used for the linear transformation of the recurrent state.
Default: `orthogonal`.
bias_initializer: Initializer for the bias vector. Default: `zeros`.
kernel_regularizer: Regularizer function applied to the `kernel` weights
matrix. Default: `None`.
recurrent_regularizer: Regularizer function applied to the
`recurrent_kernel` weights matrix. Default: `None`.
bias_regularizer: Regularizer function applied to the bias vector. Default:
`None`.
kernel_constraint: Constraint function applied to the `kernel` weights
matrix. Default: `None`.
recurrent_constraint: Constraint function applied to the `recurrent_kernel`
weights matrix. Default: `None`.
bias_constraint: Constraint function applied to the bias vector. Default:
`None`.
dropout: Float between 0 and 1. Fraction of the units to drop for the
linear transformation of the inputs. Default: 0.
recurrent_dropout: Float between 0 and 1. Fraction of the units to drop for
the linear transformation of the recurrent state. Default: 0.
implementation: Implementation mode, either 1 or 2.
Mode 1 will structure its operations as a larger number of
smaller dot products and additions, whereas mode 2 (default) will
batch them into fewer, larger operations. These modes will
have different performance profiles on different hardware and
for different applications. Default: 2.
reset_after: GRU convention (whether to apply reset gate after or
before matrix multiplication). False = ""before"",
True = ""after"" (default and CuDNN compatible).
Call arguments:
inputs: A 2D tensor, with shape of `[batch, feature]`.
states: A 2D tensor with shape of `[batch, units]`, which is the state from
the previous time step. For timestep 0, the initial state provided by user
will be feed to cell.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. Only relevant when `dropout` or
`recurrent_dropout` is used."
6597,GRU,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,188,class,"Gated Recurrent Unit - Cho et al. 2014.
See [the Keras RNN API guide](https://www.tensorflow.org/guide/keras/rnn)
for details about the usage of RNN API.
Based on available runtime hardware and constraints, this layer
will choose different implementations (cuDNN-based or pure-TensorFlow)
to maximize the performance. If a GPU is available and all
the arguments to the layer meet the requirement of the CuDNN kernel
(see below for details), the layer will use a fast cuDNN implementation.
The requirements to use the cuDNN implementation are:
1. `activation` == `tanh`
2. `recurrent_activation` == `sigmoid`
3. `recurrent_dropout` == 0
4. `unroll` is `False`
5. `use_bias` is `True`
6. `reset_after` is `True`
7. Inputs, if use masking, are strictly right-padded.
8. Eager execution is enabled in the outermost context.
There are two variants of the GRU implementation. The default one is based on
[v3](https://arxiv.org/abs/1406.1078v3) and has reset gate applied to hidden
state before matrix multiplication. The other one is based on
[original](https://arxiv.org/abs/1406.1078v1) and has the order reversed.
The second variant is compatible with CuDNNGRU (GPU-only) and allows
inference on CPU. Thus it has separate biases for `kernel` and
`recurrent_kernel`. To use this variant, set `'reset_after'=True` and
`recurrent_activation='sigmoid'`.
For example:
>>> inputs = tf.random.normal([32, 10, 8])
>>> gru = tf.keras.layers.GRU(4)
>>> output = gru(inputs)
>>> print(output.shape)
(32, 4)
>>> gru = tf.keras.layers.GRU(4, return_sequences=True, return_state=True)
>>> whole_sequence_output, final_state = gru(inputs)
>>> print(whole_sequence_output.shape)
(32, 10, 4)
>>> print(final_state.shape)
(32, 4)
Arguments:
units: Positive integer, dimensionality of the output space.
activation: Activation function to use.
Default: hyperbolic tangent (`tanh`).
If you pass `None`, no activation is applied
(ie. ""linear"" activation: `a(x) = x`).
recurrent_activation: Activation function to use
for the recurrent step.
Default: sigmoid (`sigmoid`).
If you pass `None`, no activation is applied
(ie. ""linear"" activation: `a(x) = x`).
use_bias: Boolean, (default `True`), whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix,
used for the linear transformation of the inputs. Default:
`glorot_uniform`.
recurrent_initializer: Initializer for the `recurrent_kernel`
weights matrix, used for the linear transformation of the recurrent
state. Default: `orthogonal`.
bias_initializer: Initializer for the bias vector. Default: `zeros`.
kernel_regularizer: Regularizer function applied to the `kernel` weights
matrix. Default: `None`.
recurrent_regularizer: Regularizer function applied to the
`recurrent_kernel` weights matrix. Default: `None`.
bias_regularizer: Regularizer function applied to the bias vector. Default:
`None`.
activity_regularizer: Regularizer function applied to the output of the
layer (its ""activation""). Default: `None`.
kernel_constraint: Constraint function applied to the `kernel` weights
matrix. Default: `None`.
recurrent_constraint: Constraint function applied to the `recurrent_kernel`
weights matrix. Default: `None`.
bias_constraint: Constraint function applied to the bias vector. Default:
`None`.
dropout: Float between 0 and 1. Fraction of the units to drop for the linear
transformation of the inputs. Default: 0.
recurrent_dropout: Float between 0 and 1. Fraction of the units to drop for
the linear transformation of the recurrent state. Default: 0.
implementation: Implementation mode, either 1 or 2.
Mode 1 will structure its operations as a larger number of
smaller dot products and additions, whereas mode 2 will
batch them into fewer, larger operations. These modes will
have different performance profiles on different hardware and
for different applications. Default: 2.
return_sequences: Boolean. Whether to return the last output
in the output sequence, or the full sequence. Default: `False`.
return_state: Boolean. Whether to return the last state in addition to the
output. Default: `False`.
go_backwards: Boolean (default `False`).
If True, process the input sequence backwards and return the
reversed sequence.
stateful: Boolean (default False). If True, the last state
for each sample at index i in a batch will be used as initial
state for the sample of index i in the following batch.
unroll: Boolean (default False).
If True, the network will be unrolled,
else a symbolic loop will be used.
Unrolling can speed-up a RNN,
although it tends to be more memory-intensive.
Unrolling is only suitable for short sequences.
time_major: The shape format of the `inputs` and `outputs` tensors.
If True, the inputs and outputs will be in shape
`[timesteps, batch, feature]`, whereas in the False case, it will be
`[batch, timesteps, feature]`. Using `time_major = True` is a bit more
efficient because it avoids transposes at the beginning and end of the
RNN calculation. However, most TensorFlow data is batch-major, so by
default this function accepts input and emits output in batch-major
form.
reset_after: GRU convention (whether to apply reset gate after or
before matrix multiplication). False = ""before"",
True = ""after"" (default and CuDNN compatible).
Call arguments:
inputs: A 3D tensor, with shape `[batch, timesteps, feature]`.
mask: Binary tensor of shape `[samples, timesteps]` indicating whether
a given timestep should be masked (optional, defaults to `None`).
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. This argument is passed to the cell
when calling it. This is only relevant if `dropout` or
`recurrent_dropout` is used (optional, defaults to `None`).
initial_state: List of initial state tensors to be passed to the first
call of the cell (optional, defaults to `None` which causes creation
of zero-filled initial state tensors)."
6598,build,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,389,method,
6599,call,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,402,method,
6600,step,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,424,method,
6601,standard_gru,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,508,function,"GRU with standard kernel implementation.
This implementation can be run on all types of hardware.
This implementation lifts out all the layer weights and make them function
parameters. It has same number of tensor input params as the CuDNN
counterpart. The RNN step logic has been simplified, eg dropout and mask is
removed since CuDNN implementation does not support that.
Arguments:
inputs: Input tensor of GRU layer.
init_h: Initial state tensor for the cell output.
kernel: Weights for cell kernel.
recurrent_kernel: Weights for cell recurrent kernel.
bias: Weights for cell kernel bias and recurrent bias. The bias contains the
combined input_bias and recurrent_bias.
mask: Binary tensor of shape `(samples, timesteps)` indicating whether
a given timestep should be masked.
time_major: Boolean, whether the inputs are in the format of
[time, batch, feature] or [batch, time, feature].
go_backwards: Boolean (default False). If True, process the input sequence
backwards and return the reversed sequence.
sequence_lengths: The lengths of all sequences coming from a variable length
input, such as ragged tensors. If the input has a fixed timestep size,
this should be None.
zero_output_for_mask: Boolean, whether to output zero for masked timestep.
Returns:
last_output: output tensor for the last timestep, which has shape
[batch, units].
outputs: output tensor for all timesteps, which has shape
[batch, time, units].
state_0: the cell output, which has same shape as init_h.
runtime: constant string tensor which indicate real runtime hardware. This
value is for testing purpose and should be used by user."
6602,gpu_gru,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,590,function,GRU with CuDNN implementation which is only available for GPU.
6603,gru_with_backend_selection,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,676,function,"Call the GRU with optimized backend kernel selection.
Under the hood, this function will create two TF function, one with the most
generic kernel and can run on all device condition, and the second one with
CuDNN specific kernel, which can only run on GPU.
The first function will be called with normal_lstm_params, while the second
function is not called, but only registered in the graph. The Grappler will
do the proper graph rewrite and swap the optimized TF function based on the
device placement.
Args:
inputs: Input tensor of GRU layer.
init_h: Initial state tensor for the cell output.
kernel: Weights for cell kernel.
recurrent_kernel: Weights for cell recurrent kernel.
bias: Weights for cell kernel bias and recurrent bias. Only recurrent bias
is used in this case.
mask: Boolean tensor for mask out the steps within sequence.
time_major: Boolean, whether the inputs are in the format of
[time, batch, feature] or [batch, time, feature].
go_backwards: Boolean (default False). If True, process the input sequence
backwards and return the reversed sequence.
sequence_lengths: The lengths of all sequences coming from a variable length
input, such as ragged tensors. If the input has a fixed timestep size,
this should be None.
zero_output_for_mask: Boolean, whether to output zero for masked timestep.
Returns:
List of output tensors, same as standard_gru."
6604,LSTMCell,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,791,class,"Cell class for the LSTM layer.
See [the Keras RNN API guide](https://www.tensorflow.org/guide/keras/rnn)
for details about the usage of RNN API.
This class processes one step within the whole time sequence input, whereas
`tf.keras.layer.LSTM` processes the whole sequence.
For example:
>>> inputs = tf.random.normal([32, 10, 8])
>>> rnn = tf.keras.layers.RNN(tf.keras.layers.LSTMCell(4))
>>> output = rnn(inputs)
>>> print(output.shape)
(32, 4)
>>> rnn = tf.keras.layers.RNN(
... tf.keras.layers.LSTMCell(4),
... return_sequences=True,
... return_state=True)
>>> whole_seq_output, final_memory_state, final_carry_state = rnn(inputs)
>>> print(whole_seq_output.shape)
(32, 10, 4)
>>> print(final_memory_state.shape)
(32, 4)
>>> print(final_carry_state.shape)
(32, 4)
Arguments:
units: Positive integer, dimensionality of the output space.
activation: Activation function to use. Default: hyperbolic tangent
(`tanh`). If you pass `None`, no activation is applied (ie. ""linear""
activation: `a(x) = x`).
recurrent_activation: Activation function to use for the recurrent step.
Default: sigmoid (`sigmoid`). If you pass `None`, no activation is applied
(ie. ""linear"" activation: `a(x) = x`).
use_bias: Boolean, (default `True`), whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix, used for
the linear transformation of the inputs. Default: `glorot_uniform`.
recurrent_initializer: Initializer for the `recurrent_kernel` weights
matrix, used for the linear transformation of the recurrent state.
Default: `orthogonal`.
bias_initializer: Initializer for the bias vector. Default: `zeros`.
unit_forget_bias: Boolean (default `True`). If True, add 1 to the bias of
the forget gate at initialization. Setting it to true will also force
`bias_initializer=""zeros""`. This is recommended in [Jozefowicz et
al.](http://www.jmlr.org/proceedings/papers/v37/jozefowicz15.pdf)
kernel_regularizer: Regularizer function applied to the `kernel` weights
matrix. Default: `None`.
recurrent_regularizer: Regularizer function applied to
the `recurrent_kernel` weights matrix. Default: `None`.
bias_regularizer: Regularizer function applied to the bias vector. Default:
`None`.
kernel_constraint: Constraint function applied to the `kernel` weights
matrix. Default: `None`.
recurrent_constraint: Constraint function applied to the `recurrent_kernel`
weights matrix. Default: `None`.
bias_constraint: Constraint function applied to the bias vector. Default:
`None`.
dropout: Float between 0 and 1. Fraction of the units to drop for the linear
transformation of the inputs. Default: 0.
recurrent_dropout: Float between 0 and 1. Fraction of the units to drop for
the linear transformation of the recurrent state. Default: 0.
implementation: Implementation mode, either 1 or 2.
Mode 1 will structure its operations as a larger number of smaller dot
products and additions, whereas mode 2 (default) will batch them into
fewer, larger operations. These modes will have different performance
profiles on different hardware and for different applications. Default: 2.
Call arguments:
inputs: A 2D tensor, with shape of `[batch, feature]`.
states: List of 2 tensors that corresponding to the cell's units. Both of
them have shape `[batch, units]`, the first tensor is the memory state
from previous time step, the second tensor is the carry state from
previous time step. For timestep 0, the initial state provided by user
will be feed to cell.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. Only relevant when `dropout` or
`recurrent_dropout` is used."
6605,LSTM,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,913,class,"Long Short-Term Memory layer - Hochreiter 1997.
See [the Keras RNN API guide](https://www.tensorflow.org/guide/keras/rnn)
for details about the usage of RNN API.
Based on available runtime hardware and constraints, this layer
will choose different implementations (cuDNN-based or pure-TensorFlow)
to maximize the performance. If a GPU is available and all
the arguments to the layer meet the requirement of the CuDNN kernel
(see below for details), the layer will use a fast cuDNN implementation.
The requirements to use the cuDNN implementation are:
1. `activation` == `tanh`
2. `recurrent_activation` == `sigmoid`
3. `recurrent_dropout` == 0
4. `unroll` is `False`
5. `use_bias` is `True`
6. Inputs, if use masking, are strictly right-padded.
7. Eager execution is enabled in the outermost context.
For example:
>>> inputs = tf.random.normal([32, 10, 8])
>>> lstm = tf.keras.layers.LSTM(4)
>>> output = lstm(inputs)
>>> print(output.shape)
(32, 4)
>>> lstm = tf.keras.layers.LSTM(4, return_sequences=True, return_state=True)
>>> whole_seq_output, final_memory_state, final_carry_state = lstm(inputs)
>>> print(whole_seq_output.shape)
(32, 10, 4)
>>> print(final_memory_state.shape)
(32, 4)
>>> print(final_carry_state.shape)
(32, 4)
Arguments:
units: Positive integer, dimensionality of the output space.
activation: Activation function to use.
Default: hyperbolic tangent (`tanh`). If you pass `None`, no activation
is applied (ie. ""linear"" activation: `a(x) = x`).
recurrent_activation: Activation function to use for the recurrent step.
Default: sigmoid (`sigmoid`). If you pass `None`, no activation is
applied (ie. ""linear"" activation: `a(x) = x`).
use_bias: Boolean (default `True`), whether the layer uses a bias vector.
kernel_initializer: Initializer for the `kernel` weights matrix, used for
the linear transformation of the inputs. Default: `glorot_uniform`.
recurrent_initializer: Initializer for the `recurrent_kernel` weights
matrix, used for the linear transformation of the recurrent state.
Default: `orthogonal`.
bias_initializer: Initializer for the bias vector. Default: `zeros`.
unit_forget_bias: Boolean (default `True`). If True, add 1 to the bias of
the forget gate at initialization. Setting it to true will also force
`bias_initializer=""zeros""`. This is recommended in [Jozefowicz et
al.](http://www.jmlr.org/proceedings/papers/v37/jozefowicz15.pdf).
kernel_regularizer: Regularizer function applied to the `kernel` weights
matrix. Default: `None`.
recurrent_regularizer: Regularizer function applied to the
`recurrent_kernel` weights matrix. Default: `None`.
bias_regularizer: Regularizer function applied to the bias vector. Default:
`None`.
activity_regularizer: Regularizer function applied to the output of the
layer (its ""activation""). Default: `None`.
kernel_constraint: Constraint function applied to the `kernel` weights
matrix. Default: `None`.
recurrent_constraint: Constraint function applied to the `recurrent_kernel`
weights matrix. Default: `None`.
bias_constraint: Constraint function applied to the bias vector. Default:
`None`.
dropout: Float between 0 and 1. Fraction of the units to drop for the linear
transformation of the inputs. Default: 0.
recurrent_dropout: Float between 0 and 1. Fraction of the units to drop for
the linear transformation of the recurrent state. Default: 0.
implementation: Implementation mode, either 1 or 2. Mode 1 will structure
its operations as a larger number of smaller dot products and additions,
whereas mode 2 will batch them into fewer, larger operations. These modes
will have different performance profiles on different hardware and for
different applications. Default: 2.
return_sequences: Boolean. Whether to return the last output. in the output
sequence, or the full sequence. Default: `False`.
return_state: Boolean. Whether to return the last state in addition to the
output. Default: `False`.
go_backwards: Boolean (default `False`). If True, process the input sequence
backwards and return the reversed sequence.
stateful: Boolean (default `False`). If True, the last state for each sample
at index i in a batch will be used as initial state for the sample of
index i in the following batch.
time_major: The shape format of the `inputs` and `outputs` tensors.
If True, the inputs and outputs will be in shape
`[timesteps, batch, feature]`, whereas in the False case, it will be
`[batch, timesteps, feature]`. Using `time_major = True` is a bit more
efficient because it avoids transposes at the beginning and end of the
RNN calculation. However, most TensorFlow data is batch-major, so by
default this function accepts input and emits output in batch-major
form.
unroll: Boolean (default `False`). If True, the network will be unrolled,
else a symbolic loop will be used. Unrolling can speed-up a RNN, although
it tends to be more memory-intensive. Unrolling is only suitable for short
sequences.
Call arguments:
inputs: A 3D tensor with shape `[batch, timesteps, feature]`.
mask: Binary tensor of shape `[batch, timesteps]` indicating whether
a given timestep should be masked (optional, defaults to `None`).
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. This argument is passed to the cell
when calling it. This is only relevant if `dropout` or
`recurrent_dropout` is used (optional, defaults to `None`).
initial_state: List of initial state tensors to be passed to the first
call of the cell (optional, defaults to `None` which causes creation
of zero-filled initial state tensors)."
6606,call,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,1101,method,
6607,step,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,1124,method,
6608,standard_lstm,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,1239,function,"LSTM with standard kernel implementation.
This implementation can be run on all types for hardware.
This implementation lifts out all the layer weights and make them function
parameters. It has same number of tensor input params as the CuDNN
counterpart. The RNN step logic has been simplified, eg dropout and mask is
removed since CuDNN implementation does not support that.
Note that the first half of the bias tensor should be ignored by this impl.
The CuDNN impl need an extra set of input gate bias. In order to make the both
function take same shape of parameter, that extra set of bias is also feed
here.
Args:
inputs: input tensor of LSTM layer.
init_h: initial state tensor for the cell output.
init_c: initial state tensor for the cell hidden state.
kernel: weights for cell kernel.
recurrent_kernel: weights for cell recurrent kernel.
bias: weights for cell kernel bias and recurrent bias. Only recurrent bias
is used in this case.
mask: Boolean tensor for mask out the steps within sequence.
time_major: boolean, whether the inputs are in the format of
[time, batch, feature] or [batch, time, feature].
go_backwards: Boolean (default False). If True, process the input sequence
backwards and return the reversed sequence.
sequence_lengths: The lengths of all sequences coming from a variable length
input, such as ragged tensors. If the input has a fixed timestep size,
this should be None.
zero_output_for_mask: Boolean, whether to output zero for masked timestep.
Returns:
last_output: output tensor for the last timestep, which has shape
[batch, units].
outputs: output tensor for all timesteps, which has shape
[batch, time, units].
state_0: the cell output, which has same shape as init_h.
state_1: the cell hidden state, which has same shape as init_c.
runtime: constant string tensor which indicate real runtime hardware. This
value is for testing purpose and should be used by user."
6609,gpu_lstm,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,1321,function,"LSTM with either CuDNN or ROCm implementation which is only available for GPU.
Note that currently only right padded data is supported, or the result will be
polluted by the unmasked data which should be filtered.
Args:
inputs: Input tensor of LSTM layer.
init_h: Initial state tensor for the cell output.
init_c: Initial state tensor for the cell hidden state.
kernel: Weights for cell kernel.
recurrent_kernel: Weights for cell recurrent kernel.
bias: Weights for cell kernel bias and recurrent bias. Only recurrent bias
is used in this case.
mask: Boolean tensor for mask out the steps within sequence.
time_major: Boolean, whether the inputs are in the format of [time, batch,
feature] or [batch, time, feature].
go_backwards: Boolean (default False). If True, process the input sequence
backwards and return the reversed sequence.
sequence_lengths: The lengths of all sequences coming from a variable length
input, such as ragged tensors. If the input has a fixed timestep size,
this should be None.
Returns:
last_output: Output tensor for the last timestep, which has shape
[batch, units].
outputs: Output tensor for all timesteps, which has shape
[batch, time, units].
state_0: The cell output, which has same shape as init_h.
state_1: The cell hidden state, which has same shape as init_c.
runtime: Constant string tensor which indicate real runtime hardware. This
value is for testing purpose and should not be used by user."
6610,lstm_with_backend_selection,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,1443,function,"Call the LSTM with optimized backend kernel selection.
Under the hood, this function will create two TF function, one with the most
generic kernel and can run on all device condition, and the second one with
CuDNN specific kernel, which can only run on GPU.
The first function will be called with normal_lstm_params, while the second
function is not called, but only registered in the graph. The Grappler will
do the proper graph rewrite and swap the optimized TF function based on the
device placement.
Args:
inputs: Input tensor of LSTM layer.
init_h: Initial state tensor for the cell output.
init_c: Initial state tensor for the cell hidden state.
kernel: Weights for cell kernel.
recurrent_kernel: Weights for cell recurrent kernel.
bias: Weights for cell kernel bias and recurrent bias. Only recurrent bias
is used in this case.
mask: Boolean tensor for mask out the steps within sequence.
time_major: Boolean, whether the inputs are in the format of
[time, batch, feature] or [batch, time, feature].
go_backwards: Boolean (default False). If True, process the input sequence
backwards and return the reversed sequence.
sequence_lengths: The lengths of all sequences coming from a variable length
input, such as ragged tensors. If the input has a fixed timestep size,
this should be None.
zero_output_for_mask: Boolean, whether to output zero for masked timestep.
Returns:
List of output tensors, same as standard_lstm."
6611,is_sequence_right_padded,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,1565,function,"Check the mask tensor and see if it right padded.
For CuDNN kernel, it uses the sequence length param to skip the tailing
timestep. If the data is left padded, or not a strict right padding (has
masked value in the middle of the sequence), then CuDNN kernel won't be work
properly in those cases.
Left padded data: [[False, False, True, True, True]].
Right padded data: [[True, True, True, False, False]].
Mixture of mask/unmasked data: [[True, False, True, False, False]].
Note that for the mixed data example above, the actually data RNN should see
are those 2 Trues (index 0 and 2), the index 1 False should be ignored and not
pollute the internal states.
Args:
mask: the Boolean tensor with shape [batch, timestep]
Returns:
boolean scalar tensor, whether the mask is strictly right padded."
6612,has_fully_masked_sequence,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,1594,function,
6613,is_cudnn_supported_inputs,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,1607,function,
6614,calculate_sequence_by_mask,tensorflow/tensorflow/python/keras/layers/recurrent_v2.py,1616,function,"Calculate the sequence length tensor (1-D) based on the masking tensor.
The masking tensor is a 2D boolean tensor with shape [batch, timestep]. For
any timestep that should be masked, the corresponding field will be False.
Consider the following example:
a = [[True, True, False, False],
[True, True, True, False]]
It is a (2, 4) tensor, and the corresponding sequence length result should be
1D tensor with value [2, 3]. Note that the masking tensor must be right
padded that could be checked by, e.g., `is_sequence_right_padded()`.
Args:
mask: Boolean tensor with shape [batch, timestep] or [timestep, batch] if
time_major=True.
time_major: Boolean, which indicates whether the mask is time major or batch
major.
Returns:
sequence_length: 1D int32 tensor."
6615,DropoutWrapper,tensorflow/tensorflow/python/keras/layers/rnn_cell_wrapper_v2.py,98,class,Operator adding dropout to inputs and outputs of the given cell.
6616,ResidualWrapper,tensorflow/tensorflow/python/keras/layers/rnn_cell_wrapper_v2.py,113,class,RNNCell wrapper that ensures cell inputs are added to the outputs.
6617,DeviceWrapper,tensorflow/tensorflow/python/keras/layers/rnn_cell_wrapper_v2.py,124,class,Operator that ensures an RNNCell runs on a particular device.
6618,populate_deserializable_objects,tensorflow/tensorflow/python/keras/layers/serialization.py,83,function,"Populates dict ALL_OBJECTS with every built-in layer.
"
6619,serialize,tensorflow/tensorflow/python/keras/layers/serialization.py,154,function,
6620,deserialize,tensorflow/tensorflow/python/keras/layers/serialization.py,159,function,"Instantiates a layer from a config dictionary.
Arguments:
config: dict of the form {'class_name': str, 'config': dict}
custom_objects: dict mapping class names (or function names)
of custom (non-Keras) objects to class/functions
Returns:
Layer instance (may be Model, Sequential, Network, Layer...)"
6621,SerializableInt,tensorflow/tensorflow/python/keras/layers/serialization_test.py,33,class,
6622,get_config,tensorflow/tensorflow/python/keras/layers/serialization_test.py,38,method,
6623,from_config,tensorflow/tensorflow/python/keras/layers/serialization_test.py,42,method,
6624,MyAdd,tensorflow/tensorflow/python/keras/layers/tensorflow_op_layer_test.py,190,class,
6625,call,tensorflow/tensorflow/python/keras/layers/tensorflow_op_layer_test.py,192,method,
6626,LayerWithLayer,tensorflow/tensorflow/python/keras/layers/tensorflow_op_layer_test.py,203,class,
6627,build,tensorflow/tensorflow/python/keras/layers/tensorflow_op_layer_test.py,205,method,
6628,call,tensorflow/tensorflow/python/keras/layers/tensorflow_op_layer_test.py,209,method,
6629,Wrapper,tensorflow/tensorflow/python/keras/layers/wrappers.py,40,class,"Abstract wrapper base class.
Wrappers take another layer and augment it in various ways.
Do not use this class as a layer, it is only an abstract base class.
Two usable wrappers are the `TimeDistributed` and `Bidirectional` wrappers.
Arguments:
layer: The layer to be wrapped."
6630,build,tensorflow/tensorflow/python/keras/layers/wrappers.py,56,method,
6631,activity_regularizer,tensorflow/tensorflow/python/keras/layers/wrappers.py,63,method,
6632,get_config,tensorflow/tensorflow/python/keras/layers/wrappers.py,69,method,
6633,from_config,tensorflow/tensorflow/python/keras/layers/wrappers.py,75,method,
6634,TimeDistributed,tensorflow/tensorflow/python/keras/layers/wrappers.py,85,class,"This wrapper allows to apply a layer to every temporal slice of an input.
The input should be at least 3D, and the dimension of index one
will be considered to be the temporal dimension.
Consider a batch of 32 video samples, where each sample is a 128x128 RGB image
with `channels_last` data format, across 10 timesteps.
The batch input shape is `(32, 10, 128, 128, 3)`.
You can then use `TimeDistributed` to apply a `Conv2D` layer to each of the
10 timesteps, independently:
>>> inputs = tf.keras.Input(shape=(10, 128, 128, 3))
>>> conv_2d_layer = tf.keras.layers.Conv2D(64, (3, 3))
>>> outputs = tf.keras.layers.TimeDistributed(conv_2d_layer)(inputs)
>>> outputs.shape
TensorShape([None, 10, 126, 126, 64])
Arguments:
layer: a `tf.keras.layers.Layer` instance.
Call arguments:
inputs: Input tensor.
training: Python boolean indicating whether the layer should behave in
training mode or in inference mode. This argument is passed to the
wrapped layer (only if the layer supports this argument).
mask: Binary tensor of shape `(samples, timesteps)` indicating whether
a given timestep should be masked. This argument is passed to the
wrapped layer (only if the layer supports this argument).
Raises:
ValueError: If not initialized with a `tf.keras.layers.Layer` instance."
6635,build,tensorflow/tensorflow/python/keras/layers/wrappers.py,168,method,
6636,compute_output_shape,tensorflow/tensorflow/python/keras/layers/wrappers.py,180,method,
6637,call,tensorflow/tensorflow/python/keras/layers/wrappers.py,192,method,
6638,compute_mask,tensorflow/tensorflow/python/keras/layers/wrappers.py,247,method,"Computes an output mask tensor for Embedding layer.
This is based on the inputs, mask, and the inner layer.
If batch size is specified:
Simply return the input `mask`. (An rnn-based implementation with
more than one rnn inputs is required but not supported in tf.keras yet.)
Otherwise we call `compute_mask` of the inner layer at each time step.
If the output mask at each time step is not `None`:
(E.g., inner layer is Masking or RNN)
Concatenate all of them and return the concatenation.
If the output mask at each time step is `None` and the input mask is not
`None`:(E.g., inner layer is Dense)
Reduce the input_mask to 2 dimensions and return it.
Otherwise (both the output mask and the input mask are `None`):
(E.g., `mask` is not used at all)
Return `None`.
Arguments:
inputs: Tensor with shape [batch size, timesteps, ...] indicating the
input to TimeDistributed. If static shape information is available for
""batch size"", `mask` is returned unmodified.
mask: Either None (indicating no masking) or a Tensor indicating the
input mask for TimeDistributed. The shape can be static or dynamic.
Returns:
Either None (no masking), or a [batch size, timesteps, ...] Tensor with
an output mask for the TimeDistributed layer with the shape beyond the
second dimension being the value of the input mask shape(if the computed
output mask is none), an output mask with the shape beyond the first
dimension being the value of the mask shape(if mask is not None) or
output mask with the shape beyond the first dimension being the
value of the computed output shape."
6639,step,tensorflow/tensorflow/python/keras/layers/wrappers.py,203,method,
6640,Bidirectional,tensorflow/tensorflow/python/keras/layers/wrappers.py,325,class,"Bidirectional wrapper for RNNs.
Arguments:
layer: `keras.layers.RNN` instance, such as `keras.layers.LSTM` or
`keras.layers.GRU`. It could also be a `keras.layers.Layer` instance
that meets the following criteria:
1. Be a sequence-processing layer (accepts 3D+ inputs).
2. Have a `go_backwards`, `return_sequences` and `return_state`
attribute (with the same semantics as for the `RNN` class).
3. Have an `input_spec` attribute.
4. Implement serialization via `get_config()` and `from_config()`.
Note that the recommended way to create new RNN layers is to write a
custom RNN cell and use it with `keras.layers.RNN`, instead of
subclassing `keras.layers.Layer` directly.
merge_mode: Mode by which outputs of the forward and backward RNNs will be
combined. One of {'sum', 'mul', 'concat', 'ave', None}. If None, the
outputs will not be combined, they will be returned as a list. Default
value is 'concat'.
backward_layer: Optional `keras.layers.RNN`, or `keras.layers.Layer`
instance to be used to handle backwards input processing.
If `backward_layer` is not provided, the layer instance passed as the
`layer` argument will be used to generate the backward layer
automatically.
Note that the provided `backward_layer` layer should have properties
matching those of the `layer` argument, in particular it should have the
same values for `stateful`, `return_states`, `return_sequence`, etc.
In addition, `backward_layer` and `layer` should have different
`go_backwards` argument values.
A `ValueError` will be raised if these requirements are not met.
Call arguments:
The call arguments for this layer are the same as those of the wrapped RNN
layer.
Beware that when passing the `initial_state` argument during the call of
this layer, the first half in the list of elements in the `initial_state`
list will be passed to the forward RNN call and the last half in the list
of elements will be passed to the backward RNN call.
Raises:
ValueError:
1. If `layer` or `backward_layer` is not a `Layer` instance.
2. In case of invalid `merge_mode` argument.
3. If `backward_layer` has mismatched properties compared to `layer`.
Examples:
```python
model = Sequential()
model.add(Bidirectional(LSTM(10, return_sequences=True), input_shape=(5, 10)))
model.add(Bidirectional(LSTM(10)))
model.add(Dense(5))
model.add(Activation('softmax'))
model.compile(loss='categorical_crossentropy', optimizer='rmsprop')
# With custom backward layer
model = Sequential()
forward_layer = LSTM(10, return_sequences=True)
backward_layer = LSTM(10, activation='relu', return_sequences=True,
go_backwards=True)
model.add(Bidirectional(forward_layer, backward_layer=backward_layer,
input_shape=(5, 10)))
model.add(Dense(5))
model.add(Activation('softmax'))
model.compile(loss='categorical_crossentropy', optimizer='rmsprop')
```"
6641,compute_output_shape,tensorflow/tensorflow/python/keras/layers/wrappers.py,497,method,
6642,call,tensorflow/tensorflow/python/keras/layers/wrappers.py,594,method,`Bidirectional.call` implements the same API as the wrapped `RNN`.
6643,reset_states,tensorflow/tensorflow/python/keras/layers/wrappers.py,679,method,
6644,build,tensorflow/tensorflow/python/keras/layers/wrappers.py,683,method,
6645,compute_mask,tensorflow/tensorflow/python/keras/layers/wrappers.py,690,method,
6646,constraints,tensorflow/tensorflow/python/keras/layers/wrappers.py,710,method,
6647,get_config,tensorflow/tensorflow/python/keras/layers/wrappers.py,717,method,
6648,from_config,tensorflow/tensorflow/python/keras/layers/wrappers.py,728,method,
6649,force_zero_output_for_mask,tensorflow/tensorflow/python/keras/layers/wrappers.py,436,method,
6650,ExampleWrapper,tensorflow/tensorflow/python/keras/layers/wrappers_test.py,1252,class,Simple Wrapper subclass.
6651,call,tensorflow/tensorflow/python/keras/layers/wrappers_test.py,1255,method,
6652,assert_like_rnncell,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,73,function,"Raises a TypeError if cell is not like an RNNCell.
NOTE: Do not rely on the error message (in particular in tests) which can be
subject to change to increase readability. Use
ASSERT_LIKE_RNNCELL_ERROR_REGEXP.
Args:
cell_name: A string to give a meaningful error referencing to the name of
the functionargument.
cell: The object which should behave like an RNNCell.
Raises:
TypeError: A human-friendly exception."
6653,RNNCell,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,185,class,"Abstract object representing an RNN cell.
Every `RNNCell` must have the properties below and implement `call` with
the signature `(output, next_state) = call(input, state)`. The optional
third input argument, `scope`, is allowed for backwards compatibility
purposes; but should be left off for new subclasses.
This definition of cell differs from the definition used in the literature.
In the literature, 'cell' refers to an object with a single scalar output.
This definition refers to a horizontal array of such units.
An RNN cell, in the most abstract setting, is anything that has
a state and performs some operation that takes a matrix of inputs.
This operation results in an output matrix with `self.output_size` columns.
If `self.state_size` is an integer, this operation also results in a new
state matrix with `self.state_size` columns. If `self.state_size` is a
(possibly nested tuple of) TensorShape object(s), then it should return a
matching structure of Tensors having shape `[batch_size].concatenate(s)`
for each `s` in `self.batch_size`."
6654,state_size,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,265,method,"size(s) of state(s) used by this cell.
It can be represented by an Integer, a TensorShape or a tuple of Integers
or TensorShapes."
6655,output_size,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,274,method,Integer or TensorShape: size of outputs produced by this cell.
6656,build,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,278,method,
6657,get_initial_state,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,283,method,
6658,zero_state,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,313,method,"Return zero-filled state tensor(s).
Args:
batch_size: int, float, or unit Tensor representing the batch size.
dtype: the data type to use for the state.
Returns:
If `state_size` is an int or TensorShape, then the return value is a
`N-D` tensor of shape `[batch_size, state_size]` filled with zeros.
If `state_size` is a nested list or tuple, then the return value is
a nested list or tuple (of the same structure) of `2-D` tensors with
the shapes `[batch_size, s]` for each s in `state_size`."
6659,get_config,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,345,method,
6660,LayerRNNCell,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,349,class,"Subclass of RNNCells that act like proper `tf.Layer` objects.
For backwards compatibility purposes, most `RNNCell` instances allow their
`call` methods to instantiate variables via `tf.compat.v1.get_variable`. The
underlying
variable scope thus keeps track of any variables, and returning cached
versions. This is atypical of `tf.layer` objects, which separate this
part of layer building into a `build` method that is only called once.
Here we provide a subclass for `RNNCell` objects that act exactly as
`Layer` objects do. They must provide a `build` method and their
`call` methods do not access Variables `tf.compat.v1.get_variable`."
6661,BasicRNNCell,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,392,class,"The most basic RNN cell.
Note that this cell is not optimized for performance. Please use
`tf.contrib.cudnn_rnn.CudnnRNNTanh` for better performance on GPU.
Args:
num_units: int, The number of units in the RNN cell.
activation: Nonlinearity to use. Default: `tanh`. It could also be string
that is within Keras activation function names.
reuse: (optional) Python boolean describing whether to reuse variables in an
existing scope. If not `True`, and the existing scope already has the
given variables, an error is raised.
name: String, the name of the layer. Layers with the same name will share
weights, but to avoid mistakes we require reuse=True in such cases.
dtype: Default dtype of the layer (default of `None` means use the type of
the first input). Required when `build` is called before `call`.
**kwargs: Dict, keyword named properties for common layer attributes, like
`trainable` etc when constructing the cell from configs of get_config()."
6662,state_size,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,441,method,
6663,output_size,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,445,method,
6664,build,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,449,method,
6665,call,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,466,method,Most basic RNN: output = new_state = act(W * input + U * state + B).
6666,get_config,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,475,method,
6667,GRUCell,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,486,class,"Gated Recurrent Unit cell.
Note that this cell is not optimized for performance. Please use
`tf.contrib.cudnn_rnn.CudnnGRU` for better performance on GPU, or
`tf.contrib.rnn.GRUBlockCellV2` for better performance on CPU.
Args:
num_units: int, The number of units in the GRU cell.
activation: Nonlinearity to use. Default: `tanh`.
reuse: (optional) Python boolean describing whether to reuse variables in an
existing scope. If not `True`, and the existing scope already has the
given variables, an error is raised.
kernel_initializer: (optional) The initializer to use for the weight and
projection matrices.
bias_initializer: (optional) The initializer to use for the bias.
name: String, the name of the layer. Layers with the same name will share
weights, but to avoid mistakes we require reuse=True in such cases.
dtype: Default dtype of the layer (default of `None` means use the type of
the first input). Required when `build` is called before `call`.
**kwargs: Dict, keyword named properties for common layer attributes, like
`trainable` etc when constructing the cell from configs of get_config().
References:
Learning Phrase Representations using RNN Encoder Decoder for Statistical
Machine Translation:
[Cho et al., 2014]
(https://aclanthology.coli.uni-saarland.de/papers/D14-1179/d14-1179)
([pdf](http://emnlp2014.org/papers/pdf/EMNLP2014179.pdf))"
6668,state_size,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,549,method,
6669,output_size,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,553,method,
6670,build,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,557,method,
6671,call,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,586,method,Gated recurrent unit (GRU) with nunits cells.
6672,get_config,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,607,method,
6673,LSTMStateTuple,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,623,class,"Tuple used by LSTM Cells for `state_size`, `zero_state`, and output state.
Stores two elements: `(c, h)`, in that order. Where `c` is the hidden state
and `h` is the output.
Only used when `state_is_tuple=True`."
6674,dtype,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,634,method,
6675,BasicLSTMCell,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,643,class,"DEPRECATED: Please use `tf.compat.v1.nn.rnn_cell.LSTMCell` instead.
Basic LSTM recurrent network cell.
The implementation is based on
We add forget_bias (default: 1) to the biases of the forget gate in order to
reduce the scale of forgetting in the beginning of the training.
It does not allow cell clipping, a projection layer, and does not
use peep-hole connections: it is the basic baseline.
For advanced models, please use the full `tf.compat.v1.nn.rnn_cell.LSTMCell`
that follows.
Note that this cell is not optimized for performance. Please use
`tf.contrib.cudnn_rnn.CudnnLSTM` for better performance on GPU, or
`tf.contrib.rnn.LSTMBlockCell` and `tf.contrib.rnn.LSTMBlockFusedCell` for
better performance on CPU."
6676,state_size,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,724,method,
6677,output_size,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,729,method,
6678,build,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,733,method,
6679,call,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,750,method,"Long short-term memory cell (LSTM).
Args:
inputs: `2-D` tensor with shape `[batch_size, input_size]`.
state: An `LSTMStateTuple` of state tensors, each shaped `[batch_size,
num_units]`, if `state_is_tuple` has been set to `True`. Otherwise, a
`Tensor` shaped `[batch_size, 2 * num_units]`.
Returns:
A pair containing the new hidden state, and the new state (either a
`LSTMStateTuple` or a concatenated state, depending on
`state_is_tuple`)."
6680,get_config,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,798,method,
6681,LSTMCell,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,811,class,"Long short-term memory unit (LSTM) recurrent network cell.
The default non-peephole implementation is based on (Gers et al., 1999).
The peephole implementation is based on (Sak et al., 2014).
The class uses optional peep-hole connections, optional cell clipping, and
an optional projection layer.
Note that this cell is not optimized for performance. Please use
`tf.contrib.cudnn_rnn.CudnnLSTM` for better performance on GPU, or
`tf.contrib.rnn.LSTMBlockCell` and `tf.contrib.rnn.LSTMBlockFusedCell` for
better performance on CPU.
References:
Long short-term memory recurrent neural network architectures for large
scale acoustic modeling:
[Sak et al., 2014]
(https://www.isca-speech.org/archive/interspeech_2014/i14_0338.html)
([pdf]
(https://www.isca-speech.org/archive/archive_papers/interspeech_2014/i14_0338.pdf))
Learning to forget:
[Gers et al., 1999]
(http://digital-library.theiet.org/content/conferences/10.1049/cp_19991218)
([pdf](https://arxiv.org/pdf/1409.2329.pdf))
Long Short-Term Memory:
[Hochreiter et al., 1997]
(https://www.mitpressjournals.org/doi/abs/10.1162/neco.1997.9.8.1735)
([pdf](http://ml.jku.at/publications/older/3504.pdf))"
6682,state_size,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,946,method,
6683,output_size,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,950,method,
6684,build,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,954,method,
6685,call,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,997,method,"Run one step of LSTM.
Args:
inputs: input Tensor, must be 2-D, `[batch, input_size]`.
state: if `state_is_tuple` is False, this must be a state Tensor, `2-D,
[batch, state_size]`. If `state_is_tuple` is True, this must be a tuple
of state Tensors, both `2-D`, with column sizes `c_state` and `m_state`.
Returns:
A tuple containing:
- A `2-D, [batch, output_dim]`, Tensor representing the output of the
LSTM after reading `inputs` when previous state was `state`.
Here output_dim is:
num_proj if num_proj was set,
num_units otherwise.
- Tensor(s) representing the new state of LSTM after reading `inputs` when
the previous state was `state`. Same type and shape(s) as `state`.
Raises:
ValueError: If input size cannot be inferred from inputs via
static shape inference."
6686,get_config,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,1075,method,
6687,DropoutWrapper,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,1176,class,Operator adding dropout to inputs and outputs of the given cell.
6688,ResidualWrapper,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,1187,class,RNNCell wrapper that ensures cell inputs are added to the outputs.
6689,DeviceWrapper,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,1198,class,
6690,MultiRNNCell,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,1208,class,"RNN cell composed sequentially of multiple simple cells.
Example:
```python
num_units = [128, 64]
cells = [BasicLSTMCell(num_units=n) for n in num_units]
stacked_rnn_cell = MultiRNNCell(cells)
```"
6691,state_size,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,1262,method,
6692,output_size,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,1269,method,
6693,zero_state,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,1272,method,
6694,trainable_weights,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,1282,method,
6695,non_trainable_weights,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,1292,method,
6696,call,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_impl.py,1305,method,"Run this multi-layer cell on inputs, starting from state."
6697,DropoutWrapperBase,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,38,class,Operator adding dropout to inputs and outputs of the given cell.
6698,wrapped_cell,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,186,method,
6699,state_size,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,190,method,
6700,output_size,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,194,method,
6701,build,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,197,method,
6702,zero_state,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,201,method,
6703,get_config,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,291,method,Returns the config of the dropout wrapper.
6704,from_config,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,311,method,
6705,tensor_and_const_value,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,118,method,
6706,convert_to_batch_shape,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,149,method,
6707,batch_noise,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,155,method,
6708,dropout,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,232,method,
6709,dropout,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,244,method,
6710,ResidualWrapperBase,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,323,class,RNNCell wrapper that ensures cell inputs are added to the outputs.
6711,state_size,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,341,method,
6712,output_size,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,345,method,
6713,zero_state,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,348,method,
6714,get_config,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,383,method,Returns the config of the residual wrapper.
6715,from_config,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,399,method,
6716,assert_shape_match,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,372,method,
6717,default_residual_fn,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,375,method,
6718,DeviceWrapperBase,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,411,class,Operator that ensures an RNNCell runs on a particular device.
6719,state_size,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,428,method,
6720,output_size,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,432,method,
6721,zero_state,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,435,method,
6722,get_config,tensorflow/tensorflow/python/keras/layers/legacy_rnn/rnn_cell_wrapper_impl.py,445,method,
6723,dense,tensorflow/tensorflow/python/keras/layers/ops/core.py,30,function,"Densely connected NN layer op.
Arguments:
inputs: `tf.Tensor` or `tf.SparseTensor`. Inputs to operation.
kernel: `tf.Variable`. Matrix kernel.
bias: (Optional) `tf.Variable`. Bias to add to outputs.
activation: (Optional) 1-argument callable. Activation function to apply to
outputs.
dtype: (Optional) `tf.DType`. Dtype to cast `inputs` to.
Returns:
`tf.Tensor`. Output of dense connection."
6724,CategoryCrossing,tensorflow/tensorflow/python/keras/layers/preprocessing/category_crossing.py,39,class,"Category crossing layer.
This layer concatenates multiple categorical inputs into a single categorical
output (similar to Cartesian product). The output dtype is string.
Usage:
>>> inp_1 = ['a', 'b', 'c']
>>> inp_2 = ['d', 'e', 'f']
>>> layer = tf.keras.layers.experimental.preprocessing.CategoryCrossing()
>>> layer([inp_1, inp_2])
<tf.Tensor: shape=(3, 1), dtype=string, numpy=
array([[b'a_X_d'],
[b'b_X_e'],
[b'c_X_f']], dtype=object)>
>>> inp_1 = ['a', 'b', 'c']
>>> inp_2 = ['d', 'e', 'f']
>>> layer = tf.keras.layers.experimental.preprocessing.CategoryCrossing(
... separator='-')
>>> layer([inp_1, inp_2])
<tf.Tensor: shape=(3, 1), dtype=string, numpy=
array([[b'a-d'],
[b'b-e'],
[b'c-f']], dtype=object)>
Arguments:
depth: depth of input crossing. By default None, all inputs are crossed into
one output. It can also be an int or tuple/list of ints. Passing an
integer will create combinations of crossed outputs with depth up to that
integer, i.e., [1, 2, ..., `depth`), and passing a tuple of integers will
create crossed outputs with depth for the specified values in the tuple,
i.e., `depth`=(N1, N2) will create all possible crossed outputs with depth
equal to N1 or N2. Passing `None` means a single crossed output with all
inputs. For example, with inputs `a`, `b` and `c`, `depth=2` means the
output will be [a;b;c;cross(a, b);cross(bc);cross(ca)].
separator: A string added between each input being joined. Defaults to
'_X_'.
name: Name to give to the layer.
**kwargs: Keyword arguments to construct a layer.
Input shape: a list of string or int tensors or sparse tensors of shape
`[batch_size, d1, ..., dm]`
Output shape: a single string or int tensor or sparse tensor of shape
`[batch_size, d1, ..., dm]`
Returns:
If any input is `RaggedTensor`, the output is `RaggedTensor`.
Else, if any input is `SparseTensor`, the output is `SparseTensor`.
Otherwise, the output is `Tensor`.
Example: (`depth`=None)
If the layer receives three inputs:
`a=[[1], [4]]`, `b=[[2], [5]]`, `c=[[3], [6]]`
the output will be a string tensor:
`[[b'1_X_2_X_3'], [b'4_X_5_X_6']]`
Example: (`depth` is an integer)
With the same input above, and if `depth`=2,
the output will be a list of 6 string tensors:
`[[b'1'], [b'4']]`
`[[b'2'], [b'5']]`
`[[b'3'], [b'6']]`
`[[b'1_X_2'], [b'4_X_5']]`,
`[[b'2_X_3'], [b'5_X_6']]`,
`[[b'3_X_1'], [b'6_X_4']]`
Example: (`depth` is a tuple/list of integers)
With the same input above, and if `depth`=(2, 3)
the output will be a list of 4 string tensors:
`[[b'1_X_2'], [b'4_X_5']]`,
`[[b'2_X_3'], [b'5_X_6']]`,
`[[b'3_X_1'], [b'6_X_4']]`,
`[[b'1_X_2_X_3'], [b'4_X_5_X_6']]`"
6725,partial_crossing,tensorflow/tensorflow/python/keras/layers/preprocessing/category_crossing.py,129,method,Gets the crossed output from a partial list/tuple of inputs.
6726,call,tensorflow/tensorflow/python/keras/layers/preprocessing/category_crossing.py,151,method,
6727,compute_output_shape,tensorflow/tensorflow/python/keras/layers/preprocessing/category_crossing.py,174,method,
6728,compute_output_signature,tensorflow/tensorflow/python/keras/layers/preprocessing/category_crossing.py,190,method,
6729,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/category_crossing.py,204,method,
6730,batch_wrapper,tensorflow/tensorflow/python/keras/layers/preprocessing/category_crossing_distribution_test.py,36,function,
6731,CategoryEncoding,tensorflow/tensorflow/python/keras/layers/preprocessing/category_encoding.py,56,class,"Category encoding layer.
This layer provides options for condensing data into a categorical encoding.
It accepts integer values as inputs and outputs a dense representation
(one sample = 1-index tensor of float values representing data about the
sample's tokens) of those inputs.
Examples:
>>> layer = tf.keras.layers.experimental.preprocessing.CategoryEncoding(
... max_tokens=4, output_mode=""count"")
>>> layer([[0, 1], [0, 0], [1, 2], [3, 1]])
<tf.Tensor: shape=(4, 4), dtype=float32, numpy=
array([[1., 1., 0., 0.],
[2., 0., 0., 0.],
[0., 1., 1., 0.],
[0., 1., 0., 1.]], dtype=float32)>
Examples with weighted inputs:
>>> layer = tf.keras.layers.experimental.preprocessing.CategoryEncoding(
... max_tokens=4, output_mode=""count"")
>>> count_weights = np.array([[.1, .2], [.1, .1], [.2, .3], [.4, .2]])
>>> layer([[0, 1], [0, 0], [1, 2], [3, 1]], count_weights=count_weights)
<tf.Tensor: shape=(4, 4), dtype=float64, numpy=
array([[0.1, 0.2, 0. , 0. ],
[0.2, 0. , 0. , 0. ],
[0. , 0.2, 0.3, 0. ],
[0. , 0.2, 0. , 0.4]])>
Attributes:
max_tokens: The maximum size of the vocabulary for this layer. If None,
there is no cap on the size of the vocabulary.
output_mode: Specification for the output of the layer.
Defaults to ""binary"". Values can
be ""binary"", ""count"" or ""tf-idf"", configuring the layer as follows:
""binary"": Outputs a single int array per batch, of either vocab_size or
max_tokens size, containing 1s in all elements where the token mapped
to that index exists at least once in the batch item.
""count"": As ""binary"", but the int array contains a count of the number
of times the token at that index appeared in the batch item.
""tf-idf"": As ""binary"", but the TF-IDF algorithm is applied to find the
value in each token slot.
sparse: Boolean. If true, returns a `SparseTensor` instead of a dense
`Tensor`. Defaults to `False`.
Call arguments:
inputs: A 2D tensor `(samples, timesteps)`.
count_weights: A 2D tensor in the same shape as `inputs` indicating the
weight for each sample value when summing up in `count` mode. Not used in
`binary` or `tfidf` mode."
6732,compute_output_shape,tensorflow/tensorflow/python/keras/layers/preprocessing/category_encoding.py,170,method,
6733,compute_output_signature,tensorflow/tensorflow/python/keras/layers/preprocessing/category_encoding.py,173,method,
6734,adapt,tensorflow/tensorflow/python/keras/layers/preprocessing/category_encoding.py,182,method,"Fits the state of the preprocessing layer to the dataset.
Overrides the default adapt method to apply relevant preprocessing to the
inputs before passing to the combiner.
Arguments:
data: The data to train on. It can be passed either as a tf.data Dataset,
or as a numpy array.
reset_state: Optional argument specifying whether to clear the state of
the layer at the start of the call to `adapt`. This must be True for
this layer, which does not support repeated calls to `adapt`.
Raises:
RuntimeError: if the layer cannot be adapted at this time."
6735,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/category_encoding.py,214,method,
6736,set_num_elements,tensorflow/tensorflow/python/keras/layers/preprocessing/category_encoding.py,241,method,
6737,set_tfidf_data,tensorflow/tensorflow/python/keras/layers/preprocessing/category_encoding.py,253,method,
6738,call,tensorflow/tensorflow/python/keras/layers/preprocessing/category_encoding.py,270,method,
6739,batch_wrapper,tensorflow/tensorflow/python/keras/layers/preprocessing/category_encoding_distribution_test.py,36,function,
6740,get_layer_class,tensorflow/tensorflow/python/keras/layers/preprocessing/category_encoding_test.py,43,function,
6741,CategoryEncoding,tensorflow/tensorflow/python/keras/layers/preprocessing/category_encoding_v1.py,27,class,"CategoryEncoding layer.
This layer provides options for condensing input data into denser
representations. It accepts either integer values or strings as inputs,
allows users to map those inputs into a contiguous integer space, and
outputs either those integer values (one sample = 1D tensor of integer token
indices) or a dense representation (one sample = 1D tensor of float values
representing data about the sample's tokens).
If desired, the user can call this layer's adapt() method on a dataset.
When this layer is adapted, it will analyze the dataset, determine the
frequency of individual integer or string values, and create a 'vocabulary'
from them. This vocabulary can have unlimited size or be capped, depending
on the configuration options for this layer; if there are more unique
values in the input than the maximum vocabulary size, the most frequent
terms will be used to create the vocabulary.
Attributes:
max_elements: The maximum size of the vocabulary for this layer. If None,
there is no cap on the size of the vocabulary.
output_mode: Optional specification for the output of the layer. Values can
be ""int"", ""binary"", ""count"" or ""tf-idf"", configuring the layer as follows:
""int"": Outputs integer indices, one integer index per split string
token.
""binary"": Outputs a single int array per batch, of either vocab_size or
max_elements size, containing 1s in all elements where the token
mapped to that index exists at least once in the batch item.
""count"": As ""binary"", but the int array contains a count of the number
of times the token at that index appeared in the batch item.
""tf-idf"": As ""binary"", but the TF-IDF algorithm is applied to find the
value in each token slot.
output_sequence_length: Only valid in INT mode. If set, the output will have
its time dimension padded or truncated to exactly `output_sequence_length`
values, resulting in a tensor of shape [batch_size,
output_sequence_length] regardless of the input shape.
pad_to_max_elements: Only valid in ""binary"", ""count"", and ""tf-idf"" modes.
If True, the output will have its feature axis padded to `max_elements`
even if the number of unique values in the vocabulary is less than
max_elements, resulting in a tensor of shape [batch_size, max_elements]
regardless of vocabulary size. Defaults to False."
6742,Discretization,tensorflow/tensorflow/python/keras/layers/preprocessing/discretization.py,32,class,"Buckets data into discrete ranges.
This layer will place each element of its input data into one of several
contiguous ranges and output an integer index indicating which range each
element was placed in.
Input shape:
Any `tf.Tensor` or `tf.RaggedTensor` of dimension 2 or higher.
Output shape:
Same as input shape.
Attributes:
bins: Optional boundary specification. Bins include the left boundary and
exclude the right boundary, so `bins=[0., 1., 2.]` generates bins
`(-inf, 0.)`, `[0., 1.)`, `[1., 2.)`, and `[2., +inf)`.
Examples:
Bucketize float values based on provided buckets.
>>> input = np.array([[-1.5, 1.0, 3.4, .5], [0.0, 3.0, 1.3, 0.0]])
>>> layer = tf.keras.layers.experimental.preprocessing.Discretization(
... bins=[0., 1., 2.])
>>> layer(input)
<tf.Tensor: shape=(2, 4), dtype=int32, numpy=
array([[0, 2, 3, 1],
[1, 3, 2, 1]], dtype=int32)>"
6743,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/discretization.py,67,method,
6744,compute_output_shape,tensorflow/tensorflow/python/keras/layers/preprocessing/discretization.py,74,method,
6745,compute_output_signature,tensorflow/tensorflow/python/keras/layers/preprocessing/discretization.py,77,method,
6746,call,tensorflow/tensorflow/python/keras/layers/preprocessing/discretization.py,85,method,
6747,Hashing,tensorflow/tensorflow/python/keras/layers/preprocessing/hashing.py,44,class,"Implements categorical feature hashing, also known as ""hashing trick"".
This layer transforms single or multiple categorical inputs to hashed output.
It converts a sequence of int or string to a sequence of int. The stable hash
function uses tensorflow::ops::Fingerprint to produce universal output that
is consistent across platforms.
This layer uses [FarmHash64](https://github.com/google/farmhash) by default,
which provides a consistent hashed output across different platforms and is
stable across invocations, regardless of device and context, by mixing the
input bits thoroughly.
If you want to obfuscate the hashed output, you can also pass a random `salt`
argument in the constructor. In that case, the layer will use the
[SipHash64](https://github.com/google/highwayhash) hash function, with
the `salt` value serving as additional input to the hash function.
Example (FarmHash64):
>>> layer = tf.keras.layers.experimental.preprocessing.Hashing(num_bins=3)
>>> inp = [['A'], ['B'], ['C'], ['D'], ['E']]
>>> layer(inp)
<tf.Tensor: shape=(5, 1), dtype=int64, numpy=
array([[1],
[0],
[1],
[1],
[2]])>
Example (FarmHash64) with list of inputs:
>>> layer = tf.keras.layers.experimental.preprocessing.Hashing(num_bins=3)
>>> inp_1 = [['A'], ['B'], ['C'], ['D'], ['E']]
>>> inp_2 = np.asarray([[5], [4], [3], [2], [1]])
>>> layer([inp_1, inp_2])
<tf.Tensor: shape=(5, 1), dtype=int64, numpy=
array([[1],
[1],
[0],
[2],
[0]])>
Example (SipHash64):
>>> layer = tf.keras.layers.experimental.preprocessing.Hashing(num_bins=3,
... salt=[133, 137])
>>> inp = [['A'], ['B'], ['C'], ['D'], ['E']]
>>> layer(inp)
<tf.Tensor: shape=(5, 1), dtype=int64, numpy=
array([[1],
[2],
[1],
[0],
[2]])>
Example (Siphash64 with a single integer, same as `salt=[133, 133]`
>>> layer = tf.keras.layers.experimental.preprocessing.Hashing(num_bins=3,
... salt=133)
>>> inp = [['A'], ['B'], ['C'], ['D'], ['E']]
>>> layer(inp)
<tf.Tensor: shape=(5, 1), dtype=int64, numpy=
array([[0],
[0],
[2],
[1],
[0]])>
Reference: [SipHash with salt](https://www.131002.net/siphash/siphash.pdf)
Arguments:
num_bins: Number of hash bins.
salt: A single unsigned integer or None.
If passed, the hash function used will be SipHash64, with these values
used as an additional input (known as a ""salt"" in cryptography).
These should be non-zero. Defaults to `None` (in that
case, the FarmHash64 hash function is used). It also supports
tuple/list of 2 unsigned integer numbers, see reference paper for details.
name: Name to give to the layer.
**kwargs: Keyword arguments to construct a layer.
Input shape: A single or list of string, int32 or int64 `Tensor`,
`SparseTensor` or `RaggedTensor` of shape `[batch_size, ...,]`
Output shape: An int64 `Tensor`, `SparseTensor` or `RaggedTensor` of shape
`[batch_size, ...]`. If any input is `RaggedTensor` then output is
`RaggedTensor`, otherwise if any input is `SparseTensor` then output is
`SparseTensor`, otherwise the output is `Tensor`."
6748,call,tensorflow/tensorflow/python/keras/layers/preprocessing/hashing.py,169,method,
6749,compute_output_shape,tensorflow/tensorflow/python/keras/layers/preprocessing/hashing.py,239,method,
6750,compute_output_signature,tensorflow/tensorflow/python/keras/layers/preprocessing/hashing.py,254,method,
6751,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/hashing.py,276,method,
6752,check_fill_mode_and_interpolation,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,60,function,
6753,Resizing,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,71,class,"Image resizing layer.
Resize the batched image input to target height and width. The input should
be a 4-D tensor in the format of NHWC.
Arguments:
height: Integer, the height of the output shape.
width: Integer, the width of the output shape.
interpolation: String, the interpolation method. Defaults to `bilinear`.
Supports `bilinear`, `nearest`, `bicubic`, `area`, `lanczos3`, `lanczos5`,
`gaussian`, `mitchellcubic`
name: A string, the name of the layer."
6754,call,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,100,method,
6755,compute_output_shape,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,107,method,
6756,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,112,method,
6757,CenterCrop,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,123,class,"Crop the central portion of the images to target height and width.
Input shape:
4D tensor with shape:
`(samples, height, width, channels)`, data_format='channels_last'.
Output shape:
4D tensor with shape:
`(samples, target_height, target_width, channels)`.
If the input height/width is even and the target height/width is odd (or
inversely), the input image is left-padded by 1 pixel.
Arguments:
height: Integer, the height of the output shape.
width: Integer, the width of the output shape.
name: A string, the name of the layer."
6758,call,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,150,method,
6759,compute_output_shape,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,176,method,
6760,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,181,method,
6761,RandomCrop,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,191,class,"Randomly crop the images to target height and width.
This layer will crop all the images in the same batch to the same cropping
location.
By default, random cropping is only applied during training. At inference
time, the images will be first rescaled to preserve the shorter side, and
center cropped. If you need to apply random cropping at inference time,
set `training` to True when calling the layer.
Input shape:
4D tensor with shape:
`(samples, height, width, channels)`, data_format='channels_last'.
Output shape:
4D tensor with shape:
`(samples, target_height, target_width, channels)`.
Arguments:
height: Integer, the height of the output shape.
width: Integer, the width of the output shape.
seed: Integer. Used to create a random seed.
name: A string, the name of the layer."
6762,call,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,225,method,
6763,compute_output_shape,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,283,method,
6764,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,288,method,
6765,random_cropped_inputs,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,229,method,Cropped inputs with stateless random ops.
6766,resize_and_center_cropped_inputs,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,247,method,Deterministically resize to shorter side and center crop.
6767,Rescaling,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,299,class,"Multiply inputs by `scale` and adds `offset`.
For instance:
1. To rescale an input in the `[0, 255]` range
to be in the `[0, 1]` range, you would pass `scale=1./255`.
2. To rescale an input in the `[0, 255]` range to be in the `[-1, 1]` range,
you would pass `scale=1./127.5, offset=-1`.
The rescaling is applied both during training and inference.
Input shape:
Arbitrary.
Output shape:
Same as input.
Arguments:
scale: Float, the scale to apply to the inputs.
offset: Float, the offset to apply to the inputs.
name: A string, the name of the layer."
6768,call,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,330,method,
6769,compute_output_shape,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,336,method,
6770,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,339,method,
6771,RandomFlip,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,354,class,"Randomly flip each image horizontally and vertically.
This layer will flip the images based on the `mode` attribute.
During inference time, the output will be identical to input. Call the layer
with `training=True` to flip the input.
Input shape:
4D tensor with shape:
`(samples, height, width, channels)`, data_format='channels_last'.
Output shape:
4D tensor with shape:
`(samples, height, width, channels)`, data_format='channels_last'.
Attributes:
mode: String indicating which flip mode to use. Can be ""horizontal"",
""vertical"", or ""horizontal_and_vertical"". Defaults to
""horizontal_and_vertical"". ""horizontal"" is a left-right flip and
""vertical"" is a top-bottom flip.
seed: Integer. Used to create a random seed.
name: A string, the name of the layer."
6772,call,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,402,method,
6773,compute_output_shape,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,421,method,
6774,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,424,method,
6775,random_flipped_inputs,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,406,method,
6776,RandomTranslation,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,435,class,"Randomly translate each image during training.
Arguments:
height_factor: a float represented as fraction of value, or a tuple
of size 2 representing lower and upper bound for shifting vertically.
A negative value means shifting image up, while a positive value
means shifting image down. When represented as a single positive float,
this value is used for both the upper and lower bound. For instance,
`height_factor=(-0.2, 0.3)` results in an output shifted by a random
amount in the range [-20%, +30%].
`height_factor=0.2` results in an output height shifted by a random
amount in the range [-20%, +20%].
width_factor: a float represented as fraction of value, or a tuple
of size 2 representing lower and upper bound for shifting horizontally.
A negative value means shifting image left, while a positive value
means shifting image right. When represented as a single positive float,
this value is used for both the upper and lower bound. For instance,
`width_factor=(-0.2, 0.3)` results in an output shifted left by 20%, and
shifted right by 30%.
`width_factor=0.2` results in an output height shifted left or right
by 20%.
fill_mode: Points outside the boundaries of the input are filled according
to the given mode (one of `{'constant', 'reflect', 'wrap', 'nearest'}`).
- *reflect*: `(d c b a | a b c d | d c b a)`
The input is extended by reflecting about the edge of the last pixel.
- *constant*: `(k k k k | a b c d | k k k k)`
The input is extended by filling all values beyond the edge with the
same constant value k = 0.
- *wrap*: `(a b c d | a b c d | a b c d)`
The input is extended by wrapping around to the opposite edge.
- *nearest*: `(a a a a | a b c d | d d d d)`
The input is extended by the nearest pixel.
interpolation: Interpolation mode. Supported values: ""nearest"", ""bilinear"".
seed: Integer. Used to create a random seed.
name: A string, the name of the layer.
Input shape:
4D tensor with shape: `(samples, height, width, channels)`,
data_format='channels_last'.
Output shape:
4D tensor with shape: `(samples, height, width, channels)`,
data_format='channels_last'.
Raise:
ValueError: if either bound is not between [0, 1], or upper bound is
less than lower bound."
6777,call,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,531,method,
6778,compute_output_shape,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,568,method,
6779,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,571,method,
6780,random_translated_inputs,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,535,method,Translated inputs with random ops.
6781,get_translation_matrix,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,583,function,"Returns projective transform(s) for the given translation(s).
Args:
translations: A matrix of 2-element lists representing [dx, dy] to translate
for each image (for a batch of images).
name: The name of the op.
Returns:
A tensor of shape (num_images, 8) projective transforms which can be given
to `transform`."
6782,transform,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,616,function,"Applies the given transform(s) to the image(s).
Args:
images: A tensor of shape (num_images, num_rows, num_columns, num_channels)
(NHWC), (num_rows, num_columns, num_channels) (HWC), or (num_rows,
num_columns) (HW). The rank must be statically known (the shape is not
`TensorShape(None)`.
transforms: Projective transform matrix/matrices. A vector of length 8 or
tensor of size N x 8. If one row of transforms is [a0, a1, a2, b0, b1, b2,
c0, c1], then it maps the *output* point `(x, y)` to a transformed *input*
point `(x', y') = ((a0 x + a1 y + a2) / k, (b0 x + b1 y + b2) / k)`, where
`k = c0 x + c1 y + 1`. The transforms are *inverted* compared to the
transform mapping input points to output points. Note that gradients are
not backpropagated into transformation parameters.
fill_mode: Points outside the boundaries of the input are filled according
to the given mode (one of `{'constant', 'reflect', 'wrap', 'nearest'}`).
interpolation: Interpolation mode. Supported values: ""nearest"", ""bilinear"".
output_shape: Output dimesion after the transform, [height, width]. If None,
output is the same size as input image.
name: The name of the op.
## Fill mode.
Behavior for each valid value is as follows:
reflect (d c b a | a b c d | d c b a)
The input is extended by reflecting about the edge of the last pixel.
constant (k k k k | a b c d | k k k k)
The input is extended by filling all values beyond the edge with the same
constant value k = 0.
wrap (a b c d | a b c d | a b c d)
The input is extended by wrapping around to the opposite edge.
nearest (a a a a | a b c d | d d d d)
The input is extended by the nearest pixel.
Input shape:
4D tensor with shape: `(samples, height, width, channels)`,
data_format='channels_last'.
Output shape:
4D tensor with shape: `(samples, height, width, channels)`,
data_format='channels_last'.
Returns:
Image(s) with the same type and shape as `images`, with the given
transform(s) applied. Transformed coordinates outside of the input image
will be filled with zeros.
Raises:
TypeError: If `image` is an invalid type.
ValueError: If output shape is not 1-D int32 Tensor."
6783,get_rotation_matrix,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,699,function,"Returns projective transform(s) for the given angle(s).
Args:
angles: A scalar angle to rotate all images by, or (for batches of images) a
vector with an angle to rotate each image in the batch. The rank must be
statically known (the shape is not `TensorShape(None)`).
image_height: Height of the image(s) to be transformed.
image_width: Width of the image(s) to be transformed.
name: The name of the op.
Returns:
A tensor of shape (num_images, 8). Projective transforms which can be given
to operation `image_projective_transform_v2`. If one row of transforms is
[a0, a1, a2, b0, b1, b2, c0, c1], then it maps the *output* point
`(x, y)` to a transformed *input* point
`(x', y') = ((a0 x + a1 y + a2) / k, (b0 x + b1 y + b2) / k)`,
where `k = c0 x + c1 y + 1`."
6784,RandomRotation,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,740,class,"Randomly rotate each image.
By default, random rotations are only applied during training.
At inference time, the layer does nothing. If you need to apply random
rotations at inference time, set `training` to True when calling the layer.
Input shape:
4D tensor with shape:
`(samples, height, width, channels)`, data_format='channels_last'.
Output shape:
4D tensor with shape:
`(samples, height, width, channels)`, data_format='channels_last'.
Attributes:
factor: a float represented as fraction of 2pi, or a tuple of size
2 representing lower and upper bound for rotating clockwise and
counter-clockwise. A positive values means rotating counter clock-wise,
while a negative value means clock-wise. When represented as a single
float, this value is used for both the upper and lower bound. For
instance, `factor=(-0.2, 0.3)` results in an output
rotation by a random amount in the range `[-20% * 2pi, 30% * 2pi]`.
`factor=0.2` results in an output rotating by a random amount in the range
`[-20% * 2pi, 20% * 2pi]`.
fill_mode: Points outside the boundaries of the input are filled according
to the given mode (one of `{'constant', 'reflect', 'wrap', 'nearest'}`).
- *reflect*: `(d c b a | a b c d | d c b a)`
The input is extended by reflecting about the edge of the last pixel.
- *constant*: `(k k k k | a b c d | k k k k)`
The input is extended by filling all values beyond the edge with the
same constant value k = 0.
- *wrap*: `(a b c d | a b c d | a b c d)`
The input is extended by wrapping around to the opposite edge.
- *nearest*: `(a a a a | a b c d | d d d d)`
The input is extended by the nearest pixel.
interpolation: Interpolation mode. Supported values: ""nearest"", ""bilinear"".
seed: Integer. Used to create a random seed.
name: A string, the name of the layer.
Input shape:
4D tensor with shape: `(samples, height, width, channels)`,
data_format='channels_last'.
Output shape:
4D tensor with shape: `(samples, height, width, channels)`,
data_format='channels_last'.
Raise:
ValueError: if either bound is not between [0, 1], or upper bound is
less than lower bound."
6785,call,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,818,method,
6786,compute_output_shape,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,843,method,
6787,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,846,method,
6788,random_rotated_inputs,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,822,method,Rotated inputs with random ops.
6789,RandomZoom,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,858,class,"Randomly zoom each image during training.
Arguments:
height_factor: a float represented as fraction of value, or a tuple
of size 2 representing lower and upper bound for zooming vertically.
When represented as a single float, this value is used for both the
upper and lower bound. A positive value means zooming out, while a
negative value means zooming in.
For instance, `height_factor=(0.2, 0.3)` result in an output zoomed out
by a random amount in the range [+20%, +30%].
`height_factor=(-0.3, -0.2)` result in an output zoomed in by a random
amount in the range [+20%, +30%].
width_factor: a float represented as fraction of value, or a tuple
of size 2 representing lower and upper bound for zooming horizontally.
When represented as a single float, this value is used for both the
upper and lower bound.
For instance, `width_factor=(0.2, 0.3)` result in an output zooming out
between 20% to 30%.
`width_factor=(-0.3, -0.2)` result in an output zooming in between 20%
to 30%. Defaults to `None`, i.e., zooming vertical and horizontal
directions by preserving the aspect ratio.
fill_mode: Points outside the boundaries of the input are filled according
to the given mode (one of `{'constant', 'reflect', 'wrap', 'nearest'}`).
- *reflect*: `(d c b a | a b c d | d c b a)`
The input is extended by reflecting about the edge of the last pixel.
- *constant*: `(k k k k | a b c d | k k k k)`
The input is extended by filling all values beyond the edge with the
same constant value k = 0.
- *wrap*: `(a b c d | a b c d | a b c d)`
The input is extended by wrapping around to the opposite edge.
- *nearest*: `(a a a a | a b c d | d d d d)`
The input is extended by the nearest pixel.
interpolation: Interpolation mode. Supported values: ""nearest"", ""bilinear"".
seed: Integer. Used to create a random seed.
name: A string, the name of the layer.
Example:
>>> input_img = np.random.random((32, 224, 224, 3))
>>> layer = tf.keras.layers.experimental.preprocessing.RandomZoom(.5, .2)
>>> out_img = layer(input_img)
>>> out_img.shape
TensorShape([32, 224, 224, 3])
Input shape:
4D tensor with shape:
`(samples, height, width, channels)`, data_format='channels_last'.
Output shape:
4D tensor with shape:
`(samples, height, width, channels)`, data_format='channels_last'.
Raise:
ValueError: if lower bound is not between [0, 1], or upper bound is
negative."
6790,call,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,960,method,
6791,compute_output_shape,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,994,method,
6792,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,997,method,
6793,random_zoomed_inputs,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,964,method,Zoomed inputs with random ops.
6794,get_zoom_matrix,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,1009,function,"Returns projective transform(s) for the given zoom(s).
Args:
zooms: A matrix of 2-element lists representing [zx, zy] to zoom
for each image (for a batch of images).
image_height: Height of the image(s) to be transformed.
image_width: Width of the image(s) to be transformed.
name: The name of the op.
Returns:
A tensor of shape (num_images, 8). Projective transforms which can be given
to operation `image_projective_transform_v2`. If one row of transforms is
[a0, a1, a2, b0, b1, b2, c0, c1], then it maps the *output* point
`(x, y)` to a transformed *input* point
`(x', y') = ((a0 x + a1 y + a2) / k, (b0 x + b1 y + b2) / k)`,
where `k = c0 x + c1 y + 1`."
6795,RandomContrast,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,1051,class,"Adjust the contrast of an image or images by a random factor.
Contrast is adjusted independently for each channel of each image during
training.
For each channel, this layer computes the mean of the image pixels in the
channel and then adjusts each component `x` of each pixel to
`(x - mean) * contrast_factor + mean`.
Input shape:
4D tensor with shape:
`(samples, height, width, channels)`, data_format='channels_last'.
Output shape:
4D tensor with shape:
`(samples, height, width, channels)`, data_format='channels_last'.
Attributes:
factor: a positive float represented as fraction of value, or a tuple of
size 2 representing lower and upper bound. When represented as a single
float, lower = upper. The contrast factor will be randomly picked between
[1.0 - lower, 1.0 + upper].
seed: Integer. Used to create a random seed.
name: A string, the name of the layer.
Raise:
ValueError: if lower bound is not between [0, 1], or upper bound is
negative."
6796,call,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,1097,method,
6797,compute_output_shape,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,1110,method,
6798,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,1113,method,
6799,random_contrasted_inputs,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,1101,method,
6800,RandomHeight,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,1123,class,"Randomly vary the height of a batch of images during training.
Adjusts the height of a batch of images by a random factor. The input
should be a 4-D tensor in the ""channels_last"" image data format.
By default, this layer is inactive during inference.
Arguments:
factor: A positive float (fraction of original height), or a tuple of size 2
representing lower and upper bound for resizing vertically. When
represented as a single float, this value is used for both the upper and
lower bound. For instance, `factor=(0.2, 0.3)` results in an output with
height changed by a random amount in the range `[20%, 30%]`.
`factor=(-0.2, 0.3)` results in an output with height changed by a random
amount in the range `[-20%, +30%]. `factor=0.2` results in an output with
height changed by a random amount in the range `[-20%, +20%]`.
interpolation: String, the interpolation method. Defaults to `bilinear`.
Supports `bilinear`, `nearest`, `bicubic`, `area`, `lanczos3`, `lanczos5`,
`gaussian`, `mitchellcubic`
seed: Integer. Used to create a random seed.
name: A string, the name of the layer.
Input shape:
4D tensor with shape: `(samples, height, width, channels)`
(data_format='channels_last').
Output shape:
4D tensor with shape: `(samples, random_height, width, channels)`."
6801,call,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,1181,method,
6802,compute_output_shape,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,1205,method,
6803,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,1210,method,
6804,random_height_inputs,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,1185,method,Inputs height-adjusted with random ops.
6805,RandomWidth,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,1221,class,"Randomly vary the width of a batch of images during training.
Adjusts the width of a batch of images by a random factor. The input
should be a 4-D tensor in the ""channels_last"" image data format.
By default, this layer is inactive during inference.
Arguments:
factor: A positive float (fraction of original height), or a tuple of size 2
representing lower and upper bound for resizing vertically. When
represented as a single float, this value is used for both the upper and
lower bound. For instance, `factor=(0.2, 0.3)` results in an output with
width changed by a random amount in the range `[20%, 30%]`.
`factor=(-0.2, 0.3)` results in an output with width changed by a random
amount in the range `[-20%, +30%]. `factor=0.2` results in an output with
width changed by a random amount in the range `[-20%, +20%]`.
interpolation: String, the interpolation method. Defaults to `bilinear`.
Supports `bilinear`, `nearest`, `bicubic`, `area`, `lanczos3`, `lanczos5`,
`gaussian`, `mitchellcubic`
seed: Integer. Used to create a random seed.
name: A string, the name of the layer.
Input shape:
4D tensor with shape:
`(samples, height, width, channels)` (data_format='channels_last').
Output shape:
4D tensor with shape:
`(samples, height, random_width, channels)`."
6806,call,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,1280,method,
6807,compute_output_shape,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,1304,method,
6808,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,1309,method,
6809,random_width_inputs,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,1284,method,Inputs width-adjusted with random ops.
6810,make_generator,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,1352,function,
6811,get_interpolation,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing.py,1359,function,
6812,get_numpy_center_crop,tensorflow/tensorflow/python/keras/layers/preprocessing/image_preprocessing_test.py,131,function,
6813,IndexLookup,tensorflow/tensorflow/python/keras/layers/preprocessing/index_lookup.py,44,class,"Maps values from a vocabulary to integer indices.
This layer translates a set of arbitrary hashables into an integer output via
a table-based lookup, with optional out-of-vocabulary handling. This is the
basis layer for both IntegerLookup and IndexLookup; it holds the common
logic but is not intended to be exported as part of the Keras API.
If desired, the user can call this layer's `adapt()` method on a data set,
which will analyze the data set, determine the frequency of individual string
values, and create a vocabulary from them. This vocabulary can have
unlimited size or be capped, depending on the configuration options for this
layer; if there are more unique values in the input than the maximum
vocabulary size, the most frequent terms will be used to create the
vocabulary.
Attributes:
max_tokens: The maximum size of the vocabulary for this layer. If None,
there is no cap on the size of the vocabulary. Note that this vocabulary
includes the OOV and mask tokens, so the effective number of tokens is
(max_tokens - num_oov_indices - (1 if mask_token else 0))
num_oov_indices: The number of out-of-vocabulary tokens to use. If this
value is more than 1, OOV inputs are hashed to determine their OOV value;
if this value is 0, passing an OOV input will result in a '-1' being
returned for that value in the output tensor. (Note that, because the
value is -1 and not 0, this will allow you to effectively drop OOV values
from categorical encodings.)
mask_token: A token that represents masked values, and which is mapped to
index 0. If set to None, no mask term will be added and the OOV tokens, if
any, will be indexed from (0...num_oov_indices) instead of
(1...num_oov_indices+1).
oov_token: The token representing an out-of-vocabulary value. This token is
only used when performing an inverse lookup.
vocabulary: An optional list of vocabulary terms. If the list contains the
same token multiple times, an error will be thrown.
invert: If true, this layer will map indices to vocabulary items instead
of mapping vocabulary items to indices."
6814,compute_output_shape,tensorflow/tensorflow/python/keras/layers/preprocessing/index_lookup.py,170,method,
6815,compute_output_signature,tensorflow/tensorflow/python/keras/layers/preprocessing/index_lookup.py,173,method,
6816,adapt,tensorflow/tensorflow/python/keras/layers/preprocessing/index_lookup.py,178,method,"Fits the state of the preprocessing layer to the dataset.
Overrides the default adapt method to apply relevant preprocessing to the
inputs before passing to the combiner.
Arguments:
data: The data to train on. It can be passed either as a tf.data Dataset,
or as a numpy array.
reset_state: Optional argument specifying whether to clear the state of
the layer at the start of the call to `adapt`. This must be True for
this layer, which does not support repeated calls to `adapt`."
6817,get_vocabulary,tensorflow/tensorflow/python/keras/layers/preprocessing/index_lookup.py,195,method,
6818,vocab_size,tensorflow/tensorflow/python/keras/layers/preprocessing/index_lookup.py,208,method,
6819,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/index_lookup.py,211,method,
6820,count_params,tensorflow/tensorflow/python/keras/layers/preprocessing/index_lookup.py,222,method,
6821,set_vocabulary,tensorflow/tensorflow/python/keras/layers/preprocessing/index_lookup.py,345,method,"Sets vocabulary data for this layer with inverse=False.
This method sets the vocabulary for this layer directly, instead of
analyzing a dataset through 'adapt'. It should be used whenever the vocab
information is already known. If vocabulary data is already present in the
layer, this method will either replace it
Arguments:
vocab: An array of string tokens.
Raises:
ValueError: If there are too many inputs, the inputs do not match, or
input data is missing."
6822,call,tensorflow/tensorflow/python/keras/layers/preprocessing/index_lookup.py,370,method,
6823,get_layer_class,tensorflow/tensorflow/python/keras/layers/preprocessing/index_lookup_distribution_test.py,37,function,
6824,get_layer_class,tensorflow/tensorflow/python/keras/layers/preprocessing/index_lookup_test.py,48,function,
6825,IndexLookup,tensorflow/tensorflow/python/keras/layers/preprocessing/index_lookup_v1.py,26,class,"IndexLookup layer.
This layer translates a set of arbitray strings or integers into an integer
output via a table-based lookup, with optional out-of-vocabulary handling.
If desired, the user can call this layer's adapt() method on a data set.
When this layer is adapted, it will analyze the dataset, determine the
frequency of individual string or integer values, and create a vocabulary
from them. This vocabulary can have unlimited size or be capped, depending on
the configuration options for this layer; if there are more unique values in
the input than the maximum vocabulary size, the most frequent terms will be
used to create the vocabulary.
Attributes:
max_vocab_size: The maximum size of the vocabulary for this layer. If None,
there is no cap on the size of the vocabulary. Note that the vocabulary
does include OOV buckets, so the effective number of unique values in the
vocabulary is (max_vocab_size - num_oov_buckets) when this value is set.
num_oov_buckets: The number of out-of-vocabulary tokens to use; defaults to
1. If this value is more than 1, OOV inputs are hashed to determine their
OOV value; if this value is 0, passing an OOV input will result in a
runtime error.
reserve_zero: Whether to reserve the index '0', which has a special meaning
in the Keras masking system. If True, the output of this layer will be in
the range [1...max_vocab_size+1); if False, the output will be in the
range [0...max_vocab_size). Defaults to True.
mask_inputs: If True, input values of 0 (for integers) and """" (for strings)
will be treated as masked values and assigned an output value of 0. If
this option is set, reserve_zero must also be set. Defaults to False."
6826,IntegerLookup,tensorflow/tensorflow/python/keras/layers/preprocessing/integer_lookup.py,28,class,"Maps integers from a vocabulary to integer indices.
This layer translates a set of arbitrary integers into an integer output via a
table-based lookup, with optional out-of-vocabulary handling.
If desired, the user can call this layer's `adapt()` method on a data set,
which will analyze the data set, determine the frequency of individual string
values, and create a vocabulary from them. This vocabulary can have
unlimited size or be capped, depending on the configuration options for this
layer; if there are more unique values in the input than the maximum
vocabulary size, the most frequent terms will be used to create the
vocabulary.
Attributes:
max_values: The maximum size of the vocabulary for this layer. If None,
there is no cap on the size of the vocabulary. Note that this vocabulary
includes the OOV and mask values, so the effective number of values is
(max_values - num_oov_values - (1 if mask_token else 0))
num_oov_indices: The number of out-of-vocabulary values to use; defaults to
1. If this value is more than 1, OOV inputs are modulated to determine
their OOV value; if this value is 0, passing an OOV input will result in
a '-1' being returned for that value in the output tensor. (Note that,
because the value is -1 and not 0, this will allow you to effectively drop
OOV values from categorical encodings.)
mask_value: A value that represents masked inputs, and which is mapped to
index 0. Defaults to 0. If set to None, no mask term will be added and the
OOV values, if any, will be indexed from (0...num_oov_values) instead of
(1...num_oov_values+1).
oov_value: The value representing an out-of-vocabulary value. Defaults to
-1.
vocabulary: An optional list of values, or a path to a text file containing
a vocabulary to load into this layer. The file should contain one value
per line. If the list or file contains the same token multiple times, an
error will be thrown.
invert: If true, this layer will map indices to vocabulary items instead
of mapping vocabulary items to indices.
Examples:
Creating a lookup layer with a known vocabulary
This example creates a lookup layer with a pre-existing vocabulary.
>>> vocab = [12, 36, 1138, 42]
>>> data = tf.constant([[12, 1138, 42], [42, 1000, 36]])
>>> layer = IntegerLookup(vocabulary=vocab)
>>> layer(data)
<tf.Tensor: shape=(2, 3), dtype=int64, numpy=
array([[2, 4, 5],
[5, 1, 3]])>
Creating a lookup layer with an adapted vocabulary
This example creates a lookup layer and generates the vocabulary by analyzing
the dataset.
>>> data = tf.constant([[12, 1138, 42], [42, 1000, 36]])
>>> layer = IntegerLookup()
>>> layer.adapt(data)
>>> layer.get_vocabulary()
[0, -1, 42, 1138, 1000, 36, 12]
Note how the mask value 0 and the OOV value -1 have been added to the
vocabulary. The remaining values are sorted by frequency (1138, which has
2 occurrences, is first) then by inverse sort order.
>>> data = tf.constant([[12, 1138, 42], [42, 1000, 36]])
>>> layer = IntegerLookup()
>>> layer.adapt(data)
>>> layer(data)
<tf.Tensor: shape=(2, 3), dtype=int64, numpy=
array([[6, 3, 2],
[2, 4, 5]])>
Lookups with multiple OOV tokens.
This example demonstrates how to use a lookup layer with multiple OOV tokens.
When a layer is created with more than one OOV token, any OOV values are
hashed into the number of OOV buckets, distributing OOV values in a
deterministic fashion across the set.
>>> vocab = [12, 36, 1138, 42]
>>> data = tf.constant([[12, 1138, 42], [37, 1000, 36]])
>>> layer = IntegerLookup(vocabulary=vocab, num_oov_indices=2)
>>> layer(data)
<tf.Tensor: shape=(2, 3), dtype=int64, numpy=
array([[3, 5, 6],
[2, 1, 4]])>
Note that the output for OOV value 37 is 2, while the output for OOV value
1000 is 1. The in-vocab terms have their output index increased by 1 from
earlier examples (12 maps to 3, etc) in order to make space for the extra OOV
value.
Inverse lookup
This example demonstrates how to map indices to values using this layer. (You
can also use adapt() with inverse=True, but for simplicity we'll pass the
vocab in this example.)
>>> vocab = [12, 36, 1138, 42]
>>> data = tf.constant([[1, 3, 4], [4, 5, 2]])
>>> layer = IntegerLookup(vocabulary=vocab, invert=True)
>>> layer(data)
<tf.Tensor: shape=(2, 3), dtype=int64, numpy=
array([[ 12, 1138, 42],
[ 42, -1, 36]])>
Note that the integer 5, which is out of the vocabulary space, returns an OOV
token.
Forward and inverse lookup pairs
This example demonstrates how to use the vocabulary of a standard lookup
layer to create an inverse lookup layer.
>>> vocab = [12, 36, 1138, 42]
>>> data = tf.constant([[12, 1138, 42], [42, 1000, 36]])
>>> layer = IntegerLookup(vocabulary=vocab)
>>> i_layer = IntegerLookup(vocabulary=layer.get_vocabulary(), invert=True)
>>> int_data = layer(data)
>>> i_layer(int_data)
<tf.Tensor: shape=(2, 3), dtype=int64, numpy=
array([[ 12, 1138, 42],
[ 42, -1, 36]])>
In this example, the input value 1000 resulted in an output of -1, since
1000 was not in the vocabulary - it got represented as an OOV, and all OOV
values are returned as -1 in the inverse layer. Also, note that for the
inverse to work, you must have already set the forward layer vocabulary
either directly or via fit() before calling get_vocabulary()."
6827,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/integer_lookup.py,207,method,
6828,get_layer_class,tensorflow/tensorflow/python/keras/layers/preprocessing/integer_lookup_test.py,48,function,
6829,IntegerLookup,tensorflow/tensorflow/python/keras/layers/preprocessing/integer_lookup_v1.py,27,class,Maps integers from a vocabulary to integer indices.
6830,Normalization,tensorflow/tensorflow/python/keras/layers/preprocessing/normalization.py,42,class,"Feature-wise normalization of the data.
This layer will coerce its inputs into a distribution centered around
0 with standard deviation 1. It accomplishes this by precomputing the mean and
variance of the data, and calling (input-mean)/sqrt(var) at runtime.
What happens in `adapt`: Compute mean and variance of the data and store them
as the layer's weights. `adapt` should be called before `fit`, `evaluate`,
or `predict`.
Attributes:
axis: Integer or tuple of integers, the axis or axes that should be
""kept"". These axes are not be summed over when calculating the
normalization statistics. By default the last axis, the `features` axis
is kept and any `space` or `time` axes are summed. Each element in the
the axes that are kept is normalized independently. If `axis` is set to
'None', the layer will perform scalar normalization (diving the input
by a single scalar value). The `batch` axis, 0, is always summed over
(`axis=0` is not allowed).
Examples:
Calculate the mean and variance by analyzing the dataset in `adapt`.
>>> adapt_data = np.array([[1.], [2.], [3.], [4.], [5.]], dtype=np.float32)
>>> input_data = np.array([[1.], [2.], [3.]], np.float32)
>>> layer = Normalization()
>>> layer.adapt(adapt_data)
>>> layer(input_data)
<tf.Tensor: shape=(3, 1), dtype=float32, numpy=
array([[-1.4142135 ],
[-0.70710677],
[ 0. ]], dtype=float32)>"
6831,build,tensorflow/tensorflow/python/keras/layers/preprocessing/normalization.py,100,method,
6832,call,tensorflow/tensorflow/python/keras/layers/preprocessing/normalization.py,147,method,
6833,compute_output_shape,tensorflow/tensorflow/python/keras/layers/preprocessing/normalization.py,162,method,
6834,compute_output_signature,tensorflow/tensorflow/python/keras/layers/preprocessing/normalization.py,165,method,
6835,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/normalization.py,168,method,
6836,set_weights,tensorflow/tensorflow/python/keras/layers/preprocessing/normalization.py,173,method,Override for set_weights to ensure we can set just mean/var weights.
6837,get_layer_class,tensorflow/tensorflow/python/keras/layers/preprocessing/normalization_distribution_test.py,35,function,
6838,get_layer_class,tensorflow/tensorflow/python/keras/layers/preprocessing/normalization_test.py,37,function,
6839,Normalization,tensorflow/tensorflow/python/keras/layers/preprocessing/normalization_v1.py,28,class,
6840,PreprocessingStage,tensorflow/tensorflow/python/keras/layers/preprocessing/preprocessing_stage.py,30,class,"A sequential preprocessing stage.
This preprocessing stage wraps a list of preprocessing layers into a
Sequential-like object that enables you to `adapt()` the whole list via
a single `adapt()` call on the preprocessing stage.
Arguments:
layers: List of layers. Can include layers that aren't preprocessing layers.
name: String. Optional name for the preprocessing stage object."
6841,adapt,tensorflow/tensorflow/python/keras/layers/preprocessing/preprocessing_stage.py,43,method,"Adapt the state of the layers of the preprocessing stage to the data.
Arguments:
data: A batched Dataset object, or a NumPy array, or an EagerTensor.
Data to be iterated over to adapt the state of the layers in this
preprocessing stage.
reset_state: Whether this call to `adapt` should reset the state of
the layers in this preprocessing stage."
6842,map_fn,tensorflow/tensorflow/python/keras/layers/preprocessing/preprocessing_stage.py,73,method,"Maps `PreprocessingStage` inputs to inputs at `current_layer_index`.
Args:
x: Batch of inputs seen in entry of the `PreprocessingStage` instance.
Returns:
Batch of inputs to be processed by layer
`self.layers[current_layer_index]`"
6843,get_reduce_op,tensorflow/tensorflow/python/keras/layers/preprocessing/reduction.py,28,function,Translate a reduction string name to a reduction op.
6844,Reduction,tensorflow/tensorflow/python/keras/layers/preprocessing/reduction.py,45,class,"Performs an optionally-weighted reduction.
This layer performs a reduction across one axis of its input data. This
data may optionally be weighted by passing in an identical float tensor.
Arguments:
reduction: The type of reduction to perform. Can be one of the following:
""max"", ""mean"", ""min"", ""prod"", or ""sum"". This layer uses the Tensorflow
reduce op which corresponds to that reduction (so, for ""mean"", we use
""reduce_mean"").
axis: The axis to reduce along. Defaults to '-2', which is usually the axis
that contains embeddings (but is not within the embedding itself).
Input shape:
A tensor of 2 or more dimensions of any numeric dtype.
Output:
A tensor of 1 less dimension than the input tensor, of the same dtype.
Call arguments:
inputs: The data to reduce.
weights: An optional tensor or constant of the same shape as inputs that
will weight the input data before it is reduced."
6845,call,tensorflow/tensorflow/python/keras/layers/preprocessing/reduction.py,79,method,
6846,StringLookup,tensorflow/tensorflow/python/keras/layers/preprocessing/string_lookup.py,28,class,"Maps strings from a vocabulary to integer indices.
This layer translates a set of arbitrary strings into an integer output via a
table-based lookup, with optional out-of-vocabulary handling.
If desired, the user can call this layer's `adapt()` method on a data set,
which will analyze the data set, determine the frequency of individual string
values, and create a vocabulary from them. This vocabulary can have
unlimited size or be capped, depending on the configuration options for this
layer; if there are more unique values in the input than the maximum
vocabulary size, the most frequent terms will be used to create the
vocabulary.
Attributes:
max_tokens: The maximum size of the vocabulary for this layer. If None,
there is no cap on the size of the vocabulary. Note that this vocabulary
includes the OOV and mask tokens, so the effective number of tokens is
(max_tokens - num_oov_indices - (1 if mask_token else 0))
num_oov_indices: The number of out-of-vocabulary tokens to use; defaults to
1. If this value is more than 1, OOV inputs are hashed to determine their
OOV value; if this value is 0, passing an OOV input will result in a '-1'
being returned for that value in the output tensor. (Note that, because
the value is -1 and not 0, this will allow you to effectively drop OOV
values from categorical encodings.)
mask_token: A token that represents masked values, and which is mapped to
index 0. Defaults to the empty string """". If set to None, no mask term
will be added and the OOV tokens, if any, will be indexed from
(0...num_oov_indices) instead of (1...num_oov_indices+1).
oov_token: The token representing an out-of-vocabulary value. Defaults to
""[UNK]"".
vocabulary: An optional list of vocabulary terms, or a path to a text file
containing a vocabulary to load into this layer. The file should contain
one token per line. If the list or file contains the same token multiple
times, an error will be thrown.
encoding: The Python string encoding to use. Defaults to `'utf-8'`.
invert: If true, this layer will map indices to vocabulary items instead
of mapping vocabulary items to indices.
Examples:
Creating a lookup layer with a known vocabulary
This example creates a lookup layer with a pre-existing vocabulary.
>>> vocab = [""a"", ""b"", ""c"", ""d""]
>>> data = tf.constant([[""a"", ""c"", ""d""], [""d"", ""z"", ""b""]])
>>> layer = StringLookup(vocabulary=vocab)
>>> layer(data)
<tf.Tensor: shape=(2, 3), dtype=int64, numpy=
array([[2, 4, 5],
[5, 1, 3]])>
Creating a lookup layer with an adapted vocabulary
This example creates a lookup layer and generates the vocabulary by analyzing
the dataset.
>>> data = tf.constant([[""a"", ""c"", ""d""], [""d"", ""z"", ""b""]])
>>> layer = StringLookup()
>>> layer.adapt(data)
>>> layer.get_vocabulary()
['', '[UNK]', 'd', 'z', 'c', 'b', 'a']
Note how the mask token '' and the OOV token [UNK] have been added to the
vocabulary. The remaining tokens are sorted by frequency ('d', which has
2 occurrences, is first) then by inverse sort order.
>>> data = tf.constant([[""a"", ""c"", ""d""], [""d"", ""z"", ""b""]])
>>> layer = StringLookup()
>>> layer.adapt(data)
>>> layer(data)
<tf.Tensor: shape=(2, 3), dtype=int64, numpy=
array([[6, 4, 2],
[2, 3, 5]])>
Lookups with multiple OOV tokens.
This example demonstrates how to use a lookup layer with multiple OOV tokens.
When a layer is created with more than one OOV token, any OOV values are
hashed into the number of OOV buckets, distributing OOV values in a
deterministic fashion across the set.
>>> vocab = [""a"", ""b"", ""c"", ""d""]
>>> data = tf.constant([[""a"", ""c"", ""d""], [""m"", ""z"", ""b""]])
>>> layer = StringLookup(vocabulary=vocab, num_oov_indices=2)
>>> layer(data)
<tf.Tensor: shape=(2, 3), dtype=int64, numpy=
array([[3, 5, 6],
[1, 2, 4]])>
Note that the output for OOV value 'm' is 1, while the output for OOV value
'z' is 2. The in-vocab terms have their output index increased by 1 from
earlier examples (a maps to 3, etc) in order to make space for the extra OOV
value.
Inverse lookup
This example demonstrates how to map indices to strings using this layer. (You
can also use adapt() with inverse=True, but for simplicity we'll pass the
vocab in this example.)
>>> vocab = [""a"", ""b"", ""c"", ""d""]
>>> data = tf.constant([[1, 3, 4], [4, 5, 2]])
>>> layer = StringLookup(vocabulary=vocab, invert=True)
>>> layer(data)
<tf.Tensor: shape=(2, 3), dtype=string, numpy=
array([[b'a', b'c', b'd'],
[b'd', b'[UNK]', b'b']], dtype=object)>
Note that the integer 5, which is out of the vocabulary space, returns an OOV
token.
Forward and inverse lookup pairs
This example demonstrates how to use the vocabulary of a standard lookup
layer to create an inverse lookup layer.
>>> vocab = [""a"", ""b"", ""c"", ""d""]
>>> data = tf.constant([[""a"", ""c"", ""d""], [""d"", ""z"", ""b""]])
>>> layer = StringLookup(vocabulary=vocab)
>>> i_layer = StringLookup(vocabulary=layer.get_vocabulary(), invert=True)
>>> int_data = layer(data)
>>> i_layer(int_data)
<tf.Tensor: shape=(2, 3), dtype=string, numpy=
array([[b'a', b'c', b'd'],
[b'd', b'[UNK]', b'b']], dtype=object)>
In this example, the input value 'z' resulted in an output of '[UNK]', since
1000 was not in the vocabulary - it got represented as an OOV, and all OOV
values are returned as '[OOV}' in the inverse layer. Also, note that for the
inverse to work, you must have already set the forward layer vocabulary
either directly or via fit() before calling get_vocabulary()."
6847,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/string_lookup.py,202,method,
6848,get_vocabulary,tensorflow/tensorflow/python/keras/layers/preprocessing/string_lookup.py,207,method,
6849,get_layer_class,tensorflow/tensorflow/python/keras/layers/preprocessing/string_lookup_test.py,44,function,
6850,StringLookup,tensorflow/tensorflow/python/keras/layers/preprocessing/string_lookup_v1.py,27,class,Maps strings from a vocabulary to integer indices.
6851,TableHandler,tensorflow/tensorflow/python/keras/layers/preprocessing/table_utils.py,36,class,Wrapper object that holds a lookup table and provides accessors.
6852,data,tensorflow/tensorflow/python/keras/layers/preprocessing/table_utils.py,49,method,
6853,vocab_size,tensorflow/tensorflow/python/keras/layers/preprocessing/table_utils.py,53,method,
6854,clear,tensorflow/tensorflow/python/keras/layers/preprocessing/table_utils.py,56,method,
6855,insert,tensorflow/tensorflow/python/keras/layers/preprocessing/table_utils.py,60,method,
6856,lookup,tensorflow/tensorflow/python/keras/layers/preprocessing/table_utils.py,123,method,Perform a table lookup.
6857,get_vocabulary_from_file,tensorflow/tensorflow/python/keras/layers/preprocessing/table_utils.py,151,function,Read a vocabulary in from a file.
6858,validate_vocabulary_is_unique,tensorflow/tensorflow/python/keras/layers/preprocessing/table_utils.py,171,function,Validate that a vocabulary contains no repeated tokens.
6859,assert_same_type,tensorflow/tensorflow/python/keras/layers/preprocessing/table_utils.py,184,function,Assert that 'values' is of type 'expected_type'.
6860,convert_to_ndarray,tensorflow/tensorflow/python/keras/layers/preprocessing/table_utils.py,191,function,Convert 'x' to a numpy array.
6861,get_table,tensorflow/tensorflow/python/keras/layers/preprocessing/table_utils_test.py,34,function,
6862,TextVectorization,tensorflow/tensorflow/python/keras/layers/preprocessing/text_vectorization.py,74,class,"Text vectorization layer.
This layer has basic options for managing text in a Keras model. It
transforms a batch of strings (one sample = one string) into either a list of
token indices (one sample = 1D tensor of integer token indices) or a dense
representation (one sample = 1D tensor of float values representing data about
the sample's tokens).
If desired, the user can call this layer's adapt() method on a dataset.
When this layer is adapted, it will analyze the dataset, determine the
frequency of individual string values, and create a 'vocabulary' from them.
This vocabulary can have unlimited size or be capped, depending on the
configuration options for this layer; if there are more unique values in the
input than the maximum vocabulary size, the most frequent terms will be used
to create the vocabulary.
The processing of each sample contains the following steps:
1. standardize each sample (usually lowercasing + punctuation stripping)
2. split each sample into substrings (usually words)
3. recombine substrings into tokens (usually ngrams)
4. index tokens (associate a unique int value with each token)
5. transform each sample using this index, either into a vector of ints or
a dense float vector.
Some notes on passing Callables to customize splitting and normalization for
this layer:
1. Any callable can be passed to this Layer, but if you want to serialize
this object you should only pass functions that are registered Keras
serializables (see `tf.keras.utils.register_keras_serializable` for more
details).
2. When using a custom callable for `standardize`, the data received
by the callable will be exactly as passed to this layer. The callable
should return a tensor of the same shape as the input.
3. When using a custom callable for `split`, the data received by the
callable will have the 1st dimension squeezed out - instead of
`[[""string to split""], [""another string to split""]]`, the Callable will
see `[""string to split"", ""another string to split""]`. The callable should
return a Tensor with the first dimension containing the split tokens -
in this example, we should see something like `[[""string"", ""to"", ""split],
[""another"", ""string"", ""to"", ""split""]]`. This makes the callable site
natively compatible with `tf.strings.split()`.
Attributes:
max_tokens: The maximum size of the vocabulary for this layer. If None,
there is no cap on the size of the vocabulary. Note that this vocabulary
contains 1 OOV token, so the effective number of tokens is `(max_tokens -
1 - (1 if output == ""int"" else 0))`.
standardize: Optional specification for standardization to apply to the
input text. Values can be None (no standardization),
'lower_and_strip_punctuation' (lowercase and remove punctuation) or a
Callable. Default is 'lower_and_strip_punctuation'.
split: Optional specification for splitting the input text. Values can be
None (no splitting), 'whitespace' (split on ASCII whitespace), or a
Callable. The default is 'whitespace'.
ngrams: Optional specification for ngrams to create from the possibly-split
input text. Values can be None, an integer or tuple of integers; passing
an integer will create ngrams up to that integer, and passing a tuple of
integers will create ngrams for the specified values in the tuple. Passing
None means that no ngrams will be created.
output_mode: Optional specification for the output of the layer. Values can
be ""int"", ""binary"", ""count"" or ""tf-idf"", configuring the layer as follows:
""int"": Outputs integer indices, one integer index per split string
token. When output == ""int"", 0 is reserved for masked locations;
this reduces the vocab size to max_tokens-2 instead of max_tokens-1
""binary"": Outputs a single int array per batch, of either vocab_size or
max_tokens size, containing 1s in all elements where the token mapped
to that index exists at least once in the batch item.
""count"": As ""binary"", but the int array contains a count of the number
of times the token at that index appeared in the batch item.
""tf-idf"": As ""binary"", but the TF-IDF algorithm is applied to find the
value in each token slot.
output_sequence_length: Only valid in INT mode. If set, the output will have
its time dimension padded or truncated to exactly `output_sequence_length`
values, resulting in a tensor of shape [batch_size,
output_sequence_length] regardless of how many tokens resulted from the
splitting step. Defaults to None.
pad_to_max_tokens: Only valid in ""binary"", ""count"", and ""tf-idf"" modes. If
True, the output will have its feature axis padded to `max_tokens` even if
the number of unique tokens in the vocabulary is less than max_tokens,
resulting in a tensor of shape [batch_size, max_tokens] regardless of
vocabulary size. Defaults to True.
Example:
This example instantiates a TextVectorization layer that lowercases text,
splits on whitespace, strips punctuation, and outputs integer vocab indices.
>>> text_dataset = tf.data.Dataset.from_tensor_slices([""foo"", ""bar"", ""baz""])
>>> max_features = 5000 # Maximum vocab size.
>>> max_len = 4 # Sequence length to pad the outputs to.
>>> embedding_dims = 2
>>>
>>> # Create the layer.
>>> vectorize_layer = TextVectorization(
... max_tokens=max_features,
... output_mode='int',
... output_sequence_length=max_len)
>>>
>>> # Now that the vocab layer has been created, call `adapt` on the text-only
>>> # dataset to create the vocabulary. You don't have to batch, but for large
>>> # datasets this means we're not keeping spare copies of the dataset.
>>> vectorize_layer.adapt(text_dataset.batch(64))
>>>
>>> # Create the model that uses the vectorize text layer
>>> model = tf.keras.models.Sequential()
>>>
>>> # Start by creating an explicit input layer. It needs to have a shape of
>>> # (1,) (because we need to guarantee that there is exactly one string
>>> # input per batch), and the dtype needs to be 'string'.
>>> model.add(tf.keras.Input(shape=(1,), dtype=tf.string))
>>>
>>> # The first layer in our model is the vectorization layer. After this
>>> # layer, we have a tensor of shape (batch_size, max_len) containing vocab
>>> # indices.
>>> model.add(vectorize_layer)
>>>
>>> # Now, the model can map strings to integers, and you can add an embedding
>>> # layer to map these integers to learned embeddings.
>>> input_data = [[""foo qux bar""], [""qux baz""]]
>>> model.predict(input_data)
array([[2, 1, 4, 0],
[1, 3, 0, 0]])"
6863,compute_output_shape,tensorflow/tensorflow/python/keras/layers/preprocessing/text_vectorization.py,328,method,
6864,compute_output_signature,tensorflow/tensorflow/python/keras/layers/preprocessing/text_vectorization.py,345,method,
6865,adapt,tensorflow/tensorflow/python/keras/layers/preprocessing/text_vectorization.py,350,method,"Fits the state of the preprocessing layer to the dataset.
Overrides the default adapt method to apply relevant preprocessing to the
inputs before passing to the combiner.
Arguments:
data: The data to train on. It can be passed either as a tf.data Dataset,
as a NumPy array, a string tensor, or as a list of texts.
reset_state: Optional argument specifying whether to clear the state of
the layer at the start of the call to `adapt`. This must be True for
this layer, which does not support repeated calls to `adapt`."
6866,get_vocabulary,tensorflow/tensorflow/python/keras/layers/preprocessing/text_vectorization.py,403,method,
6867,get_config,tensorflow/tensorflow/python/keras/layers/preprocessing/text_vectorization.py,406,method,
6868,count_params,tensorflow/tensorflow/python/keras/layers/preprocessing/text_vectorization.py,419,method,
6869,set_vocabulary,tensorflow/tensorflow/python/keras/layers/preprocessing/text_vectorization.py,426,method,"Sets vocabulary (and optionally document frequency) data for this layer.
This method sets the vocabulary and DF data for this layer directly, instead
of analyzing a dataset through 'adapt'. It should be used whenever the vocab
(and optionally document frequency) information is already known. If
vocabulary data is already present in the layer, this method will replace
it.
Arguments:
vocab: An array of string tokens.
df_data: An array of document frequency data. Only necessary if the layer
output_mode is TFIDF.
oov_df_value: The document frequency of the OOV token. Only necessary if
output_mode is TFIDF.
Raises:
ValueError: If there are too many inputs, the inputs do not match, or
input data is missing.
RuntimeError: If the vocabulary cannot be set when this function is
called. This happens when ""binary"", ""count"", and ""tfidf"" modes,
if ""pad_to_max_tokens"" is False and the layer itself has already been
called."
6870,build,tensorflow/tensorflow/python/keras/layers/preprocessing/text_vectorization.py,492,method,
6871,call,tensorflow/tensorflow/python/keras/layers/preprocessing/text_vectorization.py,567,method,
6872,get_layer_class,tensorflow/tensorflow/python/keras/layers/preprocessing/text_vectorization_distribution_test.py,37,function,
6873,get_layer_class,tensorflow/tensorflow/python/keras/layers/preprocessing/text_vectorization_test.py,50,function,
6874,custom_standardize_fn,tensorflow/tensorflow/python/keras/layers/preprocessing/text_vectorization_test.py,1427,function,
6875,custom_split_fn,tensorflow/tensorflow/python/keras/layers/preprocessing/text_vectorization_test.py,1432,function,
6876,TextVectorization,tensorflow/tensorflow/python/keras/layers/preprocessing/text_vectorization_v1.py,30,class,"Text vectorization layer.
This layer has basic options for managing text in a Keras model. It
transforms a batch of strings (one sample = one string) into either a list of
token indices (one sample = 1D tensor of integer token indices) or a dense
representation (one sample = 1D tensor of float values representing data about
the sample's tokens).
The processing of each sample contains the following steps:
1) standardize each sample (usually lowercasing + punctuation stripping)
2) split each sample into substrings (usually words)
3) recombine substrings into tokens (usually ngrams)
4) index tokens (associate a unique int value with each token)
5) transform each sample using this index, either into a vector of ints or
a dense float vector.
Attributes:
max_tokens: The maximum size of the vocabulary for this layer. If None,
there is no cap on the size of the vocabulary.
standardize: Optional specification for standardization to apply to the
input text. Values can be None (no standardization),
LOWER_AND_STRIP_PUNCTUATION (lowercase and remove punctuation) or a
Callable.
split: Optional specification for splitting the input text. Values can be
None (no splitting), SPLIT_ON_WHITESPACE (split on ASCII whitespace), or a
Callable.
ngrams: Optional specification for ngrams to create from the possibly-split
input text. Values can be None, an integer or tuple of integers; passing
an integer will create ngrams up to that integer, and passing a tuple of
integers will create ngrams for the specified values in the tuple. Passing
None means that no ngrams will be created.
output_mode: Optional specification for the output of the layer. Values can
be INT, BINARY, COUNT or TFIDF, which control the outputs as follows:
INT: Outputs integer indices, one integer index per split string token.
BINARY: Outputs a single int array per batch, of either vocab_size or
max_tokens size, containing 1s in all elements where the token mapped
to that index exists at least once in the batch item.
COUNT: As BINARY, but the int array contains a count of the number of
times the token at that index appeared in the batch item.
TFIDF: As BINARY, but the TF-IDF algorithm is applied to find the value
in each token slot.
output_sequence_length: Optional length for the output tensor. If set, the
output will be padded or truncated to this value in INT mode.
pad_to_max_tokens: If True, BINARY, COUNT, and TFIDF modes will have their
outputs padded to max_tokens, even if the number of unique tokens in the
vocabulary is less than max_tokens."
6877,int_gen,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/category_crossing_benchmark.py,43,function,
6878,BenchmarkLayer,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/category_crossing_benchmark.py,48,class,Benchmark the layer forward pass.
6879,run_dataset_implementation,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/category_crossing_benchmark.py,51,method,
6880,bm_layer_implementation,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/category_crossing_benchmark.py,74,method,
6881,benchmark_vocab_size_by_batch,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/category_crossing_benchmark.py,110,method,
6882,BenchmarkLayer,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/category_encoding_benchmark.py,39,class,Benchmark the layer forward pass.
6883,run_dataset_implementation,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/category_encoding_benchmark.py,42,method,
6884,benchmark_vocab_size_by_batch,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/category_encoding_benchmark.py,75,method,
6885,word_gen,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/hashing_benchmark.py,45,function,
6886,BenchmarkLayer,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/hashing_benchmark.py,50,class,Benchmark the layer forward pass.
6887,run_dataset_implementation,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/hashing_benchmark.py,53,method,
6888,bm_layer_implementation,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/hashing_benchmark.py,75,method,
6889,benchmark_vocab_size_by_batch,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/hashing_benchmark.py,109,method,
6890,rotate,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/image_preproc_benchmark.py,48,function,rotate image.
6891,zoom,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/image_preproc_benchmark.py,62,function,zoom image.
6892,image_augmentation,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/image_preproc_benchmark.py,78,function,image augmentation.
6893,BenchmarkLayer,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/image_preproc_benchmark.py,88,class,Benchmark the layer forward pass.
6894,run_dataset_implementation,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/image_preproc_benchmark.py,91,method,
6895,bm_layer_implementation,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/image_preproc_benchmark.py,116,method,
6896,benchmark_vocab_size_by_batch,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/image_preproc_benchmark.py,157,method,
6897,word_gen,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/index_lookup_adapt_benchmark.py,45,function,
6898,get_top_k,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/index_lookup_adapt_benchmark.py,50,function,Python implementation of vocabulary building using a defaultdict.
6899,BenchmarkAdapt,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/index_lookup_adapt_benchmark.py,66,class,Benchmark adapt.
6900,run_numpy_implementation,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/index_lookup_adapt_benchmark.py,69,method,Test the python implementation.
6901,bm_adapt_implementation,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/index_lookup_adapt_benchmark.py,89,method,Test the KPL adapt implementation.
6902,benchmark_vocab_size_by_batch,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/index_lookup_adapt_benchmark.py,117,method,
6903,reduce_fn,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/normalization_adapt_benchmark.py,41,function,tf.data.Dataset-friendly implementation of mean and variance.
6904,BenchmarkAdapt,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/normalization_adapt_benchmark.py,63,class,Benchmark adapt.
6905,run_dataset_implementation,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/normalization_adapt_benchmark.py,66,method,
6906,bm_adapt_implementation,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/normalization_adapt_benchmark.py,93,method,Test the KPL adapt implementation.
6907,benchmark_vocab_size_by_batch,tensorflow/tensorflow/python/keras/layers/preprocessing/benchmarks/normalization_adapt_benchmark.py,126,method,
6908,keras_style_scope,tensorflow/tensorflow/python/keras/legacy_tf_layers/base.py,47,function,"Use Keras-style variable management.
All tf.layers and tf RNN cells created in this scope use Keras-style
variable management. Creating such layers with a scope= argument is
disallowed, and reuse=True is disallowed.
The purpose of this scope is to allow users of existing layers to
slowly transition to a Keras layers API without breaking existing
functionality.
One example of this is when using TensorFlow's RNN classes with Keras
Models or Networks. Because Keras models do not properly set variable
scopes, users of RNNs may either accidentally share scopes between two
different models, or get errors about variables that already exist.
Example:
```python
class RNNModel(tf.keras.Model):
def __init__(self, name):
super(RNNModel, self).__init__(name=name)
self.rnn = tf.compat.v1.nn.rnn_cell.MultiRNNCell(
[tf.compat.v1.nn.rnn_cell.LSTMCell(64) for _ in range(2)])
def call(self, input, state):
return self.rnn(input, state)
model_1 = RNNModel(""model_1"")
model_2 = RNNModel(""model_2"")
# OK
output_1, next_state_1 = model_1(input, state)
# Raises an error about trying to create an already existing variable.
output_2, next_state_2 = model_2(input, state)
```
The solution is to wrap the model construction and execution in a keras-style
scope:
```python
with keras_style_scope():
model_1 = RNNModel(""model_1"")
model_2 = RNNModel(""model_2"")
# model_1 and model_2 are guaranteed to create their own variables.
output_1, next_state_1 = model_1(input, state)
output_2, next_state_2 = model_2(input, state)
assert len(model_1.weights) > 0
assert len(model_2.weights) > 0
assert(model_1.weights != model_2.weights)
```
Yields:
A keras layer style scope."
6909,set_keras_style,tensorflow/tensorflow/python/keras/legacy_tf_layers/base.py,115,function,"Use Keras-style variable management.
All tf.layers and tf RNN cells created after keras style ha been enabled
use Keras-style variable management. Creating such layers with a
scope= argument is disallowed, and reuse=True is disallowed.
The purpose of this function is to allow users of existing layers to
slowly transition to Keras layers API without breaking existing
functionality.
For more details, see the documentation for `keras_style_scope`.
Note, once keras style has been set, it is set globally for the entire
program and cannot be unset.
Example:
```python
set_keras_style()
model_1 = RNNModel(name=""model_1"")
model_2 = RNNModel(name=""model_2"")
# model_1 and model_2 are guaranteed to create their own variables.
output_1, next_state_1 = model_1(input, state)
output_2, next_state_2 = model_2(input, state)
assert len(model_1.weights) > 0
assert len(model_2.weights) > 0
assert(model_1.weights != model_2.weights)
```"
6910,Layer,tensorflow/tensorflow/python/keras/legacy_tf_layers/base.py,158,class,"Base layer class.
It is considered legacy, and we recommend the use of `tf.keras.layers.Layer`
instead.
Arguments:
trainable: Boolean, whether the layer's variables should be trainable.
name: String name of the layer.
dtype: Default dtype of the layer's weights (default of `None` means use the
type of the first input).
Read-only properties:
name: The name of the layer (string).
dtype: Default dtype of the layer's weights (default of `None` means use the
type of the first input).
trainable_variables: List of trainable variables.
non_trainable_variables: List of non-trainable variables.
variables: List of all variables of this layer, trainable and
non-trainable.
updates: List of update ops of this layer.
losses: List of losses added by this layer.
trainable_weights: List of variables to be included in backprop.
non_trainable_weights: List of variables that should not be
included in backprop.
weights: The concatenation of the lists trainable_weights and
non_trainable_weights (in this order).
Mutable properties:
trainable: Whether the layer should be trained (boolean).
input_spec: Optional (list of) `InputSpec` object(s) specifying the
constraints on inputs that can be accepted by the layer."
6911,graph,tensorflow/tensorflow/python/keras/legacy_tf_layers/base.py,244,method,
6912,scope_name,tensorflow/tensorflow/python/keras/legacy_tf_layers/base.py,273,method,
6913,add_loss,tensorflow/tensorflow/python/keras/legacy_tf_layers/base.py,282,method,
6914,add_weight,tensorflow/tensorflow/python/keras/legacy_tf_layers/base.py,317,method,"Adds a new variable to the layer, or gets an existing one; returns it.
Arguments:
name: variable name.
shape: variable shape.
dtype: The type of the variable. Defaults to `self.dtype` or `float32`.
initializer: initializer instance (callable).
regularizer: regularizer instance (callable).
trainable: whether the variable should be part of the layer's
""trainable_variables"" (e.g. variables, biases)
or ""non_trainable_variables"" (e.g. BatchNorm mean, stddev).
Note, if the current variable scope is marked as non-trainable
then this parameter is ignored and any added variables are also
marked as non-trainable. `trainable` defaults to `True` unless
`synchronization` is set to `ON_READ`.
constraint: constraint instance (callable).
use_resource: Whether to use `ResourceVariable`.
synchronization: Indicates when a distributed a variable will be
aggregated. Accepted values are constants defined in the class
`tf.VariableSynchronization`. By default the synchronization is set to
`AUTO` and the current `DistributionStrategy` chooses
when to synchronize. If `synchronization` is set to `ON_READ`,
`trainable` must not be set to `True`.
aggregation: Indicates how a distributed variable will be aggregated.
Accepted values are constants defined in the class
`tf.VariableAggregation`.
partitioner: (optional) partitioner instance (callable). If
provided, when the requested variable is created it will be split
into multiple partitions according to `partitioner`. In this case,
an instance of `PartitionedVariable` is returned. Available
partitioners include `tf.compat.v1.fixed_size_partitioner` and
`tf.compat.v1.variable_axis_size_partitioner`. For more details, see
the documentation of `tf.compat.v1.get_variable` and the ""Variable
Partitioners and Sharding"" section of the API guide.
**kwargs: Additional keyword arguments.
Returns:
The created variable. Usually either a `Variable` or `ResourceVariable`
instance. If `partitioner` is not `None`, a `PartitionedVariable`
instance is returned.
Raises:
RuntimeError: If called with partitioned variable regularization and
eager execution is enabled.
ValueError: When trainable has been set to True with synchronization
set as `ON_READ`."
6915,IdentityLayer,tensorflow/tensorflow/python/keras/legacy_tf_layers/base_test.py,647,class,A layer returns the identity of it's input.
6916,call,tensorflow/tensorflow/python/keras/legacy_tf_layers/base_test.py,650,method,
6917,Conv1D,tensorflow/tensorflow/python/keras/legacy_tf_layers/convolutional.py,30,class,"1D convolution layer (e.g. temporal convolution).
This layer creates a convolution kernel that is convolved
(actually cross-correlated) with the layer input to produce a tensor of
outputs. If `use_bias` is True (and a `bias_initializer` is provided),
a bias vector is created and added to the outputs. Finally, if
`activation` is not `None`, it is applied to the outputs as well.
Arguments:
filters: Integer, the dimensionality of the output space (i.e. the number
of filters in the convolution).
kernel_size: An integer or tuple/list of a single integer, specifying the
length of the 1D convolution window.
strides: An integer or tuple/list of a single integer,
specifying the stride length of the convolution.
Specifying any stride value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, length, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, length)`.
dilation_rate: An integer or tuple/list of a single integer, specifying
the dilation rate to use for dilated convolution.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any `strides` value != 1.
activation: Activation function. Set it to None to maintain a
linear activation.
use_bias: Boolean, whether the layer uses a bias.
kernel_initializer: An initializer for the convolution kernel.
bias_initializer: An initializer for the bias vector. If None, the default
initializer will be used.
kernel_regularizer: Optional regularizer for the convolution kernel.
bias_regularizer: Optional regularizer for the bias vector.
activity_regularizer: Optional regularizer function for the output.
kernel_constraint: Optional projection function to be applied to the
kernel after being updated by an `Optimizer` (e.g. used to implement
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training.
bias_constraint: Optional projection function to be applied to the
bias after being updated by an `Optimizer`.
trainable: Boolean, if `True` also add variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see `tf.Variable`).
name: A string, the name of the layer."
6918,conv1d,tensorflow/tensorflow/python/keras/legacy_tf_layers/convolutional.py,125,function,"Functional interface for 1D convolution layer (e.g. temporal convolution).
This layer creates a convolution kernel that is convolved
(actually cross-correlated) with the layer input to produce a tensor of
outputs. If `use_bias` is True (and a `bias_initializer` is provided),
a bias vector is created and added to the outputs. Finally, if
`activation` is not `None`, it is applied to the outputs as well.
Arguments:
inputs: Tensor input.
filters: Integer, the dimensionality of the output space (i.e. the number
of filters in the convolution).
kernel_size: An integer or tuple/list of a single integer, specifying the
length of the 1D convolution window.
strides: An integer or tuple/list of a single integer,
specifying the stride length of the convolution.
Specifying any stride value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, length, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, length)`.
dilation_rate: An integer or tuple/list of a single integer, specifying
the dilation rate to use for dilated convolution.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any `strides` value != 1.
activation: Activation function. Set it to None to maintain a
linear activation.
use_bias: Boolean, whether the layer uses a bias.
kernel_initializer: An initializer for the convolution kernel.
bias_initializer: An initializer for the bias vector. If None, the default
initializer will be used.
kernel_regularizer: Optional regularizer for the convolution kernel.
bias_regularizer: Optional regularizer for the bias vector.
activity_regularizer: Optional regularizer function for the output.
kernel_constraint: Optional projection function to be applied to the
kernel after being updated by an `Optimizer` (e.g. used to implement
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training.
bias_constraint: Optional projection function to be applied to the
bias after being updated by an `Optimizer`.
trainable: Boolean, if `True` also add variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see `tf.Variable`).
name: A string, the name of the layer.
reuse: Boolean, whether to reuse the weights of a previous layer
by the same name.
Returns:
Output tensor.
Raises:
ValueError: if eager execution is enabled."
6919,Conv2D,tensorflow/tensorflow/python/keras/legacy_tf_layers/convolutional.py,228,class,"2D convolution layer (e.g. spatial convolution over images).
This layer creates a convolution kernel that is convolved
(actually cross-correlated) with the layer input to produce a tensor of
outputs. If `use_bias` is True (and a `bias_initializer` is provided),
a bias vector is created and added to the outputs. Finally, if
`activation` is not `None`, it is applied to the outputs as well.
Arguments:
filters: Integer, the dimensionality of the output space (i.e. the number
of filters in the convolution).
kernel_size: An integer or tuple/list of 2 integers, specifying the
height and width of the 2D convolution window.
Can be a single integer to specify the same value for
all spatial dimensions.
strides: An integer or tuple/list of 2 integers,
specifying the strides of the convolution along the height and width.
Can be a single integer to specify the same value for
all spatial dimensions.
Specifying any stride value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, height, width, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, height, width)`.
dilation_rate: An integer or tuple/list of 2 integers, specifying
the dilation rate to use for dilated convolution.
Can be a single integer to specify the same value for
all spatial dimensions.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any stride value != 1.
activation: Activation function. Set it to None to maintain a
linear activation.
use_bias: Boolean, whether the layer uses a bias.
kernel_initializer: An initializer for the convolution kernel.
bias_initializer: An initializer for the bias vector. If None, the default
initializer will be used.
kernel_regularizer: Optional regularizer for the convolution kernel.
bias_regularizer: Optional regularizer for the bias vector.
activity_regularizer: Optional regularizer function for the output.
kernel_constraint: Optional projection function to be applied to the
kernel after being updated by an `Optimizer` (e.g. used to implement
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training.
bias_constraint: Optional projection function to be applied to the
bias after being updated by an `Optimizer`.
trainable: Boolean, if `True` also add variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see `tf.Variable`).
name: A string, the name of the layer."
6920,conv2d,tensorflow/tensorflow/python/keras/legacy_tf_layers/convolutional.py,330,function,"Functional interface for the 2D convolution layer.
This layer creates a convolution kernel that is convolved
(actually cross-correlated) with the layer input to produce a tensor of
outputs. If `use_bias` is True (and a `bias_initializer` is provided),
a bias vector is created and added to the outputs. Finally, if
`activation` is not `None`, it is applied to the outputs as well.
Arguments:
inputs: Tensor input.
filters: Integer, the dimensionality of the output space (i.e. the number
of filters in the convolution).
kernel_size: An integer or tuple/list of 2 integers, specifying the
height and width of the 2D convolution window.
Can be a single integer to specify the same value for
all spatial dimensions.
strides: An integer or tuple/list of 2 integers,
specifying the strides of the convolution along the height and width.
Can be a single integer to specify the same value for
all spatial dimensions.
Specifying any stride value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, height, width, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, height, width)`.
dilation_rate: An integer or tuple/list of 2 integers, specifying
the dilation rate to use for dilated convolution.
Can be a single integer to specify the same value for
all spatial dimensions.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any stride value != 1.
activation: Activation function. Set it to None to maintain a
linear activation.
use_bias: Boolean, whether the layer uses a bias.
kernel_initializer: An initializer for the convolution kernel.
bias_initializer: An initializer for the bias vector. If None, the default
initializer will be used.
kernel_regularizer: Optional regularizer for the convolution kernel.
bias_regularizer: Optional regularizer for the bias vector.
activity_regularizer: Optional regularizer function for the output.
kernel_constraint: Optional projection function to be applied to the
kernel after being updated by an `Optimizer` (e.g. used to implement
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training.
bias_constraint: Optional projection function to be applied to the
bias after being updated by an `Optimizer`.
trainable: Boolean, if `True` also add variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see `tf.Variable`).
name: A string, the name of the layer.
reuse: Boolean, whether to reuse the weights of a previous layer
by the same name.
Returns:
Output tensor.
Raises:
ValueError: if eager execution is enabled."
6921,Conv3D,tensorflow/tensorflow/python/keras/legacy_tf_layers/convolutional.py,440,class,"3D convolution layer (e.g. spatial convolution over volumes).
This layer creates a convolution kernel that is convolved
(actually cross-correlated) with the layer input to produce a tensor of
outputs. If `use_bias` is True (and a `bias_initializer` is provided),
a bias vector is created and added to the outputs. Finally, if
`activation` is not `None`, it is applied to the outputs as well.
Arguments:
filters: Integer, the dimensionality of the output space (i.e. the number
of filters in the convolution).
kernel_size: An integer or tuple/list of 3 integers, specifying the
depth, height and width of the 3D convolution window.
Can be a single integer to specify the same value for
all spatial dimensions.
strides: An integer or tuple/list of 3 integers,
specifying the strides of the convolution along the depth,
height and width.
Can be a single integer to specify the same value for
all spatial dimensions.
Specifying any stride value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, depth, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch, channels, depth, height, width)`.
dilation_rate: An integer or tuple/list of 3 integers, specifying
the dilation rate to use for dilated convolution.
Can be a single integer to specify the same value for
all spatial dimensions.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any stride value != 1.
activation: Activation function. Set it to None to maintain a
linear activation.
use_bias: Boolean, whether the layer uses a bias.
kernel_initializer: An initializer for the convolution kernel.
bias_initializer: An initializer for the bias vector. If None, the default
initializer will be used.
kernel_regularizer: Optional regularizer for the convolution kernel.
bias_regularizer: Optional regularizer for the bias vector.
activity_regularizer: Optional regularizer function for the output.
kernel_constraint: Optional projection function to be applied to the
kernel after being updated by an `Optimizer` (e.g. used to implement
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training.
bias_constraint: Optional projection function to be applied to the
bias after being updated by an `Optimizer`.
trainable: Boolean, if `True` also add variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see `tf.Variable`).
name: A string, the name of the layer."
6922,conv3d,tensorflow/tensorflow/python/keras/legacy_tf_layers/convolutional.py,543,function,"Functional interface for the 3D convolution layer.
This layer creates a convolution kernel that is convolved
(actually cross-correlated) with the layer input to produce a tensor of
outputs. If `use_bias` is True (and a `bias_initializer` is provided),
a bias vector is created and added to the outputs. Finally, if
`activation` is not `None`, it is applied to the outputs as well.
Arguments:
inputs: Tensor input.
filters: Integer, the dimensionality of the output space (i.e. the number
of filters in the convolution).
kernel_size: An integer or tuple/list of 3 integers, specifying the
depth, height and width of the 3D convolution window.
Can be a single integer to specify the same value for
all spatial dimensions.
strides: An integer or tuple/list of 3 integers,
specifying the strides of the convolution along the depth,
height and width.
Can be a single integer to specify the same value for
all spatial dimensions.
Specifying any stride value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, depth, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch, channels, depth, height, width)`.
dilation_rate: An integer or tuple/list of 3 integers, specifying
the dilation rate to use for dilated convolution.
Can be a single integer to specify the same value for
all spatial dimensions.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any stride value != 1.
activation: Activation function. Set it to None to maintain a
linear activation.
use_bias: Boolean, whether the layer uses a bias.
kernel_initializer: An initializer for the convolution kernel.
bias_initializer: An initializer for the bias vector. If None, the default
initializer will be used.
kernel_regularizer: Optional regularizer for the convolution kernel.
bias_regularizer: Optional regularizer for the bias vector.
activity_regularizer: Optional regularizer function for the output.
kernel_constraint: Optional projection function to be applied to the
kernel after being updated by an `Optimizer` (e.g. used to implement
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training.
bias_constraint: Optional projection function to be applied to the
bias after being updated by an `Optimizer`.
trainable: Boolean, if `True` also add variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see `tf.Variable`).
name: A string, the name of the layer.
reuse: Boolean, whether to reuse the weights of a previous layer
by the same name.
Returns:
Output tensor.
Raises:
ValueError: if eager execution is enabled."
6923,SeparableConv1D,tensorflow/tensorflow/python/keras/legacy_tf_layers/convolutional.py,654,class,"Depthwise separable 1D convolution.
This layer performs a depthwise convolution that acts separately on
channels, followed by a pointwise convolution that mixes channels.
If `use_bias` is True and a bias initializer is provided,
it adds a bias vector to the output.
It then optionally applies an activation function to produce the final output.
Arguments:
filters: Integer, the dimensionality of the output space (i.e. the number
of filters in the convolution).
kernel_size: A single integer specifying the spatial
dimensions of the filters.
strides: A single integer specifying the strides
of the convolution.
Specifying any `stride` value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, length, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, length)`.
dilation_rate: A single integer, specifying
the dilation rate to use for dilated convolution.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any stride value != 1.
depth_multiplier: The number of depthwise convolution output channels for
each input channel. The total number of depthwise convolution output
channels will be equal to `num_filters_in * depth_multiplier`.
activation: Activation function. Set it to None to maintain a
linear activation.
use_bias: Boolean, whether the layer uses a bias.
depthwise_initializer: An initializer for the depthwise convolution kernel.
pointwise_initializer: An initializer for the pointwise convolution kernel.
bias_initializer: An initializer for the bias vector. If None, the default
initializer will be used.
depthwise_regularizer: Optional regularizer for the depthwise
convolution kernel.
pointwise_regularizer: Optional regularizer for the pointwise
convolution kernel.
bias_regularizer: Optional regularizer for the bias vector.
activity_regularizer: Optional regularizer function for the output.
depthwise_constraint: Optional projection function to be applied to the
depthwise kernel after being updated by an `Optimizer` (e.g. used for
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training.
pointwise_constraint: Optional projection function to be applied to the
pointwise kernel after being updated by an `Optimizer`.
bias_constraint: Optional projection function to be applied to the
bias after being updated by an `Optimizer`.
trainable: Boolean, if `True` also add variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see `tf.Variable`).
name: A string, the name of the layer."
6924,SeparableConv2D,tensorflow/tensorflow/python/keras/legacy_tf_layers/convolutional.py,764,class,"Depthwise separable 2D convolution.
This layer performs a depthwise convolution that acts separately on
channels, followed by a pointwise convolution that mixes channels.
If `use_bias` is True and a bias initializer is provided,
it adds a bias vector to the output.
It then optionally applies an activation function to produce the final output.
Arguments:
filters: Integer, the dimensionality of the output space (i.e. the number
of filters in the convolution).
kernel_size: A tuple or list of 2 integers specifying the spatial
dimensions of the filters. Can be a single integer to specify the same
value for all spatial dimensions.
strides: A tuple or list of 2 positive integers specifying the strides
of the convolution. Can be a single integer to specify the same value for
all spatial dimensions.
Specifying any `stride` value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, height, width, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, height, width)`.
dilation_rate: An integer or tuple/list of 2 integers, specifying
the dilation rate to use for dilated convolution.
Can be a single integer to specify the same value for
all spatial dimensions.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any stride value != 1.
depth_multiplier: The number of depthwise convolution output channels for
each input channel. The total number of depthwise convolution output
channels will be equal to `num_filters_in * depth_multiplier`.
activation: Activation function. Set it to None to maintain a
linear activation.
use_bias: Boolean, whether the layer uses a bias.
depthwise_initializer: An initializer for the depthwise convolution kernel.
pointwise_initializer: An initializer for the pointwise convolution kernel.
bias_initializer: An initializer for the bias vector. If None, the default
initializer will be used.
depthwise_regularizer: Optional regularizer for the depthwise
convolution kernel.
pointwise_regularizer: Optional regularizer for the pointwise
convolution kernel.
bias_regularizer: Optional regularizer for the bias vector.
activity_regularizer: Optional regularizer function for the output.
depthwise_constraint: Optional projection function to be applied to the
depthwise kernel after being updated by an `Optimizer` (e.g. used for
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training.
pointwise_constraint: Optional projection function to be applied to the
pointwise kernel after being updated by an `Optimizer`.
bias_constraint: Optional projection function to be applied to the
bias after being updated by an `Optimizer`.
trainable: Boolean, if `True` also add variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see `tf.Variable`).
name: A string, the name of the layer."
6925,separable_conv1d,tensorflow/tensorflow/python/keras/legacy_tf_layers/convolutional.py,882,function,"Functional interface for the depthwise separable 1D convolution layer.
This layer performs a depthwise convolution that acts separately on
channels, followed by a pointwise convolution that mixes channels.
If `use_bias` is True and a bias initializer is provided,
it adds a bias vector to the output.
It then optionally applies an activation function to produce the final output.
Arguments:
inputs: Input tensor.
filters: Integer, the dimensionality of the output space (i.e. the number
of filters in the convolution).
kernel_size: A single integer specifying the spatial
dimensions of the filters.
strides: A single integer specifying the strides
of the convolution.
Specifying any `stride` value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, length, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, length)`.
dilation_rate: A single integer, specifying
the dilation rate to use for dilated convolution.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any stride value != 1.
depth_multiplier: The number of depthwise convolution output channels for
each input channel. The total number of depthwise convolution output
channels will be equal to `num_filters_in * depth_multiplier`.
activation: Activation function. Set it to None to maintain a
linear activation.
use_bias: Boolean, whether the layer uses a bias.
depthwise_initializer: An initializer for the depthwise convolution kernel.
pointwise_initializer: An initializer for the pointwise convolution kernel.
bias_initializer: An initializer for the bias vector. If None, the default
initializer will be used.
depthwise_regularizer: Optional regularizer for the depthwise
convolution kernel.
pointwise_regularizer: Optional regularizer for the pointwise
convolution kernel.
bias_regularizer: Optional regularizer for the bias vector.
activity_regularizer: Optional regularizer function for the output.
depthwise_constraint: Optional projection function to be applied to the
depthwise kernel after being updated by an `Optimizer` (e.g. used for
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training.
pointwise_constraint: Optional projection function to be applied to the
pointwise kernel after being updated by an `Optimizer`.
bias_constraint: Optional projection function to be applied to the
bias after being updated by an `Optimizer`.
trainable: Boolean, if `True` also add variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see `tf.Variable`).
name: A string, the name of the layer.
reuse: Boolean, whether to reuse the weights of a previous layer
by the same name.
Returns:
Output tensor.
Raises:
ValueError: if eager execution is enabled."
6926,separable_conv2d,tensorflow/tensorflow/python/keras/legacy_tf_layers/convolutional.py,1005,function,"Functional interface for the depthwise separable 2D convolution layer.
This layer performs a depthwise convolution that acts separately on
channels, followed by a pointwise convolution that mixes channels.
If `use_bias` is True and a bias initializer is provided,
it adds a bias vector to the output.
It then optionally applies an activation function to produce the final output.
Arguments:
inputs: Input tensor.
filters: Integer, the dimensionality of the output space (i.e. the number
of filters in the convolution).
kernel_size: A tuple or list of 2 integers specifying the spatial
dimensions of the filters. Can be a single integer to specify the same
value for all spatial dimensions.
strides: A tuple or list of 2 positive integers specifying the strides
of the convolution. Can be a single integer to specify the same value for
all spatial dimensions.
Specifying any `stride` value != 1 is incompatible with specifying
any `dilation_rate` value != 1.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, height, width, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, height, width)`.
dilation_rate: An integer or tuple/list of 2 integers, specifying
the dilation rate to use for dilated convolution.
Can be a single integer to specify the same value for
all spatial dimensions.
Currently, specifying any `dilation_rate` value != 1 is
incompatible with specifying any stride value != 1.
depth_multiplier: The number of depthwise convolution output channels for
each input channel. The total number of depthwise convolution output
channels will be equal to `num_filters_in * depth_multiplier`.
activation: Activation function. Set it to None to maintain a
linear activation.
use_bias: Boolean, whether the layer uses a bias.
depthwise_initializer: An initializer for the depthwise convolution kernel.
pointwise_initializer: An initializer for the pointwise convolution kernel.
bias_initializer: An initializer for the bias vector. If None, the default
initializer will be used.
depthwise_regularizer: Optional regularizer for the depthwise
convolution kernel.
pointwise_regularizer: Optional regularizer for the pointwise
convolution kernel.
bias_regularizer: Optional regularizer for the bias vector.
activity_regularizer: Optional regularizer function for the output.
depthwise_constraint: Optional projection function to be applied to the
depthwise kernel after being updated by an `Optimizer` (e.g. used for
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training.
pointwise_constraint: Optional projection function to be applied to the
pointwise kernel after being updated by an `Optimizer`.
bias_constraint: Optional projection function to be applied to the
bias after being updated by an `Optimizer`.
trainable: Boolean, if `True` also add variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see `tf.Variable`).
name: A string, the name of the layer.
reuse: Boolean, whether to reuse the weights of a previous layer
by the same name.
Returns:
Output tensor.
Raises:
ValueError: if eager execution is enabled."
6927,Conv2DTranspose,tensorflow/tensorflow/python/keras/legacy_tf_layers/convolutional.py,1130,class,"Transposed 2D convolution layer (sometimes called 2D Deconvolution).
The need for transposed convolutions generally arises
from the desire to use a transformation going in the opposite direction
of a normal convolution, i.e., from something that has the shape of the
output of some convolution to something that has the shape of its input
while maintaining a connectivity pattern that is compatible with
said convolution.
Arguments:
filters: Integer, the dimensionality of the output space (i.e. the number
of filters in the convolution).
kernel_size: A tuple or list of 2 positive integers specifying the spatial
dimensions of the filters. Can be a single integer to specify the same
value for all spatial dimensions.
strides: A tuple or list of 2 positive integers specifying the strides
of the convolution. Can be a single integer to specify the same value for
all spatial dimensions.
padding: one of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, height, width, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, height, width)`.
activation: Activation function. Set it to None to maintain a
linear activation.
use_bias: Boolean, whether the layer uses a bias.
kernel_initializer: An initializer for the convolution kernel.
bias_initializer: An initializer for the bias vector. If None, the default
initializer will be used.
kernel_regularizer: Optional regularizer for the convolution kernel.
bias_regularizer: Optional regularizer for the bias vector.
activity_regularizer: Optional regularizer function for the output.
kernel_constraint: Optional projection function to be applied to the
kernel after being updated by an `Optimizer` (e.g. used to implement
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training.
bias_constraint: Optional projection function to be applied to the
bias after being updated by an `Optimizer`.
trainable: Boolean, if `True` also add variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see `tf.Variable`).
name: A string, the name of the layer."
6928,conv2d_transpose,tensorflow/tensorflow/python/keras/legacy_tf_layers/convolutional.py,1221,function,"Functional interface for transposed 2D convolution layer.
The need for transposed convolutions generally arises
from the desire to use a transformation going in the opposite direction
of a normal convolution, i.e., from something that has the shape of the
output of some convolution to something that has the shape of its input
while maintaining a connectivity pattern that is compatible with
said convolution.
Arguments:
inputs: Input tensor.
filters: Integer, the dimensionality of the output space (i.e. the number
of filters in the convolution).
kernel_size: A tuple or list of 2 positive integers specifying the spatial
dimensions of the filters. Can be a single integer to specify the same
value for all spatial dimensions.
strides: A tuple or list of 2 positive integers specifying the strides
of the convolution. Can be a single integer to specify the same value for
all spatial dimensions.
padding: one of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, height, width, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, height, width)`.
activation: Activation function. Set it to `None` to maintain a
linear activation.
use_bias: Boolean, whether the layer uses a bias.
kernel_initializer: An initializer for the convolution kernel.
bias_initializer: An initializer for the bias vector. If `None`, the default
initializer will be used.
kernel_regularizer: Optional regularizer for the convolution kernel.
bias_regularizer: Optional regularizer for the bias vector.
activity_regularizer: Optional regularizer function for the output.
kernel_constraint: Optional projection function to be applied to the
kernel after being updated by an `Optimizer` (e.g. used to implement
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training.
bias_constraint: Optional projection function to be applied to the
bias after being updated by an `Optimizer`.
trainable: Boolean, if `True` also add variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see `tf.Variable`).
name: A string, the name of the layer.
reuse: Boolean, whether to reuse the weights of a previous layer
by the same name.
Returns:
Output tensor.
Raises:
ValueError: if eager execution is enabled."
6929,Conv3DTranspose,tensorflow/tensorflow/python/keras/legacy_tf_layers/convolutional.py,1319,class,"Transposed 3D convolution layer (sometimes called 3D Deconvolution).
Arguments:
filters: Integer, the dimensionality of the output space (i.e. the number
of filters in the convolution).
kernel_size: An integer or tuple/list of 3 integers, specifying the
depth, height and width of the 3D convolution window.
Can be a single integer to specify the same value for all spatial
dimensions.
strides: An integer or tuple/list of 3 integers, specifying the strides
of the convolution along the depth, height and width.
Can be a single integer to specify the same value for all spatial
dimensions.
padding: One of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, depth, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch, channels, depth, height, width)`.
activation: Activation function. Set it to `None` to maintain a
linear activation.
use_bias: Boolean, whether the layer uses a bias.
kernel_initializer: An initializer for the convolution kernel.
bias_initializer: An initializer for the bias vector. If `None`, the default
initializer will be used.
kernel_regularizer: Optional regularizer for the convolution kernel.
bias_regularizer: Optional regularizer for the bias vector.
activity_regularizer: Optional regularizer function for the output.
kernel_constraint: Optional projection function to be applied to the
kernel after being updated by an `Optimizer` (e.g. used to implement
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training.
bias_constraint: Optional projection function to be applied to the
bias after being updated by an `Optimizer`.
trainable: Boolean, if `True` also add variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see `tf.Variable`).
name: A string, the name of the layer."
6930,conv3d_transpose,tensorflow/tensorflow/python/keras/legacy_tf_layers/convolutional.py,1407,function,"Functional interface for transposed 3D convolution layer.
Arguments:
inputs: Input tensor.
filters: Integer, the dimensionality of the output space (i.e. the number
of filters in the convolution).
kernel_size: A tuple or list of 3 positive integers specifying the spatial
dimensions of the filters. Can be a single integer to specify the same
value for all spatial dimensions.
strides: A tuple or list of 3 positive integers specifying the strides
of the convolution. Can be a single integer to specify the same value for
all spatial dimensions.
padding: one of `""valid""` or `""same""` (case-insensitive).
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, depth, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch, channels, depth, height, width)`.
activation: Activation function. Set it to None to maintain a
linear activation.
use_bias: Boolean, whether the layer uses a bias.
kernel_initializer: An initializer for the convolution kernel.
bias_initializer: An initializer for the bias vector. If None, the default
initializer will be used.
kernel_regularizer: Optional regularizer for the convolution kernel.
bias_regularizer: Optional regularizer for the bias vector.
activity_regularizer: Optional regularizer function for the output.
kernel_constraint: Optional projection function to be applied to the
kernel after being updated by an `Optimizer` (e.g. used to implement
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training.
bias_constraint: Optional projection function to be applied to the
bias after being updated by an `Optimizer`.
trainable: Boolean, if `True` also add variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see `tf.Variable`).
name: A string, the name of the layer.
reuse: Boolean, whether to reuse the weights of a previous layer
by the same name.
Returns:
Output tensor.
Raises:
ValueError: if eager execution is enabled."
6931,Dense,tensorflow/tensorflow/python/keras/legacy_tf_layers/core.py,33,class,"Densely-connected layer class.
This layer implements the operation:
`outputs = activation(inputs * kernel + bias)`
Where `activation` is the activation function passed as the `activation`
argument (if not `None`), `kernel` is a weights matrix created by the layer,
and `bias` is a bias vector created by the layer
(only if `use_bias` is `True`).
Arguments:
units: Integer or Long, dimensionality of the output space.
activation: Activation function (callable). Set it to None to maintain a
linear activation.
use_bias: Boolean, whether the layer uses a bias.
kernel_initializer: Initializer function for the weight matrix.
If `None` (default), weights are initialized using the default
initializer used by `tf.compat.v1.get_variable`.
bias_initializer: Initializer function for the bias.
kernel_regularizer: Regularizer function for the weight matrix.
bias_regularizer: Regularizer function for the bias.
activity_regularizer: Regularizer function for the output.
kernel_constraint: An optional projection function to be applied to the
kernel after being updated by an `Optimizer` (e.g. used to implement
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training.
bias_constraint: An optional projection function to be applied to the
bias after being updated by an `Optimizer`.
trainable: Boolean, if `True` also add variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see `tf.Variable`).
name: String, the name of the layer. Layers with the same name will
share weights, but to avoid mistakes we require reuse=True in such cases.
_reuse: Boolean, whether to reuse the weights of a previous layer
by the same name.
Properties:
units: Python integer, dimensionality of the output space.
activation: Activation function (callable).
use_bias: Boolean, whether the layer uses a bias.
kernel_initializer: Initializer instance (or name) for the kernel matrix.
bias_initializer: Initializer instance (or name) for the bias.
kernel_regularizer: Regularizer instance for the kernel matrix (callable)
bias_regularizer: Regularizer instance for the bias (callable).
activity_regularizer: Regularizer instance for the output (callable)
kernel_constraint: Constraint function for the kernel matrix.
bias_constraint: Constraint function for the bias.
kernel: Weight matrix (TensorFlow variable or tensor).
bias: Bias vector, if applicable (TensorFlow variable or tensor)."
6932,dense,tensorflow/tensorflow/python/keras/legacy_tf_layers/core.py,116,function,"Functional interface for the densely-connected layer.
This layer implements the operation:
`outputs = activation(inputs * kernel + bias)`
where `activation` is the activation function passed as the `activation`
argument (if not `None`), `kernel` is a weights matrix created by the layer,
and `bias` is a bias vector created by the layer
(only if `use_bias` is `True`).
Arguments:
inputs: Tensor input.
units: Integer or Long, dimensionality of the output space.
activation: Activation function (callable). Set it to None to maintain a
linear activation.
use_bias: Boolean, whether the layer uses a bias.
kernel_initializer: Initializer function for the weight matrix.
If `None` (default), weights are initialized using the default
initializer used by `tf.compat.v1.get_variable`.
bias_initializer: Initializer function for the bias.
kernel_regularizer: Regularizer function for the weight matrix.
bias_regularizer: Regularizer function for the bias.
activity_regularizer: Regularizer function for the output.
kernel_constraint: An optional projection function to be applied to the
kernel after being updated by an `Optimizer` (e.g. used to implement
norm constraints or value constraints for layer weights). The function
must take as input the unprojected variable and must return the
projected variable (which must have the same shape). Constraints are
not safe to use when doing asynchronous distributed training.
bias_constraint: An optional projection function to be applied to the
bias after being updated by an `Optimizer`.
trainable: Boolean, if `True` also add variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see `tf.Variable`).
name: String, the name of the layer.
reuse: Boolean, whether to reuse the weights of a previous layer
by the same name.
Returns:
Output tensor the same shape as `inputs` except the last dimension is of
size `units`.
Raises:
ValueError: if eager execution is enabled."
6933,Dropout,tensorflow/tensorflow/python/keras/legacy_tf_layers/core.py,191,class,"Applies Dropout to the input.
Dropout consists in randomly setting a fraction `rate` of input units to 0
at each update during training time, which helps prevent overfitting.
The units that are kept are scaled by `1 / (1 - rate)`, so that their
sum is unchanged at training time and inference time.
Arguments:
rate: The dropout rate, between 0 and 1. E.g. `rate=0.1` would drop out
10% of input units.
noise_shape: 1D tensor of type `int32` representing the shape of the
binary dropout mask that will be multiplied with the input.
For instance, if your inputs have shape
`(batch_size, timesteps, features)`, and you want the dropout mask
to be the same for all timesteps, you can use
`noise_shape=[batch_size, 1, features]`.
seed: A Python integer. Used to create random seeds. See
`tf.compat.v1.set_random_seed`.
for behavior.
name: The name of the layer (string)."
6934,call,tensorflow/tensorflow/python/keras/legacy_tf_layers/core.py,225,method,
6935,dropout,tensorflow/tensorflow/python/keras/legacy_tf_layers/core.py,233,function,"Applies Dropout to the input.
Dropout consists in randomly setting a fraction `rate` of input units to 0
at each update during training time, which helps prevent overfitting.
The units that are kept are scaled by `1 / (1 - rate)`, so that their
sum is unchanged at training time and inference time.
Arguments:
inputs: Tensor input.
rate: The dropout rate, between 0 and 1. E.g. ""rate=0.1"" would drop out
10% of input units.
noise_shape: 1D tensor of type `int32` representing the shape of the
binary dropout mask that will be multiplied with the input.
For instance, if your inputs have shape
`(batch_size, timesteps, features)`, and you want the dropout mask
to be the same for all timesteps, you can use
`noise_shape=[batch_size, 1, features]`.
seed: A Python integer. Used to create random seeds. See
`tf.compat.v1.set_random_seed`
for behavior.
training: Either a Python boolean, or a TensorFlow boolean scalar tensor
(e.g. a placeholder). Whether to return the output in training mode
(apply dropout) or in inference mode (return the input untouched).
name: The name of the layer (string).
Returns:
Output tensor.
Raises:
ValueError: if eager execution is enabled."
6936,Flatten,tensorflow/tensorflow/python/keras/legacy_tf_layers/core.py,275,class,"Flattens an input tensor while preserving the batch axis (axis 0).
Arguments:
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, ..., channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, ...)`.
Examples:
```
x = tf.compat.v1.placeholder(shape=(None, 4, 4), dtype='float32')
y = Flatten()(x)
# now `y` has shape `(None, 16)`
x = tf.compat.v1.placeholder(shape=(None, 3, None), dtype='float32')
y = Flatten()(x)
# now `y` has shape `(None, None)`
```"
6937,flatten,tensorflow/tensorflow/python/keras/legacy_tf_layers/core.py,304,function,"Flattens an input tensor while preserving the batch axis (axis 0).
Arguments:
inputs: Tensor input.
name: The name of the layer (string).
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, height, width, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, height, width)`.
Returns:
Reshaped tensor.
Examples:
```
x = tf.compat.v1.placeholder(shape=(None, 4, 4), dtype='float32')
y = flatten(x)
# now `y` has shape `(None, 16)`
x = tf.compat.v1.placeholder(shape=(None, 3, None), dtype='float32')
y = flatten(x)
# now `y` has shape `(None, None)`
```"
6938,BatchNormalization,tensorflow/tensorflow/python/keras/legacy_tf_layers/normalization.py,31,class,"Batch Normalization layer from (Ioffe et al., 2015).
Keras APIs handle BatchNormalization updates to the moving_mean and
moving_variance as part of their `fit()` and `evaluate()` loops. However, if a
custom training loop is used with an instance of `Model`, these updates need
to be explicitly included. Here's a simple example of how it can be done:
```python
# model is an instance of Model that contains BatchNormalization layer.
update_ops = model.get_updates_for(None) + model.get_updates_for(features)
train_op = optimizer.minimize(loss)
train_op = tf.group([train_op, update_ops])
```
Arguments:
axis: An `int` or list of `int`, the axis or axes that should be normalized,
typically the features axis/axes. For instance, after a `Conv2D` layer
with `data_format=""channels_first""`, set `axis=1`. If a list of axes is
provided, each axis in `axis` will be normalized
simultaneously. Default is `-1` which uses the last axis. Note: when
using multi-axis batch norm, the `beta`, `gamma`, `moving_mean`, and
`moving_variance` variables are the same rank as the input Tensor,
with dimension size 1 in all reduced (non-axis) dimensions).
momentum: Momentum for the moving average.
epsilon: Small float added to variance to avoid dividing by zero.
center: If True, add offset of `beta` to normalized tensor. If False, `beta`
is ignored.
scale: If True, multiply by `gamma`. If False, `gamma` is not used. When the
next layer is linear (also e.g. `nn.relu`), this can be disabled since the
scaling can be done by the next layer.
beta_initializer: Initializer for the beta weight.
gamma_initializer: Initializer for the gamma weight.
moving_mean_initializer: Initializer for the moving mean.
moving_variance_initializer: Initializer for the moving variance.
beta_regularizer: Optional regularizer for the beta weight.
gamma_regularizer: Optional regularizer for the gamma weight.
beta_constraint: An optional projection function to be applied to the `beta`
weight after being updated by an `Optimizer` (e.g. used to implement norm
constraints or value constraints for layer weights). The function must
take as input the unprojected variable and must return the projected
variable (which must have the same shape). Constraints are not safe to use
when doing asynchronous distributed training.
gamma_constraint: An optional projection function to be applied to the
`gamma` weight after being updated by an `Optimizer`.
renorm: Whether to use Batch Renormalization (Ioffe, 2017). This adds extra
variables during training. The inference is the same for either value of
this parameter.
renorm_clipping: A dictionary that may map keys 'rmax', 'rmin', 'dmax' to
scalar `Tensors` used to clip the renorm correction. The correction `(r,
d)` is used as `corrected_value = normalized_value * r + d`, with `r`
clipped to [rmin, rmax], and `d` to [-dmax, dmax]. Missing rmax, rmin,
dmax are set to inf, 0, inf, respectively.
renorm_momentum: Momentum used to update the moving means and standard
deviations with renorm. Unlike `momentum`, this affects training and
should be neither too small (which would add noise) nor too large (which
would give stale estimates). Note that `momentum` is still applied to get
the means and variances for inference.
fused: if `None` or `True`, use a faster, fused implementation if possible.
If `False`, use the system recommended implementation.
trainable: Boolean, if `True` also add variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see tf.Variable).
virtual_batch_size: An `int`. By default, `virtual_batch_size` is `None`,
which means batch normalization is performed across the whole batch. When
`virtual_batch_size` is not `None`, instead perform ""Ghost Batch
Normalization"", which creates virtual sub-batches which are each
normalized separately (with shared gamma, beta, and moving statistics).
Must divide the actual batch size during execution.
adjustment: A function taking the `Tensor` containing the (dynamic) shape of
the input tensor and returning a pair (scale, bias) to apply to the
normalized values (before gamma and beta), only during training. For
example, if axis==-1,
`adjustment = lambda shape: (
tf.random.uniform(shape[-1:], 0.93, 1.07),
tf.random.uniform(shape[-1:], -0.1, 0.1))` will scale the normalized
value by up to 7% up or down, then shift the result by up to 0.1
(with independent scaling and bias for each feature but shared
across all examples), and finally apply gamma and/or beta. If
`None`, no adjustment is applied. Cannot be specified if
virtual_batch_size is specified.
name: A string, the name of the layer.
References:
Batch Normalization - Accelerating Deep Network Training by Reducing
Internal Covariate Shift:
[Ioffe et al., 2015](http://proceedings.mlr.press/v37/ioffe15.html)
([pdf](http://proceedings.mlr.press/v37/ioffe15.pdf))
Batch Renormalization - Towards Reducing Minibatch Dependence in
Batch-Normalized Models:
[Ioffe,
2017](http://papers.nips.cc/paper/6790-batch-renormalization-towards-reducing-minibatch-dependence-in-batch-normalized-models)
([pdf](http://papers.nips.cc/paper/6790-batch-renormalization-towards-reducing-minibatch-dependence-in-batch-normalized-models.pdf))"
6939,call,tensorflow/tensorflow/python/keras/legacy_tf_layers/normalization.py,171,method,
6940,batch_normalization,tensorflow/tensorflow/python/keras/legacy_tf_layers/normalization.py,181,function,"Functional interface for the batch normalization layer from_config(Ioffe et al., 2015).
Note: when training, the moving_mean and moving_variance need to be updated.
By default the update ops are placed in `tf.GraphKeys.UPDATE_OPS`, so they
need to be executed alongside the `train_op`. Also, be sure to add any
batch_normalization ops before getting the update_ops collection. Otherwise,
update_ops will be empty, and training/inference will not work properly. For
example:
```python
x_norm = tf.compat.v1.layers.batch_normalization(x, training=training)
# ...
update_ops = tf.compat.v1.get_collection(tf.GraphKeys.UPDATE_OPS)
train_op = optimizer.minimize(loss)
train_op = tf.group([train_op, update_ops])
```
Arguments:
inputs: Tensor input.
axis: An `int`, the axis that should be normalized (typically the features
axis). For instance, after a `Convolution2D` layer with
`data_format=""channels_first""`, set `axis=1` in `BatchNormalization`.
momentum: Momentum for the moving average.
epsilon: Small float added to variance to avoid dividing by zero.
center: If True, add offset of `beta` to normalized tensor. If False, `beta`
is ignored.
scale: If True, multiply by `gamma`. If False, `gamma` is not used. When the
next layer is linear (also e.g. `nn.relu`), this can be disabled since the
scaling can be done by the next layer.
beta_initializer: Initializer for the beta weight.
gamma_initializer: Initializer for the gamma weight.
moving_mean_initializer: Initializer for the moving mean.
moving_variance_initializer: Initializer for the moving variance.
beta_regularizer: Optional regularizer for the beta weight.
gamma_regularizer: Optional regularizer for the gamma weight.
beta_constraint: An optional projection function to be applied to the `beta`
weight after being updated by an `Optimizer` (e.g. used to implement norm
constraints or value constraints for layer weights). The function must
take as input the unprojected variable and must return the projected
variable (which must have the same shape). Constraints are not safe to use
when doing asynchronous distributed training.
gamma_constraint: An optional projection function to be applied to the
`gamma` weight after being updated by an `Optimizer`.
training: Either a Python boolean, or a TensorFlow boolean scalar tensor
(e.g. a placeholder). Whether to return the output in training mode
(normalized with statistics of the current batch) or in inference mode
(normalized with moving statistics). **NOTE**: make sure to set this
parameter correctly, or else your training/inference will not work
properly.
trainable: Boolean, if `True` also add variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES` (see tf.Variable).
name: String, the name of the layer.
reuse: Boolean, whether to reuse the weights of a previous layer by the same
name.
renorm: Whether to use Batch Renormalization (Ioffe, 2017). This adds extra
variables during training. The inference is the same for either value of
this parameter.
renorm_clipping: A dictionary that may map keys 'rmax', 'rmin', 'dmax' to
scalar `Tensors` used to clip the renorm correction. The correction `(r,
d)` is used as `corrected_value = normalized_value * r + d`, with `r`
clipped to [rmin, rmax], and `d` to [-dmax, dmax]. Missing rmax, rmin,
dmax are set to inf, 0, inf, respectively.
renorm_momentum: Momentum used to update the moving means and standard
deviations with renorm. Unlike `momentum`, this affects training and
should be neither too small (which would add noise) nor too large (which
would give stale estimates). Note that `momentum` is still applied to get
the means and variances for inference.
fused: if `None` or `True`, use a faster, fused implementation if possible.
If `False`, use the system recommended implementation.
virtual_batch_size: An `int`. By default, `virtual_batch_size` is `None`,
which means batch normalization is performed across the whole batch. When
`virtual_batch_size` is not `None`, instead perform ""Ghost Batch
Normalization"", which creates virtual sub-batches which are each
normalized separately (with shared gamma, beta, and moving statistics).
Must divide the actual batch size during execution.
adjustment: A function taking the `Tensor` containing the (dynamic) shape of
the input tensor and returning a pair (scale, bias) to apply to the
normalized values (before gamma and beta), only during training. For
example, if axis==-1,
`adjustment = lambda shape: (
tf.random.uniform(shape[-1:], 0.93, 1.07),
tf.random.uniform(shape[-1:], -0.1, 0.1))` will scale the normalized
value by up to 7% up or down, then shift the result by up to 0.1
(with independent scaling and bias for each feature but shared
across all examples), and finally apply gamma and/or beta. If
`None`, no adjustment is applied. Cannot be specified if
virtual_batch_size is specified.
Returns:
Output tensor.
Raises:
ValueError: if eager execution is enabled.
References:
Batch Normalization - Accelerating Deep Network Training by Reducing
Internal Covariate Shift:
[Ioffe et al., 2015](http://proceedings.mlr.press/v37/ioffe15.html)
([pdf](http://proceedings.mlr.press/v37/ioffe15.pdf))
Batch Renormalization - Towards Reducing Minibatch Dependence in
Batch-Normalized Models:
[Ioffe,
2017](http://papers.nips.cc/paper/6790-batch-renormalization-towards-reducing-minibatch-dependence-in-batch-normalized-models)
([pdf](http://papers.nips.cc/paper/6790-batch-renormalization-towards-reducing-minibatch-dependence-in-batch-normalized-models.pdf))"
6941,AveragePooling1D,tensorflow/tensorflow/python/keras/legacy_tf_layers/pooling.py,29,class,"Average Pooling layer for 1D inputs.
Arguments:
pool_size: An integer or tuple/list of a single integer,
representing the size of the pooling window.
strides: An integer or tuple/list of a single integer, specifying the
strides of the pooling operation.
padding: A string. The padding method, either 'valid' or 'same'.
Case-insensitive.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, length, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, length)`.
name: A string, the name of the layer."
6942,average_pooling1d,tensorflow/tensorflow/python/keras/legacy_tf_layers/pooling.py,64,function,"Average Pooling layer for 1D inputs.
Arguments:
inputs: The tensor over which to pool. Must have rank 3.
pool_size: An integer or tuple/list of a single integer,
representing the size of the pooling window.
strides: An integer or tuple/list of a single integer, specifying the
strides of the pooling operation.
padding: A string. The padding method, either 'valid' or 'same'.
Case-insensitive.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, length, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, length)`.
name: A string, the name of the layer.
Returns:
The output tensor, of rank 3.
Raises:
ValueError: if eager execution is enabled."
6943,MaxPooling1D,tensorflow/tensorflow/python/keras/legacy_tf_layers/pooling.py,99,class,"Max Pooling layer for 1D inputs.
Arguments:
pool_size: An integer or tuple/list of a single integer,
representing the size of the pooling window.
strides: An integer or tuple/list of a single integer, specifying the
strides of the pooling operation.
padding: A string. The padding method, either 'valid' or 'same'.
Case-insensitive.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, length, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, length)`.
name: A string, the name of the layer."
6944,max_pooling1d,tensorflow/tensorflow/python/keras/legacy_tf_layers/pooling.py,134,function,"Max Pooling layer for 1D inputs.
Arguments:
inputs: The tensor over which to pool. Must have rank 3.
pool_size: An integer or tuple/list of a single integer,
representing the size of the pooling window.
strides: An integer or tuple/list of a single integer, specifying the
strides of the pooling operation.
padding: A string. The padding method, either 'valid' or 'same'.
Case-insensitive.
data_format: A string, one of `channels_last` (default) or `channels_first`.
The ordering of the dimensions in the inputs.
`channels_last` corresponds to inputs with shape
`(batch, length, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, length)`.
name: A string, the name of the layer.
Returns:
The output tensor, of rank 3.
Raises:
ValueError: if eager execution is enabled."
6945,AveragePooling2D,tensorflow/tensorflow/python/keras/legacy_tf_layers/pooling.py,169,class,"Average pooling layer for 2D inputs (e.g. images).
Arguments:
pool_size: An integer or tuple/list of 2 integers: (pool_height, pool_width)
specifying the size of the pooling window.
Can be a single integer to specify the same value for
all spatial dimensions.
strides: An integer or tuple/list of 2 integers,
specifying the strides of the pooling operation.
Can be a single integer to specify the same value for
all spatial dimensions.
padding: A string. The padding method, either 'valid' or 'same'.
Case-insensitive.
data_format: A string. The ordering of the dimensions in the inputs.
`channels_last` (default) and `channels_first` are supported.
`channels_last` corresponds to inputs with shape
`(batch, height, width, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, height, width)`.
name: A string, the name of the layer."
6946,average_pooling2d,tensorflow/tensorflow/python/keras/legacy_tf_layers/pooling.py,204,function,"Average pooling layer for 2D inputs (e.g. images).
Arguments:
inputs: The tensor over which to pool. Must have rank 4.
pool_size: An integer or tuple/list of 2 integers: (pool_height, pool_width)
specifying the size of the pooling window.
Can be a single integer to specify the same value for
all spatial dimensions.
strides: An integer or tuple/list of 2 integers,
specifying the strides of the pooling operation.
Can be a single integer to specify the same value for
all spatial dimensions.
padding: A string. The padding method, either 'valid' or 'same'.
Case-insensitive.
data_format: A string. The ordering of the dimensions in the inputs.
`channels_last` (default) and `channels_first` are supported.
`channels_last` corresponds to inputs with shape
`(batch, height, width, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, height, width)`.
name: A string, the name of the layer.
Returns:
Output tensor.
Raises:
ValueError: if eager execution is enabled."
6947,MaxPooling2D,tensorflow/tensorflow/python/keras/legacy_tf_layers/pooling.py,242,class,"Max pooling layer for 2D inputs (e.g. images).
Arguments:
pool_size: An integer or tuple/list of 2 integers: (pool_height, pool_width)
specifying the size of the pooling window.
Can be a single integer to specify the same value for
all spatial dimensions.
strides: An integer or tuple/list of 2 integers,
specifying the strides of the pooling operation.
Can be a single integer to specify the same value for
all spatial dimensions.
padding: A string. The padding method, either 'valid' or 'same'.
Case-insensitive.
data_format: A string. The ordering of the dimensions in the inputs.
`channels_last` (default) and `channels_first` are supported.
`channels_last` corresponds to inputs with shape
`(batch, height, width, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, height, width)`.
name: A string, the name of the layer."
6948,max_pooling2d,tensorflow/tensorflow/python/keras/legacy_tf_layers/pooling.py,277,function,"Max pooling layer for 2D inputs (e.g. images).
Arguments:
inputs: The tensor over which to pool. Must have rank 4.
pool_size: An integer or tuple/list of 2 integers: (pool_height, pool_width)
specifying the size of the pooling window.
Can be a single integer to specify the same value for
all spatial dimensions.
strides: An integer or tuple/list of 2 integers,
specifying the strides of the pooling operation.
Can be a single integer to specify the same value for
all spatial dimensions.
padding: A string. The padding method, either 'valid' or 'same'.
Case-insensitive.
data_format: A string. The ordering of the dimensions in the inputs.
`channels_last` (default) and `channels_first` are supported.
`channels_last` corresponds to inputs with shape
`(batch, height, width, channels)` while `channels_first` corresponds to
inputs with shape `(batch, channels, height, width)`.
name: A string, the name of the layer.
Returns:
Output tensor.
Raises:
ValueError: if eager execution is enabled."
6949,AveragePooling3D,tensorflow/tensorflow/python/keras/legacy_tf_layers/pooling.py,315,class,"Average pooling layer for 3D inputs (e.g. volumes).
Arguments:
pool_size: An integer or tuple/list of 3 integers:
(pool_depth, pool_height, pool_width)
specifying the size of the pooling window.
Can be a single integer to specify the same value for
all spatial dimensions.
strides: An integer or tuple/list of 3 integers,
specifying the strides of the pooling operation.
Can be a single integer to specify the same value for
all spatial dimensions.
padding: A string. The padding method, either 'valid' or 'same'.
Case-insensitive.
data_format: A string. The ordering of the dimensions in the inputs.
`channels_last` (default) and `channels_first` are supported.
`channels_last` corresponds to inputs with shape
`(batch, depth, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch, channels, depth, height, width)`.
name: A string, the name of the layer."
6950,average_pooling3d,tensorflow/tensorflow/python/keras/legacy_tf_layers/pooling.py,352,function,"Average pooling layer for 3D inputs (e.g. volumes).
Arguments:
inputs: The tensor over which to pool. Must have rank 5.
pool_size: An integer or tuple/list of 3 integers:
(pool_depth, pool_height, pool_width)
specifying the size of the pooling window.
Can be a single integer to specify the same value for
all spatial dimensions.
strides: An integer or tuple/list of 3 integers,
specifying the strides of the pooling operation.
Can be a single integer to specify the same value for
all spatial dimensions.
padding: A string. The padding method, either 'valid' or 'same'.
Case-insensitive.
data_format: A string. The ordering of the dimensions in the inputs.
`channels_last` (default) and `channels_first` are supported.
`channels_last` corresponds to inputs with shape
`(batch, depth, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch, channels, depth, height, width)`.
name: A string, the name of the layer.
Returns:
Output tensor.
Raises:
ValueError: if eager execution is enabled."
6951,MaxPooling3D,tensorflow/tensorflow/python/keras/legacy_tf_layers/pooling.py,392,class,"Max pooling layer for 3D inputs (e.g. volumes).
Arguments:
pool_size: An integer or tuple/list of 3 integers:
(pool_depth, pool_height, pool_width)
specifying the size of the pooling window.
Can be a single integer to specify the same value for
all spatial dimensions.
strides: An integer or tuple/list of 3 integers,
specifying the strides of the pooling operation.
Can be a single integer to specify the same value for
all spatial dimensions.
padding: A string. The padding method, either 'valid' or 'same'.
Case-insensitive.
data_format: A string. The ordering of the dimensions in the inputs.
`channels_last` (default) and `channels_first` are supported.
`channels_last` corresponds to inputs with shape
`(batch, depth, height, width, channels)` while `channels_first`
corresponds to inputs with shape
`(batch, channels, depth, height, width)`.
name: A string, the name of the layer."
6952,max_pooling3d,tensorflow/tensorflow/python/keras/legacy_tf_layers/pooling.py,429,function,"Max pooling layer for 3D inputs (e.g.
volumes).
Arguments:
inputs: The tensor over which to pool. Must have rank 5.
pool_size: An integer or tuple/list of 3 integers: (pool_depth, pool_height,
pool_width) specifying the size of the pooling window. Can be a single
integer to specify the same value for all spatial dimensions.
strides: An integer or tuple/list of 3 integers, specifying the strides of
the pooling operation. Can be a single integer to specify the same value
for all spatial dimensions.
padding: A string. The padding method, either 'valid' or 'same'.
Case-insensitive.
data_format: A string. The ordering of the dimensions in the inputs.
`channels_last` (default) and `channels_first` are supported.
`channels_last` corresponds to inputs with shape `(batch, depth, height,
width, channels)` while `channels_first` corresponds to inputs with shape
`(batch, channels, depth, height, width)`.
name: A string, the name of the layer.
Returns:
Output tensor.
Raises:
ValueError: if eager execution is enabled."
6953,AutoCastVariable,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,30,class,"Variable that will cast itself to a different dtype in applicable contexts.
This class wraps a floating-point `tf.Variable`. It emulates the variable
interface and delegates to the wrapped variable, but it additionally will cast
the wrapped variable under a `Graph._enable_auto_casting_variables(dtype)`
context manager.
For example:
>>> v = tf.Variable(1.0, dtype=tf.float32)
>>> v = AutoCastVariable(v)
>>> tf.identity(v).dtype
tf.float32
>>> with ops.get_default_graph()._enable_auto_casting_variables(tf.float16):
... tf.identity(v).dtype
tf.float16
>>> with ops.get_default_graph()._enable_auto_casting_variables(tf.float16):
... v.dtype # v.dtype also changes under the context manager
tf.float16
The purpose of this class is to allow Keras layers to create variables in
float32, and automatically cast them to float16 or bfloat16 when the layer is
called."
6954,dtype,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,82,method,The dtype this variable will be casted to when read.
6955,true_dtype,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,90,method,"The dtype of the underlying variable, before any casts are done."
6956,value,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,94,method,
6957,read_value,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,100,method,
6958,sparse_read,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,104,method,"Reads the value of this variable sparsely, using `gather`."
6959,gather_nd,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,109,method,Gather slices of the variable into a Tensor.
6960,set_shape,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,162,method,
6961,trainable,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,166,method,
6962,synchronization,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,170,method,
6963,aggregation,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,174,method,
6964,eval,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,177,method,
6965,initialized_value,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,180,method,
6966,initial_value,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,184,method,
6967,constraint,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,188,method,
6968,assign,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,191,method,
6969,assign_add,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,195,method,
6970,assign_sub,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,199,method,
6971,scatter_sub,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,203,method,
6972,scatter_add,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,207,method,
6973,scatter_max,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,211,method,
6974,scatter_min,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,215,method,
6975,scatter_mul,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,219,method,
6976,scatter_div,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,223,method,
6977,scatter_update,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,227,method,
6978,batch_scatter_update,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,231,method,
6979,scatter_nd_sub,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,235,method,
6980,scatter_nd_add,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,239,method,
6981,scatter_nd_update,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,243,method,
6982,load,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,247,method,
6983,name,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,251,method,
6984,initializer,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,259,method,
6985,device,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,263,method,
6986,op,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,267,method,
6987,graph,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,271,method,
6988,shape,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,275,method,
6989,get_shape,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,278,method,
6990,to_proto,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,297,method,
6991,from_proto,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,300,method,
6992,create_autocast_variable,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable.py,431,function,"Creates an AutoCastVariable that wraps another variable.
This typically just returns `AutoCastVariable(variable)`. But, if the variable
is a DistributedVariable or one of its subclasses, we instead dynamically
create a class that subclasses from both AutoCastVariable and
variable.__class__. This is so the returned variable will still pass
`isinstance(variable, variable.__class__)`, which is required for
DistributedVariables and its subclasses to work properly.
Args:
variable: A floating-point resource variable to wrap.
Returns:
An AutoCastVariable that wraps the variable."
6993,get_var,tensorflow/tensorflow/python/keras/mixed_precision/experimental/autocast_variable_test.py,51,function,
6994,log_device_compatibility_check,tensorflow/tensorflow/python/keras/mixed_precision/experimental/device_compatibility_check.py,135,function,"Logs a compatibility check if the devices support the policy.
Currently only logs for the policy mixed_float16. A log is shown only the
first time this function is called.
Args:
policy_name: The name of the dtype policy."
6995,device_details,tensorflow/tensorflow/python/keras/mixed_precision/experimental/device_compatibility_check_test.py,29,function,
6996,get_layer_policy,tensorflow/tensorflow/python/keras/mixed_precision/experimental/get_layer_policy.py,29,function,"Returns the dtype policy of a layer.
Args:
layer: A `tf.keras.layers.Layer`.
Returns:
The `tf.keras.mixed_precision.experimental.Policy` of the layer."
6997,MultiplyLayerWithoutAutoCast,tensorflow/tensorflow/python/keras/mixed_precision/experimental/keras_test.py,65,class,"Same as MultiplyLayer, but does not use AutoCastVariables."
6998,build,tensorflow/tensorflow/python/keras/mixed_precision/experimental/keras_test.py,68,method,
6999,call,tensorflow/tensorflow/python/keras/mixed_precision/experimental/keras_test.py,80,method,
7000,MultiplyLayerWithFunction,tensorflow/tensorflow/python/keras/mixed_precision/experimental/keras_test.py,86,class,"Same as MultiplyLayer, but _multiply is decorated with a tf.function."
7001,create_mirrored_strategy,tensorflow/tensorflow/python/keras/mixed_precision/experimental/keras_test.py,99,function,"Create a MirroredStrategy, using a GPU if it is available."
7002,create_central_storage_strategy,tensorflow/tensorflow/python/keras/mixed_precision/experimental/keras_test.py,107,function,"Create a CentralStorageStrategy, using a GPU if it is available."
7003,create_mirrored_strategy,tensorflow/tensorflow/python/keras/mixed_precision/experimental/layer_correctness_test.py,49,function,
7004,serialize,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale.py,30,function,
7005,deserialize,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale.py,34,function,
7006,get,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale.py,48,function,Get a loss scale object.
7007,LossScaleBenchmark,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_benchmark.py,46,class,Benchmark for loss scaling.
7008,benchmark_optimizer,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_benchmark.py,164,method,
7009,benchmark_gradient_tape,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_benchmark.py,170,method,
7010,get_loss,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_benchmark.py,111,method,
7011,run_fn,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_benchmark.py,138,method,
7012,minimize_fn,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_benchmark.py,117,method,
7013,minimize_fn,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_benchmark.py,126,method,
7014,LossScaleOptimizer,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,176,class,"An optimizer that applies loss scaling.
Loss scaling is a process that multiplies the loss by a multiplier called the
loss scale, and divides each gradient by the same multiplier. The pseudocode
for this process is:
```
loss = ...
loss *= loss_scale
grads = gradients(loss, vars)
grads /= loss_scale
```
Mathematically, loss scaling has no effect, but can help avoid numerical
underflow in intermediate gradients when float16 tensors are used. By
multiplying the loss, each intermediate gradient will have the same multiplier
applied.
The loss scale can either be a fixed constant, chosen by the user, or be
dynamically determined. Dynamically determining the loss scale is convenient
as a loss scale does not have to be explicitly chosen. However it reduces
performance.
This optimizer wraps another optimizer and applies loss scaling to it via a
`LossScale`. Loss scaling is applied whenever gradients are
computed, either through `minimize()` or `get_gradients()`. The loss scale is
updated via `LossScale.update()` whenever gradients are applied, either
through `minimize()` or `apply_gradients()`. For example:
>>> opt = tf.keras.optimizers.SGD(0.25)
>>> opt = tf.keras.mixed_precision.experimental.LossScaleOptimizer(opt,
... ""dynamic"")
>>> var = tf.Variable(1.)
>>> loss_fn = lambda: var ** 2
>>> # 'minimize' applies loss scaling to the loss and updates the loss sale.
>>> opt.minimize(loss_fn, var_list=var)
>>> var.numpy()
0.5
If a `tf.GradientTape` is used to compute gradients instead of
`LossScaleOptimizer.minimize` or `LossScaleOptimizer.get_gradients`, the loss
and gradients must be scaled manually. This can be done by calling
`LossScaleOptimizer.get_scaled_loss` before passing the loss to
`tf.GradientTape`, and `LossScaleOptimizer.get_unscaled_gradients` after
computing the gradients with `tf.GradientTape`. For example:
>>> with tf.GradientTape() as tape:
... loss = loss_fn()
... scaled_loss = opt.get_scaled_loss(loss)
>>> scaled_grad = tape.gradient(scaled_loss, var)
>>> (grad,) = opt.get_unscaled_gradients([scaled_grad])
>>> opt.apply_gradients([(grad, var)]) # Loss scale is updated here
>>> var.numpy()
0.25"
7015,loss_scale,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,290,method,The `LossScale` instance associated with this optimizer.
7016,get_scaled_loss,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,294,method,"Scales the loss by the loss scale.
This method is only needed if you compute gradients manually, e.g. with
`tf.GradientTape`. In that case, call this method to scale the loss before
passing the loss to `tf.GradientTape`. If you use
`LossScaleOptimizer.minimize` or `LossScaleOptimizer.get_gradients`, loss
scaling is automatically applied and this method is unneeded.
If this method is called, `get_unscaled_gradients` should also be called.
See the `tf.keras.mixed_precision.experimental.LossScaleOptimizer` doc for
an example.
Args:
loss: The loss, which will be multiplied by the loss scale. Can either be
a tensor or a callable returning a tensor.
Returns:
`loss` multiplied by `LossScaleOptimizer.loss_scale()`."
7017,get_unscaled_gradients,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,323,method,"Unscales the gradients by the loss scale.
This method is only needed if you compute gradients manually, e.g. with
`tf.GradientTape`. In that case, call this method to unscale the gradients
after computing them with `tf.GradientTape`. If you use
`LossScaleOptimizer.minimize` or `LossScaleOptimizer.get_gradients`, loss
scaling is automatically applied and this method is unneeded.
If this method is called, `get_scaled_loss` should also be called. See
the `tf.keras.mixed_precision.experimental.LossScaleOptimizer` doc for an
example.
Args:
grads: A list of tensors, each which will be divided by the loss scale.
Can have None values, which are ignored.
Returns:
A new list the same size as `grads`, where every non-None value in `grads`
is divided by `LossScaleOptimizer.loss_scale()`."
7018,get_gradients,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,360,method,
7019,apply_gradients,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,368,method,
7020,get_config,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,423,method,
7021,from_config,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,432,method,
7022,iterations,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,461,method,
7023,iterations,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,465,method,
7024,get_slot_names,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,468,method,
7025,variables,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,471,method,
7026,weights,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,475,method,
7027,get_weights,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,478,method,
7028,set_weights,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,481,method,
7029,get_slot,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,496,method,
7030,add_slot,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,499,method,
7031,learning_rate,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,508,method,
7032,learning_rate,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,512,method,
7033,lr,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,516,method,
7034,lr,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,520,method,
7035,apply_fn,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,388,method,
7036,do_not_apply_fn,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,399,method,
7037,new_loss,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,316,method,
7038,FakeOptimizerForRestoration,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,545,class,"A fake optimizer used to support restoring TensorFlow 2.2 checkpoints.
The checkpoint format for LossScaleOptimizers changed after TF 2.2. This class
exists to support restoring TF 2.2 checkpoints in newer version of TensorFlow.
In TF 2.2, LossScaleOptimizer would track the wrapped optimizer by calling the
following in LossScaleOptimizer.__init__
```
self._track_trackable(self._optimizer, 'base_optimizer')
```
This means a dependency from the LossScaleOptimizer to the wrapped optimizer
would be stored in the checkpoint. However now, the checkpoint format with a
LossScaleOptimizer is the same as the format without a LossScaleOptimizer,
except the loss scale is also stored. This means there is no dependency from
the LossScaleOptimizer to the wrapped optimizer. Instead, the
LossScaleOptimizer acts as if it is the wrapped optimizer, from a checkpoint's
perspective, by overriding all Trackable methods and delegating them to the
wrapped optimizer.
To allow restoring TF 2.2. checkpoints, LossScaleOptimizer adds a dependency
on this class instead of the inner optimizer. When restored, this class will
instead restore the slot variables of the inner optimizer. Since this class
has no variables, it does not affect the checkpoint when saved."
7039,get_slot_names,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,576,method,
7040,strategy_supports_loss_scaling,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer.py,602,function,Returns True if the current Strategy supports loss scaling.
7041,create_mirrored_strategy,tensorflow/tensorflow/python/keras/mixed_precision/experimental/loss_scale_optimizer_test.py,55,function,
7042,Policy,tensorflow/tensorflow/python/keras/mixed_precision/experimental/policy.py,41,class,"A dtype policy for a Keras layer.
A dtype policy determines dtype-related aspects of a layer, such as its
computation and variable dtypes. Each layer has a policy. Policies can be
passed to the `dtype` argument of layer constructors, or a global policy can
be set with `tf.keras.mixed_precision.experimental.set_policy`. A layer will
default to the global policy if no policy is passed to it's constructor.
For many models, each layer's policy will have the same compute dtype and
variable dtype, which will typically be float32. In this case, we refer to the
singular dtype as the layer's dtype, which can be queried by the property
`tf.keras.layers.Layer.dtype`.
When mixed precision training is used, most layers will instead have a float16
or bfloat16 compute dtype and a float32 variable dtype, and so the layer does
not have a single dtype. When the variable dtype does not match the compute
dtype, variables will be automatically casted to the compute dtype to avoid
type errors. In this case, `tf.keras.layers.Layer.dtype` refers to the
variable dtype, not the compute dtype. See [the mixed precision guide](
https://www.tensorflow.org/guide/keras/mixed_precision) for more
information on how to use mixed precision.
Certain policies also have a `tf.mixed_precision.experimental.LossScale`
instance, which is used by `tf.keras.Model`s to performance loss scaling. Loss
scaling is a technique used with mixed precision to avoid numerical underflow
in float16 gradients. Loss scaling is only done by Models in `Model.fit`,
`Model.train_on_batch`, and similar methods. Layers which are not Models
ignore the loss scale.
Policies are constructed by passing a string to the constructor, e.g.
`tf.keras.mixed_precision.experimental.Policy('float32')`. The string
determines the compute and variable dtypes. It can be one of the following:
* Any dtype name, such as 'float32' or 'float64'. Both the variable and
compute dtypes will be that dtype. No loss scaling is done by default.
* 'mixed_float16' or 'mixed_bfloat16': The compute dtype is float16 or
bfloat16, while the variable dtype is float32. These policies are used for
mixed precision training. With 'mixed_float16', a dynamic loss scale is
used by default. 'mixed_bfloat16' does no loss scaling by default, as loss
scaling is unnecessary with bfloat16.
### How to use mixed precision in a Keras model
To use mixed precision in a Keras model, the `'mixed_float16'` or
`'mixed_bfloat16'` policy can be used.
`tf.keras.mixed_precision.experimental.set_policy` can be used to set the
default policy for layers if no policy is passed to them. For example:
>>> tf.keras.mixed_precision.experimental.set_policy('mixed_float16')
>>> model = tf.keras.models.Sequential([
... tf.keras.layers.Input((100,)),
... # Dense layers use global policy of 'mixed_float16', which does
... # computations in float16 while keeping variables in float32.
... tf.keras.layers.Dense(10),
... tf.keras.layers.Dense(10),
... # Softmax should be done in float32 for numeric stability. We pass
... # dtype='float32' to use float32 instead of the global policy.
... tf.keras.layers.Activation('softmax', dtype='float32')
... ])
Alternatively, the policy can be passed to individual layers instead of
setting the global policy with `set_policy`:
>>> policy = tf.keras.mixed_precision.experimental.Policy('mixed_float16')
>>> model = tf.keras.models.Sequential([
... tf.keras.layers.Input((100,)),
... tf.keras.layers.Dense(10, dtype=policy),
... tf.keras.layers.Dense(10, dtype=policy),
... # Softmax should be done in float32 for numeric stability.
... tf.keras.layers.Activation('softmax', dtype='float32')
... ])
Note the `'mixed_float16'` policy will apply loss scaling by default in
`Model.fit`, `Model.train_on_batch`, and other training methods. If no such
method is used (e.g., a custom training loop is used) and `'mixed_float16'` is
used, the loss scale must be manually applied. See
`tf.keras.mixed_precision.experimental.LossScaleOptimizer` for details. For
`'mixed_bfloat16'`, no loss scaling is done and loss scaling never needs to be
manually applied.
See [the mixed precision guide](
https://www.tensorflow.org/guide/keras/mixed_precision) for more
information on using mixed precision
### How to use float64 in a Keras model
Using float64 is similar to mixed precision. Either the global policy can be
set to float64, or `dtype='float64'` can be passed to individual layers. For
example, to set the global policy:
>>> tf.keras.mixed_precision.experimental.set_policy('float64')
>>> model = tf.keras.models.Sequential([
... tf.keras.layers.Input((100,)),
... # All layers use global policy of 'float64', which does computations
... # and creates variables in float64.
... tf.keras.layers.Dense(10),
... tf.keras.layers.Dense(10),
... tf.keras.layers.Activation('softmax')
... ])
>>> # Optionaly set policy back to float32 if any other models use float32
>>> tf.keras.mixed_precision.experimental.set_policy('float32')
### How a layer uses its policy's compute dtype
A layer will cast its inputs to its compute dtype in TensorFlow 2. For
example:
>>> x = tf.ones((4, 4, 4, 4), dtype='float64')
>>> # `layer`'s policy defaults to float32.
>>> layer = tf.keras.layers.Conv2D(filters=4, kernel_size=2)
>>> # `layer` casts it's inputs to its compute dtype, which is float32, and
>>> # does computations in float32.
>>> y = layer(x)
>>> y.dtype
tf.float32
Note that the base `tf.keras.layers.Layer` class inserts the casts. If
subclassing your own layer, you do not have to insert any casts.
Currently, only tensors in the first argument to the layer's `call` method are
casted. For example:
>>> class MyLayer(tf.keras.layers.Layer):
... # Bug! `b` will not be casted.
... def call(self, a, b):
... return a + 1., b + 1.
>>> a = tf.constant(1., dtype=""float32"")
>>> b = tf.constant(1., dtype=""float32"")
>>> layer = MyLayer(dtype=""float64"")
>>> x, y = layer(a, b)
>>> x.dtype
tf.float64
>>> y.dtype
tf.float32
If writing your own layer, it is recommended to accept tensors only in the
first argument. This way, all tensors are casted to the layer's compute dtype.
`MyLayer` should therefore be written as:
>>> class MyLayer(tf.keras.layers.Layer):
... # Now, all tensor inputs will be casted.
... def call(self, inputs):
... a, b = inputs
... return a + 1., b + 1.
>>> a = tf.constant(1., dtype=""float32"")
>>> b = tf.constant(1., dtype=""float32"")
>>> layer = MyLayer(dtype=""float64"")
>>> x, y = layer((a, b))
>>> x.dtype
tf.float64
>>> y.dtype
tf.float64
Other arguments are not automatically casted for technical reasons, but this
may change in a future minor release.
The casting only occurs in TensorFlow 2, but can be enabled if
`tf.compat.v1.disable_v2_behavior()` has been called with
`tf.compat.v1.keras.layers.enable_v2_dtype_behavior()`.
A layer subclass can prevent its inputs from being autocasted by passing
`autocast=False` to the layer constructor. For example:
>>> class NonAutoCastingLayer(tf.keras.layers.Layer):
... def __init__(self, **kwargs):
... kwargs['autocast'] = False
... super(NonAutoCastingLayer, self).__init__(**kwargs)
... def call(self, inp):
... return inp
>>> x = tf.ones((4, 4, 4, 4), dtype='float32')
>>> layer = NonAutoCastingLayer(dtype='float64')
>>> y = layer(x) # Will not cast inputs to it's compute dtype of float64
>>> y.dtype
tf.float32
### How a layer uses its policy's variable dtype
The default dtype of variables created by `tf.keras.layers.Layer.add_weight`
is the layer's policy's variable dtype.
If a layer's compute and variable dtypes differ, `add_weight` will wrap
floating-point variables with a special wrapper called an `AutoCastVariable`.
This wrapper is identical to the original variable except it casts itself to
the layer's compute dtype when used within `Layer.call`. Outside `Layer.call`,
the variable is not casted.
A layer author can prevent a variable from being wrapped with an
`AutoCastVariable` by passing `experimental_autocast=False` to `add_weight`:
>>> class MyLayer(tf.keras.layers.Layer):
... def build(self, input_shape):
... self.x = self.add_weight('x')
... self.y = self.add_weight('y', experimental_autocast=False)
>>> policy = tf.keras.mixed_precision.experimental.Policy('mixed_float16')
>>> layer = MyLayer(dtype=policy)
>>> layer.build((2, 2))
>>> layer.x
<AutoCastVariable 'x:0' shape=() dtype=float32 true_dtype=float32, numpy=...>
>>> layer.y
<tf.Variable 'y:0' shape=() dtype=float32, numpy=...>
Passing `experimental_autocast=False` is useful for layers which may
internally do some math in the variable dtype instead of the compute dtype.
For example, you may wish to compute variable statistics, such as mean and
variance, in the variable dtype.
### How to write a layer that supports mixed precision and float64.
For the most part, layers will automatically support mixed precision and
float64 without any additional work, due to the fact the base layer
automatically casts inputs, creates variables of the correct type, and in the
case of mixed precision, wraps variables with `AutoCastVariables`.
For example, this simple dense layer does not require any additional work to
support mixed precision or float64. Keras automatically casts the inputs and
variable to the appropriate dtype.
>>> class MyDense(tf.keras.layers.Layer):
... def build(self, input_shape):
... self.kernel = self.add_weight('kernel', (input_shape[-1], 10))
... def call(self, inputs):
... return tf.matmul(inputs, self.kernel)
>>> policy = tf.keras.mixed_precision.experimental.Policy('mixed_float16')
>>> layer = MyDense(dtype=policy)
>>> x = np.random.rand(10, 10)
>>> y = layer(x)
>>> y.dtype
tf.float16
The primary case where you need extra work to support mixed precision or
float64 is when you create a new tensor, such as with `tf.ones` or
`tf.constant`. In such cases, you must create the tensor of the correct dtype.
For example, suppose you modify the `MyDense` layer to add a random number to
the output using `tf.random.normal`. You must pass the input dtype to
`tf.random.normal` to ensure the dtypes match.
>>> class MyDense(tf.keras.layers.Layer):
... def build(self, input_shape):
... self.kernel = self.add_weight('kernel', (input_shape[-1], 10))
... def call(self, inputs):
... rand = tf.random.normal(shape=inputs.shape, dtype=inputs.dtype)
... return tf.matmul(inputs, self.kernel) + rand
>>>
>>> layer = MyDense(dtype=policy)
>>> y = layer(x)
>>> y.dtype
tf.float16
If you did not pass `dtype=inputs.dtype` to `tf.random.normal`, a `TypeError`
would have occurred. This is because the dtype defaults to `""float32""`, so the
layer would only work if the inputs were float32."
7043,variable_dtype,tensorflow/tensorflow/python/keras/mixed_precision/experimental/policy.py,393,method,"The variable dtype of this policy.
This is the dtype layers will create their variables in, unless a layer
explicitly chooses a different dtype. If this is different than
`Policy.compute_dtype`, Layers will cast variables to the compute dtype to
avoid type errors.
Returns:
The variable dtype of this policy."
7044,compute_dtype,tensorflow/tensorflow/python/keras/mixed_precision/experimental/policy.py,407,method,"The compute dtype of this policy.
This is the dtype layers will do their computations in.
Note that even if the compute dtype is float16 or bfloat16, hardware devices
may not do individual adds, multiplies, and other fundamental operations in
[b]float16, but instead may do some of them in float32 for numeric
stability. The compute dtype is the dtype of the inputs and outputs of the
TensorFlow ops that the layer executes. Internally, many TensorFlow ops will
do certain internal calculations in float32, or some other device-internal
intermediate format with higher precision than [b]float16, to increase
numeric stability.
For example, a `tf.keras.layers.Dense` layer, when run on a GPU with a
float16 compute dtype, will pass float16 inputs to tf.matmul. But, tf.matmul
will do use float32 intermediate math. The performance benefit of float16 is
still apparent, due to increased memory bandwidth and the fact modern GPUs
have specialized hardware for computing matmuls on float16 while still
keeping intermediate computations in float32.
Returns:
The compute dtype of this policy."
7045,should_cast_variables,tensorflow/tensorflow/python/keras/mixed_precision/experimental/policy.py,434,method,"Returns True if variables should be casted.
This is true if the variable dtype is not the same as the compute dtype.
Returns:
True, if variables should be casted."
7046,loss_scale,tensorflow/tensorflow/python/keras/mixed_precision/experimental/policy.py,445,method,"Returns the loss scale of this Policy.
Returns:
A `tf.mixed_precision.experimental.LossScale`, or None."
7047,name,tensorflow/tensorflow/python/keras/mixed_precision/experimental/policy.py,454,method,Returns the name of this policy.
7048,get_config,tensorflow/tensorflow/python/keras/mixed_precision/experimental/policy.py,461,method,
7049,from_config,tensorflow/tensorflow/python/keras/mixed_precision/experimental/policy.py,473,method,
7050,global_policy,tensorflow/tensorflow/python/keras/mixed_precision/experimental/policy.py,489,function,"Returns the global Policy.
The global policy is the default policy used for layers, if no policy is
passed to the layer constructor. If no policy has been set with
`keras.mixed_precision.experimental.set_policy`, this will return a policy
constructed from `tf.keras.backend.floatx()` (floatx defaults to float32).
If TensorFlow 2 behavior has been disabled with
`tf.compat.v1.disable_v2_behavior()`, this will instead return a special
""_infer"" policy which infers the dtype from the dtype of the first input the
first time the layer is called. This behavior matches the behavior that
existed in TensorFlow 1.
See `tf.keras.mixed_precision.experimental.Policy` for more information on
policies.
Returns:
The global Policy."
7051,policy_defaults_to_floatx,tensorflow/tensorflow/python/keras/mixed_precision/experimental/policy.py,517,function,Returns True if `global_policy()` will use the current value of floatx.
7052,set_policy,tensorflow/tensorflow/python/keras/mixed_precision/experimental/policy.py,539,function,"Sets the global Policy.
The global policy is the default policy used for layers, if no policy is
passed to the layer constructor. If no global policy is set, layers will
instead default to a Policy constructed from `tf.keras.backend.floatx()`.
See `keras.mixed_precision.experimental.Policy` for more information.
Args:
policy: A Policy, or a string that will be converted to a Policy.."
7053,policy_scope,tensorflow/tensorflow/python/keras/mixed_precision/experimental/policy.py,568,function,"A context manager that sets the global Policy under it.
Args:
policy: A Policy, or a string that will be converted to a Policy..
Yields:
Nothing."
7054,serialize,tensorflow/tensorflow/python/keras/mixed_precision/experimental/policy.py,615,function,
7055,deserialize,tensorflow/tensorflow/python/keras/mixed_precision/experimental/policy.py,624,function,
7056,create_identity_with_grad_check_fn,tensorflow/tensorflow/python/keras/mixed_precision/experimental/test_util.py,32,function,"Returns a function that asserts it's gradient has a certain value.
This serves as a hook to assert intermediate gradients have a certain value.
This returns an identity function. The identity's gradient function is also
the identity function, except it asserts that the gradient equals
`expected_gradient` and has dtype `expected_dtype`.
Args:
expected_gradient: The gradient function asserts that the gradient is this
value.
expected_dtype: The gradient function asserts the gradient has this dtype.
Returns:
An identity function whose gradient function asserts the gradient has a
certain value."
7057,create_identity_with_nan_gradients_fn,tensorflow/tensorflow/python/keras/mixed_precision/experimental/test_util.py,76,function,"Returns a function that optionally has NaN gradients.
This serves as a hook to introduce NaN gradients to a model. This returns an
identity function. The identity's gradient function will check if the boolean
tensor `have_nan_gradients` is True. If so, the gradient will be NaN.
Otherwise, the gradient will also be the identity.
Args:
have_nan_gradients: A scalar boolean tensor. If True, gradients will be NaN.
Otherwise, the gradient function is the identity function.
Returns:
An identity function whose gradient function will return NaNs, if
`have_nan_gradients` is True."
7058,AssertTypeLayer,tensorflow/tensorflow/python/keras/mixed_precision/experimental/test_util.py,110,class,A layer which asserts it's inputs are a certain type.
7059,assert_input_types,tensorflow/tensorflow/python/keras/mixed_precision/experimental/test_util.py,118,method,Asserts `inputs` are of the correct type. Should be called in call().
7060,MultiplyLayer,tensorflow/tensorflow/python/keras/mixed_precision/experimental/test_util.py,128,class,A layer which multiplies its input by a scalar variable.
7061,build,tensorflow/tensorflow/python/keras/mixed_precision/experimental/test_util.py,155,method,
7062,call,tensorflow/tensorflow/python/keras/mixed_precision/experimental/test_util.py,160,method,
7063,get_config,tensorflow/tensorflow/python/keras/mixed_precision/experimental/test_util.py,171,method,
7064,IdentityRegularizer,tensorflow/tensorflow/python/keras/mixed_precision/experimental/test_util.py,180,class,
7065,get_config,tensorflow/tensorflow/python/keras/mixed_precision/experimental/test_util.py,186,method,
7066,Adadelta,tensorflow/tensorflow/python/keras/optimizer_v2/adadelta.py,32,class,"Optimizer that implements the Adadelta algorithm.
Adadelta optimization is a stochastic gradient descent method that is based on
adaptive learning rate per dimension to address two drawbacks:
- The continual decay of learning rates throughout training
- The need for a manually selected global learning rate
Adadelta is a more robust extension of Adagrad that adapts learning rates
based on a moving window of gradient updates, instead of accumulating all
past gradients. This way, Adadelta continues learning even when many updates
have been done. Compared to Adagrad, in the original version of Adadelta you
don't have to set an initial learning rate. In this version, initial
learning rate can be set, as in most other Keras optimizers.
According to section 4.3 (""Effective Learning rates""), near the end of
training step sizes converge to 1 which is effectively a high learning
rate which would cause divergence. This occurs only near the end of the
training as gradients and step sizes are small, and the epsilon constant
in the numerator and denominator dominate past gradients and parameter
updates which converge the learning rate to 1.
According to section 4.4(""Speech Data""),where a large neural network with
4 hidden layers was trained on a corpus of US English data, ADADELTA was
used with 100 network replicas.The epsilon used is 1e-6 with rho=0.95
which converged faster than ADAGRAD, by the following construction:
def __init__(self, lr=1.0, rho=0.95, epsilon=1e-6, decay=0., **kwargs):
Args:
learning_rate: A `Tensor`, floating point value, or a schedule that is a
`tf.keras.optimizers.schedules.LearningRateSchedule`. The learning rate.
To match the exact form in the original paper use 1.0.
rho: A `Tensor` or a floating point value. The decay rate.
epsilon: A `Tensor` or a floating point value. A constant epsilon used
to better conditioning the grad update.
name: Optional name prefix for the operations created when applying
gradients. Defaults to `""Adadelta""`.
**kwargs: Keyword arguments. Allowed to be one of
`""clipnorm""` or `""clipvalue""`.
`""clipnorm""` (float) clips gradients by norm; `""clipvalue""` (float) clips
gradients by value.
Reference:
- [Zeiler, 2012](http://arxiv.org/abs/1212.5701)"
7067,set_weights,tensorflow/tensorflow/python/keras/optimizer_v2/adadelta.py,107,method,
7068,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/adadelta.py,151,method,
7069,Adagrad,tensorflow/tensorflow/python/keras/optimizer_v2/adagrad.py,34,class,"Optimizer that implements the Adagrad algorithm.
Adagrad is an optimizer with parameter-specific learning rates,
which are adapted relative to how frequently a parameter gets
updated during training. The more updates a parameter receives,
the smaller the updates.
Args:
learning_rate: A `Tensor`, floating point value, or a schedule that is a
`tf.keras.optimizers.schedules.LearningRateSchedule`. The learning rate.
initial_accumulator_value: A floating point value.
Starting value for the accumulators, must be non-negative.
epsilon: A small floating point value to avoid zero denominator.
name: Optional name prefix for the operations created when applying
gradients. Defaults to `""Adagrad""`.
**kwargs: Keyword arguments. Allowed to be one of
`""clipnorm""` or `""clipvalue""`.
`""clipnorm""` (float) clips gradients by norm; `""clipvalue""` (float) clips
gradients by value.
Reference:
- [Duchi et al., 2011](
http://www.jmlr.org/papers/volume12/duchi11a/duchi11a.pdf)."
7070,set_weights,tensorflow/tensorflow/python/keras/optimizer_v2/adagrad.py,94,method,
7071,from_config,tensorflow/tensorflow/python/keras/optimizer_v2/adagrad.py,104,method,"Creates an optimizer from its config.
This method is the reverse of `get_config`,
capable of instantiating the same optimizer from the config
dictionary.
Arguments:
config: A Python dictionary, typically the output of get_config.
custom_objects: A Python dictionary mapping names to additional Python
objects used to create this optimizer, such as a function used for a
hyperparameter.
Returns:
An optimizer instance."
7072,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/adagrad.py,155,method,
7073,adagrad_update_numpy,tensorflow/tensorflow/python/keras/optimizer_v2/adagrad_test.py,45,function,
7074,sparse_adagrad_update_numpy,tensorflow/tensorflow/python/keras/optimizer_v2/adagrad_test.py,51,function,
7075,Adam,tensorflow/tensorflow/python/keras/optimizer_v2/adam.py,34,class,"Optimizer that implements the Adam algorithm.
Adam optimization is a stochastic gradient descent method that is based on
adaptive estimation of first-order and second-order moments.
According to
[Kingma et al., 2014](http://arxiv.org/abs/1412.6980),
the method is ""*computationally
efficient, has little memory requirement, invariant to diagonal rescaling of
gradients, and is well suited for problems that are large in terms of
data/parameters*"".
Args:
learning_rate: A `Tensor`, floating point value, or a schedule that is a
`tf.keras.optimizers.schedules.LearningRateSchedule`, or a callable
that takes no arguments and returns the actual value to use, The
learning rate. Defaults to 0.001.
beta_1: A float value or a constant float tensor, or a callable
that takes no arguments and returns the actual value to use. The
exponential decay rate for the 1st moment estimates. Defaults to 0.9.
beta_2: A float value or a constant float tensor, or a callable
that takes no arguments and returns the actual value to use, The
exponential decay rate for the 2nd moment estimates. Defaults to 0.999.
epsilon: A small constant for numerical stability. This epsilon is
""epsilon hat"" in the Kingma and Ba paper (in the formula just before
Section 2.1), not the epsilon in Algorithm 1 of the paper. Defaults to
1e-7.
amsgrad: Boolean. Whether to apply AMSGrad variant of this algorithm from
the paper ""On the Convergence of Adam and beyond"". Defaults to `False`.
name: Optional name for the operations created when applying gradients.
Defaults to `""Adam""`.
**kwargs: Keyword arguments. Allowed to be one of
`""clipnorm""` or `""clipvalue""`.
`""clipnorm""` (float) clips gradients by norm; `""clipvalue""` (float) clips
gradients by value.
Usage:
>>> opt = tf.keras.optimizers.Adam(learning_rate=0.1)
>>> var1 = tf.Variable(10.0)
>>> loss = lambda: (var1 ** 2)/2.0 # d(loss)/d(var1) == var1
>>> step_count = opt.minimize(loss, [var1]).numpy()
>>> # The first step is `-learning_rate*sign(grad)`
>>> var1.numpy()
9.9
Reference:
- [Kingma et al., 2014](http://arxiv.org/abs/1412.6980)
- [Reddi et al., 2018](
https://openreview.net/pdf?id=ryQu7f-RZ) for `amsgrad`.
Notes:
The default value of 1e-7 for epsilon might not be a good default in
general. For example, when training an Inception network on ImageNet a
current good choice is 1.0 or 0.1. Note that since Adam uses the
formulation just before Section 2.1 of the Kingma and Ba paper rather than
the formulation in Algorithm 1, the ""epsilon"" referred to here is ""epsilon
hat"" in the paper.
The sparse implementation of this algorithm (used when the gradient is an
IndexedSlices object, typically because of `tf.gather` or an embedding
lookup in the forward pass) does apply momentum to variable slices even if
they were not used in the forward pass (meaning they have a gradient equal
to zero). Momentum decay (beta1) is also applied to the entire momentum
accumulator. This means that the sparse behavior is equivalent to the dense
behavior (in contrast to some momentum implementations which ignore momentum
unless a variable slice was actually used)."
7076,set_weights,tensorflow/tensorflow/python/keras/optimizer_v2/adam.py,155,method,
7077,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/adam.py,242,method,
7078,NonFusedAdam,tensorflow/tensorflow/python/keras/optimizer_v2/adam.py,255,class,"Optimizer that implements the Adam algorithm without fused kernels.
Adam optimization is a stochastic gradient descent method that is based on
adaptive estimation of first-order and second-order moments.
According to the paper
[Adam: A Method for Stochastic Optimization. Kingma et al.,
2014](http://arxiv.org/abs/1412.6980), the method is ""*computationally
efficient, has little memory requirement, invariant to diagonal rescaling of
gradients, and is well suited for problems that are large in terms of
data/parameters*"".
For AMSGrad see [On The Convergence Of Adam And Beyond.
Reddi et al., 5-8](https://openreview.net/pdf?id=ryQu7f-RZ).
**If amsgrad = False**:
initialize $m_0$ as 1st moment vector
initialize $v_0$ as 2nd moment vector
The update rule for $\theta$ with gradient $g$ uses an optimization
described at the end of section 2 of the paper:
$$lr_t = \mathrm{learning\_rate} *
\sqrt{1 - \beta_2^t} / (1 - \beta_1^t)$$
$$m_t = \beta_1 * m_{t-1} + (1 - \beta_1) * g$$
$$v_t = \beta_2 * v_{t-1} + (1 - \beta_2) * g^2$$
$$\theta_t = \theta_{t-1} - lr_t * m_t / (\sqrt{v_t} + \epsilon)$$
**If amsgrad = True**:
initialize $m_0$ as 1st moment vector
initialize $v_0$ as 2nd moment vector
initialize $\hat{v}_0$ as 2nd moment vector
The update rule for $\theta$ with gradient $g$ uses an optimization
described at the end of section 2 of the paper:
$$lr_t = \mathrm{learning\_rate} *
\sqrt{1 - \beta_2^t} / (1 - \beta_1^t)$$
$$m_t = \beta_1 * m_{t-1} + (1 - \beta_1) * g$$
$$v_t = \beta_2 * v_{t-1} + (1 - \beta_2) * g^2$$
$$\hat{v}_t = \max(\hat{v}_{t-1}, v_t)$$
$$\theta_t = \theta_{t-1} - lr_t * m_t / (\sqrt{\hat{v}_t} + \epsilon)$$
The default value of 1e-7 for epsilon might not be a good default in
general. For example, when training an Inception network on ImageNet a
current good choice is 1.0 or 0.1. Note that since Adam uses the
formulation just before Section 2.1 of the Kingma and Ba paper rather than
the formulation in Algorithm 1, the ""epsilon"" referred to here is ""epsilon
hat"" in the paper.
The sparse implementation of this algorithm (used when the gradient is an
IndexedSlices object, typically because of `tf.gather` or an embedding
lookup in the forward pass) does apply momentum to variable slices even if
they were not used in the forward pass (meaning they have a gradient equal
to zero). Momentum decay (beta1) is also applied to the entire momentum
accumulator. This means that the sparse behavior is equivalent to the dense
behavior (in contrast to some momentum implementations which ignore momentum
unless a variable slice was actually used).
Usage:
>>> opt = tf.keras.optimizers.Adam(learning_rate=0.1)
>>> var1 = tf.Variable(10.0)
>>> loss = lambda: (var1 ** 2)/2.0 # d(loss)/d(var1) == var1
>>> step_count = opt.minimize(loss, [var1]).numpy()
>>> # The first step is `-learning_rate*sign(grad)`
>>> var1.numpy()
9.9"
7079,set_weights,tensorflow/tensorflow/python/keras/optimizer_v2/adam.py,407,method,
7080,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/adam.py,465,method,
7081,adam_update_numpy,tensorflow/tensorflow/python/keras/optimizer_v2/adam_test.py,38,function,
7082,adam_update_numpy_amsgrad,tensorflow/tensorflow/python/keras/optimizer_v2/adam_test.py,56,function,
7083,adam_sparse_update_numpy_amsgrad,tensorflow/tensorflow/python/keras/optimizer_v2/adam_test.py,76,function,
7084,get_beta_accumulators,tensorflow/tensorflow/python/keras/optimizer_v2/adam_test.py,102,function,
7085,Adamax,tensorflow/tensorflow/python/keras/optimizer_v2/adamax.py,33,class,"Optimizer that implements the Adamax algorithm.
It is a variant of Adam based on the infinity norm.
Default parameters follow those provided in the paper.
Adamax is sometimes superior to adam, specially in models with embeddings.
Initialization:
```python
m = 0 # Initialize initial 1st moment vector
v = 0 # Initialize the exponentially weighted infinity norm
t = 0 # Initialize timestep
```
The update rule for parameter `w` with gradient `g` is
described at the end of section 7.1 of the paper:
```python
t += 1
m = beta1 * m + (1 - beta) * g
v = max(beta2 * v, abs(g))
current_lr = learning_rate / (1 - beta1 ** t)
w = w - current_lr * m / (v + epsilon)
```
Similarly to `Adam`, the epsilon is added for numerical stability
(especially to get rid of division by zero when `v_t == 0`).
In contrast to `Adam`, the sparse implementation of this algorithm
(used when the gradient is an IndexedSlices object, typically because of
`tf.gather` or an embedding lookup in the forward pass) only updates
variable slices and corresponding `m_t`, `v_t` terms when that part of
the variable was used in the forward pass. This means that the sparse
behavior is contrast to the dense behavior (similar to some momentum
implementations which ignore momentum unless a variable slice was actually
used).
Args:
learning_rate: A `Tensor`, floating point value, or a schedule that is a
`tf.keras.optimizers.schedules.LearningRateSchedule`. The learning rate.
beta_1: A float value or a constant float tensor. The exponential decay
rate for the 1st moment estimates.
beta_2: A float value or a constant float tensor. The exponential decay
rate for the exponentially weighted infinity norm.
epsilon: A small constant for numerical stability.
name: Optional name for the operations created when applying gradients.
Defaults to `""Adamax""`.
**kwargs: Keyword arguments. Allowed to be one of
`""clipnorm""` or `""clipvalue""`.
`""clipnorm""` (float) clips gradients by norm; `""clipvalue""` (float) clips
gradients by value.
Reference:
- [Kingma et al., 2014](http://arxiv.org/abs/1412.6980)"
7086,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/adamax.py,178,method,
7087,adamax_update_numpy,tensorflow/tensorflow/python/keras/optimizer_v2/adamax_test.py,36,function,
7088,adamax_sparse_update_numpy,tensorflow/tensorflow/python/keras/optimizer_v2/adamax_test.py,51,function,
7089,get_beta_accumulators,tensorflow/tensorflow/python/keras/optimizer_v2/adamax_test.py,72,function,
7090,Ftrl,tensorflow/tensorflow/python/keras/optimizer_v2/ftrl.py,30,class,"Optimizer that implements the FTRL algorithm.
See Algorithm 1 of this [paper](
https://www.eecs.tufts.edu/~dsculley/papers/ad-click-prediction.pdf).
This version has support for both online L2 (the L2 penalty given in the paper
above) and shrinkage-type L2 (which is the addition of an L2 penalty to the
loss function).
Args:
learning_rate: A `Tensor`, floating point value, or a schedule that is a
`tf.keras.optimizers.schedules.LearningRateSchedule`. The learning rate.
learning_rate_power: A float value, must be less or equal to zero.
Controls how the learning rate decreases during training. Use zero for
a fixed learning rate.
initial_accumulator_value: The starting value for accumulators.
Only zero or positive values are allowed.
l1_regularization_strength: A float value, must be greater than or
equal to zero.
l2_regularization_strength: A float value, must be greater than or
equal to zero.
name: Optional name prefix for the operations created when applying
gradients. Defaults to `""Ftrl""`.
l2_shrinkage_regularization_strength: A float value, must be greater than
or equal to zero. This differs from L2 above in that the L2 above is a
stabilization penalty, whereas this L2 shrinkage is a magnitude penalty.
When input is sparse shrinkage will only happen on the active weights.
**kwargs: Keyword arguments. Allowed to be one of
`""clipnorm""` or `""clipvalue""`.
`""clipnorm""` (float) clips gradients by norm; `""clipvalue""` (float) clips
gradients by value.
Reference:
- [paper](
https://www.eecs.tufts.edu/~dsculley/papers/ad-click-prediction.pdf)"
7091,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/ftrl.py,195,method,
7092,SGD,tensorflow/tensorflow/python/keras/optimizer_v2/gradient_descent.py,30,class,"Gradient descent (with momentum) optimizer.
Update rule for parameter `w` with gradient `g` when `momentum` is 0:
```python
w = w - learning_rate * g
```
Update rule when `momentum` is larger than 0:
```python
velocity = momentum * velocity - learning_rate * g
w = w * velocity
```
When `nesterov=False`, this rule becomes:
```python
velocity = momentum * velocity - learning_rate * g
w = w + momentum * velocity - learning_rate * g
```
Args:
learning_rate: A `Tensor`, floating point value, or a schedule that is a
`tf.keras.optimizers.schedules.LearningRateSchedule`, or a callable
that takes no arguments and returns the actual value to use. The
learning rate. Defaults to 0.01.
momentum: float hyperparameter >= 0 that accelerates gradient descent
in the relevant
direction and dampens oscillations. Defaults to 0, i.e., vanilla gradient
descent.
nesterov: boolean. Whether to apply Nesterov momentum.
Defaults to `False`.
name: Optional name prefix for the operations created when applying
gradients. Defaults to `""SGD""`.
**kwargs: Keyword arguments. Allowed to be one of
`""clipnorm""` or `""clipvalue""`.
`""clipnorm""` (float) clips gradients by norm; `""clipvalue""` (float) clips
gradients by value.
Usage:
>>> opt = tf.keras.optimizers.SGD(learning_rate=0.1)
>>> var = tf.Variable(1.0)
>>> loss = lambda: (var ** 2)/2.0 # d(loss)/d(var1) = var1
>>> step_count = opt.minimize(loss, [var]).numpy()
>>> # Step is `- learning_rate * grad`
>>> var.numpy()
0.9
>>> opt = tf.keras.optimizers.SGD(learning_rate=0.1, momentum=0.9)
>>> var = tf.Variable(1.0)
>>> val0 = var.value()
>>> loss = lambda: (var ** 2)/2.0 # d(loss)/d(var1) = var1
>>> # First step is `- learning_rate * grad`
>>> step_count = opt.minimize(loss, [var]).numpy()
>>> val1 = var.value()
>>> (val0 - val1).numpy()
0.1
>>> # On later steps, step-size increases because of momentum
>>> step_count = opt.minimize(loss, [var]).numpy()
>>> val2 = var.value()
>>> (val1 - val2).numpy()
0.18
Reference:
- For `nesterov=True`, See [Sutskever et al., 2013](
http://jmlr.org/proceedings/papers/v28/sutskever13.pdf)."
7093,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/gradient_descent.py,186,method,
7094,LearningRateSchedule,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,34,class,"A serializable learning rate decay schedule.
`LearningRateSchedule`s can be passed in as the learning rate of optimizers in
`tf.keras.optimizers`. They can be serialized and deserialized using
`tf.keras.optimizers.schedules.serialize` and
`tf.keras.optimizers.schedules.deserialize`."
7095,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,48,method,
7096,from_config,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,52,method,"Instantiates a `LearningRateSchedule` from its config.
Args:
config: Output of `get_config()`.
Returns:
A `LearningRateSchedule` instance."
7097,ExponentialDecay,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,65,class,"A LearningRateSchedule that uses an exponential decay schedule.
When training a model, it is often recommended to lower the learning rate as
the training progresses. This schedule applies an exponential decay function
to an optimizer step, given a provided initial learning rate.
The schedule a 1-arg callable that produces a decayed learning
rate when passed the current optimizer step. This can be useful for changing
the learning rate value across different invocations of optimizer functions.
It is computed as:
```python
def decayed_learning_rate(step):
return initial_learning_rate * decay_rate ^ (step / decay_steps)
```
If the argument `staircase` is `True`, then `step / decay_steps` is
an integer division and the decayed learning rate follows a
staircase function.
You can pass this schedule directly into a `tf.keras.optimizers.Optimizer`
as the learning rate.
Example: When fitting a Keras model, decay every 100000 steps with a base
of 0.96:
```python
initial_learning_rate = 0.1
lr_schedule = tf.keras.optimizers.schedules.ExponentialDecay(
initial_learning_rate,
decay_steps=100000,
decay_rate=0.96,
staircase=True)
model.compile(optimizer=tf.keras.optimizers.SGD(learning_rate=lr_schedule),
loss='sparse_categorical_crossentropy',
metrics=['accuracy'])
model.fit(data, labels, epochs=5)
```
The learning rate schedule is also serializable and deserializable using
`tf.keras.optimizers.schedules.serialize` and
`tf.keras.optimizers.schedules.deserialize`.
Returns:
A 1-arg callable learning rate schedule that takes the current optimizer
step and outputs the decayed learning rate, a scalar `Tensor` of the same
type as `initial_learning_rate`."
7098,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,159,method,
7099,PiecewiseConstantDecay,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,170,class,"A LearningRateSchedule that uses a piecewise constant decay schedule.
The function returns a 1-arg callable to compute the piecewise constant
when passed the current optimizer step. This can be useful for changing the
learning rate value across different invocations of optimizer functions.
Example: use a learning rate that's 1.0 for the first 100001 steps, 0.5
for the next 10000 steps, and 0.1 for any additional steps.
```python
step = tf.Variable(0, trainable=False)
boundaries = [100000, 110000]
values = [1.0, 0.5, 0.1]
learning_rate_fn = keras.optimizers.schedules.PiecewiseConstantDecay(
boundaries, values)
# Later, whenever we perform an optimization step, we pass in the step.
learning_rate = learning_rate_fn(step)
```
You can pass this schedule directly into a `tf.keras.optimizers.Optimizer`
as the learning rate. The learning rate schedule is also serializable and
deserializable using `tf.keras.optimizers.schedules.serialize` and
`tf.keras.optimizers.schedules.deserialize`.
Returns:
A 1-arg callable learning rate schedule that takes the current optimizer
step and outputs the decayed learning rate, a scalar `Tensor` of the same
type as the boundary tensors.
The output of the 1-arg function that takes the `step`
is `values[0]` when `step <= boundaries[0]`,
`values[1]` when `step > boundaries[0]` and `step <= boundaries[1]`, ...,
and values[-1] when `step > boundaries[-1]`."
7100,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,263,method,
7101,PolynomialDecay,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,272,class,"A LearningRateSchedule that uses a polynomial decay schedule.
It is commonly observed that a monotonically decreasing learning rate, whose
degree of change is carefully chosen, results in a better performing model.
This schedule applies a polynomial decay function to an optimizer step,
given a provided `initial_learning_rate`, to reach an `end_learning_rate`
in the given `decay_steps`.
It requires a `step` value to compute the decayed learning rate. You
can just pass a TensorFlow variable that you increment at each training
step.
The schedule is a 1-arg callable that produces a decayed learning rate
when passed the current optimizer step. This can be useful for changing the
learning rate value across different invocations of optimizer functions.
It is computed as:
```python
def decayed_learning_rate(step):
step = min(step, decay_steps)
return ((initial_learning_rate - end_learning_rate) *
(1 - step / decay_steps) ^ (power)
) + end_learning_rate
```
If `cycle` is True then a multiple of `decay_steps` is used, the first one
that is bigger than `step`.
```python
def decayed_learning_rate(step):
decay_steps = decay_steps * ceil(step / decay_steps)
return ((initial_learning_rate - end_learning_rate) *
(1 - step / decay_steps) ^ (power)
) + end_learning_rate
```
You can pass this schedule directly into a `tf.keras.optimizers.Optimizer`
as the learning rate.
Example: Fit a model while decaying from 0.1 to 0.01 in 10000 steps using
sqrt (i.e. power=0.5):
```python
...
starter_learning_rate = 0.1
end_learning_rate = 0.01
decay_steps = 10000
learning_rate_fn = tf.keras.optimizers.schedules.PolynomialDecay(
starter_learning_rate,
decay_steps,
end_learning_rate,
power=0.5)
model.compile(optimizer=tf.keras.optimizers.SGD(
learning_rate=learning_rate_fn),
loss='sparse_categorical_crossentropy',
metrics=['accuracy'])
model.fit(data, labels, epochs=5)
```
The learning rate schedule is also serializable and deserializable using
`tf.keras.optimizers.schedules.serialize` and
`tf.keras.optimizers.schedules.deserialize`.
Returns:
A 1-arg callable learning rate schedule that takes the current optimizer
step and outputs the decayed learning rate, a scalar `Tensor` of the same
type as `initial_learning_rate`."
7102,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,404,method,
7103,InverseTimeDecay,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,416,class,"A LearningRateSchedule that uses an inverse time decay schedule.
When training a model, it is often recommended to lower the learning rate as
the training progresses. This schedule applies the inverse decay function
to an optimizer step, given a provided initial learning rate.
It requires a `step` value to compute the decayed learning rate. You can
just pass a TensorFlow variable that you increment at each training step.
The schedule a 1-arg callable that produces a decayed learning
rate when passed the current optimizer step. This can be useful for changing
the learning rate value across different invocations of optimizer functions.
It is computed as:
```python
def decayed_learning_rate(step):
return initial_learning_rate / (1 + decay_rate * step / decay_step)
```
or, if `staircase` is `True`, as:
```python
def decayed_learning_rate(step):
return initial_learning_rate / (1 + decay_rate * floor(step / decay_step))
```
You can pass this schedule directly into a `tf.keras.optimizers.Optimizer`
as the learning rate.
Example: Fit a Keras model when decaying 1/t with a rate of 0.5:
```python
...
initial_learning_rate = 0.1
decay_steps = 1.0
decay_rate = 0.5
learning_rate_fn = keras.optimizers.schedules.InverseTimeDecay(
initial_learning_rate, decay_steps, decay_rate)
model.compile(optimizer=tf.keras.optimizers.SGD(
learning_rate=learning_rate_fn),
loss='sparse_categorical_crossentropy',
metrics=['accuracy'])
model.fit(data, labels, epochs=5)
```
Returns:
A 1-arg callable learning rate schedule that takes the current optimizer
step and outputs the decayed learning rate, a scalar `Tensor` of the same
type as `initial_learning_rate`."
7104,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,511,method,
7105,CosineDecay,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,522,class,"A LearningRateSchedule that uses a cosine decay schedule.
See [Loshchilov & Hutter, ICLR2016], SGDR: Stochastic Gradient Descent
with Warm Restarts. https://arxiv.org/abs/1608.03983
When training a model, it is often recommended to lower the learning rate as
the training progresses. This schedule applies a cosine decay function
to an optimizer step, given a provided initial learning rate.
It requires a `step` value to compute the decayed learning rate. You can
just pass a TensorFlow variable that you increment at each training step.
The schedule a 1-arg callable that produces a decayed learning
rate when passed the current optimizer step. This can be useful for changing
the learning rate value across different invocations of optimizer functions.
It is computed as:
```python
def decayed_learning_rate(step):
step = min(step, decay_steps)
cosine_decay = 0.5 * (1 + cos(pi * step / decay_steps))
decayed = (1 - alpha) * cosine_decay + alpha
return initial_learning_rate * decayed
```
Example usage:
```python
decay_steps = 1000
lr_decayed_fn = tf.keras.experimental.CosineDecay(
initial_learning_rate, decay_steps)
```
You can pass this schedule directly into a `tf.keras.optimizers.Optimizer`
as the learning rate. The learning rate schedule is also serializable and
deserializable using `tf.keras.optimizers.schedules.serialize` and
`tf.keras.optimizers.schedules.deserialize`.
Returns:
A 1-arg callable learning rate schedule that takes the current optimizer
step and outputs the decayed learning rate, a scalar `Tensor` of the same
type as `initial_learning_rate`."
7106,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,605,method,
7107,CosineDecayRestarts,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,615,class,"A LearningRateSchedule that uses a cosine decay schedule with restarts.
See [Loshchilov & Hutter, ICLR2016], SGDR: Stochastic Gradient Descent
with Warm Restarts. https://arxiv.org/abs/1608.03983
When training a model, it is often recommended to lower the learning rate as
the training progresses. This schedule applies a cosine decay function with
restarts to an optimizer step, given a provided initial learning rate.
It requires a `step` value to compute the decayed learning rate. You can
just pass a TensorFlow variable that you increment at each training step.
The schedule a 1-arg callable that produces a decayed learning
rate when passed the current optimizer step. This can be useful for changing
the learning rate value across different invocations of optimizer functions.
The learning rate multiplier first decays
from 1 to `alpha` for `first_decay_steps` steps. Then, a warm
restart is performed. Each new warm restart runs for `t_mul` times more
steps and with `m_mul` times smaller initial learning rate.
Example usage:
```python
first_decay_steps = 1000
lr_decayed_fn = (
tf.keras.experimental.CosineDecayRestarts(
initial_learning_rate,
first_decay_steps))
```
You can pass this schedule directly into a `tf.keras.optimizers.Optimizer`
as the learning rate. The learning rate schedule is also serializable and
deserializable using `tf.keras.optimizers.schedules.serialize` and
`tf.keras.optimizers.schedules.deserialize`.
Returns:
A 1-arg callable learning rate schedule that takes the current optimizer
step and outputs the decayed learning rate, a scalar `Tensor` of the same
type as `initial_learning_rate`."
7108,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,729,method,
7109,compute_step,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,701,method,Helper for `cond` operation.
7110,LinearCosineDecay,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,741,class,"A LearningRateSchedule that uses a linear cosine decay schedule.
See [Bello et al., ICML2017] Neural Optimizer Search with RL.
https://arxiv.org/abs/1709.07417
For the idea of warm starts here controlled by `num_periods`,
see [Loshchilov & Hutter, ICLR2016] SGDR: Stochastic Gradient Descent
with Warm Restarts. https://arxiv.org/abs/1608.03983
Note that linear cosine decay is more aggressive than cosine decay and
larger initial learning rates can typically be used.
When training a model, it is often recommended to lower the learning rate as
the training progresses. This schedule applies a linear cosine decay
function to an optimizer step, given a provided initial learning rate.
It requires a `step` value to compute the decayed learning rate. You can
just pass a TensorFlow variable that you increment at each training step.
The schedule a 1-arg callable that produces a decayed learning
rate when passed the current optimizer step. This can be useful for changing
the learning rate value across different invocations of optimizer functions.
It is computed as:
```python
def decayed_learning_rate(step):
step = min(step, decay_steps)
linear_decay = (decay_steps - step) / decay_steps
cosine_decay = 0.5 * (
1 + cos(pi * 2 * num_periods * step / decay_steps))
decayed = (alpha + linear_decay) * cosine_decay + beta
return initial_learning_rate * decayed
```
Example usage:
```python
decay_steps = 1000
lr_decayed_fn = (
tf.keras.experimental.LinearCosineDecay(
initial_learning_rate, decay_steps))
```
You can pass this schedule directly into a `tf.keras.optimizers.Optimizer`
as the learning rate. The learning rate schedule is also serializable and
deserializable using `tf.keras.optimizers.schedules.serialize` and
`tf.keras.optimizers.schedules.deserialize`.
Returns:
A 1-arg callable learning rate schedule that takes the current optimizer
step and outputs the decayed learning rate, a scalar `Tensor` of the same
type as `initial_learning_rate`."
7111,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,847,method,
7112,NoisyLinearCosineDecay,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,859,class,"A LearningRateSchedule that uses a noisy linear cosine decay schedule.
See [Bello et al., ICML2017] Neural Optimizer Search with RL.
https://arxiv.org/abs/1709.07417
For the idea of warm starts here controlled by `num_periods`,
see [Loshchilov & Hutter, ICLR2016] SGDR: Stochastic Gradient Descent
with Warm Restarts. https://arxiv.org/abs/1608.03983
Note that linear cosine decay is more aggressive than cosine decay and
larger initial learning rates can typically be used.
When training a model, it is often recommended to lower the learning rate as
the training progresses. This schedule applies a noisy linear cosine decay
function to an optimizer step, given a provided initial learning rate.
It requires a `step` value to compute the decayed learning rate. You can
just pass a TensorFlow variable that you increment at each training step.
The schedule a 1-arg callable that produces a decayed learning
rate when passed the current optimizer step. This can be useful for changing
the learning rate value across different invocations of optimizer functions.
It is computed as:
```python
def decayed_learning_rate(step):
step = min(step, decay_steps)
linear_decay = (decay_steps - step) / decay_steps)
cosine_decay = 0.5 * (
1 + cos(pi * 2 * num_periods * step / decay_steps))
decayed = (alpha + linear_decay + eps_t) * cosine_decay + beta
return initial_learning_rate * decayed
```
where eps_t is 0-centered gaussian noise with variance
initial_variance / (1 + global_step) ** variance_decay
Example usage:
```python
decay_steps = 1000
lr_decayed_fn = (
tf.keras.experimental.NoisyLinearCosineDecay(
initial_learning_rate, decay_steps))
```
You can pass this schedule directly into a `tf.keras.optimizers.Optimizer`
as the learning rate. The learning rate schedule is also serializable and
deserializable using `tf.keras.optimizers.schedules.serialize` and
`tf.keras.optimizers.schedules.deserialize`.
Returns:
A 1-arg callable learning rate schedule that takes the current optimizer
step and outputs the decayed learning rate, a scalar `Tensor` of the same
type as `initial_learning_rate`."
7113,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,983,method,
7114,serialize,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,997,function,
7115,deserialize,tensorflow/tensorflow/python/keras/optimizer_v2/learning_rate_schedule.py,1002,function,
7116,exponential_decay,tensorflow/tensorflow/python/keras/optimizer_v2/legacy_learning_rate_decay.py,32,function,"Applies exponential decay to the learning rate.
When training a model, it is often recommended to lower the learning rate as
the training progresses. This function applies an exponential decay function
to a provided initial learning rate. It requires a `global_step` value to
compute the decayed learning rate. You can just pass a TensorFlow variable
that you increment at each training step.
The function returns the decayed learning rate. It is computed as:
```python
decayed_learning_rate = learning_rate *
decay_rate ^ (global_step / decay_steps)
```
If the argument `staircase` is `True`, then `global_step / decay_steps` is an
integer division and the decayed learning rate follows a staircase function.
Example: decay every 100000 steps with a base of 0.96:
```python
...
global_step = tf.Variable(0, trainable=False)
starter_learning_rate = 0.1
learning_rate = tf.compat.v1.train.exponential_decay(starter_learning_rate,
global_step,
100000, 0.96, staircase=True)
# Passing global_step to minimize() will increment it at each step.
learning_step = (
tf.compat.v1.train.GradientDescentOptimizer(learning_rate)
.minimize(...my loss..., global_step=global_step)
)
```
Args:
learning_rate: A scalar `float32` or `float64` `Tensor` or a Python number.
The initial learning rate.
global_step: A scalar `int32` or `int64` `Tensor` or a Python number. Global
step to use for the decay computation. Must not be negative.
decay_steps: A scalar `int32` or `int64` `Tensor` or a Python number. Must
be positive. See the decay computation above.
decay_rate: A scalar `float32` or `float64` `Tensor` or a Python number.
The decay rate.
staircase: Boolean. If `True` decay the learning rate at discrete intervals
name: String. Optional name of the operation. Defaults to
'ExponentialDecay'.
Returns:
A scalar `Tensor` of the same type as `learning_rate`. The decayed
learning rate.
Raises:
ValueError: if `global_step` is not supplied.
@compatibility(eager)
When eager execution is enabled, this function returns a function which in
turn returns the decayed learning rate Tensor. This can be useful for changing
the learning rate value across different invocations of optimizer functions.
@end_compatibility"
7117,piecewise_constant,tensorflow/tensorflow/python/keras/optimizer_v2/legacy_learning_rate_decay.py,108,function,"Piecewise constant from boundaries and interval values.
Example: use a learning rate that's 1.0 for the first 100001 steps, 0.5
for the next 10000 steps, and 0.1 for any additional steps.
```python
global_step = tf.Variable(0, trainable=False)
boundaries = [100000, 110000]
values = [1.0, 0.5, 0.1]
learning_rate = tf.compat.v1.train.piecewise_constant(global_step, boundaries,
values)
# Later, whenever we perform an optimization step, we increment global_step.
```
Args:
x: A 0-D scalar `Tensor`. Must be one of the following types: `float32`,
`float64`, `uint8`, `int8`, `int16`, `int32`, `int64`.
boundaries: A list of `Tensor`s or `int`s or `float`s with strictly
increasing entries, and with all elements having the same type as `x`.
values: A list of `Tensor`s or `float`s or `int`s that specifies the values
for the intervals defined by `boundaries`. It should have one more element
than `boundaries`, and all elements should have the same type.
name: A string. Optional name of the operation. Defaults to
'PiecewiseConstant'.
Returns:
A 0-D Tensor. Its value is `values[0]` when `x <= boundaries[0]`,
`values[1]` when `x > boundaries[0]` and `x <= boundaries[1]`, ...,
and values[-1] when `x > boundaries[-1]`.
Raises:
ValueError: if types of `x` and `boundaries` do not match, or types of all
`values` do not match or
the number of elements in the lists does not match.
@compatibility(eager)
When eager execution is enabled, this function returns a function which in
turn returns the decayed learning rate Tensor. This can be useful for changing
the learning rate value across different invocations of optimizer functions.
@end_compatibility"
7118,polynomial_decay,tensorflow/tensorflow/python/keras/optimizer_v2/legacy_learning_rate_decay.py,185,function,"Applies a polynomial decay to the learning rate.
It is commonly observed that a monotonically decreasing learning rate, whose
degree of change is carefully chosen, results in a better performing model.
This function applies a polynomial decay function to a provided initial
`learning_rate` to reach an `end_learning_rate` in the given `decay_steps`.
It requires a `global_step` value to compute the decayed learning rate. You
can just pass a TensorFlow variable that you increment at each training step.
The function returns the decayed learning rate. It is computed as:
```python
global_step = min(global_step, decay_steps)
decayed_learning_rate = (learning_rate - end_learning_rate) *
(1 - global_step / decay_steps) ^ (power) +
end_learning_rate
```
If `cycle` is True then a multiple of `decay_steps` is used, the first one
that is bigger than `global_steps`.
```python
decay_steps = decay_steps * ceil(global_step / decay_steps)
decayed_learning_rate = (learning_rate - end_learning_rate) *
(1 - global_step / decay_steps) ^ (power) +
end_learning_rate
```
Example: decay from 0.1 to 0.01 in 10000 steps using sqrt (i.e. power=0.5):
```python
...
global_step = tf.Variable(0, trainable=False)
starter_learning_rate = 0.1
end_learning_rate = 0.01
decay_steps = 10000
learning_rate = tf.compat.v1.train.polynomial_decay(starter_learning_rate,
global_step,
decay_steps, end_learning_rate,
power=0.5)
# Passing global_step to minimize() will increment it at each step.
learning_step = (
tf.compat.v1.train.GradientDescentOptimizer(learning_rate)
.minimize(...my loss..., global_step=global_step)
)
```
Args:
learning_rate: A scalar `float32` or `float64` `Tensor` or a Python number.
The initial learning rate.
global_step: A scalar `int32` or `int64` `Tensor` or a Python number. Global
step to use for the decay computation. Must not be negative.
decay_steps: A scalar `int32` or `int64` `Tensor` or a Python number. Must
be positive. See the decay computation above.
end_learning_rate: A scalar `float32` or `float64` `Tensor` or a Python
number. The minimal end learning rate.
power: A scalar `float32` or `float64` `Tensor` or a Python number. The
power of the polynomial. Defaults to linear, 1.0.
cycle: A boolean, whether or not it should cycle beyond decay_steps.
name: String. Optional name of the operation. Defaults to
'PolynomialDecay'.
Returns:
A scalar `Tensor` of the same type as `learning_rate`. The decayed
learning rate.
Raises:
ValueError: if `global_step` is not supplied.
@compatibility(eager)
When eager execution is enabled, this function returns a function which in
turn returns the decayed learning rate Tensor. This can be useful for changing
the learning rate value across different invocations of optimizer functions.
@end_compatibility"
7119,natural_exp_decay,tensorflow/tensorflow/python/keras/optimizer_v2/legacy_learning_rate_decay.py,286,function,"Applies natural exponential decay to the initial learning rate.
When training a model, it is often recommended to lower the learning rate as
the training progresses. This function applies an exponential decay function
to a provided initial learning rate. It requires an `global_step` value to
compute the decayed learning rate. You can just pass a TensorFlow variable
that you increment at each training step.
The function returns the decayed learning rate. It is computed as:
```python
decayed_learning_rate = learning_rate * exp(-decay_rate * global_step /
decay_step)
```
or, if `staircase` is `True`, as:
```python
decayed_learning_rate = learning_rate * exp(-decay_rate * floor(global_step /
decay_step))
```
Example: decay exponentially with a base of 0.96:
```python
...
global_step = tf.Variable(0, trainable=False)
learning_rate = 0.1
decay_steps = 5
k = 0.5
learning_rate = tf.compat.v1.train.natural_exp_decay(learning_rate,
global_step,
decay_steps, k)
# Passing global_step to minimize() will increment it at each step.
learning_step = (
tf.compat.v1.train.GradientDescentOptimizer(learning_rate)
.minimize(...my loss..., global_step=global_step)
)
```
Args:
learning_rate: A scalar `float32` or `float64` `Tensor` or a Python number.
The initial learning rate.
global_step: A Python number. Global step to use for the decay computation.
Must not be negative.
decay_steps: How often to apply decay.
decay_rate: A Python number. The decay rate.
staircase: Whether to apply decay in a discrete staircase, as opposed to
continuous, fashion.
name: String. Optional name of the operation. Defaults to
'ExponentialTimeDecay'.
Returns:
A scalar `Tensor` of the same type as `learning_rate`. The decayed
learning rate.
Raises:
ValueError: if `global_step` is not supplied.
@compatibility(eager)
When eager execution is enabled, this function returns a function which in
turn returns the decayed learning rate Tensor. This can be useful for changing
the learning rate value across different invocations of optimizer functions.
@end_compatibility"
7120,inverse_time_decay,tensorflow/tensorflow/python/keras/optimizer_v2/legacy_learning_rate_decay.py,374,function,"Applies inverse time decay to the initial learning rate.
When training a model, it is often recommended to lower the learning rate as
the training progresses. This function applies an inverse decay function
to a provided initial learning rate. It requires an `global_step` value to
compute the decayed learning rate. You can just pass a TensorFlow variable
that you increment at each training step.
The function returns the decayed learning rate. It is computed as:
```python
decayed_learning_rate = learning_rate / (1 + decay_rate * global_step /
decay_step)
```
or, if `staircase` is `True`, as:
```python
decayed_learning_rate = learning_rate / (1 + decay_rate * floor(global_step /
decay_step))
```
Example: decay 1/t with a rate of 0.5:
```python
...
global_step = tf.Variable(0, trainable=False)
learning_rate = 0.1
decay_steps = 1.0
decay_rate = 0.5
learning_rate = tf.compat.v1.train.inverse_time_decay(learning_rate,
global_step,
decay_steps, decay_rate)
# Passing global_step to minimize() will increment it at each step.
learning_step = (
tf.compat.v1.train.GradientDescentOptimizer(learning_rate)
.minimize(...my loss..., global_step=global_step)
)
```
Args:
learning_rate: A scalar `float32` or `float64` `Tensor` or a Python number.
The initial learning rate.
global_step: A Python number. Global step to use for the decay computation.
Must not be negative.
decay_steps: How often to apply decay.
decay_rate: A Python number. The decay rate.
staircase: Whether to apply decay in a discrete staircase, as opposed to
continuous, fashion.
name: String. Optional name of the operation. Defaults to
'InverseTimeDecay'.
Returns:
A scalar `Tensor` of the same type as `learning_rate`. The decayed
learning rate.
Raises:
ValueError: if `global_step` is not supplied.
@compatibility(eager)
When eager execution is enabled, this function returns a function which in
turn returns the decayed learning rate Tensor. This can be useful for changing
the learning rate value across different invocations of optimizer functions.
@end_compatibility"
7121,cosine_decay,tensorflow/tensorflow/python/keras/optimizer_v2/legacy_learning_rate_decay.py,457,function,"Applies cosine decay to the learning rate.
When training a model, it is often recommended to lower the learning rate as
the training progresses. This function applies a cosine decay function
to a provided initial learning rate. It requires a `global_step` value to
compute the decayed learning rate. You can just pass a TensorFlow variable
that you increment at each training step.
The function returns the decayed learning rate. It is computed as:
```python
global_step = min(global_step, decay_steps)
cosine_decay = 0.5 * (1 + cos(pi * global_step / decay_steps))
decayed = (1 - alpha) * cosine_decay + alpha
decayed_learning_rate = learning_rate * decayed
```
Example usage:
```python
decay_steps = 1000
lr_decayed = cosine_decay(learning_rate, global_step, decay_steps)
```
Args:
learning_rate: A scalar `float32` or `float64` Tensor or a Python number.
The initial learning rate.
global_step: A scalar `int32` or `int64` `Tensor` or a Python number. Global
step to use for the decay computation.
decay_steps: A scalar `int32` or `int64` `Tensor` or a Python number. Number
of steps to decay over.
alpha: A scalar `float32` or `float64` Tensor or a Python number. Minimum
learning rate value as a fraction of learning_rate.
name: String. Optional name of the operation. Defaults to 'CosineDecay'.
Returns:
A scalar `Tensor` of the same type as `learning_rate`. The decayed
learning rate.
Raises:
ValueError: if `global_step` is not supplied.
References:
Stochastic Gradient Descent with Warm Restarts:
[Loshchilov et al., 2017]
(https://openreview.net/forum?id=Skq89Scxx¬eId=Skq89Scxx)
([pdf](https://openreview.net/pdf?id=Skq89Scxx))
@compatibility(eager)
When eager execution is enabled, this function returns a function which in
turn returns the decayed learning rate Tensor. This can be useful for changing
the learning rate value across different invocations of optimizer functions.
@end_compatibility"
7122,cosine_decay_restarts,tensorflow/tensorflow/python/keras/optimizer_v2/legacy_learning_rate_decay.py,520,function,"Applies cosine decay with restarts to the learning rate.
When training a model, it is often recommended to lower the learning rate as
the training progresses. This function applies a cosine decay function with
restarts to a provided initial learning rate. It requires a `global_step`
value to compute the decayed learning rate. You can just pass a TensorFlow
variable that you increment at each training step.
The function returns the decayed learning rate while taking into account
possible warm restarts. The learning rate multiplier first decays
from 1 to `alpha` for `first_decay_steps` steps. Then, a warm
restart is performed. Each new warm restart runs for `t_mul` times more steps
and with `m_mul` times smaller initial learning rate.
Example usage:
```python
first_decay_steps = 1000
lr_decayed = cosine_decay_restarts(learning_rate, global_step,
first_decay_steps)
```
Args:
learning_rate: A scalar `float32` or `float64` Tensor or a Python number.
The initial learning rate.
global_step: A scalar `int32` or `int64` `Tensor` or a Python number. Global
step to use for the decay computation.
first_decay_steps: A scalar `int32` or `int64` `Tensor` or a Python number.
Number of steps to decay over.
t_mul: A scalar `float32` or `float64` `Tensor` or a Python number. Used to
derive the number of iterations in the i-th period
m_mul: A scalar `float32` or `float64` `Tensor` or a Python number.
Used to derive the initial learning rate of the i-th period:
alpha: A scalar `float32` or `float64` Tensor or a Python number. Minimum
learning rate value as a fraction of the learning_rate.
name: String. Optional name of the operation. Defaults to 'SGDRDecay'.
Returns:
A scalar `Tensor` of the same type as `learning_rate`. The decayed
learning rate.
Raises:
ValueError: if `global_step` is not supplied.
References:
Stochastic Gradient Descent with Warm Restarts:
[Loshchilov et al., 2017]
(https://openreview.net/forum?id=Skq89Scxx¬eId=Skq89Scxx)
([pdf](https://openreview.net/pdf?id=Skq89Scxx))
@compatibility(eager)
When eager execution is enabled, this function returns a function which in
turn returns the decayed learning rate Tensor. This can be useful for changing
the learning rate value across different invocations of optimizer functions.
@end_compatibility"
7123,linear_cosine_decay,tensorflow/tensorflow/python/keras/optimizer_v2/legacy_learning_rate_decay.py,597,function,"Applies linear cosine decay to the learning rate.
Note that linear cosine decay is more aggressive than cosine decay and
larger initial learning rates can typically be used.
When training a model, it is often recommended to lower the learning rate as
the training progresses. This function applies a linear cosine decay function
to a provided initial learning rate. It requires a `global_step` value to
compute the decayed learning rate. You can just pass a TensorFlow variable
that you increment at each training step.
The function returns the decayed learning rate. It is computed as:
```python
global_step = min(global_step, decay_steps)
linear_decay = (decay_steps - global_step) / decay_steps)
cosine_decay = 0.5 * (
1 + cos(pi * 2 * num_periods * global_step / decay_steps))
decayed = (alpha + linear_decay) * cosine_decay + beta
decayed_learning_rate = learning_rate * decayed
```
Example usage:
```python
decay_steps = 1000
lr_decayed = linear_cosine_decay(learning_rate, global_step, decay_steps)
```
Args:
learning_rate: A scalar `float32` or `float64` Tensor or a Python number.
The initial learning rate.
global_step: A scalar `int32` or `int64` `Tensor` or a Python number. Global
step to use for the decay computation.
decay_steps: A scalar `int32` or `int64` `Tensor` or a Python number. Number
of steps to decay over.
num_periods: Number of periods in the cosine part of the decay. See
computation above.
alpha: See computation above.
beta: See computation above.
name: String. Optional name of the operation. Defaults to
'LinearCosineDecay'.
Returns:
A scalar `Tensor` of the same type as `learning_rate`. The decayed
learning rate.
Raises:
ValueError: if `global_step` is not supplied.
References:
Neural Optimizer Search with Reinforcement Learning:
[Bello et al., 2017](http://proceedings.mlr.press/v70/bello17a.html)
([pdf](http://proceedings.mlr.press/v70/bello17a/bello17a.pdf))
Stochastic Gradient Descent with Warm Restarts:
[Loshchilov et al., 2017]
(https://openreview.net/forum?id=Skq89Scxx¬eId=Skq89Scxx)
([pdf](https://openreview.net/pdf?id=Skq89Scxx))
@compatibility(eager)
When eager execution is enabled, this function returns a function which in
turn returns the decayed learning rate Tensor. This can be useful for changing
the learning rate value across different invocations of optimizer functions.
@end_compatibility"
7124,noisy_linear_cosine_decay,tensorflow/tensorflow/python/keras/optimizer_v2/legacy_learning_rate_decay.py,682,function,"Applies noisy linear cosine decay to the learning rate.
Note that linear cosine decay is more aggressive than cosine decay and
larger initial learning rates can typically be used.
When training a model, it is often recommended to lower the learning rate as
the training progresses. This function applies a noisy linear
cosine decay function to a provided initial learning rate.
It requires a `global_step` value to compute the decayed learning rate.
You can just pass a TensorFlow variable that you increment at each
training step.
The function returns the decayed learning rate. It is computed as:
```python
global_step = min(global_step, decay_steps)
linear_decay = (decay_steps - global_step) / decay_steps)
cosine_decay = 0.5 * (
1 + cos(pi * 2 * num_periods * global_step / decay_steps))
decayed = (alpha + linear_decay + eps_t) * cosine_decay + beta
decayed_learning_rate = learning_rate * decayed
```
where eps_t is 0-centered gaussian noise with variance
initial_variance / (1 + global_step) ** variance_decay
Example usage:
```python
decay_steps = 1000
lr_decayed = noisy_linear_cosine_decay(
learning_rate, global_step, decay_steps)
```
Args:
learning_rate: A scalar `float32` or `float64` Tensor or a Python number.
The initial learning rate.
global_step: A scalar `int32` or `int64` `Tensor` or a Python number. Global
step to use for the decay computation.
decay_steps: A scalar `int32` or `int64` `Tensor` or a Python number. Number
of steps to decay over.
initial_variance: initial variance for the noise. See computation above.
variance_decay: decay for the noise's variance. See computation above.
num_periods: Number of periods in the cosine part of the decay. See
computation above.
alpha: See computation above.
beta: See computation above.
name: String. Optional name of the operation. Defaults to
'NoisyLinearCosineDecay'.
Returns:
A scalar `Tensor` of the same type as `learning_rate`. The decayed
learning rate.
Raises:
ValueError: if `global_step` is not supplied.
References:
Neural Optimizer Search with Reinforcement Learning:
[Bello et al., 2017](http://proceedings.mlr.press/v70/bello17a.html)
([pdf](http://proceedings.mlr.press/v70/bello17a/bello17a.pdf))
Stochastic Gradient Descent with Warm Restarts:
[Loshchilov et al., 2017]
(https://openreview.net/forum?id=Skq89Scxx¬eId=Skq89Scxx)
([pdf](https://openreview.net/pdf?id=Skq89Scxx))
@compatibility(eager)
When eager execution is enabled, this function returns a function which in
turn returns the decayed learning rate Tensor. This can be useful for changing
the learning rate value across different invocations of optimizer functions.
@end_compatibility"
7125,Nadam,tensorflow/tensorflow/python/keras/optimizer_v2/nadam.py,34,class,"Optimizer that implements the NAdam algorithm.
Much like Adam is essentially RMSprop with momentum, Nadam is Adam with
Nesterov momentum.
Args:
learning_rate: A Tensor or a floating point value. The learning rate.
beta_1: A float value or a constant float tensor. The exponential decay
rate for the 1st moment estimates.
beta_2: A float value or a constant float tensor. The exponential decay
rate for the exponentially weighted infinity norm.
epsilon: A small constant for numerical stability.
name: Optional name for the operations created when applying gradients.
Defaults to `""Nadam""`.
**kwargs: Keyword arguments. Allowed to be one of
`""clipnorm""` or `""clipvalue""`.
`""clipnorm""` (float) clips gradients by norm; `""clipvalue""` (float) clips
gradients by value.
Reference:
- [Dozat, 2015](http://cs229.stanford.edu/proj2015/054_report.pdf)."
7126,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/nadam.py,206,method,
7127,get_beta_accumulators,tensorflow/tensorflow/python/keras/optimizer_v2/nadam_test.py,32,function,
7128,update_m_cache,tensorflow/tensorflow/python/keras/optimizer_v2/nadam_test.py,41,function,
7129,nadam_update_numpy,tensorflow/tensorflow/python/keras/optimizer_v2/nadam_test.py,47,function,
7130,OptimizerV2,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,81,class,"Base class for Keras optimizers.
You should not use this class directly, but instead instantiate one of its
subclasses such as `tf.keras.optimizers.SGD`, `tf.keras.optimizers.Adam`, etc.
### Usage
```python
# Create an optimizer with the desired parameters.
opt = tf.keras.optimizers.SGD(learning_rate=0.1)
# `loss` is a callable that takes no argument and returns the value
# to minimize.
loss = lambda: 3 * var1 * var1 + 2 * var2 * var2
# In graph mode, returns op that minimizes the loss by updating the listed
# variables.
opt_op = opt.minimize(loss, var_list=[var1, var2])
opt_op.run()
# In eager mode, simply call minimize to update the list of variables.
opt.minimize(loss, var_list=[var1, var2])
```
### Usage in custom training loops
In Keras models, sometimes variables are created when the model is first
called, instead of construction time. Examples include 1) sequential models
without input shape pre-defined, or 2) subclassed models. Pass var_list as
callable in these cases.
Example:
```python
opt = tf.keras.optimizers.SGD(learning_rate=0.1)
model = tf.keras.Sequential()
model.add(tf.keras.layers.Dense(num_hidden, activation='relu'))
model.add(tf.keras.layers.Dense(num_classes, activation='sigmoid'))
loss_fn = lambda: tf.keras.losses.mse(model(input), output)
var_list_fn = lambda: model.trainable_weights
for input, output in data:
opt.minimize(loss_fn, var_list_fn)
```
### Processing gradients before applying them
Calling `minimize()` takes care of both computing the gradients and
applying them to the variables. If you want to process the gradients
before applying them you can instead use the optimizer in three steps:
1. Compute the gradients with `tf.GradientTape`.
2. Process the gradients as you wish.
3. Apply the processed gradients with `apply_gradients()`.
Example:
```python
# Create an optimizer.
opt = tf.keras.optimizers.SGD(learning_rate=0.1)
# Compute the gradients for a list of variables.
with tf.GradientTape() as tape:
loss = <call_loss_function>
vars = <list_of_variables>
grads = tape.gradient(loss, vars)
# Process the gradients, for example cap them, etc.
# capped_grads = [MyCapper(g) for g in grads]
processed_grads = [process_gradient(g) for g in grads]
# Ask the optimizer to apply the processed gradients.
opt.apply_gradients(zip(processed_grads, var_list))
```
### Use with `tf.distribute.Strategy`
This optimizer class is `tf.distribute.Strategy` aware, which means it
automatically sums gradients across all replicas. To average gradients,
you divide your loss by the global batch size, which is done
automatically if you use `tf.keras` built-in training or evaluation loops.
See the `reduction` argument of your loss which should be set to
`tf.keras.losses.Reduction.SUM_OVER_BATCH_SIZE` for averaging or
`tf.keras.losses.Reduction.SUM` for not.
To aggregate gradients yourself, call `apply_gradients` with
`experimental_aggregate_gradients` set to False. This is useful if you need to
process aggregated gradients.
If you are not using these and you want to average gradients, you should use
`tf.math.reduce_sum` to add up your per-example losses and then divide by the
global batch size. Note that when using `tf.distribute.Strategy`, the first
component of a tensor's shape is the *replica-local* batch size, which is off
by a factor equal to the number of replicas being used to compute a single
step. As a result, using `tf.math.reduce_mean` will give the wrong answer,
resulting in gradients that can be many times too big.
### Variable Constraints
All Keras optimizers respect variable constraints. If constraint function is
passed to any variable, the constraint will be applied to the variable after
the gradient has been applied to the variable.
Important: If gradient is sparse tensor, variable constraint is not supported.
### Thread Compatibility
The entire optimizer is currently thread compatible, not thread-safe. The user
needs to perform synchronization if necessary.
### Slots
Many optimizer subclasses, such as `Adam` and `Adagrad` allocate and manage
additional variables associated with the variables to train. These are called
<i>Slots</i>. Slots have names and you can ask the optimizer for the names of
the slots that it uses. Once you have a slot name you can ask the optimizer
for the variable it created to hold the slot value.
This can be useful if you want to log debug a training algorithm, report stats
about the slots, etc.
### Hyperparameters
These are arguments passed to the optimizer subclass constructor
(the `__init__` method), and then passed to `self._set_hyper()`.
They can be either regular Python values (like 1.0), tensors, or
callables. If they are callable, the callable will be called during
`apply_gradients()` to get the value for the hyper parameter.
Hyperparameters can be overwritten through user code:
Example:
```python
# Create an optimizer with the desired parameters.
opt = tf.keras.optimizers.SGD(learning_rate=0.1)
# `loss` is a callable that takes no argument and returns the value
# to minimize.
loss = lambda: 3 * var1 + 2 * var2
# In eager mode, simply call minimize to update the list of variables.
opt.minimize(loss, var_list=[var1, var2])
# update learning rate
opt.learning_rate = 0.05
opt.minimize(loss, var_list=[var1, var2])
```
### Callable learning rate
Optimizer accepts a callable learning rate in two ways. The first way is
through built-in or customized
`tf.keras.optimizers.schedules.LearningRateSchedule`. The schedule will be
called on each iteration with `schedule(iteration)`, a `tf.Variable`
owned by the optimizer.
Example:
>>> var = tf.Variable(np.random.random(size=(1,)))
>>> learning_rate = tf.keras.optimizers.schedules.ExponentialDecay(
... initial_learning_rate=.01, decay_steps=20, decay_rate=.1)
>>> opt = tf.keras.optimizers.SGD(learning_rate=learning_rate)
>>> loss = lambda: 3 * var
>>> opt.minimize(loss, var_list=[var])
<tf.Variable...
The second way is through a callable function that
does not accept any arguments.
Example:
>>> var = tf.Variable(np.random.random(size=(1,)))
>>> def lr_callable():
... return .1
>>> opt = tf.keras.optimizers.SGD(learning_rate=lr_callable)
>>> loss = lambda: 3 * var
>>> opt.minimize(loss, var_list=[var])
<tf.Variable...
### Creating a custom optimizer
If you intend to create your own optimization algorithm, simply inherit from
this class and override the following methods:
- `_resource_apply_dense` (update variable given gradient tensor is dense)
- `_resource_apply_sparse` (update variable given gradient tensor is sparse)
- `_create_slots`
(if your optimizer algorithm requires additional variables)
- `get_config`
(serialization of the optimizer, include all hyper parameters)"
7131,minimize,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,348,method,"Minimize `loss` by updating `var_list`.
This method simply computes gradient using `tf.GradientTape` and calls
`apply_gradients()`. If you want to process the gradient before applying
then call `tf.GradientTape` and `apply_gradients()` explicitly instead
of using this function.
Args:
loss: A callable taking no arguments which returns the value to minimize.
var_list: list or tuple of `Variable` objects to update to minimize
`loss`, or a callable returning the list or tuple of `Variable` objects.
Use callable when the variable list would otherwise be incomplete before
`minimize` since the variables are created at the first time `loss` is
called.
grad_loss: Optional. A `Tensor` holding the gradient computed for `loss`.
name: Optional name for the returned operation.
Returns:
An `Operation` that updates the variables in `var_list`. The `iterations`
will be automatically increased by 1.
Raises:
ValueError: If some of the variables are not `Variable` objects."
7132,get_gradients,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,445,method,"Returns gradients of `loss` with respect to `params`.
Arguments:
loss: Loss tensor.
params: List of variables.
Returns:
List of gradient tensors.
Raises:
ValueError: In case any gradient cannot be computed (e.g. if gradient
function not implemented)."
7133,apply_gradients,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,473,method,"Apply gradients to variables.
This is the second part of `minimize()`. It returns an `Operation` that
applies gradients.
The method sums gradients from all replicas in the presence of
`tf.distribute.Strategy` by default. You can aggregate gradients yourself by
passing `experimental_aggregate_gradients=False`.
Example:
```python
grads = tape.gradient(loss, vars)
grads = tf.distribute.get_replica_context().all_reduce('sum', grads)
# Processing aggregated gradients.
optimizer.apply_gradients(zip(grads, vars),
experimental_aggregate_gradients=False)
```
Args:
grads_and_vars: List of (gradient, variable) pairs.
name: Optional name for the returned operation. Default to the name passed
to the `Optimizer` constructor.
experimental_aggregate_gradients: Whether to sum gradients from different
replicas in the presense of `tf.distribute.Strategy`. If False, it's
user responsibility to aggregate the gradients. Default to True.
Returns:
An `Operation` that applies the specified gradients. The `iterations`
will be automatically increased by 1.
Raises:
TypeError: If `grads_and_vars` is malformed.
ValueError: If none of the variables have gradients."
7134,get_updates,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,647,method,
7135,get_slot_names,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,731,method,A list of names for this optimizer's slots.
7136,add_slot,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,735,method,Add a new slot variable for `var`.
7137,get_slot,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,773,method,
7138,iterations,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,830,method,Variable. The number of training steps this Optimizer has run.
7139,iterations,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,844,method,
7140,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,864,method,"Returns the config of the optimizer.
An optimizer config is a Python dictionary (serializable)
containing the configuration of an optimizer.
The same optimizer can be reinstantiated later
(without any saved state) from this configuration.
Returns:
Python dictionary."
7141,from_config,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,883,method,"Creates an optimizer from its config.
This method is the reverse of `get_config`,
capable of instantiating the same optimizer from the config
dictionary.
Arguments:
config: A Python dictionary, typically the output of get_config.
custom_objects: A Python dictionary mapping names to additional Python
objects used to create this optimizer, such as a function used for a
hyperparameter.
Returns:
An optimizer instance."
7142,variables,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,918,method,Returns variables of this Optimizer based on the order created.
7143,weights,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,923,method,Returns variables of this Optimizer based on the order created.
7144,get_weights,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,927,method,"Returns the current weights of the optimizer.
The weights of an optimizer are its state (ie, variables).
This function returns the weight values associated with this
optimizer as a list of Numpy arrays. The first value is always the
iterations count of the optimizer, followed by the optimizer's state
variables in the order they were created. The returned list can in turn
be used to load state into similarly parameterized optimizers.
For example, the RMSprop optimizer for this simple model returns a list of
three values-- the iteration count, followed by the root-mean-square value
of the kernel and bias of the single Dense layer:
>>> opt = tf.keras.optimizers.RMSprop()
>>> m = tf.keras.models.Sequential([tf.keras.layers.Dense(10)])
>>> m.compile(opt, loss='mse')
>>> data = np.arange(100).reshape(5, 20)
>>> labels = np.zeros(5)
>>> print('Training'); results = m.fit(data, labels)
Training ...
>>> len(opt.get_weights())
3
Returns:
Weights values as a list of numpy arrays."
7145,set_weights,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,958,method,"Set the weights of the optimizer.
The weights of an optimizer are its state (ie, variables).
This function takes the weight values associated with this
optimizer as a list of Numpy arrays. The first value is always the
iterations count of the optimizer, followed by the optimizer's state
variables in the order they are created. The passed values are used to set
the new state of the optimizer.
For example, the RMSprop optimizer for this simple model takes a list of
three values-- the iteration count, followed by the root-mean-square value
of the kernel and bias of the single Dense layer:
>>> opt = tf.keras.optimizers.RMSprop()
>>> m = tf.keras.models.Sequential([tf.keras.layers.Dense(10)])
>>> m.compile(opt, loss='mse')
>>> data = np.arange(100).reshape(5, 20)
>>> labels = np.zeros(5)
>>> print('Training'); results = m.fit(data, labels)
Training ...
>>> new_weights = [np.array(10), np.ones([20, 10]), np.zeros([10])]
>>> opt.set_weights(new_weights)
>>> opt.iterations
<tf.Variable 'RMSprop/iter:0' shape=() dtype=int64, numpy=10>
Arguments:
weights: weight values as a list of numpy arrays."
7146,add_weight,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,1006,method,
7147,all_reduce_fn,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,563,method,
7148,apply_grad_to_update_var,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,591,method,Apply gradient to variable.
7149,RestoredOptimizer,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,1315,class,"A non-functional Optimizer implementation for checkpoint compatibility.
Holds slot variables and hyperparameters when an optimizer is restored from a
SavedModel. These variables may be referenced in functions along with ops
created by the original optimizer, but currently we do not support using the
optimizer object iself (e.g. through `apply_gradients`)."
7150,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2.py,1330,method,
7151,get_inputs,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2_test.py,928,function,
7152,strip_name,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2_test.py,935,function,
7153,topological_sort,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2_test.py,941,function,
7154,identify_redundant_ops,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2_test.py,981,function,"Implements basic common subexpression elimination.
This is not intended to replicate the graph semantics of TensorFlow Graphs
(for instance it does not handle stateful op ordering), nor is it intended to
replace the common subexpression elimination Grappler pass. Rather, it
provides a high level sanity check that clearly redundant ops are not being
created.
Args:
graph: The graph to be analyzed.
Returns:
A count of the duplicate ops and a description of the structure of each."
7155,make_model,tensorflow/tensorflow/python/keras/optimizer_v2/optimizer_v2_test.py,1064,function,"Constructs a simple ensemble of weak learners model.
--------- --------- --------- ---------
| Input | | Input | ... | Input | | Input |
--------- --------- --------- ---------
| | | |
V V V V
--------- --------- --------- ---------
| Embed | | Embed | ... | Embed | | Embed |
--------- --------- --------- ---------
| | | |
V V V V
--------- --------- --------- ---------
| Dense | | Dense | ... | Dense | | Dense |
--------- --------- --------- ---------
\ | | /
\ | | /
---------------------------------------------
|
---------
| Dense |
---------
This topology is chosen because it exercises both dense and sparse update
paths.
Returns:
A model for testing optimizer coefficient reuse."
7156,RMSprop,tensorflow/tensorflow/python/keras/optimizer_v2/rmsprop.py,35,class,"Optimizer that implements the RMSprop algorithm.
The gist of RMSprop is to:
- Maintain a moving (discounted) average of the square of gradients
- Divide the gradient by the root of this average
This implementation of RMSprop uses plain momentum, not Nesterov momentum.
The centered version additionally maintains a moving average of the
gradients, and uses that average to estimate the variance.
Args:
learning_rate: A `Tensor`, floating point value, or a schedule that is a
`tf.keras.optimizers.schedules.LearningRateSchedule`, or a callable
that takes no arguments and returns the actual value to use. The
learning rate. Defeaults to 0.001.
rho: Discounting factor for the history/coming gradient. Defaults to 0.9.
momentum: A scalar or a scalar `Tensor`. Defaults to 0.0.
epsilon: A small constant for numerical stability. This epsilon is
""epsilon hat"" in the Kingma and Ba paper (in the formula just before
Section 2.1), not the epsilon in Algorithm 1 of the paper. Defaults to
1e-7.
centered: Boolean. If `True`, gradients are normalized by the estimated
variance of the gradient; if False, by the uncentered second moment.
Setting this to `True` may help with training, but is slightly more
expensive in terms of computation and memory. Defaults to `False`.
name: Optional name prefix for the operations created when applying
gradients. Defaults to `""RMSprop""`.
**kwargs: Keyword arguments. Allowed to be one of
`""clipnorm""` or `""clipvalue""`.
`""clipnorm""` (float) clips gradients by norm; `""clipvalue""` (float) clips
gradients by value.
Note that in the dense implementation of this algorithm, variables and their
corresponding accumulators (momentum, gradient moving average, square
gradient moving average) will be updated even if the gradient is zero
(i.e. accumulators will decay, momentum will be applied). The sparse
implementation (used when the gradient is an `IndexedSlices` object,
typically because of `tf.gather` or an embedding lookup in the forward pass)
will not update variable slices or their accumulators unless those slices
were used in the forward pass (nor is there an ""eventual"" correction to
account for these omitted updates). This leads to more efficient updates for
large embedding lookup tables (where most of the slices are not accessed in
a particular graph execution), but differs from the published algorithm.
Usage:
>>> opt = tf.keras.optimizers.RMSprop(learning_rate=0.1)
>>> var1 = tf.Variable(10.0)
>>> loss = lambda: (var1 ** 2) / 2.0 # d(loss) / d(var1) = var1
>>> step_count = opt.minimize(loss, [var1]).numpy()
>>> var1.numpy()
9.683772
Reference:
- [Hinton, 2012](
http://www.cs.toronto.edu/~tijmen/csc321/slides/lecture_slides_lec6.pdf)"
7157,set_weights,tensorflow/tensorflow/python/keras/optimizer_v2/rmsprop.py,279,method,
7158,get_config,tensorflow/tensorflow/python/keras/optimizer_v2/rmsprop.py,288,method,
7159,LinearModel,tensorflow/tensorflow/python/keras/premade/linear.py,32,class,"Linear Model for regression and classification problems.
This model approximates the following function:
$$y = \beta + \sum_{i=1}^{N} w_{i} * x_{i}$$
where $$\beta$$ is the bias and $$w_{i}$$ is the weight for each feature.
Example:
```python
model = LinearModel()
model.compile(optimizer='sgd', loss='mse')
model.fit(x, y, epochs=epochs)
```
This model accepts sparse float inputs as well:
Example:
```python
model = LinearModel()
opt = tf.keras.optimizers.Adam()
loss_fn = tf.keras.losses.MeanSquaredError()
with tf.GradientTape() as tape:
output = model(sparse_input)
loss = tf.reduce_mean(loss_fn(target, output))
grads = tape.gradient(loss, model.weights)
opt.apply_gradients(zip(grads, model.weights))
```"
7160,build,tensorflow/tensorflow/python/keras/premade/linear.py,98,method,
7161,call,tensorflow/tensorflow/python/keras/premade/linear.py,129,method,
7162,get_config,tensorflow/tensorflow/python/keras/premade/linear.py,149,method,
7163,from_config,tensorflow/tensorflow/python/keras/premade/linear.py,163,method,
7164,WideDeepModel,tensorflow/tensorflow/python/keras/premade/wide_deep.py,34,class,"Wide & Deep Model for regression and classification problems.
This model jointly train a linear and a dnn model.
Example:
```python
linear_model = LinearModel()
dnn_model = keras.Sequential([keras.layers.Dense(units=64),
keras.layers.Dense(units=1)])
combined_model = WideDeepModel(linear_model, dnn_model)
combined_model.compile(optimizer=['sgd', 'adam'], 'mse', ['mse'])
# define dnn_inputs and linear_inputs as separate numpy arrays or
# a single numpy array if dnn_inputs is same as linear_inputs.
combined_model.fit([linear_inputs, dnn_inputs], y, epochs)
# or define a single `tf.data.Dataset` that contains a single tensor or
# separate tensors for dnn_inputs and linear_inputs.
dataset = tf.data.Dataset.from_tensors(([linear_inputs, dnn_inputs], y))
combined_model.fit(dataset, epochs)
```
Both linear and dnn model can be pre-compiled and trained separately
before jointly training:
Example:
```python
linear_model = LinearModel()
linear_model.compile('adagrad', 'mse')
linear_model.fit(linear_inputs, y, epochs)
dnn_model = keras.Sequential([keras.layers.Dense(units=1)])
dnn_model.compile('rmsprop', 'mse')
dnn_model.fit(dnn_inputs, y, epochs)
combined_model = WideDeepModel(linear_model, dnn_model)
combined_model.compile(optimizer=['sgd', 'adam'], 'mse', ['mse'])
combined_model.fit([linear_inputs, dnn_inputs], y, epochs)
```"
7165,call,tensorflow/tensorflow/python/keras/premade/wide_deep.py,93,method,
7166,train_step,tensorflow/tensorflow/python/keras/premade/wide_deep.py,112,method,
7167,get_config,tensorflow/tensorflow/python/keras/premade/wide_deep.py,196,method,
7168,from_config,tensorflow/tensorflow/python/keras/premade/wide_deep.py,208,method,
7169,index_directory,tensorflow/tensorflow/python/keras/preprocessing/dataset_utils.py,31,function,"Make list of all files in the subdirs of `directory`, with their labels.
Args:
directory: The target directory (string).
labels: Either ""inferred""
(labels are generated from the directory structure),
or a list/tuple of integer labels of the same size as the number of
valid files found in the directory. Labels should be sorted according
to the alphanumeric order of the image file paths
(obtained via `os.walk(directory)` in Python).
formats: Allowlist of file extensions to index (e.g. "".jpg"", "".txt"").
class_names: Only valid if ""labels"" is ""inferred"". This is the explict
list of class names (must match names of subdirectories). Used
to control the order of the classes
(otherwise alphanumerical order is used).
shuffle: Whether to shuffle the data. Default: True.
If set to False, sorts the data in alphanumeric order.
seed: Optional random seed for shuffling.
follow_links: Whether to visits subdirectories pointed to by symlinks.
Returns:
tuple (file_paths, labels, class_names).
file_paths: list of file paths (strings).
labels: list of matching integer labels (same length as file_paths)
class_names: names of the classes corresponding to these labels, in order."
7170,iter_valid_files,tensorflow/tensorflow/python/keras/preprocessing/dataset_utils.py,123,function,
7171,index_subdirectory,tensorflow/tensorflow/python/keras/preprocessing/dataset_utils.py,131,function,"Recursively walks directory and list image paths and their class index.
Arguments:
directory: string, target directory.
class_indices: dict mapping class names to their index.
follow_links: boolean, whether to recursively follow subdirectories
(if False, we only list top-level images in `directory`).
formats: Allowlist of file extensions to index (e.g. "".jpg"", "".txt"").
Returns:
tuple `(filenames, labels)`. `filenames` is a list of relative file
paths, and `labels` is a list of integer labels corresponding to these
files."
7172,get_training_or_validation_split,tensorflow/tensorflow/python/keras/preprocessing/dataset_utils.py,159,function,"Potentially restict samples & labels to a training or validation split.
Args:
samples: List of elements.
labels: List of corresponding labels.
validation_split: Float, fraction of data to reserve for validation.
subset: Subset of the data to return.
Either ""training"", ""validation"", or None. If None, we return all of the
data.
Returns:
tuple (samples, labels), potentially restricted to the specified subset."
7173,labels_to_dataset,tensorflow/tensorflow/python/keras/preprocessing/dataset_utils.py,191,function,
7174,check_validation_split_arg,tensorflow/tensorflow/python/keras/preprocessing/dataset_utils.py,201,function,Raise errors in case of invalid argument values.
7175,smart_resize,tensorflow/tensorflow/python/keras/preprocessing/image.py,55,function,"Resize images to a target size without aspect ratio distortion.
TensorFlow image datasets typically yield images that have each a different
size. However, these images need to be batched before they can be
processed by Keras layers. To be batched, images need to share the same height
and width.
You could simply do:
````python
size = (200, 200)
ds = ds.map(lambda img: tf.image.resize(img, size))
```
However, if you do this, you distort the aspect ratio of your images, since
in general they do not all have the same aspect ratio as `size`. This is
fine in many cases, but not always (e.g. for GANs this can be a problem).
Note that passing the argument `preserve_aspect_ratio=True` to `resize`
will preserve the aspect ratio, but at the cost of no longer respecting the
provided target size. Because `tf.image.resize` doesn't crop images,
your output images will still have different sizes.
This calls for:
```python
size = (200, 200)
ds = ds.map(lambda img: smart_resize(img, size))
```
Your output images will actually be `(200, 200)`, and will not be distorted.
Instead, the parts of the image that do not fit within the target size
get cropped out.
The resizing process is:
1. Take the largest centered crop of the image that has the same aspect ratio
as the target size. For instance, if `size=(200, 200)` and the input image has
size `(340, 500)`, we take a crop of `(340, 340)` centered along the width.
2. Resize the cropped image to the target size. In the example above,
we resize the `(340, 340)` crop to `(200, 200)`.
Arguments:
x: Input image (as a tensor or NumPy array). Must be in format
`(height, width, channels)`.
size: Tuple of `(height, width)` integer. Target size.
interpolation: String, interpolation to use for resizing.
Defaults to `'bilinear'`. Supports `bilinear`, `nearest`, `bicubic`,
`area`, `lanczos3`, `lanczos5`, `gaussian`, `mitchellcubic`.
Returns:
Array with shape `(size[0], size[1], channels)`. If the input image was a
NumPy array, the output is a NumPy array, and if it was a TF tensor,
the output is a TF tensor."
7176,array_to_img,tensorflow/tensorflow/python/keras/preprocessing/image.py,152,function,"Converts a 3D Numpy array to a PIL Image instance.
Usage:
```python
from PIL import Image
img = np.random.random(size=(100, 100, 3))
pil_img = tf.keras.preprocessing.image.array_to_img(img)
```
Arguments:
x: Input Numpy array.
data_format: Image data format, can be either ""channels_first"" or
""channels_last"". Defaults to `None`, in which case the global setting
`tf.keras.backend.image_data_format()` is used (unless you changed it,
it defaults to ""channels_last"").
scale: Whether to rescale image values to be within `[0, 255]`. Defaults
to `True`.
dtype: Dtype to use. Default to `None`, in which case the global setting
`tf.keras.backend.floatx()` is used (unless you changed it, it defaults
to ""float32"")
Returns:
A PIL Image instance.
Raises:
ImportError: if PIL is not available.
ValueError: if invalid `x` or `data_format` is passed."
7177,img_to_array,tensorflow/tensorflow/python/keras/preprocessing/image.py,195,function,"Converts a PIL Image instance to a Numpy array.
Usage:
```python
from PIL import Image
img_data = np.random.random(size=(100, 100, 3))
img = tf.keras.preprocessing.image.array_to_img(img_data)
array = tf.keras.preprocessing.image.img_to_array(img)
```
Arguments:
img: Input PIL Image instance.
data_format: Image data format, can be either ""channels_first"" or
""channels_last"". Defaults to `None`, in which case the global setting
`tf.keras.backend.image_data_format()` is used (unless you changed it,
it defaults to ""channels_last"").
dtype: Dtype to use. Default to `None`, in which case the global setting
`tf.keras.backend.floatx()` is used (unless you changed it, it defaults
to ""float32"")
Returns:
A 3D Numpy array.
Raises:
ValueError: if invalid `img` or `data_format` is passed."
7178,save_img,tensorflow/tensorflow/python/keras/preprocessing/image.py,236,function,"Saves an image stored as a Numpy array to a path or file object.
Arguments:
path: Path or file object.
x: Numpy array.
data_format: Image data format,
either ""channels_first"" or ""channels_last"".
file_format: Optional file format override. If omitted, the
format to use is determined from the filename extension.
If a file object was used instead of a filename, this
parameter should always be used.
scale: Whether to rescale image values to be within `[0, 255]`.
**kwargs: Additional keyword arguments passed to `PIL.Image.save()`."
7179,load_img,tensorflow/tensorflow/python/keras/preprocessing/image.py,265,function,"Loads an image into PIL format.
Usage:
```
image = tf.keras.preprocessing.image.load_img(image_path)
input_arr = keras.preprocessing.image.img_to_array(image)
input_arr = np.array([input_arr]) # Convert single image to a batch.
predictions = model.predict(input_arr)
```
Arguments:
path: Path to image file.
grayscale: DEPRECATED use `color_mode=""grayscale""`.
color_mode: One of ""grayscale"", ""rgb"", ""rgba"". Default: ""rgb"".
The desired image format.
target_size: Either `None` (default to original size)
or tuple of ints `(img_height, img_width)`.
interpolation: Interpolation method used to resample the image if the
target size is different from that of the loaded image.
Supported methods are ""nearest"", ""bilinear"", and ""bicubic"".
If PIL version 1.1.3 or newer is installed, ""lanczos"" is also
supported. If PIL version 3.4.0 or newer is installed, ""box"" and
""hamming"" are also supported. By default, ""nearest"" is used.
Returns:
A PIL Image instance.
Raises:
ImportError: if PIL is not available.
ValueError: if interpolation method is not supported."
7180,Iterator,tensorflow/tensorflow/python/keras/preprocessing/image.py,304,class,
7181,DirectoryIterator,tensorflow/tensorflow/python/keras/preprocessing/image.py,309,class,"Iterator capable of reading images from a directory on disk.
Arguments:
directory: Path to the directory to read images from.
Each subdirectory in this directory will be
considered to contain images from one class,
or alternatively you could specify class subdirectories
via the `classes` argument.
image_data_generator: Instance of `ImageDataGenerator`
to use for random transformations and normalization.
target_size: tuple of integers, dimensions to resize input images to.
color_mode: One of `""rgb""`, `""rgba""`, `""grayscale""`.
Color mode to read images.
classes: Optional list of strings, names of subdirectories
containing images from each class (e.g. `[""dogs"", ""cats""]`).
It will be computed automatically if not set.
class_mode: Mode for yielding the targets:
`""binary""`: binary targets (if there are only two classes),
`""categorical""`: categorical targets,
`""sparse""`: integer targets,
`""input""`: targets are images identical to input images (mainly
used to work with autoencoders),
`None`: no targets get yielded (only input images are yielded).
batch_size: Integer, size of a batch.
shuffle: Boolean, whether to shuffle the data between epochs.
seed: Random seed for data shuffling.
data_format: String, one of `channels_first`, `channels_last`.
save_to_dir: Optional directory where to save the pictures
being yielded, in a viewable format. This is useful
for visualizing the random transformations being
applied, for debugging purposes.
save_prefix: String prefix to use for saving sample
images (if `save_to_dir` is set).
save_format: Format to use for saving sample images
(if `save_to_dir` is set).
subset: Subset of data (`""training""` or `""validation""`) if
validation_split is set in ImageDataGenerator.
interpolation: Interpolation method used to resample the image if the
target size is different from that of the loaded image.
Supported methods are ""nearest"", ""bilinear"", and ""bicubic"".
If PIL version 1.1.3 or newer is installed, ""lanczos"" is also
supported. If PIL version 3.4.0 or newer is installed, ""box"" and
""hamming"" are also supported. By default, ""nearest"" is used.
dtype: Dtype to use for generated arrays."
7182,NumpyArrayIterator,tensorflow/tensorflow/python/keras/preprocessing/image.py,400,class,"Iterator yielding data from a Numpy array.
Arguments:
x: Numpy array of input data or tuple.
If tuple, the second elements is either
another numpy array or a list of numpy arrays,
each of which gets passed
through as an output without any modifications.
y: Numpy array of targets data.
image_data_generator: Instance of `ImageDataGenerator`
to use for random transformations and normalization.
batch_size: Integer, size of a batch.
shuffle: Boolean, whether to shuffle the data between epochs.
sample_weight: Numpy array of sample weights.
seed: Random seed for data shuffling.
data_format: String, one of `channels_first`, `channels_last`.
save_to_dir: Optional directory where to save the pictures
being yielded, in a viewable format. This is useful
for visualizing the random transformations being
applied, for debugging purposes.
save_prefix: String prefix to use for saving sample
images (if `save_to_dir` is set).
save_format: Format to use for saving sample images
(if `save_to_dir` is set).
subset: Subset of data (`""training""` or `""validation""`) if
validation_split is set in ImageDataGenerator.
dtype: Dtype to use for the generated arrays."
7183,DataFrameIterator,tensorflow/tensorflow/python/keras/preprocessing/image.py,463,class,"Iterator capable of reading images from a directory on disk as a dataframe.
Arguments:
dataframe: Pandas dataframe containing the filepaths relative to
`directory` (or absolute paths if `directory` is None) of the images in
a string column. It should include other column/s
depending on the `class_mode`: - if `class_mode` is `""categorical""`
(default value) it must include the `y_col` column with the class/es
of each image. Values in column can be string/list/tuple if a single
class or list/tuple if multiple classes. - if `class_mode` is
`""binary""` or `""sparse""` it must include the given `y_col` column
with class values as strings. - if `class_mode` is `""raw""` or
`""multi_output""` it should contain the columns specified in `y_col`.
- if `class_mode` is `""input""` or `None` no extra column is needed.
directory: string, path to the directory to read images from. If `None`,
data in `x_col` column should be absolute paths.
image_data_generator: Instance of `ImageDataGenerator` to use for random
transformations and normalization. If None, no transformations and
normalizations are made.
x_col: string, column in `dataframe` that contains the filenames (or
absolute paths if `directory` is `None`).
y_col: string or list, column/s in `dataframe` that has the target data.
weight_col: string, column in `dataframe` that contains the sample
weights. Default: `None`.
target_size: tuple of integers, dimensions to resize input images to.
color_mode: One of `""rgb""`, `""rgba""`, `""grayscale""`. Color mode to read
images.
classes: Optional list of strings, classes to use (e.g. `[""dogs"",
""cats""]`). If None, all classes in `y_col` will be used.
class_mode: one of ""binary"", ""categorical"", ""input"", ""multi_output"",
""raw"", ""sparse"" or None. Default: ""categorical"".
Mode for yielding the targets:
- `""binary""`: 1D numpy array of binary labels,
- `""categorical""`: 2D numpy array of one-hot encoded labels. Supports
multi-label output.
- `""input""`: images identical to input images (mainly used to work
with autoencoders),
- `""multi_output""`: list with the values of the different columns,
- `""raw""`: numpy array of values in `y_col` column(s),
- `""sparse""`: 1D numpy array of integer labels, - `None`, no targets
are returned (the generator will only yield batches of image data,
which is useful to use in `model.predict()`).
batch_size: Integer, size of a batch.
shuffle: Boolean, whether to shuffle the data between epochs.
seed: Random seed for data shuffling.
data_format: String, one of `channels_first`, `channels_last`.
save_to_dir: Optional directory where to save the pictures being yielded,
in a viewable format. This is useful for visualizing the random
transformations being applied, for debugging purposes.
save_prefix: String prefix to use for saving sample images (if
`save_to_dir` is set).
save_format: Format to use for saving sample images (if `save_to_dir` is
set).
subset: Subset of data (`""training""` or `""validation""`) if
validation_split is set in ImageDataGenerator.
interpolation: Interpolation method used to resample the image if the
target size is different from that of the loaded image. Supported
methods are ""nearest"", ""bilinear"", and ""bicubic"". If PIL version 1.1.3
or newer is installed, ""lanczos"" is also supported. If PIL version 3.4.0
or newer is installed, ""box"" and ""hamming"" are also supported. By
default, ""nearest"" is used.
dtype: Dtype to use for the generated arrays.
validate_filenames: Boolean, whether to validate image filenames in
`x_col`. If `True`, invalid images will be ignored. Disabling this
option
can lead to speed-up in the instantiation of this class. Default: `True`."
7184,ImageDataGenerator,tensorflow/tensorflow/python/keras/preprocessing/image.py,581,class,"Generate batches of tensor image data with real-time data augmentation.
The data will be looped over (in batches).
Arguments:
featurewise_center: Boolean.
Set input mean to 0 over the dataset, feature-wise.
samplewise_center: Boolean. Set each sample mean to 0.
featurewise_std_normalization: Boolean.
Divide inputs by std of the dataset, feature-wise.
samplewise_std_normalization: Boolean. Divide each input by its std.
zca_epsilon: epsilon for ZCA whitening. Default is 1e-6.
zca_whitening: Boolean. Apply ZCA whitening.
rotation_range: Int. Degree range for random rotations.
width_shift_range: Float, 1-D array-like or int
- float: fraction of total width, if < 1, or pixels if >= 1.
- 1-D array-like: random elements from the array.
- int: integer number of pixels from interval
`(-width_shift_range, +width_shift_range)`
- With `width_shift_range=2` possible values
are integers `[-1, 0, +1]`,
same as with `width_shift_range=[-1, 0, +1]`,
while with `width_shift_range=1.0` possible values are floats
in the interval [-1.0, +1.0).
height_shift_range: Float, 1-D array-like or int
- float: fraction of total height, if < 1, or pixels if >= 1.
- 1-D array-like: random elements from the array.
- int: integer number of pixels from interval
`(-height_shift_range, +height_shift_range)`
- With `height_shift_range=2` possible values
are integers `[-1, 0, +1]`,
same as with `height_shift_range=[-1, 0, +1]`,
while with `height_shift_range=1.0` possible values are floats
in the interval [-1.0, +1.0).
brightness_range: Tuple or list of two floats. Range for picking
a brightness shift value from.
shear_range: Float. Shear Intensity
(Shear angle in counter-clockwise direction in degrees)
zoom_range: Float or [lower, upper]. Range for random zoom.
If a float, `[lower, upper] = [1-zoom_range, 1+zoom_range]`.
channel_shift_range: Float. Range for random channel shifts.
fill_mode: One of {""constant"", ""nearest"", ""reflect"" or ""wrap""}.
Default is 'nearest'.
Points outside the boundaries of the input are filled
according to the given mode:
- 'constant': kkkkkkkk|abcd|kkkkkkkk (cval=k)
- 'nearest': aaaaaaaa|abcd|dddddddd
- 'reflect': abcddcba|abcd|dcbaabcd
- 'wrap': abcdabcd|abcd|abcdabcd
cval: Float or Int.
Value used for points outside the boundaries
when `fill_mode = ""constant""`.
horizontal_flip: Boolean. Randomly flip inputs horizontally.
vertical_flip: Boolean. Randomly flip inputs vertically.
rescale: rescaling factor. Defaults to None.
If None or 0, no rescaling is applied,
otherwise we multiply the data by the value provided
(after applying all other transformations).
preprocessing_function: function that will be applied on each input.
The function will run after the image is resized and augmented.
The function should take one argument:
one image (Numpy tensor with rank 3),
and should output a Numpy tensor with the same shape.
data_format: Image data format,
either ""channels_first"" or ""channels_last"".
""channels_last"" mode means that the images should have shape
`(samples, height, width, channels)`,
""channels_first"" mode means that the images should have shape
`(samples, channels, height, width)`.
It defaults to the `image_data_format` value found in your
Keras config file at `~/.keras/keras.json`.
If you never set it, then it will be ""channels_last"".
validation_split: Float. Fraction of images reserved for validation
(strictly between 0 and 1).
dtype: Dtype to use for the generated arrays.
Examples:
Example of using `.flow(x, y)`:
```python
(x_train, y_train), (x_test, y_test) = cifar10.load_data()
y_train = np_utils.to_categorical(y_train, num_classes)
y_test = np_utils.to_categorical(y_test, num_classes)
datagen = ImageDataGenerator(
featurewise_center=True,
featurewise_std_normalization=True,
rotation_range=20,
width_shift_range=0.2,
height_shift_range=0.2,
horizontal_flip=True)
# compute quantities required for featurewise normalization
# (std, mean, and principal components if ZCA whitening is applied)
datagen.fit(x_train)
# fits the model on batches with real-time data augmentation:
model.fit(datagen.flow(x_train, y_train, batch_size=32),
steps_per_epoch=len(x_train) / 32, epochs=epochs)
# here's a more ""manual"" example
for e in range(epochs):
print('Epoch', e)
batches = 0
for x_batch, y_batch in datagen.flow(x_train, y_train, batch_size=32):
model.fit(x_batch, y_batch)
batches += 1
if batches >= len(x_train) / 32:
# we need to break the loop by hand because
# the generator loops indefinitely
break
```
Example of using `.flow_from_directory(directory)`:
```python
train_datagen = ImageDataGenerator(
rescale=1./255,
shear_range=0.2,
zoom_range=0.2,
horizontal_flip=True)
test_datagen = ImageDataGenerator(rescale=1./255)
train_generator = train_datagen.flow_from_directory(
'data/train',
target_size=(150, 150),
batch_size=32,
class_mode='binary')
validation_generator = test_datagen.flow_from_directory(
'data/validation',
target_size=(150, 150),
batch_size=32,
class_mode='binary')
model.fit(
train_generator,
steps_per_epoch=2000,
epochs=50,
validation_data=validation_generator,
validation_steps=800)
```
Example of transforming images and masks together.
```python
# we create two instances with the same arguments
data_gen_args = dict(featurewise_center=True,
featurewise_std_normalization=True,
rotation_range=90,
width_shift_range=0.1,
height_shift_range=0.1,
zoom_range=0.2)
image_datagen = ImageDataGenerator(**data_gen_args)
mask_datagen = ImageDataGenerator(**data_gen_args)
# Provide the same seed and keyword arguments to the fit and flow methods
seed = 1
image_datagen.fit(images, augment=True, seed=seed)
mask_datagen.fit(masks, augment=True, seed=seed)
image_generator = image_datagen.flow_from_directory(
'data/images',
class_mode=None,
seed=seed)
mask_generator = mask_datagen.flow_from_directory(
'data/masks',
class_mode=None,
seed=seed)
# combine generators into one which yields image and masks
train_generator = zip(image_generator, mask_generator)
model.fit(
train_generator,
steps_per_epoch=2000,
epochs=50)
```"
7185,flow,tensorflow/tensorflow/python/keras/preprocessing/image.py,807,method,"Takes data & label arrays, generates batches of augmented data.
Arguments:
x: Input data. Numpy array of rank 4 or a tuple. If tuple, the first
element should contain the images and the second element another numpy
array or a list of numpy arrays that gets passed to the output without
any modifications. Can be used to feed the model miscellaneous data
along with the images. In case of grayscale data, the channels axis of
the image array should have value 1, in case of RGB data, it should
have value 3, and in case of RGBA data, it should have value 4.
y: Labels.
batch_size: Int (default: 32).
shuffle: Boolean (default: True).
sample_weight: Sample weights.
seed: Int (default: None).
save_to_dir: None or str (default: None). This allows you to optionally
specify a directory to which to save the augmented pictures being
generated (useful for visualizing what you are doing).
save_prefix: Str (default: `''`). Prefix to use for filenames of saved
pictures (only relevant if `save_to_dir` is set).
save_format: one of ""png"", ""jpeg""
(only relevant if `save_to_dir` is set). Default: ""png"".
subset: Subset of data (`""training""` or `""validation""`) if
`validation_split` is set in `ImageDataGenerator`.
Returns:
An `Iterator` yielding tuples of `(x, y)`
where `x` is a numpy array of image data
(in the case of a single image input) or a list
of numpy arrays (in the case with
additional inputs) and `y` is a numpy array
of corresponding labels. If 'sample_weight' is not None,
the yielded tuples are of the form `(x, y, sample_weight)`.
If `y` is None, only the numpy array `x` is returned."
7186,flow_from_directory,tensorflow/tensorflow/python/keras/preprocessing/image.py,867,method,"Takes the path to a directory & generates batches of augmented data.
Arguments:
directory: string, path to the target directory. It should contain one
subdirectory per class. Any PNG, JPG, BMP, PPM or TIF images inside
each of the subdirectories directory tree will be included in the
generator. See [this script](
https://gist.github.com/fchollet/0830affa1f7f19fd47b06d4cf89ed44d)
for more details.
target_size: Tuple of integers `(height, width)`, defaults to `(256,
256)`. The dimensions to which all images found will be resized.
color_mode: One of ""grayscale"", ""rgb"", ""rgba"". Default: ""rgb"". Whether
the images will be converted to have 1, 3, or 4 channels.
classes: Optional list of class subdirectories
(e.g. `['dogs', 'cats']`). Default: None. If not provided, the list
of classes will be automatically inferred from the subdirectory
names/structure under `directory`, where each subdirectory will be
treated as a different class (and the order of the classes, which
will map to the label indices, will be alphanumeric). The
dictionary containing the mapping from class names to class
indices can be obtained via the attribute `class_indices`.
class_mode: One of ""categorical"", ""binary"", ""sparse"",
""input"", or None. Default: ""categorical"".
Determines the type of label arrays that are returned: -
""categorical"" will be 2D one-hot encoded labels, - ""binary"" will
be 1D binary labels, ""sparse"" will be 1D integer labels, - ""input""
will be images identical to input images (mainly used to work with
autoencoders). - If None, no labels are returned (the generator
will only yield batches of image data, which is useful to use with
`model.predict()`). Please note that in case of
class_mode None, the data still needs to reside in a subdirectory
of `directory` for it to work correctly.
batch_size: Size of the batches of data (default: 32).
shuffle: Whether to shuffle the data (default: True) If set to False,
sorts the data in alphanumeric order.
seed: Optional random seed for shuffling and transformations.
save_to_dir: None or str (default: None). This allows you to optionally
specify a directory to which to save the augmented pictures being
generated (useful for visualizing what you are doing).
save_prefix: Str. Prefix to use for filenames of saved pictures (only
relevant if `save_to_dir` is set).
save_format: One of ""png"", ""jpeg""
(only relevant if `save_to_dir` is set). Default: ""png"".
follow_links: Whether to follow symlinks inside
class subdirectories (default: False).
subset: Subset of data (`""training""` or `""validation""`) if
`validation_split` is set in `ImageDataGenerator`.
interpolation: Interpolation method used to resample the image if the
target size is different from that of the loaded image. Supported
methods are `""nearest""`, `""bilinear""`, and `""bicubic""`. If PIL version
1.1.3 or newer is installed, `""lanczos""` is also supported. If PIL
version 3.4.0 or newer is installed, `""box""` and `""hamming""` are also
supported. By default, `""nearest""` is used.
Returns:
A `DirectoryIterator` yielding tuples of `(x, y)`
where `x` is a numpy array containing a batch
of images with shape `(batch_size, *target_size, channels)`
and `y` is a numpy array of corresponding labels."
7187,flow_from_dataframe,tensorflow/tensorflow/python/keras/preprocessing/image.py,960,method,"Takes the dataframe and the path to a directory + generates batches.
The generated batches contain augmented/normalized data.
**A simple tutorial can be found **[here](
http://bit.ly/keras_flow_from_dataframe).
Arguments:
dataframe: Pandas dataframe containing the filepaths relative to
`directory` (or absolute paths if `directory` is None) of the images
in a string column. It should include other column/s
depending on the `class_mode`: - if `class_mode` is `""categorical""`
(default value) it must include the `y_col` column with the
class/es of each image. Values in column can be string/list/tuple
if a single class or list/tuple if multiple classes. - if
`class_mode` is `""binary""` or `""sparse""` it must include the given
`y_col` column with class values as strings. - if `class_mode` is
`""raw""` or `""multi_output""` it should contain the columns
specified in `y_col`. - if `class_mode` is `""input""` or `None` no
extra column is needed.
directory: string, path to the directory to read images from. If `None`,
data in `x_col` column should be absolute paths.
x_col: string, column in `dataframe` that contains the filenames (or
absolute paths if `directory` is `None`).
y_col: string or list, column/s in `dataframe` that has the target data.
weight_col: string, column in `dataframe` that contains the sample
weights. Default: `None`.
target_size: tuple of integers `(height, width)`, default: `(256, 256)`.
The dimensions to which all images found will be resized.
color_mode: one of ""grayscale"", ""rgb"", ""rgba"". Default: ""rgb"". Whether
the images will be converted to have 1 or 3 color channels.
classes: optional list of classes (e.g. `['dogs', 'cats']`). Default is
None. If not provided, the list of classes will be automatically
inferred from the `y_col`, which will map to the label indices, will
be alphanumeric). The dictionary containing the mapping from class
names to class indices can be obtained via the attribute
`class_indices`.
class_mode: one of ""binary"", ""categorical"", ""input"", ""multi_output"",
""raw"", sparse"" or None. Default: ""categorical"".
Mode for yielding the targets:
- `""binary""`: 1D numpy array of binary labels,
- `""categorical""`: 2D numpy array of one-hot encoded labels.
Supports multi-label output.
- `""input""`: images identical to input images (mainly used to work
with autoencoders),
- `""multi_output""`: list with the values of the different columns,
- `""raw""`: numpy array of values in `y_col` column(s),
- `""sparse""`: 1D numpy array of integer labels, - `None`, no targets
are returned (the generator will only yield batches of image data,
which is useful to use in `model.predict()`).
batch_size: size of the batches of data (default: 32).
shuffle: whether to shuffle the data (default: True)
seed: optional random seed for shuffling and transformations.
save_to_dir: None or str (default: None). This allows you to optionally
specify a directory to which to save the augmented pictures being
generated (useful for visualizing what you are doing).
save_prefix: str. Prefix to use for filenames of saved pictures (only
relevant if `save_to_dir` is set).
save_format: one of ""png"", ""jpeg""
(only relevant if `save_to_dir` is set). Default: ""png"".
subset: Subset of data (`""training""` or `""validation""`) if
`validation_split` is set in `ImageDataGenerator`.
interpolation: Interpolation method used to resample the image if the
target size is different from that of the loaded image. Supported
methods are `""nearest""`, `""bilinear""`, and `""bicubic""`. If PIL version
1.1.3 or newer is installed, `""lanczos""` is also supported. If PIL
version 3.4.0 or newer is installed, `""box""` and `""hamming""` are also
supported. By default, `""nearest""` is used.
validate_filenames: Boolean, whether to validate image filenames in
`x_col`. If `True`, invalid images will be ignored. Disabling this
option can lead to speed-up in the execution of this function.
Defaults to `True`.
**kwargs: legacy arguments for raising deprecation warnings.
Returns:
A `DataFrameIterator` yielding tuples of `(x, y)`
where `x` is a numpy array containing a batch
of images with shape `(batch_size, *target_size, channels)`
and `y` is a numpy array of corresponding labels."
7188,image_dataset_from_directory,tensorflow/tensorflow/python/keras/preprocessing/image_dataset.py,35,function,"Generates a `tf.data.Dataset` from image files in a directory.
If your directory structure is:
```
main_directory/
...class_a/
......a_image_1.jpg
......a_image_2.jpg
...class_b/
......b_image_1.jpg
......b_image_2.jpg
```
Then calling `image_dataset_from_directory(main_directory, labels='inferred')`
will return a `tf.data.Dataset` that yields batches of images from
the subdirectories `class_a` and `class_b`, together with labels
0 and 1 (0 corresponding to `class_a` and 1 corresponding to `class_b`).
Supported image formats: jpeg, png, bmp, gif.
Animated gifs are truncated to the first frame.
Arguments:
directory: Directory where the data is located.
If `labels` is ""inferred"", it should contain
subdirectories, each containing images for a class.
Otherwise, the directory structure is ignored.
labels: Either ""inferred""
(labels are generated from the directory structure),
or a list/tuple of integer labels of the same size as the number of
image files found in the directory. Labels should be sorted according
to the alphanumeric order of the image file paths
(obtained via `os.walk(directory)` in Python).
label_mode:
- 'int': means that the labels are encoded as integers
(e.g. for `sparse_categorical_crossentropy` loss).
- 'categorical' means that the labels are
encoded as a categorical vector
(e.g. for `categorical_crossentropy` loss).
- 'binary' means that the labels (there can be only 2)
are encoded as `float32` scalars with values 0 or 1
(e.g. for `binary_crossentropy`).
- None (no labels).
class_names: Only valid if ""labels"" is ""inferred"". This is the explict
list of class names (must match names of subdirectories). Used
to control the order of the classes
(otherwise alphanumerical order is used).
color_mode: One of ""grayscale"", ""rgb"", ""rgba"". Default: ""rgb"".
Whether the images will be converted to
have 1, 3, or 4 channels.
batch_size: Size of the batches of data. Default: 32.
image_size: Size to resize images to after they are read from disk.
Defaults to `(256, 256)`.
Since the pipeline processes batches of images that must all have
the same size, this must be provided.
shuffle: Whether to shuffle the data. Default: True.
If set to False, sorts the data in alphanumeric order.
seed: Optional random seed for shuffling and transformations.
validation_split: Optional float between 0 and 1,
fraction of data to reserve for validation.
subset: One of ""training"" or ""validation"".
Only used if `validation_split` is set.
interpolation: String, the interpolation method used when resizing images.
Defaults to `bilinear`. Supports `bilinear`, `nearest`, `bicubic`,
`area`, `lanczos3`, `lanczos5`, `gaussian`, `mitchellcubic`.
follow_links: Whether to visits subdirectories pointed to by symlinks.
Defaults to False.
Returns:
A `tf.data.Dataset` object.
- If `label_mode` is None, it yields `float32` tensors of shape
`(batch_size, image_size[0], image_size[1], num_channels)`,
encoding images (see below for rules regarding `num_channels`).
- Otherwise, it yields a tuple `(images, labels)`, where `images`
has shape `(batch_size, image_size[0], image_size[1], num_channels)`,
and `labels` follows the format described below.
Rules regarding labels format:
- if `label_mode` is `int`, the labels are an `int32` tensor of shape
`(batch_size,)`.
- if `label_mode` is `binary`, the labels are a `float32` tensor of
1s and 0s of shape `(batch_size, 1)`.
- if `label_mode` is `categorial`, the labels are a `float32` tensor
of shape `(batch_size, num_classes)`, representing a one-hot
encoding of the class index.
Rules regarding number of channels in the yielded images:
- if `color_mode` is `grayscale`,
there's 1 channel in the image tensors.
- if `color_mode` is `rgb`,
there are 3 channel in the image tensors.
- if `color_mode` is `rgba`,
there are 4 channel in the image tensors."
7189,paths_and_labels_to_dataset,tensorflow/tensorflow/python/keras/preprocessing/image_dataset.py,209,function,Constructs a dataset of images and labels.
7190,path_to_image,tensorflow/tensorflow/python/keras/preprocessing/image_dataset.py,227,function,
7191,TimeseriesGenerator,tensorflow/tensorflow/python/keras/preprocessing/sequence.py,34,class,"Utility class for generating batches of temporal data.
This class takes in a sequence of data-points gathered at
equal intervals, along with time series parameters such as
stride, length of history, etc., to produce batches for
training/validation.
# Arguments
data: Indexable generator (such as list or Numpy array)
containing consecutive data points (timesteps).
The data should be at 2D, and axis 0 is expected
to be the time dimension.
targets: Targets corresponding to timesteps in `data`.
It should have same length as `data`.
length: Length of the output sequences (in number of timesteps).
sampling_rate: Period between successive individual timesteps
within sequences. For rate `r`, timesteps
`data[i]`, `data[i-r]`, ... `data[i - length]`
are used for create a sample sequence.
stride: Period between successive output sequences.
For stride `s`, consecutive output samples would
be centered around `data[i]`, `data[i+s]`, `data[i+2*s]`, etc.
start_index: Data points earlier than `start_index` will not be used
in the output sequences. This is useful to reserve part of the
data for test or validation.
end_index: Data points later than `end_index` will not be used
in the output sequences. This is useful to reserve part of the
data for test or validation.
shuffle: Whether to shuffle output samples,
or instead draw them in chronological order.
reverse: Boolean: if `true`, timesteps in each output sample will be
in reverse chronological order.
batch_size: Number of timeseries samples in each batch
(except maybe the last one).
# Returns
A [Sequence](/utils/#sequence) instance.
# Examples
```python
from keras.preprocessing.sequence import TimeseriesGenerator
import numpy as np
data = np.array([[i] for i in range(50)])
targets = np.array([[i] for i in range(50)])
data_gen = TimeseriesGenerator(data, targets,
length=10, sampling_rate=2,
batch_size=2)
assert len(data_gen) == 20
batch_0 = data_gen[0]
x, y = batch_0
assert np.array_equal(x,
np.array([[[0], [2], [4], [6], [8]],
[[1], [3], [5], [7], [9]]]))
assert np.array_equal(y,
np.array([[10], [11]]))
```"
7192,pad_sequences,tensorflow/tensorflow/python/keras/preprocessing/sequence.py,93,function,"Pads sequences to the same length.
This function transforms a list (of length `num_samples`)
of sequences (lists of integers)
into a 2D Numpy array of shape `(num_samples, num_timesteps)`.
`num_timesteps` is either the `maxlen` argument if provided,
or the length of the longest sequence in the list.
Sequences that are shorter than `num_timesteps`
are padded with `value` until they are `num_timesteps` long.
Sequences longer than `num_timesteps` are truncated
so that they fit the desired length.
The position where padding or truncation happens is determined by
the arguments `padding` and `truncating`, respectively.
Pre-padding or removing values from the beginning of the sequence is the
default.
>>> sequence = [[1], [2, 3], [4, 5, 6]]
>>> tf.keras.preprocessing.sequence.pad_sequences(sequence)
array([[0, 0, 1],
[0, 2, 3],
[4, 5, 6]], dtype=int32)
>>> tf.keras.preprocessing.sequence.pad_sequences(sequence, value=-1)
array([[-1, -1, 1],
[-1, 2, 3],
[ 4, 5, 6]], dtype=int32)
>>> tf.keras.preprocessing.sequence.pad_sequences(sequence, padding='post')
array([[1, 0, 0],
[2, 3, 0],
[4, 5, 6]], dtype=int32)
>>> tf.keras.preprocessing.sequence.pad_sequences(sequence, maxlen=2)
array([[0, 1],
[2, 3],
[5, 6]], dtype=int32)
Arguments:
sequences: List of sequences (each sequence is a list of integers).
maxlen: Optional Int, maximum length of all sequences. If not provided,
sequences will be padded to the length of the longest individual
sequence.
dtype: (Optional, defaults to int32). Type of the output sequences.
To pad sequences with variable length strings, you can use `object`.
padding: String, 'pre' or 'post' (optional, defaults to 'pre'):
pad either before or after each sequence.
truncating: String, 'pre' or 'post' (optional, defaults to 'pre'):
remove values from sequences larger than
`maxlen`, either at the beginning or at the end of the sequences.
value: Float or String, padding value. (Optional, defaults to 0.)
Returns:
Numpy array with shape `(len(sequences), maxlen)`
Raises:
ValueError: In case of invalid values for `truncating` or `padding`,
or in case of invalid shape for a `sequences` entry."
7193,text_to_word_sequence,tensorflow/tensorflow/python/keras/preprocessing/text.py,32,function,"Converts a text to a sequence of words (or tokens).
This function transforms a string of text into a list of words
while ignoring `filters` which include punctuations by default.
>>> sample_text = 'This is a sample sentence.'
>>> tf.keras.preprocessing.text.text_to_word_sequence(sample_text)
['this', 'is', 'a', 'sample', 'sentence']
Arguments:
input_text: Input text (string).
filters: list (or concatenation) of characters to filter out, such as
punctuation. Default: `'!""#$%&()*+,-./:;<=>?@[\]^_`{|}~\t\n'`,
includes basic punctuation, tabs, and newlines.
lower: boolean. Whether to convert the input to lowercase.
split: str. Separator for word splitting.
Returns:
A list of words (or tokens)."
7194,one_hot,tensorflow/tensorflow/python/keras/preprocessing/text.py,61,function,"One-hot encodes a text into a list of word indexes of size `n`.
This function receives as input a string of text and returns a
list of encoded integers each corresponding to a word (or token)
in the given input string.
Arguments:
input_text: Input text (string).
n: int. Size of vocabulary.
filters: list (or concatenation) of characters to filter out, such as
punctuation. Default: ``!""#$%&()*+,-./:;<=>?@[\]^_`{|}~\t\n``,
includes basic punctuation, tabs, and newlines.
lower: boolean. Whether to set the text to lowercase.
split: str. Separator for word splitting.
Returns:
List of integers in `[1, n]`. Each integer encodes a word
(unicity non-guaranteed)."
7195,text_dataset_from_directory,tensorflow/tensorflow/python/keras/preprocessing/text_dataset.py,30,function,"Generates a `tf.data.Dataset` from text files in a directory.
If your directory structure is:
```
main_directory/
...class_a/
......a_text_1.txt
......a_text_2.txt
...class_b/
......b_text_1.txt
......b_text_2.txt
```
Then calling `text_dataset_from_directory(main_directory, labels='inferred')`
will return a `tf.data.Dataset` that yields batches of texts from
the subdirectories `class_a` and `class_b`, together with labels
0 and 1 (0 corresponding to `class_a` and 1 corresponding to `class_b`).
Only `.txt` files are supported at this time.
Arguments:
directory: Directory where the data is located.
If `labels` is ""inferred"", it should contain
subdirectories, each containing text files for a class.
Otherwise, the directory structure is ignored.
labels: Either ""inferred""
(labels are generated from the directory structure),
or a list/tuple of integer labels of the same size as the number of
text files found in the directory. Labels should be sorted according
to the alphanumeric order of the text file paths
(obtained via `os.walk(directory)` in Python).
label_mode:
- 'int': means that the labels are encoded as integers
(e.g. for `sparse_categorical_crossentropy` loss).
- 'categorical' means that the labels are
encoded as a categorical vector
(e.g. for `categorical_crossentropy` loss).
- 'binary' means that the labels (there can be only 2)
are encoded as `float32` scalars with values 0 or 1
(e.g. for `binary_crossentropy`).
- None (no labels).
class_names: Only valid if ""labels"" is ""inferred"". This is the explict
list of class names (must match names of subdirectories). Used
to control the order of the classes
(otherwise alphanumerical order is used).
batch_size: Size of the batches of data. Default: 32.
max_length: Maximum size of a text string. Texts longer than this will
be truncated to `max_length`.
shuffle: Whether to shuffle the data. Default: True.
If set to False, sorts the data in alphanumeric order.
seed: Optional random seed for shuffling and transformations.
validation_split: Optional float between 0 and 1,
fraction of data to reserve for validation.
subset: One of ""training"" or ""validation"".
Only used if `validation_split` is set.
follow_links: Whether to visits subdirectories pointed to by symlinks.
Defaults to False.
Returns:
A `tf.data.Dataset` object.
- If `label_mode` is None, it yields `string` tensors of shape
`(batch_size,)`, containing the contents of a batch of text files.
- Otherwise, it yields a tuple `(texts, labels)`, where `texts`
has shape `(batch_size,)` and `labels` follows the format described
below.
Rules regarding labels format:
- if `label_mode` is `int`, the labels are an `int32` tensor of shape
`(batch_size,)`.
- if `label_mode` is `binary`, the labels are a `float32` tensor of
1s and 0s of shape `(batch_size, 1)`.
- if `label_mode` is `categorial`, the labels are a `float32` tensor
of shape `(batch_size, num_classes)`, representing a one-hot
encoding of the class index."
7196,paths_and_labels_to_dataset,tensorflow/tensorflow/python/keras/preprocessing/text_dataset.py,171,function,Constructs a dataset of text strings and labels.
7197,path_to_string_content,tensorflow/tensorflow/python/keras/preprocessing/text_dataset.py,186,function,
7198,timeseries_dataset_from_array,tensorflow/tensorflow/python/keras/preprocessing/timeseries.py,30,function,"Creates a dataset of sliding windows over a timeseries provided as array.
This function takes in a sequence of data-points gathered at
equal intervals, along with time series parameters such as
length of the sequences/windows, spacing between two sequence/windows, etc.,
to produce batches of timeseries inputs and targets.
Arguments:
data: Numpy array or eager tensor
containing consecutive data points (timesteps).
Axis 0 is expected to be the time dimension.
targets: Targets corresponding to timesteps in `data`.
It should have same length as `data`. `targets[i]` should be the target
corresponding to the window that starts at index `i`
(see example 2 below).
Pass None if you don't have target data (in this case the dataset will
only yield the input data).
sequence_length: Length of the output sequences (in number of timesteps).
sequence_stride: Period between successive output sequences.
For stride `s`, output samples would
start at index `data[i]`, `data[i + s]`, `data[i + 2 * s]`, etc.
sampling_rate: Period between successive individual timesteps
within sequences. For rate `r`, timesteps
`data[i], data[i + r], ... data[i + sequence_length]`
are used for create a sample sequence.
batch_size: Number of timeseries samples in each batch
(except maybe the last one).
shuffle: Whether to shuffle output samples,
or instead draw them in chronological order.
seed: Optional int; random seed for shuffling.
start_index: Optional int; data points earlier (exclusive)
than `start_index` will not be used
in the output sequences. This is useful to reserve part of the
data for test or validation.
end_index: Optional int; data points later (exclusive) than `end_index`
will not be used in the output sequences.
This is useful to reserve part of the data for test or validation.
Returns:
A tf.data.Dataset instance. If `targets` was passed, the dataset yields
tuple `(batch_of_sequences, batch_of_targets)`. If not, the dataset yields
only `batch_of_sequences`.
Example 1:
Consider indices `[0, 1, ... 99]`.
With `sequence_length=10, sampling_rate=2, sequence_stride=3`,
`shuffle=False`, the dataset will yield batches of sequences
composed of the following indices:
```
First sequence: [0 2 4 6 8 10 12 14 16 18]
Second sequence: [3 5 7 9 11 13 15 17 19 21]
Third sequence: [6 8 10 12 14 16 18 20 22 24]
...
Last sequence: [78 80 82 84 86 88 90 92 94 96]
```
In this case the last 3 data points are discarded since no full sequence
can be generated to include them (the next sequence would have started
at index 81, and thus its last step would have gone over 99).
Example 2: temporal regression. Consider an array `data` of scalar
values, of shape `(steps,)`. To generate a dataset that uses the past 10
timesteps to predict the next timestep, you would use:
```python
input_data = data[:-10]
targets = data[10:]
dataset = tf.keras.preprocessing.timeseries_dataset_from_array(
input_data, targets, sequence_length=10)
for batch in dataset:
inputs, targets = batch
assert np.array_equal(inputs[0], data[:10]) # First sequence: steps [0-9]
assert np.array_equal(targets[0], data[10]) # Corresponding target: step 10
break
```"
7199,sequences_from_indices,tensorflow/tensorflow/python/keras/preprocessing/timeseries.py,202,function,
7200,save_model_to_hdf5,tensorflow/tensorflow/python/keras/saving/hdf5_format.py,56,function,"Saves a model to a HDF5 file.
The saved model contains:
- the model's configuration (topology)
- the model's weights
- the model's optimizer's state (if any)
Thus the saved model can be reinstantiated in
the exact same state, without any of the code
used for model definition or training.
Arguments:
model: Keras model instance to be saved.
filepath: One of the following:
- String, path where to save the model
- `h5py.File` object where to save the model
overwrite: Whether we should overwrite any existing
model at the target location, or instead
ask the user with a manual prompt.
include_optimizer: If True, save optimizer's state together.
Raises:
ImportError: if h5py is not available."
7201,load_model_from_hdf5,tensorflow/tensorflow/python/keras/saving/hdf5_format.py,133,function,"Loads a model saved via `save_model_to_hdf5`.
Arguments:
filepath: One of the following:
- String, path to the saved model
- `h5py.File` object from which to load the model
custom_objects: Optional dictionary mapping names
(strings) to custom classes or functions to be
considered during deserialization.
compile: Boolean, whether to compile the model
after loading.
Returns:
A Keras model instance. If an optimizer was found
as part of the saved model, the model is already
compiled. Otherwise, the model is uncompiled and
a warning will be displayed. When `compile` is set
to False, the compilation is omitted without any
warning.
Raises:
ImportError: if h5py is not available.
ValueError: In case of an invalid savefile."
7202,preprocess_weights_for_loading,tensorflow/tensorflow/python/keras/saving/hdf5_format.py,221,function,"Preprocess layer weights between different Keras formats.
Converts layers weights from Keras 1 format to Keras 2 and also weights of
CuDNN layers in Keras 2.
Arguments:
layer: Layer instance.
weights: List of weights values (Numpy arrays).
original_keras_version: Keras version for the weights, as a string.
original_backend: Keras backend the weights were trained with,
as a string.
Returns:
A list of weights values (Numpy arrays)."
7203,save_optimizer_weights_to_hdf5_group,tensorflow/tensorflow/python/keras/saving/hdf5_format.py,578,function,"Saves optimizer weights of a optimizer to a HDF5 group.
Arguments:
hdf5_group: HDF5 group.
optimizer: optimizer instance."
7204,load_optimizer_weights_from_hdf5_group,tensorflow/tensorflow/python/keras/saving/hdf5_format.py,602,function,"Load optimizer weights from a HDF5 group.
Arguments:
hdf5_group: A pointer to a HDF5 group.
Returns:
data: List of optimizer weight names."
7205,save_weights_to_hdf5_group,tensorflow/tensorflow/python/keras/saving/hdf5_format.py,617,function,"Saves the weights of a list of layers to a HDF5 group.
Arguments:
f: HDF5 group.
layers: List of layer instances."
7206,load_weights_from_hdf5_group,tensorflow/tensorflow/python/keras/saving/hdf5_format.py,648,function,"Implements topological (order-based) weight loading.
Arguments:
f: A pointer to a HDF5 group.
layers: a list of target layers.
Raises:
ValueError: in case of mismatch between provided layers
and weights file."
7207,load_weights_from_hdf5_group_by_name,tensorflow/tensorflow/python/keras/saving/hdf5_format.py,711,function,"Implements name-based weight loading.
(instead of topological weight loading).
Layers that have no matching name are skipped.
Arguments:
f: A pointer to a HDF5 group.
layers: a list of target layers.
skip_mismatch: Boolean, whether to skip loading of layers
where there is a mismatch in the number of weights,
or a mismatch in the shape of the weights.
Raises:
ValueError: in case of mismatch between provided layers
and weights file and skip_match=False."
7208,save_attributes_to_hdf5_group,tensorflow/tensorflow/python/keras/saving/hdf5_format.py,793,function,"Saves attributes (data) of the specified name into the HDF5 group.
This method deals with an inherent problem of HDF5 file which is not
able to store data larger than HDF5_OBJECT_HEADER_LIMIT bytes.
Arguments:
group: A pointer to a HDF5 group.
name: A name of the attributes to save.
data: Attributes data to store.
Raises:
RuntimeError: If any single attribute is too large to be saved."
7209,load_attributes_from_hdf5_group,tensorflow/tensorflow/python/keras/saving/hdf5_format.py,835,function,"Loads attributes of the specified name from the HDF5 group.
This method deals with an inherent problem
of HDF5 file which is not able to store
data larger than HDF5_OBJECT_HEADER_LIMIT bytes.
Arguments:
group: A pointer to a HDF5 group.
name: A name of the attributes to load.
Returns:
data: Attributes data."
7210,SubclassedModel,tensorflow/tensorflow/python/keras/saving/hdf5_format_test.py,973,class,
7211,call,tensorflow/tensorflow/python/keras/saving/hdf5_format_test.py,980,method,
7212,DummySubclassModel,tensorflow/tensorflow/python/keras/saving/hdf5_format_test.py,1274,class,
7213,MyMeanAbsoluteError,tensorflow/tensorflow/python/keras/saving/losses_serialization_test.py,45,class,
7214,my_mae,tensorflow/tensorflow/python/keras/saving/losses_serialization_test.py,55,function,
7215,LossesSerialization,tensorflow/tensorflow/python/keras/saving/losses_serialization_test.py,117,class,
7216,setUp,tensorflow/tensorflow/python/keras/saving/losses_serialization_test.py,119,method,
7217,MyMeanAbsoluteError,tensorflow/tensorflow/python/keras/saving/metrics_serialization_test.py,45,class,
7218,MetricsSerialization,tensorflow/tensorflow/python/keras/saving/metrics_serialization_test.py,150,class,
7219,setUp,tensorflow/tensorflow/python/keras/saving/metrics_serialization_test.py,152,method,
7220,get_instance,tensorflow/tensorflow/python/keras/saving/metrics_serialization_test.py,163,method,
7221,get_instance,tensorflow/tensorflow/python/keras/saving/metrics_serialization_test.py,208,method,
7222,model_from_config,tensorflow/tensorflow/python/keras/saving/model_config.py,35,function,"Instantiates a Keras model from its config.
Arguments:
config: Configuration dictionary.
custom_objects: Optional dictionary mapping names
(strings) to custom classes or functions to be
considered during deserialization.
Returns:
A Keras model instance (uncompiled).
Raises:
TypeError: if `config` is not a dictionary."
7223,model_from_yaml,tensorflow/tensorflow/python/keras/saving/model_config.py,59,function,"Parses a yaml model configuration file and returns a model instance.
Usage:
>>> model = tf.keras.Sequential([
... tf.keras.layers.Dense(5, input_shape=(3,)),
... tf.keras.layers.Softmax()])
>>> try:
... import yaml
... config = model.to_yaml()
... loaded_model = tf.keras.models.model_from_yaml(config)
... except ImportError:
... pass
Arguments:
yaml_string: YAML string or open file encoding a model configuration.
custom_objects: Optional dictionary mapping names
(strings) to custom classes or functions to be
considered during deserialization.
Returns:
A Keras model instance (uncompiled).
Raises:
ImportError: if yaml module is not found."
7224,model_from_json,tensorflow/tensorflow/python/keras/saving/model_config.py,100,function,"Parses a JSON model configuration string and returns a model instance.
Usage:
>>> model = tf.keras.Sequential([
... tf.keras.layers.Dense(5, input_shape=(3,)),
... tf.keras.layers.Softmax()])
>>> config = model.to_json()
>>> loaded_model = tf.keras.models.model_from_json(config)
Arguments:
json_string: JSON string encoding a model configuration.
custom_objects: Optional dictionary mapping names
(strings) to custom classes or functions to be
considered during deserialization.
Returns:
A Keras model instance (uncompiled)."
7225,save_model,tensorflow/tensorflow/python/keras/saving/save.py,49,function,"Saves a model as a TensorFlow SavedModel or HDF5 file.
Usage:
>>> model = tf.keras.Sequential([
... tf.keras.layers.Dense(5, input_shape=(3,)),
... tf.keras.layers.Softmax()])
>>> model.save('/tmp/model')
>>> loaded_model = tf.keras.models.load_model('/tmp/model')
>>> x = tf.random.uniform((10, 3))
>>> assert np.allclose(model.predict(x), loaded_model.predict(x))
The saved model contains:
- the model's configuration (topology)
- the model's weights
- the model's optimizer's state (if any)
Thus the saved model can be reinstantiated in
the exact same state, without any of the code
used for model definition or training.
Note that the model weights may have different scoped names after being
loaded. Scoped names include the model/layer names, such as
`""dense_1/kernel:0""`. It is recommended that you use the layer properties to
access specific variables, e.g. `model.get_layer(""dense_1"").kernel`.
_SavedModel serialization_
The SavedModel serialization path uses `tf.saved_model.save` to save the model
and all trackable objects attached to the model (e.g. layers and variables).
`@tf.function`-decorated methods are also saved. Additional trackable objects
and functions are added to the SavedModel to allow the model to be
loaded back as a Keras Model object.
Arguments:
model: Keras model instance to be saved.
filepath: One of the following:
- String or `pathlib.Path` object, path where to save the model
- `h5py.File` object where to save the model
overwrite: Whether we should overwrite any existing model at the target
location, or instead ask the user with a manual prompt.
include_optimizer: If True, save optimizer's state together.
save_format: Either 'tf' or 'h5', indicating whether to save the model
to Tensorflow SavedModel or HDF5. Defaults to 'tf' in TF 2.X, and 'h5'
in TF 1.X.
signatures: Signatures to save with the SavedModel. Applicable to the 'tf'
format only. Please see the `signatures` argument in
`tf.saved_model.save` for details.
options: Optional `tf.saved_model.SaveOptions` object that specifies
options for saving to SavedModel.
Raises:
ImportError: If save format is hdf5, and h5py is not available."
7226,load_model,tensorflow/tensorflow/python/keras/saving/save.py,139,function,"Loads a model saved via `model.save()`.
Usage:
>>> model = tf.keras.Sequential([
... tf.keras.layers.Dense(5, input_shape=(3,)),
... tf.keras.layers.Softmax()])
>>> model.save('/tmp/model')
>>> loaded_model = tf.keras.models.load_model('/tmp/model')
>>> x = tf.random.uniform((10, 3))
>>> assert np.allclose(model.predict(x), loaded_model.predict(x))
Note that the model weights may have different scoped names after being
loaded. Scoped names include the model/layer names, such as
`""dense_1/kernel:0""`. It is recommended that you use the layer properties to
access specific variables, e.g. `model.get_layer(""dense_1"").kernel`.
Arguments:
filepath: One of the following:
- String or `pathlib.Path` object, path to the saved model
- `h5py.File` object from which to load the model
custom_objects: Optional dictionary mapping names
(strings) to custom classes or functions to be
considered during deserialization.
compile: Boolean, whether to compile the model
after loading.
options: Optional `tf.saved_model.LoadOptions` object that specifies
options for loading from SavedModel.
Returns:
A Keras model instance. If the original model was compiled, and saved with
the optimizer, then the returned model will be compiled. Otherwise, the
model will be left uncompiled. In the case that an uncompiled model is
returned, a warning is displayed if the `compile` argument is set to
`True`.
Raises:
ImportError: if loading from an hdf5 file and h5py is not available.
IOError: In case of an invalid savefile."
7227,export_saved_model,tensorflow/tensorflow/python/keras/saving/saved_model_experimental.py,69,function,"Exports a `tf.keras.Model` as a Tensorflow SavedModel.
Note that at this time, subclassed models can only be saved using
`serving_only=True`.
The exported `SavedModel` is a standalone serialization of Tensorflow objects,
and is supported by TF language APIs and the Tensorflow Serving system.
To load the model, use the function
`tf.keras.experimental.load_from_saved_model`.
The `SavedModel` contains:
1. a checkpoint containing the model weights.
2. a `SavedModel` proto containing the Tensorflow backend graph. Separate
graphs are saved for prediction (serving), train, and evaluation. If
the model has not been compiled, then only the graph computing predictions
will be exported.
3. the model's json config. If the model is subclassed, this will only be
included if the model's `get_config()` method is overwritten.
Example:
```python
import tensorflow as tf
# Create a tf.keras model.
model = tf.keras.Sequential()
model.add(tf.keras.layers.Dense(1, input_shape=[10]))
model.summary()
# Save the tf.keras model in the SavedModel format.
path = '/tmp/simple_keras_model'
tf.keras.experimental.export_saved_model(model, path)
# Load the saved keras model back.
new_model = tf.keras.experimental.load_from_saved_model(path)
new_model.summary()
```
Args:
model: A `tf.keras.Model` to be saved. If the model is subclassed, the flag
`serving_only` must be set to True.
saved_model_path: a string specifying the path to the SavedModel directory.
custom_objects: Optional dictionary mapping string names to custom classes
or functions (e.g. custom loss functions).
as_text: bool, `False` by default. Whether to write the `SavedModel` proto
in text format. Currently unavailable in serving-only mode.
input_signature: A possibly nested sequence of `tf.TensorSpec` objects, used
to specify the expected model inputs. See `tf.function` for more details.
serving_only: bool, `False` by default. When this is true, only the
prediction graph is saved.
Raises:
NotImplementedError: If the model is a subclassed model, and serving_only is
False.
ValueError: If the input signature cannot be inferred from the model.
AssertionError: If the SavedModel directory already exists and isn't empty."
7228,create_placeholder,tensorflow/tensorflow/python/keras/saving/saved_model_experimental.py,230,function,
7229,load_from_saved_model,tensorflow/tensorflow/python/keras/saving/saved_model_experimental.py,379,function,"Loads a keras Model from a SavedModel created by `export_saved_model()`.
This function reinstantiates model state by:
1) loading model topology from json (this will eventually come
from metagraph).
2) loading model weights from checkpoint.
Example:
```python
import tensorflow as tf
# Create a tf.keras model.
model = tf.keras.Sequential()
model.add(tf.keras.layers.Dense(1, input_shape=[10]))
model.summary()
# Save the tf.keras model in the SavedModel format.
path = '/tmp/simple_keras_model'
tf.keras.experimental.export_saved_model(model, path)
# Load the saved keras model back.
new_model = tf.keras.experimental.load_from_saved_model(path)
new_model.summary()
```
Args:
saved_model_path: a string specifying the path to an existing SavedModel.
custom_objects: Optional dictionary mapping names
(strings) to custom classes or functions to be
considered during deserialization.
Returns:
a keras.Model instance."
7230,LayerWithLearningPhase,tensorflow/tensorflow/python/keras/saving/saved_model_experimental_test.py,203,class,
7231,build,tensorflow/tensorflow/python/keras/saving/saved_model_experimental_test.py,205,method,
7232,call,tensorflow/tensorflow/python/keras/saving/saved_model_experimental_test.py,209,method,
7233,compute_output_shape,tensorflow/tensorflow/python/keras/saving/saved_model_experimental_test.py,218,method,
7234,functional_model,tensorflow/tensorflow/python/keras/saving/saved_model_experimental_test.py,222,function,
7235,sequential_model,tensorflow/tensorflow/python/keras/saving/saved_model_experimental_test.py,231,function,
7236,sequential_model_without_input_shape,tensorflow/tensorflow/python/keras/saving/saved_model_experimental_test.py,240,function,
7237,Subclassed,tensorflow/tensorflow/python/keras/saving/saved_model_experimental_test.py,249,class,
7238,call,tensorflow/tensorflow/python/keras/saving/saved_model_experimental_test.py,256,method,
7239,subclassed_model,tensorflow/tensorflow/python/keras/saving/saved_model_experimental_test.py,262,function,
7240,load_model,tensorflow/tensorflow/python/keras/saving/saved_model_experimental_test.py,266,function,
7241,extract_model_metrics,tensorflow/tensorflow/python/keras/saving/saving_utils.py,37,function,"Convert metrics from a Keras model `compile` API to dictionary.
This is used for converting Keras models to Estimators and SavedModels.
Args:
model: A `tf.keras.Model` object.
Returns:
Dictionary mapping metric names to metric instances. May return `None` if
the model does not contain any metrics."
7242,model_input_signature,tensorflow/tensorflow/python/keras/saving/saving_utils.py,57,function,"Inspect model to get its input signature.
The model's input signature is a list with a single (possibly-nested) object.
This is due to the Keras-enforced restriction that tensor inputs must be
passed in as the first argument.
For example, a model with input {'feature1': <Tensor>, 'feature2': <Tensor>}
will have input signature: [{'feature1': TensorSpec, 'feature2': TensorSpec}]
Args:
model: Keras Model object.
keep_original_batch_size: A boolean indicating whether we want to keep using
the original batch size or set it to None. Default is `False`, which means
that the batch dim of the returned input signature will always be set to
`None`.
Returns:
A list containing either a single TensorSpec or an object with nested
TensorSpecs. This list does not contain the `training` argument."
7243,raise_model_input_error,tensorflow/tensorflow/python/keras/saving/saving_utils.py,92,function,
7244,trace_model_call,tensorflow/tensorflow/python/keras/saving/saving_utils.py,100,function,"Trace the model call to create a tf.function for exporting a Keras model.
Args:
model: A Keras model.
input_signature: optional, a list of tf.TensorSpec objects specifying the
inputs to the model.
Returns:
A tf.function wrapping the model's call function with input signatures set.
Raises:
ValueError: if input signature cannot be inferred from the model."
7245,model_metadata,tensorflow/tensorflow/python/keras/saving/saving_utils.py,147,function,Returns a dictionary containing the model metadata.
7246,should_overwrite,tensorflow/tensorflow/python/keras/saving/saving_utils.py,196,function,Returns whether the filepath should be overwritten.
7247,compile_args_from_training_config,tensorflow/tensorflow/python/keras/saving/saving_utils.py,204,function,Return model.compile arguments from training config.
7248,try_build_compiled_arguments,tensorflow/tensorflow/python/keras/saving/saving_utils.py,313,function,
7249,SavedModelSaver,tensorflow/tensorflow/python/keras/saving/saved_model/base_serialization.py,29,class,"Saver defining the methods and properties used to serialize Keras objects.
"
7250,object_identifier,tensorflow/tensorflow/python/keras/saving/saved_model/base_serialization.py,37,method,"String stored in object identifier field in the SavedModel proto.
Returns:
A string with the object identifier, which is used at load time."
7251,tracking_metadata,tensorflow/tensorflow/python/keras/saving/saved_model/base_serialization.py,46,method,"String stored in metadata field in the SavedModel proto.
Returns:
A serialized JSON storing information necessary for recreating this layer."
7252,list_extra_dependencies_for_serialization,tensorflow/tensorflow/python/keras/saving/saved_model/base_serialization.py,56,method,"Lists extra dependencies to serialize to SavedModel.
By overriding this method, extra dependencies can be attached to the
serialized Layer. For example, this is used to save the list of `variables`
and `trainable_variables`, which are python properties in a Layer object,
but are represented as a static list in the SavedModel.
Args:
serialization_cache: A dictionary shared between all objects in the same
object graph. This object is passed to both
`_list_extra_dependencies_for_serialization` and
`_list_functions_for_serialization`.
Returns:
A dictionary mapping attribute names to trackable objects. The entire list
of attributes are listed in the `saved_model._LayerAttributes` class."
7253,list_functions_for_serialization,tensorflow/tensorflow/python/keras/saving/saved_model/base_serialization.py,76,method,"Lists extra functions to serialize to the SavedModel.
Args:
serialization_cache: Dictionary passed to all objects in the same object
graph during serialization.
Returns:
A dictionary mapping attribute names to `Function` or
`ConcreteFunction`."
7254,python_properties,tensorflow/tensorflow/python/keras/saving/saved_model/base_serialization.py,98,method,"Returns dictionary of python properties to save in the metadata.
This dictionary must be serializable and deserializable to/from JSON.
When loading, the items in this dict are used to initialize the object and
define attributes in the revived object."
7255,objects_to_serialize,tensorflow/tensorflow/python/keras/saving/saved_model/base_serialization.py,109,method,"Returns dictionary of extra checkpointable objects to serialize.
See `functions_to_serialize` for an explanation of this function's
effects.
Args:
serialization_cache: Dictionary passed to all objects in the same object
graph during serialization.
Returns:
A dictionary mapping attribute names to checkpointable objects."
7256,functions_to_serialize,tensorflow/tensorflow/python/keras/saving/saved_model/base_serialization.py,125,method,"Returns extra functions to include when serializing a Keras object.
Normally, when calling exporting an object to SavedModel, only the
functions and objects defined by the user are saved. For example:
```
obj = tf.Module()
obj.v = tf.Variable(1.)
@tf.function
def foo(...): ...
obj.foo = foo
w = tf.Variable(1.)
tf.saved_model.save(obj, 'path/to/saved/model')
loaded = tf.saved_model.load('path/to/saved/model')
loaded.v # Variable with the same value as obj.v
loaded.foo # Equivalent to obj.foo
loaded.w # AttributeError
```
Assigning trackable objects to attributes creates a graph, which is used for
both checkpointing and SavedModel serialization.
When the graph generated from attribute tracking is insufficient, extra
objects and functions may be added at serialization time. For example,
most models do not have their call function wrapped with a @tf.function
decorator. This results in `model.call` not being saved. Since Keras objects
should be revivable from the SavedModel format, the call function is added
as an extra function to serialize.
This function and `objects_to_serialize` is called multiple times when
exporting to SavedModel. Please use the cache to avoid generating new
functions and objects. A fresh cache is created for each SavedModel export.
Args:
serialization_cache: Dictionary passed to all objects in the same object
graph during serialization.
Returns:
A dictionary mapping attribute names to `Function` or
`ConcreteFunction`."
7257,Encoder,tensorflow/tensorflow/python/keras/saving/saved_model/json_utils.py,43,class,JSON encoder and decoder that handles TensorShapes and tuples.
7258,default,tensorflow/tensorflow/python/keras/saving/saved_model/json_utils.py,46,method,
7259,encode,tensorflow/tensorflow/python/keras/saving/saved_model/json_utils.py,52,method,
7260,decode,tensorflow/tensorflow/python/keras/saving/saved_model/json_utils.py,68,function,
7261,get_json_type,tensorflow/tensorflow/python/keras/saving/saved_model/json_utils.py,84,function,"Serializes any object to a JSON-serializable structure.
Arguments:
obj: the object to serialize
Returns:
JSON-serializable structure representing `obj`.
Raises:
TypeError: if `obj` cannot be serialized."
7262,LayerSavedModelSaver,tensorflow/tensorflow/python/keras/saving/saved_model/layer_serialization.py,31,class,Implements Layer SavedModel serialization.
7263,object_identifier,tensorflow/tensorflow/python/keras/saving/saved_model/layer_serialization.py,35,method,
7264,python_properties,tensorflow/tensorflow/python/keras/saving/saved_model/layer_serialization.py,39,method,
7265,objects_to_serialize,tensorflow/tensorflow/python/keras/saving/saved_model/layer_serialization.py,73,method,
7266,functions_to_serialize,tensorflow/tensorflow/python/keras/saving/saved_model/layer_serialization.py,77,method,
7267,get_config,tensorflow/tensorflow/python/keras/saving/saved_model/layer_serialization.py,113,function,
7268,InputLayerSavedModelSaver,tensorflow/tensorflow/python/keras/saving/saved_model/layer_serialization.py,125,class,InputLayer serialization.
7269,object_identifier,tensorflow/tensorflow/python/keras/saving/saved_model/layer_serialization.py,129,method,
7270,python_properties,tensorflow/tensorflow/python/keras/saving/saved_model/layer_serialization.py,133,method,
7271,objects_to_serialize,tensorflow/tensorflow/python/keras/saving/saved_model/layer_serialization.py,144,method,
7272,functions_to_serialize,tensorflow/tensorflow/python/keras/saving/saved_model/layer_serialization.py,147,method,
7273,RNNSavedModelSaver,tensorflow/tensorflow/python/keras/saving/saved_model/layer_serialization.py,151,class,RNN layer serialization.
7274,object_identifier,tensorflow/tensorflow/python/keras/saving/saved_model/layer_serialization.py,155,method,
7275,load,tensorflow/tensorflow/python/keras/saving/saved_model/load.py,93,function,"Loads Keras objects from a SavedModel.
Any Keras layer or model saved to the SavedModel will be loaded back
as Keras objects. Other objects are loaded as regular trackable objects (same
as `tf.saved_model.load`).
Currently, Keras saving/loading only retains the Keras object's weights,
losses, and call function.
The loaded model can be re-compiled, but the original optimizer, compiled loss
functions, and metrics are not retained. This is temporary, and `model.save`
will soon be able to serialize compiled models.
Args:
path: Path to SavedModel.
compile: If true, compile the model after loading it.
options: Optional `tf.saved_model.LoadOptions` object that specifies
options for loading from SavedModel.
Returns:
Object loaded from SavedModel."
7276,KerasObjectLoader,tensorflow/tensorflow/python/keras/saving/saved_model/load.py,157,class,"Loader that recreates Keras objects (e.g. layers, models).
Layers and models are revived from either the config or SavedModel following
these rules:
1. If object is a graph network (i.e. Sequential or Functional) then it will
be initialized using the structure from the config only after the children
layers have been created. Graph networks must be initialized with inputs
and outputs, so all child layers must be created beforehand.
2. If object's config exists and the class can be found, then revive from
config.
3. Object may have already been created if its parent was revived from config.
In this case, do nothing.
4. If nothing of the above applies, compose the various artifacts from the
SavedModel to create a subclassed layer or model. At this time, custom
metrics are not supported."
7277,setattr_wrapper,tensorflow/tensorflow/python/keras/saving/saved_model/load.py,675,method,
7278,revive_custom_object,tensorflow/tensorflow/python/keras/saving/saved_model/load.py,820,function,Revives object from SavedModel.
7279,RevivedLayer,tensorflow/tensorflow/python/keras/saving/saved_model/load.py,860,class,Keras layer loaded from a SavedModel.
7280,keras_api,tensorflow/tensorflow/python/keras/saving/saved_model/load.py,898,method,
7281,get_config,tensorflow/tensorflow/python/keras/saving/saved_model/load.py,901,method,
7282,RevivedInputLayer,tensorflow/tensorflow/python/keras/saving/saved_model/load.py,930,class,InputLayer loaded from a SavedModel.
7283,get_config,tensorflow/tensorflow/python/keras/saving/saved_model/load.py,948,method,
7284,recursively_deserialize_keras_object,tensorflow/tensorflow/python/keras/saving/saved_model/load.py,952,function,Deserialize Keras object from a nested structure.
7285,infer_inputs_from_restored_call_function,tensorflow/tensorflow/python/keras/saving/saved_model/load.py,969,function,"Returns TensorSpec of inputs from a restored call function.
Args:
fn: Restored layer call function. It is assumed that the inputs are entirely
in the first argument.
Returns:
TensorSpec of call function inputs."
7286,RevivedNetwork,tensorflow/tensorflow/python/keras/saving/saved_model/load.py,989,class,Keras network of layers loaded from a SavedModel.
7287,MetricSavedModelSaver,tensorflow/tensorflow/python/keras/saving/saved_model/metric_serialization.py,25,class,Metric serialization.
7288,object_identifier,tensorflow/tensorflow/python/keras/saving/saved_model/metric_serialization.py,29,method,
7289,ModelSavedModelSaver,tensorflow/tensorflow/python/keras/saving/saved_model/model_serialization.py,27,class,Model SavedModel serialization.
7290,object_identifier,tensorflow/tensorflow/python/keras/saving/saved_model/model_serialization.py,31,method,
7291,SequentialSavedModelSaver,tensorflow/tensorflow/python/keras/saving/saved_model/model_serialization.py,62,class,
7292,object_identifier,tensorflow/tensorflow/python/keras/saving/saved_model/model_serialization.py,65,method,
7293,NetworkSavedModelSaver,tensorflow/tensorflow/python/keras/saving/saved_model/network_serialization.py,25,class,Network serialization.
7294,object_identifier,tensorflow/tensorflow/python/keras/saving/saved_model/network_serialization.py,29,method,
7295,SubclassedModelNoConfig,tensorflow/tensorflow/python/keras/saving/saved_model/revive_test.py,47,class,
7296,build,tensorflow/tensorflow/python/keras/saving/saved_model/revive_test.py,57,method,
7297,call,tensorflow/tensorflow/python/keras/saving/saved_model/revive_test.py,68,method,
7298,SubclassedModelWithConfig,tensorflow/tensorflow/python/keras/saving/saved_model/revive_test.py,75,class,
7299,get_config,tensorflow/tensorflow/python/keras/saving/saved_model/revive_test.py,77,method,
7300,from_config,tensorflow/tensorflow/python/keras/saving/saved_model/revive_test.py,82,method,
7301,CustomLayerNoConfig,tensorflow/tensorflow/python/keras/saving/saved_model/revive_test.py,86,class,
7302,build,tensorflow/tensorflow/python/keras/saving/saved_model/revive_test.py,98,method,
7303,call,tensorflow/tensorflow/python/keras/saving/saved_model/revive_test.py,102,method,
7304,a_regularizer,tensorflow/tensorflow/python/keras/saving/saved_model/revive_test.py,92,method,
7305,CustomLayerWithConfig,tensorflow/tensorflow/python/keras/saving/saved_model/revive_test.py,110,class,
7306,get_config,tensorflow/tensorflow/python/keras/saving/saved_model/revive_test.py,112,method,
7307,save,tensorflow/tensorflow/python/keras/saving/saved_model/save.py,40,function,"Saves a model as a SavedModel to the filepath.
Args:
model: Keras model instance to be saved.
filepath: String path to save the model.
overwrite: whether to overwrite the existing filepath.
include_optimizer: If True, save the model's optimizer state.
signatures: Signatures to save with the SavedModel. Applicable to the 'tf'
format only. Please see the `signatures` argument in `tf.saved_model.save`
for details.
options: Optional `tf.saved_model.SaveOptions` object that specifies
options for saving to SavedModel.
Raises:
ValueError: if the model's inputs have not been defined."
7308,should_skip_serialization,tensorflow/tensorflow/python/keras/saving/saved_model/save_impl.py,72,function,Skip serializing extra objects and functions if layer inputs aren't set.
7309,wrap_layer_objects,tensorflow/tensorflow/python/keras/saving/saved_model/save_impl.py,83,function,"Returns extra trackable objects to attach to the serialized layer.
Args:
layer: Keras Layer object.
serialization_cache: Dictionary shared between all objects during
serialization.
Returns:
A dictionary containing all checkpointable objects from a
SerializedAttributes object. See LayerAttributes and ModelAttributes for
entire list of objects"
7310,wrap_layer_functions,tensorflow/tensorflow/python/keras/saving/saved_model/save_impl.py,134,function,"Returns dict of wrapped layer call function and losses in tf.functions.
Args:
layer: Keras Layer object.
serialization_cache: Dictionary shared between all objects during
serialization.
Returns:
A dictionary containing all keras tf.functions to serialize. See
LayerAttributes and ModelAttributes for the list of all attributes."
7311,default_save_signature,tensorflow/tensorflow/python/keras/saving/saved_model/save_impl.py,202,function,
7312,LayerCallCollection,tensorflow/tensorflow/python/keras/saving/saved_model/save_impl.py,329,class,"Groups wrapped layer call functions.
This is used to ensure that all layer call functions are traced with the same
inputs-
- call
- call_and_return_conditional_losses
- call_and_return_all_conditional_losses"
7313,add_trace,tensorflow/tensorflow/python/keras/saving/saved_model/save_impl.py,401,method,"Traces all functions with the same args and kwargs.
Args:
*args: Positional args passed to the original function.
**kwargs: Keyword args passed to the original function."
7314,fn_input_signature,tensorflow/tensorflow/python/keras/saving/saved_model/save_impl.py,427,method,Returns input signature for the wrapped layer call function.
7315,training_arg_was_passed,tensorflow/tensorflow/python/keras/saving/saved_model/save_impl.py,438,method,
7316,get_training_arg_value,tensorflow/tensorflow/python/keras/saving/saved_model/save_impl.py,446,method,
7317,get_input_arg_value,tensorflow/tensorflow/python/keras/saving/saved_model/save_impl.py,453,method,
7318,add_function,tensorflow/tensorflow/python/keras/saving/saved_model/save_impl.py,495,method,Adds a layer call function to the collection.
7319,wrap_with_training_arg,tensorflow/tensorflow/python/keras/saving/saved_model/save_impl.py,479,method,
7320,trace_with_training,tensorflow/tensorflow/python/keras/saving/saved_model/save_impl.py,415,method,
7321,to_tensor_spec_or_none,tensorflow/tensorflow/python/keras/saving/saved_model/save_impl.py,385,method,
7322,layer_call_wrapper,tensorflow/tensorflow/python/keras/saving/saved_model/save_impl.py,509,function,"Ensures layer losses are kept the same, and runs method in call context."
7323,LayerCall,tensorflow/tensorflow/python/keras/saving/saved_model/save_impl.py,532,class,Function that triggers traces of other functions in the same collection.
7324,get_concrete_function,tensorflow/tensorflow/python/keras/saving/saved_model/save_impl.py,546,method,
7325,LayerWithLearningPhase,tensorflow/tensorflow/python/keras/saving/saved_model/saved_model_test.py,69,class,
7326,build,tensorflow/tensorflow/python/keras/saving/saved_model/saved_model_test.py,71,method,
7327,call,tensorflow/tensorflow/python/keras/saving/saved_model/saved_model_test.py,75,method,
7328,compute_output_shape,tensorflow/tensorflow/python/keras/saving/saved_model/saved_model_test.py,84,method,
7329,LayerWithLoss,tensorflow/tensorflow/python/keras/saving/saved_model/saved_model_test.py,88,class,
7330,call,tensorflow/tensorflow/python/keras/saving/saved_model/saved_model_test.py,90,method,
7331,LayerWithUpdate,tensorflow/tensorflow/python/keras/saving/saved_model/saved_model_test.py,95,class,
7332,build,tensorflow/tensorflow/python/keras/saving/saved_model/saved_model_test.py,97,method,
7333,call,tensorflow/tensorflow/python/keras/saving/saved_model/saved_model_test.py,105,method,
7334,GlobalLayerThatShouldFailIfNotAdded,tensorflow/tensorflow/python/keras/saving/saved_model/saved_model_test.py,112,class,
7335,SerializedAttributes,tensorflow/tensorflow/python/keras/saving/saved_model/serialized_attributes.py,45,class,"Class that tracks and validates all serialization attributes.
Keras models contain many Python-defined components. For example, the
trainable_variable property lists the model's trainable variables by
recursively retrieving the trainable variables from each of the child layers.
Another example is model.call, a python function that calls child layers and
adds ops to the backend graph.
Only Tensorflow checkpointable objects and functions can be serialized to
SavedModel. Serializing a Keras model as-is results in a checkpointable object
that does not resemble a Keras model at all. Thus, extra checkpointable
objects and functions must be created during serialization.
**Defining new serialized attributes**
Child classes should be defined using:
SerializedAttributes.with_attributes(
'name', checkpointable_objects=[...], functions=[...], copy_from=[...])
This class is used to cache generated checkpointable objects and functions,
ensuring that new objects and functions are generated a single time.
**Usage during serialization**
Each Layer/Model object should have a corresponding instance of
SerializedAttributes. Create a new instance by calling
`SerializedAttributes.new(obj)`. Objects and functions may be saved using
`.set_and_validate_checkpointable_objects`/`.set_and_and_validate_functions`.
The properties `.checkpointable_objects` and `.functions` returns the cached
values.
**Adding/changing attributes to save to SavedModel**
1. Change the call to `SerializedAttributes.with_attributes` in the correct
class:
- CommonEndpoints: Base attributes to be added during serialization. If
these attributes are present in a Trackable object, it can be
deserialized to a Keras Model.
- LayerAttributes: Attributes to serialize for Layer objects.
- ModelAttributes: Attributes to serialize for Model objects.
2. Update class docstring
3. Update arguments to any calls to `set_and_validate_*`. For example, if
`call_raw_tensors` is added to the ModelAttributes function list, then
a `call_raw_tensors` function should be passed to
`set_and_validate_functions`.
**Common endpoints vs other attributes**
Only common endpoints are attached directly to the root object. Keras-specific
attributes are saved to a separate trackable object with the name ""keras_api"".
The number of objects attached to the root is limited because any naming
conflicts will cause user code to break.
Another reason is that this will only affect users who call
`tf.saved_model.load` instead of `tf.keras.models.load_model`. These are
advanced users who are likely to have defined their own tf.functions and
trackable objects. The added Keras-specific attributes are kept out of the way
in the ""keras_api"" namespace.
Properties defined in this class may be used to filter out keras-specific
attributes:
- `functions_to_serialize`: Returns dict of functions to attach to the root
object.
- `checkpointable_objects_to_serialize`: Returns dict of objects to attach to
the root object (including separate trackable object containing
keras-specific attributes)
All changes to the serialized attributes must be backwards-compatible, so
attributes should not be removed or modified without sufficient justification."
7336,with_attributes,tensorflow/tensorflow/python/keras/saving/saved_model/serialized_attributes.py,113,method,"Creates a subclass with all attributes as specified in the arguments.
Args:
name: Name of subclass
checkpointable_objects: List of checkpointable objects to be serialized
in the SavedModel.
functions: List of functions to be serialized in the SavedModel.
copy_from: List of other SerializedAttributes subclasses. The returned
class will copy checkpoint objects/functions from each subclass.
Returns:
Child class with attributes as defined in the `checkpointable_objects`
and `functions` lists."
7337,new,tensorflow/tensorflow/python/keras/saving/saved_model/serialized_attributes.py,143,method,Returns a new SerializedAttribute object.
7338,functions,tensorflow/tensorflow/python/keras/saving/saved_model/serialized_attributes.py,163,method,Returns dictionary of all functions.
7339,checkpointable_objects,tensorflow/tensorflow/python/keras/saving/saved_model/serialized_attributes.py,169,method,Returns dictionary of all checkpointable objects.
7340,functions_to_serialize,tensorflow/tensorflow/python/keras/saving/saved_model/serialized_attributes.py,175,method,Returns functions to attach to the root object during serialization.
7341,objects_to_serialize,tensorflow/tensorflow/python/keras/saving/saved_model/serialized_attributes.py,181,method,Returns objects to attach to the root object during serialization.
7342,set_and_validate_functions,tensorflow/tensorflow/python/keras/saving/saved_model/serialized_attributes.py,188,method,"Saves function dictionary, and validates dictionary values."
7343,set_and_validate_objects,tensorflow/tensorflow/python/keras/saving/saved_model/serialized_attributes.py,205,method,"Saves objects to a dictionary, and validates the values."
7344,CommonEndpoints,tensorflow/tensorflow/python/keras/saving/saved_model/serialized_attributes.py,221,class,"Common endpoints shared by all models loadable by Keras.
List of all attributes:
variables: List of all variables in the model and its sublayers.
trainable_variables: List of all trainable variables in the model and its
sublayers.
regularization_losses: List of all unconditional losses (losses not
dependent on the inputs) in the model and its sublayers.
__call__: Function that takes inputs and returns the outputs of the model
call function.
call_and_return_all_conditional_losses: Function that returns a tuple of
(call function outputs, list of all losses that depend on the inputs).
_default_save_signature: Traced model call function. This is only included
if the top level exported object is a Keras model."
7345,LayerAttributes,tensorflow/tensorflow/python/keras/saving/saved_model/serialized_attributes.py,244,class,"Layer checkpointable objects + functions that are saved to the SavedModel.
List of all attributes:
All attributes from CommonEndpoints
non_trainable_variables: List of non-trainable variables in the layer and
its sublayers.
layers: List of all sublayers.
metrics: List of all metrics in the layer and its sublayers.
call_and_return_conditional_losses: Function that takes inputs and returns a
tuple of (outputs of the call function, list of input-dependent losses).
The list of losses excludes the activity regularizer function, which is
separate to allow the deserialized Layer object to define a different
activity regularizer.
activity_regularizer_fn: Callable that returns the activity regularizer loss
layer_regularization_losses: List of losses owned only by this layer.
layer_metrics: List of metrics owned by this layer."
7346,ModelAttributes,tensorflow/tensorflow/python/keras/saving/saved_model/serialized_attributes.py,270,class,"Model checkpointable objects + functions that are saved to the SavedModel.
List of all attributes:
All attributes from LayerAttributes (including CommonEndpoints)"
7347,MetricAttributes,tensorflow/tensorflow/python/keras/saving/saved_model/serialized_attributes.py,282,class,"Attributes that are added to Metric objects when saved to SavedModel.
List of all attributes:
variables: list of all variables"
7348,RNNAttributes,tensorflow/tensorflow/python/keras/saving/saved_model/serialized_attributes.py,296,class,"RNN checkpointable objects + functions that are saved to the SavedModel.
List of all attributes:
All attributes from LayerAttributes (including CommonEndpoints)
states: List of state variables"
7349,use_wrapped_call,tensorflow/tensorflow/python/keras/saving/saved_model/utils.py,40,function,"Creates fn that adds the losses returned by call_fn & returns the outputs.
Args:
layer: A Keras layer object
call_fn: tf.function that takes layer inputs (and possibly a training arg),
and returns a tuple of (outputs, list of losses).
default_training_value: Default value of the training kwarg. If `None`, the
default is `K.learning_phase()`.
return_method: Whether to return a method bound to the layer.
Returns:
function that calls call_fn and returns the outputs. Losses returned by
call_fn are added to the layer losses."
7350,layer_uses_training_bool,tensorflow/tensorflow/python/keras/saving/saved_model/utils.py,99,function,Returns whether this layer or any of its children uses the training arg.
7351,list_all_layers,tensorflow/tensorflow/python/keras/saving/saved_model/utils.py,116,function,
7352,list_all_layers_and_sublayers,tensorflow/tensorflow/python/keras/saving/saved_model/utils.py,124,function,
7353,maybe_add_training_arg,tensorflow/tensorflow/python/keras/saving/saved_model/utils.py,131,function,"Decorate call and optionally adds training argument.
If a layer expects a training argument, this function ensures that 'training'
is present in the layer args or kwonly args, with the default training value.
Args:
original_call: Original call function.
wrapped_call: Wrapped call function.
expects_training_arg: Whether to include 'training' argument.
default_training_value: Default value of the training kwarg to include in
the arg spec. If `None`, the default is `K.learning_phase()`.
Returns:
Tuple of (
function that calls `wrapped_call` and sets the training arg,
Argspec of returned function or `None` if the argspec is unchanged)"
7354,get_training_arg_index,tensorflow/tensorflow/python/keras/saving/saved_model/utils.py,202,function,"Returns the index of 'training' in the layer call function arguments.
Args:
call_fn: Call function.
Returns:
- n: index of 'training' in the call function arguments.
- -1: if 'training' is not found in the arguments, but layer.call accepts
variable keyword arguments
- None: if layer doesn't expect a training argument."
7355,set_training_arg,tensorflow/tensorflow/python/keras/saving/saved_model/utils.py,223,function,
7356,get_training_arg,tensorflow/tensorflow/python/keras/saving/saved_model/utils.py,233,function,
7357,remove_training_arg,tensorflow/tensorflow/python/keras/saving/saved_model/utils.py,242,function,
7358,get_ctl_train_step,tensorflow/tensorflow/python/keras/tests/add_loss_correctness_test.py,44,function,
7359,get_tpu_cluster_resolver,tensorflow/tensorflow/python/keras/tests/automatic_outside_compilation_test.py,57,function,
7360,get_tpu_strategy,tensorflow/tensorflow/python/keras/tests/automatic_outside_compilation_test.py,66,function,
7361,LayerForScalarSummary,tensorflow/tensorflow/python/keras/tests/automatic_outside_compilation_test.py,73,class,A pass-through layer that only records scalar values to summary.
7362,call,tensorflow/tensorflow/python/keras/tests/automatic_outside_compilation_test.py,76,method,
7363,LayerForImageSummary,tensorflow/tensorflow/python/keras/tests/automatic_outside_compilation_test.py,81,class,A pass-through layer that only records image values to summary.
7364,call,tensorflow/tensorflow/python/keras/tests/automatic_outside_compilation_test.py,84,method,
7365,LayerForHistogramSummary,tensorflow/tensorflow/python/keras/tests/automatic_outside_compilation_test.py,89,class,A pass-through layer that records histogram values to summary.
7366,call,tensorflow/tensorflow/python/keras/tests/automatic_outside_compilation_test.py,92,method,
7367,CustomModel,tensorflow/tensorflow/python/keras/tests/automatic_outside_compilation_test.py,97,class,Custom model with summary ops in model call definition.
7368,call,tensorflow/tensorflow/python/keras/tests/automatic_outside_compilation_test.py,117,method,
7369,get_image_dataset,tensorflow/tensorflow/python/keras/tests/automatic_outside_compilation_test.py,124,function,
7370,mnist_model,tensorflow/tensorflow/python/keras/tests/automatic_outside_compilation_test.py,133,function,Creates a MNIST model.
7371,LayerWithLosses,tensorflow/tensorflow/python/keras/tests/custom_training_loop_test.py,35,class,
7372,build,tensorflow/tensorflow/python/keras/tests/custom_training_loop_test.py,37,method,
7373,call,tensorflow/tensorflow/python/keras/tests/custom_training_loop_test.py,44,method,
7374,LayerWithMetrics,tensorflow/tensorflow/python/keras/tests/custom_training_loop_test.py,49,class,
7375,build,tensorflow/tensorflow/python/keras/tests/custom_training_loop_test.py,51,method,
7376,call,tensorflow/tensorflow/python/keras/tests/custom_training_loop_test.py,54,method,
7377,LayerWithTrainingArg,tensorflow/tensorflow/python/keras/tests/custom_training_loop_test.py,61,class,
7378,call,tensorflow/tensorflow/python/keras/tests/custom_training_loop_test.py,63,method,
7379,add_loss_step,tensorflow/tensorflow/python/keras/tests/custom_training_loop_test.py,71,function,
7380,batch_norm_step,tensorflow/tensorflow/python/keras/tests/custom_training_loop_test.py,92,function,
7381,add_metric_step,tensorflow/tensorflow/python/keras/tests/custom_training_loop_test.py,115,function,
7382,SingleLayerNet,tensorflow/tensorflow/python/keras/tests/memory_test.py,35,class,Simple keras model used to ensure that there are no leaks.
7383,call,tensorflow/tensorflow/python/keras/tests/memory_test.py,42,method,
7384,basic_sequential,tensorflow/tensorflow/python/keras/tests/model_architectures.py,29,function,Basic sequential model.
7385,basic_sequential_deferred,tensorflow/tensorflow/python/keras/tests/model_architectures.py,38,function,Sequential model with deferred input shape.
7386,stacked_rnn,tensorflow/tensorflow/python/keras/tests/model_architectures.py,47,function,Stacked RNN model.
7387,lstm,tensorflow/tensorflow/python/keras/tests/model_architectures.py,57,function,LSTM model.
7388,multi_input_multi_output,tensorflow/tensorflow/python/keras/tests/model_architectures.py,68,function,Multi-input Multi-ouput model.
7389,nested_sequential_in_functional,tensorflow/tensorflow/python/keras/tests/model_architectures.py,85,function,A sequential model nested in a functional model.
7390,seq_to_seq,tensorflow/tensorflow/python/keras/tests/model_architectures.py,99,function,Sequence to sequence model.
7391,shared_layer_functional,tensorflow/tensorflow/python/keras/tests/model_architectures.py,121,function,Shared layer in a functional model.
7392,shared_sequential,tensorflow/tensorflow/python/keras/tests/model_architectures.py,140,function,Shared sequential model in a functional model.
7393,MySubclassModel,tensorflow/tensorflow/python/keras/tests/model_architectures.py,156,class,A subclass model.
7394,call,tensorflow/tensorflow/python/keras/tests/model_architectures.py,167,method,
7395,get_config,tensorflow/tensorflow/python/keras/tests/model_architectures.py,173,method,
7396,from_config,tensorflow/tensorflow/python/keras/tests/model_architectures.py,177,method,
7397,nested_subclassed_model,tensorflow/tensorflow/python/keras/tests/model_architectures.py,181,function,A subclass model nested in another subclass model.
7398,nested_subclassed_in_functional_model,tensorflow/tensorflow/python/keras/tests/model_architectures.py,203,function,A subclass model nested in a functional model.
7399,nested_functional_in_subclassed_model,tensorflow/tensorflow/python/keras/tests/model_architectures.py,214,function,A functional model nested in a subclass model.
7400,shared_layer_subclassed_model,tensorflow/tensorflow/python/keras/tests/model_architectures.py,240,function,Shared layer in a subclass model.
7401,functional_with_keyword_args,tensorflow/tensorflow/python/keras/tests/model_architectures.py,261,function,A functional model with keyword args.
7402,get_models,tensorflow/tensorflow/python/keras/tests/model_architectures.py,292,function,Get all models excluding the specificed ones.
7403,get_multi_io_subclass_model,tensorflow/tensorflow/python/keras/tests/model_subclassing_test_util.py,42,function,Creates MultiIOModel for the tests of subclass model.
7404,get_nested_model_3,tensorflow/tensorflow/python/keras/tests/model_subclassing_test_util.py,109,function,
7405,CustomCallModel,tensorflow/tensorflow/python/keras/tests/model_subclassing_test_util.py,136,class,
7406,call,tensorflow/tensorflow/python/keras/tests/model_subclassing_test_util.py,143,method,
7407,TrainingNoDefaultModel,tensorflow/tensorflow/python/keras/tests/model_subclassing_test_util.py,151,class,
7408,call,tensorflow/tensorflow/python/keras/tests/model_subclassing_test_util.py,157,method,
7409,TrainingMaskingModel,tensorflow/tensorflow/python/keras/tests/model_subclassing_test_util.py,161,class,
7410,call,tensorflow/tensorflow/python/keras/tests/model_subclassing_test_util.py,167,method,
7411,NonLayerTrackable,tensorflow/tensorflow/python/keras/tests/saver_test.py,39,class,
7412,MyModel,tensorflow/tensorflow/python/keras/tests/saver_test.py,47,class,A concrete Model for testing.
7413,call,tensorflow/tensorflow/python/keras/tests/saver_test.py,57,method,
7414,events_from_file,tensorflow/tensorflow/python/keras/tests/summary_ops_test.py,110,function,"Returns all events in a single event file.
Args:
filepath: Path to the event file.
Returns:
A list of all tf.Event protos in the event file."
7415,events_from_logdir,tensorflow/tensorflow/python/keras/tests/summary_ops_test.py,128,function,"Returns all events in the single eventfile in logdir.
Args:
logdir: The directory in which the single event file is sought.
Returns:
A list of all tf.Event protos from the single event file.
Raises:
AssertionError: If logdir does not contain exactly one file."
7416,Bias,tensorflow/tensorflow/python/keras/tests/temporal_sample_weights_correctness_test.py,32,class,Layer that add a bias to its inputs.
7417,build,tensorflow/tensorflow/python/keras/tests/temporal_sample_weights_correctness_test.py,35,method,
7418,call,tensorflow/tensorflow/python/keras/tests/temporal_sample_weights_correctness_test.py,38,method,
7419,compute_output_shape,tensorflow/tensorflow/python/keras/tests/temporal_sample_weights_correctness_test.py,41,method,
7420,get_multi_io_temporal_model,tensorflow/tensorflow/python/keras/tests/temporal_sample_weights_correctness_test.py,45,function,
7421,get_compiled_multi_io_model_temporal,tensorflow/tensorflow/python/keras/tests/temporal_sample_weights_correctness_test.py,58,function,
7422,run_with_different_sample_weight_mode_inputs,tensorflow/tensorflow/python/keras/tests/temporal_sample_weights_correctness_test.py,70,function,"Executes the given function with different sample weight mode inputs.
Args:
fn: Training or eval function to execute.
partial_sw: Boolean flag to indicate whether temporal sample weight mode
should be set partially just for one output."
7423,HasList,tensorflow/tensorflow/python/keras/tests/tracking_test.py,45,class,
7424,call,tensorflow/tensorflow/python/keras/tests/tracking_test.py,67,method,
7425,HasMapping,tensorflow/tensorflow/python/keras/tests/tracking_test.py,245,class,
7426,call,tensorflow/tensorflow/python/keras/tests/tracking_test.py,260,method,
7427,HasTuple,tensorflow/tensorflow/python/keras/tests/tracking_test.py,400,class,
7428,call,tensorflow/tensorflow/python/keras/tests/tracking_test.py,409,method,
7429,MyModel,tensorflow/tensorflow/python/keras/tests/tracking_util_test.py,56,class,A concrete Model for testing.
7430,call,tensorflow/tensorflow/python/keras/tests/tracking_util_test.py,66,method,
7431,NonLayerTrackable,tensorflow/tensorflow/python/keras/tests/tracking_util_test.py,71,class,
7432,NonLayerTrackable,tensorflow/tensorflow/python/keras/tests/tracking_util_with_v1_optimizers_test.py,49,class,
7433,MyModel,tensorflow/tensorflow/python/keras/tests/tracking_util_with_v1_optimizers_test.py,58,class,A concrete Model for testing.
7434,call,tensorflow/tensorflow/python/keras/tests/tracking_util_with_v1_optimizers_test.py,68,method,
7435,NonLayerTrackable,tensorflow/tensorflow/python/keras/tests/tracking_util_xla_test.py,32,class,
7436,Subclassed,tensorflow/tensorflow/python/keras/tests/tracking_util_xla_test.py,40,class,A concrete Model for testing.
7437,call,tensorflow/tensorflow/python/keras/tests/tracking_util_xla_test.py,50,method,
7438,Layer,tensorflow/tensorflow/python/keras/type/types.py,34,class,"This is the class from which all layers inherit.
A layer is a callable object that takes as input one or more tensors and
that outputs one or more tensors. It involves *computation*, defined
in the `call()` method, and a *state* (weight variables), defined
either in the constructor `__init__()` or in the `build()` method.
Users will just instantiate a layer and then treat it as a callable.
We recommend that descendants of `Layer` implement the following methods:
* `__init__()`: Defines custom layer attributes, and creates layer state
variables that do not depend on input shapes, using `add_weight()`.
* `build(self, input_shape)`: This method can be used to create weights that
depend on the shape(s) of the input(s), using `add_weight()`. `__call__()`
will automatically build the layer (if it has not been built yet) by
calling `build()`.
* `call(self, *args, **kwargs)`: Called in `__call__` after making sure
`build()` has been called. `call()` performs the logic of applying the
layer to the input tensors (which should be passed in as argument).
Two reserved keyword arguments you can optionally use in `call()` are:
- `training` (boolean, whether the call is in
inference mode or training mode)
- `mask` (boolean tensor encoding masked timesteps in the input, used
in RNN layers)
* `get_config(self)`: Returns a dictionary containing the configuration used
to initialize this layer. If the keys differ from the arguments
in `__init__`, then override `from_config(self)` as well.
This method is used when saving
the layer or a model that contains this layer.
Examples:
Here's a basic example: a layer with two variables, `w` and `b`,
that returns `y = w . x + b`.
It shows how to implement `build()` and `call()`.
Variables set as attributes of a layer are tracked as weights
of the layers (in `layer.weights`).
```python
class SimpleDense(Layer):
def __init__(self, units=32):
super(SimpleDense, self).__init__()
self.units = units
def build(self, input_shape): # Create the state of the layer (weights)
w_init = tf.random_normal_initializer()
self.w = tf.Variable(
initial_value=w_init(shape=(input_shape[-1], self.units),
dtype='float32'),
trainable=True)
b_init = tf.zeros_initializer()
self.b = tf.Variable(
initial_value=b_init(shape=(self.units,), dtype='float32'),
trainable=True)
def call(self, inputs): # Defines the computation from inputs to outputs
return tf.matmul(inputs, self.w) + self.b
# Instantiates the layer.
linear_layer = SimpleDense(4)
# This will also call `build(input_shape)` and create the weights.
y = linear_layer(tf.ones((2, 2)))
assert len(linear_layer.weights) == 2
# These weights are trainable, so they're listed in `trainable_weights`:
assert len(linear_layer.trainable_weights) == 2
```
Note that the method `add_weight()` offers a shortcut to create weights:
```python
class SimpleDense(Layer):
def __init__(self, units=32):
super(SimpleDense, self).__init__()
self.units = units
def build(self, input_shape):
self.w = self.add_weight(shape=(input_shape[-1], self.units),
initializer='random_normal',
trainable=True)
self.b = self.add_weight(shape=(self.units,),
initializer='random_normal',
trainable=True)
def call(self, inputs):
return tf.matmul(inputs, self.w) + self.b
```
Besides trainable weights, updated via backpropagation during training,
layers can also have non-trainable weights. These weights are meant to
be updated manually during `call()`. Here's a example layer that computes
the running sum of its inputs:
```python
class ComputeSum(Layer):
def __init__(self, input_dim):
super(ComputeSum, self).__init__()
# Create a non-trainable weight.
self.total = tf.Variable(initial_value=tf.zeros((input_dim,)),
trainable=False)
def call(self, inputs):
self.total.assign_add(tf.reduce_sum(inputs, axis=0))
return self.total
my_sum = ComputeSum(2)
x = tf.ones((2, 2))
y = my_sum(x)
print(y.numpy()) # [2. 2.]
y = my_sum(x)
print(y.numpy()) # [4. 4.]
assert my_sum.weights == [my_sum.total]
assert my_sum.non_trainable_weights == [my_sum.total]
assert my_sum.trainable_weights == []
```
For more information about creating layers, see the guide
[Writing custom layers and models with Keras](
https://www.tensorflow.org/guide/keras/custom_layers_and_models)
Arguments:
trainable: Boolean, whether the layer's variables should be trainable.
name: String name of the layer.
dtype: The dtype of the layer's computations and weights (default of
`None` means use `tf.keras.backend.floatx` in TensorFlow 2, or the type
of the first input in TensorFlow 1).
dynamic: Set this to `True` if your layer should only be run eagerly, and
should not be used to generate a static computation graph.
This would be the case for a Tree-RNN or a recursive network,
for example, or generally for any layer that manipulates tensors
using Python control flow. If `False`, we assume that the layer can
safely be used to generate a static computation graph.
Attributes:
name: The name of the layer (string).
dtype: The dtype of the layer's computations and weights. If mixed
precision is used with a `tf.keras.mixed_precision.experimental.Policy`,
this is instead just the dtype of the layer's weights, as the computations
are done in a different dtype.
updates: List of update ops of this layer.
losses: List of losses added by this layer.
trainable_weights: List of variables to be included in backprop.
non_trainable_weights: List of variables that should not be
included in backprop.
weights: The concatenation of the lists trainable_weights and
non_trainable_weights (in this order).
trainable: Whether the layer should be trained (boolean).
input_spec: Optional (list of) `InputSpec` object(s) specifying the
constraints on inputs that can be accepted by the layer.
Each layer has a dtype, which is typically the dtype of the layer's
computations and variables. A layer's dtype can be queried via the
`Layer.dtype` property. The dtype is specified with the `dtype` constructor
argument. In TensorFlow 2, the dtype defaults to `tf.keras.backend.floatx()`
if no dtype is passed. `floatx()` itself defaults to ""float32"". Additionally,
layers will cast their inputs to the layer's dtype in TensorFlow 2. When mixed
precision is used, layers may have different computation and variable dtypes.
See `tf.keras.mixed_precision.experimental.Policy` for details on layer
dtypes."
7439,ToDense,tensorflow/tensorflow/python/keras/utils/composite_tensor_support_test.py,52,class,Create a dense (standard) tensor from the given input tensor.
7440,call,tensorflow/tensorflow/python/keras/utils/composite_tensor_support_test.py,59,method,
7441,ToRagged,tensorflow/tensorflow/python/keras/utils/composite_tensor_support_test.py,78,class,Create a ragged tensor based on a given dense tensor.
7442,call,tensorflow/tensorflow/python/keras/utils/composite_tensor_support_test.py,86,method,
7443,ToSparse,tensorflow/tensorflow/python/keras/utils/composite_tensor_support_test.py,91,class,Create a sparse tensor based on a given dense tensor.
7444,call,tensorflow/tensorflow/python/keras/utils/composite_tensor_support_test.py,94,method,
7445,get_model_from_layers_with_input,tensorflow/tensorflow/python/keras/utils/composite_tensor_support_test.py,125,function,Builds a model from a sequence of layers.
7446,get_input_name,tensorflow/tensorflow/python/keras/utils/composite_tensor_support_test.py,289,function,
7447,get_kwargs,tensorflow/tensorflow/python/keras/utils/composite_tensor_support_test.py,299,function,
7448,prepare_inputs,tensorflow/tensorflow/python/keras/utils/composite_tensor_support_test.py,308,function,
7449,InXlaContext,tensorflow/tensorflow/python/keras/utils/control_flow_util.py,26,function,
7450,GraphOrParentsInXlaContext,tensorflow/tensorflow/python/keras/utils/control_flow_util.py,31,function,
7451,IsInWhileLoop,tensorflow/tensorflow/python/keras/utils/control_flow_util.py,40,function,
7452,GetContainingWhileContext,tensorflow/tensorflow/python/keras/utils/control_flow_util.py,45,function,"Returns the first ancestor WhileContext of `ctxt`.
Returns `ctxt` if `ctxt` is a WhileContext, or None if `ctxt` is not in a
while loop.
Args:
ctxt: ControlFlowContext
stop_ctxt: ControlFlowContext, optional. If provided, the search will end
if it sees stop_ctxt.
Returns:
`ctxt` if `ctxt` is a WhileContext, the most nested WhileContext containing
`ctxt`, or None if `ctxt` is not in a while loop. If `stop_ctxt` is not
`None`, this returns `ctxt` if it matches `stop_ctxt` in its traversal."
7453,GetContainingXLAContext,tensorflow/tensorflow/python/keras/utils/control_flow_util.py,67,function,"Returns the first ancestor XLAContext of `ctxt`.
Returns `ctxt` if `ctxt` is a XLAContext, or None if `ctxt` is not in a
while loop.
Args:
ctxt: ControlFlowContext
Returns:
`ctxt` if `ctxt` is a XLAContext, the most nested XLAContext containing
`ctxt`, or None if `ctxt` is not in a while loop."
7454,convert_data_format,tensorflow/tensorflow/python/keras/utils/conv_utils.py,28,function,
7455,normalize_tuple,tensorflow/tensorflow/python/keras/utils/conv_utils.py,51,function,"Transforms a single integer or iterable of integers into an integer tuple.
Arguments:
value: The value to validate and convert. Could an int, or any iterable of
ints.
n: The size of the tuple to be returned.
name: The name of the argument being validated, e.g. ""strides"" or
""kernel_size"". This is only used to format error messages.
Returns:
A tuple of n integers.
Raises:
ValueError: If something else than an int/long or iterable thereof was
passed."
7456,conv_output_length,tensorflow/tensorflow/python/keras/utils/conv_utils.py,90,function,"Determines output length of a convolution given input length.
Arguments:
input_length: integer.
filter_size: integer.
padding: one of ""same"", ""valid"", ""full"", ""causal""
stride: integer.
dilation: dilation rate, integer.
Returns:
The output length (integer)."
7457,conv_input_length,tensorflow/tensorflow/python/keras/utils/conv_utils.py,116,function,"Determines input length of a convolution given output length.
Arguments:
output_length: integer.
filter_size: integer.
padding: one of ""same"", ""valid"", ""full"".
stride: integer.
Returns:
The input length (integer)."
7458,deconv_output_length,tensorflow/tensorflow/python/keras/utils/conv_utils.py,140,function,"Determines output length of a transposed convolution given input length.
Arguments:
input_length: Integer.
filter_size: Integer.
padding: one of `""same""`, `""valid""`, `""full""`.
output_padding: Integer, amount of padding along the output dimension. Can
be set to `None` in which case the output length is inferred.
stride: Integer.
dilation: Integer.
Returns:
The output length (integer)."
7459,normalize_data_format,tensorflow/tensorflow/python/keras/utils/conv_utils.py,189,function,
7460,normalize_padding,tensorflow/tensorflow/python/keras/utils/conv_utils.py,200,function,
7461,convert_kernel,tensorflow/tensorflow/python/keras/utils/conv_utils.py,211,function,"Converts a Numpy kernel matrix from Theano format to TensorFlow format.
Also works reciprocally, since the transformation is its own inverse.
This is used for converting legacy Theano-saved model files.
Arguments:
kernel: Numpy array (3D, 4D or 5D).
Returns:
The converted kernel.
Raises:
ValueError: in case of invalid kernel shape or invalid data_format."
7462,conv_kernel_mask,tensorflow/tensorflow/python/keras/utils/conv_utils.py,236,function,"Compute a mask representing the connectivity of a convolution operation.
Assume a convolution with given parameters is applied to an input having N
spatial dimensions with `input_shape = (d_in1, ..., d_inN)` to produce an
output with shape `(d_out1, ..., d_outN)`. This method returns a boolean array
of shape `(d_in1, ..., d_inN, d_out1, ..., d_outN)` with `True` entries
indicating pairs of input and output locations that are connected by a weight.
Example:
>>> input_shape = (4,)
>>> kernel_shape = (2,)
>>> strides = (1,)
>>> padding = ""valid""
>>> conv_kernel_mask(input_shape, kernel_shape, strides, padding)
array([[ True, False, False],
[ True, True, False],
[False, True, True],
[False, False, True]])
where rows and columns correspond to inputs and outputs respectively.
Args:
input_shape: tuple of size N: `(d_in1, ..., d_inN)`, spatial shape of the
input.
kernel_shape: tuple of size N, spatial shape of the convolutional kernel /
receptive field.
strides: tuple of size N, strides along each spatial dimension.
padding: type of padding, string `""same""` or `""valid""`.
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
Returns:
A boolean 2N-D `np.ndarray` of shape
`(d_in1, ..., d_inN, d_out1, ..., d_outN)`, where `(d_out1, ..., d_outN)`
is the spatial shape of the output. `True` entries in the mask represent
pairs of input-output locations that are connected by a weight.
Raises:
ValueError: if `input_shape`, `kernel_shape` and `strides` don't have the
same number of dimensions.
NotImplementedError: if `padding` is not in {`""same""`, `""valid""`}."
7463,conv_kernel_idxs,tensorflow/tensorflow/python/keras/utils/conv_utils.py,315,function,"Yields output-input tuples of indices in a CNN layer.
The generator iterates over all `(output_idx, input_idx)` tuples, where
`output_idx` is an integer index in a flattened tensor representing a single
output image of a convolutional layer that is connected (via the layer
weights) to the respective single input image at `input_idx`
Example:
>>> input_shape = (2, 2)
>>> kernel_shape = (2, 1)
>>> strides = (1, 1)
>>> padding = ""valid""
>>> filters_in = 1
>>> filters_out = 1
>>> data_format = ""channels_last""
>>> list(conv_kernel_idxs(input_shape, kernel_shape, strides, padding,
... filters_in, filters_out, data_format))
[(0, 0), (0, 2), (1, 1), (1, 3)]
Args:
input_shape: tuple of size N: `(d_in1, ..., d_inN)`, spatial shape of the
input.
kernel_shape: tuple of size N, spatial shape of the convolutional kernel /
receptive field.
strides: tuple of size N, strides along each spatial dimension.
padding: type of padding, string `""same""` or `""valid""`.
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
filters_in: `int`, number if filters in the input to the layer.
filters_out: `int', number if filters in the output of the layer.
data_format: string, ""channels_first"" or ""channels_last"".
Yields:
The next tuple `(output_idx, input_idx)`, where
`output_idx` is an integer index in a flattened tensor representing a single
output image of a convolutional layer that is connected (via the layer
weights) to the respective single input image at `input_idx`.
Raises:
ValueError: if `data_format` is neither
`""channels_last""` nor `""channels_first""`, or if number of strides, input,
and kernel number of dimensions do not match.
NotImplementedError: if `padding` is neither `""same""` nor `""valid""`."
7464,conv_connected_inputs,tensorflow/tensorflow/python/keras/utils/conv_utils.py,409,function,"Return locations of the input connected to an output position.
Assume a convolution with given parameters is applied to an input having N
spatial dimensions with `input_shape = (d_in1, ..., d_inN)`. This method
returns N ranges specifying the input region that was convolved with the
kernel to produce the output at position
`output_position = (p_out1, ..., p_outN)`.
Example:
>>> input_shape = (4, 4)
>>> kernel_shape = (2, 1)
>>> output_position = (1, 1)
>>> strides = (1, 1)
>>> padding = ""valid""
>>> conv_connected_inputs(input_shape, kernel_shape, output_position,
... strides, padding)
[range(1, 3), range(1, 2)]
Args:
input_shape: tuple of size N: `(d_in1, ..., d_inN)`, spatial shape of the
input.
kernel_shape: tuple of size N, spatial shape of the convolutional kernel /
receptive field.
output_position: tuple of size N: `(p_out1, ..., p_outN)`, a single position
in the output of the convolution.
strides: tuple of size N, strides along each spatial dimension.
padding: type of padding, string `""same""` or `""valid""`.
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
Returns:
N ranges `[[p_in_left1, ..., p_in_right1], ...,
[p_in_leftN, ..., p_in_rightN]]` specifying the region in the
input connected to output_position."
7465,conv_output_shape,tensorflow/tensorflow/python/keras/utils/conv_utils.py,468,function,"Return the output shape of an N-D convolution.
Forces dimensions where input is empty (size 0) to remain empty.
Args:
input_shape: tuple of size N: `(d_in1, ..., d_inN)`, spatial shape of the
input.
kernel_shape: tuple of size N, spatial shape of the convolutional kernel /
receptive field.
strides: tuple of size N, strides along each spatial dimension.
padding: type of padding, string `""same""` or `""valid""`.
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
Returns:
tuple of size N: `(d_out1, ..., d_outN)`, spatial shape of the output."
7466,urlretrieve,tensorflow/tensorflow/python/keras/utils/data_utils.py,71,function,"Replacement for `urlretrieve` for Python 2.
Under Python 2, `urlretrieve` relies on `FancyURLopener` from legacy
`urllib` module, known to have issues with proxy management.
Arguments:
url: url to retrieve.
filename: where to store the retrieved data locally.
reporthook: a hook function that will be called once on establishment of
the network connection and once after each block read thereafter. The
hook will be passed three arguments; a count of blocks transferred so
far, a block size in bytes, and the total size of the file.
data: `data` argument passed to `urlopen`."
7467,is_generator_or_sequence,tensorflow/tensorflow/python/keras/utils/data_utils.py,111,function,Check if `x` is a Keras generator type.
7468,get_file,tensorflow/tensorflow/python/keras/utils/data_utils.py,169,function,"Downloads a file from a URL if it not already in the cache.
By default the file at the url `origin` is downloaded to the
cache_dir `~/.keras`, placed in the cache_subdir `datasets`,
and given the filename `fname`. The final location of a file
`example.txt` would therefore be `~/.keras/datasets/example.txt`.
Files in tar, tar.gz, tar.bz, and zip formats can also be extracted.
Passing a hash will verify the file after download. The command line
programs `shasum` and `sha256sum` can compute the hash.
Example:
```python
path_to_downloaded_file = tf.keras.utils.get_file(
""flower_photos"",
""https://storage.googleapis.com/download.tensorflow.org/example_images/flower_photos.tgz"",
untar=True)
```
Arguments:
fname: Name of the file. If an absolute path `/path/to/file.txt` is
specified the file will be saved at that location.
origin: Original URL of the file.
untar: Deprecated in favor of `extract` argument.
boolean, whether the file should be decompressed
md5_hash: Deprecated in favor of `file_hash` argument.
md5 hash of the file for verification
file_hash: The expected hash string of the file after download.
The sha256 and md5 hash algorithms are both supported.
cache_subdir: Subdirectory under the Keras cache dir where the file is
saved. If an absolute path `/path/to/folder` is
specified the file will be saved at that location.
hash_algorithm: Select the hash algorithm to verify the file.
options are `'md5'`, `'sha256'`, and `'auto'`.
The default 'auto' detects the hash algorithm in use.
extract: True tries extracting the file as an Archive, like tar or zip.
archive_format: Archive format to try for extracting the file.
Options are `'auto'`, `'tar'`, `'zip'`, and `None`.
`'tar'` includes tar, tar.gz, and tar.bz files.
The default `'auto'` corresponds to `['tar', 'zip']`.
None or an empty list will return no matches found.
cache_dir: Location to store cached files, when None it
defaults to the default directory `~/.keras/`.
Returns:
Path to the downloaded file"
7469,validate_file,tensorflow/tensorflow/python/keras/utils/data_utils.py,343,function,"Validates a file against a sha256 or md5 hash.
Arguments:
fpath: path to the file being validated
file_hash: The expected hash string of the file.
The sha256 and md5 hash algorithms are both supported.
algorithm: Hash algorithm, one of 'auto', 'sha256', or 'md5'.
The default 'auto' detects the hash algorithm in use.
chunk_size: Bytes to read at a time, important for large files.
Returns:
Whether the file is valid"
7470,ThreadsafeIter,tensorflow/tensorflow/python/keras/utils/data_utils.py,368,class,Wrap an iterator with a lock and propagate exceptions to all threads.
7471,next,tensorflow/tensorflow/python/keras/utils/data_utils.py,387,method,
7472,threadsafe_generator,tensorflow/tensorflow/python/keras/utils/data_utils.py,402,function,
7473,Sequence,tensorflow/tensorflow/python/keras/utils/data_utils.py,412,class,"Base object for fitting to a sequence of data, such as a dataset.
Every `Sequence` must implement the `__getitem__` and the `__len__` methods.
If you want to modify your dataset between epochs you may implement
`on_epoch_end`.
The method `__getitem__` should return a complete batch.
Notes:
`Sequence` are a safer way to do multiprocessing. This structure guarantees
that the network will only train once
on each sample per epoch which is not the case with generators.
Examples:
```python
from skimage.io import imread
from skimage.transform import resize
import numpy as np
import math
# Here, `x_set` is list of path to the images
# and `y_set` are the associated classes.
class CIFAR10Sequence(Sequence):
def __init__(self, x_set, y_set, batch_size):
self.x, self.y = x_set, y_set
self.batch_size = batch_size
def __len__(self):
return math.ceil(len(self.x) / self.batch_size)
def __getitem__(self, idx):
batch_x = self.x[idx * self.batch_size:(idx + 1) *
self.batch_size]
batch_y = self.y[idx * self.batch_size:(idx + 1) *
self.batch_size]
return np.array([
resize(imread(file_name), (200, 200))
for file_name in batch_x]), np.array(batch_y)
```"
7474,on_epoch_end,tensorflow/tensorflow/python/keras/utils/data_utils.py,479,method,"Method called at the end of every epoch.
"
7475,iter_sequence_infinite,tensorflow/tensorflow/python/keras/utils/data_utils.py,490,function,"Iterates indefinitely over a Sequence.
Arguments:
seq: `Sequence` instance.
Yields:
Batches of data from the `Sequence`."
7476,dont_use_multiprocessing_pool,tensorflow/tensorflow/python/keras/utils/data_utils.py,520,function,
7477,get_pool_class,tensorflow/tensorflow/python/keras/utils/data_utils.py,532,function,
7478,get_worker_id_queue,tensorflow/tensorflow/python/keras/utils/data_utils.py,543,function,Lazily create the queue to track worker ids.
7479,init_pool,tensorflow/tensorflow/python/keras/utils/data_utils.py,551,function,
7480,terminate_keras_multiprocessing_pools,tensorflow/tensorflow/python/keras/utils/data_utils.py,559,function,"Destroy Keras' multiprocessing pools to prevent deadlocks.
In general multiprocessing.Pool can interact quite badly with other, seemingly
unrelated, parts of a codebase due to Pool's reliance on fork. This method
cleans up all pools which are known to belong to Keras (and thus can be safely
terminated).
Args:
grace_period: Time (in seconds) to wait for process cleanup to propagate.
use_sigkill: Boolean of whether or not to perform a cleanup pass using
SIGKILL.
Returns:
A list of human readable strings describing all issues encountered. It is up
to the caller to decide whether to treat this as an error condition."
7481,get_index,tensorflow/tensorflow/python/keras/utils/data_utils.py,665,function,"Get the value from the Sequence `uid` at index `i`.
To allow multiple Sequences to be used at the same time, we use `uid` to
get a specific one. A single Sequence would cause the validation to
overwrite the training Sequence.
Arguments:
uid: int, Sequence identifier
i: index
Returns:
The value at index `i`."
7482,SequenceEnqueuer,tensorflow/tensorflow/python/keras/utils/data_utils.py,683,class,"Base class to enqueue inputs.
The task of an Enqueuer is to use parallelism to speed up preprocessing.
This is done with processes or threads.
Example:
```python
enqueuer = SequenceEnqueuer(...)
enqueuer.start()
datas = enqueuer.get()
for data in datas:
# Use the inputs; training, evaluating, predicting.
# ... stop sometime.
enqueuer.stop()
```
The `enqueuer.get()` should be an infinite stream of datas."
7483,is_running,tensorflow/tensorflow/python/keras/utils/data_utils.py,734,method,
7484,start,tensorflow/tensorflow/python/keras/utils/data_utils.py,737,method,"Starts the handler's workers.
Arguments:
workers: Number of workers.
max_queue_size: queue size
(when full, workers could block on `put()`)"
7485,stop,tensorflow/tensorflow/python/keras/utils/data_utils.py,762,method,"Stops running threads and wait for them to exit, if necessary.
Should be called by the same thread which called `start()`.
Arguments:
timeout: maximum time to wait on `thread.join()`"
7486,get,tensorflow/tensorflow/python/keras/utils/data_utils.py,800,method,"Creates a generator to extract data from the queue.
Skip the data if it is `None`.
# Returns
Generator yielding tuples `(inputs, targets)`
or `(inputs, targets, sample_weights)`."
7487,OrderedEnqueuer,tensorflow/tensorflow/python/keras/utils/data_utils.py,812,class,"Builds a Enqueuer from a Sequence.
Used in `fit_generator`, `evaluate_generator`, `predict_generator`.
Arguments:
sequence: A `tf.keras.utils.data_utils.Sequence` object.
use_multiprocessing: use multiprocessing if True, otherwise threading
shuffle: whether to shuffle the data at the beginning of each epoch"
7488,get,tensorflow/tensorflow/python/keras/utils/data_utils.py,879,method,"Creates a generator to extract data from the queue.
Skip the data if it is `None`.
Yields:
The next element in the queue, i.e. a tuple
`(inputs, targets)` or
`(inputs, targets, sample_weights)`."
7489,pool_fn,tensorflow/tensorflow/python/keras/utils/data_utils.py,836,method,
7490,init_pool_generator,tensorflow/tensorflow/python/keras/utils/data_utils.py,903,function,"Initializer function for pool workers.
Args:
gens: State which should be made available to worker processes.
random_seed: An optional value with which to seed child processes.
id_queue: A multiprocessing Queue of worker ids. This is used to indicate
that a worker process was created by Keras and can be terminated using
the cleanup_all_keras_forkpools utility."
7491,next_sample,tensorflow/tensorflow/python/keras/utils/data_utils.py,930,function,"Gets the next value from the generator `uid`.
To allow multiple generators to be used at the same time, we use `uid` to
get a specific one. A single generator would cause the validation to
overwrite the training generator.
Arguments:
uid: int, generator identifier
Returns:
The next value of generator `uid`."
7492,GeneratorEnqueuer,tensorflow/tensorflow/python/keras/utils/data_utils.py,947,class,"Builds a queue out of a data generator.
The provided generator can be finite in which case the class will throw
a `StopIteration` exception.
Used in `fit_generator`, `evaluate_generator`, `predict_generator`.
Arguments:
generator: a generator function which yields data
use_multiprocessing: use multiprocessing if True, otherwise threading
wait_time: time to sleep in-between calls to `put()`
random_seed: Initial seed for workers,
will be incremented by one for each worker."
7493,get,tensorflow/tensorflow/python/keras/utils/data_utils.py,997,method,"Creates a generator to extract data from the queue.
Skip the data if it is `None`.
Yields:
The next element in the queue, i.e. a tuple
`(inputs, targets)` or
`(inputs, targets, sample_weights)`."
7494,pool_fn,tensorflow/tensorflow/python/keras/utils/data_utils.py,978,method,
7495,FaultSequence,tensorflow/tensorflow/python/keras/utils/data_utils_test.py,107,class,
7496,create_generator_from_sequence_threads,tensorflow/tensorflow/python/keras/utils/data_utils_test.py,117,function,
7497,create_generator_from_sequence_pcs,tensorflow/tensorflow/python/keras/utils/data_utils_test.py,122,function,
7498,CustomObjectScope,tensorflow/tensorflow/python/keras/utils/generic_utils.py,53,class,"Exposes custom classes/functions to Keras deserialization internals.
Under a scope `with custom_object_scope(objects_dict)`, Keras methods such
as `tf.keras.models.load_model` or `tf.keras.models.model_from_config`
will be able to deserialize any custom object referenced by a
saved config (e.g. a custom layer or metric).
Example:
Consider a custom regularizer `my_regularizer`:
```python
layer = Dense(3, kernel_regularizer=my_regularizer)
config = layer.get_config() # Config contains a reference to `my_regularizer`
...
# Later:
with custom_object_scope({'my_regularizer': my_regularizer}):
layer = Dense.from_config(config)
```
Arguments:
*args: Dictionary or dictionaries of `{name: object}` pairs."
7499,get_custom_objects,tensorflow/tensorflow/python/keras/utils/generic_utils.py,94,function,"Retrieves a live reference to the global dictionary of custom objects.
Updating and clearing custom objects using `custom_object_scope`
is preferred, but `get_custom_objects` can
be used to directly access the current collection of custom objects.
Example:
```python
get_custom_objects().clear()
get_custom_objects()['MyObject'] = MyObject
```
Returns:
Global dictionary of names to classes (`_GLOBAL_CUSTOM_OBJECTS`)."
7500,serialize_keras_class_and_config,tensorflow/tensorflow/python/keras/utils/generic_utils.py,114,function,Returns the serialization of the class with the given config.
7501,register_keras_serializable,tensorflow/tensorflow/python/keras/utils/generic_utils.py,120,function,"Registers an object with the Keras serialization framework.
This decorator injects the decorated class or function into the Keras custom
object dictionary, so that it can be serialized and deserialized without
needing an entry in the user-provided custom object dict. It also injects a
function that Keras will call to get the object's serializable string key.
Note that to be serialized and deserialized, classes must implement the
`get_config()` method. Functions do not have this requirement.
The object will be registered under the key 'package>name' where `name`,
defaults to the object name if not passed.
Arguments:
package: The package that this class belongs to.
name: The name to serialize this class under in this package. If None, the
class' name will be used.
Returns:
A decorator that registers the decorated class with the passed names."
7502,get_registered_name,tensorflow/tensorflow/python/keras/utils/generic_utils.py,169,function,"Returns the name registered to an object within the Keras framework.
This function is part of the Keras serialization and deserialization
framework. It maps objects to the string names associated with those objects
for serialization/deserialization.
Args:
obj: The object to look up.
Returns:
The name associated with the object, or the default Python name if the
object is not registered."
7503,skip_failed_serialization,tensorflow/tensorflow/python/keras/utils/generic_utils.py,190,function,
7504,get_registered_object,tensorflow/tensorflow/python/keras/utils/generic_utils.py,201,function,"Returns the class associated with `name` if it is registered with Keras.
This function is part of the Keras serialization and deserialization
framework. It maps strings to the objects associated with them for
serialization/deserialization.
Example:
```
def from_config(cls, config, custom_objects=None):
if 'my_custom_object_name' in config:
config['hidden_cls'] = tf.keras.utils.get_registered_object(
config['my_custom_object_name'], custom_objects=custom_objects)
```
Args:
name: The name to look up.
custom_objects: A dictionary of custom objects to look the name up in.
Generally, custom_objects is provided by the user.
module_objects: A dictionary of custom objects to look the name up in.
Generally, module_objects is provided by midlevel library implementers.
Returns:
An instantiable class associated with 'name', or None if no such class
exists."
7505,serialize_keras_object,tensorflow/tensorflow/python/keras/utils/generic_utils.py,237,function,Serialize a Keras object into a JSON-compatible representation.
7506,get_custom_objects_by_name,tensorflow/tensorflow/python/keras/utils/generic_utils.py,275,function,Returns the item if it is in either local or global custom objects.
7507,class_and_config_for_serialized_keras_object,tensorflow/tensorflow/python/keras/utils/generic_utils.py,284,function,Returns the class name and config for a serialized keras object.
7508,deserialize_keras_object,tensorflow/tensorflow/python/keras/utils/generic_utils.py,336,function,Turns the serialized form of a Keras object back into an actual object.
7509,func_dump,tensorflow/tensorflow/python/keras/utils/generic_utils.py,393,function,"Serializes a user defined function.
Arguments:
func: the function to serialize.
Returns:
A tuple `(code, defaults, closure)`."
7510,func_load,tensorflow/tensorflow/python/keras/utils/generic_utils.py,416,function,"Deserializes a user defined function.
Arguments:
code: bytecode of the function.
defaults: defaults of the function.
closure: closure of the function.
globs: dictionary of global objects.
Returns:
A function object."
7511,has_arg,tensorflow/tensorflow/python/keras/utils/generic_utils.py,465,function,"Checks if a callable accepts a given keyword argument.
Arguments:
fn: Callable to inspect.
name: Check if `fn` can be called with `name` as a keyword argument.
accept_all: What to return if there is no parameter called `name` but the
function accepts a `**kwargs` argument.
Returns:
bool, whether `fn` accepts a `name` keyword argument."
7512,Progbar,tensorflow/tensorflow/python/keras/utils/generic_utils.py,484,class,"Displays a progress bar.
Arguments:
target: Total number of steps expected, None if unknown.
width: Progress bar width on screen.
verbose: Verbosity mode, 0 (silent), 1 (verbose), 2 (semi-verbose)
stateful_metrics: Iterable of string names of metrics that should *not* be
averaged over time. Metrics in this list will be displayed as-is. All
others will be averaged by the progbar before display.
interval: Minimum visual progress update interval (in seconds).
unit_name: Display name for step counts (usually ""step"" or ""sample"")."
7513,update,tensorflow/tensorflow/python/keras/utils/generic_utils.py,529,method,"Updates the progress bar.
Arguments:
current: Index of current step.
values: List of tuples: `(name, value_for_last_step)`. If `name` is in
`stateful_metrics`, `value_for_last_step` will be displayed as-is.
Else, an average of the metric over time will be displayed.
finalize: Whether this is the last update for the progress bar. If
`None`, defaults to `current >= self.target`."
7514,add,tensorflow/tensorflow/python/keras/utils/generic_utils.py,664,method,
7515,make_batches,tensorflow/tensorflow/python/keras/utils/generic_utils.py,668,function,"Returns a list of batch indices (tuples of indices).
Arguments:
size: Integer, total size of the data to slice into batches.
batch_size: Integer, batch size.
Returns:
A list of tuples of array indices."
7516,slice_arrays,tensorflow/tensorflow/python/keras/utils/generic_utils.py,683,function,"Slice an array or list of arrays.
This takes an array-like, or a list of
array-likes, and outputs:
- arrays[start:stop] if `arrays` is an array-like
- [x[start:stop] for x in arrays] if `arrays` is a list
Can also work on list/array of indices: `slice_arrays(x, indices)`
Arguments:
arrays: Single array or list of arrays.
start: can be an integer index (start index) or a list/array of indices
stop: integer (stop index); should be None if `start` was a list.
Returns:
A slice of the array(s).
Raises:
ValueError: If the value of start is a list and stop is not None."
7517,to_list,tensorflow/tensorflow/python/keras/utils/generic_utils.py,729,function,"Normalizes a list/tensor into a list.
If a tensor is passed, we return
a list of size 1 containing the tensor.
Arguments:
x: target object to be normalized.
Returns:
A list."
7518,to_snake_case,tensorflow/tensorflow/python/keras/utils/generic_utils.py,746,function,
7519,is_all_none,tensorflow/tensorflow/python/keras/utils/generic_utils.py,756,function,
7520,check_for_unexpected_keys,tensorflow/tensorflow/python/keras/utils/generic_utils.py,765,function,
7521,validate_kwargs,tensorflow/tensorflow/python/keras/utils/generic_utils.py,773,function,Checks that all keyword arguments are in the set of allowed keys.
7522,validate_config,tensorflow/tensorflow/python/keras/utils/generic_utils.py,782,function,Determines whether config appears to be a valid layer config.
7523,default,tensorflow/tensorflow/python/keras/utils/generic_utils.py,787,function,Decorates a method to detect overrides in subclasses.
7524,is_default,tensorflow/tensorflow/python/keras/utils/generic_utils.py,793,function,Check if a method is decorated with the `default` wrapper.
7525,populate_dict_with_module_objects,tensorflow/tensorflow/python/keras/utils/generic_utils.py,798,function,
7526,LazyLoader,tensorflow/tensorflow/python/keras/utils/generic_utils.py,806,class,"Lazily import a module, mainly to avoid pulling in large dependencies."
7527,path_to_string,tensorflow/tensorflow/python/keras/utils/io_utils.py,57,function,"Convert `PathLike` objects to their string representation.
If given a non-string typed path object, converts it to its string
representation. Depending on the python version used, this function
can handle the following arguments:
python >= 3.6: Everything supporting the fs path protocol
https://www.python.org/dev/peps/pep-0519
python >= 3.4: Only `pathlib.Path` objects
If the object passed to `path` is not among the above, then it is
returned unchanged. This allows e.g. passthrough of file objects
through this function.
Args:
path: `PathLike` object that represents a path
Returns:
A string representation of the path argument, if Python support exists."
7528,HDF5Matrix,tensorflow/tensorflow/python/keras/utils/io_utils.py,81,class,"Representation of HDF5 dataset to be used instead of a Numpy array.
THIS CLASS IS DEPRECATED.
Training with HDF5Matrix may not be optimized for performance, and might
not work with every distribution strategy.
We recommend using https://github.com/tensorflow/io to load your
HDF5 data into a tf.data Dataset and passing that dataset to Keras."
7529,shape,tensorflow/tensorflow/python/keras/utils/io_utils.py,181,method,"Gets a numpy-style shape tuple giving the dataset dimensions.
Returns:
A numpy-style shape tuple."
7530,dtype,tensorflow/tensorflow/python/keras/utils/io_utils.py,190,method,"Gets the datatype of the dataset.
Returns:
A numpy dtype string."
7531,ndim,tensorflow/tensorflow/python/keras/utils/io_utils.py,199,method,"Gets the number of dimensions (rank) of the dataset.
Returns:
An integer denoting the number of dimensions (rank) of the dataset."
7532,size,tensorflow/tensorflow/python/keras/utils/io_utils.py,208,method,"Gets the total dataset size (number of elements).
Returns:
An integer denoting the number of elements in the dataset."
7533,ask_to_proceed_with_overwrite,tensorflow/tensorflow/python/keras/utils/io_utils.py,236,function,"Produces a prompt asking about overwriting a file.
Arguments:
filepath: the path to the file to be overwritten.
Returns:
True if we can proceed with overwrite, False otherwise."
7534,create_dataset,tensorflow/tensorflow/python/keras/utils/io_utils_test.py,40,function,
7535,inner_product,tensorflow/tensorflow/python/keras/utils/kernelized_utils.py,54,function,
7536,exact_gaussian_kernel,tensorflow/tensorflow/python/keras/utils/kernelized_utils.py,60,function,"Computes exact Gaussian kernel value(s) for tensors x and y and stddev.
The Gaussian kernel for vectors u, v is defined as follows:
K(u, v) = exp(-||u-v||^2 / (2* stddev^2))
where the norm is the l2-norm. x, y can be either vectors or matrices. If they
are vectors, they must have the same dimension. If they are matrices, they
must have the same number of columns. In the latter case, the method returns
(as a matrix) K(u, v) values for all pairs (u, v) where u is a row from x and
v is a row from y.
Args:
x: a tensor of rank 1 or 2. It's shape should be either [dim] or [m, dim].
y: a tensor of rank 1 or 2. It's shape should be either [dim] or [n, dim].
stddev: The width of the Gaussian kernel.
Returns:
A single value (scalar) with shape (1, 1) (if x, y are vectors) or a matrix
of shape (m, n) with entries K(u, v) (where K is the Gaussian kernel) for
all (u,v) pairs where u, v are rows from x and y respectively.
Raises:
ValueError: if the shapes of x, y are not compatible."
7537,exact_laplacian_kernel,tensorflow/tensorflow/python/keras/utils/kernelized_utils.py,90,function,"Computes exact Laplacian kernel value(s) for tensors x and y using stddev.
The Laplacian kernel for vectors u, v is defined as follows:
K(u, v) = exp(-||u-v|| / stddev)
where the norm is the l1-norm. x, y can be either vectors or matrices. If they
are vectors, they must have the same dimension. If they are matrices, they
must have the same number of columns. In the latter case, the method returns
(as a matrix) K(u, v) values for all pairs (u, v) where u is a row from x and
v is a row from y.
Args:
x: a tensor of rank 1 or 2. It's shape should be either [dim] or [m, dim].
y: a tensor of rank 1 or 2. It's shape should be either [dim] or [n, dim].
stddev: The width of the Gaussian kernel.
Returns:
A single value (scalar) with shape (1, 1) if x, y are vectors or a matrix
of shape (m, n) with entries K(u, v) (where K is the Laplacian kernel) for
all (u,v) pairs where u, v are rows from x and y respectively.
Raises:
ValueError: if the shapes of x, y are not compatible."
7538,get_source_inputs,tensorflow/tensorflow/python/keras/utils/layer_utils.py,34,function,"Returns the list of input tensors necessary to compute `tensor`.
Output will always be a list of tensors
(potentially with 1 element).
Arguments:
tensor: The tensor to start from.
layer: Origin layer of the tensor. Will be
determined via tensor._keras_history if not provided.
node_index: Origin node index of the tensor.
Returns:
List of input tensors."
7539,validate_string_arg,tensorflow/tensorflow/python/keras/utils/layer_utils.py,72,function,Validates the correctness of a string-based arg.
7540,count_params,tensorflow/tensorflow/python/keras/utils/layer_utils.py,95,function,"Count the total number of scalars composing the weights.
Arguments:
weights: An iterable containing the weights on which to compute params
Returns:
The total number of scalars composing the weights"
7541,print_summary,tensorflow/tensorflow/python/keras/utils/layer_utils.py,112,function,"Prints a summary of a model.
Arguments:
model: Keras model instance.
line_length: Total length of printed lines
(e.g. set this to adapt the display to different
terminal window sizes).
positions: Relative or absolute positions of log elements in each line.
If not provided, defaults to `[.33, .55, .67, 1.]`.
print_fn: Print function to use.
It will be called on each line of the summary.
You can set it to a custom function
in order to capture the string summary.
It defaults to `print` (prints to stdout)."
7542,gather_trainable_weights,tensorflow/tensorflow/python/keras/utils/layer_utils.py,274,function,"Lists the trainable weights for an object with sub-layers.
Args:
trainable: Whether the object collecting the variables is trainable.
sub_layers: A flat list of Layer objects owned by this object, to collect
variables from.
extra_variables: Any extra variables to include. Their `.trainable` property
is used to categorize them.
Returns:
A list of collected trainable weights/variables."
7543,gather_non_trainable_weights,tensorflow/tensorflow/python/keras/utils/layer_utils.py,297,function,"Lists the non-trainable weights for an object with sub-layers.
Args:
trainable: Whether the object collecting the variables is trainable.
sub_layers: A flat list of Layer objects owned by this object, to collect
variables from.
extra_variables: Any extra variables to include. Their `.trainable` property
is used to categorize them.
Returns:
A list of collected non-trainable weights/variables."
7544,convert_all_kernels_in_model,tensorflow/tensorflow/python/keras/utils/layer_utils.py,333,function,"Converts all convolution kernels in a model from Theano to TensorFlow.
Also works from TensorFlow to Theano.
This is used for converting legacy Theano-saved model files.
Arguments:
model: target model for the conversion."
7545,convert_dense_weights_data_format,tensorflow/tensorflow/python/keras/utils/layer_utils.py,360,function,"Utility useful when changing a convnet's `data_format`.
When porting the weights of a convnet from one data format to the other,
if the convnet includes a `Flatten` layer
(applied to the last convolutional feature map)
followed by a `Dense` layer, the weights of that `Dense` layer
should be updated to reflect the new dimension ordering.
Arguments:
dense: The target `Dense` layer.
previous_feature_map_shape: A shape tuple of 3 integers,
e.g. `(512, 7, 7)`. The shape of the convolutional
feature map right before the `Flatten` layer that
came before the target `Dense` layer.
target_data_format: One of ""channels_last"", ""channels_first"".
Set it ""channels_last""
if converting a ""channels_first"" model to ""channels_last"",
or reciprocally."
7546,is_builtin_layer,tensorflow/tensorflow/python/keras/utils/layer_utils.py,399,function,
7547,remove_squeezable_dimensions,tensorflow/tensorflow/python/keras/utils/losses_utils.py,38,function,"Squeeze last dim if ranks differ from expected by exactly 1.
In the common case where we expect shapes to match, `expected_rank_diff`
defaults to 0, and we squeeze the last dimension of the larger rank if they
differ by 1.
But, for example, if `labels` contains class IDs and `predictions` contains 1
probability per class, we expect `predictions` to have 1 more dimension than
`labels`, so `expected_rank_diff` would be 1. In this case, we'd squeeze
`labels` if `rank(predictions) - rank(labels) == 0`, and
`predictions` if `rank(predictions) - rank(labels) == 2`.
This will use static shape if available. Otherwise, it will add graph
operations, which could result in a performance hit.
Args:
labels: Label values, a `Tensor` whose dimensions match `predictions`.
predictions: Predicted values, a `Tensor` of arbitrary dimensions.
expected_rank_diff: Expected result of `rank(predictions) - rank(labels)`.
name: Name of the op.
Returns:
Tuple of `labels` and `predictions`, possibly with last dim squeezed."
7548,squeeze_or_expand_dimensions,tensorflow/tensorflow/python/keras/utils/losses_utils.py,99,function,"Squeeze or expand last dimension if needed.
1. Squeezes last dim of `y_pred` or `y_true` if their rank differs by 1
(using `remove_squeezable_dimensions`).
2. Squeezes or expands last dim of `sample_weight` if its rank differs by 1
from the new rank of `y_pred`.
If `sample_weight` is scalar, it is kept scalar.
This will use static shape if available. Otherwise, it will add graph
operations, which could result in a performance hit.
Args:
y_pred: Predicted values, a `Tensor` of arbitrary dimensions.
y_true: Optional label `Tensor` whose dimensions match `y_pred`.
sample_weight: Optional weight scalar or `Tensor` whose dimensions match
`y_pred`.
Returns:
Tuple of `y_pred`, `y_true` and `sample_weight`. Each of them possibly has
the last dimension squeezed,
`sample_weight` could be extended by one dimension.
If `sample_weight` is None, (y_pred, y_true) is returned."
7549,reduce_weighted_loss,tensorflow/tensorflow/python/keras/utils/losses_utils.py,209,function,Reduces the individual weighted loss measurements.
7550,compute_weighted_loss,tensorflow/tensorflow/python/keras/utils/losses_utils.py,221,function,"Computes the weighted loss.
Args:
losses: `Tensor` of shape `[batch_size, d1, ... dN]`.
sample_weight: Optional `Tensor` whose rank is either 0, or the same rank as
`losses`, or be broadcastable to `losses`.
reduction: (Optional) Type of `tf.keras.losses.Reduction` to apply to loss.
Default value is `SUM_OVER_BATCH_SIZE`.
name: Optional name for the op.
Raises:
ValueError: If the shape of `sample_weight` is not compatible with `losses`.
Returns:
Weighted loss `Tensor` of the same type as `losses`. If `reduction` is
`NONE`, this has the same shape as `losses`; otherwise, it is scalar."
7551,scale_loss_for_distribution,tensorflow/tensorflow/python/keras/utils/losses_utils.py,278,function,Scales and returns the given loss value by the number of replicas.
7552,cast_losses_to_common_dtype,tensorflow/tensorflow/python/keras/utils/losses_utils.py,287,function,"Cast a list of losses to a common dtype.
If any loss is floating-point, they will all be casted to the most-precise
floating-point loss. Otherwise the losses are not casted. We also skip casting
losses if there are any complex losses.
Args:
losses: A list of losses.
Returns:
`losses`, but they have been casted to a common dtype."
7553,Reduction,tensorflow/tensorflow/python/keras/utils/metrics_utils.py,48,class,"Types of metrics reduction.
Contains the following values:
* `SUM`: Scalar sum of weighted values.
* `SUM_OVER_BATCH_SIZE`: Scalar sum of weighted values divided by
number of elements.
* `WEIGHTED_MEAN`: Scalar sum of weighted values divided by sum of weights."
7554,update_state_wrapper,tensorflow/tensorflow/python/keras/utils/metrics_utils.py,63,function,"Decorator to wrap metric `update_state()` with `add_update()`.
Args:
update_state_fn: function that accumulates metric statistics.
Returns:
Decorated function that wraps `update_state_fn()` with `add_update()`."
7555,result_wrapper,tensorflow/tensorflow/python/keras/utils/metrics_utils.py,98,function,"Decorator to wrap metric `result()` function in `merge_call()`.
Result computation is an idempotent operation that simply calculates the
metric value using the state variables.
If metric state variables are distributed across replicas/devices and
`result()` is requested from the context of one device - This function wraps
`result()` in a distribution strategy `merge_call()`. With this,
the metric state variables will be aggregated across devices.
Args:
result_fn: function that computes the metric result.
Returns:
Decorated function that wraps `result_fn()` in distribution strategy
`merge_call()`."
7556,weakmethod,tensorflow/tensorflow/python/keras/utils/metrics_utils.py,154,function,Creates a weak reference to the bound method.
7557,assert_thresholds_range,tensorflow/tensorflow/python/keras/utils/metrics_utils.py,169,function,
7558,parse_init_thresholds,tensorflow/tensorflow/python/keras/utils/metrics_utils.py,178,function,
7559,ConfusionMatrix,tensorflow/tensorflow/python/keras/utils/metrics_utils.py,185,class,
7560,AUCCurve,tensorflow/tensorflow/python/keras/utils/metrics_utils.py,192,class,Type of AUC Curve (ROC or PR).
7561,from_str,tensorflow/tensorflow/python/keras/utils/metrics_utils.py,198,method,
7562,AUCSummationMethod,tensorflow/tensorflow/python/keras/utils/metrics_utils.py,207,class,"Type of AUC summation method.
https://en.wikipedia.org/wiki/Riemann_sum)
Contains the following values:
* 'interpolation': Applies mid-point summation scheme for `ROC` curve. For
`PR` curve, interpolates (true/false) positives but not the ratio that is
precision (see Davis & Goadrich 2006 for details).
* 'minoring': Applies left summation for increasing intervals and right
summation for decreasing intervals.
* 'majoring': Applies right summation for increasing intervals and left
summation for decreasing intervals."
7563,from_str,tensorflow/tensorflow/python/keras/utils/metrics_utils.py,226,method,
7564,update_confusion_matrix_variables,tensorflow/tensorflow/python/keras/utils/metrics_utils.py,237,function,"Returns op to update the given confusion matrix variables.
For every pair of values in y_true and y_pred:
true_positive: y_true == True and y_pred > thresholds
false_negatives: y_true == True and y_pred <= thresholds
true_negatives: y_true == False and y_pred <= thresholds
false_positive: y_true == False and y_pred > thresholds
The results will be weighted and added together. When multiple thresholds are
provided, we will repeat the same for every threshold.
For estimation of these metrics over a stream of data, the function creates an
`update_op` operation that updates the given variables.
If `sample_weight` is `None`, weights default to 1.
Use weights of 0 to mask values.
Args:
variables_to_update: Dictionary with 'tp', 'fn', 'tn', 'fp' as valid keys
and corresponding variables to update as values.
y_true: A `Tensor` whose shape matches `y_pred`. Will be cast to `bool`.
y_pred: A floating point `Tensor` of arbitrary shape and whose values are in
the range `[0, 1]`.
thresholds: A float value, float tensor, python list, or tuple of float
thresholds in `[0, 1]`, or NEG_INF (used when top_k is set).
top_k: Optional int, indicates that the positive labels should be limited to
the top k predictions.
class_id: Optional int, limits the prediction and labels to the class
specified by this argument.
sample_weight: Optional `Tensor` whose rank is either 0, or the same rank as
`y_true`, and must be broadcastable to `y_true` (i.e., all dimensions must
be either `1`, or the same as the corresponding `y_true` dimension).
multi_label: Optional boolean indicating whether multidimensional
prediction/labels should be treated as multilabel responses, or flattened
into a single label. When True, the valus of `variables_to_update` must
have a second dimension equal to the number of labels in y_true and
y_pred, and those tensors must not be RaggedTensors.
label_weights: (optional) tensor of non-negative weights for multilabel
data. The weights are applied when calculating TP, FP, FN, and TN without
explicit multilabel handling (i.e. when the data is to be flattened).
Returns:
Update op.
Raises:
ValueError: If `y_pred` and `y_true` have mismatched shapes, or if
`sample_weight` is not `None` and its shape doesn't match `y_pred`, or if
`variables_to_update` contains invalid keys."
7565,ragged_assert_compatible_and_get_flat_values,tensorflow/tensorflow/python/keras/utils/metrics_utils.py,479,function,"If ragged, it checks the compatibility and then returns the flat_values.
Note: If two tensors are dense, it does not check their compatibility.
Note: Although two ragged tensors with different ragged ranks could have
identical overall rank and dimension sizes and hence be compatible,
we do not support those cases.
Args:
values: A list of potentially ragged tensor of the same ragged_rank.
mask: A potentially ragged tensor of the same ragged_rank as elements in
Values.
Returns:
A tuple in which the first element is the list of tensors and the second
is the mask tensor. ([Values], mask). Mask and the element in Values
are equal to the flat_values of the input arguments (if they were ragged)."
7566,multi_gpu_model,tensorflow/tensorflow/python/keras/utils/multi_gpu_utils.py,42,function,"Replicates a model on different GPUs.
Specifically, this function implements single-machine
multi-GPU data parallelism. It works in the following way:
- Divide the model's input(s) into multiple sub-batches.
- Apply a model copy on each sub-batch. Every model copy
is executed on a dedicated GPU.
- Concatenate the results (on CPU) into one big batch.
E.g. if your `batch_size` is 64 and you use `gpus=2`,
then we will divide the input into 2 sub-batches of 32 samples,
process each sub-batch on one GPU, then return the full
batch of 64 processed samples.
This induces quasi-linear speedup on up to 8 GPUs.
This function is only available with the TensorFlow backend
for the time being.
Arguments:
model: A Keras model instance. To avoid OOM errors,
this model could have been built on CPU, for instance
(see usage example below).
gpus: Integer >= 2, number of on GPUs on which to create
model replicas.
cpu_merge: A boolean value to identify whether to force
merging model weights under the scope of the CPU or not.
cpu_relocation: A boolean value to identify whether to
create the model's weights under the scope of the CPU.
If the model is not defined under any preceding device
scope, you can still rescue it by activating this option.
Returns:
A Keras `Model` instance which can be used just like the initial
`model` argument, but which distributes its workload on multiple GPUs.
Example 1: Training models with weights merge on CPU
```python
import tensorflow as tf
from keras.applications import Xception
from keras.utils import multi_gpu_model
import numpy as np
num_samples = 1000
height = 224
width = 224
num_classes = 1000
# Instantiate the base model (or ""template"" model).
# We recommend doing this with under a CPU device scope,
# so that the model's weights are hosted on CPU memory.
# Otherwise they may end up hosted on a GPU, which would
# complicate weight sharing.
with tf.device('/cpu:0'):
model = Xception(weights=None,
input_shape=(height, width, 3),
classes=num_classes)
# Replicates the model on 8 GPUs.
# This assumes that your machine has 8 available GPUs.
parallel_model = multi_gpu_model(model, gpus=8)
parallel_model.compile(loss='categorical_crossentropy',
optimizer='rmsprop')
# Generate dummy data.
x = np.random.random((num_samples, height, width, 3))
y = np.random.random((num_samples, num_classes))
# This `fit` call will be distributed on 8 GPUs.
# Since the batch size is 256, each GPU will process 32 samples.
parallel_model.fit(x, y, epochs=20, batch_size=256)
# Save model via the template model (which shares the same weights):
model.save('my_model.h5')
```
Example 2: Training models with weights merge on CPU using cpu_relocation
```python
..
# Not needed to change the device scope for model definition:
model = Xception(weights=None, ..)
try:
model = multi_gpu_model(model, cpu_relocation=True)
print(""Training using multiple GPUs.."")
except:
print(""Training using single GPU or CPU.."")
model.compile(..)
..
```
Example 3: Training models with weights merge on GPU (recommended for NV-link)
```python
..
# Not needed to change the device scope for model definition:
model = Xception(weights=None, ..)
try:
model = multi_gpu_model(model, cpu_merge=False)
print(""Training using multiple GPUs.."")
except:
print(""Training using single GPU or CPU.."")
model.compile(..)
..
```
Raises:
ValueError: if the `gpus` argument does not match available devices."
7567,check_if_compatible_devices,tensorflow/tensorflow/python/keras/utils/multi_gpu_utils_test.py,32,function,
7568,to_categorical,tensorflow/tensorflow/python/keras/utils/np_utils.py,25,function,"Converts a class vector (integers) to binary class matrix.
E.g. for use with categorical_crossentropy.
Arguments:
y: class vector to be converted into a matrix
(integers from 0 to num_classes).
num_classes: total number of classes. If `None`, this would be inferred
as the (largest number in `y`) + 1.
dtype: The data type expected by the input. Default: `'float32'`.
Returns:
A binary matrix representation of the input. The classes axis is placed
last.
Example:
>>> a = tf.keras.utils.to_categorical([0, 1, 2, 3], num_classes=4)
>>> a = tf.constant(a, shape=[4, 4])
>>> print(a)
tf.Tensor(
[[1. 0. 0. 0.]
[0. 1. 0. 0.]
[0. 0. 1. 0.]
[0. 0. 0. 1.]], shape=(4, 4), dtype=float32)
>>> b = tf.constant([.9, .04, .03, .03,
... .3, .45, .15, .13,
... .04, .01, .94, .05,
... .12, .21, .5, .17],
... shape=[4, 4])
>>> loss = tf.keras.backend.categorical_crossentropy(a, b)
>>> print(np.around(loss, 5))
[0.10536 0.82807 0.1011 1.77196]
>>> loss = tf.keras.backend.categorical_crossentropy(a, a)
>>> print(np.around(loss, 5))
[0. 0. 0. 0.]
Raises:
Value Error: If input contains string value"
7569,normalize,tensorflow/tensorflow/python/keras/utils/np_utils.py,85,function,"Normalizes a Numpy array.
Arguments:
x: Numpy array to normalize.
axis: axis along which to normalize.
order: Normalization order (e.g. `order=2` for L2 norm).
Returns:
A normalized copy of the array."
7570,smart_cond,tensorflow/tensorflow/python/keras/utils/tf_utils.py,44,function,"Return either `true_fn()` if predicate `pred` is true else `false_fn()`.
If `pred` is a bool or has a constant value, we return either `true_fn()`
or `false_fn()`, otherwise we use `tf.cond` to dynamically route to both.
Arguments:
pred: A scalar determining whether to return the result of `true_fn` or
`false_fn`.
true_fn: The callable to be performed if pred is true.
false_fn: The callable to be performed if pred is false.
name: Optional name prefix when using `tf.cond`.
Returns:
Tensors returned by the call to either `true_fn` or `false_fn`.
Raises:
TypeError: If `true_fn` or `false_fn` is not callable."
7571,constant_value,tensorflow/tensorflow/python/keras/utils/tf_utils.py,70,function,"Return the bool value for `pred`, or None if `pred` had a dynamic value.
Arguments:
pred: A scalar, either a Python bool or a TensorFlow boolean variable
or tensor, or the Python integer 1 or 0.
Returns:
True or False if `pred` has a constant boolean value, None otherwise.
Raises:
TypeError: If `pred` is not a Variable, Tensor or bool, or Python
integer 1 or 0."
7572,is_tensor_or_tensor_list,tensorflow/tensorflow/python/keras/utils/tf_utils.py,96,function,
7573,get_reachable_from_inputs,tensorflow/tensorflow/python/keras/utils/tf_utils.py,104,function,"Returns the set of tensors/ops reachable from `inputs`.
Stops if all targets have been found (target is optional).
Only valid in Symbolic mode, not Eager mode.
Args:
inputs: List of tensors.
targets: List of tensors.
Returns:
A set of tensors reachable from the inputs (includes the inputs themselves)."
7574,map_structure_with_atomic,tensorflow/tensorflow/python/keras/utils/tf_utils.py,159,function,"Maps the atomic elements of a nested structure.
Arguments:
is_atomic_fn: A function that determines if an element of `nested` is
atomic.
map_fn: The function to apply to atomic elements of `nested`.
nested: A nested structure.
Returns:
The nested structure, with atomic elements mapped according to `map_fn`.
Raises:
ValueError: If an element that is neither atomic nor a sequence is
encountered."
7575,get_shapes,tensorflow/tensorflow/python/keras/utils/tf_utils.py,194,function,Gets shapes from tensors.
7576,convert_shapes,tensorflow/tensorflow/python/keras/utils/tf_utils.py,202,function,"Converts nested shape representations to desired format.
Performs:
TensorShapes -> tuples if `to_tuples=True`.
tuples of int or None -> TensorShapes if `to_tuples=False`.
Valid objects to be converted are:
- TensorShapes
- tuples with elements of type int or None.
- ints
- None
Arguments:
input_shape: A nested structure of objects to be converted to TensorShapes.
to_tuples: If `True`, converts all TensorShape to tuples. Otherwise converts
all tuples representing shapes to TensorShapes.
Returns:
Nested structure of shapes in desired format.
Raises:
ValueError: when the input tensor shape can't be converted to tuples, eg
unknown tensor shape."
7577,ListWrapper,tensorflow/tensorflow/python/keras/utils/tf_utils.py,253,class,A wrapper for lists to be treated as elements for `nest`.
7578,as_list,tensorflow/tensorflow/python/keras/utils/tf_utils.py,259,method,
7579,convert_inner_node_data,tensorflow/tensorflow/python/keras/utils/tf_utils.py,263,function,"Either wraps or unwraps innermost node data lists in `ListWrapper` objects.
Arguments:
nested: A nested data structure.
wrap: If `True`, wrap innermost lists in `ListWrapper` objects. If `False`,
unwraps `ListWrapper` objects into lists.
Returns:
Structure of same type as nested, with lists wrapped/unwrapped."
7580,shape_type_conversion,tensorflow/tensorflow/python/keras/utils/tf_utils.py,308,function,"Decorator that handles tuple/TensorShape conversion.
Used in `compute_output_shape` and `build`.
Arguments:
fn: function to wrap.
Returns:
Wrapped function."
7581,are_all_symbolic_tensors,tensorflow/tensorflow/python/keras/utils/tf_utils.py,334,function,
7582,is_symbolic_tensor,tensorflow/tensorflow/python/keras/utils/tf_utils.py,341,function,"Returns whether a tensor is symbolic (from a TF graph) or an eager tensor.
A Variable can be seen as either: it is considered symbolic
when we are in a graph scope, and eager when we are in an eager scope.
Arguments:
tensor: A tensor instance to test.
Returns:
True for symbolic tensors, False for eager tensors."
7583,register_symbolic_tensor_type,tensorflow/tensorflow/python/keras/utils/tf_utils.py,373,function,"Allows users to specify types regarded as symbolic `Tensor`s.
Used in conjunction with `tf.register_tensor_conversion_function`, calling
`tf.keras.utils.register_symbolic_tensor_type(cls)` allows non-`Tensor`
objects to be plumbed through Keras layers.
Example:
```python
# One-time setup.
class Foo(object):
def __init__(self, input_):
self._input = input_
def value(self):
return tf.constant(42.)
tf.register_tensor_conversion_function(
Foo, lambda x, *args, **kwargs: x.value())
tf.keras.utils.register_symbolic_tensor_type(Foo)
# User-land.
layer = tf.keras.layers.Lambda(lambda input_: Foo(input_))
```
Arguments:
cls: A `class` type which shall be regarded as a symbolic `Tensor`."
7584,type_spec_from_value,tensorflow/tensorflow/python/keras/utils/tf_utils.py,406,function,Grab type_spec without converting array-likes to tensors.
7585,is_ragged,tensorflow/tensorflow/python/keras/utils/tf_utils.py,418,function,Returns true if `tensor` is a ragged tensor or ragged tensor value.
7586,is_tensor_or_variable,tensorflow/tensorflow/python/keras/utils/tf_utils.py,425,function,
7587,assert_no_legacy_layers,tensorflow/tensorflow/python/keras/utils/tf_utils.py,429,function,"Prevent tf.layers.Layers from being used with Keras.
Certain legacy layers inherit from their keras analogs; however they are
not supported with keras and can lead to subtle and hard to diagnose bugs.
Args:
layers: A list of layers to check
Raises:
TypeError: If any elements of layers are tf.layers.Layers"
7588,maybe_init_scope,tensorflow/tensorflow/python/keras/utils/tf_utils.py,455,function,"Open an `init_scope` if in V2 mode and using the keras graph.
Arguments:
layer: The Layer/Model that is currently active.
Yields:
None"
7589,graph_context_for_symbolic_tensors,tensorflow/tensorflow/python/keras/utils/tf_utils.py,474,function,Returns graph context manager if any of the inputs is a symbolic tensor.
7590,dataset_is_infinite,tensorflow/tensorflow/python/keras/utils/tf_utils.py,483,function,True if the passed dataset is infinite.
7591,get_tensor_spec,tensorflow/tensorflow/python/keras/utils/tf_utils.py,493,function,Returns a `TensorSpec` given a single `Tensor` or `TensorSpec`.
7592,to_numpy_or_python_type,tensorflow/tensorflow/python/keras/utils/tf_utils.py,522,function,"Converts a structure of `Tensor`s to `NumPy` arrays or Python scalar types.
For each tensor, it calls `tensor.numpy()`. If the result is a scalar value,
it converts it to a Python type, such as a float or int, by calling
`result.item()`.
Numpy scalars are converted, as Python types are often more convenient to deal
with. This is especially useful for bfloat16 Numpy scalars, which don't
support as many operations as other Numpy values.
Args:
tensors: A structure of tensors.
Returns:
`tensors`, but scalar tensors are converted to Python types and non-scalar
tensors are converted to Numpy arrays."
7593,ModelVersionSelector,tensorflow/tensorflow/python/keras/utils/version_utils.py,51,class,Chooses between Keras v1 and v2 Model class.
7594,LayerVersionSelector,tensorflow/tensorflow/python/keras/utils/version_utils.py,60,class,Chooses between Keras v1 and v2 Layer class.
7595,TensorBoardVersionSelector,tensorflow/tensorflow/python/keras/utils/version_utils.py,69,class,Chooses between Keras v1 and v2 TensorBoard callback class.
7596,should_use_v2,tensorflow/tensorflow/python/keras/utils/version_utils.py,84,function,Determine if v1 or v2 version should be used.
7597,swap_class,tensorflow/tensorflow/python/keras/utils/version_utils.py,98,function,Swaps in v2_cls or v1_cls depending on graph mode.
7598,disallow_legacy_graph,tensorflow/tensorflow/python/keras/utils/version_utils.py,114,function,
7599,is_v1_layer_or_model,tensorflow/tensorflow/python/keras/utils/version_utils.py,125,function,
7600,check_pydot,tensorflow/tensorflow/python/keras/utils/vis_utils.py,44,function,Returns True if PyDot and Graphviz are available.
7601,is_wrapped_model,tensorflow/tensorflow/python/keras/utils/vis_utils.py,57,function,
7602,add_edge,tensorflow/tensorflow/python/keras/utils/vis_utils.py,64,function,
7603,model_to_dot,tensorflow/tensorflow/python/keras/utils/vis_utils.py,70,function,"Convert a Keras model to dot format.
Arguments:
model: A Keras model instance.
show_shapes: whether to display shape information.
show_dtype: whether to display layer dtypes.
show_layer_names: whether to display layer names.
rankdir: `rankdir` argument passed to PyDot,
a string specifying the format of the plot:
'TB' creates a vertical plot;
'LR' creates a horizontal plot.
expand_nested: whether to expand nested models into clusters.
dpi: Dots per inch.
subgraph: whether to return a `pydot.Cluster` instance.
Returns:
A `pydot.Dot` instance representing the Keras model or
a `pydot.Cluster` instance representing nested model if
`subgraph=True`.
Raises:
ImportError: if graphviz or pydot are not available."
7604,plot_model,tensorflow/tensorflow/python/keras/utils/vis_utils.py,281,function,"Converts a Keras model to dot format and save to a file.
Example:
```python
input = tf.keras.Input(shape=(100,), dtype='int32', name='input')
x = tf.keras.layers.Embedding(
output_dim=512, input_dim=10000, input_length=100)(input)
x = tf.keras.layers.LSTM(32)(x)
x = tf.keras.layers.Dense(64, activation='relu')(x)
x = tf.keras.layers.Dense(64, activation='relu')(x)
x = tf.keras.layers.Dense(64, activation='relu')(x)
output = tf.keras.layers.Dense(1, activation='sigmoid', name='output')(x)
model = tf.keras.Model(inputs=[input], outputs=[output])
dot_img_file = '/tmp/model_1.png'
tf.keras.utils.plot_model(model, to_file=dot_img_file, show_shapes=True)
```
Arguments:
model: A Keras model instance
to_file: File name of the plot image.
show_shapes: whether to display shape information.
show_dtype: whether to display layer dtypes.
show_layer_names: whether to display layer names.
rankdir: `rankdir` argument passed to PyDot,
a string specifying the format of the plot:
'TB' creates a vertical plot;
'LR' creates a horizontal plot.
expand_nested: Whether to expand nested models into clusters.
dpi: Dots per inch.
Returns:
A Jupyter notebook Image object if Jupyter is installed.
This enables in-line display of the model plots in notebooks."
7605,BaseWrapper,tensorflow/tensorflow/python/keras/wrappers/scikit_learn.py,33,class,"Base class for the Keras scikit-learn wrapper.
Warning: This class should not be used directly.
Use descendant classes instead.
Arguments:
build_fn: callable function or class instance
**sk_params: model parameters & fitting parameters
The `build_fn` should construct, compile and return a Keras model, which
will then be used to fit/predict. One of the following
three values could be passed to `build_fn`:
1. A function
2. An instance of a class that implements the `__call__` method
3. None. This means you implement a class that inherits from either
`KerasClassifier` or `KerasRegressor`. The `__call__` method of the
present class will then be treated as the default `build_fn`.
`sk_params` takes both model parameters and fitting parameters. Legal model
parameters are the arguments of `build_fn`. Note that like all other
estimators in scikit-learn, `build_fn` should provide default values for
its arguments, so that you could create the estimator without passing any
values to `sk_params`.
`sk_params` could also accept parameters for calling `fit`, `predict`,
`predict_proba`, and `score` methods (e.g., `epochs`, `batch_size`).
fitting (predicting) parameters are selected in the following order:
1. Values passed to the dictionary arguments of
`fit`, `predict`, `predict_proba`, and `score` methods
2. Values passed to `sk_params`
3. The default values of the `keras.models.Sequential`
`fit`, `predict`, `predict_proba` and `score` methods
When using scikit-learn's `grid_search` API, legal tunable parameters are
those you could pass to `sk_params`, including fitting parameters.
In other words, you could use `grid_search` to search for the best
`batch_size` or `epochs` as well as the model parameters."
7606,check_params,tensorflow/tensorflow/python/keras/wrappers/scikit_learn.py,79,method,"Checks for user typos in `params`.
Arguments:
params: dictionary; the parameters to be checked
Raises:
ValueError: if any member of `params` is not a valid argument."
7607,get_params,tensorflow/tensorflow/python/keras/wrappers/scikit_learn.py,108,method,"Gets parameters for this estimator.
Arguments:
**params: ignored (exists for API compatibility).
Returns:
Dictionary of parameter names mapped to their values."
7608,set_params,tensorflow/tensorflow/python/keras/wrappers/scikit_learn.py,121,method,"Sets the parameters of this estimator.
Arguments:
**params: Dictionary of parameter names mapped to their values.
Returns:
self"
7609,fit,tensorflow/tensorflow/python/keras/wrappers/scikit_learn.py,134,method,"Constructs a new model with `build_fn` & fit the model to `(x, y)`.
Arguments:
x : array-like, shape `(n_samples, n_features)`
Training samples where `n_samples` is the number of samples
and `n_features` is the number of features.
y : array-like, shape `(n_samples,)` or `(n_samples, n_outputs)`
True labels for `x`.
**kwargs: dictionary arguments
Legal arguments are the arguments of `Sequential.fit`
Returns:
history : object
details about the training history at each epoch."
7610,filter_sk_params,tensorflow/tensorflow/python/keras/wrappers/scikit_learn.py,170,method,"Filters `sk_params` and returns those in `fn`'s arguments.
Arguments:
fn : arbitrary function
override: dictionary, values to override `sk_params`
Returns:
res : dictionary containing variables
in both `sk_params` and `fn`'s arguments."
7611,KerasClassifier,tensorflow/tensorflow/python/keras/wrappers/scikit_learn.py,191,class,"Implementation of the scikit-learn classifier API for Keras.
"
7612,fit,tensorflow/tensorflow/python/keras/wrappers/scikit_learn.py,195,method,"Constructs a new model with `build_fn` & fit the model to `(x, y)`.
Arguments:
x : array-like, shape `(n_samples, n_features)`
Training samples where `n_samples` is the number of samples
and `n_features` is the number of features.
y : array-like, shape `(n_samples,)` or `(n_samples, n_outputs)`
True labels for `x`.
**kwargs: dictionary arguments
Legal arguments are the arguments of `Sequential.fit`
Returns:
history : object
details about the training history at each epoch.
Raises:
ValueError: In case of invalid shape for `y` argument."
7613,predict,tensorflow/tensorflow/python/keras/wrappers/scikit_learn.py,225,method,"Returns the class predictions for the given test data.
Arguments:
x: array-like, shape `(n_samples, n_features)`
Test samples where `n_samples` is the number of samples
and `n_features` is the number of features.
**kwargs: dictionary arguments
Legal arguments are the arguments
of `Sequential.predict_classes`.
Returns:
preds: array-like, shape `(n_samples,)`
Class predictions."
7614,predict_proba,tensorflow/tensorflow/python/keras/wrappers/scikit_learn.py,244,method,"Returns class probability estimates for the given test data.
Arguments:
x: array-like, shape `(n_samples, n_features)`
Test samples where `n_samples` is the number of samples
and `n_features` is the number of features.
**kwargs: dictionary arguments
Legal arguments are the arguments
of `Sequential.predict_classes`.
Returns:
proba: array-like, shape `(n_samples, n_outputs)`
Class probability estimates.
In the case of binary classification,
to match the scikit-learn API,
will return an array of shape `(n_samples, 2)`
(instead of `(n_sample, 1)` as in Keras)."
7615,score,tensorflow/tensorflow/python/keras/wrappers/scikit_learn.py,272,method,"Returns the mean accuracy on the given test data and labels.
Arguments:
x: array-like, shape `(n_samples, n_features)`
Test samples where `n_samples` is the number of samples
and `n_features` is the number of features.
y: array-like, shape `(n_samples,)` or `(n_samples, n_outputs)`
True labels for `x`.
**kwargs: dictionary arguments
Legal arguments are the arguments of `Sequential.evaluate`.
Returns:
score: float
Mean accuracy of predictions on `x` wrt. `y`.
Raises:
ValueError: If the underlying model isn't configured to
compute accuracy. You should pass `metrics=[""accuracy""]` to
the `.compile()` method of the model."
7616,KerasRegressor,tensorflow/tensorflow/python/keras/wrappers/scikit_learn.py,314,class,"Implementation of the scikit-learn regressor API for Keras.
"
7617,predict,tensorflow/tensorflow/python/keras/wrappers/scikit_learn.py,318,method,"Returns predictions for the given test data.
Arguments:
x: array-like, shape `(n_samples, n_features)`
Test samples where `n_samples` is the number of samples
and `n_features` is the number of features.
**kwargs: dictionary arguments
Legal arguments are the arguments of `Sequential.predict`.
Returns:
preds: array-like, shape `(n_samples,)`
Predictions."
7618,score,tensorflow/tensorflow/python/keras/wrappers/scikit_learn.py,335,method,"Returns the mean loss on the given test data and labels.
Arguments:
x: array-like, shape `(n_samples, n_features)`
Test samples where `n_samples` is the number of samples
and `n_features` is the number of features.
y: array-like, shape `(n_samples,)`
True labels for `x`.
**kwargs: dictionary arguments
Legal arguments are the arguments of `Sequential.evaluate`.
Returns:
score: float
Mean accuracy of predictions on `x` wrt. `y`."
7619,build_fn_clf,tensorflow/tensorflow/python/keras/wrappers/scikit_learn_test.py,37,function,
7620,assert_classification_works,tensorflow/tensorflow/python/keras/wrappers/scikit_learn_test.py,50,function,
7621,build_fn_reg,tensorflow/tensorflow/python/keras/wrappers/scikit_learn_test.py,73,function,
7622,assert_regression_works,tensorflow/tensorflow/python/keras/wrappers/scikit_learn_test.py,86,function,
7623,StridedSliceChecker,tensorflow/tensorflow/python/kernel_tests/array_ops_test.py,566,class,Check a given tensor against the numpy result.
7624,eval_if_tensor,tensorflow/tensorflow/python/kernel_tests/array_ops_test.py,586,method,
7625,StridedSliceShapeChecker,tensorflow/tensorflow/python/kernel_tests/array_ops_test.py,851,class,
7626,GradSliceChecker,tensorflow/tensorflow/python/kernel_tests/array_ops_test.py,917,class,Tests that we can compute a gradient for var^2.
7627,BenchmarkSlice,tensorflow/tensorflow/python/kernel_tests/array_ops_test.py,1057,class,
7628,StridedSliceBenchmark,tensorflow/tensorflow/python/kernel_tests/array_ops_test.py,1066,class,Benchmark new strided slice operation on non-trivial case.
7629,run_and_time,tensorflow/tensorflow/python/kernel_tests/array_ops_test.py,1069,method,
7630,make_variable,tensorflow/tensorflow/python/kernel_tests/array_ops_test.py,1080,method,
7631,benchmark_strided_slice_skip,tensorflow/tensorflow/python/kernel_tests/array_ops_test.py,1089,method,
7632,benchmark_strided_slice_easy,tensorflow/tensorflow/python/kernel_tests/array_ops_test.py,1096,method,
7633,benchmark_slice_easy,tensorflow/tensorflow/python/kernel_tests/array_ops_test.py,1103,method,
7634,StridedSliceAssignChecker,tensorflow/tensorflow/python/kernel_tests/array_ops_test.py,1110,class,
7635,upsample_filters,tensorflow/tensorflow/python/kernel_tests/atrous_convolution_test.py,36,function,"Upsamples the filters by a factor of rate along the spatial dimensions.
Args:
filters: spatial_shape + [in_channels, out_channels]
Original filters.
rate: A list of len(spatial_shape) positive ints, specifying the
upsampling rate.
Returns:
filters_up: output_spatial_shape + [in_channels, out_channels].
Upsampled filters with
output_spatial_shape[i] = (spatial_shape[i] - 1) * rate[i] + 1
containing (rate[i] - 1) zeros between consecutive filter values along
spatial dimension i."
7636,GetRandomNormalInput,tensorflow/tensorflow/python/kernel_tests/batch_matmul_op_test.py,35,function,
7637,BatchMatMulBenchmark,tensorflow/tensorflow/python/kernel_tests/batch_matmul_op_test.py,212,class,
7638,benchmarkBatchMatMulBroadcast,tensorflow/tensorflow/python/kernel_tests/batch_matmul_op_test.py,229,method,
7639,PythonOpImpl,tensorflow/tensorflow/python/kernel_tests/batchtospace_op_test.py,37,class,
7640,batch_to_space,tensorflow/tensorflow/python/kernel_tests/batchtospace_op_test.py,40,method,
7641,CppOpImpl,tensorflow/tensorflow/python/kernel_tests/batchtospace_op_test.py,44,class,
7642,batch_to_space,tensorflow/tensorflow/python/kernel_tests/batchtospace_op_test.py,47,method,
7643,BatchToSpaceDepthToSpace,tensorflow/tensorflow/python/kernel_tests/batchtospace_op_test.py,51,class,
7644,BatchToSpaceDepthToSpaceCpp,tensorflow/tensorflow/python/kernel_tests/batchtospace_op_test.py,69,class,
7645,SomeRandomBenchmark,tensorflow/tensorflow/python/kernel_tests/benchmark_test.py,41,class,This Benchmark should automatically be registered in the registry.
7646,notBenchmarkMethod,tensorflow/tensorflow/python/kernel_tests/benchmark_test.py,47,method,
7647,benchmark1,tensorflow/tensorflow/python/kernel_tests/benchmark_test.py,50,method,
7648,benchmark2,tensorflow/tensorflow/python/kernel_tests/benchmark_test.py,53,method,
7649,AssertV2Asserts,tensorflow/tensorflow/python/kernel_tests/check_ops_test.py,45,class,
7650,failing_fn,tensorflow/tensorflow/python/kernel_tests/check_ops_test.py,99,method,
7651,EnsureShapeBenchmark,tensorflow/tensorflow/python/kernel_tests/check_ops_test.py,1023,class,
7652,benchmark_const_op,tensorflow/tensorflow/python/kernel_tests/check_ops_test.py,1057,method,
7653,benchmark_single_ensure_op,tensorflow/tensorflow/python/kernel_tests/check_ops_test.py,1064,method,
7654,benchmark_n_ops,tensorflow/tensorflow/python/kernel_tests/check_ops_test.py,1077,method,
7655,benchmark_n_ensure_ops,tensorflow/tensorflow/python/kernel_tests/check_ops_test.py,1083,method,
7656,TriAngSolveCompositeGrad,tensorflow/tensorflow/python/kernel_tests/cholesky_op_test.py,54,function,
7657,MatrixInverseCompositeGrad,tensorflow/tensorflow/python/kernel_tests/cholesky_op_test.py,75,function,
7658,TriAngInvCompositeGrad,tensorflow/tensorflow/python/kernel_tests/cholesky_op_test.py,80,function,
7659,CholeskyBenchmark,tensorflow/tensorflow/python/kernel_tests/cholesky_op_test.py,300,class,
7660,benchmarkCholeskyOp,tensorflow/tensorflow/python/kernel_tests/cholesky_op_test.py,326,method,
7661,benchmarkGradVariants,tensorflow/tensorflow/python/kernel_tests/cholesky_op_test.py,355,method,
7662,check_consumers,tensorflow/tensorflow/python/kernel_tests/control_flow_ops_py_test.py,90,function,Sanity check on the consumer list of the tensors.
7663,all_fetchables,tensorflow/tensorflow/python/kernel_tests/control_flow_ops_py_test.py,104,function,
7664,all_feedables,tensorflow/tensorflow/python/kernel_tests/control_flow_ops_py_test.py,114,function,
7665,opt_cfg,tensorflow/tensorflow/python/kernel_tests/control_flow_ops_py_test.py,124,function,
7666,isum,tensorflow/tensorflow/python/kernel_tests/control_flow_ops_py_test.py,134,function,
7667,enqueue_print_op,tensorflow/tensorflow/python/kernel_tests/control_flow_ops_py_test.py,143,function,Enqueues an op that prints a message to be captured in the test.
7668,tf_function_in_tf2,tensorflow/tensorflow/python/kernel_tests/control_flow_ops_py_test.py,154,function,
7669,WhileOpBenchmark,tensorflow/tensorflow/python/kernel_tests/control_flow_ops_py_test.py,4903,class,Evaluate the performance of while_loop op.
7670,benchmarkWhileOpCrossDevicePlacement,tensorflow/tensorflow/python/kernel_tests/control_flow_ops_py_test.py,4988,method,
7671,benchmarkWhileOpSameDevicePlacement,tensorflow/tensorflow/python/kernel_tests/control_flow_ops_py_test.py,4995,method,
7672,benchmarkWhileOpUnrollCrossDevicePlacement,tensorflow/tensorflow/python/kernel_tests/control_flow_ops_py_test.py,5002,method,
7673,benchmarkWhileOpUnrollSameDevicePlacement,tensorflow/tensorflow/python/kernel_tests/control_flow_ops_py_test.py,5009,method,
7674,loop_body,tensorflow/tensorflow/python/kernel_tests/control_flow_ops_py_test.py,4945,method,
7675,GetShrunkInceptionShapes,tensorflow/tensorflow/python/kernel_tests/conv_ops_test.py,52,function,"Iterator for smaller versions of convolution shapes in 2015 Inception.
Relative to inception, each depth value is `depth // shrink`.
Args:
shrink: Factor to shrink each depth value by relative to Inception.
Yields:
Tuple (input_size, filter_size, out_size, stride, padding), the convolution
parameters of Inception layers."
7676,Conv2DBenchmark,tensorflow/tensorflow/python/kernel_tests/conv_ops_test.py,2924,class,
7677,benchmarkGPUConvStackFirst,tensorflow/tensorflow/python/kernel_tests/conv_ops_test.py,2926,method,
7678,benchmarkExplicitVsManualPadding,tensorflow/tensorflow/python/kernel_tests/conv_ops_test.py,2966,method,"Compare performance of EXPLICIT padding and calling tf.pad.
A Conv2D op with EXPLICIT padding is benchmarked, and a tf.pad with the same
padding followed by an equivalent Conv2D op is benchmarked."
7679,benchmarkExplicitVsSamePaddingGraph,tensorflow/tensorflow/python/kernel_tests/conv_ops_test.py,3016,method,"Compare performance of EXPLICIT and SAME padding in graph mode.
A Conv2D op with SAME padding is benchmarked, and an equivalent Conv2D op
with explicit padding is benchmarked, where the padding is the same as in
the SAME case. The purpose is to ensure EXPLICIT padding is just as
efficient as the SAME case"
7680,benchmarkExplicitVsSamePaddingEager,tensorflow/tensorflow/python/kernel_tests/conv_ops_test.py,3065,method,"Compare performance of EXPLICIT and SAME padding in eager mode.
A Conv2D op with SAME padding is benchmarked, and an equivalent Conv2D op
with explicit padding is benchmarked, where the padding is the same as in
the SAME case. Currently, EXPLICIT padding is slightly slower, due to the
fact the Python padding list must be checked and processed before the Conv2D
op can run."
7681,grouper,tensorflow/tensorflow/python/kernel_tests/ctc_decoder_ops_test.py,34,function,Collect data into fixed-length chunks or blocks.
7682,flatten,tensorflow/tensorflow/python/kernel_tests/ctc_decoder_ops_test.py,41,function,Flatten one level of nesting.
7683,SimpleSparseTensorFrom,tensorflow/tensorflow/python/kernel_tests/ctc_loss_op_test.py,43,function,"Create a very simple SparseTensor with dimensions (batch, time).
Args:
x: a list of lists of type int
Returns:
x_ix and x_val, the indices and values of the SparseTensor<2>."
7684,DecodeJpegBenchmark,tensorflow/tensorflow/python/kernel_tests/decode_jpeg_op_test.py,37,class,Evaluate tensorflow DecodeJpegOp performance.
7685,benchmarkDecodeJpegSmall,tensorflow/tensorflow/python/kernel_tests/decode_jpeg_op_test.py,123,method,Evaluate single DecodeImageOp for small size image.
7686,benchmarkDecodeJpegMedium,tensorflow/tensorflow/python/kernel_tests/decode_jpeg_op_test.py,147,method,Evaluate single DecodeImageOp for medium size image.
7687,benchmarkDecodeJpegLarge,tensorflow/tensorflow/python/kernel_tests/decode_jpeg_op_test.py,171,method,Evaluate single DecodeImageOp for large size image.
7688,MatrixDeterminantBenchmark,tensorflow/tensorflow/python/kernel_tests/determinant_op_test.py,166,class,
7689,benchmarkMatrixDeterminantOp,tensorflow/tensorflow/python/kernel_tests/determinant_op_test.py,191,method,
7690,zip_to_first_list_length,tensorflow/tensorflow/python/kernel_tests/diag_op_test.py,39,function,
7691,repack_diagonals,tensorflow/tensorflow/python/kernel_tests/diag_op_test.py,45,function,
7692,square_cases,tensorflow/tensorflow/python/kernel_tests/diag_op_test.py,99,function,
7693,tall_cases,tensorflow/tensorflow/python/kernel_tests/diag_op_test.py,177,function,
7694,fat_cases,tensorflow/tensorflow/python/kernel_tests/diag_op_test.py,264,function,
7695,ConstantOf,tensorflow/tensorflow/python/kernel_tests/edit_distance_op_test.py,30,function,
7696,SortEigenValues,tensorflow/tensorflow/python/kernel_tests/eig_op_test.py,96,function,
7697,SortEigenDecomposition,tensorflow/tensorflow/python/kernel_tests/eig_op_test.py,101,function,
7698,EquilibrateEigenVectorPhases,tensorflow/tensorflow/python/kernel_tests/eig_op_test.py,108,function,"Equilibrate the phase of the Eigenvectors in the columns of `x` and `y`.
Eigenvectors are only unique up to an arbitrary phase. This function rotates x
such that it matches y. Precondition: The columns of x and y differ by a
multiplicative complex phase factor only.
Args:
x: `np.ndarray` with Eigenvectors
y: `np.ndarray` with Eigenvectors
Returns:
`np.ndarray` containing an equilibrated version of x."
7699,EinsumBenchmark,tensorflow/tensorflow/python/kernel_tests/einsum_op_test.py,397,class,
7700,benchmarkEinsum,tensorflow/tensorflow/python/kernel_tests/einsum_op_test.py,426,method,
7701,ExtractImagePatches,tensorflow/tensorflow/python/kernel_tests/extract_image_patches_op_test.py,28,class,Functional tests for ExtractImagePatches op.
7702,ExtractVolumePatches,tensorflow/tensorflow/python/kernel_tests/extract_volume_patches_op_test.py,27,class,Functional tests for ExtractVolumePatches op.
7703,FIFOQueueBenchmark,tensorflow/tensorflow/python/kernel_tests/fifo_queue_test.py,1738,class,Benchmark FIFOQueue operations.
7704,simple_scoped_fn,tensorflow/tensorflow/python/kernel_tests/functional_ops_test.py,50,function,"Simple function: (a, x) -> 2(x+a), but with ""2"" as a variable in scope."
7705,GatherNdOpBenchmark,tensorflow/tensorflow/python/kernel_tests/gather_nd_op_test.py,387,class,
7706,benchmark_gather_nd_op,tensorflow/tensorflow/python/kernel_tests/gather_nd_op_test.py,389,method,
7707,duplicated_initializer,tensorflow/tensorflow/python/kernel_tests/init_ops_test.py,65,function,"Tests duplicated random initializer within the same graph.
This test generates two random kernels from the same initializer to the same
graph, and checks if the results are close enough. Even given the same global,
seed, two different instances of random kernels should generate different
results.
Args:
tc: An instance of TensorFlowTestCase.
init: An Initializer that generates a tensor of a given shape
graph_seed: A graph-level seed to use.
shape: Shape of the tensor to initialize or `None` to use a vector of length
100.
Returns:
True or False as determined by test."
7708,make_tensor_hiding_attributes,tensorflow/tensorflow/python/kernel_tests/linalg_ops_test.py,415,function,
7709,LUReconstructStatic,tensorflow/tensorflow/python/kernel_tests/linalg_ops_test.py,456,class,
7710,LUReconstructDynamic,tensorflow/tensorflow/python/kernel_tests/linalg_ops_test.py,461,class,
7711,LUMatrixInverseStatic,tensorflow/tensorflow/python/kernel_tests/linalg_ops_test.py,500,class,
7712,LUMatrixInverseDynamic,tensorflow/tensorflow/python/kernel_tests/linalg_ops_test.py,505,class,
7713,LUSolveStatic,tensorflow/tensorflow/python/kernel_tests/linalg_ops_test.py,556,class,
7714,LUSolveDynamic,tensorflow/tensorflow/python/kernel_tests/linalg_ops_test.py,561,class,
7715,IndexTableFromFile,tensorflow/tensorflow/python/kernel_tests/lookup_ops_test.py,1913,class,
7716,IndexTableFromTensor,tensorflow/tensorflow/python/kernel_tests/lookup_ops_test.py,2177,class,
7717,MutableHashTableBenchmark,tensorflow/tensorflow/python/kernel_tests/lookup_ops_test.py,3405,class,
7718,benchmark_single_repeated_scalar_insert_scalar,tensorflow/tensorflow/python/kernel_tests/lookup_ops_test.py,3410,method,
7719,benchmark_many_repeated_scalar_insert_scalar,tensorflow/tensorflow/python/kernel_tests/lookup_ops_test.py,3420,method,
7720,benchmark_single_repeated_batch_32_insert_scalar,tensorflow/tensorflow/python/kernel_tests/lookup_ops_test.py,3431,method,
7721,benchmark_many_repeated_batch_32_insert_scalar,tensorflow/tensorflow/python/kernel_tests/lookup_ops_test.py,3441,method,
7722,DenseHashTableBenchmark,tensorflow/tensorflow/python/kernel_tests/lookup_ops_test.py,3453,class,
7723,LuBenchmark,tensorflow/tensorflow/python/kernel_tests/lu_op_test.py,238,class,
7724,benchmarkLuOp,tensorflow/tensorflow/python/kernel_tests/lu_op_test.py,264,method,
7725,simple_scoped_fn,tensorflow/tensorflow/python/kernel_tests/map_fn_test.py,41,function,"Simple function: (a, x) -> 2(x+a), but with ""2"" as a variable in scope."
7726,infix_matmul,tensorflow/tensorflow/python/kernel_tests/matmul_op_test.py,179,function,
7727,MatrixBandPartBenchmark,tensorflow/tensorflow/python/kernel_tests/matrix_band_part_op_test.py,93,class,
7728,benchmarkMatrixBandPartOp,tensorflow/tensorflow/python/kernel_tests/matrix_band_part_op_test.py,112,method,
7729,np_expm,tensorflow/tensorflow/python/kernel_tests/matrix_exponential_op_test.py,38,function,Slow but accurate Taylor series matrix exponential.
7730,MatrixExponentialBenchmark,tensorflow/tensorflow/python/kernel_tests/matrix_exponential_op_test.py,163,class,
7731,benchmarkMatrixExponentialOp,tensorflow/tensorflow/python/kernel_tests/matrix_exponential_op_test.py,188,method,
7732,MatrixInverseBenchmark,tensorflow/tensorflow/python/kernel_tests/matrix_inverse_op_test.py,156,class,
7733,benchmarkMatrixInverseOp,tensorflow/tensorflow/python/kernel_tests/matrix_inverse_op_test.py,181,method,
7734,MatrixLogarithmBenchmark,tensorflow/tensorflow/python/kernel_tests/matrix_logarithm_op_test.py,148,class,
7735,benchmarkMatrixLogarithmOp,tensorflow/tensorflow/python/kernel_tests/matrix_logarithm_op_test.py,173,method,
7736,MatrixSolveLsBenchmark,tensorflow/tensorflow/python/kernel_tests/matrix_solve_ls_op_test.py,289,class,
7737,benchmarkMatrixSolveLsOp,tensorflow/tensorflow/python/kernel_tests/matrix_solve_ls_op_test.py,327,method,
7738,MatrixSolveBenchmark,tensorflow/tensorflow/python/kernel_tests/matrix_solve_op_test.py,145,class,
7739,benchmarkMatrixSolveOp,tensorflow/tensorflow/python/kernel_tests/matrix_solve_op_test.py,175,method,
7740,empty_sparse,tensorflow/tensorflow/python/kernel_tests/parse_single_example_op_test.py,50,function,
7741,flatten,tensorflow/tensorflow/python/kernel_tests/parse_single_example_op_test.py,57,function,Flatten one level of nesting.
7742,flatten_values_tensors_or_sparse,tensorflow/tensorflow/python/kernel_tests/parse_single_example_op_test.py,62,function,Flatten each SparseTensor object into 3 Tensors for session.run().
7743,flatten,tensorflow/tensorflow/python/kernel_tests/parsing_ops_test.py,61,function,Flatten one level of nesting.
7744,pool_direct_single_axis,tensorflow/tensorflow/python/kernel_tests/pool_test.py,34,function,"Numpy implementation of pooling along a single axis.
This is intended for testing only, and therefore isn't particularly efficient.
See pool_direct below for the meaning of the arguments.
Args:
input: numpy array.
axis: axis along which to perform pooling.
window_size: int >= 1. Size of pooling window within axis.
pooling_type: either ""MAX"" or ""AVG"".
padding: either ""SAME"" or ""VALID"".
dilation_rate: int >= 1. Dilation factor for window, i.e. stride at which
to sample input.
stride: int >= 1. Stride at which to generate output.
Returns:
pooling output array of rank N+2.
Raises:
ValueError: if arguments are invalid."
7745,pool_direct,tensorflow/tensorflow/python/kernel_tests/pool_test.py,99,function,"Numpy implementation of pooling.
This is intended for testing only, and therefore isn't particularly efficient.
See tensorflow.nn.pool.
Args:
input: numpy array of rank N+2.
window_shape: Sequence of N ints >= 1.
pooling_type: either ""MAX"" or ""AVG"".
padding: either ""SAME"" or ""VALID"".
dilation_rate: Sequence of N ints >= 1.
strides: Sequence of N ints >= 1.
data_format: If specified and starts with ""NC"", indicates that second
dimension, rather than the last dimension, specifies the channel.
Returns:
pooling output array of rank N+2.
Raises:
ValueError: if arguments are invalid."
7746,GetDeviceScope,tensorflow/tensorflow/python/kernel_tests/pooling_ops_test.py,44,function,
7747,GetShrunkInceptionMaxPoolShapes,tensorflow/tensorflow/python/kernel_tests/pooling_ops_test.py,80,function,"Iterator for some of the max pool ops in the Inception 2015 model.
Args:
shrink: Factor to shrink depth relative to Inception.
Yields:
Tuple (name, input_size, filter_size, out_size, strides, padding)"
7748,np_func,tensorflow/tensorflow/python/kernel_tests/py_func_test.py,48,function,
7749,matmul,tensorflow/tensorflow/python/kernel_tests/py_func_test.py,52,function,
7750,QRBenchmark,tensorflow/tensorflow/python/kernel_tests/qr_op_test.py,226,class,
7751,benchmarkQROp,tensorflow/tensorflow/python/kernel_tests/qr_op_test.py,248,method,
7752,ReduceBenchmarks,tensorflow/tensorflow/python/kernel_tests/reduce_benchmark_test.py,38,class,Benchmarks for reductions.
7753,benchmark_reduce_sum_grad_eager,tensorflow/tensorflow/python/kernel_tests/reduce_benchmark_test.py,54,method,
7754,benchmark_reduce_sum_grad_eager_cpu,tensorflow/tensorflow/python/kernel_tests/reduce_benchmark_test.py,63,method,
7755,benchmark_reduce_sum_grad_graph,tensorflow/tensorflow/python/kernel_tests/reduce_benchmark_test.py,72,method,
7756,benchmark_reduce_sum_grad_graph_cpu,tensorflow/tensorflow/python/kernel_tests/reduce_benchmark_test.py,88,method,
7757,fn,tensorflow/tensorflow/python/kernel_tests/reduce_benchmark_test.py,58,method,
7758,fn,tensorflow/tensorflow/python/kernel_tests/reduce_benchmark_test.py,67,method,
7759,fn,tensorflow/tensorflow/python/kernel_tests/reduce_benchmark_test.py,83,method,
7760,fn,tensorflow/tensorflow/python/kernel_tests/reduce_benchmark_test.py,100,method,
7761,ReductionUnknownShape,tensorflow/tensorflow/python/kernel_tests/reduction_ops_test.py,100,class,
7762,as_string,tensorflow/tensorflow/python/kernel_tests/regex_replace_op_test.py,94,function,
7763,as_tensor,tensorflow/tensorflow/python/kernel_tests/regex_replace_op_test.py,98,function,
7764,Plus1RNNCell,tensorflow/tensorflow/python/kernel_tests/rnn_cell_test.py,54,class,"RNN Cell generating (output, new_state) = (input + 1, state + 1)."
7765,output_size,tensorflow/tensorflow/python/kernel_tests/rnn_cell_test.py,58,method,
7766,state_size,tensorflow/tensorflow/python/kernel_tests/rnn_cell_test.py,62,method,
7767,DummyMultiDimensionalLSTM,tensorflow/tensorflow/python/kernel_tests/rnn_cell_test.py,69,class,"LSTM Cell generating (output, new_state) = (input + 1, state + 1).
The input to this cell may have an arbitrary number of dimensions that follow
the preceding 'Time' and 'Batch' dimensions."
7768,output_size,tensorflow/tensorflow/python/kernel_tests/rnn_cell_test.py,92,method,
7769,state_size,tensorflow/tensorflow/python/kernel_tests/rnn_cell_test.py,96,method,
7770,NestedRNNCell,tensorflow/tensorflow/python/kernel_tests/rnn_cell_test.py,104,class,"RNN Cell generating (output, new_state) = (input + 1, state + 1).
The input, output and state of this cell is a tuple of two tensors."
7771,output_size,tensorflow/tensorflow/python/kernel_tests/rnn_cell_test.py,111,method,
7772,state_size,tensorflow/tensorflow/python/kernel_tests/rnn_cell_test.py,115,method,
7773,DeviceWrapperCell,tensorflow/tensorflow/python/kernel_tests/rnn_cell_test.py,2356,class,Class to ensure cell calculation happens on a specific device.
7774,output_size,tensorflow/tensorflow/python/kernel_tests/rnn_cell_test.py,2364,method,
7775,state_size,tensorflow/tensorflow/python/kernel_tests/rnn_cell_test.py,2368,method,
7776,Plus1RNNCell,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,52,class,"RNN Cell generating (output, new_state) = (input + 1, state + 1)."
7777,output_size,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,56,method,
7778,state_size,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,60,method,
7779,call,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,63,method,
7780,ScalarStateRNNCell,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,67,class,"RNN Cell generating (output, new_state) = (input + 1, state + 1)."
7781,output_size,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,71,method,
7782,state_size,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,75,method,
7783,zero_state,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,78,method,
7784,call,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,81,method,
7785,UnbalancedOutputRNNCell,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,85,class,"RNN Cell generating (output, new_state) = (input + 1, state + 1)."
7786,output_size,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,89,method,
7787,state_size,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,93,method,
7788,zero_state,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,96,method,
7789,call,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,99,method,
7790,TensorArrayStateRNNCell,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,104,class,RNN Cell its state as a TensorArray.
7791,output_size,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,108,method,
7792,state_size,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,112,method,
7793,zero_state,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,115,method,
7794,call,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,120,method,
7795,graph_creation_static_vs_dynamic_rnn_benchmark,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,403,function,
7796,static_vs_dynamic_rnn_benchmark,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,455,function,
7797,half_seq_len_vs_unroll_half_rnn_benchmark,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,517,function,
7798,concat_state_vs_tuple_state_rnn_benchmark,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,585,function,
7799,dynamic_rnn_swap_memory_benchmark,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,651,function,
7800,rnn_long_sequence_benchmark,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,685,function,
7801,BenchmarkRNN,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,723,class,
7802,benchmarkGraphCreationStaticVsDynamicLSTM,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,725,method,
7803,benchmarkStaticUnrollVsDynamicFlowLSTM,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,739,method,
7804,benchmarkDynamicLSTMNoMemorySwapVsMemorySwap,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,761,method,
7805,benchmarkStaticUnrollHalfSequenceLengthVsHalfUnroll,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,780,method,
7806,benchmarkStaticUnrollStateConcatVsStateTuple,tensorflow/tensorflow/python/kernel_tests/rnn_test.py,804,method,
7807,numpy_reverse,tensorflow/tensorflow/python/kernel_tests/scan_ops_test.py,33,function,
7808,handle_options,tensorflow/tensorflow/python/kernel_tests/scan_ops_test.py,44,function,Adds tf options to numpy scan ops.
7809,SegmentReductionHelper,tensorflow/tensorflow/python/kernel_tests/segment_reduction_ops_test.py,37,class,
7810,SparseSegmentReductionHelper,tensorflow/tensorflow/python/kernel_tests/segment_reduction_ops_test.py,492,class,
7811,SegmentReductionOpBenchmark,tensorflow/tensorflow/python/kernel_tests/segment_reduction_ops_test.py,938,class,
7812,benchmarkSegmentSumGPU,tensorflow/tensorflow/python/kernel_tests/segment_reduction_ops_test.py,983,method,
7813,benchmarkUnsortedSegmentSumGPU,tensorflow/tensorflow/python/kernel_tests/segment_reduction_ops_test.py,991,method,
7814,SortEigenDecomposition,tensorflow/tensorflow/python/kernel_tests/self_adjoint_eig_op_test.py,95,function,
7815,EquilibrateEigenVectorPhases,tensorflow/tensorflow/python/kernel_tests/self_adjoint_eig_op_test.py,103,function,"Equilibrate the phase of the Eigenvectors in the columns of `x` and `y`.
Eigenvectors are only unique up to an arbitrary phase. This function rotates x
such that it matches y. Precondition: The columns of x and y differ by a
multiplicative complex phase factor only.
Args:
x: `np.ndarray` with Eigenvectors
y: `np.ndarray` with Eigenvectors
Returns:
`np.ndarray` containing an equilibrated version of x."
7816,space_to_batch_direct,tensorflow/tensorflow/python/kernel_tests/spacetobatch_op_test.py,34,function,"Direct Python implementation of space-to-batch conversion.
This is used for tests only.
Args:
input_array: N-D array
block_shape: 1-D array of shape [num_block_dims].
paddings: 2-D array of shape [num_block_dims, 2].
Returns:
Converted tensor."
7817,PythonOpImpl,tensorflow/tensorflow/python/kernel_tests/spacetobatch_op_test.py,75,class,
7818,space_to_batch,tensorflow/tensorflow/python/kernel_tests/spacetobatch_op_test.py,78,method,
7819,batch_to_space,tensorflow/tensorflow/python/kernel_tests/spacetobatch_op_test.py,82,method,
7820,CppOpImpl,tensorflow/tensorflow/python/kernel_tests/spacetobatch_op_test.py,86,class,
7821,space_to_batch,tensorflow/tensorflow/python/kernel_tests/spacetobatch_op_test.py,89,method,
7822,batch_to_space,tensorflow/tensorflow/python/kernel_tests/spacetobatch_op_test.py,93,method,
7823,SpaceToBatchSpaceToDepth,tensorflow/tensorflow/python/kernel_tests/spacetobatch_op_test.py,317,class,
7824,SpaceToBatchSpaceToDepthCpp,tensorflow/tensorflow/python/kernel_tests/spacetobatch_op_test.py,334,class,
7825,SparseAddBenchmark,tensorflow/tensorflow/python/kernel_tests/sparse_add_op_test.py,239,class,
7826,benchmarkSparseAddDense,tensorflow/tensorflow/python/kernel_tests/sparse_add_op_test.py,241,method,
7827,RandMatrix,tensorflow/tensorflow/python/kernel_tests/sparse_matmul_op_test.py,31,function,
7828,sparse_tensor_dense_vs_dense_matmul_benchmark,tensorflow/tensorflow/python/kernel_tests/sparse_tensor_dense_matmul_op_test.py,310,function,
7829,BenchmarkSparseTensorsMapVsSerialization,tensorflow/tensorflow/python/kernel_tests/sparse_tensors_map_ops_test.py,190,class,
7830,benchmarkVeryLarge2DFloatSparseTensor,tensorflow/tensorflow/python/kernel_tests/sparse_tensors_map_ops_test.py,192,method,
7831,sparse_vs_dense_xent_benchmark,tensorflow/tensorflow/python/kernel_tests/sparse_xent_op_test.py,313,function,
7832,np_split_squeeze,tensorflow/tensorflow/python/kernel_tests/stack_op_test.py,34,function,
7833,events_from_file,tensorflow/tensorflow/python/kernel_tests/summary_ops_test.py,1210,function,"Returns all events in a single event file.
Args:
filepath: Path to the event file.
Returns:
A list of all tf.Event protos in the event file."
7834,events_from_logdir,tensorflow/tensorflow/python/kernel_tests/summary_ops_test.py,1228,function,"Returns all events in the single eventfile in logdir.
Args:
logdir: The directory in which the single event file is sought.
Returns:
A list of all tf.Event protos from the single event file.
Raises:
AssertionError: If logdir does not contain exactly one file."
7835,to_numpy,tensorflow/tensorflow/python/kernel_tests/summary_ops_test.py,1246,function,
7836,SVDBenchmark,tensorflow/tensorflow/python/kernel_tests/svd_op_test.py,314,class,
7837,benchmarkSVDOp,tensorflow/tensorflow/python/kernel_tests/svd_op_test.py,337,method,
7838,variable_scoped_function,tensorflow/tensorflow/python/kernel_tests/template_test.py,39,function,
7839,internally_variable_scoped_function,tensorflow/tensorflow/python/kernel_tests/template_test.py,45,function,
7840,function_with_create,tensorflow/tensorflow/python/kernel_tests/template_test.py,51,function,Creates a variable as a side effect using tf.Variable.
7841,function_with_side_create,tensorflow/tensorflow/python/kernel_tests/template_test.py,58,function,Creates a variable as a side effect using tf.get_variable.
7842,variable_scoped_function_with_local_variable,tensorflow/tensorflow/python/kernel_tests/template_test.py,65,function,
7843,TensorArrayBenchmark,tensorflow/tensorflow/python/kernel_tests/tensor_array_ops_test.py,1798,class,
7844,benchmarkWriteInWhile,tensorflow/tensorflow/python/kernel_tests/tensor_array_ops_test.py,1814,method,
7845,benchmarkWriteInWhileWithControlFlowV2,tensorflow/tensorflow/python/kernel_tests/tensor_array_ops_test.py,1818,method,
7846,benchmarkWriteInDatasetMapFn,tensorflow/tensorflow/python/kernel_tests/tensor_array_ops_test.py,1821,method,
7847,benchmarkWriteInDatasetParallelMapFn,tensorflow/tensorflow/python/kernel_tests/tensor_array_ops_test.py,1827,method,
7848,TopKBenchmark,tensorflow/tensorflow/python/kernel_tests/topk_op_test.py,224,class,
7849,benchmarkTopK,tensorflow/tensorflow/python/kernel_tests/topk_op_test.py,226,method,
7850,flags,tensorflow/tensorflow/python/kernel_tests/tridiagonal_solve_op_test.py,53,function,
7851,decor,tensorflow/tensorflow/python/kernel_tests/tridiagonal_solve_op_test.py,746,function,
7852,UnicodeScriptBenchmarks,tensorflow/tensorflow/python/kernel_tests/unicode_script_op_test.py,61,class,
7853,benchmark_unicode_script,tensorflow/tensorflow/python/kernel_tests/unicode_script_op_test.py,80,method,
7854,np_split_squeeze,tensorflow/tensorflow/python/kernel_tests/unstack_op_test.py,31,function,
7855,run_inside_wrap_function_in_eager_mode,tensorflow/tensorflow/python/kernel_tests/variable_scope_test.py,48,function,"Decorator to execute the same graph code in eager and graph modes.
In graph mode, we just execute the graph_function passed as argument. In eager
mode, we wrap the function using wrap_function and then execute the wrapped
result.
Args:
graph_function: python function containing graph code to be wrapped
Returns:
decorated function"
7856,axis0_into1_partitioner,tensorflow/tensorflow/python/kernel_tests/variable_scope_test.py,1375,function,
7857,axis0_into2_partitioner,tensorflow/tensorflow/python/kernel_tests/variable_scope_test.py,1380,function,
7858,axis0_into3_partitioner,tensorflow/tensorflow/python/kernel_tests/variable_scope_test.py,1386,function,
7859,WhereBenchmark,tensorflow/tensorflow/python/kernel_tests/where_op_test.py,271,class,
7860,benchmarkWhere,tensorflow/tensorflow/python/kernel_tests/where_op_test.py,273,method,
7861,benchmarkBatchSelect,tensorflow/tensorflow/python/kernel_tests/where_op_test.py,298,method,
7862,random_gamma,tensorflow/tensorflow/python/kernel_tests/while_v2_test.py,56,function,
7863,random_gamma_with_alpha_beta,tensorflow/tensorflow/python/kernel_tests/while_v2_test.py,60,function,
7864,random_poisson_v2,tensorflow/tensorflow/python/kernel_tests/while_v2_test.py,65,function,
7865,random_poisson_v2_with_lam,tensorflow/tensorflow/python/kernel_tests/while_v2_test.py,69,function,
7866,fill,tensorflow/tensorflow/python/kernel_tests/while_v2_test.py,73,function,
7867,ScalarShape,tensorflow/tensorflow/python/kernel_tests/while_v2_test.py,1814,function,
7868,GetOptimizedGraph,tensorflow/tensorflow/python/kernel_tests/while_v2_test.py,1818,function,
7869,XentBenchmark,tensorflow/tensorflow/python/kernel_tests/xent_op_test.py,329,class,
7870,benchmarkZeroDimension,tensorflow/tensorflow/python/kernel_tests/xent_op_test.py,331,method,
7871,benchmarkSingleClass,tensorflow/tensorflow/python/kernel_tests/xent_op_test.py,356,method,
7872,BestMultiDimFeatureSplitMultiClassV2Op,tensorflow/tensorflow/python/kernel_tests/boosted_trees/stats_ops_test.py,1673,class,Tests multi-class/multi-regression for best splits using V2 op.
7873,frobenius,tensorflow/tensorflow/python/kernel_tests/boosted_trees/stats_ops_test.py,1729,method,
7874,try_import,tensorflow/tensorflow/python/kernel_tests/distributions/bernoulli_test.py,36,function,
7875,make_bernoulli,tensorflow/tensorflow/python/kernel_tests/distributions/bernoulli_test.py,48,function,
7876,entropy,tensorflow/tensorflow/python/kernel_tests/distributions/bernoulli_test.py,54,function,
7877,try_import,tensorflow/tensorflow/python/kernel_tests/distributions/beta_test.py,36,function,
7878,IntentionallyMissingError,tensorflow/tensorflow/python/kernel_tests/distributions/bijector_test.py,83,class,
7879,BrokenBijector,tensorflow/tensorflow/python/kernel_tests/distributions/bijector_test.py,87,class,Forward and inverse are not inverses of each other.
7880,ExpOnlyJacobian,tensorflow/tensorflow/python/kernel_tests/distributions/bijector_test.py,196,class,Only used for jacobian calculations.
7881,ConstantJacobian,tensorflow/tensorflow/python/kernel_tests/distributions/bijector_test.py,213,class,Only used for jacobian calculations.
7882,make_categorical,tensorflow/tensorflow/python/kernel_tests/distributions/categorical_test.py,40,function,
7883,try_import,tensorflow/tensorflow/python/kernel_tests/distributions/dirichlet_test.py,35,function,
7884,try_import,tensorflow/tensorflow/python/kernel_tests/distributions/exponential_test.py,34,function,
7885,try_import,tensorflow/tensorflow/python/kernel_tests/distributions/gamma_test.py,36,function,
7886,try_import,tensorflow/tensorflow/python/kernel_tests/distributions/laplace_test.py,35,function,
7887,try_import,tensorflow/tensorflow/python/kernel_tests/distributions/normal_test.py,42,function,
7888,try_import,tensorflow/tensorflow/python/kernel_tests/distributions/special_math_test.py,39,function,
7889,try_import,tensorflow/tensorflow/python/kernel_tests/distributions/student_t_test.py,37,function,
7890,try_import,tensorflow/tensorflow/python/kernel_tests/distributions/uniform_test.py,37,function,
7891,try_import,tensorflow/tensorflow/python/kernel_tests/distributions/util_test.py,43,function,
7892,ReduceWeightedLogSumExp,tensorflow/tensorflow/python/kernel_tests/distributions/util_test.py,809,class,
7893,LinearOperatorLowRankUpdateBroadcastsShape,tensorflow/tensorflow/python/kernel_tests/linalg/linear_operator_low_rank_update_test.py,273,class,Test that the operator's shape is the broadcast of arguments.
7894,LinearOperatorShape,tensorflow/tensorflow/python/kernel_tests/linalg/linear_operator_test.py,37,class,LinearOperator that implements the methods ._shape and _shape_tensor.
7895,LinearOperatorMatmulSolve,tensorflow/tensorflow/python/kernel_tests/linalg/linear_operator_test.py,65,class,LinearOperator that wraps a [batch] matrix and implements matmul/solve.
7896,DummyOperatorWithHint,tensorflow/tensorflow/python/kernel_tests/linalg/linear_operator_util_test.py,305,class,
7897,dense_to_csr_sparse_matrix,tensorflow/tensorflow/python/kernel_tests/linalg/sparse/csr_sparse_matrix_dense_mat_mul_grad_test.py,36,function,
7898,dense_to_csr_sparse_matrix,tensorflow/tensorflow/python/kernel_tests/linalg/sparse/csr_sparse_matrix_grad_test.py,35,function,
7899,dense_to_csr_sparse_matrix,tensorflow/tensorflow/python/kernel_tests/linalg/sparse/csr_sparse_matrix_ops_test.py,53,function,
7900,twist_matrix,tensorflow/tensorflow/python/kernel_tests/linalg/sparse/csr_sparse_matrix_ops_test.py,63,function,Permute the rows and columns of a 2D or (batched) 3D Tensor.
7901,CSRSparseMatrixOpsBenchmark,tensorflow/tensorflow/python/kernel_tests/linalg/sparse/csr_sparse_matrix_ops_test.py,1336,class,
7902,benchmark_sparse_matrix_mat_mul_gpu,tensorflow/tensorflow/python/kernel_tests/linalg/sparse/csr_sparse_matrix_ops_test.py,1338,method,
7903,benchmark_sparse_matrix_mat_vec_mul,tensorflow/tensorflow/python/kernel_tests/linalg/sparse/csr_sparse_matrix_ops_test.py,1405,method,
7904,benchmark_sparse_matrix_sparse_matmul,tensorflow/tensorflow/python/kernel_tests/linalg/sparse/csr_sparse_matrix_ops_test.py,1478,method,
7905,benchmark_sparse_dense_conversion,tensorflow/tensorflow/python/kernel_tests/linalg/sparse/csr_sparse_matrix_ops_test.py,1529,method,
7906,benchmark_sparse_cholesky,tensorflow/tensorflow/python/kernel_tests/linalg/sparse/csr_sparse_matrix_ops_test.py,1582,method,
7907,dense_to_csr_sparse_matrix,tensorflow/tensorflow/python/kernel_tests/linalg/sparse/csr_sparse_matrix_sparse_mat_mul_grad_test.py,36,function,
7908,composed_sampler,tensorflow/tensorflow/python/kernel_tests/random/multinomial_op_test.py,40,function,
7909,native_op_vs_composed_ops,tensorflow/tensorflow/python/kernel_tests/random/multinomial_op_test.py,216,function,
7910,MultinomialBenchmark,tensorflow/tensorflow/python/kernel_tests/random/multinomial_op_test.py,237,class,
7911,benchmarkNativeOpVsComposedOps,tensorflow/tensorflow/python/kernel_tests/random/multinomial_op_test.py,239,method,
7912,TruncatedNormalMoments,tensorflow/tensorflow/python/kernel_tests/random/parameterized_truncated_normal_op_test.py,51,class,
7913,calculate_moments,tensorflow/tensorflow/python/kernel_tests/random/parameterized_truncated_normal_op_test.py,96,function,
7914,parameterized_vs_naive,tensorflow/tensorflow/python/kernel_tests/random/parameterized_truncated_normal_op_test.py,394,function,
7915,randn_sampler_switchover,tensorflow/tensorflow/python/kernel_tests/random/parameterized_truncated_normal_op_test.py,419,function,
7916,TruncatedNormalBenchmark,tensorflow/tensorflow/python/kernel_tests/random/parameterized_truncated_normal_op_test.py,471,class,
7917,benchmarkParameterizedOpVsNaiveOpCpu,tensorflow/tensorflow/python/kernel_tests/random/parameterized_truncated_normal_op_test.py,473,method,
7918,benchmarkParameterizedOpVsNaiveOpGpu,tensorflow/tensorflow/python/kernel_tests/random/parameterized_truncated_normal_op_test.py,476,method,
7919,benchmarkRandnSamplerCPU,tensorflow/tensorflow/python/kernel_tests/random/parameterized_truncated_normal_op_test.py,498,method,
7920,benchmarkRandnSamplerGPU,tensorflow/tensorflow/python/kernel_tests/random/parameterized_truncated_normal_op_test.py,501,method,
7921,invert_philox,tensorflow/tensorflow/python/kernel_tests/random/stateless_random_ops_test.py,38,function,Invert the Philox bijection.
7922,chi_squared,tensorflow/tensorflow/python/kernel_tests/random/util.py,79,function,Pearson's Chi-squared test.
7923,normal_cdf,tensorflow/tensorflow/python/kernel_tests/random/util.py,88,function,Cumulative distribution function for a standard normal distribution.
7924,anderson_darling,tensorflow/tensorflow/python/kernel_tests/random/util.py,93,function,Anderson-Darling test for a standard normal distribution.
7925,try_import,tensorflow/tensorflow/python/kernel_tests/signal/dct_ops_test.py,33,function,
7926,hertz_to_mel,tensorflow/tensorflow/python/kernel_tests/signal/mel_ops_test.py,40,function,"Convert frequencies to mel scale using HTK formula.
Copied from
https://github.com/tensorflow/models/blob/master/research/audioset/mel_features.py.
Args:
frequencies_hertz: Scalar or np.array of frequencies in hertz.
Returns:
Object of same size as frequencies_hertz containing corresponding values
on the mel scale."
7927,spectrogram_to_mel_matrix,tensorflow/tensorflow/python/kernel_tests/signal/mel_ops_test.py,57,function,"Return a matrix that can post-multiply spectrogram rows to make mel.
Copied from
https://github.com/tensorflow/models/blob/master/research/audioset/mel_features.py.
Returns a np.array matrix A that can be used to post-multiply a matrix S of
spectrogram values (STFT magnitudes) arranged as frames x bins to generate a
""mel spectrogram"" M of frames x num_mel_bins. M = S A.
The classic HTK algorithm exploits the complementarity of adjacent mel bands
to multiply each FFT bin by only one mel weight, then add it, with positive
and negative signs, to the two adjacent mel bands to which that bin
contributes. Here, by expressing this operation as a matrix multiply, we go
from num_fft multiplies per frame (plus around 2*num_fft adds) to around
num_fft^2 multiplies and adds. However, because these are all presumably
accomplished in a single call to np.dot(), it's not clear which approach is
faster in Python. The matrix multiplication has the attraction of being more
general and flexible, and much easier to read.
Args:
num_mel_bins: How many bands in the resulting mel spectrum. This is
the number of columns in the output matrix.
num_spectrogram_bins: How many bins there are in the source spectrogram
data, which is understood to be fft_size/2 + 1, i.e. the spectrogram
only contains the nonredundant FFT bins.
audio_sample_rate: Samples per second of the audio at the input to the
spectrogram. We need this to figure out the actual frequencies for
each spectrogram bin, which dictates how they are mapped into mel.
lower_edge_hertz: Lower bound on the frequencies to be included in the mel
spectrum. This corresponds to the lower edge of the lowest triangular
band.
upper_edge_hertz: The desired top edge of the highest frequency band.
Returns:
An np.array with shape (num_spectrogram_bins, num_mel_bins).
Raises:
ValueError: if frequency edges are incorrectly ordered."
7928,grappler_optimize,tensorflow/tensorflow/python/kernel_tests/signal/test_util.py,29,function,"Tries to optimize the provided graph using grappler.
Args:
graph: A `tf.Graph` instance containing the graph to optimize.
fetches: An optional list of `Tensor`s to fetch (i.e. not optimize away).
Grappler uses the 'train_op' collection to look for fetches, so if not
provided this collection should be non-empty.
config_proto: An optional `tf.compat.v1.ConfigProto` to use when rewriting
the graph.
Returns:
A `tf.compat.v1.GraphDef` containing the rewritten graph."
7929,tflite_convert,tensorflow/tensorflow/python/kernel_tests/signal/test_util.py,53,function,"Converts the provided fn to tf.lite model.
Args:
fn: A callable that expects a list of inputs like input_templates that
returns a tensor or structure of tensors.
input_templates: A list of Tensors, ndarrays or TensorSpecs describing the
inputs that fn expects. The actual values of the Tensors or ndarrays are
unused.
Returns:
The serialized tf.lite model."
7930,evaluate_tflite_model,tensorflow/tensorflow/python/kernel_tests/signal/test_util.py,72,function,"Evaluates the provided tf.lite model with the given input ndarrays.
Args:
tflite_model: bytes. The serialized tf.lite model.
input_ndarrays: A list of NumPy arrays to feed as input to the model.
Returns:
A list of ndarrays produced by the model.
Raises:
ValueError: If the number of input arrays does not match the number of
inputs the model expects."
7931,convert_data_format,tensorflow/tensorflow/python/layers/utils.py,26,function,
7932,normalize_tuple,tensorflow/tensorflow/python/layers/utils.py,49,function,"Transforms a single integer or iterable of integers into an integer tuple.
Arguments:
value: The value to validate and convert. Could an int, or any iterable
of ints.
n: The size of the tuple to be returned.
name: The name of the argument being validated, e.g. ""strides"" or
""kernel_size"". This is only used to format error messages.
Returns:
A tuple of n integers.
Raises:
ValueError: If something else than an int/long or iterable thereof was
passed."
7933,normalize_data_format,tensorflow/tensorflow/python/layers/utils.py,88,function,
7934,normalize_padding,tensorflow/tensorflow/python/layers/utils.py,97,function,
7935,conv_output_length,tensorflow/tensorflow/python/layers/utils.py,105,function,"Determines output length of a convolution given input length.
Arguments:
input_length: integer.
filter_size: integer.
padding: one of ""same"", ""valid"", ""full"".
stride: integer.
dilation: dilation rate, integer.
Returns:
The output length (integer)."
7936,conv_input_length,tensorflow/tensorflow/python/layers/utils.py,131,function,"Determines input length of a convolution given output length.
Arguments:
output_length: integer.
filter_size: integer.
padding: one of ""same"", ""valid"", ""full"".
stride: integer.
Returns:
The input length (integer)."
7937,deconv_output_length,tensorflow/tensorflow/python/layers/utils.py,155,function,"Determines output length of a transposed convolution given input length.
Arguments:
input_length: integer.
filter_size: integer.
padding: one of ""same"", ""valid"", ""full"".
stride: integer.
Returns:
The output length (integer)."
7938,smart_cond,tensorflow/tensorflow/python/layers/utils.py,177,function,"Return either `true_fn()` if predicate `pred` is true else `false_fn()`.
If `pred` is a bool or has a constant value, we return either `true_fn()`
or `false_fn()`, otherwise we use `tf.cond` to dynamically route to both.
Arguments:
pred: A scalar determining whether to return the result of `true_fn` or
`false_fn`.
true_fn: The callable to be performed if pred is true.
false_fn: The callable to be performed if pred is false.
name: Optional name prefix when using `tf.cond`.
Returns:
Tensors returned by the call to either `true_fn` or `false_fn`.
Raises:
TypeError: If `true_fn` or `false_fn` is not callable."
7939,constant_value,tensorflow/tensorflow/python/layers/utils.py,203,function,"Return the bool value for `pred`, or None if `pred` had a dynamic value.
Arguments:
pred: A scalar, either a Python bool or a TensorFlow boolean variable
or tensor, or the Python integer 1 or 0.
Returns:
True or False if `pred` has a constant boolean value, None otherwise.
Raises:
TypeError: If `pred` is not a Variable, Tensor or bool, or Python
integer 1 or 0."
7940,float_values,tensorflow/tensorflow/python/lib/core/bfloat16_test.py,35,function,Returns values that should round trip exactly to float and back.
7941,FileIO,tensorflow/tensorflow/python/lib/io/file_io.py,37,class,"FileIO class that exposes methods to read / write to / from files.
The constructor takes the following arguments:
name: [path-like object](https://docs.python.org/3/glossary.html#term-path-like-object)
giving the pathname of the file to be opened.
mode: one of `r`, `w`, `a`, `r+`, `w+`, `a+`. Append `b` for bytes mode.
Can be used as an iterator to iterate over lines in the file.
The default buffer size used for the BufferedInputStream used for reading
the file line by line is 1024 * 512 bytes."
7942,name,tensorflow/tensorflow/python/lib/io/file_io.py,65,method,Returns the file name.
7943,mode,tensorflow/tensorflow/python/lib/io/file_io.py,70,method,Returns the mode in which the file was opened.
7944,size,tensorflow/tensorflow/python/lib/io/file_io.py,96,method,Returns the size of the file.
7945,write,tensorflow/tensorflow/python/lib/io/file_io.py,100,method,Writes file_content to the file. Appends to the end of the file.
7946,read,tensorflow/tensorflow/python/lib/io/file_io.py,105,method,"Returns the contents of a file as a string.
Starts reading from current position in file.
Args:
n: Read `n` bytes if `n != -1`. If `n = -1`, reads to end of file.
Returns:
`n` bytes of the file (or whole file) in bytes mode or `n` bytes of the
string if in string (regular) mode."
7947,seek,tensorflow/tensorflow/python/lib/io/file_io.py,127,method,"Seeks to the offset in the file.
Args:
offset: The byte count relative to the whence argument.
whence: Valid values for whence are:
0: start of the file (default)
1: relative to the current position of the file
2: relative to the end of file. `offset` is usually negative."
7948,readline,tensorflow/tensorflow/python/lib/io/file_io.py,168,method,"Reads the next line, keeping \n. At EOF, returns ''."
7949,readlines,tensorflow/tensorflow/python/lib/io/file_io.py,173,method,Returns all lines from the file in a list.
7950,tell,tensorflow/tensorflow/python/lib/io/file_io.py,184,method,Returns the current position in the file.
7951,next,tensorflow/tensorflow/python/lib/io/file_io.py,211,method,
7952,flush,tensorflow/tensorflow/python/lib/io/file_io.py,214,method,"Flushes the Writable file.
This only ensures that the data has made its way out of the process without
any guarantees on whether it's written to disk. This means that the
data would survive an application crash but not necessarily an OS crash."
7953,close,tensorflow/tensorflow/python/lib/io/file_io.py,224,method,Closes FileIO. Should be called for the WritableFile to be flushed.
7954,seekable,tensorflow/tensorflow/python/lib/io/file_io.py,231,method,Returns True as FileIO supports random access ops of seek()/tell()
7955,file_exists,tensorflow/tensorflow/python/lib/io/file_io.py,237,function,"Determines whether a path exists or not.
Args:
filename: string, a path
Returns:
True if the path exists, whether it's a file or a directory.
False if the path does not exist and there are no filesystem errors.
Raises:
errors.OpError: Propagates any errors reported by the FileSystem API."
7956,file_exists_v2,tensorflow/tensorflow/python/lib/io/file_io.py,254,function,"Determines whether a path exists or not.
Args:
path: string, a path
Returns:
True if the path exists, whether it's a file or a directory.
False if the path does not exist and there are no filesystem errors.
Raises:
errors.OpError: Propagates any errors reported by the FileSystem API."
7957,delete_file,tensorflow/tensorflow/python/lib/io/file_io.py,275,function,"Deletes the file located at 'filename'.
Args:
filename: string, a filename
Raises:
errors.OpError: Propagates any errors reported by the FileSystem API. E.g.,
`NotFoundError` if the file does not exist."
7958,delete_file_v2,tensorflow/tensorflow/python/lib/io/file_io.py,289,function,"Deletes the path located at 'path'.
Args:
path: string, a path
Raises:
errors.OpError: Propagates any errors reported by the FileSystem API. E.g.,
`NotFoundError` if the path does not exist."
7959,read_file_to_string,tensorflow/tensorflow/python/lib/io/file_io.py,302,function,"Reads the entire contents of a file to a string.
Args:
filename: string, path to a file
binary_mode: whether to open the file in binary mode or not. This changes
the type of the object returned.
Returns:
contents of the file as a string or bytes.
Raises:
errors.OpError: Raises variety of errors that are subtypes e.g.
`NotFoundError` etc."
7960,write_string_to_file,tensorflow/tensorflow/python/lib/io/file_io.py,324,function,"Writes a string to a given file.
Args:
filename: string, path to a file
file_content: string, contents that need to be written to the file
Raises:
errors.OpError: If there are errors during the operation."
7961,get_matching_files,tensorflow/tensorflow/python/lib/io/file_io.py,339,function,"Returns a list of files that match the given pattern(s).
Args:
filename: string or iterable of strings. The glob pattern(s).
Returns:
A list of strings containing filenames that match the given pattern(s).
Raises:
* errors.OpError: If there are filesystem / directory listing errors."
7962,get_matching_files_v2,tensorflow/tensorflow/python/lib/io/file_io.py,355,function,"Returns a list of files that match the given pattern(s).
The patterns are defined as strings. Supported patterns are defined
here. Note that the pattern can be a Python iteratable of string patterns.
The format definition of the pattern is:
**pattern**: `{ term }`
**term**:
* `'*'`: matches any sequence of non-'/' characters
* `'?'`: matches a single non-'/' character
* `'[' [ '^' ] { match-list } ']'`: matches any single
character (not) on the list
* `c`: matches character `c` where `c != '*', '?', '\\', '['`
* `'\\' c`: matches character `c`
**character range**:
* `c`: matches character `c` while `c != '\\', '-', ']'`
* `'\\' c`: matches character `c`
* `lo '-' hi`: matches character `c` for `lo <= c <= hi`
Examples:
>>> tf.io.gfile.glob(""*.py"")
... # For example, ['__init__.py']
>>> tf.io.gfile.glob(""__init__.??"")
... # As above
>>> files = {""*.py""}
>>> the_iterator = iter(files)
>>> tf.io.gfile.glob(the_iterator)
... # As above
See the C++ function `GetMatchingPaths` in
[`core/platform/file_system.h`]
(../../../core/platform/file_system.h)
for implementation details.
Args:
pattern: string or iterable of strings. The glob pattern(s).
Returns:
A list of strings containing filenames that match the given pattern(s).
Raises:
errors.OpError: If there are filesystem / directory listing errors."
7963,create_dir,tensorflow/tensorflow/python/lib/io/file_io.py,423,function,"Creates a directory with the name `dirname`.
Args:
dirname: string, name of the directory to be created
Notes: The parent directories need to exist. Use `tf.io.gfile.makedirs`
instead if there is the possibility that the parent dirs don't exist.
Raises:
errors.OpError: If the operation fails."
7964,create_dir_v2,tensorflow/tensorflow/python/lib/io/file_io.py,439,function,"Creates a directory with the name given by `path`.
Args:
path: string, name of the directory to be created
Notes: The parent directories need to exist. Use `tf.io.gfile.makedirs`
instead if there is the possibility that the parent dirs don't exist.
Raises:
errors.OpError: If the operation fails."
7965,recursive_create_dir,tensorflow/tensorflow/python/lib/io/file_io.py,455,function,"Creates a directory and all parent/intermediate directories.
It succeeds if dirname already exists and is writable.
Args:
dirname: string, name of the directory to be created
Raises:
errors.OpError: If the operation fails."
7966,recursive_create_dir_v2,tensorflow/tensorflow/python/lib/io/file_io.py,470,function,"Creates a directory and all parent/intermediate directories.
It succeeds if path already exists and is writable.
Args:
path: string, name of the directory to be created
Raises:
errors.OpError: If the operation fails."
7967,copy,tensorflow/tensorflow/python/lib/io/file_io.py,485,function,"Copies data from `oldpath` to `newpath`.
Args:
oldpath: string, name of the file who's contents need to be copied
newpath: string, name of the file to which to copy to
overwrite: boolean, if false it's an error for `newpath` to be occupied by
an existing file.
Raises:
errors.OpError: If the operation fails."
7968,copy_v2,tensorflow/tensorflow/python/lib/io/file_io.py,501,function,"Copies data from `src` to `dst`.
Args:
src: string, name of the file whose contents need to be copied
dst: string, name of the file to which to copy to
overwrite: boolean, if false it's an error for `dst` to be occupied by an
existing file.
Raises:
errors.OpError: If the operation fails."
7969,rename,tensorflow/tensorflow/python/lib/io/file_io.py,518,function,"Rename or move a file / directory.
Args:
oldname: string, pathname for a file
newname: string, pathname to which the file needs to be moved
overwrite: boolean, if false it's an error for `newname` to be occupied by
an existing file.
Raises:
errors.OpError: If the operation fails."
7970,rename_v2,tensorflow/tensorflow/python/lib/io/file_io.py,534,function,"Rename or move a file / directory.
Args:
src: string, pathname for a file
dst: string, pathname to which the file needs to be moved
overwrite: boolean, if false it's an error for `dst` to be occupied by an
existing file.
Raises:
errors.OpError: If the operation fails."
7971,atomic_write_string_to_file,tensorflow/tensorflow/python/lib/io/file_io.py,550,function,"Writes to `filename` atomically.
This means that when `filename` appears in the filesystem, it will contain
all of `contents`. With write_string_to_file, it is possible for the file
to appear in the filesystem with `contents` only partially written.
Accomplished by writing to a temp file and then renaming it.
Args:
filename: string, pathname for a file
contents: string, contents that need to be written to the file
overwrite: boolean, if false it's an error for `filename` to be occupied by
an existing file."
7972,delete_recursively,tensorflow/tensorflow/python/lib/io/file_io.py,578,function,"Deletes everything under dirname recursively.
Args:
dirname: string, a path to a directory
Raises:
errors.OpError: If the operation fails."
7973,delete_recursively_v2,tensorflow/tensorflow/python/lib/io/file_io.py,591,function,"Deletes everything under path recursively.
Args:
path: string, a path
Raises:
errors.OpError: If the operation fails."
7974,is_directory,tensorflow/tensorflow/python/lib/io/file_io.py,604,function,"Returns whether the path is a directory or not.
Args:
dirname: string, path to a potential directory
Returns:
True, if the path is a directory; False otherwise"
7975,is_directory_v2,tensorflow/tensorflow/python/lib/io/file_io.py,617,function,"Returns whether the path is a directory or not.
Args:
path: string, path to a potential directory
Returns:
True, if the path is a directory; False otherwise"
7976,has_atomic_move,tensorflow/tensorflow/python/lib/io/file_io.py,632,function,"Checks whether the file system supports atomic moves.
Returns whether or not the file system of the given path supports the atomic
move operation for a file or folder. If atomic move is supported, it is
recommended to use a temp location for writing and then move to the final
location.
Args:
path: string, path to a file
Returns:
True, if the path is on a file system that supports atomic move
False, if the file system does not support atomic move. In such cases
we need to be careful about using moves. In some cases it is safer
not to use temporary locations in this case."
7977,list_directory,tensorflow/tensorflow/python/lib/io/file_io.py,657,function,"Returns a list of entries contained within a directory.
The list is in arbitrary order. It does not contain the special entries "".""
and "".."".
Args:
dirname: string, path to a directory
Returns:
[filename1, filename2, ... filenameN] as strings
Raises:
errors.NotFoundError if directory doesn't exist"
7978,list_directory_v2,tensorflow/tensorflow/python/lib/io/file_io.py,676,function,"Returns a list of entries contained within a directory.
The list is in arbitrary order. It does not contain the special entries "".""
and "".."".
Args:
path: string, path to a directory
Returns:
[filename1, filename2, ... filenameN] as strings
Raises:
errors.NotFoundError if directory doesn't exist"
7979,walk,tensorflow/tensorflow/python/lib/io/file_io.py,706,function,"Recursive directory tree generator for directories.
Args:
top: string, a Directory name
in_order: bool, Traverse in order if True, post order if False. Errors that
happen while listing directories are ignored.
Yields:
Each yield is a 3-tuple: the pathname of a directory, followed by lists of
all its subdirectories and leaf files. That is, each yield looks like:
`(dirname, [subdirname, subdirname, ...], [filename, filename, ...])`.
Each item is a string."
7980,walk_v2,tensorflow/tensorflow/python/lib/io/file_io.py,724,function,"Recursive directory tree generator for directories.
Args:
top: string, a Directory name
topdown: bool, Traverse pre order if True, post order if False.
onerror: optional handler for errors. Should be a function, it will be
called with the error as argument. Rethrowing the error aborts the walk.
Errors that happen while listing directories are ignored.
Yields:
Each yield is a 3-tuple: the pathname of a directory, followed by lists of
all its subdirectories and leaf files. That is, each yield looks like:
`(dirname, [subdirname, subdirname, ...], [filename, filename, ...])`.
Each item is a string."
7981,stat,tensorflow/tensorflow/python/lib/io/file_io.py,782,function,"Returns file statistics for a given path.
Args:
filename: string, path to a file
Returns:
FileStatistics struct that contains information about the path
Raises:
errors.OpError: If the operation fails."
7982,stat_v2,tensorflow/tensorflow/python/lib/io/file_io.py,798,function,"Returns file statistics for a given path.
Args:
path: string, path to a file
Returns:
FileStatistics struct that contains information about the path
Raises:
errors.OpError: If the operation fails."
7983,filecmp,tensorflow/tensorflow/python/lib/io/file_io.py,813,function,"Compare two files, returning True if they are the same, False otherwise.
We check size first and return False quickly if the files are different sizes.
If they are the same size, we continue to generating a crc for the whole file.
You might wonder: why not use Python's `filecmp.cmp()` instead? The answer is
that the builtin library is not robust to the many different filesystems
TensorFlow runs on, and so we here perform a similar comparison with
the more robust FileIO.
Args:
filename_a: string path to the first file.
filename_b: string path to the second file.
Returns:
True if the files are the same, False otherwise."
7984,file_crc32,tensorflow/tensorflow/python/lib/io/file_io.py,842,function,"Get the crc32 of the passed file.
The crc32 of a file can be used for error checking; two files with the same
crc32 are considered equivalent. Note that the entire file must be read
to produce the crc32.
Args:
filename: string, path to a file
block_size: Integer, process the files by reading blocks of `block_size`
bytes. Use -1 to read the file as once.
Returns:
hexadecimal as string, the crc32 of the passed file."
7985,PathLike,tensorflow/tensorflow/python/lib/io/file_io_test.py,33,class,Backport of pathlib.Path for Python < 3.6
7986,TFRecordCompressionType,tensorflow/tensorflow/python/lib/io/tf_record.py,32,class,The type of compression for the record.
7987,TFRecordOptions,tensorflow/tensorflow/python/lib/io/tf_record.py,43,class,Options used for manipulating TFRecord files.
7988,get_compression_type_string,tensorflow/tensorflow/python/lib/io/tf_record.py,102,method,"Convert various option types to a unified string.
Args:
options: `TFRecordOption`, `TFRecordCompressionType`, or string.
Returns:
Compression type as string (e.g. `'ZLIB'`, `'GZIP'`, or `''`).
Raises:
ValueError: If compression_type is invalid."
7989,tf_record_iterator,tensorflow/tensorflow/python/lib/io/tf_record.py,157,function,"An iterator that read the records from a TFRecords file.
Args:
path: The path to the TFRecords file.
options: (optional) A TFRecordOptions object.
Returns:
An iterator of serialized TFRecords.
Raises:
IOError: If `path` cannot be opened for reading."
7990,tf_record_random_reader,tensorflow/tensorflow/python/lib/io/tf_record.py,174,function,"Creates a reader that allows random-access reads from a TFRecords file.
The created reader object has the following method:
- `read(offset)`, which returns a tuple of `(record, ending_offset)`, where
`record` is the TFRecord read at the offset, and
`ending_offset` is the ending offset of the read record.
The method throws a `tf.errors.DataLossError` if data is corrupted at
the given offset. The method throws `IndexError` if the offset is out of
range for the TFRecords file.
Usage example:
```py
reader = tf_record_random_reader(file_path)
record_1, offset_1 = reader.read(0) # 0 is the initial offset.
# offset_1 is the ending offset of the 1st record and the starting offset of
# the next.
record_2, offset_2 = reader.read(offset_1)
# offset_2 is the ending offset of the 2nd record and the starting offset of
# the next.
# We can jump back and read the first record again if so desired.
reader.read(0)
```
Args:
path: The path to the TFRecords file.
Returns:
An object that supports random-access reading of the serialized TFRecords.
Raises:
IOError: If `path` cannot be opened for reading."
7991,TFRecordWriter,tensorflow/tensorflow/python/lib/io/tf_record.py,218,class,"A class to write records to a TFRecords file.
[TFRecords tutorial](https://www.tensorflow.org/tutorials/load_data/tfrecord)
TFRecords is a binary format which is optimized for high throughput data
retrieval, generally in conjunction with `tf.data`. `TFRecordWriter` is used
to write serialized examples to a file for later consumption. The key steps
are:
Ahead of time:
- [Convert data into a serialized format](
https://www.tensorflow.org/tutorials/load_data/tfrecord#tfexample)
- [Write the serialized data to one or more files](
https://www.tensorflow.org/tutorials/load_data/tfrecord#tfrecord_files_in_python)
During training or evaluation:
- [Read serialized examples into memory](
https://www.tensorflow.org/tutorials/load_data/tfrecord#reading_a_tfrecord_file)
- [Parse (deserialize) examples](
https://www.tensorflow.org/tutorials/load_data/tfrecord#reading_a_tfrecord_file)
A minimal example is given below:
>>> import tempfile
>>> example_path = os.path.join(tempfile.gettempdir(), ""example.tfrecords"")
>>> np.random.seed(0)
>>> # Write the records to a file.
... with tf.io.TFRecordWriter(example_path) as file_writer:
... for _ in range(4):
... x, y = np.random.random(), np.random.random()
...
... record_bytes = tf.train.Example(features=tf.train.Features(feature={
... ""x"": tf.train.Feature(float_list=tf.train.FloatList(value=[x])),
... ""y"": tf.train.Feature(float_list=tf.train.FloatList(value=[y])),
... })).SerializeToString()
... file_writer.write(record_bytes)
>>> # Read the data back out.
>>> def decode_fn(record_bytes):
... return tf.io.parse_single_example(
... # Data
... record_bytes,
...
... # Schema
... {""x"": tf.io.FixedLenFeature([], dtype=tf.float32),
... ""y"": tf.io.FixedLenFeature([], dtype=tf.float32)}
... )
>>> for batch in tf.data.TFRecordDataset([example_path]).map(decode_fn):
... print(""x = {x:.4f}, y = {y:.4f}"".format(**batch))
x = 0.5488, y = 0.7152
x = 0.6028, y = 0.5449
x = 0.4237, y = 0.6459
x = 0.4376, y = 0.8918
This class implements `__enter__` and `__exit__`, and can be used
in `with` blocks like a normal file. (See the usage example above.)"
7992,write,tensorflow/tensorflow/python/lib/io/tf_record.py,307,method,"Write a string record to the file.
Args:
record: str"
7993,flush,tensorflow/tensorflow/python/lib/io/tf_record.py,315,method,Flush the file.
7994,close,tensorflow/tensorflow/python/lib/io/tf_record.py,319,method,Close the file.
7995,Module,tensorflow/tensorflow/python/module/module.py,35,class,"Base neural network module class.
A module is a named container for `tf.Variable`s, other `tf.Module`s and
functions which apply to user input. For example a dense layer in a neural
network might be implemented as a `tf.Module`:
>>> class Dense(tf.Module):
... def __init__(self, input_dim, output_size, name=None):
... super(Dense, self).__init__(name=name)
... self.w = tf.Variable(
... tf.random.normal([input_dim, output_size]), name='w')
... self.b = tf.Variable(tf.zeros([output_size]), name='b')
... def __call__(self, x):
... y = tf.matmul(x, self.w) + self.b
... return tf.nn.relu(y)
You can use the Dense layer as you would expect:
>>> d = Dense(input_dim=3, output_size=2)
>>> d(tf.ones([1, 3]))
<tf.Tensor: shape=(1, 2), dtype=float32, numpy=..., dtype=float32)>
By subclassing `tf.Module` instead of `object` any `tf.Variable` or
`tf.Module` instances assigned to object properties can be collected using
the `variables`, `trainable_variables` or `submodules` property:
>>> d.variables
(<tf.Variable 'b:0' shape=(2,) dtype=float32, numpy=...,
dtype=float32)>,
<tf.Variable 'w:0' shape=(3, 2) dtype=float32, numpy=..., dtype=float32)>)
Subclasses of `tf.Module` can also take advantage of the `_flatten` method
which can be used to implement tracking of any other types.
All `tf.Module` classes have an associated `tf.name_scope` which can be used
to group operations in TensorBoard and create hierarchies for variable names
which can help with debugging. We suggest using the name scope when creating
nested submodules/parameters or for forward methods whose graph you might want
to inspect in TensorBoard. You can enter the name scope explicitly using
`with self.name_scope:` or you can annotate methods (apart from `__init__`)
with `@tf.Module.with_name_scope`.
>>> class MLP(tf.Module):
... def __init__(self, input_size, sizes, name=None):
... super(MLP, self).__init__(name=name)
... self.layers = []
... with self.name_scope:
... for size in sizes:
... self.layers.append(Dense(input_dim=input_size, output_size=size))
... input_size = size
... @tf.Module.with_name_scope
... def __call__(self, x):
... for layer in self.layers:
... x = layer(x)
... return x
>>> module = MLP(input_size=5, sizes=[5, 5])
>>> module.variables
(<tf.Variable 'mlp/b:0' shape=(5,) dtype=float32, numpy=..., dtype=float32)>,
<tf.Variable 'mlp/w:0' shape=(5, 5) dtype=float32, numpy=...,
dtype=float32)>,
<tf.Variable 'mlp/b:0' shape=(5,) dtype=float32, numpy=..., dtype=float32)>,
<tf.Variable 'mlp/w:0' shape=(5, 5) dtype=float32, numpy=...,
dtype=float32)>)"
7996,name,tensorflow/tensorflow/python/module/module.py,130,method,"Returns the name of this module as passed or determined in the ctor.
NOTE: This is not the same as the `self.name_scope.name` which includes
parent module names."
7997,name_scope,tensorflow/tensorflow/python/module/module.py,139,method,Returns a `tf.name_scope` instance for this class.
7998,variables,tensorflow/tensorflow/python/module/module.py,148,method,"Sequence of variables owned by this module and its submodules.
Note: this method uses reflection to find variables on the current instance
and submodules. For performance reasons you may wish to cache the result
of calling this method if you don't expect the return value to change.
Returns:
A sequence of variables for the current module (sorted by attribute
name) followed by variables from all submodules recursively (breadth
first)."
7999,trainable_variables,tensorflow/tensorflow/python/module/module.py,163,method,"Sequence of trainable variables owned by this module and its submodules.
Note: this method uses reflection to find variables on the current instance
and submodules. For performance reasons you may wish to cache the result
of calling this method if you don't expect the return value to change.
Returns:
A sequence of variables for the current module (sorted by attribute
name) followed by variables from all submodules recursively (breadth
first)."
8000,submodules,tensorflow/tensorflow/python/module/module.py,178,method,"Sequence of all sub-modules.
Submodules are modules which are properties of this module, or found as
properties of modules which are properties of this module (and so on).
>>> a = tf.Module()
>>> b = tf.Module()
>>> c = tf.Module()
>>> a.b = b
>>> b.c = c
>>> list(a.submodules) == [b, c]
True
>>> list(b.submodules) == [c]
True
>>> list(c.submodules) == []
True
Returns:
A sequence of all submodules."
8001,with_name_scope,tensorflow/tensorflow/python/module/module.py,267,method,"Decorator to automatically enter the module name scope.
>>> class MyModule(tf.Module):
... @tf.Module.with_name_scope
... def __call__(self, x):
... if not hasattr(self, 'w'):
... self.w = tf.Variable(tf.random.normal([x.shape[1], 3]))
... return tf.matmul(x, self.w)
Using the above module would produce `tf.Variable`s and `tf.Tensor`s whose
names included the module name:
>>> mod = MyModule()
>>> mod(tf.ones([1, 2]))
<tf.Tensor: shape=(1, 3), dtype=float32, numpy=..., dtype=float32)>
>>> mod.w
<tf.Variable 'my_module/Variable:0' shape=(2, 3) dtype=float32,
numpy=..., dtype=float32)>
Args:
method: The method to wrap.
Returns:
The original method wrapped such that it enters the module's name scope."
8002,method_with_name_scope,tensorflow/tensorflow/python/module/module.py,293,method,
8003,valid_identifier,tensorflow/tensorflow/python/module/module.py,315,function,
8004,camel_to_snake,tensorflow/tensorflow/python/module/module.py,319,function,
8005,get_name_scope,tensorflow/tensorflow/python/module/module_test.py,317,function,
8006,ErrorModuleError,tensorflow/tensorflow/python/module/module_test.py,323,class,
8007,ErrorModule,tensorflow/tensorflow/python/module/module_test.py,327,class,
8008,RecursiveModule,tensorflow/tensorflow/python/module/module_test.py,339,class,
8009,AbstractModule,tensorflow/tensorflow/python/module/module_test.py,351,class,
8010,ConcreteModule,tensorflow/tensorflow/python/module/module_test.py,358,class,
8011,TreeModule,tensorflow/tensorflow/python/module/module_test.py,365,class,
8012,new_leaf,tensorflow/tensorflow/python/module/module_test.py,372,method,
8013,ReturnsNameScopeModule,tensorflow/tensorflow/python/module/module_test.py,378,class,
8014,alternative_forward,tensorflow/tensorflow/python/module/module_test.py,381,method,
8015,SubclassedReturnsNameScopeModule,tensorflow/tensorflow/python/module/module_test.py,389,class,
8016,alternative_alternative_forward,tensorflow/tensorflow/python/module/module_test.py,392,method,
8017,PropertyThrowsWhenCalledModule,tensorflow/tensorflow/python/module/module_test.py,396,class,
8018,raise_assertion_error,tensorflow/tensorflow/python/module/module_test.py,399,method,
8019,ModuleOverridingNameScope,tensorflow/tensorflow/python/module/module_test.py,403,class,
8020,name_scope,tensorflow/tensorflow/python/module/module_test.py,406,method,
8021,ModuleWithFunctionAnnotatedCall,tensorflow/tensorflow/python/module/module_test.py,410,class,
8022,forward,tensorflow/tensorflow/python/module/module_test.py,414,method,
8023,forward_ag,tensorflow/tensorflow/python/module/module_test.py,419,method,
8024,PropertyModule,tensorflow/tensorflow/python/module/module_test.py,423,class,
8025,some_property,tensorflow/tensorflow/python/module/module_test.py,431,method,
8026,some_property,tensorflow/tensorflow/python/module/module_test.py,437,method,
8027,no_name_scope_property,tensorflow/tensorflow/python/module/module_test.py,441,method,
8028,no_name_scope_property,tensorflow/tensorflow/python/module/module_test.py,446,method,
8029,LayerModule,tensorflow/tensorflow/python/module/module_test.py,523,class,
8030,variables,tensorflow/tensorflow/python/module/module_test.py,538,method,
8031,key_function,tensorflow/tensorflow/python/module/module_test.py,539,method,
8032,MemberType,tensorflow/tensorflow/python/module/module_test.py,549,class,A simple type to search for.
8033,SimpleModule,tensorflow/tensorflow/python/module/module_test.py,554,class,
8034,AccumulateNBenchmark,tensorflow/tensorflow/python/ops/accumulate_n_benchmark.py,39,class,
8035,benchmarkAccumulateN,tensorflow/tensorflow/python/ops/accumulate_n_benchmark.py,130,method,
8036,reshape,tensorflow/tensorflow/python/ops/array_ops.py,61,function,"Reshapes a tensor.
Given `tensor`, this operation returns a new `tf.Tensor` that has the same
values as `tensor` in the same order, except with a new shape given by
`shape`.
>>> t1 = [[1, 2, 3],
... [4, 5, 6]]
>>> print(tf.shape(t1).numpy())
[2 3]
>>> t2 = tf.reshape(t1, [6])
>>> t2
<tf.Tensor: shape=(6,), dtype=int32,
numpy=array([1, 2, 3, 4, 5, 6], dtype=int32)>
>>> tf.reshape(t2, [3, 2])
<tf.Tensor: shape=(3, 2), dtype=int32, numpy=
array([[1, 2],
[3, 4],
[5, 6]], dtype=int32)>
The `tf.reshape` does not change the order of or the total number of elements
in the tensor, and so it can reuse the underlying data buffer. This makes it
a fast operation independent of how big of a tensor it is operating on.
>>> tf.reshape([1, 2, 3], [2, 2])
Traceback (most recent call last):
...
InvalidArgumentError: Input to reshape is a tensor with 3 values, but the
requested shape has 4
To instead reorder the data to rearrange the dimensions of a tensor, see
`tf.transpose`.
>>> t = [[1, 2, 3],
... [4, 5, 6]]
>>> tf.reshape(t, [3, 2]).numpy()
array([[1, 2],
[3, 4],
[5, 6]], dtype=int32)
>>> tf.transpose(t, perm=[1, 0]).numpy()
array([[1, 4],
[2, 5],
[3, 6]], dtype=int32)
If one component of `shape` is the special value -1, the size of that
dimension is computed so that the total size remains constant. In particular,
a `shape` of `[-1]` flattens into 1-D. At most one component of `shape` can
be -1.
>>> t = [[1, 2, 3],
... [4, 5, 6]]
>>> tf.reshape(t, [-1])
<tf.Tensor: shape=(6,), dtype=int32,
numpy=array([1, 2, 3, 4, 5, 6], dtype=int32)>
>>> tf.reshape(t, [3, -1])
<tf.Tensor: shape=(3, 2), dtype=int32, numpy=
array([[1, 2],
[3, 4],
[5, 6]], dtype=int32)>
>>> tf.reshape(t, [-1, 2])
<tf.Tensor: shape=(3, 2), dtype=int32, numpy=
array([[1, 2],
[3, 4],
[5, 6]], dtype=int32)>
`tf.reshape(t, [])` reshapes a tensor `t` with one element to a scalar.
>>> tf.reshape([7], []).numpy()
7
More examples:
>>> t = [1, 2, 3, 4, 5, 6, 7, 8, 9]
>>> print(tf.shape(t).numpy())
[9]
>>> tf.reshape(t, [3, 3])
<tf.Tensor: shape=(3, 3), dtype=int32, numpy=
array([[1, 2, 3],
[4, 5, 6],
[7, 8, 9]], dtype=int32)>
>>> t = [[[1, 1], [2, 2]],
... [[3, 3], [4, 4]]]
>>> print(tf.shape(t).numpy())
[2 2 2]
>>> tf.reshape(t, [2, 4])
<tf.Tensor: shape=(2, 4), dtype=int32, numpy=
array([[1, 1, 2, 2],
[3, 3, 4, 4]], dtype=int32)>
>>> t = [[[1, 1, 1],
... [2, 2, 2]],
... [[3, 3, 3],
... [4, 4, 4]],
... [[5, 5, 5],
... [6, 6, 6]]]
>>> print(tf.shape(t).numpy())
[3 2 3]
>>> # Pass '[-1]' to flatten 't'.
>>> tf.reshape(t, [-1])
<tf.Tensor: shape=(18,), dtype=int32,
numpy=array([1, 1, 1, 2, 2, 2, 3, 3, 3, 4, 4, 4, 5, 5, 5, 6, 6, 6],
dtype=int32)>
>>> # -- Using -1 to infer the shape --
>>> # Here -1 is inferred to be 9:
>>> tf.reshape(t, [2, -1])
<tf.Tensor: shape=(2, 9), dtype=int32, numpy=
array([[1, 1, 1, 2, 2, 2, 3, 3, 3],
[4, 4, 4, 5, 5, 5, 6, 6, 6]], dtype=int32)>
>>> # -1 is inferred to be 2:
>>> tf.reshape(t, [-1, 9])
<tf.Tensor: shape=(2, 9), dtype=int32, numpy=
array([[1, 1, 1, 2, 2, 2, 3, 3, 3],
[4, 4, 4, 5, 5, 5, 6, 6, 6]], dtype=int32)>
>>> # -1 is inferred to be 3:
>>> tf.reshape(t, [ 2, -1, 3])
<tf.Tensor: shape=(2, 3, 3), dtype=int32, numpy=
array([[[1, 1, 1],
[2, 2, 2],
[3, 3, 3]],
[[4, 4, 4],
[5, 5, 5],
[6, 6, 6]]], dtype=int32)>
Args:
tensor: A `Tensor`.
shape: A `Tensor`. Must be one of the following types: `int32`, `int64`.
Defines the shape of the output tensor.
name: Optional string. A name for the operation.
Returns:
A `Tensor`. Has the same type as `tensor`."
8037,fill,tensorflow/tensorflow/python/ops/array_ops.py,202,function,"Creates a tensor filled with a scalar value.
See also `tf.ones`, `tf.zeros`, `tf.one_hot`, `tf.eye`.
This operation creates a tensor of shape `dims` and fills it with `value`.
For example:
>>> tf.fill([2, 3], 9)
<tf.Tensor: shape=(2, 3), dtype=int32, numpy=
array([[9, 9, 9],
[9, 9, 9]], dtype=int32)>
`tf.fill` evaluates at graph runtime and supports dynamic shapes based on
other runtime `tf.Tensors`, unlike `tf.constant(value, shape=dims)`, which
embeds the value as a `Const` node.
Args:
dims: A 1-D sequence of non-negative numbers. Represents the shape of the
output `tf.Tensor`. Entries should be of type: `int32`, `int64`.
value: A value to fill the returned `tf.Tensor`.
name: Optional string. The name of the output `tf.Tensor`.
Returns:
A `tf.Tensor` with shape `dims` and the same dtype as `value`.
Raises:
InvalidArgumentError: `dims` contains negative entries.
NotFoundError: `dims` contains non-integer entries.
@compatibility(numpy)
Similar to `np.full`. In `numpy`, more parameters are supported. Passing a
number argument as the shape (`np.full(5, value)`) is valid in `numpy` for
specifying a 1-D shaped result, while TensorFlow does not support this syntax.
@end_compatibility"
8038,identity,tensorflow/tensorflow/python/ops/array_ops.py,246,function,"Return a Tensor with the same shape and contents as input.
The return value is not the same Tensor as the original, but contains the same
values. This operation is fast when used on the same device.
For example:
>>> a = tf.constant([0.78])
>>> a_identity = tf.identity(a)
>>> a.numpy()
array([0.78], dtype=float32)
>>> a_identity.numpy()
array([0.78], dtype=float32)
Calling `tf.identity` on a variable will make a Tensor that represents the
value of that variable at the time it is called. This is equivalent to calling
`<variable>.read_value()`.
>>> a = tf.Variable(5)
>>> a_identity = tf.identity(a)
>>> a.assign_add(1)
<tf.Variable ... shape=() dtype=int32, numpy=6>
>>> a.numpy()
6
>>> a_identity.numpy()
5
Args:
input: A `Tensor`.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `input`."
8039,expand_dims,tensorflow/tensorflow/python/ops/array_ops.py,298,function,"Returns a tensor with a length 1 axis inserted at index `axis`.
Given a tensor `input`, this operation inserts a dimension of length 1 at the
dimension index `axis` of `input`'s shape. The dimension index follows Python
indexing rules: It's zero-based, a negative index it is counted backward
from the end.
This operation is useful to:
* Add an outer ""batch"" dimension to a single element.
* Align axes for broadcasting.
* To add an inner vector length axis to a tensor of scalars.
For example:
If you have a single image of shape `[height, width, channels]`:
>>> image = tf.zeros([10,10,3])
You can add an outer `batch` axis by passing `axis=0`:
>>> tf.expand_dims(image, axis=0).shape.as_list()
[1, 10, 10, 3]
The new axis location matches Python `list.insert(axis, 1)`:
>>> tf.expand_dims(image, axis=1).shape.as_list()
[10, 1, 10, 3]
Following standard Python indexing rules, a negative `axis` counts from the
end so `axis=-1` adds an inner most dimension:
>>> tf.expand_dims(image, -1).shape.as_list()
[10, 10, 3, 1]
This operation requires that `axis` is a valid index for `input.shape`,
following Python indexing rules:
```
-1-tf.rank(input) <= axis <= tf.rank(input)
```
This operation is related to:
* `tf.squeeze`, which removes dimensions of size 1.
* `tf.reshape`, which provides more flexible reshaping capability.
* `tf.sparse.expand_dims`, which provides this functionality for
`tf.SparseTensor`
Args:
input: A `Tensor`.
axis: 0-D (scalar). Specifies the dimension index at which to expand the
shape of `input`. Must be in the range `[-rank(input) - 1, rank(input)]`.
name: The name of the output `Tensor` (optional).
dim: 0-D (scalar). Equivalent to `axis`, to be deprecated.
Returns:
A `Tensor` with the same data as `input`, but its shape has an additional
dimension of size 1 added.
Raises:
ValueError: if either both or neither of `dim` and `axis` are specified."
8040,expand_dims_v2,tensorflow/tensorflow/python/ops/array_ops.py,370,function,"Returns a tensor with a length 1 axis inserted at index `axis`.
Given a tensor `input`, this operation inserts a dimension of length 1 at the
dimension index `axis` of `input`'s shape. The dimension index follows Python
indexing rules: It's zero-based, a negative index it is counted backward
from the end.
This operation is useful to:
* Add an outer ""batch"" dimension to a single element.
* Align axes for broadcasting.
* To add an inner vector length axis to a tensor of scalars.
For example:
If you have a single image of shape `[height, width, channels]`:
>>> image = tf.zeros([10,10,3])
You can add an outer `batch` axis by passing `axis=0`:
>>> tf.expand_dims(image, axis=0).shape.as_list()
[1, 10, 10, 3]
The new axis location matches Python `list.insert(axis, 1)`:
>>> tf.expand_dims(image, axis=1).shape.as_list()
[10, 1, 10, 3]
Following standard Python indexing rules, a negative `axis` counts from the
end so `axis=-1` adds an inner most dimension:
>>> tf.expand_dims(image, -1).shape.as_list()
[10, 10, 3, 1]
This operation requires that `axis` is a valid index for `input.shape`,
following Python indexing rules:
```
-1-tf.rank(input) <= axis <= tf.rank(input)
```
This operation is related to:
* `tf.squeeze`, which removes dimensions of size 1.
* `tf.reshape`, which provides more flexible reshaping capability.
* `tf.sparse.expand_dims`, which provides this functionality for
`tf.SparseTensor`
Args:
input: A `Tensor`.
axis: Integer specifying the dimension index at which to expand the
shape of `input`. Given an input of D dimensions, `axis` must be in range
`[-(D+1), D]` (inclusive).
name: Optional string. The name of the output `Tensor`.
Returns:
A tensor with the same data as `input`, with an additional dimension
inserted at the index specified by `axis`.
Raises:
ValueError: If `axis` is not specified.
InvalidArgumentError: If `axis` is out of range `[-(D+1), D]`."
8041,listdiff,tensorflow/tensorflow/python/ops/array_ops.py,446,function,
8042,setdiff1d,tensorflow/tensorflow/python/ops/array_ops.py,461,function,"Computes the difference between two lists of numbers or strings.
Given a list x and a list y, this operation returns a list out that
represents all values that are in x but not in y. The returned list
out is sorted in the same order that the numbers appear in x
(duplicates are preserved). This operation also returns a list idx
that represents the position of each out element in x.
In other words:
```python
out[i] = x[idx[i]] for i in [0, 1, ..., len(out) - 1]
```
Example usage:
>>> x = [1, 2, 3, 4, 5, 6]
>>> y = [1, 3, 5]
>>> setdiff1d(x,y)
ListDiff(out=<tf.Tensor: id=2, shape=(3,), dtype=int32,
numpy=array([2, 4, 6], dtype=int32)>, idx=<tf.Tensor: id=3,
shape=(3,), dtype=int32, numpy=array([1, 3, 5], dtype=int32)>)
Args:
x: A Tensor. 1-D. Values to keep.
y: A Tensor. Must have the same type as x. 1-D. Values to remove.
out_idx: An optional tf.DType from: tf.int32, tf.int64. Defaults to
tf.int32.
name: A name for the operation (optional).
Returns:
A tuple of Tensor objects (out, idx).
out: A Tensor. Has the same type as x.
idx: A Tensor of type out_idx."
8043,broadcast_dynamic_shape,tensorflow/tensorflow/python/ops/array_ops.py,505,function,"Computes the shape of a broadcast given symbolic shapes.
When `shape_x` and `shape_y` are Tensors representing shapes (i.e. the result
of calling tf.shape on another Tensor) this computes a Tensor which is the
shape of the result of a broadcasting op applied in tensors of shapes
`shape_x` and `shape_y`.
This is useful when validating the result of a broadcasting operation when the
tensors do not have statically known shapes.
Example:
>>> shape_x = (1, 2, 3)
>>> shape_y = (5, 1, 3)
>>> tf.broadcast_dynamic_shape(shape_x, shape_y)
<tf.Tensor: shape=(3,), dtype=int32, numpy=array([5, 2, 3], ...>
Args:
shape_x: A rank 1 integer `Tensor`, representing the shape of x.
shape_y: A rank 1 integer `Tensor`, representing the shape of y.
Returns:
A rank 1 integer `Tensor` representing the broadcasted shape.
Raises:
InvalidArgumentError: If the two shapes are incompatible for
broadcasting."
8044,broadcast_static_shape,tensorflow/tensorflow/python/ops/array_ops.py,539,function,"Computes the shape of a broadcast given known shapes.
When `shape_x` and `shape_y` are fully known `TensorShape`s this computes a
`TensorShape` which is the shape of the result of a broadcasting op applied in
tensors of shapes `shape_x` and `shape_y`.
For example, if shape_x is `TensorShape([1, 2, 3])` and shape_y is
`TensorShape([5, 1, 3])`, the result is a TensorShape whose value is
`TensorShape([5, 2, 3])`.
This is useful when validating the result of a broadcasting operation when the
tensors have statically known shapes.
Example:
>>> shape_x = tf.TensorShape([1, 2, 3])
>>> shape_y = tf.TensorShape([5, 1 ,3])
>>> tf.broadcast_static_shape(shape_x, shape_y)
TensorShape([5, 2, 3])
Args:
shape_x: A `TensorShape`
shape_y: A `TensorShape`
Returns:
A `TensorShape` representing the broadcasted shape.
Raises:
ValueError: If the two shapes can not be broadcasted."
8045,shape_v2,tensorflow/tensorflow/python/ops/array_ops.py,575,function,"Returns the shape of a tensor.
See also `tf.size`, `tf.rank`.
`tf.shape` returns a 1-D integer tensor representing the shape of `input`.
For example:
>>> t = tf.constant([[[1, 1, 1], [2, 2, 2]], [[3, 3, 3], [4, 4, 4]]])
>>> tf.shape(t)
<tf.Tensor: shape=(3,), dtype=int32, numpy=array([2, 2, 3], dtype=int32)>
Note: When using symbolic tensors, such as when using the Keras API,
tf.shape() will return the shape of the symbolic tensor.
>>> a = tf.keras.layers.Input((None, 10))
>>> tf.shape(a)
<... shape=(3,) dtype=int32...>
In these cases, using `tf.Tensor.shape` will return more informative results.
>>> a.shape
TensorShape([None, None, 10])
(The first `None` represents the as yet unknown batch size.)
`tf.shape` and `Tensor.shape` should be identical in eager mode. Within
`tf.function` or within a `compat.v1` context, not all dimensions may be
known until execution time. Hence when defining custom layers and models
for graph mode, prefer the dynamic `tf.shape(x)` over the static `x.shape`.
Args:
input: A `Tensor` or `SparseTensor`.
out_type: (Optional) The specified output type of the operation (`int32` or
`int64`). Defaults to `tf.int32`.
name: A name for the operation (optional).
Returns:
A `Tensor` of type `out_type`."
8046,shape,tensorflow/tensorflow/python/ops/array_ops.py,622,function,"Returns the shape of a tensor.
This operation returns a 1-D integer tensor representing the shape of `input`.
For example:
```python
t = tf.constant([[[1, 1, 1], [2, 2, 2]], [[3, 3, 3], [4, 4, 4]]])
tf.shape(t) # [2, 2, 3]
```
Args:
input: A `Tensor` or `SparseTensor`.
name: A name for the operation (optional).
out_type: (Optional) The specified output type of the operation (`int32`
or `int64`). Defaults to `tf.int32`.
Returns:
A `Tensor` of type `out_type`."
8047,shape_internal,tensorflow/tensorflow/python/ops/array_ops.py,647,function,"Returns the shape of a tensor.
Args:
input: A `Tensor` or `SparseTensor`.
name: A name for the operation (optional).
optimize: if true, encode the shape as a constant when possible.
out_type: (Optional) The specified output type of the operation (`int32` or
`int64`). Defaults to tf.int32.
Returns:
A `Tensor` of type `out_type`."
8048,shape_n,tensorflow/tensorflow/python/ops/array_ops.py,677,function,"Returns shape of tensors.
Args:
input: A list of at least 1 `Tensor` object with the same type.
out_type: The specified output type of the operation (`int32` or `int64`).
Defaults to `tf.int32`(optional).
name: A name for the operation (optional).
Returns:
A list with the same length as `input` of `Tensor` objects with
type `out_type`."
8049,size_v2,tensorflow/tensorflow/python/ops/array_ops.py,697,function,"Returns the size of a tensor.
See also `tf.shape`.
Returns a 0-D `Tensor` representing the number of elements in `input`
of type `out_type`. Defaults to tf.int32.
For example:
>>> t = tf.constant([[[1, 1, 1], [2, 2, 2]], [[3, 3, 3], [4, 4, 4]]])
>>> tf.size(t)
<tf.Tensor: shape=(), dtype=int32, numpy=12>
Args:
input: A `Tensor` or `SparseTensor`.
name: A name for the operation (optional).
out_type: (Optional) The specified non-quantized numeric output type of the
operation. Defaults to `tf.int32`.
Returns:
A `Tensor` of type `out_type`. Defaults to `tf.int32`.
@compatibility(numpy)
Equivalent to np.size()
@end_compatibility"
8050,size,tensorflow/tensorflow/python/ops/array_ops.py,731,function,"Returns the size of a tensor.
Returns a 0-D `Tensor` representing the number of elements in `input`
of type `out_type`. Defaults to tf.int32.
For example:
```python
t = tf.constant([[[1, 1, 1], [2, 2, 2]], [[3, 3, 3], [4, 4, 4]]])
tf.size(t) # 12
```
Args:
input: A `Tensor` or `SparseTensor`.
name: A name for the operation (optional).
out_type: (Optional) The specified non-quantized numeric output type of the
operation. Defaults to `tf.int32`.
Returns:
A `Tensor` of type `out_type`. Defaults to `tf.int32`.
@compatibility(numpy)
Equivalent to np.size()
@end_compatibility"
8051,size_internal,tensorflow/tensorflow/python/ops/array_ops.py,761,function,"Returns the size of a tensor.
Args:
input: A `Tensor` or `SparseTensor`.
name: A name for the operation (optional).
optimize: if true, encode the size as a constant when possible.
out_type: (Optional) The specified non-quantized numeric output type of the
operation. Defaults to `tf.int32`.
Returns:
A `Tensor` of type `out_type`. Defaults to `tf.int32`."
8052,rank,tensorflow/tensorflow/python/ops/array_ops.py,801,function,"Returns the rank of a tensor.
See also `tf.shape`.
Returns a 0-D `int32` `Tensor` representing the rank of `input`.
For example:
```python
# shape of tensor 't' is [2, 2, 3]
t = tf.constant([[[1, 1, 1], [2, 2, 2]], [[3, 3, 3], [4, 4, 4]]])
tf.rank(t) # 3
```
**Note**: The rank of a tensor is not the same as the rank of a matrix. The
rank of a tensor is the number of indices required to uniquely select each
element of the tensor. Rank is also known as ""order"", ""degree"", or ""ndims.""
Args:
input: A `Tensor` or `SparseTensor`.
name: A name for the operation (optional).
Returns:
A `Tensor` of type `int32`.
@compatibility(numpy)
Equivalent to np.ndim
@end_compatibility"
8053,rank_internal,tensorflow/tensorflow/python/ops/array_ops.py,835,function,"Returns the rank of a tensor.
Args:
input: A `Tensor` or `SparseTensor`.
name: A name for the operation (optional).
optimize: if true, encode the rank as a constant when possible.
Returns:
A `Tensor` of type `int32`."
8054,slice,tensorflow/tensorflow/python/ops/array_ops.py,1048,function,"Extracts a slice from a tensor.
See also `tf.strided_slice`.
This operation extracts a slice of size `size` from a tensor `input_` starting
at the location specified by `begin`. The slice `size` is represented as a
tensor shape, where `size[i]` is the number of elements of the 'i'th dimension
of `input_` that you want to slice. The starting location (`begin`) for the
slice is represented as an offset in each dimension of `input_`. In other
words, `begin[i]` is the offset into the i'th dimension of `input_` that you
want to slice from.
Note that `tf.Tensor.__getitem__` is typically a more pythonic way to
perform slices, as it allows you to write `foo[3:7, :-2]` instead of
`tf.slice(foo, [3, 0], [4, foo.get_shape()[1]-2])`.
`begin` is zero-based; `size` is one-based. If `size[i]` is -1,
all remaining elements in dimension i are included in the
slice. In other words, this is equivalent to setting:
`size[i] = input_.dim_size(i) - begin[i]`
This operation requires that:
`0 <= begin[i] <= begin[i] + size[i] <= Di for i in [0, n]`
For example:
```python
t = tf.constant([[[1, 1, 1], [2, 2, 2]],
[[3, 3, 3], [4, 4, 4]],
[[5, 5, 5], [6, 6, 6]]])
tf.slice(t, [1, 0, 0], [1, 1, 3]) # [[[3, 3, 3]]]
tf.slice(t, [1, 0, 0], [1, 2, 3]) # [[[3, 3, 3],
# [4, 4, 4]]]
tf.slice(t, [1, 0, 0], [2, 1, 3]) # [[[3, 3, 3]],
# [[5, 5, 5]]]
```
Args:
input_: A `Tensor`.
begin: An `int32` or `int64` `Tensor`.
size: An `int32` or `int64` `Tensor`.
name: A name for the operation (optional).
Returns:
A `Tensor` the same type as `input_`."
8055,strided_slice,tensorflow/tensorflow/python/ops/array_ops.py,1104,function,"Extracts a strided slice of a tensor (generalized Python array indexing).
See also `tf.slice`.
**Instead of calling this op directly most users will want to use the
NumPy-style slicing syntax (e.g. `tensor[..., 3:4:-1, tf.newaxis, 3]`), which
is supported via `tf.Tensor.__getitem__` and `tf.Variable.__getitem__`.**
The interface of this op is a low-level encoding of the slicing syntax.
Roughly speaking, this op extracts a slice of size `(end-begin)/stride`
from the given `input_` tensor. Starting at the location specified by `begin`
the slice continues by adding `stride` to the index until all dimensions are
not less than `end`.
Note that a stride can be negative, which causes a reverse slice.
Given a Python slice `input[spec0, spec1, ..., specn]`,
this function will be called as follows.
`begin`, `end`, and `strides` will be vectors of length n.
n in general is not equal to the rank of the `input_` tensor.
In each mask field (`begin_mask`, `end_mask`, `ellipsis_mask`,
`new_axis_mask`, `shrink_axis_mask`) the ith bit will correspond to
the ith spec.
If the ith bit of `begin_mask` is set, `begin[i]` is ignored and
the fullest possible range in that dimension is used instead.
`end_mask` works analogously, except with the end range.
`foo[5:,:,:3]` on a 7x8x9 tensor is equivalent to `foo[5:7,0:8,0:3]`.
`foo[::-1]` reverses a tensor with shape 8.
If the ith bit of `ellipsis_mask` is set, as many unspecified dimensions
as needed will be inserted between other dimensions. Only one
non-zero bit is allowed in `ellipsis_mask`.
For example `foo[3:5,...,4:5]` on a shape 10x3x3x10 tensor is
equivalent to `foo[3:5,:,:,4:5]` and
`foo[3:5,...]` is equivalent to `foo[3:5,:,:,:]`.
If the ith bit of `new_axis_mask` is set, then `begin`,
`end`, and `stride` are ignored and a new length 1 dimension is
added at this point in the output tensor.
For example,
`foo[:4, tf.newaxis, :2]` would produce a shape `(4, 1, 2)` tensor.
If the ith bit of `shrink_axis_mask` is set, it implies that the ith
specification shrinks the dimensionality by 1, taking on the value at index
`begin[i]`. `end[i]` and `strides[i]` are ignored in this case. For example in
Python one might do `foo[:, 3, :]` which would result in `shrink_axis_mask`
equal to 2.
NOTE: `begin` and `end` are zero-indexed.
`strides` entries must be non-zero.
```python
t = tf.constant([[[1, 1, 1], [2, 2, 2]],
[[3, 3, 3], [4, 4, 4]],
[[5, 5, 5], [6, 6, 6]]])
tf.strided_slice(t, [1, 0, 0], [2, 1, 3], [1, 1, 1]) # [[[3, 3, 3]]]
tf.strided_slice(t, [1, 0, 0], [2, 2, 3], [1, 1, 1]) # [[[3, 3, 3],
# [4, 4, 4]]]
tf.strided_slice(t, [1, -1, 0], [2, -3, 3], [1, -1, 1]) # [[[4, 4, 4],
# [3, 3, 3]]]
```
Args:
input_: A `Tensor`.
begin: An `int32` or `int64` `Tensor`.
end: An `int32` or `int64` `Tensor`.
strides: An `int32` or `int64` `Tensor`.
begin_mask: An `int32` mask.
end_mask: An `int32` mask.
ellipsis_mask: An `int32` mask.
new_axis_mask: An `int32` mask.
shrink_axis_mask: An `int32` mask.
var: The variable corresponding to `input_` or None
name: A name for the operation (optional).
Returns:
A `Tensor` the same type as `input`."
8056,parallel_stack,tensorflow/tensorflow/python/ops/array_ops.py,1296,function,"Stacks a list of rank-`R` tensors into one rank-`(R+1)` tensor in parallel.
Requires that the shape of inputs be known at graph construction time.
Packs the list of tensors in `values` into a tensor with rank one higher than
each tensor in `values`, by packing them along the first dimension.
Given a list of length `N` of tensors of shape `(A, B, C)`; the `output`
tensor will have the shape `(N, A, B, C)`.
For example:
```python
x = tf.constant([1, 4])
y = tf.constant([2, 5])
z = tf.constant([3, 6])
tf.parallel_stack([x, y, z]) # [[1, 4], [2, 5], [3, 6]]
```
The difference between `stack` and `parallel_stack` is that `stack` requires
all the inputs be computed before the operation will begin but doesn't require
that the input shapes be known during graph construction.
`parallel_stack` will copy pieces of the input into the output as they become
available, in some situations this can provide a performance benefit.
Unlike `stack`, `parallel_stack` does NOT support backpropagation.
This is the opposite of unstack. The numpy equivalent is
tf.parallel_stack([x, y, z]) = np.asarray([x, y, z])
Args:
values: A list of `Tensor` objects with the same shape and type.
name: A name for this operation (optional).
Returns:
output: A stacked `Tensor` with the same type as `values`."
8057,stack,tensorflow/tensorflow/python/ops/array_ops.py,1348,function,"Stacks a list of rank-`R` tensors into one rank-`(R+1)` tensor.
See also `tf.concat`, `tf.tile`, `tf.repeat`.
Packs the list of tensors in `values` into a tensor with rank one higher than
each tensor in `values`, by packing them along the `axis` dimension.
Given a list of length `N` of tensors of shape `(A, B, C)`;
if `axis == 0` then the `output` tensor will have the shape `(N, A, B, C)`.
if `axis == 1` then the `output` tensor will have the shape `(A, N, B, C)`.
Etc.
For example:
>>> x = tf.constant([1, 4])
>>> y = tf.constant([2, 5])
>>> z = tf.constant([3, 6])
>>> tf.stack([x, y, z])
<tf.Tensor: shape=(3, 2), dtype=int32, numpy=
array([[1, 4],
[2, 5],
[3, 6]], dtype=int32)>
>>> tf.stack([x, y, z], axis=1)
<tf.Tensor: shape=(2, 3), dtype=int32, numpy=
array([[1, 2, 3],
[4, 5, 6]], dtype=int32)>
This is the opposite of unstack. The numpy equivalent is `np.stack`
>>> np.array_equal(np.stack([x, y, z]), tf.stack([x, y, z]))
True
Args:
values: A list of `Tensor` objects with the same shape and type.
axis: An `int`. The axis to stack along. Defaults to the first dimension.
Negative values wrap around, so the valid range is `[-(R+1), R+1)`.
name: A name for this operation (optional).
Returns:
output: A stacked `Tensor` with the same type as `values`.
Raises:
ValueError: If `axis` is out of the range [-(R+1), R+1)."
8058,unstack,tensorflow/tensorflow/python/ops/array_ops.py,1533,function,"Unpacks the given dimension of a rank-`R` tensor into rank-`(R-1)` tensors.
Unpacks `num` tensors from `value` by chipping it along the `axis` dimension.
If `num` is not specified (the default), it is inferred from `value`'s shape.
If `value.shape[axis]` is not known, `ValueError` is raised.
For example, given a tensor of shape `(A, B, C, D)`;
If `axis == 0` then the i'th tensor in `output` is the slice
`value[i, :, :, :]` and each tensor in `output` will have shape `(B, C, D)`.
(Note that the dimension unpacked along is gone, unlike `split`).
If `axis == 1` then the i'th tensor in `output` is the slice
`value[:, i, :, :]` and each tensor in `output` will have shape `(A, C, D)`.
Etc.
This is the opposite of stack.
Args:
value: A rank `R > 0` `Tensor` to be unstacked.
num: An `int`. The length of the dimension `axis`. Automatically inferred if
`None` (the default).
axis: An `int`. The axis to unstack along. Defaults to the first dimension.
Negative values wrap around, so the valid range is `[-R, R)`.
name: A name for the operation (optional).
Returns:
The list of `Tensor` objects unstacked from `value`.
Raises:
ValueError: If `num` is unspecified and cannot be inferred.
ValueError: If `axis` is out of the range [-R, R)."
8059,concat,tensorflow/tensorflow/python/ops/array_ops.py,1582,function,"Concatenates tensors along one dimension.
See also `tf.tile`, `tf.stack`, `tf.repeat`.
Concatenates the list of tensors `values` along dimension `axis`. If
`values[i].shape = [D0, D1, ... Daxis(i), ...Dn]`, the concatenated
result has shape
[D0, D1, ... Raxis, ...Dn]
where
Raxis = sum(Daxis(i))
That is, the data from the input tensors is joined along the `axis`
dimension.
The number of dimensions of the input tensors must match, and all dimensions
except `axis` must be equal.
For example:
>>> t1 = [[1, 2, 3], [4, 5, 6]]
>>> t2 = [[7, 8, 9], [10, 11, 12]]
>>> tf.concat([t1, t2], 0)
<tf.Tensor: shape=(4, 3), dtype=int32, numpy=
array([[ 1, 2, 3],
[ 4, 5, 6],
[ 7, 8, 9],
[10, 11, 12]], dtype=int32)>
>>> tf.concat([t1, t2], 1)
<tf.Tensor: shape=(2, 6), dtype=int32, numpy=
array([[ 1, 2, 3, 7, 8, 9],
[ 4, 5, 6, 10, 11, 12]], dtype=int32)>
As in Python, the `axis` could also be negative numbers. Negative `axis`
are interpreted as counting from the end of the rank, i.e.,
`axis + rank(values)`-th dimension.
For example:
>>> t1 = [[[1, 2], [2, 3]], [[4, 4], [5, 3]]]
>>> t2 = [[[7, 4], [8, 4]], [[2, 10], [15, 11]]]
>>> tf.concat([t1, t2], -1)
<tf.Tensor: shape=(2, 2, 4), dtype=int32, numpy=
array([[[ 1, 2, 7, 4],
[ 2, 3, 8, 4]],
[[ 4, 4, 2, 10],
[ 5, 3, 15, 11]]], dtype=int32)>
Note: If you are concatenating along a new axis consider using stack.
E.g.
```python
tf.concat([tf.expand_dims(t, axis) for t in tensors], axis)
```
can be rewritten as
```python
tf.stack(tensors, axis=axis)
```
Args:
values: A list of `Tensor` objects or a single `Tensor`.
axis: 0-D `int32` `Tensor`. Dimension along which to concatenate. Must be
in the range `[-rank(values), rank(values))`. As in Python, indexing for
axis is 0-based. Positive axis in the rage of `[0, rank(values))` refers
to `axis`-th dimension. And negative axis refers to `axis +
rank(values)`-th dimension.
name: A name for the operation (optional).
Returns:
A `Tensor` resulting from concatenation of the input tensors."
8060,boolean_mask,tensorflow/tensorflow/python/ops/array_ops.py,1677,function,"Apply boolean mask to tensor.
Numpy equivalent is `tensor[mask]`.
In general, `0 < dim(mask) = K <= dim(tensor)`, and `mask`'s shape must match
the first K dimensions of `tensor`'s shape. We then have:
`boolean_mask(tensor, mask)[i, j1,...,jd] = tensor[i1,...,iK,j1,...,jd]`
where `(i1,...,iK)` is the ith `True` entry of `mask` (row-major order).
The `axis` could be used with `mask` to indicate the axis to mask from.
In that case, `axis + dim(mask) <= dim(tensor)` and `mask`'s shape must match
the first `axis + dim(mask)` dimensions of `tensor`'s shape.
See also: `tf.ragged.boolean_mask`, which can be applied to both dense and
ragged tensors, and can be used if you need to preserve the masked dimensions
of `tensor` (rather than flattening them, as `tf.boolean_mask` does).
Examples:
```python
# 1-D example
tensor = [0, 1, 2, 3]
mask = np.array([True, False, True, False])
tf.boolean_mask(tensor, mask) # [0, 2]
# 2-D example
tensor = [[1, 2], [3, 4], [5, 6]]
mask = np.array([True, False, True])
tf.boolean_mask(tensor, mask) # [[1, 2], [5, 6]]
```
Args:
tensor: N-D Tensor.
mask: K-D boolean Tensor, K <= N and K must be known statically.
name: A name for this operation (optional).
axis: A 0-D int Tensor representing the axis in `tensor` to mask from. By
default, axis is 0 which will mask from the first dimension. Otherwise K +
axis <= N.
Returns:
(N-K+1)-dimensional tensor populated by entries in `tensor` corresponding
to `True` values in `mask`.
Raises:
ValueError: If shapes do not conform."
8061,boolean_mask_v2,tensorflow/tensorflow/python/ops/array_ops.py,1771,function,"Apply boolean mask to tensor.
Numpy equivalent is `tensor[mask]`.
In general, `0 < dim(mask) = K <= dim(tensor)`, and `mask`'s shape must match
the first K dimensions of `tensor`'s shape. We then have:
`boolean_mask(tensor, mask)[i, j1,...,jd] = tensor[i1,...,iK,j1,...,jd]`
where `(i1,...,iK)` is the ith `True` entry of `mask` (row-major order).
The `axis` could be used with `mask` to indicate the axis to mask from.
In that case, `axis + dim(mask) <= dim(tensor)` and `mask`'s shape must match
the first `axis + dim(mask)` dimensions of `tensor`'s shape.
See also: `tf.ragged.boolean_mask`, which can be applied to both dense and
ragged tensors, and can be used if you need to preserve the masked dimensions
of `tensor` (rather than flattening them, as `tf.boolean_mask` does).
Examples:
>>> tensor = [0, 1, 2, 3] # 1-D example
>>> mask = np.array([True, False, True, False])
>>> tf.boolean_mask(tensor, mask)
<tf.Tensor: shape=(2,), dtype=int32, numpy=array([0, 2], dtype=int32)>
>>> tensor = [[1, 2], [3, 4], [5, 6]] # 2-D example
>>> mask = np.array([True, False, True])
>>> tf.boolean_mask(tensor, mask)
<tf.Tensor: shape=(2, 2), dtype=int32, numpy=
array([[1, 2],
[5, 6]], dtype=int32)>
Args:
tensor: N-D Tensor.
mask: K-D boolean Tensor, K <= N and K must be known statically.
axis: A 0-D int Tensor representing the axis in `tensor` to mask from. By
default, axis is 0 which will mask from the first dimension. Otherwise K +
axis <= N.
name: A name for this operation (optional).
Returns:
(N-K+1)-dimensional tensor populated by entries in `tensor` corresponding
to `True` values in `mask`.
Raises:
ValueError: If shapes do not conform.
Examples:
```python
# 2-D example
tensor = [[1, 2], [3, 4], [5, 6]]
mask = np.array([True, False, True])
boolean_mask(tensor, mask) # [[1, 2], [5, 6]]
```"
8062,sparse_mask,tensorflow/tensorflow/python/ops/array_ops.py,1831,function,"Masks elements of `IndexedSlices`.
Given an `IndexedSlices` instance `a`, returns another `IndexedSlices` that
contains a subset of the slices of `a`. Only the slices at indices not
specified in `mask_indices` are returned.
This is useful when you need to extract a subset of slices in an
`IndexedSlices` object.
For example:
```python
# `a` contains slices at indices [12, 26, 37, 45] from a large tensor
# with shape [1000, 10]
a.indices # [12, 26, 37, 45]
tf.shape(a.values) # [4, 10]
# `b` will be the subset of `a` slices at its second and third indices, so
# we want to mask its first and last indices (which are at absolute
# indices 12, 45)
b = tf.sparse.mask(a, [12, 45])
b.indices # [26, 37]
tf.shape(b.values) # [2, 10]
```
Args:
a: An `IndexedSlices` instance.
mask_indices: Indices of elements to mask.
name: A name for the operation (optional).
Returns:
The masked `IndexedSlices` instance."
8063,unique,tensorflow/tensorflow/python/ops/array_ops.py,1875,function,"Finds unique elements in a 1-D tensor.
See also `tf.unique_with_counts`.
This operation returns a tensor `y` containing all of the unique elements
of `x` sorted in the same order that they occur in `x`. This operation
also returns a tensor `idx` the same size as `x` that contains the index
of each value of `x` in the unique output `y`. In other words:
y[idx[i]] = x[i] for i in [0, 1,...,rank(x) - 1]
Example usage:
>>> x = tf.constant([1, 1, 2, 4, 4, 4, 7, 8, 8])
>>> y, idx = unique(x)
>>> y
<tf.Tensor: id=5, shape=(5,), dtype=int32,
numpy=array([1, 2, 4, 7, 8], dtype=int32)>
>>> idx
<tf.Tensor: id=6, shape=(9,), dtype=int32,
numpy=array([0, 0, 1, 2, 2, 2, 3, 4, 4], dtype=int32)>
Args:
x: A Tensor. 1-D.
out_idx: An optional tf.DType from: tf.int32, tf.int64. Defaults to
tf.int32.
name: A name for the operation (optional).
Returns:
A tuple of Tensor objects (y, idx).
y: A Tensor. Has the same type as x.
idx: A Tensor of type out_idx."
8064,unique_with_counts,tensorflow/tensorflow/python/ops/array_ops.py,1923,function,"Finds unique elements in a 1-D tensor.
See also `tf.unique`.
This operation returns a tensor `y` containing all of the unique elements
of `x` sorted in the same order that they occur in `x`. This operation
also returns a tensor `idx` the same size as `x` that contains the index
of each value of `x` in the unique output `y`. Finally, it returns a
third tensor `count` that contains the count of each element of `y`
in `x`. In other words:
y[idx[i]] = x[i] for i in [0, 1,...,rank(x) - 1]
Example usage:
>>> x = tf.constant([1, 1, 2, 4, 4, 4, 7, 8, 8])
>>> y, idx, count = unique_with_counts(x)
>>> y
<tf.Tensor: id=8, shape=(5,), dtype=int32,
numpy=array([1, 2, 4, 7, 8], dtype=int32)>
>>> idx
<tf.Tensor: id=9, shape=(9,), dtype=int32,
numpy=array([0, 0, 1, 2, 2, 2, 3, 4, 4], dtype=int32)>
>>> count
<tf.Tensor: id=10, shape=(5,), dtype=int32,
numpy=array([2, 1, 3, 1, 2], dtype=int32)>
Args:
x: A Tensor. 1-D.
out_idx: An optional tf.DType from: tf.int32, tf.int64. Defaults to
tf.int32.
name: A name for the operation (optional).
Returns:
A tuple of Tensor objects (y, idx, count).
y: A Tensor. Has the same type as x.
idx: A Tensor of type out_idx.
count: A Tensor of type out_idx."
8065,split,tensorflow/tensorflow/python/ops/array_ops.py,1976,function,"Splits a tensor `value` into a list of sub tensors.
See also `tf.unstack`.
If `num_or_size_splits` is an integer, then `value` is split along the
dimension `axis` into `num_or_size_splits` smaller tensors. This requires that
`value.shape[axis]` is divisible by `num_or_size_splits`.
If `num_or_size_splits` is a 1-D Tensor (or list), then `value` is split into
`len(num_or_size_splits)` elements. The shape of the `i`-th
element has the same size as the `value` except along dimension `axis` where
the size is `num_or_size_splits[i]`.
For example:
>>> x = tf.Variable(tf.random.uniform([5, 30], -1, 1))
>>>
>>> # Split `x` into 3 tensors along dimension 1
>>> s0, s1, s2 = tf.split(x, num_or_size_splits=3, axis=1)
>>> tf.shape(s0).numpy()
array([ 5, 10], dtype=int32)
>>>
>>> # Split `x` into 3 tensors with sizes [4, 15, 11] along dimension 1
>>> split0, split1, split2 = tf.split(x, [4, 15, 11], 1)
>>> tf.shape(split0).numpy()
array([5, 4], dtype=int32)
>>> tf.shape(split1).numpy()
array([ 5, 15], dtype=int32)
>>> tf.shape(split2).numpy()
array([ 5, 11], dtype=int32)
Args:
value: The `Tensor` to split.
num_or_size_splits: Either an integer indicating the number of splits along
`axis` or a 1-D integer `Tensor` or Python list containing the sizes of
each output tensor along `axis`. If a scalar, then it must evenly divide
`value.shape[axis]`; otherwise the sum of sizes along the split axis
must match that of the `value`.
axis: An integer or scalar `int32` `Tensor`. The dimension along which to
split. Must be in the range `[-rank(value), rank(value))`. Defaults to 0.
num: Optional, used to specify the number of outputs when it cannot be
inferred from the shape of `size_splits`.
name: A name for the operation (optional).
Returns:
if `num_or_size_splits` is a scalar returns a list of `num_or_size_splits`
`Tensor` objects; if `num_or_size_splits` is a 1-D Tensor returns
`num_or_size_splits.get_shape[0]` `Tensor` objects resulting from splitting
`value`.
Raises:
ValueError: If `num` is unspecified and cannot be inferred."
8066,transpose_v2,tensorflow/tensorflow/python/ops/array_ops.py,2054,function,"Transposes `a`, where `a` is a Tensor.
Permutes the dimensions according to the value of `perm`.
The returned tensor's dimension `i` will correspond to the input dimension
`perm[i]`. If `perm` is not given, it is set to (n-1...0), where n is the rank
of the input tensor. Hence by default, this operation performs a regular
matrix transpose on 2-D input Tensors.
If conjugate is `True` and `a.dtype` is either `complex64` or `complex128`
then the values of `a` are conjugated and transposed.
@compatibility(numpy)
In `numpy` transposes are memory-efficient constant time operations as they
simply return a new view of the same data with adjusted `strides`.
TensorFlow does not support strides, so `transpose` returns a new tensor with
the items permuted.
@end_compatibility
For example:
>>> x = tf.constant([[1, 2, 3], [4, 5, 6]])
>>> tf.transpose(x)
<tf.Tensor: shape=(3, 2), dtype=int32, numpy=
array([[1, 4],
[2, 5],
[3, 6]], dtype=int32)>
Equivalently, you could call `tf.transpose(x, perm=[1, 0])`.
If `x` is complex, setting conjugate=True gives the conjugate transpose:
>>> x = tf.constant([[1 + 1j, 2 + 2j, 3 + 3j],
... [4 + 4j, 5 + 5j, 6 + 6j]])
>>> tf.transpose(x, conjugate=True)
<tf.Tensor: shape=(3, 2), dtype=complex128, numpy=
array([[1.-1.j, 4.-4.j],
[2.-2.j, 5.-5.j],
[3.-3.j, 6.-6.j]])>
'perm' is more useful for n-dimensional tensors where n > 2:
>>> x = tf.constant([[[ 1, 2, 3],
... [ 4, 5, 6]],
... [[ 7, 8, 9],
... [10, 11, 12]]])
As above, simply calling `tf.transpose` will default to `perm=[2,1,0]`.
To take the transpose of the matrices in dimension-0 (such as when you are
transposing matrices where 0 is the batch dimesnion), you would set
`perm=[0,2,1]`.
>>> tf.transpose(x, perm=[0, 2, 1])
<tf.Tensor: shape=(2, 3, 2), dtype=int32, numpy=
array([[[ 1, 4],
[ 2, 5],
[ 3, 6]],
[[ 7, 10],
[ 8, 11],
[ 9, 12]]], dtype=int32)>
Note: This has a shorthand `linalg.matrix_transpose`):
Args:
a: A `Tensor`.
perm: A permutation of the dimensions of `a`. This should be a vector.
conjugate: Optional bool. Setting it to `True` is mathematically equivalent
to tf.math.conj(tf.transpose(input)).
name: A name for the operation (optional).
Returns:
A transposed `Tensor`."
8067,transpose,tensorflow/tensorflow/python/ops/array_ops.py,2135,function,"Transposes `a`.
Permutes the dimensions according to `perm`.
The returned tensor's dimension i will correspond to the input dimension
`perm[i]`. If `perm` is not given, it is set to (n-1...0), where n is
the rank of the input tensor. Hence by default, this operation performs a
regular matrix transpose on 2-D input Tensors. If conjugate is True and
`a.dtype` is either `complex64` or `complex128` then the values of `a`
are conjugated and transposed.
@compatibility(numpy)
In `numpy` transposes are memory-efficient constant time operations as they
simply return a new view of the same data with adjusted `strides`.
TensorFlow does not support strides, so `transpose` returns a new tensor with
the items permuted.
@end_compatibility
For example:
```python
x = tf.constant([[1, 2, 3], [4, 5, 6]])
tf.transpose(x) # [[1, 4]
# [2, 5]
# [3, 6]]
# Equivalently
tf.transpose(x, perm=[1, 0]) # [[1, 4]
# [2, 5]
# [3, 6]]
# If x is complex, setting conjugate=True gives the conjugate transpose
x = tf.constant([[1 + 1j, 2 + 2j, 3 + 3j],
[4 + 4j, 5 + 5j, 6 + 6j]])
tf.transpose(x, conjugate=True) # [[1 - 1j, 4 - 4j],
# [2 - 2j, 5 - 5j],
# [3 - 3j, 6 - 6j]]
# 'perm' is more useful for n-dimensional tensors, for n > 2
x = tf.constant([[[ 1, 2, 3],
[ 4, 5, 6]],
[[ 7, 8, 9],
[10, 11, 12]]])
# Take the transpose of the matrices in dimension-0
# (this common operation has a shorthand `linalg.matrix_transpose`)
tf.transpose(x, perm=[0, 2, 1]) # [[[1, 4],
# [2, 5],
# [3, 6]],
# [[7, 10],
# [8, 11],
# [9, 12]]]
```
Args:
a: A `Tensor`.
perm: A permutation of the dimensions of `a`.
name: A name for the operation (optional).
conjugate: Optional bool. Setting it to `True` is mathematically equivalent
to tf.math.conj(tf.transpose(input)).
Returns:
A transposed `Tensor`."
8068,matrix_transpose,tensorflow/tensorflow/python/ops/array_ops.py,2227,function,"Transposes last two dimensions of tensor `a`.
For example:
```python
x = tf.constant([[1, 2, 3], [4, 5, 6]])
tf.linalg.matrix_transpose(x) # [[1, 4],
# [2, 5],
# [3, 6]]
x = tf.constant([[1 + 1j, 2 + 2j, 3 + 3j],
[4 + 4j, 5 + 5j, 6 + 6j]])
tf.linalg.matrix_transpose(x, conjugate=True) # [[1 - 1j, 4 - 4j],
# [2 - 2j, 5 - 5j],
# [3 - 3j, 6 - 6j]]
# Matrix with two batch dimensions.
# x.shape is [1, 2, 3, 4]
# tf.linalg.matrix_transpose(x) is shape [1, 2, 4, 3]
```
Note that `tf.matmul` provides kwargs allowing for transpose of arguments.
This is done with minimal cost, and is preferable to using this function. E.g.
```python
# Good! Transpose is taken at minimal additional cost.
tf.matmul(matrix, b, transpose_b=True)
# Inefficient!
tf.matmul(matrix, tf.linalg.matrix_transpose(b))
```
@compatibility(numpy)
In `numpy` transposes are memory-efficient constant time operations as they
simply return a new view of the same data with adjusted `strides`.
TensorFlow does not support strides, `linalg.matrix_transpose` returns a new
tensor with the items permuted.
@end_compatibility
Args:
a: A `Tensor` with `rank >= 2`.
name: A name for the operation (optional).
conjugate: Optional bool. Setting it to `True` is mathematically equivalent
to tf.math.conj(tf.linalg.matrix_transpose(input)).
Returns:
A transposed batch matrix `Tensor`.
Raises:
ValueError: If `a` is determined statically to have `rank < 2`."
8069,matrix_diag,tensorflow/tensorflow/python/ops/array_ops.py,2306,function,"Returns a batched diagonal tensor with given batched diagonal values.
Returns a tensor with the contents in `diagonal` as `k[0]`-th to `k[1]`-th
diagonals of a matrix, with everything else padded with `padding`. `num_rows`
and `num_cols` specify the dimension of the innermost matrix of the output. If
both are not specified, the op assumes the innermost matrix is square and
infers its size from `k` and the innermost dimension of `diagonal`. If only
one of them is specified, the op assumes the unspecified value is the smallest
possible based on other criteria.
Let `diagonal` have `r` dimensions `[I, J, ..., L, M, N]`. The output tensor
has rank `r+1` with shape `[I, J, ..., L, M, num_rows, num_cols]` when only
one diagonal is given (`k` is an integer or `k[0] == k[1]`). Otherwise, it has
rank `r` with shape `[I, J, ..., L, num_rows, num_cols]`.
The second innermost dimension of `diagonal` has double meaning. When `k` is
scalar or `k[0] == k[1]`, `M` is part of the batch size [I, J, ..., M], and
the output tensor is:
```
output[i, j, ..., l, m, n]
= diagonal[i, j, ..., l, n-max(d_upper, 0)] ; if n - m == d_upper
padding_value ; otherwise
```
Otherwise, `M` is treated as the number of diagonals for the matrix in the
same batch (`M = k[1]-k[0]+1`), and the output tensor is:
```
output[i, j, ..., l, m, n]
= diagonal[i, j, ..., l, diag_index, index_in_diag] ; if k[0] <= d <= k[1]
padding_value ; otherwise
```
where `d = n - m`, `diag_index = k[1] - d`, and
`index_in_diag = n - max(d, 0) + offset`.
`offset` is zero except when the alignment of the diagonal is to the right.
```
offset = max_diag_len - diag_len(d) ; if (`align` in {RIGHT_LEFT, RIGHT_RIGHT}
and `d >= 0`) or
(`align` in {LEFT_RIGHT, RIGHT_RIGHT}
and `d <= 0`)
0 ; otherwise
```
where `diag_len(d) = min(cols - max(d, 0), rows + min(d, 0))`.
For example:
```
# The main diagonal.
diagonal = np.array([[1, 2, 3, 4], # Input shape: (2, 4)
[5, 6, 7, 8]])
tf.matrix_diag(diagonal) ==> [[[1, 0, 0, 0], # Output shape: (2, 4, 4)
[0, 2, 0, 0],
[0, 0, 3, 0],
[0, 0, 0, 4]],
[[5, 0, 0, 0],
[0, 6, 0, 0],
[0, 0, 7, 0],
[0, 0, 0, 8]]]
# A superdiagonal (per batch).
diagonal = np.array([[1, 2, 3], # Input shape: (2, 3)
[4, 5, 6]])
tf.matrix_diag(diagonal, k = 1)
==> [[[0, 1, 0, 0], # Output shape: (2, 4, 4)
[0, 0, 2, 0],
[0, 0, 0, 3],
[0, 0, 0, 0]],
[[0, 4, 0, 0],
[0, 0, 5, 0],
[0, 0, 0, 6],
[0, 0, 0, 0]]]
# A tridiagonal band (per batch).
diagonals = np.array([[[8, 9, 0], # Input shape: (2, 2, 3)
[1, 2, 3],
[0, 4, 5]],
[[2, 3, 0],
[6, 7, 9],
[0, 9, 1]]])
tf.matrix_diag(diagonals, k = (-1, 1))
==> [[[1, 8, 0], # Output shape: (2, 3, 3)
[4, 2, 9],
[0, 5, 3]],
[[6, 2, 0],
[9, 7, 3],
[0, 1, 9]]]
# RIGHT_LEFT alignment.
diagonals = np.array([[[0, 8, 9], # Input shape: (2, 2, 3)
[1, 2, 3],
[4, 5, 0]],
[[0, 2, 3],
[6, 7, 9],
[9, 1, 0]]])
tf.matrix_diag(diagonals, k = (-1, 1), align=""RIGHT_LEFT"")
==> [[[1, 8, 0], # Output shape: (2, 3, 3)
[4, 2, 9],
[0, 5, 3]],
[[6, 2, 0],
[9, 7, 3],
[0, 1, 9]]]
# Rectangular matrix.
diagonal = np.array([1, 2]) # Input shape: (2)
tf.matrix_diag(diagonal, k = -1, num_rows = 3, num_cols = 4)
==> [[0, 0, 0, 0], # Output shape: (3, 4)
[1, 0, 0, 0],
[0, 2, 0, 0]]
# Rectangular matrix with inferred num_cols and padding_value = 9.
tf.matrix_diag(diagonal, k = -1, num_rows = 3, padding_value = 9)
==> [[9, 9], # Output shape: (3, 2)
[1, 9],
[9, 2]]
```
Args:
diagonal: A `Tensor` with `rank k >= 1`.
name: A name for the operation (optional).
k: Diagonal offset(s). Positive value means superdiagonal, 0 refers to the
main diagonal, and negative value means subdiagonals. `k` can be a single
integer (for a single diagonal) or a pair of integers specifying the low
and high ends of a matrix band. `k[0]` must not be larger than `k[1]`.
num_rows: The number of rows of the output matrix. If it is not provided,
the op assumes the output matrix is a square matrix and infers the matrix
size from `d_lower`, `d_upper`, and the innermost dimension of `diagonal`.
num_cols: The number of columns of the output matrix. If it is not provided,
the op assumes the output matrix is a square matrix and infers the matrix
size from `d_lower`, `d_upper`, and the innermost dimension of `diagonal`.
padding_value: The value to fill the area outside the specified diagonal
band with. Default is 0.
align: Some diagonals are shorter than `max_diag_len` and need to be padded.
`align` is a string specifying how superdiagonals and subdiagonals should
be aligned, respectively. There are four possible alignments: ""RIGHT_LEFT""
(default), ""LEFT_RIGHT"", ""LEFT_LEFT"", and ""RIGHT_RIGHT"". ""RIGHT_LEFT""
aligns superdiagonals to the right (left-pads the row) and subdiagonals to
the left (right-pads the row). It is the packing format LAPACK uses.
cuSPARSE uses ""LEFT_RIGHT"", which is the opposite alignment.
Returns:
A Tensor. Has the same type as `diagonal`."
8070,matrix_diag_part,tensorflow/tensorflow/python/ops/array_ops.py,2476,function,"Returns the batched diagonal part of a batched tensor.
Returns a tensor with the `k[0]`-th to `k[1]`-th diagonals of the batched
`input`.
Assume `input` has `r` dimensions `[I, J, ..., L, M, N]`.
Let `max_diag_len` be the maximum length among all diagonals to be extracted,
`max_diag_len = min(M + min(k[1], 0), N + min(-k[0], 0))`
Let `num_diags` be the number of diagonals to extract,
`num_diags = k[1] - k[0] + 1`.
If `num_diags == 1`, the output tensor is of rank `r - 1` with shape
`[I, J, ..., L, max_diag_len]` and values:
```
diagonal[i, j, ..., l, n]
= input[i, j, ..., l, n+y, n+x] ; if 0 <= n+y < M and 0 <= n+x < N,
padding_value ; otherwise.
```
where `y = max(-k[1], 0)`, `x = max(k[1], 0)`.
Otherwise, the output tensor has rank `r` with dimensions
`[I, J, ..., L, num_diags, max_diag_len]` with values:
```
diagonal[i, j, ..., l, m, n]
= input[i, j, ..., l, n+y, n+x] ; if 0 <= n+y < M and 0 <= n+x < N,
padding_value ; otherwise.
```
where `d = k[1] - m`, `y = max(-d, 0) - offset`, and `x = max(d, 0) - offset`.
`offset` is zero except when the alignment of the diagonal is to the right.
```
offset = max_diag_len - diag_len(d) ; if (`align` in {RIGHT_LEFT, RIGHT_RIGHT}
and `d >= 0`) or
(`align` in {LEFT_RIGHT, RIGHT_RIGHT}
and `d <= 0`)
0 ; otherwise
```
where `diag_len(d) = min(cols - max(d, 0), rows + min(d, 0))`.
The input must be at least a matrix.
For example:
```
input = np.array([[[1, 2, 3, 4], # Input shape: (2, 3, 4)
[5, 6, 7, 8],
[9, 8, 7, 6]],
[[5, 4, 3, 2],
[1, 2, 3, 4],
[5, 6, 7, 8]]])
# A main diagonal from each batch.
tf.linalg.diag_part(input) ==> [[1, 6, 7], # Output shape: (2, 3)
[5, 2, 7]]
# A superdiagonal from each batch.
tf.linalg.diag_part(input, k = 1)
==> [[2, 7, 6], # Output shape: (2, 3)
[4, 3, 8]]
# A band from each batch.
tf.linalg.diag_part(input, k = (-1, 2))
==> [[[3, 8, 0], # Output shape: (2, 4, 3)
[2, 7, 6],
[1, 6, 7],
[0, 5, 8]],
[[3, 4, 0],
[4, 3, 8],
[5, 2, 7],
[0, 1, 6]]]
# RIGHT_LEFT alignment.
tf.linalg.diag_part(input, k = (-1, 2), align=""RIGHT_LEFT"")
==> [[[0, 3, 8], # Output shape: (2, 4, 3)
[2, 7, 6],
[1, 6, 7],
[5, 8, 0]],
[[0, 3, 4],
[4, 3, 8],
[5, 2, 7],
[1, 6, 0]]]
# max_diag_len can be shorter than the main diagonal.
tf.linalg.diag_part(input, k = (-2, -1))
==> [[[5, 8],
[0, 9]],
[[1, 6],
[0, 5]]]
# padding_value = 9
tf.linalg.diag_part(input, k = (1, 3), padding_value = 9)
==> [[[4, 9, 9], # Output shape: (2, 3, 3)
[3, 8, 9],
[2, 7, 6]],
[[2, 9, 9],
[3, 4, 9],
[4, 3, 8]]]
```
Args:
input: A `Tensor` with `rank k >= 2`.
name: A name for the operation (optional).
k: Diagonal offset(s). Positive value means superdiagonal, 0 refers to the
main diagonal, and negative value means subdiagonals. `k` can be a single
integer (for a single diagonal) or a pair of integers specifying the low
and high ends of a matrix band. `k[0]` must not be larger than `k[1]`.
padding_value: The value to fill the area outside the specified diagonal
band with. Default is 0.
align: Some diagonals are shorter than `max_diag_len` and need to be padded.
`align` is a string specifying how superdiagonals and subdiagonals should
be aligned, respectively. There are four possible alignments: ""RIGHT_LEFT""
(default), ""LEFT_RIGHT"", ""LEFT_LEFT"", and ""RIGHT_RIGHT"". ""RIGHT_LEFT""
aligns superdiagonals to the right (left-pads the row) and subdiagonals to
the left (right-pads the row). It is the packing format LAPACK uses.
cuSPARSE uses ""LEFT_RIGHT"", which is the opposite alignment.
Returns:
A Tensor containing diagonals of `input`. Has the same type as `input`."
8071,matrix_set_diag,tensorflow/tensorflow/python/ops/array_ops.py,2616,function,"Returns a batched matrix tensor with new batched diagonal values.
Given `input` and `diagonal`, this operation returns a tensor with the
same shape and values as `input`, except for the specified diagonals of the
innermost matrices. These will be overwritten by the values in `diagonal`.
`input` has `r+1` dimensions `[I, J, ..., L, M, N]`. When `k` is scalar or
`k[0] == k[1]`, `diagonal` has `r` dimensions `[I, J, ..., L, max_diag_len]`.
Otherwise, it has `r+1` dimensions `[I, J, ..., L, num_diags, max_diag_len]`.
`num_diags` is the number of diagonals, `num_diags = k[1] - k[0] + 1`.
`max_diag_len` is the longest diagonal in the range `[k[0], k[1]]`,
`max_diag_len = min(M + min(k[1], 0), N + min(-k[0], 0))`
The output is a tensor of rank `k+1` with dimensions `[I, J, ..., L, M, N]`.
If `k` is scalar or `k[0] == k[1]`:
```
output[i, j, ..., l, m, n]
= diagonal[i, j, ..., l, n-max(k[1], 0)] ; if n - m == k[1]
input[i, j, ..., l, m, n] ; otherwise
```
Otherwise,
```
output[i, j, ..., l, m, n]
= diagonal[i, j, ..., l, diag_index, index_in_diag] ; if k[0] <= d <= k[1]
input[i, j, ..., l, m, n] ; otherwise
```
where `d = n - m`, `diag_index = k[1] - d`, and
`index_in_diag = n - max(d, 0) + offset`.
`offset` is zero except when the alignment of the diagonal is to the right.
```
offset = max_diag_len - diag_len(d) ; if (`align` in {RIGHT_LEFT, RIGHT_RIGHT}
and `d >= 0`) or
(`align` in {LEFT_RIGHT, RIGHT_RIGHT}
and `d <= 0`)
0 ; otherwise
```
where `diag_len(d) = min(cols - max(d, 0), rows + min(d, 0))`.
For example:
```
# The main diagonal.
input = np.array([[[7, 7, 7, 7], # Input shape: (2, 3, 4)
[7, 7, 7, 7],
[7, 7, 7, 7]],
[[7, 7, 7, 7],
[7, 7, 7, 7],
[7, 7, 7, 7]]])
diagonal = np.array([[1, 2, 3], # Diagonal shape: (2, 3)
[4, 5, 6]])
tf.matrix_set_diag(input, diagonal)
==> [[[1, 7, 7, 7], # Output shape: (2, 3, 4)
[7, 2, 7, 7],
[7, 7, 3, 7]],
[[4, 7, 7, 7],
[7, 5, 7, 7],
[7, 7, 6, 7]]]
# A superdiagonal (per batch).
tf.matrix_set_diag(input, diagonal, k = 1)
==> [[[7, 1, 7, 7], # Output shape: (2, 3, 4)
[7, 7, 2, 7],
[7, 7, 7, 3]],
[[7, 4, 7, 7],
[7, 7, 5, 7],
[7, 7, 7, 6]]]
# A band of diagonals.
diagonals = np.array([[[9, 1, 0], # Diagonal shape: (2, 4, 3)
[6, 5, 8],
[1, 2, 3],
[0, 4, 5]],
[[1, 2, 0],
[5, 6, 4],
[6, 1, 2],
[0, 3, 4]]])
tf.matrix_set_diag(input, diagonals, k = (-1, 2))
==> [[[1, 6, 9, 7], # Output shape: (2, 3, 4)
[4, 2, 5, 1],
[7, 5, 3, 8]],
[[6, 5, 1, 7],
[3, 1, 6, 2],
[7, 4, 2, 4]]]
# RIGHT_LEFT alignment.
diagonals = np.array([[[0, 9, 1], # Diagonal shape: (2, 4, 3)
[6, 5, 8],
[1, 2, 3],
[4, 5, 0]],
[[0, 1, 2],
[5, 6, 4],
[6, 1, 2],
[3, 4, 0]]])
tf.matrix_set_diag(input, diagonals, k = (-1, 2), align=""RIGHT_LEFT"")
==> [[[1, 6, 9, 7], # Output shape: (2, 3, 4)
[4, 2, 5, 1],
[7, 5, 3, 8]],
[[6, 5, 1, 7],
[3, 1, 6, 2],
[7, 4, 2, 4]]]
```
Args:
input: A `Tensor` with rank `k + 1`, where `k >= 1`.
diagonal: A `Tensor` with rank `k`, when `d_lower == d_upper`, or `k + 1`,
otherwise. `k >= 1`.
name: A name for the operation (optional).
k: Diagonal offset(s). Positive value means superdiagonal, 0 refers to the
main diagonal, and negative value means subdiagonals. `k` can be a single
integer (for a single diagonal) or a pair of integers specifying the low
and high ends of a matrix band. `k[0]` must not be larger than `k[1]`.
align: Some diagonals are shorter than `max_diag_len` and need to be padded.
`align` is a string specifying how superdiagonals and subdiagonals should
be aligned, respectively. There are four possible alignments: ""RIGHT_LEFT""
(default), ""LEFT_RIGHT"", ""LEFT_LEFT"", and ""RIGHT_RIGHT"". ""RIGHT_LEFT""
aligns superdiagonals to the right (left-pads the row) and subdiagonals to
the left (right-pads the row). It is the packing format LAPACK uses.
cuSPARSE uses ""LEFT_RIGHT"", which is the opposite alignment."
8072,zeros,tensorflow/tensorflow/python/ops/array_ops.py,2780,function,"Creates a tensor with all elements set to zero.
See also `tf.zeros_like`, `tf.ones`, `tf.fill`, `tf.eye`.
This operation returns a tensor of type `dtype` with shape `shape` and
all elements set to zero.
>>> tf.zeros([3, 4], tf.int32)
<tf.Tensor: shape=(3, 4), dtype=int32, numpy=
array([[0, 0, 0, 0],
[0, 0, 0, 0],
[0, 0, 0, 0]], dtype=int32)>
Args:
shape: A `list` of integers, a `tuple` of integers, or
a 1-D `Tensor` of type `int32`.
dtype: The DType of an element in the resulting `Tensor`.
name: Optional string. A name for the operation.
Returns:
A `Tensor` with all elements set to zero."
8073,zeros_like,tensorflow/tensorflow/python/ops/array_ops.py,2836,function,"Creates a tensor with all elements set to zero.
See also `tf.zeros`.
Given a single tensor (`tensor`), this operation returns a tensor of the
same type and shape as `tensor` with all elements set to zero. Optionally,
you can use `dtype` to specify a new type for the returned tensor.
Examples:
>>> tensor = tf.constant([[1, 2, 3], [4, 5, 6]])
>>> tf.zeros_like(tensor)
<tf.Tensor: shape=(2, 3), dtype=int32, numpy=
array([[0, 0, 0],
[0, 0, 0]], dtype=int32)>
>>> tf.zeros_like(tensor, dtype=tf.float32)
<tf.Tensor: shape=(2, 3), dtype=float32, numpy=
array([[0., 0., 0.],
[0., 0., 0.]], dtype=float32)>
Args:
tensor: A `Tensor`.
dtype: A type for the returned `Tensor`. Must be `float16`, `float32`,
`float64`, `int8`, `uint8`, `int16`, `uint16`, `int32`, `int64`,
`complex64`, `complex128`, `bool` or `string`. (optional)
name: A name for the operation (optional).
optimize: if `True`, attempt to statically determine the shape of `tensor`
and encode it as a constant. (optional, defaults to `True`)
Returns:
A `Tensor` with all elements set to zero."
8074,zeros_like_v2,tensorflow/tensorflow/python/ops/array_ops.py,2875,function,"Creates a tensor with all elements set to zero.
See also `tf.zeros`.
Given a single tensor or array-like object (`input`), this operation returns
a tensor of the same type and shape as `input` with all elements set to zero.
Optionally, you can use `dtype` to specify a new type for the returned tensor.
Examples:
>>> tensor = tf.constant([[1, 2, 3], [4, 5, 6]])
>>> tf.zeros_like(tensor)
<tf.Tensor: shape=(2, 3), dtype=int32, numpy=
array([[0, 0, 0],
[0, 0, 0]], dtype=int32)>
>>> tf.zeros_like(tensor, dtype=tf.float32)
<tf.Tensor: shape=(2, 3), dtype=float32, numpy=
array([[0., 0., 0.],
[0., 0., 0.]], dtype=float32)>
>>> tf.zeros_like([[1, 2, 3], [4, 5, 6]])
<tf.Tensor: shape=(2, 3), dtype=int32, numpy=
array([[0, 0, 0],
[0, 0, 0]], dtype=int32)>
Args:
input: A `Tensor` or array-like object.
dtype: A type for the returned `Tensor`. Must be `float16`, `float32`,
`float64`, `int8`, `uint8`, `int16`, `uint16`, `int32`, `int64`,
`complex64`, `complex128`, `bool` or `string` (optional).
name: A name for the operation (optional).
Returns:
A `Tensor` with all elements set to zero."
8075,zeros_like_impl,tensorflow/tensorflow/python/ops/array_ops.py,2919,function,Internal implementation for the v1/v2 zeros_like API calls.
8076,ones_like,tensorflow/tensorflow/python/ops/array_ops.py,2951,function,"Creates a tensor with all elements set to 1.
See also `tf.ones`.
Given a single tensor (`tensor`), this operation returns a tensor of the same
type and shape as `tensor` with all elements set to 1. Optionally, you can
specify a new type (`dtype`) for the returned tensor.
For example:
```python
tensor = tf.constant([[1, 2, 3], [4, 5, 6]])
tf.ones_like(tensor) # [[1, 1, 1], [1, 1, 1]]
```
Args:
tensor: A `Tensor`.
dtype: A type for the returned `Tensor`. Must be `float32`, `float64`,
`int8`, `uint8`, `int16`, `uint16`, `int32`, `int64`, `complex64`,
`complex128` or `bool`.
name: A name for the operation (optional).
optimize: if true, attempt to statically determine the shape of 'tensor' and
encode it as a constant.
Returns:
A `Tensor` with all elements set to 1."
8077,ones_like_v2,tensorflow/tensorflow/python/ops/array_ops.py,2984,function,"Creates a tensor of all ones that has the same shape as the input.
See also `tf.ones`.
Given a single tensor (`tensor`), this operation returns a tensor of the
same type and shape as `tensor` with all elements set to 1. Optionally,
you can use `dtype` to specify a new type for the returned tensor.
For example:
>>> tensor = tf.constant([[1, 2, 3], [4, 5, 6]])
>>> tf.ones_like(tensor)
<tf.Tensor: shape=(2, 3), dtype=int32, numpy=
array([[1, 1, 1],
[1, 1, 1]], dtype=int32)>
Args:
input: A `Tensor`.
dtype: A type for the returned `Tensor`. Must be `float16`, `float32`,
`float64`, `int8`, `uint8`, `int16`, `uint16`, `int32`, `int64`,
`complex64`, `complex128`, `bool` or `string`.
name: A name for the operation (optional).
Returns:
A `Tensor` with all elements set to one."
8078,ones_like_impl,tensorflow/tensorflow/python/ops/array_ops.py,3017,function,Internal implementation for the v1/v2 ones_like API calls.
8079,ones,tensorflow/tensorflow/python/ops/array_ops.py,3032,function,"Creates a tensor with all elements set to one (1).
See also `tf.ones_like`, `tf.zeros`, `tf.fill`, `tf.eye`.
This operation returns a tensor of type `dtype` with shape `shape` and
all elements set to one.
>>> tf.ones([3, 4], tf.int32)
<tf.Tensor: shape=(3, 4), dtype=int32, numpy=
array([[1, 1, 1, 1],
[1, 1, 1, 1],
[1, 1, 1, 1]], dtype=int32)>
Args:
shape: A `list` of integers, a `tuple` of integers, or
a 1-D `Tensor` of type `int32`.
dtype: Optional DType of an element in the resulting `Tensor`. Default is
`tf.float32`.
name: Optional string. A name for the operation.
Returns:
A `Tensor` with all elements set to one (1)."
8080,placeholder,tensorflow/tensorflow/python/ops/array_ops.py,3087,function,"Inserts a placeholder for a tensor that will be always fed.
**Important**: This tensor will produce an error if evaluated. Its value must
be fed using the `feed_dict` optional argument to `Session.run()`,
`Tensor.eval()`, or `Operation.run()`.
For example:
```python
x = tf.compat.v1.placeholder(tf.float32, shape=(1024, 1024))
y = tf.matmul(x, x)
with tf.compat.v1.Session() as sess:
print(sess.run(y)) # ERROR: will fail because x was not fed.
rand_array = np.random.rand(1024, 1024)
print(sess.run(y, feed_dict={x: rand_array})) # Will succeed.
```
@compatibility(eager)
Placeholders are not compatible with eager execution.
@end_compatibility
Args:
dtype: The type of elements in the tensor to be fed.
shape: The shape of the tensor to be fed (optional). If the shape is not
specified, you can feed a tensor of any shape.
name: A name for the operation (optional).
Returns:
A `Tensor` that may be used as a handle for feeding a value, but not
evaluated directly.
Raises:
RuntimeError: if eager execution is enabled"
8081,placeholder_with_default,tensorflow/tensorflow/python/ops/array_ops.py,3132,function,"A placeholder op that passes through `input` when its output is not fed.
Args:
input: A `Tensor`. The default value to produce when output is not fed.
shape: A `tf.TensorShape` or list of `int`s. The (possibly partial) shape of
the tensor.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `input`."
8082,sparse_placeholder,tensorflow/tensorflow/python/ops/array_ops.py,3149,function,"Inserts a placeholder for a sparse tensor that will be always fed.
**Important**: This sparse tensor will produce an error if evaluated.
Its value must be fed using the `feed_dict` optional argument to
`Session.run()`, `Tensor.eval()`, or `Operation.run()`.
For example:
```python
x = tf.compat.v1.sparse.placeholder(tf.float32)
y = tf.sparse.reduce_sum(x)
with tf.compat.v1.Session() as sess:
print(sess.run(y)) # ERROR: will fail because x was not fed.
indices = np.array([[3, 2, 0], [4, 5, 1]], dtype=np.int64)
values = np.array([1.0, 2.0], dtype=np.float32)
shape = np.array([7, 9, 2], dtype=np.int64)
print(sess.run(y, feed_dict={
x: tf.compat.v1.SparseTensorValue(indices, values, shape)})) # Will
succeed.
print(sess.run(y, feed_dict={
x: (indices, values, shape)})) # Will succeed.
sp = tf.sparse.SparseTensor(indices=indices, values=values,
dense_shape=shape)
sp_value = sp.eval(session=sess)
print(sess.run(y, feed_dict={x: sp_value})) # Will succeed.
```
@compatibility{eager} Placeholders are not compatible with eager execution.
Args:
dtype: The type of `values` elements in the tensor to be fed.
shape: The shape of the tensor to be fed (optional). If the shape is not
specified, you can feed a sparse tensor of any shape.
name: A name for prefixing the operations (optional).
Returns:
A `SparseTensor` that may be used as a handle for feeding a value, but not
evaluated directly.
Raises:
RuntimeError: if eager execution is enabled"
8083,pad_v2,tensorflow/tensorflow/python/ops/array_ops.py,3250,function,"Pads a tensor.
This operation pads a `tensor` according to the `paddings` you specify.
`paddings` is an integer tensor with shape `[n, 2]`, where n is the rank of
`tensor`. For each dimension D of `input`, `paddings[D, 0]` indicates how
many values to add before the contents of `tensor` in that dimension, and
`paddings[D, 1]` indicates how many values to add after the contents of
`tensor` in that dimension. If `mode` is ""REFLECT"" then both `paddings[D, 0]`
and `paddings[D, 1]` must be no greater than `tensor.dim_size(D) - 1`. If
`mode` is ""SYMMETRIC"" then both `paddings[D, 0]` and `paddings[D, 1]` must be
no greater than `tensor.dim_size(D)`.
The padded size of each dimension D of the output is:
`paddings[D, 0] + tensor.dim_size(D) + paddings[D, 1]`
For example:
```python
t = tf.constant([[1, 2, 3], [4, 5, 6]])
paddings = tf.constant([[1, 1,], [2, 2]])
# 'constant_values' is 0.
# rank of 't' is 2.
tf.pad(t, paddings, ""CONSTANT"") # [[0, 0, 0, 0, 0, 0, 0],
# [0, 0, 1, 2, 3, 0, 0],
# [0, 0, 4, 5, 6, 0, 0],
# [0, 0, 0, 0, 0, 0, 0]]
tf.pad(t, paddings, ""REFLECT"") # [[6, 5, 4, 5, 6, 5, 4],
# [3, 2, 1, 2, 3, 2, 1],
# [6, 5, 4, 5, 6, 5, 4],
# [3, 2, 1, 2, 3, 2, 1]]
tf.pad(t, paddings, ""SYMMETRIC"") # [[2, 1, 1, 2, 3, 3, 2],
# [2, 1, 1, 2, 3, 3, 2],
# [5, 4, 4, 5, 6, 6, 5],
# [5, 4, 4, 5, 6, 6, 5]]
```
Args:
tensor: A `Tensor`.
paddings: A `Tensor` of type `int32`.
mode: One of ""CONSTANT"", ""REFLECT"", or ""SYMMETRIC"" (case-insensitive)
constant_values: In ""CONSTANT"" mode, the scalar pad value to use. Must be
same type as `tensor`.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `tensor`.
Raises:
ValueError: When mode is not one of ""CONSTANT"", ""REFLECT"", or ""SYMMETRIC""."
8084,pad,tensorflow/tensorflow/python/ops/array_ops.py,3309,function,"Pads a tensor.
This operation pads a `tensor` according to the `paddings` you specify.
`paddings` is an integer tensor with shape `[n, 2]`, where n is the rank of
`tensor`. For each dimension D of `input`, `paddings[D, 0]` indicates how
many values to add before the contents of `tensor` in that dimension, and
`paddings[D, 1]` indicates how many values to add after the contents of
`tensor` in that dimension. If `mode` is ""REFLECT"" then both `paddings[D, 0]`
and `paddings[D, 1]` must be no greater than `tensor.dim_size(D) - 1`. If
`mode` is ""SYMMETRIC"" then both `paddings[D, 0]` and `paddings[D, 1]` must be
no greater than `tensor.dim_size(D)`.
The padded size of each dimension D of the output is:
`paddings[D, 0] + tensor.dim_size(D) + paddings[D, 1]`
For example:
```python
t = tf.constant([[1, 2, 3], [4, 5, 6]])
paddings = tf.constant([[1, 1,], [2, 2]])
# 'constant_values' is 0.
# rank of 't' is 2.
tf.pad(t, paddings, ""CONSTANT"") # [[0, 0, 0, 0, 0, 0, 0],
# [0, 0, 1, 2, 3, 0, 0],
# [0, 0, 4, 5, 6, 0, 0],
# [0, 0, 0, 0, 0, 0, 0]]
tf.pad(t, paddings, ""REFLECT"") # [[6, 5, 4, 5, 6, 5, 4],
# [3, 2, 1, 2, 3, 2, 1],
# [6, 5, 4, 5, 6, 5, 4],
# [3, 2, 1, 2, 3, 2, 1]]
tf.pad(t, paddings, ""SYMMETRIC"") # [[2, 1, 1, 2, 3, 3, 2],
# [2, 1, 1, 2, 3, 3, 2],
# [5, 4, 4, 5, 6, 6, 5],
# [5, 4, 4, 5, 6, 6, 5]]
```
Args:
tensor: A `Tensor`.
paddings: A `Tensor` of type `int32`.
mode: One of ""CONSTANT"", ""REFLECT"", or ""SYMMETRIC"" (case-insensitive)
name: A name for the operation (optional).
constant_values: In ""CONSTANT"" mode, the scalar pad value to use. Must be
same type as `tensor`.
Returns:
A `Tensor`. Has the same type as `tensor`.
Raises:
ValueError: When mode is not one of ""CONSTANT"", ""REFLECT"", or ""SYMMETRIC""."
8085,meshgrid,tensorflow/tensorflow/python/ops/array_ops.py,3427,function,"Broadcasts parameters for evaluation on an N-D grid.
Given N one-dimensional coordinate arrays `*args`, returns a list `outputs`
of N-D coordinate arrays for evaluating expressions on an N-D grid.
Notes:
`meshgrid` supports cartesian ('xy') and matrix ('ij') indexing conventions.
When the `indexing` argument is set to 'xy' (the default), the broadcasting
instructions for the first two dimensions are swapped.
Examples:
Calling `X, Y = meshgrid(x, y)` with the tensors
```python
x = [1, 2, 3]
y = [4, 5, 6]
X, Y = tf.meshgrid(x, y)
# X = [[1, 2, 3],
# [1, 2, 3],
# [1, 2, 3]]
# Y = [[4, 4, 4],
# [5, 5, 5],
# [6, 6, 6]]
```
Args:
*args: `Tensor`s with rank 1.
**kwargs:
- indexing: Either 'xy' or 'ij' (optional, default: 'xy').
- name: A name for the operation (optional).
Returns:
outputs: A list of N `Tensor`s with rank N.
Raises:
TypeError: When no keyword arguments (kwargs) are passed.
ValueError: When indexing keyword argument is not one of `xy` or `ij`."
8086,edit_distance,tensorflow/tensorflow/python/ops/array_ops.py,3571,function,"Computes the Levenshtein distance between sequences.
This operation takes variable-length sequences (`hypothesis` and `truth`),
each provided as a `SparseTensor`, and computes the Levenshtein distance.
You can normalize the edit distance by length of `truth` by setting
`normalize` to true.
For example:
Given the following input,
* `hypothesis` is a `tf.SparseTensor` of shape `[2, 1, 1]`
* `truth` is a `tf.SparseTensor` of shape `[2, 2, 2]`
>>> hypothesis = tf.SparseTensor(
... [[0, 0, 0],
... [1, 0, 0]],
... [""a"", ""b""],
... (2, 1, 1))
>>> truth = tf.SparseTensor(
... [[0, 1, 0],
... [1, 0, 0],
... [1, 0, 1],
... [1, 1, 0]],
... [""a"", ""b"", ""c"", ""a""],
... (2, 2, 2))
>>> tf.edit_distance(hypothesis, truth, normalize=True)
<tf.Tensor: shape=(2, 2), dtype=float32, numpy=
array([[inf, 1. ],
[0.5, 1. ]], dtype=float32)>
The operaton returns a dense Tensor of shape `[2, 2]` with
edit distances normalized by `truth` lengths.
**Note**: It is possible to calculate edit distance between two
sparse tensors with variable-length values. However, attempting to create
them while eager execution is enabled will result in a `ValueError`.
For the following inputs,
```python
# 'hypothesis' is a tensor of shape `[2, 1]` with variable-length values:
# (0,0) = [""a""]
# (1,0) = [""b""]
hypothesis = tf.sparse.SparseTensor(
[[0, 0, 0],
[1, 0, 0]],
[""a"", ""b""],
(2, 1, 1))
# 'truth' is a tensor of shape `[2, 2]` with variable-length values:
# (0,0) = []
# (0,1) = [""a""]
# (1,0) = [""b"", ""c""]
# (1,1) = [""a""]
truth = tf.sparse.SparseTensor(
[[0, 1, 0],
[1, 0, 0],
[1, 0, 1],
[1, 1, 0]],
[""a"", ""b"", ""c"", ""a""],
(2, 2, 2))
normalize = True
# The output would be a dense Tensor of shape `(2,)`, with edit distances
noramlized by 'truth' lengths.
# output => array([0., 0.5], dtype=float32)
```
Args:
hypothesis: A `SparseTensor` containing hypothesis sequences.
truth: A `SparseTensor` containing truth sequences.
normalize: A `bool`. If `True`, normalizes the Levenshtein distance by
length of `truth.`
name: A name for the operation (optional).
Returns:
A dense `Tensor` with rank `R - 1`, where R is the rank of the
`SparseTensor` inputs `hypothesis` and `truth`.
Raises:
TypeError: If either `hypothesis` or `truth` are not a `SparseTensor`."
8087,required_space_to_batch_paddings,tensorflow/tensorflow/python/ops/array_ops.py,3711,function,"Calculate padding required to make block_shape divide input_shape.
This function can be used to calculate a suitable paddings argument for use
with space_to_batch_nd and batch_to_space_nd.
Args:
input_shape: int32 Tensor of shape [N].
block_shape: int32 Tensor of shape [N].
base_paddings: Optional int32 Tensor of shape [N, 2]. Specifies the minimum
amount of padding to use. All elements must be >= 0. If not specified,
defaults to 0.
name: string. Optional name prefix.
Returns:
(paddings, crops), where:
`paddings` and `crops` are int32 Tensors of rank 2 and shape [N, 2]
satisfying:
paddings[i, 0] = base_paddings[i, 0].
0 <= paddings[i, 1] - base_paddings[i, 1] < block_shape[i]
(input_shape[i] + paddings[i, 0] + paddings[i, 1]) % block_shape[i] == 0
crops[i, 0] = 0
crops[i, 1] = paddings[i, 1] - base_paddings[i, 1]
Raises: ValueError if called with incompatible shapes."
8088,space_to_batch,tensorflow/tensorflow/python/ops/array_ops.py,3792,function,
8089,space_to_batch_v2,tensorflow/tensorflow/python/ops/array_ops.py,3815,function,
8090,space_to_depth,tensorflow/tensorflow/python/ops/array_ops.py,3825,function,
8091,space_to_depth_v2,tensorflow/tensorflow/python/ops/array_ops.py,3834,function,
8092,depth_to_space,tensorflow/tensorflow/python/ops/array_ops.py,3844,function,
8093,depth_to_space_v2,tensorflow/tensorflow/python/ops/array_ops.py,3853,function,
8094,batch_to_space,tensorflow/tensorflow/python/ops/array_ops.py,3862,function,
8095,batch_to_space_v2,tensorflow/tensorflow/python/ops/array_ops.py,3880,function,"BatchToSpace for N-D tensors of type T.
This operation reshapes the ""batch"" dimension 0 into `M + 1` dimensions of
shape `block_shape + [batch]`, interleaves these blocks back into the grid
defined by the spatial dimensions `[1, ..., M]`, to obtain a result with the
same rank as the input. The spatial dimensions of this intermediate result
are then optionally cropped according to `crops` to produce the output. This
is the reverse of SpaceToBatch (see `tf.space_to_batch`).
Args:
input: A N-D `Tensor` with shape `input_shape = [batch] + spatial_shape +
remaining_shape`, where `spatial_shape` has M dimensions.
block_shape: A 1-D `Tensor` with shape [M]. Must be one of the following
types: `int32`, `int64`. All values must be >= 1. For backwards
compatibility with TF 1.0, this parameter may be an int, in which case it
is converted to
`numpy.array([block_shape, block_shape],
dtype=numpy.int64)`.
crops: A 2-D `Tensor` with shape `[M, 2]`. Must be one of the
following types: `int32`, `int64`. All values must be >= 0.
`crops[i] = [crop_start, crop_end]` specifies the amount to crop from
input dimension `i + 1`, which corresponds to spatial dimension `i`.
It is required that
`crop_start[i] + crop_end[i] <= block_shape[i] * input_shape[i + 1]`.
This operation is equivalent to the following steps:
1. Reshape `input` to `reshaped` of shape: [block_shape[0], ...,
block_shape[M-1], batch / prod(block_shape), input_shape[1], ...,
input_shape[N-1]]
2. Permute dimensions of `reshaped` to produce `permuted` of shape
[batch / prod(block_shape), input_shape[1], block_shape[0], ...,
input_shape[M], block_shape[M-1], input_shape[M+1],
..., input_shape[N-1]]
3. Reshape `permuted` to produce `reshaped_permuted` of shape
[batch / prod(block_shape), input_shape[1] * block_shape[0], ...,
input_shape[M] * block_shape[M-1], input_shape[M+1], ...,
input_shape[N-1]]
4. Crop the start and end of dimensions `[1, ..., M]` of
`reshaped_permuted` according to `crops` to produce the output
of shape:
[batch / prod(block_shape), input_shape[1] *
block_shape[0] - crops[0,0] - crops[0,1], ..., input_shape[M] *
block_shape[M-1] - crops[M-1,0] - crops[M-1,1], input_shape[M+1],
..., input_shape[N-1]]
name: A name for the operation (optional).
Examples:
(1) For the following input of shape `[4, 1, 1, 1]`,
`block_shape = [2, 2]`, and `crops = [[0, 0], [0, 0]]`:
```python
[[[[1]]],
[[[2]]],
[[[3]]],
[[[4]]]]
```
The output tensor has shape `[1, 2, 2, 1]` and value:
```
x = [[[[1], [2]],
[[3], [4]]]]
```
(2) For the following input of shape `[4, 1, 1, 3]`,
`block_shape = [2, 2]`, and `crops = [[0, 0], [0, 0]]`:
```python
[[[1, 2, 3]],
[[4, 5, 6]],
[[7, 8, 9]],
[[10, 11, 12]]]
```
The output tensor has shape `[1, 2, 2, 3]` and value:
```python
x = [[[[1, 2, 3], [4, 5, 6 ]],
[[7, 8, 9], [10, 11, 12]]]]
```
(3) For the following
input of shape `[4, 2, 2, 1]`,
`block_shape = [2, 2]`, and `crops = [[0, 0], [0, 0]]`:
```python
x = [[[[1], [3]], [[ 9], [11]]],
[[[2], [4]], [[10], [12]]],
[[[5], [7]], [[13], [15]]],
[[[6], [8]], [[14], [16]]]]
```
The output tensor has shape `[1, 4, 4, 1]` and value:
```python
x = [[[1], [2], [ 3], [ 4]],
[[5], [6], [ 7], [ 8]],
[[9], [10], [11], [12]],
[[13], [14], [15], [16]]]
```
(4) For the following input of shape
`[8, 1, 3, 1]`,
`block_shape = [2, 2]`, and `crops = [[0, 0], [2, 0]]`:
```python
x = [[[[0], [ 1], [ 3]]],
[[[0], [ 9], [11]]],
[[[0], [ 2], [ 4]]],
[[[0], [10], [12]]],
[[[0], [ 5], [ 7]]],
[[[0], [13], [15]]],
[[[0], [ 6], [ 8]]],
[[[0], [14], [16]]]]
```
The output tensor has shape `[2, 2, 4, 1]` and value:
```python
x = [[[[ 1], [ 2], [ 3], [ 4]],
[[ 5], [ 6], [ 7], [ 8]]],
[[[ 9], [10], [11], [12]],
[[13], [14], [15], [16]]]]
```
Returns:
A `Tensor`. Has the same type as `input`."
8096,one_hot,tensorflow/tensorflow/python/ops/array_ops.py,4018,function,"Returns a one-hot tensor.
See also `tf.fill`, `tf.eye`.
The locations represented by indices in `indices` take value `on_value`,
while all other locations take value `off_value`.
`on_value` and `off_value` must have matching data types. If `dtype` is also
provided, they must be the same data type as specified by `dtype`.
If `on_value` is not provided, it will default to the value `1` with type
`dtype`
If `off_value` is not provided, it will default to the value `0` with type
`dtype`
If the input `indices` is rank `N`, the output will have rank `N+1`. The
new axis is created at dimension `axis` (default: the new axis is appended
at the end).
If `indices` is a scalar the output shape will be a vector of length `depth`
If `indices` is a vector of length `features`, the output shape will be:
```
features x depth if axis == -1
depth x features if axis == 0
```
If `indices` is a matrix (batch) with shape `[batch, features]`, the output
shape will be:
```
batch x features x depth if axis == -1
batch x depth x features if axis == 1
depth x batch x features if axis == 0
```
If `indices` is a RaggedTensor, the 'axis' argument must be positive and refer
to a non-ragged axis. The output will be equivalent to applying 'one_hot' on
the values of the RaggedTensor, and creating a new RaggedTensor from the
result.
If `dtype` is not provided, it will attempt to assume the data type of
`on_value` or `off_value`, if one or both are passed in. If none of
`on_value`, `off_value`, or `dtype` are provided, `dtype` will default to the
value `tf.float32`.
Note: If a non-numeric data type output is desired (`tf.string`, `tf.bool`,
etc.), both `on_value` and `off_value` _must_ be provided to `one_hot`.
For example:
```python
indices = [0, 1, 2]
depth = 3
tf.one_hot(indices, depth) # output: [3 x 3]
# [[1., 0., 0.],
# [0., 1., 0.],
# [0., 0., 1.]]
indices = [0, 2, -1, 1]
depth = 3
tf.one_hot(indices, depth,
on_value=5.0, off_value=0.0,
axis=-1) # output: [4 x 3]
# [[5.0, 0.0, 0.0], # one_hot(0)
# [0.0, 0.0, 5.0], # one_hot(2)
# [0.0, 0.0, 0.0], # one_hot(-1)
# [0.0, 5.0, 0.0]] # one_hot(1)
indices = [[0, 2], [1, -1]]
depth = 3
tf.one_hot(indices, depth,
on_value=1.0, off_value=0.0,
axis=-1) # output: [2 x 2 x 3]
# [[[1.0, 0.0, 0.0], # one_hot(0)
# [0.0, 0.0, 1.0]], # one_hot(2)
# [[0.0, 1.0, 0.0], # one_hot(1)
# [0.0, 0.0, 0.0]]] # one_hot(-1)
indices = tf.ragged.constant([[0, 1], [2]])
depth = 3
tf.one_hot(indices, depth) # output: [2 x None x 3]
# [[[1., 0., 0.],
# [0., 1., 0.]],
# [[0., 0., 1.]]]
```
Args:
indices: A `Tensor` of indices.
depth: A scalar defining the depth of the one hot dimension.
on_value: A scalar defining the value to fill in output when `indices[j]
= i`. (default: 1)
off_value: A scalar defining the value to fill in output when `indices[j]
!= i`. (default: 0)
axis: The axis to fill (default: -1, a new inner-most axis).
dtype: The data type of the output tensor.
name: A name for the operation (optional).
Returns:
output: The one-hot tensor.
Raises:
TypeError: If dtype of either `on_value` or `off_value` don't match `dtype`
TypeError: If dtype of `on_value` and `off_value` don't match one another"
8097,sequence_mask,tensorflow/tensorflow/python/ops/array_ops.py,4196,function,"Returns a mask tensor representing the first N positions of each cell.
If `lengths` has shape `[d_1, d_2, ..., d_n]` the resulting tensor `mask` has
dtype `dtype` and shape `[d_1, d_2, ..., d_n, maxlen]`, with
```
mask[i_1, i_2, ..., i_n, j] = (j < lengths[i_1, i_2, ..., i_n])
```
Examples:
```python
tf.sequence_mask([1, 3, 2], 5) # [[True, False, False, False, False],
# [True, True, True, False, False],
# [True, True, False, False, False]]
tf.sequence_mask([[1, 3],[2,0]]) # [[[True, False, False],
# [True, True, True]],
# [[True, True, False],
# [False, False, False]]]
```
Args:
lengths: integer tensor, all its values <= maxlen.
maxlen: scalar integer tensor, size of last dimension of returned tensor.
Default is the maximum value in `lengths`.
dtype: output type of the resulting tensor.
name: name of the op.
Returns:
A mask tensor of shape `lengths.shape + (maxlen,)`, cast to specified dtype.
Raises:
ValueError: if `maxlen` is not a scalar."
8098,squeeze,tensorflow/tensorflow/python/ops/array_ops.py,4264,function,"Removes dimensions of size 1 from the shape of a tensor.
Given a tensor `input`, this operation returns a tensor of the same type with
all dimensions of size 1 removed. If you don't want to remove all size 1
dimensions, you can remove specific size 1 dimensions by specifying
`axis`.
For example:
>>> # 't' is a tensor of shape [1, 2, 1, 3, 1, 1]
>>> t = tf.ones([1, 2, 1, 3, 1, 1])
>>> print(tf.shape(tf.squeeze(t)).numpy())
[2 3]
Or, to remove specific size 1 dimensions:
>>> # 't' is a tensor of shape [1, 2, 1, 3, 1, 1]
>>> t = tf.ones([1, 2, 1, 3, 1, 1])
>>> print(tf.shape(tf.squeeze(t, [2, 4])).numpy())
[1 2 3 1]
Note: if `input` is a `tf.RaggedTensor`, then this operation takes `O(N)`
time, where `N` is the number of elements in the squeezed dimensions.
Args:
input: A `Tensor`. The `input` to squeeze.
axis: An optional list of `ints`. Defaults to `[]`. If specified, only
squeezes the dimensions listed. The dimension index starts at 0. It is an
error to squeeze a dimension that is not 1. Must be in the range
`[-rank(input), rank(input))`. Must be specified if `input` is a
`RaggedTensor`.
name: A name for the operation (optional).
squeeze_dims: Deprecated keyword argument that is now axis.
Returns:
A `Tensor`. Has the same type as `input`.
Contains the same data as `input`, but has one or more dimensions of
size 1 removed.
Raises:
ValueError: When both `squeeze_dims` and `axis` are specified."
8099,squeeze_v2,tensorflow/tensorflow/python/ops/array_ops.py,4317,function,"Removes dimensions of size 1 from the shape of a tensor.
Given a tensor `input`, this operation returns a tensor of the same type with
all dimensions of size 1 removed. If you don't want to remove all size 1
dimensions, you can remove specific size 1 dimensions by specifying
`axis`.
For example:
```python
# 't' is a tensor of shape [1, 2, 1, 3, 1, 1]
tf.shape(tf.squeeze(t)) # [2, 3]
```
Or, to remove specific size 1 dimensions:
```python
# 't' is a tensor of shape [1, 2, 1, 3, 1, 1]
tf.shape(tf.squeeze(t, [2, 4])) # [1, 2, 3, 1]
```
Unlike the older op `tf.compat.v1.squeeze`, this op does not accept a
deprecated `squeeze_dims` argument.
Note: if `input` is a `tf.RaggedTensor`, then this operation takes `O(N)`
time, where `N` is the number of elements in the squeezed dimensions.
Args:
input: A `Tensor`. The `input` to squeeze.
axis: An optional list of `ints`. Defaults to `[]`. If specified, only
squeezes the dimensions listed. The dimension index starts at 0. It is an
error to squeeze a dimension that is not 1. Must be in the range
`[-rank(input), rank(input))`. Must be specified if `input` is a
`RaggedTensor`.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `input`.
Contains the same data as `input`, but has one or more dimensions of
size 1 removed.
Raises:
ValueError: The input cannot be converted to a tensor, or the specified
axis cannot be squeezed."
8100,where,tensorflow/tensorflow/python/ops/array_ops.py,4369,function,"Return the elements, either from `x` or `y`, depending on the `condition`.
If both `x` and `y` are None, then this operation returns the coordinates of
true elements of `condition`. The coordinates are returned in a 2-D tensor
where the first dimension (rows) represents the number of true elements, and
the second dimension (columns) represents the coordinates of the true
elements. Keep in mind, the shape of the output tensor can vary depending on
how many true values there are in input. Indices are output in row-major
order.
If both non-None, `x` and `y` must have the same shape.
The `condition` tensor must be a scalar if `x` and `y` are scalar.
If `x` and `y` are tensors of higher rank, then `condition` must be either a
vector with size matching the first dimension of `x`, or must have the same
shape as `x`.
The `condition` tensor acts as a mask that chooses, based on the value at each
element, whether the corresponding element / row in the output should be taken
from `x` (if true) or `y` (if false).
If `condition` is a vector and `x` and `y` are higher rank matrices, then it
chooses which row (outer dimension) to copy from `x` and `y`. If `condition`
has the same shape as `x` and `y`, then it chooses which element to copy from
`x` and `y`.
Args:
condition: A `Tensor` of type `bool`
x: A Tensor which may have the same shape as `condition`. If `condition` is
rank 1, `x` may have higher rank, but its first dimension must match the
size of `condition`.
y: A `tensor` with the same shape and type as `x`.
name: A name of the operation (optional)
Returns:
A `Tensor` with the same type and shape as `x`, `y` if they are non-None.
Otherwise, a `Tensor` with shape `(num_true, rank(condition))`.
Raises:
ValueError: When exactly one of `x` or `y` is non-None."
8101,where_v2,tensorflow/tensorflow/python/ops/array_ops.py,4423,function,"Return the elements where `condition` is `True` (multiplexing `x` and `y`).
This operator has two modes: in one mode both `x` and `y` are provided, in
another mode neither are provided. `condition` is always expected to be a
`tf.Tensor` of type `bool`.
#### Retrieving indices of `True` elements
If `x` and `y` are not provided (both are None):
`tf.where` will return the indices of `condition` that are `True`, in
the form of a 2-D tensor with shape (n, d).
(Where n is the number of matching indices in `condition`,
and d is the number of dimensions in `condition`).
Indices are output in row-major order.
>>> tf.where([True, False, False, True])
<tf.Tensor: shape=(2, 1), dtype=int64, numpy=
array([[0],
[3]])>
>>> tf.where([[True, False], [False, True]])
<tf.Tensor: shape=(2, 2), dtype=int64, numpy=
array([[0, 0],
[1, 1]])>
>>> tf.where([[[True, False], [False, True], [True, True]]])
<tf.Tensor: shape=(4, 3), dtype=int64, numpy=
array([[0, 0, 0],
[0, 1, 1],
[0, 2, 0],
[0, 2, 1]])>
#### Multiplexing between `x` and `y`
If `x` and `y` are provided (both have non-None values):
`tf.where` will choose an output shape from the shapes of `condition`, `x`,
and `y` that all three shapes are
[broadcastable](https://docs.scipy.org/doc/numpy/reference/ufuncs.html) to.
The `condition` tensor acts as a mask that chooses whether the corresponding
element / row in the output should be taken from `x`
(if the element in `condition` is True) or `y` (if it is false).
>>> tf.where([True, False, False, True], [1,2,3,4], [100,200,300,400])
<tf.Tensor: shape=(4,), dtype=int32, numpy=array([ 1, 200, 300, 4],
dtype=int32)>
>>> tf.where([True, False, False, True], [1,2,3,4], [100])
<tf.Tensor: shape=(4,), dtype=int32, numpy=array([ 1, 100, 100, 4],
dtype=int32)>
>>> tf.where([True, False, False, True], [1,2,3,4], 100)
<tf.Tensor: shape=(4,), dtype=int32, numpy=array([ 1, 100, 100, 4],
dtype=int32)>
>>> tf.where([True, False, False, True], 1, 100)
<tf.Tensor: shape=(4,), dtype=int32, numpy=array([ 1, 100, 100, 1],
dtype=int32)>
>>> tf.where(True, [1,2,3,4], 100)
<tf.Tensor: shape=(4,), dtype=int32, numpy=array([1, 2, 3, 4],
dtype=int32)>
>>> tf.where(False, [1,2,3,4], 100)
<tf.Tensor: shape=(4,), dtype=int32, numpy=array([100, 100, 100, 100],
dtype=int32)>
Args:
condition: A `tf.Tensor` of type `bool`
x: If provided, a Tensor which is of the same type as `y`, and has a shape
broadcastable with `condition` and `y`.
y: If provided, a Tensor which is of the same type as `x`, and has a shape
broadcastable with `condition` and `x`.
name: A name of the operation (optional).
Returns:
If `x` and `y` are provided:
A `Tensor` with the same type as `x` and `y`, and shape that
is broadcast from `condition`, `x`, and `y`.
Otherwise, a `Tensor` with shape `(num_true, dim_size(condition))`.
Raises:
ValueError: When exactly one of `x` or `y` is non-None, or the shapes
are not all broadcastable."
8102,reverse_sequence,tensorflow/tensorflow/python/ops/array_ops.py,4527,function,"Reverses variable length slices.
This op first slices `input` along the dimension `batch_axis`, and for
each slice `i`, reverses the first `seq_lengths[i]` elements along the
dimension `seq_axis`.
The elements of `seq_lengths` must obey `seq_lengths[i] <=
input.dims[seq_axis]`, and `seq_lengths` must be a vector of length
`input.dims[batch_axis]`.
The output slice `i` along dimension `batch_axis` is then given by
input slice `i`, with the first `seq_lengths[i]` slices along
dimension `seq_axis` reversed.
Example usage:
>>> seq_lengths = [7, 2, 3, 5]
>>> input = [[1, 2, 3, 4, 5, 0, 0, 0], [1, 2, 0, 0, 0, 0, 0, 0],
... [1, 2, 3, 4, 0, 0, 0, 0], [1, 2, 3, 4, 5, 6, 7, 8]]
>>> output = tf.reverse_sequence(input, seq_lengths, seq_axis=1, batch_axis=0)
>>> output
<tf.Tensor: shape=(4, 8), dtype=int32, numpy=
array([[0, 0, 5, 4, 3, 2, 1, 0],
[2, 1, 0, 0, 0, 0, 0, 0],
[3, 2, 1, 4, 0, 0, 0, 0],
[5, 4, 3, 2, 1, 6, 7, 8]], dtype=int32)>
Args:
input: A `Tensor`. The input to reverse.
seq_lengths: A `Tensor`. Must be one of the following types: `int32`,
`int64`. 1-D with length `input.dims(batch_axis)` and `max(seq_lengths) <=
input.dims(seq_axis)`
seq_axis: An `int`. The dimension which is partially reversed.
batch_axis: An optional `int`. Defaults to `0`. The dimension along which
reversal is performed.
name: A name for the operation (optional).
Returns:
A Tensor. Has the same type as input."
8103,reverse_sequence_v2,tensorflow/tensorflow/python/ops/array_ops.py,4587,function,"Reverses variable length slices.
This op first slices `input` along the dimension `batch_axis`, and for
each slice `i`, reverses the first `seq_lengths[i]` elements along the
dimension `seq_axis`.
The elements of `seq_lengths` must obey `seq_lengths[i] <=
input.dims[seq_axis]`, and `seq_lengths` must be a vector of length
`input.dims[batch_axis]`.
The output slice `i` along dimension `batch_axis` is then given by
input slice `i`, with the first `seq_lengths[i]` slices along
dimension `seq_axis` reversed.
Example usage:
>>> seq_lengths = [7, 2, 3, 5]
>>> input = [[1, 2, 3, 4, 5, 0, 0, 0], [1, 2, 0, 0, 0, 0, 0, 0],
... [1, 2, 3, 4, 0, 0, 0, 0], [1, 2, 3, 4, 5, 6, 7, 8]]
>>> output = tf.reverse_sequence(input, seq_lengths, seq_axis=1, batch_axis=0)
>>> output
<tf.Tensor: shape=(4, 8), dtype=int32, numpy=
array([[0, 0, 5, 4, 3, 2, 1, 0],
[2, 1, 0, 0, 0, 0, 0, 0],
[3, 2, 1, 4, 0, 0, 0, 0],
[5, 4, 3, 2, 1, 6, 7, 8]], dtype=int32)>
Args:
input: A `Tensor`. The input to reverse.
seq_lengths: A `Tensor`. Must be one of the following types: `int32`,
`int64`. 1-D with length `input.dims(batch_axis)` and `max(seq_lengths) <=
input.dims(seq_axis)`
seq_axis: An `int`. The dimension which is partially reversed.
batch_axis: An optional `int`. Defaults to `0`. The dimension along which
reversal is performed.
name: A name for the operation (optional).
Returns:
A Tensor. Has the same type as input."
8104,gather,tensorflow/tensorflow/python/ops/array_ops.py,4644,function,"Gather slices from params axis `axis` according to indices.
Gather slices from params axis `axis` according to `indices`. `indices` must
be an integer tensor of any dimension (usually 0-D or 1-D).
For 0-D (scalar) `indices`:
$$\begin{align*}
output[p_0, ..., p_{axis-1}, && &&& p_{axis + 1}, ..., p_{N-1}] = \\
params[p_0, ..., p_{axis-1}, && indices, &&& p_{axis + 1}, ..., p_{N-1}]
\end{align*}$$
Where *N* = `ndims(params)`.
For 1-D (vector) `indices` with `batch_dims=0`:
$$\begin{align*}
output[p_0, ..., p_{axis-1}, && &i, &&p_{axis + 1}, ..., p_{N-1}] =\\
params[p_0, ..., p_{axis-1}, && indices[&i], &&p_{axis + 1}, ..., p_{N-1}]
\end{align*}$$
In the general case, produces an output tensor where:
$$\begin{align*}
output[p_0, &..., p_{axis-1}, &
&i_{B}, ..., i_{M-1}, &
p_{axis + 1}, &..., p_{N-1}] = \\
params[p_0, &..., p_{axis-1}, &
indices[p_0, ..., p_{B-1}, &i_{B}, ..., i_{M-1}], &
p_{axis + 1}, &..., p_{N-1}]
\end{align*}$$
Where *N* = `ndims(params)`, *M* = `ndims(indices)`, and *B* = `batch_dims`.
Note that `params.shape[:batch_dims]` must be identical to
`indices.shape[:batch_dims]`.
The shape of the output tensor is:
> `output.shape = params.shape[:axis] + indices.shape[batch_dims:] +
> params.shape[axis + 1:]`.
Note that on CPU, if an out of bound index is found, an error is returned.
On GPU, if an out of bound index is found, a 0 is stored in the corresponding
output value.
See also `tf.gather_nd`.
<div style=""width:70%; margin:auto; margin-bottom:10px; margin-top:20px;"">
<img style=""width:100%"" src=""https://www.tensorflow.org/images/Gather.png""
alt>
</div>
Args:
params: The `Tensor` from which to gather values. Must be at least rank
`axis + 1`.
indices: The index `Tensor`. Must be one of the following types: `int32`,
`int64`. Must be in range `[0, params.shape[axis])`.
validate_indices: Deprecated, does nothing.
axis: A `Tensor`. Must be one of the following types: `int32`, `int64`. The
`axis` in `params` to gather `indices` from. Must be greater than or equal
to `batch_dims`. Defaults to the first non-batch dimension. Supports
negative indexes.
batch_dims: An `integer`. The number of batch dimensions. Must be less
than or equal to `rank(indices)`.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `params`."
8105,gather_v2,tensorflow/tensorflow/python/ops/array_ops.py,4736,function,
8106,batch_gather,tensorflow/tensorflow/python/ops/array_ops.py,4759,function,Gather slices from params according to indices with leading batch dims.
8107,gather_nd,tensorflow/tensorflow/python/ops/array_ops.py,4895,function,"Gather slices from `params` into a Tensor with shape specified by `indices`.
`indices` is an K-dimensional integer tensor, best thought of as a
(K-1)-dimensional tensor of indices into `params`, where each element defines
a slice of `params`:
output[\\(i_0, ..., i_{K-2}\\)] = params[indices[\\(i_0, ..., i_{K-2}\\)]]
Whereas in `tf.gather` `indices` defines slices into the first
dimension of `params`, in `tf.gather_nd`, `indices` defines slices into the
first `N` dimensions of `params`, where `N = indices.shape[-1]`.
The last dimension of `indices` can be at most the rank of
`params`:
indices.shape[-1] <= params.rank
The last dimension of `indices` corresponds to elements
(if `indices.shape[-1] == params.rank`) or slices
(if `indices.shape[-1] < params.rank`) along dimension `indices.shape[-1]`
of `params`. The output tensor has shape
indices.shape[:-1] + params.shape[indices.shape[-1]:]
Additionally both 'params' and 'indices' can have M leading batch
dimensions that exactly match. In this case 'batch_dims' must be M.
Note that on CPU, if an out of bound index is found, an error is returned.
On GPU, if an out of bound index is found, a 0 is stored in the
corresponding output value.
Some examples below.
Simple indexing into a matrix:
```python
indices = [[0, 0], [1, 1]]
params = [['a', 'b'], ['c', 'd']]
output = ['a', 'd']
```
Slice indexing into a matrix:
```python
indices = [[1], [0]]
params = [['a', 'b'], ['c', 'd']]
output = [['c', 'd'], ['a', 'b']]
```
Indexing into a 3-tensor:
```python
indices = [[1]]
params = [[['a0', 'b0'], ['c0', 'd0']],
[['a1', 'b1'], ['c1', 'd1']]]
output = [[['a1', 'b1'], ['c1', 'd1']]]
indices = [[0, 1], [1, 0]]
params = [[['a0', 'b0'], ['c0', 'd0']],
[['a1', 'b1'], ['c1', 'd1']]]
output = [['c0', 'd0'], ['a1', 'b1']]
indices = [[0, 0, 1], [1, 0, 1]]
params = [[['a0', 'b0'], ['c0', 'd0']],
[['a1', 'b1'], ['c1', 'd1']]]
output = ['b0', 'b1']
```
The examples below are for the case when only indices have leading extra
dimensions. If both 'params' and 'indices' have leading batch dimensions, use
the 'batch_dims' parameter to run gather_nd in batch mode.
Batched indexing into a matrix:
```python
indices = [[[0, 0]], [[0, 1]]]
params = [['a', 'b'], ['c', 'd']]
output = [['a'], ['b']]
```
Batched slice indexing into a matrix:
```python
indices = [[[1]], [[0]]]
params = [['a', 'b'], ['c', 'd']]
output = [[['c', 'd']], [['a', 'b']]]
```
Batched indexing into a 3-tensor:
```python
indices = [[[1]], [[0]]]
params = [[['a0', 'b0'], ['c0', 'd0']],
[['a1', 'b1'], ['c1', 'd1']]]
output = [[[['a1', 'b1'], ['c1', 'd1']]],
[[['a0', 'b0'], ['c0', 'd0']]]]
indices = [[[0, 1], [1, 0]], [[0, 0], [1, 1]]]
params = [[['a0', 'b0'], ['c0', 'd0']],
[['a1', 'b1'], ['c1', 'd1']]]
output = [[['c0', 'd0'], ['a1', 'b1']],
[['a0', 'b0'], ['c1', 'd1']]]
indices = [[[0, 0, 1], [1, 0, 1]], [[0, 1, 1], [1, 1, 0]]]
params = [[['a0', 'b0'], ['c0', 'd0']],
[['a1', 'b1'], ['c1', 'd1']]]
output = [['b0', 'b1'], ['d0', 'c1']]
```
Examples with batched 'params' and 'indices':
```python
batch_dims = 1
indices = [[1], [0]]
params = [[['a0', 'b0'], ['c0', 'd0']],
[['a1', 'b1'], ['c1', 'd1']]]
output = [['c0', 'd0'], ['a1', 'b1']]
batch_dims = 1
indices = [[[1]], [[0]]]
params = [[['a0', 'b0'], ['c0', 'd0']],
[['a1', 'b1'], ['c1', 'd1']]]
output = [[['c0', 'd0']], [['a1', 'b1']]]
batch_dims = 1
indices = [[[1, 0]], [[0, 1]]]
params = [[['a0', 'b0'], ['c0', 'd0']],
[['a1', 'b1'], ['c1', 'd1']]]
output = [['c0'], ['b1']]
```
See also `tf.gather`.
Args:
params: A `Tensor`. The tensor from which to gather values.
indices: A `Tensor`. Must be one of the following types: `int32`, `int64`.
Index tensor.
name: A name for the operation (optional).
batch_dims: An integer or a scalar 'Tensor'. The number of batch dimensions.
Returns:
A `Tensor`. Has the same type as `params`."
8108,gather_nd_v2,tensorflow/tensorflow/python/ops/array_ops.py,5058,function,
8109,batch_gather_nd,tensorflow/tensorflow/python/ops/array_ops.py,5065,function,gather_nd implementation with batch support.
8110,quantize_v2,tensorflow/tensorflow/python/ops/array_ops.py,5153,function,
8111,quantize,tensorflow/tensorflow/python/ops/array_ops.py,5204,function,Quantize the input tensor.
8112,dequantize,tensorflow/tensorflow/python/ops/array_ops.py,5244,function,
8113,quantize_and_dequantize,tensorflow/tensorflow/python/ops/array_ops.py,5279,function,"Quantizes then dequantizes a tensor.
Args:
input: A `Tensor` to quantize and dequantize.
input_min: If range_given=True, the minimum input value, that needs to be
represented in the quantized representation. If axis is specified, this
should be a vector of minimum values for each slice along axis.
input_max: If range_given=True, the maximum input value that needs to be
represented in the quantized representation. If axis is specified, this
should be a vector of maximum values for each slice along axis.
signed_input: True if the quantization is signed or unsigned.
num_bits: The bitwidth of the quantization.
range_given: If true use `input_min` and `input_max` for the range of the
input, otherwise determine min and max from the input `Tensor`.
round_mode: Rounding mode when rounding from float values to quantized ones.
one of ['HALF_TO_EVEN', 'HALF_UP']
name: Optional name for the operation.
narrow_range: If true, then the absolute value of the quantized minimum
value is the same as the quantized maximum value, instead of 1 greater.
i.e. for 8 bit quantization, the minimum value is -127 instead of -128.
axis: Integer. If specified, refers to a dimension of the input tensor, such
that quantization will be per slice along that dimension.
Returns:
A `Tensor`. Each element is the result of quantizing and dequantizing the
corresponding element of `input`."
8114,searchsorted,tensorflow/tensorflow/python/ops/array_ops.py,5339,function,"Searches input tensor for values on the innermost dimension.
A 2-D example:
```
sorted_sequence = [[0, 3, 9, 9, 10],
[1, 2, 3, 4, 5]]
values = [[2, 4, 9],
[0, 2, 6]]
result = searchsorted(sorted_sequence, values, side=""left"")
result == [[1, 2, 2],
[0, 1, 5]]
result = searchsorted(sorted_sequence, values, side=""right"")
result == [[1, 2, 4],
[0, 2, 5]]
```
Args:
sorted_sequence: N-D `Tensor` containing a sorted sequence.
values: N-D `Tensor` containing the search values.
side: 'left' or 'right'; 'left' corresponds to lower_bound and 'right' to
upper_bound.
out_type: The output type (`int32` or `int64`). Default is `tf.int32`.
name: Optional name for the operation.
Returns:
An N-D `Tensor` the size of values containing the result of applying either
lower_bound or upper_bound (depending on side) to each value. The result
is not a global index to the entire `Tensor`, but the index in the last
dimension.
Raises:
ValueError: If the last dimension of `sorted_sequence >= 2^31-1` elements.
If the total size of values exceeds `2^31 - 1` elements.
If the first `N-1` dimensions of the two tensors don't match."
8115,extract_image_patches_v2,tensorflow/tensorflow/python/ops/array_ops.py,5404,function,"Extract `patches` from `images`.
This op collects patches from the input image, as if applying a
convolution. All extracted patches are stacked in the depth (last) dimension
of the output.
Specifically, the op extracts patches of shape `sizes` which are `strides`
apart in the input image. The output is subsampled using the `rates` argument,
in the same manner as ""atrous"" or ""dilated"" convolutions.
The result is a 4D tensor which is indexed by batch, row, and column.
`output[i, x, y]` contains a flattened patch of size `sizes[1], sizes[2]`
which is taken from the input starting at
`images[i, x*strides[1], y*strides[2]]`.
Each output patch can be reshaped to `sizes[1], sizes[2], depth`, where
`depth` is `images.shape[3]`.
The output elements are taken from the input at intervals given by the `rate`
argument, as in dilated convolutions.
The `padding` argument has no effect on the size of each patch, it determines
how many patches are extracted. If `VALID`, only patches which are fully
contained in the input image are included. If `SAME`, all patches whose
starting point is inside the input are included, and areas outside the input
default to zero.
Example:
```
n = 10
# images is a 1 x 10 x 10 x 1 array that contains the numbers 1 through 100
images = [[[[x * n + y + 1] for y in range(n)] for x in range(n)]]
# We generate two outputs as follows:
# 1. 3x3 patches with stride length 5
# 2. Same as above, but the rate is increased to 2
tf.image.extract_patches(images=images,
sizes=[1, 3, 3, 1],
strides=[1, 5, 5, 1],
rates=[1, 1, 1, 1],
padding='VALID')
# Yields:
[[[[ 1 2 3 11 12 13 21 22 23]
[ 6 7 8 16 17 18 26 27 28]]
[[51 52 53 61 62 63 71 72 73]
[56 57 58 66 67 68 76 77 78]]]]
```
If we mark the pixels in the input image which are taken for the output with
`*`, we see the pattern:
```
* * * 4 5 * * * 9 10
* * * 14 15 * * * 19 20
* * * 24 25 * * * 29 30
31 32 33 34 35 36 37 38 39 40
41 42 43 44 45 46 47 48 49 50
* * * 54 55 * * * 59 60
* * * 64 65 * * * 69 70
* * * 74 75 * * * 79 80
81 82 83 84 85 86 87 88 89 90
91 92 93 94 95 96 97 98 99 100
```
```
tf.image.extract_patches(images=images,
sizes=[1, 3, 3, 1],
strides=[1, 5, 5, 1],
rates=[1, 2, 2, 1],
padding='VALID')
# Yields:
[[[[ 1 3 5 21 23 25 41 43 45]
[ 6 8 10 26 28 30 46 48 50]]
[[ 51 53 55 71 73 75 91 93 95]
[ 56 58 60 76 78 80 96 98 100]]]]
```
We can again draw the effect, this time using the symbols `*`, `x`, `+` and
`o` to distinguish the patches:
```
* 2 * 4 * x 7 x 9 x
11 12 13 14 15 16 17 18 19 20
* 22 * 24 * x 27 x 29 x
31 32 33 34 35 36 37 38 39 40
* 42 * 44 * x 47 x 49 x
+ 52 + 54 + o 57 o 59 o
61 62 63 64 65 66 67 68 69 70
+ 72 + 74 + o 77 o 79 o
81 82 83 84 85 86 87 88 89 90
+ 92 + 94 + o 97 o 99 o
```
Args:
images: A 4-D Tensor with shape `[batch, in_rows, in_cols, depth]
sizes: The size of the extracted patches. Must be [1, size_rows, size_cols,
1].
strides: A 1-D Tensor of length 4. How far the centers of two consecutive
patches are in the images. Must be: `[1, stride_rows, stride_cols, 1]`.
rates: A 1-D Tensor of length 4. Must be: `[1, rate_rows, rate_cols, 1]`.
This is the input stride, specifying how far two consecutive patch samples
are in the input. Equivalent to extracting patches with `patch_sizes_eff =
patch_sizes + (patch_sizes - 1) * (rates - 1)`, followed by subsampling
them spatially by a factor of `rates`. This is equivalent to `rate` in
dilated (a.k.a. Atrous) convolutions.
padding: The type of padding algorithm to use.
name: A name for the operation (optional).
Returns:
A 4-D Tensor of the same type as the input."
8116,extract_image_patches,tensorflow/tensorflow/python/ops/array_ops.py,5528,function,"Extract patches from images and put them in the ""depth"" output dimension.
Args:
`images`: A `Tensor`. Must be one of the following types: `float32`,
`float64`, `int32`, `uint8`, `int16`, `int8`, `int64`, `bfloat16`,
`uint16`, `half`, `uint32`, `uint64`. 4-D Tensor with shape
`[batch, in_rows, in_cols, depth]`. `ksizes`: A list of `ints` that has
length `>= 4`. The size of the sliding window for each
dimension of `images`. `strides`: A list of `ints` that has length `>= 4`.
1-D of length 4. How far the centers of two consecutive
patches are in the images. Must be:
`[1, stride_rows, stride_cols, 1]`. `rates`: A list of `ints`
that has length `>= 4`. 1-D of length 4. Must be: `[1, rate_rows, rate_cols,
1]`. This is the input stride, specifying how far two consecutive patch
samples are in the input. Equivalent to extracting patches with
`patch_sizes_eff = patch_sizes + (patch_sizes - 1) * (rates - 1)`,
followed by subsampling them spatially by a factor of `rates`. This is
equivalent to `rate` in dilated (a.k.a. Atrous) convolutions.
`padding`: A `string` from: ""SAME"", ""VALID"". The type of padding algorithm
to use.
We specify the size-related attributes as: ``` ksizes = [1, ksize_rows,
ksize_cols, 1] strides = [1, strides_rows, strides_cols, 1] rates = [1,
rates_rows, rates_cols, 1]
name: A name for the operation (optional). ```
Returns:
A Tensor. Has the same type as images."
8117,fingerprint,tensorflow/tensorflow/python/ops/array_ops.py,5575,function,"Generates fingerprint values.
Generates fingerprint values of `data`.
Fingerprint op considers the first dimension of `data` as the batch dimension,
and `output[i]` contains the fingerprint value generated from contents in
`data[i, ...]` for all `i`.
Fingerprint op writes fingerprint values as byte arrays. For example, the
default method `farmhash64` generates a 64-bit fingerprint value at a time.
This 8-byte value is written out as an `tf.uint8` array of size 8, in
little-endian order.
For example, suppose that `data` has data type `tf.int32` and shape (2, 3, 4),
and that the fingerprint method is `farmhash64`. In this case, the output
shape is (2, 8), where 2 is the batch dimension size of `data`, and 8 is the
size of each fingerprint value in bytes. `output[0, :]` is generated from
12 integers in `data[0, :, :]` and similarly `output[1, :]` is generated from
other 12 integers in `data[1, :, :]`.
Note that this op fingerprints the raw underlying buffer, and it does not
fingerprint Tensor's metadata such as data type and/or shape. For example, the
fingerprint values are invariant under reshapes and bitcasts as long as the
batch dimension remain the same:
```python
tf.fingerprint(data) == tf.fingerprint(tf.reshape(data, ...))
tf.fingerprint(data) == tf.fingerprint(tf.bitcast(data, ...))
```
For string data, one should expect `tf.fingerprint(data) !=
tf.fingerprint(tf.string.reduce_join(data))` in general.
Args:
data: A `Tensor`. Must have rank 1 or higher.
method: A `Tensor` of type `tf.string`. Fingerprint method used by this op.
Currently available method is `farmhash64`.
name: A name for the operation (optional).
Returns:
A two-dimensional `Tensor` of type `tf.uint8`. The first dimension equals to
`data`'s first dimension, and the second dimension size depends on the
fingerprint algorithm."
8118,convert_to_int_tensor,tensorflow/tensorflow/python/ops/array_ops.py,5623,function,Converts the given value to an integer Tensor.
8119,get_positive_axis,tensorflow/tensorflow/python/ops/array_ops.py,5634,function,"Validate an `axis` parameter, and normalize it to be positive.
If `ndims` is known (i.e., not `None`), then check that `axis` is in the
range `-ndims <= axis < ndims`, and return `axis` (if `axis >= 0`) or
`axis + ndims` (otherwise).
If `ndims` is not known, and `axis` is positive, then return it as-is.
If `ndims` is not known, and `axis` is negative, then report an error.
Args:
axis: An integer constant
ndims: An integer constant, or `None`
axis_name: The name of `axis` (for error messages).
ndims_name: The name of `ndims` (for error messages).
Returns:
The normalized `axis` value.
Raises:
ValueError: If `axis` is out-of-bounds, or if `axis` is negative and
`ndims is None`."
8120,repeat_with_axis,tensorflow/tensorflow/python/ops/array_ops.py,5680,function,"Repeats elements of `data`.
Args:
data: An `N`-dimensional tensor.
repeats: A 1-D integer tensor specifying how many times each element in
`axis` should be repeated. `len(repeats)` must equal `data.shape[axis]`.
Supports broadcasting from a scalar value.
axis: `int`. The axis along which to repeat values. Must be less than
`max(N, 1)`.
name: A name for the operation.
Returns:
A tensor with `max(N, 1)` dimensions. Has the same shape as `data`,
except that dimension `axis` has size `sum(repeats)`.
Example usage:
>>> repeat(['a', 'b', 'c'], repeats=[3, 0, 2], axis=0)
<tf.Tensor: shape=(5,), dtype=string,
numpy=array([b'a', b'a', b'a', b'c', b'c'], dtype=object)>
>>> repeat([[1, 2], [3, 4]], repeats=[2, 3], axis=0)
<tf.Tensor: shape=(5, 2), dtype=int32, numpy=
array([[1, 2],
[1, 2],
[3, 4],
[3, 4],
[3, 4]], dtype=int32)>
>>> repeat([[1, 2], [3, 4]], repeats=[2, 3], axis=1)
<tf.Tensor: shape=(2, 5), dtype=int32, numpy=
array([[1, 1, 2, 2, 2],
[3, 3, 4, 4, 4]], dtype=int32)>"
8121,tile_one_dimension,tensorflow/tensorflow/python/ops/array_ops.py,5794,function,Tiles a single dimension of a tensor.
8122,repeat,tensorflow/tensorflow/python/ops/array_ops.py,5822,function,"Repeat elements of `input`.
See also `tf.concat`, `tf.stack`, `tf.tile`.
Args:
input: An `N`-dimensional Tensor.
repeats: An 1-D `int` Tensor. The number of repetitions for each element.
repeats is broadcasted to fit the shape of the given axis. `len(repeats)`
must equal `input.shape[axis]` if axis is not None.
axis: An int. The axis along which to repeat values. By default (axis=None),
use the flattened input array, and return a flat output array.
name: A name for the operation.
Returns:
A Tensor which has the same shape as `input`, except along the given axis.
If axis is None then the output array is flattened to match the flattened
input array.
Example usage:
>>> repeat(['a', 'b', 'c'], repeats=[3, 0, 2], axis=0)
<tf.Tensor: shape=(5,), dtype=string,
numpy=array([b'a', b'a', b'a', b'c', b'c'], dtype=object)>
>>> repeat([[1, 2], [3, 4]], repeats=[2, 3], axis=0)
<tf.Tensor: shape=(5, 2), dtype=int32, numpy=
array([[1, 2],
[1, 2],
[3, 4],
[3, 4],
[3, 4]], dtype=int32)>
>>> repeat([[1, 2], [3, 4]], repeats=[2, 3], axis=1)
<tf.Tensor: shape=(2, 5), dtype=int32, numpy=
array([[1, 1, 2, 2, 2],
[3, 3, 4, 4, 4]], dtype=int32)>
>>> repeat(3, repeats=4)
<tf.Tensor: shape=(4,), dtype=int32, numpy=array([3, 3, 3, 3], dtype=int32)>
>>> repeat([[1,2], [3,4]], repeats=2)
<tf.Tensor: shape=(8,), dtype=int32,
numpy=array([1, 1, 2, 2, 3, 3, 4, 4], dtype=int32)>"
8123,batch_norm_op,tensorflow/tensorflow/python/ops/batch_norm_benchmark.py,40,function,Fused kernel for batch normalization.
8124,batch_norm_py,tensorflow/tensorflow/python/ops/batch_norm_benchmark.py,55,function,Python implementation of batch normalization.
8125,batch_norm_slow,tensorflow/tensorflow/python/ops/batch_norm_benchmark.py,61,function,
8126,build_graph,tensorflow/tensorflow/python/ops/batch_norm_benchmark.py,68,function,"Build a graph containing a sequence of batch normalizations.
Args:
device: string, the device to run on.
input_shape: shape of the input tensor.
axes: axes that are to be normalized across.
num_layers: number of batch normalization layers in the graph.
mode: ""op"", ""py"" or ""slow"" depending on the implementation.
scale: scale after normalization.
train: if true, also run backprop.
Returns:
An array of tensors to run()"
8127,print_difference,tensorflow/tensorflow/python/ops/batch_norm_benchmark.py,117,function,Print the difference in timing between two runs.
8128,BatchNormBenchmark,tensorflow/tensorflow/python/ops/batch_norm_benchmark.py,123,class,Benchmark batch normalization.
8129,benchmark_batch_norm,tensorflow/tensorflow/python/ops/batch_norm_benchmark.py,177,method,
8130,batch_function,tensorflow/tensorflow/python/ops/batch_ops.py,32,function,"Batches the computation done by the decorated function.
So, for example, in the following code
```python
@batch_function(1, 2, 3)
def layer(a):
return tf.matmul(a, a)
b = layer(w)
```
if more than one session.run call is simultaneously trying to compute `b`
the values of `w` will be gathered, non-deterministically concatenated
along the first axis, and only one thread will run the computation. See the
documentation of the `Batch` op for more details.
Assumes that all arguments of the decorated function are Tensors which will
be batched along their first dimension.
SparseTensor is not supported. The return value of the decorated function
must be a Tensor or a list/tuple of Tensors.
Args:
num_batch_threads: Number of scheduling threads for processing batches
of work. Determines the number of batches processed in parallel.
max_batch_size: Batch sizes will never be bigger than this.
batch_timeout_micros: Maximum number of microseconds to wait before
outputting an incomplete batch.
allowed_batch_sizes: Optional list of allowed batch sizes. If left empty,
does nothing. Otherwise, supplies a list of batch sizes, causing the op
to pad batches up to one of those sizes. The entries must increase
monotonically, and the final entry must equal max_batch_size.
max_enqueued_batches: The maximum depth of the batch queue. Defaults to 10.
autograph: Whether to use autograph to compile python and eager style code
for efficient graph-mode execution.
Returns:
The decorated function will return the unbatched computation output Tensors."
8131,delayed_plus1,tensorflow/tensorflow/python/ops/batch_ops_test.py,37,function,Sleeps for 100ms then returns x+1.
8132,bincount,tensorflow/tensorflow/python/ops/bincount_ops.py,36,function,"Counts the number of occurrences of each value in an integer array.
If `minlength` and `maxlength` are not given, returns a vector with length
`tf.reduce_max(arr) + 1` if `arr` is non-empty, and length 0 otherwise.
If `weights` are non-None, then index `i` of the output stores the sum of the
value in `weights` at each index where the corresponding value in `arr` is
`i`.
```python
values = tf.constant([1,1,2,3,2,4,4,5])
tf.math.bincount(values) #[0 2 2 1 2 1]
```
Vector length = Maximum element in vector `values` is 5. Adding 1, which is 6
will be the vector length.
Each bin value in the output indicates number of occurrences of the particular
index. Here, index 1 in output has a value 2. This indicates value 1 occurs
two times in `values`.
```python
values = tf.constant([1,1,2,3,2,4,4,5])
weights = tf.constant([1,5,0,1,0,5,4,5])
tf.math.bincount(values, weights=weights) #[0 6 0 1 9 5]
```
Bin will be incremented by the corresponding weight instead of 1.
Here, index 1 in output has a value 6. This is the summation of weights
corresponding to the value in `values`.
**Bin-counting on a certain axis**
This example takes a 2 dimensional input and returns a `Tensor` with
bincounting on each sample.
>>> data = np.array([[1, 2, 3, 0], [0, 0, 1, 2]], dtype=np.int32)
>>> tf.math.bincount(data, axis=-1)
<tf.Tensor: shape=(2, 4), dtype=int32, numpy=
array([[1, 1, 1, 1],
[2, 1, 1, 0]], dtype=int32)>
**Bin-counting with binary_output**
This example gives binary output instead of counting the occurrence.
>>> data = np.array([[1, 2, 3, 0], [0, 0, 1, 2]], dtype=np.int32)
>>> tf.math.bincount(data, axis=-1, binary_output=True)
<tf.Tensor: shape=(2, 4), dtype=int32, numpy=
array([[1, 1, 1, 1],
[1, 1, 1, 0]], dtype=int32)>
Args:
arr: A Tensor, RaggedTensor, or SparseTensor whose values should be counted.
These tensors must have a rank of 2 if `axis=-1`.
weights: If non-None, must be the same shape as arr. For each value in
`arr`, the bin will be incremented by the corresponding weight instead of
1.
minlength: If given, ensures the output has length at least `minlength`,
padding with zeros at the end if necessary.
maxlength: If given, skips values in `arr` that are equal or greater than
`maxlength`, ensuring that the output has length at most `maxlength`.
dtype: If `weights` is None, determines the type of the output bins.
name: A name scope for the associated operations (optional).
axis: The axis to slice over. Axes at and below `axis` will be flattened
before bin counting. Currently, only `0`, and `-1` are supported. If None,
all axes will be flattened (identical to passing `0`).
binary_output: If True, this op will output 1 instead of the number of times
a token appears (equivalent to one_hot + reduce_any instead of one_hot +
reduce_add). Defaults to False.
Returns:
A vector with the same dtype as `weights` or the given `dtype`. The bin
values.
Raises:
`InvalidArgumentError` if negative values are provided as an input."
8133,bincount_v1,tensorflow/tensorflow/python/ops/bincount_ops.py,223,function,"Counts the number of occurrences of each value in an integer array.
If `minlength` and `maxlength` are not given, returns a vector with length
`tf.reduce_max(arr) + 1` if `arr` is non-empty, and length 0 otherwise.
If `weights` are non-None, then index `i` of the output stores the sum of the
value in `weights` at each index where the corresponding value in `arr` is
`i`.
Args:
arr: An int32 tensor of non-negative values.
weights: If non-None, must be the same shape as arr. For each value in
`arr`, the bin will be incremented by the corresponding weight instead of
1.
minlength: If given, ensures the output has length at least `minlength`,
padding with zeros at the end if necessary.
maxlength: If given, skips values in `arr` that are equal or greater than
`maxlength`, ensuring that the output has length at most `maxlength`.
dtype: If `weights` is None, determines the type of the output bins.
Returns:
A vector with the same dtype as `weights` or the given `dtype`. The bin
values."
8134,sparse_bincount,tensorflow/tensorflow/python/ops/bincount_ops.py,255,function,"Count the number of times an integer value appears in a tensor.
This op takes an N-dimensional `Tensor`, `RaggedTensor`, or `SparseTensor`,
and returns an N-dimensional int64 SparseTensor where element
`[i0...i[axis], j]` contains the number of times the value `j` appears in
slice `[i0...i[axis], :]` of the input tensor. Currently, only N=0 and
N=-1 are supported.
Args:
values: A Tensor, RaggedTensor, or SparseTensor whose values should be
counted. These tensors must have a rank of 2 if `axis=-1`.
weights: If non-None, must be the same shape as arr. For each value in
`value`, the bin will be incremented by the corresponding weight instead
of 1.
axis: The axis to slice over. Axes at and below `axis` will be flattened
before bin counting. Currently, only `0`, and `-1` are supported. If None,
all axes will be flattened (identical to passing `0`).
minlength: If given, ensures the output has length at least `minlength`,
padding with zeros at the end if necessary.
maxlength: If given, skips values in `values` that are equal or greater than
`maxlength`, ensuring that the output has length at most `maxlength`.
binary_output: If True, this op will output 1 instead of the number of times
a token appears (equivalent to one_hot + reduce_any instead of one_hot +
reduce_add). Defaults to False.
name: A name for this op.
Returns:
A SparseTensor with `output.shape = values.shape[:axis] + [N]`, where `N` is
* `maxlength` (if set);
* `minlength` (if set, and `minlength > reduce_max(values)`);
* `0` (if `values` is empty);
* `reduce_max(values) + 1` otherwise.
Examples:
**Bin-counting every item in individual batches**
This example takes an input (which could be a Tensor, RaggedTensor, or
SparseTensor) and returns a SparseTensor where the value of (i,j) is the
number of times value j appears in batch i.
>>> data = np.array([[10, 20, 30, 20], [11, 101, 11, 10001]], dtype=np.int64)
>>> output = tf.sparse.bincount(data, axis=-1)
>>> print(output)
SparseTensor(indices=tf.Tensor(
[[ 0 10]
[ 0 20]
[ 0 30]
[ 1 11]
[ 1 101]
[ 1 10001]], shape=(6, 2), dtype=int64),
values=tf.Tensor([1 2 1 2 1 1], shape=(6,), dtype=int64),
dense_shape=tf.Tensor([ 2 10002], shape=(2,), dtype=int64))
**Bin-counting with defined output shape**
This example takes an input (which could be a Tensor, RaggedTensor, or
SparseTensor) and returns a SparseTensor where the value of (i,j) is the
number of times value j appears in batch i. However, all values of j
above 'maxlength' are ignored. The dense_shape of the output sparse tensor
is set to 'minlength'. Note that, while the input is identical to the
example above, the value '10001' in batch item 2 is dropped, and the
dense shape is [2, 500] instead of [2,10002] or [2, 102].
>>> minlength = maxlength = 500
>>> data = np.array([[10, 20, 30, 20], [11, 101, 11, 10001]], dtype=np.int64)
>>> output = tf.sparse.bincount(
... data, axis=-1, minlength=minlength, maxlength=maxlength)
>>> print(output)
SparseTensor(indices=tf.Tensor(
[[ 0 10]
[ 0 20]
[ 0 30]
[ 1 11]
[ 1 101]], shape=(5, 2), dtype=int64),
values=tf.Tensor([1 2 1 2 1], shape=(5,), dtype=int64),
dense_shape=tf.Tensor([ 2 500], shape=(2,), dtype=int64))
**Binary bin-counting**
This example takes an input (which could be a Tensor, RaggedTensor, or
SparseTensor) and returns a SparseTensor where (i,j) is 1 if the value j
appears in batch i at least once and is 0 otherwise. Note that, even though
some values (like 20 in batch 1 and 11 in batch 2) appear more than once,
the 'values' tensor is all 1s.
>>> data = np.array([[10, 20, 30, 20], [11, 101, 11, 10001]], dtype=np.int64)
>>> output = tf.sparse.bincount(data, binary_output=True, axis=-1)
>>> print(output)
SparseTensor(indices=tf.Tensor(
[[ 0 10]
[ 0 20]
[ 0 30]
[ 1 11]
[ 1 101]
[ 1 10001]], shape=(6, 2), dtype=int64),
values=tf.Tensor([1 1 1 1 1 1], shape=(6,), dtype=int64),
dense_shape=tf.Tensor([ 2 10002], shape=(2,), dtype=int64))
**Weighted bin-counting**
This example takes two inputs - a values tensor and a weights tensor. These
tensors must be identically shaped, and have the same row splits or indices
in the case of RaggedTensors or SparseTensors. When performing a weighted
count, the op will output a SparseTensor where the value of (i, j) is the
sum of the values in the weight tensor's batch i in the locations where
the values tensor has the value j. In this case, the output dtype is the
same as the dtype of the weights tensor.
>>> data = np.array([[10, 20, 30, 20], [11, 101, 11, 10001]], dtype=np.int64)
>>> weights = [[2, 0.25, 15, 0.5], [2, 17, 3, 0.9]]
>>> output = tf.sparse.bincount(data, weights=weights, axis=-1)
>>> print(output)
SparseTensor(indices=tf.Tensor(
[[ 0 10]
[ 0 20]
[ 0 30]
[ 1 11]
[ 1 101]
[ 1 10001]], shape=(6, 2), dtype=int64),
values=tf.Tensor([2. 0.75 15. 5. 17. 0.9], shape=(6,), dtype=float32),
dense_shape=tf.Tensor([ 2 10002], shape=(2,), dtype=int64))"
8135,validate_dense_weights,tensorflow/tensorflow/python/ops/bincount_ops.py,454,function,Validates the passed weight tensor or creates an empty one.
8136,validate_sparse_weights,tensorflow/tensorflow/python/ops/bincount_ops.py,468,function,Validates the passed weight tensor or creates an empty one.
8137,validate_ragged_weights,tensorflow/tensorflow/python/ops/bincount_ops.py,502,function,Validates the passed weight tensor or creates an empty one.
8138,PruningMode,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,55,class,Class for working with Pruning modes.
8139,from_str,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,62,method,
8140,QuantileAccumulatorSaveable,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,70,class,SaveableObject implementation for QuantileAccumulator.
8141,restore,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,92,method,
8142,make_save_spec,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,82,method,
8143,QuantileAccumulator,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,99,class,"SaveableObject implementation for QuantileAccumulator.
The bucket boundaries are serialized and deserialized from checkpointing."
8144,initializer,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,137,method,
8145,is_initialized,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,142,method,
8146,saveable,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,146,method,
8147,add_summaries,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,152,method,
8148,flush,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,158,method,
8149,get_bucket_boundaries,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,161,method,
8150,TreeEnsemble,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,210,class,Creates TreeEnsemble resource.
8151,initializer,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,244,method,
8152,is_initialized,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,249,method,
8153,get_stamp_token,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,257,method,Returns the current stamp token of the resource.
8154,get_states,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,264,method,"Returns states of the tree ensemble.
Returns:
stamp_token, num_trees, num_finalized_trees, num_attempted_layers and
range of the nodes in the latest layer."
8155,serialize,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,283,method,"Serializes the ensemble into proto and returns the serialized proto.
Returns:
stamp_token: int64 scalar Tensor to denote the stamp of the resource.
serialized_proto: string scalar Tensor of the serialized proto."
8156,deserialize,tensorflow/tensorflow/python/ops/boosted_trees_ops.py,293,method,"Deserialize the input proto and resets the ensemble from it.
Args:
stamp_token: int64 scalar Tensor to denote the stamp of the resource.
serialized_proto: string scalar Tensor of the serialized proto.
Returns:
Operation (for dependencies)."
8157,uniform_candidate_sampler,tensorflow/tensorflow/python/ops/candidate_sampling_ops.py,36,function,"Samples a set of classes using a uniform base distribution.
This operation randomly samples a tensor of sampled classes
(`sampled_candidates`) from the range of integers `[0, range_max)`.
The elements of `sampled_candidates` are drawn without replacement
(if `unique=True`) or with replacement (if `unique=False`) from
the base distribution.
The base distribution for this operation is the uniform distribution
over the range of integers `[0, range_max)`.
In addition, this operation returns tensors `true_expected_count`
and `sampled_expected_count` representing the number of times each
of the target classes (`true_classes`) and the sampled
classes (`sampled_candidates`) is expected to occur in an average
tensor of sampled classes. These values correspond to `Q(y|x)`
defined in [this
document](http://www.tensorflow.org/extras/candidate_sampling.pdf).
If `unique=True`, then these are post-rejection probabilities and we
compute them approximately.
Args:
true_classes: A `Tensor` of type `int64` and shape `[batch_size,
num_true]`. The target classes.
num_true: An `int`. The number of target classes per training example.
num_sampled: An `int`. The number of classes to randomly sample. The
`sampled_candidates` return value will have shape `[num_sampled]`. If
`unique=True`, `num_sampled` must be less than or equal to `range_max`.
unique: A `bool`. Determines whether all sampled classes in a batch are
unique.
range_max: An `int`. The number of possible classes.
seed: An `int`. An operation-specific seed. Default is 0.
name: A name for the operation (optional).
Returns:
sampled_candidates: A tensor of type `int64` and shape `[num_sampled]`. The
sampled classes, either with possible duplicates (`unique=False`) or all
unique (`unique=True`). In either case, `sampled_candidates` is
independent of the true classes.
true_expected_count: A tensor of type `float`. Same shape as
`true_classes`. The expected counts under the sampling distribution
of each of `true_classes`.
sampled_expected_count: A tensor of type `float`. Same shape as
`sampled_candidates`. The expected counts under the sampling distribution
of each of `sampled_candidates`."
8158,log_uniform_candidate_sampler,tensorflow/tensorflow/python/ops/candidate_sampling_ops.py,99,function,"Samples a set of classes using a log-uniform (Zipfian) base distribution.
This operation randomly samples a tensor of sampled classes
(`sampled_candidates`) from the range of integers `[0, range_max)`.
The elements of `sampled_candidates` are drawn without replacement
(if `unique=True`) or with replacement (if `unique=False`) from
the base distribution.
The base distribution for this operation is an approximately log-uniform
or Zipfian distribution:
`P(class) = (log(class + 2) - log(class + 1)) / log(range_max + 1)`
This sampler is useful when the target classes approximately follow such
a distribution - for example, if the classes represent words in a lexicon
sorted in decreasing order of frequency. If your classes are not ordered by
decreasing frequency, do not use this op.
In addition, this operation returns tensors `true_expected_count`
and `sampled_expected_count` representing the number of times each
of the target classes (`true_classes`) and the sampled
classes (`sampled_candidates`) is expected to occur in an average
tensor of sampled classes. These values correspond to `Q(y|x)`
defined in [this
document](http://www.tensorflow.org/extras/candidate_sampling.pdf).
If `unique=True`, then these are post-rejection probabilities and we
compute them approximately.
Args:
true_classes: A `Tensor` of type `int64` and shape `[batch_size,
num_true]`. The target classes.
num_true: An `int`. The number of target classes per training example.
num_sampled: An `int`. The number of classes to randomly sample.
unique: A `bool`. Determines whether all sampled classes in a batch are
unique.
range_max: An `int`. The number of possible classes.
seed: An `int`. An operation-specific seed. Default is 0.
name: A name for the operation (optional).
Returns:
sampled_candidates: A tensor of type `int64` and shape `[num_sampled]`.
The sampled classes.
true_expected_count: A tensor of type `float`. Same shape as
`true_classes`. The expected counts under the sampling distribution
of each of `true_classes`.
sampled_expected_count: A tensor of type `float`. Same shape as
`sampled_candidates`. The expected counts under the sampling distribution
of each of `sampled_candidates`."
8159,learned_unigram_candidate_sampler,tensorflow/tensorflow/python/ops/candidate_sampling_ops.py,162,function,"Samples a set of classes from a distribution learned during training.
This operation randomly samples a tensor of sampled classes
(`sampled_candidates`) from the range of integers `[0, range_max)`.
The elements of `sampled_candidates` are drawn without replacement
(if `unique=True`) or with replacement (if `unique=False`) from
the base distribution.
The base distribution for this operation is constructed on the fly
during training. It is a unigram distribution over the target
classes seen so far during training. Every integer in `[0, range_max)`
begins with a weight of 1, and is incremented by 1 each time it is
seen as a target class. The base distribution is not saved to checkpoints,
so it is reset when the model is reloaded.
In addition, this operation returns tensors `true_expected_count`
and `sampled_expected_count` representing the number of times each
of the target classes (`true_classes`) and the sampled
classes (`sampled_candidates`) is expected to occur in an average
tensor of sampled classes. These values correspond to `Q(y|x)`
defined in [this
document](http://www.tensorflow.org/extras/candidate_sampling.pdf).
If `unique=True`, then these are post-rejection probabilities and we
compute them approximately.
Args:
true_classes: A `Tensor` of type `int64` and shape `[batch_size,
num_true]`. The target classes.
num_true: An `int`. The number of target classes per training example.
num_sampled: An `int`. The number of classes to randomly sample.
unique: A `bool`. Determines whether all sampled classes in a batch are
unique.
range_max: An `int`. The number of possible classes.
seed: An `int`. An operation-specific seed. Default is 0.
name: A name for the operation (optional).
Returns:
sampled_candidates: A tensor of type `int64` and shape `[num_sampled]`.
The sampled classes.
true_expected_count: A tensor of type `float`. Same shape as
`true_classes`. The expected counts under the sampling distribution
of each of `true_classes`.
sampled_expected_count: A tensor of type `float`. Same shape as
`sampled_candidates`. The expected counts under the sampling distribution
of each of `sampled_candidates`."
8160,fixed_unigram_candidate_sampler,tensorflow/tensorflow/python/ops/candidate_sampling_ops.py,221,function,"Samples a set of classes using the provided (fixed) base distribution.
This operation randomly samples a tensor of sampled classes
(`sampled_candidates`) from the range of integers `[0, range_max)`.
The elements of `sampled_candidates` are drawn without replacement
(if `unique=True`) or with replacement (if `unique=False`) from
the base distribution.
The base distribution is read from a file or passed in as an
in-memory array. There is also an option to skew the distribution by
applying a distortion power to the weights.
In addition, this operation returns tensors `true_expected_count`
and `sampled_expected_count` representing the number of times each
of the target classes (`true_classes`) and the sampled
classes (`sampled_candidates`) is expected to occur in an average
tensor of sampled classes. These values correspond to `Q(y|x)`
defined in [this
document](http://www.tensorflow.org/extras/candidate_sampling.pdf).
If `unique=True`, then these are post-rejection probabilities and we
compute them approximately.
Args:
true_classes: A `Tensor` of type `int64` and shape `[batch_size,
num_true]`. The target classes.
num_true: An `int`. The number of target classes per training example.
num_sampled: An `int`. The number of classes to randomly sample.
unique: A `bool`. Determines whether all sampled classes in a batch are
unique.
range_max: An `int`. The number of possible classes.
vocab_file: Each valid line in this file (which should have a CSV-like
format) corresponds to a valid word ID. IDs are in sequential order,
starting from num_reserved_ids. The last entry in each line is expected
to be a value corresponding to the count or relative probability. Exactly
one of `vocab_file` and `unigrams` needs to be passed to this operation.
distortion: The distortion is used to skew the unigram probability
distribution. Each weight is first raised to the distortion's power
before adding to the internal unigram distribution. As a result,
`distortion = 1.0` gives regular unigram sampling (as defined by the vocab
file), and `distortion = 0.0` gives a uniform distribution.
num_reserved_ids: Optionally some reserved IDs can be added in the range
`[0, num_reserved_ids)` by the users. One use case is that a special
unknown word token is used as ID 0. These IDs will have a sampling
probability of 0.
num_shards: A sampler can be used to sample from a subset of the original
range in order to speed up the whole computation through parallelism. This
parameter (together with `shard`) indicates the number of partitions that
are being used in the overall computation.
shard: A sampler can be used to sample from a subset of the original range
in order to speed up the whole computation through parallelism. This
parameter (together with `num_shards`) indicates the particular partition
number of the operation, when partitioning is being used.
unigrams: A list of unigram counts or probabilities, one per ID in
sequential order. Exactly one of `vocab_file` and `unigrams` should be
passed to this operation.
seed: An `int`. An operation-specific seed. Default is 0.
name: A name for the operation (optional).
Returns:
sampled_candidates: A tensor of type `int64` and shape `[num_sampled]`.
The sampled classes.
true_expected_count: A tensor of type `float`. Same shape as
`true_classes`. The expected counts under the sampling distribution
of each of `true_classes`.
sampled_expected_count: A tensor of type `float`. Same shape as
`sampled_candidates`. The expected counts under the sampling distribution
of each of `sampled_candidates`."
8161,all_candidate_sampler,tensorflow/tensorflow/python/ops/candidate_sampling_ops.py,313,function,"Generate the set of all classes.
Deterministically generates and returns the set of all possible classes.
For testing purposes. There is no need to use this, since you might as
well use full softmax or full logistic regression.
Args:
true_classes: A `Tensor` of type `int64` and shape `[batch_size,
num_true]`. The target classes.
num_true: An `int`. The number of target classes per training example.
num_sampled: An `int`. The number of possible classes.
unique: A `bool`. Ignored.
unique.
seed: An `int`. An operation-specific seed. Default is 0.
name: A name for the operation (optional).
Returns:
sampled_candidates: A tensor of type `int64` and shape `[num_sampled]`.
This operation deterministically returns the entire range
`[0, num_sampled]`.
true_expected_count: A tensor of type `float`. Same shape as
`true_classes`. The expected counts under the sampling distribution
of each of `true_classes`. All returned values are 1.0.
sampled_expected_count: A tensor of type `float`. Same shape as
`sampled_candidates`. The expected counts under the sampling distribution
of each of `sampled_candidates`. All returned values are 1.0."
8162,compute_accidental_hits,tensorflow/tensorflow/python/ops/candidate_sampling_ops.py,350,function,"Compute the position ids in `sampled_candidates` matching `true_classes`.
In Candidate Sampling, this operation facilitates virtually removing
sampled classes which happen to match target classes. This is done
in Sampled Softmax and Sampled Logistic.
See our [Candidate Sampling Algorithms
Reference](http://www.tensorflow.org/extras/candidate_sampling.pdf).
We presuppose that the `sampled_candidates` are unique.
We call it an 'accidental hit' when one of the target classes
matches one of the sampled classes. This operation reports
accidental hits as triples `(index, id, weight)`, where `index`
represents the row number in `true_classes`, `id` represents the
position in `sampled_candidates`, and weight is `-FLOAT_MAX`.
The result of this op should be passed through a `sparse_to_dense`
operation, then added to the logits of the sampled classes. This
removes the contradictory effect of accidentally sampling the true
target classes as noise classes for the same example.
Args:
true_classes: A `Tensor` of type `int64` and shape `[batch_size,
num_true]`. The target classes.
sampled_candidates: A tensor of type `int64` and shape `[num_sampled]`.
The sampled_candidates output of CandidateSampler.
num_true: An `int`. The number of target classes per training example.
seed: An `int`. An operation-specific seed. Default is 0.
name: A name for the operation (optional).
Returns:
indices: A `Tensor` of type `int32` and shape `[num_accidental_hits]`.
Values indicate rows in `true_classes`.
ids: A `Tensor` of type `int64` and shape `[num_accidental_hits]`.
Values indicate positions in `sampled_candidates`.
weights: A `Tensor` of type `float` and shape `[num_accidental_hits]`.
Each value is `-FLOAT_MAX`."
8163,assert_proper_iterable,tensorflow/tensorflow/python/ops/check_ops.py,381,function,"Static assert that values is a ""proper"" iterable.
`Ops` that expect iterables of `Tensor` can call this to validate input.
Useful since `Tensor`, `ndarray`, byte/text type are all iterables themselves.
Args:
values: Object to be checked.
Raises:
TypeError: If `values` is not iterable or is one of
`Tensor`, `SparseTensor`, `np.array`, `tf.compat.bytes_or_text_types`."
8164,assert_negative_v2,tensorflow/tensorflow/python/ops/check_ops.py,410,function,"Assert the condition `x < 0` holds element-wise.
This Op checks that `x[i] < 0` holds for every element of `x`. If `x` is
empty, this is trivially satisfied.
If `x` is not negative everywhere, `message`, as well as the first `summarize`
entries of `x` are printed, and `InvalidArgumentError` is raised.
Args:
x: Numeric `Tensor`.
message: A string to prefix to the default message.
summarize: Print this many entries of each tensor.
name: A name for this operation (optional). Defaults to ""assert_negative"".
Returns:
Op raising `InvalidArgumentError` unless `x` is all negative. This can be
used with `tf.control_dependencies` inside of `tf.function`s to block
followup computation until the check has executed.
@compatibility(eager)
returns None
@end_compatibility
Raises:
InvalidArgumentError: if the check can be performed immediately and
`x[i] < 0` is False. The check can be performed immediately during eager
execution or if `x` is statically known."
8165,assert_negative,tensorflow/tensorflow/python/ops/check_ops.py,445,function,
8166,assert_positive_v2,tensorflow/tensorflow/python/ops/check_ops.py,464,function,"Assert the condition `x > 0` holds element-wise.
This Op checks that `x[i] > 0` holds for every element of `x`. If `x` is
empty, this is trivially satisfied.
If `x` is not positive everywhere, `message`, as well as the first `summarize`
entries of `x` are printed, and `InvalidArgumentError` is raised.
Args:
x: Numeric `Tensor`.
message: A string to prefix to the default message.
summarize: Print this many entries of each tensor.
name: A name for this operation (optional). Defaults to ""assert_positive"".
Returns:
Op raising `InvalidArgumentError` unless `x` is all positive. This can be
used with `tf.control_dependencies` inside of `tf.function`s to block
followup computation until the check has executed.
@compatibility(eager)
returns None
@end_compatibility
Raises:
InvalidArgumentError: if the check can be performed immediately and
`x[i] > 0` is False. The check can be performed immediately during eager
execution or if `x` is statically known."
8167,assert_positive,tensorflow/tensorflow/python/ops/check_ops.py,499,function,
8168,assert_non_negative_v2,tensorflow/tensorflow/python/ops/check_ops.py,517,function,"Assert the condition `x >= 0` holds element-wise.
This Op checks that `x[i] >= 0` holds for every element of `x`. If `x` is
empty, this is trivially satisfied.
If `x` is not >= 0 everywhere, `message`, as well as the first `summarize`
entries of `x` are printed, and `InvalidArgumentError` is raised.
Args:
x: Numeric `Tensor`.
message: A string to prefix to the default message.
summarize: Print this many entries of each tensor.
name: A name for this operation (optional). Defaults to
""assert_non_negative"".
Returns:
Op raising `InvalidArgumentError` unless `x` is all non-negative. This can
be used with `tf.control_dependencies` inside of `tf.function`s to block
followup computation until the check has executed.
@compatibility(eager)
returns None
@end_compatibility
Raises:
InvalidArgumentError: if the check can be performed immediately and
`x[i] >= 0` is False. The check can be performed immediately during eager
execution or if `x` is statically known."
8169,assert_non_negative,tensorflow/tensorflow/python/ops/check_ops.py,554,function,
8170,assert_non_positive_v2,tensorflow/tensorflow/python/ops/check_ops.py,573,function,"Assert the condition `x <= 0` holds element-wise.
This Op checks that `x[i] <= 0` holds for every element of `x`. If `x` is
empty, this is trivially satisfied.
If `x` is not <= 0 everywhere, `message`, as well as the first `summarize`
entries of `x` are printed, and `InvalidArgumentError` is raised.
Args:
x: Numeric `Tensor`.
message: A string to prefix to the default message.
summarize: Print this many entries of each tensor.
name: A name for this operation (optional). Defaults to
""assert_non_positive"".
Returns:
Op raising `InvalidArgumentError` unless `x` is all non-positive. This can
be used with `tf.control_dependencies` inside of `tf.function`s to block
followup computation until the check has executed.
@compatibility(eager)
returns None
@end_compatibility
Raises:
InvalidArgumentError: if the check can be performed immediately and
`x[i] <= 0` is False. The check can be performed immediately during eager
execution or if `x` is statically known."
8171,assert_non_positive,tensorflow/tensorflow/python/ops/check_ops.py,610,function,
8172,assert_equal_v2,tensorflow/tensorflow/python/ops/check_ops.py,629,function,"Assert the condition `x == y` holds element-wise.
This Op checks that `x[i] == y[i]` holds for every pair of (possibly
broadcast) elements of `x` and `y`. If both `x` and `y` are empty, this is
trivially satisfied.
If `x` and `y` are not equal, `message`, as well as the first `summarize`
entries of `x` and `y` are printed, and `InvalidArgumentError` is raised.
Args:
x: Numeric `Tensor`.
y: Numeric `Tensor`, same dtype as and broadcastable to `x`.
message: A string to prefix to the default message.
summarize: Print this many entries of each tensor.
name: A name for this operation (optional). Defaults to ""assert_equal"".
Returns:
Op that raises `InvalidArgumentError` if `x == y` is False. This can be
used with `tf.control_dependencies` inside of `tf.function`s to block
followup computation until the check has executed.
@compatibility(eager)
returns None
@end_compatibility
Raises:
InvalidArgumentError: if the check can be performed immediately and
`x == y` is False. The check can be performed immediately during eager
execution or if `x` and `y` are statically known."
8173,assert_equal,tensorflow/tensorflow/python/ops/check_ops.py,665,function,
8174,assert_none_equal_v2,tensorflow/tensorflow/python/ops/check_ops.py,676,function,"Assert the condition `x != y` holds for all elements.
This Op checks that `x[i] != y[i]` holds for every pair of (possibly
broadcast) elements of `x` and `y`. If both `x` and `y` are empty, this is
trivially satisfied.
If any elements of `x` and `y` are equal, `message`, as well as the first
`summarize` entries of `x` and `y` are printed, and `InvalidArgumentError`
is raised.
Args:
x: Numeric `Tensor`.
y: Numeric `Tensor`, same dtype as and broadcastable to `x`.
summarize: Print this many entries of each tensor.
message: A string to prefix to the default message.
name: A name for this operation (optional). Defaults to
""assert_none_equal"".
Returns:
Op that raises `InvalidArgumentError` if `x != y` is ever False. This can
be used with `tf.control_dependencies` inside of `tf.function`s to block
followup computation until the check has executed.
@compatibility(eager)
returns None
@end_compatibility
Raises:
InvalidArgumentError: if the check can be performed immediately and
`x != y` is False for any pair of elements in `x` and `y`. The check can
be performed immediately during eager execution or if `x` and `y` are
statically known."
8175,assert_none_equal,tensorflow/tensorflow/python/ops/check_ops.py,717,function,
8176,assert_near_v2,tensorflow/tensorflow/python/ops/check_ops.py,725,function,"Assert the condition `x` and `y` are close element-wise.
This Op checks that `x[i] - y[i] < atol + rtol * tf.abs(y[i])` holds for every
pair of (possibly broadcast) elements of `x` and `y`. If both `x` and `y` are
empty, this is trivially satisfied.
If any elements of `x` and `y` are not close, `message`, as well as the first
`summarize` entries of `x` and `y` are printed, and `InvalidArgumentError`
is raised.
The default `atol` and `rtol` is `10 * eps`, where `eps` is the smallest
representable positive number such that `1 + eps != 1`. This is about
`1.2e-6` in `32bit`, `2.22e-15` in `64bit`, and `0.00977` in `16bit`.
See `numpy.finfo`.
Args:
x: Float or complex `Tensor`.
y: Float or complex `Tensor`, same dtype as and broadcastable to `x`.
rtol: `Tensor`. Same `dtype` as, and broadcastable to, `x`.
The relative tolerance. Default is `10 * eps`.
atol: `Tensor`. Same `dtype` as, and broadcastable to, `x`.
The absolute tolerance. Default is `10 * eps`.
message: A string to prefix to the default message.
summarize: Print this many entries of each tensor.
name: A name for this operation (optional). Defaults to ""assert_near"".
Returns:
Op that raises `InvalidArgumentError` if `x` and `y` are not close enough.
This can be used with `tf.control_dependencies` inside of `tf.function`s
to block followup computation until the check has executed.
@compatibility(eager)
returns None
@end_compatibility
Raises:
InvalidArgumentError: if the check can be performed immediately and
`x != y` is False for any pair of elements in `x` and `y`. The check can
be performed immediately during eager execution or if `x` and `y` are
statically known.
@compatibility(numpy)
Similar to `numpy.testing.assert_allclose`, except tolerance depends on data
type. This is due to the fact that `TensorFlow` is often used with `32bit`,
`64bit`, and even `16bit` data.
@end_compatibility"
8177,assert_near,tensorflow/tensorflow/python/ops/check_ops.py,780,function,"Assert the condition `x` and `y` are close element-wise.
Example of adding a dependency to an operation:
```python
with tf.control_dependencies([tf.compat.v1.assert_near(x, y)]):
output = tf.reduce_sum(x)
```
This condition holds if for every pair of (possibly broadcast) elements
`x[i]`, `y[i]`, we have
```tf.abs(x[i] - y[i]) <= atol + rtol * tf.abs(y[i])```.
If both `x` and `y` are empty, this is trivially satisfied.
The default `atol` and `rtol` is `10 * eps`, where `eps` is the smallest
representable positive number such that `1 + eps != 1`. This is about
`1.2e-6` in `32bit`, `2.22e-15` in `64bit`, and `0.00977` in `16bit`.
See `numpy.finfo`.
Args:
x: Float or complex `Tensor`.
y: Float or complex `Tensor`, same `dtype` as, and broadcastable to, `x`.
rtol: `Tensor`. Same `dtype` as, and broadcastable to, `x`.
The relative tolerance. Default is `10 * eps`.
atol: `Tensor`. Same `dtype` as, and broadcastable to, `x`.
The absolute tolerance. Default is `10 * eps`.
data: The tensors to print out if the condition is False. Defaults to
error message and first few entries of `x`, `y`.
summarize: Print this many entries of each tensor.
message: A string to prefix to the default message.
name: A name for this operation (optional). Defaults to ""assert_near"".
Returns:
Op that raises `InvalidArgumentError` if `x` and `y` are not close enough.
@compatibility(numpy)
Similar to `numpy.testing.assert_allclose`, except tolerance depends on data
type. This is due to the fact that `TensorFlow` is often used with `32bit`,
`64bit`, and even `16bit` data.
@end_compatibility"
8178,assert_less_v2,tensorflow/tensorflow/python/ops/check_ops.py,862,function,"Assert the condition `x < y` holds element-wise.
This Op checks that `x[i] < y[i]` holds for every pair of (possibly
broadcast) elements of `x` and `y`. If both `x` and `y` are empty, this is
trivially satisfied.
If `x` is not less than `y` element-wise, `message`, as well as the first
`summarize` entries of `x` and `y` are printed, and `InvalidArgumentError` is
raised.
Args:
x: Numeric `Tensor`.
y: Numeric `Tensor`, same dtype as and broadcastable to `x`.
message: A string to prefix to the default message.
summarize: Print this many entries of each tensor.
name: A name for this operation (optional). Defaults to ""assert_less"".
Returns:
Op that raises `InvalidArgumentError` if `x < y` is False.
This can be used with `tf.control_dependencies` inside of `tf.function`s
to block followup computation until the check has executed.
@compatibility(eager)
returns None
@end_compatibility
Raises:
InvalidArgumentError: if the check can be performed immediately and
`x < y` is False. The check can be performed immediately during eager
execution or if `x` and `y` are statically known."
8179,assert_less,tensorflow/tensorflow/python/ops/check_ops.py,899,function,
8180,assert_less_equal_v2,tensorflow/tensorflow/python/ops/check_ops.py,906,function,"Assert the condition `x <= y` holds element-wise.
This Op checks that `x[i] <= y[i]` holds for every pair of (possibly
broadcast) elements of `x` and `y`. If both `x` and `y` are empty, this is
trivially satisfied.
If `x` is not less or equal than `y` element-wise, `message`, as well as the
first `summarize` entries of `x` and `y` are printed, and
`InvalidArgumentError` is raised.
Args:
x: Numeric `Tensor`.
y: Numeric `Tensor`, same dtype as and broadcastable to `x`.
message: A string to prefix to the default message.
summarize: Print this many entries of each tensor.
name: A name for this operation (optional). Defaults to ""assert_less_equal"".
Returns:
Op that raises `InvalidArgumentError` if `x <= y` is False. This can be
used with `tf.control_dependencies` inside of `tf.function`s to block
followup computation until the check has executed.
@compatibility(eager)
returns None
@end_compatibility
Raises:
InvalidArgumentError: if the check can be performed immediately and
`x <= y` is False. The check can be performed immediately during eager
execution or if `x` and `y` are statically known."
8181,assert_less_equal,tensorflow/tensorflow/python/ops/check_ops.py,945,function,
8182,assert_greater_v2,tensorflow/tensorflow/python/ops/check_ops.py,952,function,"Assert the condition `x > y` holds element-wise.
This Op checks that `x[i] > y[i]` holds for every pair of (possibly
broadcast) elements of `x` and `y`. If both `x` and `y` are empty, this is
trivially satisfied.
If `x` is not greater than `y` element-wise, `message`, as well as the first
`summarize` entries of `x` and `y` are printed, and `InvalidArgumentError` is
raised.
Args:
x: Numeric `Tensor`.
y: Numeric `Tensor`, same dtype as and broadcastable to `x`.
message: A string to prefix to the default message.
summarize: Print this many entries of each tensor.
name: A name for this operation (optional). Defaults to ""assert_greater"".
Returns:
Op that raises `InvalidArgumentError` if `x > y` is False. This can be
used with `tf.control_dependencies` inside of `tf.function`s to block
followup computation until the check has executed.
@compatibility(eager)
returns None
@end_compatibility
Raises:
InvalidArgumentError: if the check can be performed immediately and
`x > y` is False. The check can be performed immediately during eager
execution or if `x` and `y` are statically known."
8183,assert_greater,tensorflow/tensorflow/python/ops/check_ops.py,990,function,
8184,assert_greater_equal_v2,tensorflow/tensorflow/python/ops/check_ops.py,997,function,"Assert the condition `x >= y` holds element-wise.
This Op checks that `x[i] >= y[i]` holds for every pair of (possibly
broadcast) elements of `x` and `y`. If both `x` and `y` are empty, this is
trivially satisfied.
If `x` is not greater or equal to `y` element-wise, `message`, as well as the
first `summarize` entries of `x` and `y` are printed, and
`InvalidArgumentError` is raised.
Args:
x: Numeric `Tensor`.
y: Numeric `Tensor`, same dtype as and broadcastable to `x`.
message: A string to prefix to the default message.
summarize: Print this many entries of each tensor.
name: A name for this operation (optional). Defaults to
""assert_greater_equal"".
Returns:
Op that raises `InvalidArgumentError` if `x >= y` is False. This can be
used with `tf.control_dependencies` inside of `tf.function`s to block
followup computation until the check has executed.
@compatibility(eager)
returns None
@end_compatibility
Raises:
InvalidArgumentError: if the check can be performed immediately and
`x >= y` is False. The check can be performed immediately during eager
execution or if `x` and `y` are statically known."
8185,assert_greater_equal,tensorflow/tensorflow/python/ops/check_ops.py,1037,function,
8186,assert_rank_v2,tensorflow/tensorflow/python/ops/check_ops.py,1093,function,"Assert that `x` has rank equal to `rank`.
This Op checks that the rank of `x` is equal to `rank`.
If `x` has a different rank, `message`, as well as the shape of `x` are
printed, and `InvalidArgumentError` is raised.
Args:
x: `Tensor`.
rank: Scalar integer `Tensor`.
message: A string to prefix to the default message.
name: A name for this operation (optional). Defaults to
""assert_rank"".
Returns:
Op raising `InvalidArgumentError` unless `x` has specified rank.
If static checks determine `x` has correct rank, a `no_op` is returned.
This can be used with `tf.control_dependencies` inside of `tf.function`s
to block followup computation until the check has executed.
@compatibility(eager)
returns None
@end_compatibility
Raises:
InvalidArgumentError: if the check can be performed immediately and
`x` does not have rank `rank`. The check can be performed immediately
during eager execution or if the shape of `x` is statically known."
8187,assert_rank,tensorflow/tensorflow/python/ops/check_ops.py,1127,function,"Assert `x` has rank equal to `rank`.
Example of adding a dependency to an operation:
```python
with tf.control_dependencies([tf.compat.v1.assert_rank(x, 2)]):
output = tf.reduce_sum(x)
```
Args:
x: Numeric `Tensor`.
rank: Scalar integer `Tensor`.
data: The tensors to print out if the condition is False. Defaults to
error message and the shape of `x`.
summarize: Print this many entries of each tensor.
message: A string to prefix to the default message.
name: A name for this operation (optional). Defaults to ""assert_rank"".
Returns:
Op raising `InvalidArgumentError` unless `x` has specified rank.
If static checks determine `x` has correct rank, a `no_op` is returned.
Raises:
ValueError: If static checks determine `x` has wrong rank."
8188,assert_rank_at_least_v2,tensorflow/tensorflow/python/ops/check_ops.py,1190,function,"Assert that `x` has rank of at least `rank`.
This Op checks that the rank of `x` is greater or equal to `rank`.
If `x` has a rank lower than `rank`, `message`, as well as the shape of `x`
are printed, and `InvalidArgumentError` is raised.
Args:
x: `Tensor`.
rank: Scalar integer `Tensor`.
message: A string to prefix to the default message.
name: A name for this operation (optional). Defaults to
""assert_rank_at_least"".
Returns:
Op raising `InvalidArgumentError` unless `x` has specified rank or higher.
If static checks determine `x` has correct rank, a `no_op` is returned.
This can be used with `tf.control_dependencies` inside of `tf.function`s
to block followup computation until the check has executed.
@compatibility(eager)
returns None
@end_compatibility
Raises:
InvalidArgumentError: `x` does not have rank at least `rank`, but the rank
cannot be statically determined.
ValueError: If static checks determine `x` has mismatched rank."
8189,assert_rank_at_least,tensorflow/tensorflow/python/ops/check_ops.py,1225,function,"Assert `x` has rank equal to `rank` or higher.
Example of adding a dependency to an operation:
```python
with tf.control_dependencies([tf.compat.v1.assert_rank_at_least(x, 2)]):
output = tf.reduce_sum(x)
```
Args:
x: Numeric `Tensor`.
rank: Scalar `Tensor`.
data: The tensors to print out if the condition is False. Defaults to
error message and first few entries of `x`.
summarize: Print this many entries of each tensor.
message: A string to prefix to the default message.
name: A name for this operation (optional).
Defaults to ""assert_rank_at_least"".
Returns:
Op raising `InvalidArgumentError` unless `x` has specified rank or higher.
If static checks determine `x` has correct rank, a `no_op` is returned.
Raises:
ValueError: If static checks determine `x` has wrong rank."
8190,assert_rank_in_v2,tensorflow/tensorflow/python/ops/check_ops.py,1357,function,"Assert that `x` has a rank in `ranks`.
This Op checks that the rank of `x` is in `ranks`.
If `x` has a different rank, `message`, as well as the shape of `x` are
printed, and `InvalidArgumentError` is raised.
Args:
x: `Tensor`.
ranks: `Iterable` of scalar `Tensor` objects.
message: A string to prefix to the default message.
name: A name for this operation (optional). Defaults to ""assert_rank_in"".
Returns:
Op raising `InvalidArgumentError` unless rank of `x` is in `ranks`.
If static checks determine `x` has matching rank, a `no_op` is returned.
This can be used with `tf.control_dependencies` inside of `tf.function`s
to block followup computation until the check has executed.
@compatibility(eager)
returns None
@end_compatibility
Raises:
InvalidArgumentError: `x` does not have rank in `ranks`, but the rank cannot
be statically determined.
ValueError: If static checks determine `x` has mismatched rank."
8191,assert_rank_in,tensorflow/tensorflow/python/ops/check_ops.py,1391,function,"Assert `x` has rank in `ranks`.
Example of adding a dependency to an operation:
```python
with tf.control_dependencies([tf.compat.v1.assert_rank_in(x, (2, 4))]):
output = tf.reduce_sum(x)
```
Args:
x: Numeric `Tensor`.
ranks: Iterable of scalar `Tensor` objects.
data: The tensors to print out if the condition is False. Defaults to
error message and first few entries of `x`.
summarize: Print this many entries of each tensor.
message: A string to prefix to the default message.
name: A name for this operation (optional).
Defaults to ""assert_rank_in"".
Returns:
Op raising `InvalidArgumentError` unless rank of `x` is in `ranks`.
If static checks determine `x` has matching rank, a `no_op` is returned.
Raises:
ValueError: If static checks determine `x` has mismatched rank."
8192,assert_integer_v2,tensorflow/tensorflow/python/ops/check_ops.py,1454,function,"Assert that `x` is of integer dtype.
If `x` has a non-integer type, `message`, as well as the dtype of `x` are
printed, and `InvalidArgumentError` is raised.
This can always be checked statically, so this method returns nothing.
Args:
x: A `Tensor`.
message: A string to prefix to the default message.
name: A name for this operation (optional). Defaults to ""assert_integer"".
Raises:
TypeError: If `x.dtype` is not a non-quantized integer type."
8193,assert_integer,tensorflow/tensorflow/python/ops/check_ops.py,1476,function,"Assert that `x` is of integer dtype.
Example of adding a dependency to an operation:
```python
with tf.control_dependencies([tf.compat.v1.assert_integer(x)]):
output = tf.reduce_sum(x)
```
Args:
x: `Tensor` whose basetype is integer and is not quantized.
message: A string to prefix to the default message.
name: A name for this operation (optional). Defaults to ""assert_integer"".
Raises:
TypeError: If `x.dtype` is anything other than non-quantized integer.
Returns:
A `no_op` that does nothing. Type can be determined statically."
8194,assert_type_v2,tensorflow/tensorflow/python/ops/check_ops.py,1515,function,"Asserts that the given `Tensor` is of the specified type.
This can always be checked statically, so this method returns nothing.
Args:
tensor: A `Tensor` or `SparseTensor`.
tf_type: A tensorflow type (`dtypes.float32`, `tf.int64`, `dtypes.bool`,
etc).
message: A string to prefix to the default message.
name: A name for this operation. Defaults to ""assert_type""
Raises:
TypeError: If the tensor's data type doesn't match `tf_type`."
8195,assert_type,tensorflow/tensorflow/python/ops/check_ops.py,1536,function,"Statically asserts that the given `Tensor` is of the specified type.
Args:
tensor: A `Tensor` or `SparseTensor`.
tf_type: A tensorflow type (`dtypes.float32`, `tf.int64`, `dtypes.bool`,
etc).
message: A string to prefix to the default message.
name: A name to give this `Op`. Defaults to ""assert_type""
Raises:
TypeError: If the tensors data type doesn't match `tf_type`.
Returns:
A `no_op` that does nothing. Type can be determined statically."
8196,assert_shapes_v2,tensorflow/tensorflow/python/ops/check_ops.py,1627,function,"Assert tensor shapes and dimension size relationships between tensors.
This Op checks that a collection of tensors shape relationships
satisfies given constraints.
Example:
>>> n = 10
>>> q = 3
>>> d = 7
>>> x = tf.zeros([n,q])
>>> y = tf.ones([n,d])
>>> param = tf.Variable([1.0, 2.0, 3.0])
>>> scalar = 1.0
>>> tf.debugging.assert_shapes([
... (x, ('N', 'Q')),
... (y, ('N', 'D')),
... (param, ('Q',)),
... (scalar, ()),
... ])
>>> tf.debugging.assert_shapes([
... (x, ('N', 'D')),
... (y, ('N', 'D'))
... ])
Traceback (most recent call last):
...
ValueError: ...
If `x`, `y`, `param` or `scalar` does not have a shape that satisfies
all specified constraints, `message`, as well as the first `summarize` entries
of the first encountered violating tensor are printed, and
`InvalidArgumentError` is raised.
Size entries in the specified shapes are checked against other entries by
their __hash__, except:
- a size entry is interpreted as an explicit size if it can be parsed as an
integer primitive.
- a size entry is interpreted as *any* size if it is None or '.'.
If the first entry of a shape is `...` (type `Ellipsis`) or '*' that indicates
a variable number of outer dimensions of unspecified size, i.e. the constraint
applies to the inner-most dimensions only.
Scalar tensors and specified shapes of length zero (excluding the 'inner-most'
prefix) are both treated as having a single dimension of size one.
Args:
shapes: dictionary with (`Tensor` to shape) items, or a list of
(`Tensor`, shape) tuples. A shape must be an iterable.
data: The tensors to print out if the condition is False. Defaults to error
message and first few entries of the violating tensor.
summarize: Print this many entries of the tensor.
message: A string to prefix to the default message.
name: A name for this operation (optional). Defaults to ""assert_shapes"".
Raises:
ValueError: If static checks determine any shape constraint is violated."
8197,assert_shapes,tensorflow/tensorflow/python/ops/check_ops.py,1694,function,"Assert tensor shapes and dimension size relationships between tensors.
This Op checks that a collection of tensors shape relationships
satisfies given constraints.
Example:
>>> n = 10
>>> q = 3
>>> d = 7
>>> x = tf.zeros([n,q])
>>> y = tf.ones([n,d])
>>> param = tf.Variable([1.0, 2.0, 3.0])
>>> scalar = 1.0
>>> tf.debugging.assert_shapes([
... (x, ('N', 'Q')),
... (y, ('N', 'D')),
... (param, ('Q',)),
... (scalar, ()),
... ])
>>> tf.debugging.assert_shapes([
... (x, ('N', 'D')),
... (y, ('N', 'D'))
... ])
Traceback (most recent call last):
...
ValueError: ...
Example of adding a dependency to an operation:
```python
with tf.control_dependencies([tf.assert_shapes(shapes)]):
output = tf.matmul(x, y, transpose_a=True)
```
If `x`, `y`, `param` or `scalar` does not have a shape that satisfies
all specified constraints, `message`, as well as the first `summarize` entries
of the first encountered violating tensor are printed, and
`InvalidArgumentError` is raised.
Size entries in the specified shapes are checked against other entries by
their __hash__, except:
- a size entry is interpreted as an explicit size if it can be parsed as an
integer primitive.
- a size entry is interpreted as *any* size if it is None or '.'.
If the first entry of a shape is `...` (type `Ellipsis`) or '*' that indicates
a variable number of outer dimensions of unspecified size, i.e. the constraint
applies to the inner-most dimensions only.
Scalar tensors and specified shapes of length zero (excluding the 'inner-most'
prefix) are both treated as having a single dimension of size one.
Args:
shapes: A list of (`Tensor`, `shape`) tuples, wherein `shape` is the
expected shape of `Tensor`. See the example code above. The `shape` must
be an iterable. Each element of the iterable can be either a concrete
integer value or a string that abstractly represents the dimension.
For example,
- `('N', 'Q')` specifies a 2D shape wherein the first and second
dimensions of shape may or may not be equal.
- `('N', 'N', 'Q')` specifies a 3D shape wherein the first and second
dimensions are equal.
- `(1, 'N')` specifies a 2D shape wherein the first dimension is
exactly 1 and the second dimension can be any value.
Note that the abstract dimension letters take effect across different
tuple elements of the list. For example,
`tf.debugging.assert_shapes([(x, ('N', 'A')), (y, ('N', 'B'))]` asserts
that both `x` and `y` are rank-2 tensors and their first dimensions are
equal (`N`).
`shape` can also be a `tf.TensorShape`.
data: The tensors to print out if the condition is False. Defaults to error
message and first few entries of the violating tensor.
summarize: Print this many entries of the tensor.
message: A string to prefix to the default message.
name: A name for this operation (optional). Defaults to ""assert_shapes"".
Returns:
Op raising `InvalidArgumentError` unless all shape constraints are
satisfied.
If static checks determine all constraints are satisfied, a `no_op` is
returned.
Raises:
ValueError: If static checks determine any shape constraint is violated."
8198,is_numeric_tensor,tensorflow/tensorflow/python/ops/check_ops.py,1969,function,"Returns `True` if the elements of `tensor` are numbers.
Specifically, returns `True` if the dtype of `tensor` is one of the following:
* `tf.float32`
* `tf.float64`
* `tf.int8`
* `tf.int16`
* `tf.int32`
* `tf.int64`
* `tf.uint8`
* `tf.qint8`
* `tf.qint32`
* `tf.quint8`
* `tf.complex64`
Returns `False` if `tensor` is of a non-numeric type or if `tensor` is not
a `tf.Tensor` object."
8199,is_non_decreasing,tensorflow/tensorflow/python/ops/check_ops.py,2001,function,"Returns `True` if `x` is non-decreasing.
Elements of `x` are compared in row-major order. The tensor `[x[0],...]`
is non-decreasing if for every adjacent pair we have `x[i] <= x[i+1]`.
If `x` has less than two elements, it is trivially non-decreasing.
See also: `is_strictly_increasing`
>>> x1 = tf.constant([1.0, 1.0, 3.0])
>>> tf.math.is_non_decreasing(x1)
<tf.Tensor: shape=(), dtype=bool, numpy=True>
>>> x2 = tf.constant([3.0, 1.0, 2.0])
>>> tf.math.is_non_decreasing(x2)
<tf.Tensor: shape=(), dtype=bool, numpy=False>
Args:
x: Numeric `Tensor`.
name: A name for this operation (optional). Defaults to ""is_non_decreasing""
Returns:
Boolean `Tensor`, equal to `True` iff `x` is non-decreasing.
Raises:
TypeError: if `x` is not a numeric tensor."
8200,is_strictly_increasing,tensorflow/tensorflow/python/ops/check_ops.py,2043,function,"Returns `True` if `x` is strictly increasing.
Elements of `x` are compared in row-major order. The tensor `[x[0],...]`
is strictly increasing if for every adjacent pair we have `x[i] < x[i+1]`.
If `x` has less than two elements, it is trivially strictly increasing.
See also: `is_non_decreasing`
>>> x1 = tf.constant([1.0, 2.0, 3.0])
>>> tf.math.is_strictly_increasing(x1)
<tf.Tensor: shape=(), dtype=bool, numpy=True>
>>> x2 = tf.constant([3.0, 1.0, 2.0])
>>> tf.math.is_strictly_increasing(x2)
<tf.Tensor: shape=(), dtype=bool, numpy=False>
Args:
x: Numeric `Tensor`.
name: A name for this operation (optional).
Defaults to ""is_strictly_increasing""
Returns:
Boolean `Tensor`, equal to `True` iff `x` is strictly increasing.
Raises:
TypeError: if `x` is not a numeric tensor."
8201,assert_same_float_dtype,tensorflow/tensorflow/python/ops/check_ops.py,2129,function,"Validate and return float type based on `tensors` and `dtype`.
For ops such as matrix multiplication, inputs and weights must be of the
same float type. This function validates that all `tensors` are the same type,
validates that type is `dtype` (if supplied), and returns the type. Type must
be a floating point type. If neither `tensors` nor `dtype` is supplied,
the function will return `dtypes.float32`.
Args:
tensors: Tensors of input values. Can include `None` elements, which will be
ignored.
dtype: Expected type.
Returns:
Validated type.
Raises:
ValueError: if neither `tensors` nor `dtype` is supplied, or result is not
float, or the common type of the inputs is not a floating point type."
8202,assert_scalar_v2,tensorflow/tensorflow/python/ops/check_ops.py,2161,function,"Asserts that the given `tensor` is a scalar.
This function raises `ValueError` unless it can be certain that the given
`tensor` is a scalar. `ValueError` is also raised if the shape of `tensor` is
unknown.
This is always checked statically, so this method returns nothing.
Args:
tensor: A `Tensor`.
message: A string to prefix to the default message.
name: A name for this operation. Defaults to ""assert_scalar""
Raises:
ValueError: If the tensor is not scalar (rank 0), or if its shape is
unknown."
8203,assert_scalar,tensorflow/tensorflow/python/ops/check_ops.py,2185,function,"Asserts that the given `tensor` is a scalar (i.e. zero-dimensional).
This function raises `ValueError` unless it can be certain that the given
`tensor` is a scalar. `ValueError` is also raised if the shape of `tensor` is
unknown.
Args:
tensor: A `Tensor`.
name: A name for this operation. Defaults to ""assert_scalar""
message: A string to prefix to the default message.
Returns:
The input tensor (potentially converted to a `Tensor`).
Raises:
ValueError: If the tensor is not scalar (rank 0), or if its shape is
unknown."
8204,ensure_shape,tensorflow/tensorflow/python/ops/check_ops.py,2219,function,"Updates the shape of a tensor and checks at runtime that the shape holds.
For example:
>>> @tf.function(input_signature=[tf.TensorSpec(shape=None, dtype=tf.float32)])
... def f(tensor):
... return tf.ensure_shape(tensor, [3, 3])
>>>
>>> f(tf.zeros([3, 3])) # Passes
<tf.Tensor: shape=(3, 3), dtype=float32, numpy=
array([[0., 0., 0.],
[0., 0., 0.],
[0., 0., 0.]], dtype=float32)>
>>> f([1, 2, 3]) # fails
Traceback (most recent call last):
...
InvalidArgumentError: Shape of tensor x [3] is not compatible with expected shape [3,3].
The above example raises `tf.errors.InvalidArgumentError`,
because the shape (3,) is not compatible with the shape (None, 3, 3)
With eager execution this is a shape assertion, that returns the input:
>>> x = tf.constant([1,2,3])
>>> print(x.shape)
(3,)
>>> x = tf.ensure_shape(x, [3])
>>> x = tf.ensure_shape(x, [5])
Traceback (most recent call last):
...
tf.errors.InvalidArgumentError: Shape of tensor dummy_input [3] is not
compatible with expected shape [5]. [Op:EnsureShape]
Inside a `tf.function` or `v1.Graph` context it checks both the buildtime and
runtime shapes. This is stricter than `tf.Tensor.set_shape` which only
checks the buildtime shape.
Note: This differs from `tf.Tensor.set_shape` in that it sets the static shape
of the resulting tensor and enforces it at runtime, raising an error if the
tensor's runtime shape is incompatible with the specified shape.
`tf.Tensor.set_shape` sets the static shape of the tensor without enforcing it
at runtime, which may result in inconsistencies between the statically-known
shape of tensors and the runtime value of tensors.
For example, of loading images of a known size:
>>> @tf.function
... def decode_image(png):
... image = tf.image.decode_png(png, channels=3)
... # the `print` executes during tracing.
... print(""Initial shape: "", image.shape)
... image = tf.ensure_shape(image,[28, 28, 3])
... print(""Final shape: "", image.shape)
... return image
When tracing a function, no ops are being executed, shapes may be unknown.
See the [Concrete Functions Guide](https://www.tensorflow.org/guide/concrete_function)
for details.
>>> concrete_decode = decode_image.get_concrete_function(
... tf.TensorSpec([], dtype=tf.string))
Initial shape: (None, None, 3)
Final shape: (28, 28, 3)
>>> image = tf.random.uniform(maxval=255, shape=[28, 28, 3], dtype=tf.int32)
>>> image = tf.cast(image,tf.uint8)
>>> png = tf.image.encode_png(image)
>>> image2 = concrete_decode(png)
>>> print(image2.shape)
(28, 28, 3)
>>> image = tf.concat([image,image], axis=0)
>>> print(image.shape)
(56, 28, 3)
>>> png = tf.image.encode_png(image)
>>> image2 = concrete_decode(png)
Traceback (most recent call last):
...
tf.errors.InvalidArgumentError: Shape of tensor DecodePng [56,28,3] is not
compatible with expected shape [28,28,3].
Caution: if you don't use the result of `tf.ensure_shape` the check may not
run.
>>> @tf.function
... def bad_decode_image(png):
... image = tf.image.decode_png(png, channels=3)
... # the `print` executes during tracing.
... print(""Initial shape: "", image.shape)
... # BAD: forgot to use the returned tensor.
... tf.ensure_shape(image,[28, 28, 3])
... print(""Final shape: "", image.shape)
... return image
>>> image = bad_decode_image(png)
Initial shape: (None, None, 3)
Final shape: (None, None, 3)
>>> print(image.shape)
(56, 28, 3)
Args:
x: A `Tensor`.
shape: A `TensorShape` representing the shape of this tensor, a
`TensorShapeProto`, a list, a tuple, or None.
name: A name for this operation (optional). Defaults to ""EnsureShape"".
Returns:
A `Tensor`. Has the same type and contents as `x`.
Raises:
tf.errors.InvalidArgumentError: If `shape` is incompatible with the shape
of `x`."
8205,clip_by_value,tensorflow/tensorflow/python/ops/clip_ops.py,38,function,"Clips tensor values to a specified min and max.
Given a tensor `t`, this operation returns a tensor of the same type and
shape as `t` with its values clipped to `clip_value_min` and `clip_value_max`.
Any values less than `clip_value_min` are set to `clip_value_min`. Any values
greater than `clip_value_max` are set to `clip_value_max`.
Note: `clip_value_min` needs to be smaller or equal to `clip_value_max` for
correct results.
For example:
Basic usage passes a scalar as the min and max value.
>>> t = tf.constant([[-10., -1., 0.], [0., 2., 10.]])
>>> t2 = tf.clip_by_value(t, clip_value_min=-1, clip_value_max=1)
>>> t2.numpy()
array([[-1., -1., 0.],
[ 0., 1., 1.]], dtype=float32)
The min and max can be the same size as `t`, or broadcastable to that size.
>>> t = tf.constant([[-1, 0., 10.], [-1, 0, 10]])
>>> clip_min = [[2],[1]]
>>> t3 = tf.clip_by_value(t, clip_value_min=clip_min, clip_value_max=100)
>>> t3.numpy()
array([[ 2., 2., 10.],
[ 1., 1., 10.]], dtype=float32)
Broadcasting fails, intentionally, if you would expand the dimensions of `t`
>>> t = tf.constant([[-1, 0., 10.], [-1, 0, 10]])
>>> clip_min = [[[2, 1]]] # Has a third axis
>>> t4 = tf.clip_by_value(t, clip_value_min=clip_min, clip_value_max=100)
Traceback (most recent call last):
...
InvalidArgumentError: Incompatible shapes: [2,3] vs. [1,1,2]
It throws a `TypeError` if you try to clip an `int` to a `float` value
(`tf.cast` the input to `float` first).
>>> t = tf.constant([[1, 2], [3, 4]], dtype=tf.int32)
>>> t5 = tf.clip_by_value(t, clip_value_min=-3.1, clip_value_max=3.1)
Traceback (most recent call last):
...
TypeError: Cannot convert ...
Args:
t: A `Tensor` or `IndexedSlices`.
clip_value_min: The minimum value to clip to. A scalar `Tensor` or one that
is broadcastable to the shape of `t`.
clip_value_max: The maximum value to clip to. A scalar `Tensor` or one that
is broadcastable to the shape of `t`.
name: A name for the operation (optional).
Returns:
A clipped `Tensor` or `IndexedSlices`.
Raises:
`tf.errors.InvalidArgumentError`: If the clip tensors would trigger array
broadcasting that would make the returned tensor larger than the input.
TypeError: If dtype of the input is `int32` and dtype of
the `clip_value_min` or `clip_value_max` is `float32`"
8206,clip_by_norm,tensorflow/tensorflow/python/ops/clip_ops.py,156,function,"Clips tensor values to a maximum L2-norm.
Given a tensor `t`, and a maximum clip value `clip_norm`, this operation
normalizes `t` so that its L2-norm is less than or equal to `clip_norm`,
along the dimensions given in `axes`. Specifically, in the default case
where all dimensions are used for calculation, if the L2-norm of `t` is
already less than or equal to `clip_norm`, then `t` is not modified. If
the L2-norm is greater than `clip_norm`, then this operation returns a
tensor of the same type and shape as `t` with its values set to:
`t * clip_norm / l2norm(t)`
In this case, the L2-norm of the output tensor is `clip_norm`.
As another example, if `t` is a matrix and `axes == [1]`, then each row
of the output will have L2-norm less than or equal to `clip_norm`. If
`axes == [0]` instead, each column of the output will be clipped.
Code example:
>>> some_nums = tf.constant([[1, 2, 3, 4, 5]], dtype=tf.float32)
>>> tf.clip_by_norm(some_nums, 2.0).numpy()
array([[0.26967996, 0.5393599 , 0.80903983, 1.0787199 , 1.3483998 ]],
dtype=float32)
This operation is typically used to clip gradients before applying them with
an optimizer. Most gradient data is a collection of different shaped tensors
for different parts of the model. Thus, this is a common usage:
```
# Get your gradients after training
loss_value, grads = grad(model, features, labels)
# Apply some clipping
grads = [tf.clip_by_norm(g, norm)
for g in grads]
# Continue on with training
optimizer.apply_gradients(grads)
```
Args:
t: A `Tensor` or `IndexedSlices`. This must be a floating point type.
clip_norm: A 0-D (scalar) `Tensor` > 0. A maximum clipping value, also
floating point
axes: A 1-D (vector) `Tensor` of type int32 containing the dimensions
to use for computing the L2-norm. If `None` (the default), uses all
dimensions.
name: A name for the operation (optional).
Returns:
A clipped `Tensor` or `IndexedSlices`.
Raises:
ValueError: If the clip_norm tensor is not a 0-D scalar tensor.
TypeError: If dtype of the input is not a floating point or
complex type."
8207,global_norm,tensorflow/tensorflow/python/ops/clip_ops.py,241,function,"Computes the global norm of multiple tensors.
Given a tuple or list of tensors `t_list`, this operation returns the
global norm of the elements in all tensors in `t_list`. The global norm is
computed as:
`global_norm = sqrt(sum([l2norm(t)**2 for t in t_list]))`
Any entries in `t_list` that are of type None are ignored.
Args:
t_list: A tuple or list of mixed `Tensors`, `IndexedSlices`, or None.
name: A name for the operation (optional).
Returns:
A 0-D (scalar) `Tensor` of type `float`.
Raises:
TypeError: If `t_list` is not a sequence."
8208,clip_by_global_norm,tensorflow/tensorflow/python/ops/clip_ops.py,291,function,"Clips values of multiple tensors by the ratio of the sum of their norms.
Given a tuple or list of tensors `t_list`, and a clipping ratio `clip_norm`,
this operation returns a list of clipped tensors `list_clipped`
and the global norm (`global_norm`) of all tensors in `t_list`. Optionally,
if you've already computed the global norm for `t_list`, you can specify
the global norm with `use_norm`.
To perform the clipping, the values `t_list[i]` are set to:
t_list[i] * clip_norm / max(global_norm, clip_norm)
where:
global_norm = sqrt(sum([l2norm(t)**2 for t in t_list]))
If `clip_norm > global_norm` then the entries in `t_list` remain as they are,
otherwise they're all shrunk by the global ratio.
If `global_norm == infinity` then the entries in `t_list` are all set to `NaN`
to signal that an error occurred.
Any of the entries of `t_list` that are of type `None` are ignored.
This is the correct way to perform gradient clipping (Pascanu et al., 2012).
However, it is slower than `clip_by_norm()` because all the parameters must be
ready before the clipping operation can be performed.
Args:
t_list: A tuple or list of mixed `Tensors`, `IndexedSlices`, or None.
clip_norm: A 0-D (scalar) `Tensor` > 0. The clipping ratio.
use_norm: A 0-D (scalar) `Tensor` of type `float` (optional). The global
norm to use. If not provided, `global_norm()` is used to compute the norm.
name: A name for the operation (optional).
Returns:
list_clipped: A list of `Tensors` of the same type as `list_t`.
global_norm: A 0-D (scalar) `Tensor` representing the global norm.
Raises:
TypeError: If `t_list` is not a sequence.
References:
On the difficulty of training Recurrent Neural Networks:
[Pascanu et al., 2012](http://proceedings.mlr.press/v28/pascanu13.html)
([pdf](http://proceedings.mlr.press/v28/pascanu13.pdf))"
8209,clip_by_average_norm,tensorflow/tensorflow/python/ops/clip_ops.py,389,function,"Clips tensor values to a maximum average L2-norm.
Given a tensor `t`, and a maximum clip value `clip_norm`, this operation
normalizes `t` so that its average L2-norm is less than or equal to
`clip_norm`. Specifically, if the average L2-norm is already less than or
equal to `clip_norm`, then `t` is not modified. If the average L2-norm is
greater than `clip_norm`, then this operation returns a tensor of the same
type and shape as `t` with its values set to:
`t * clip_norm / l2norm_avg(t)`
In this case, the average L2-norm of the output tensor is `clip_norm`.
This operation is typically used to clip gradients before applying them with
an optimizer.
Args:
t: A `Tensor`.
clip_norm: A 0-D (scalar) `Tensor` > 0. A maximum clipping value.
name: A name for the operation (optional).
Returns:
A clipped `Tensor`."
8210,KMeans,tensorflow/tensorflow/python/ops/clustering_ops.py,56,class,Creates the graph for k-means clustering.
8211,training_graph,tensorflow/tensorflow/python/ops/clustering_ops.py,327,method,"Generate a training graph for kmeans algorithm.
This returns, among other things, an op that chooses initial centers
(init_op), a boolean variable that is set to True when the initial centers
are chosen (cluster_centers_initialized), and an op to perform either an
entire Lloyd iteration or a mini-batch of a Lloyd iteration (training_op).
The caller should use these components as follows. A single worker should
execute init_op multiple times until cluster_centers_initialized becomes
True. Then multiple workers may execute training_op any number of times.
Returns:
A tuple consisting of:
all_scores: A matrix (or list of matrices) of dimensions (num_input,
num_clusters) where the value is the distance of an input vector and a
cluster center.
cluster_idx: A vector (or list of vectors). Each element in the vector
corresponds to an input row in 'inp' and specifies the cluster id
corresponding to the input.
scores: Similar to cluster_idx but specifies the distance to the
assigned cluster instead.
cluster_centers_initialized: scalar indicating whether clusters have been
initialized.
init_op: an op to initialize the clusters.
training_op: an op that runs an iteration of training."
8212,all_reduce,tensorflow/tensorflow/python/ops/collective_ops.py,23,function,"Reduces tensors collectively, across devices.
Args:
t: the tensor to be reduced.
group_size: the total number of tensors to be collectively reduced.
Each must reside on a different device. Should be a positive integer.
group_key: an integer identifying the group of devices.
instance_key: an integer identifying the participating group of Ops.
merge_op: string naming the binary Op to be applied to compute each
partial reduction.
final_op: string naming the unary Op to be applied to each fully
reduced value. Can be 'Id' for no operation.
subdiv_offsets: a list of integer offsets into the tensor at which each
independent subdivision should begin. Use [0] if no subdivision should
be done.
communication_hint: preferred collective communication. The implementation
may fall back to another mechanism. Options include `auto`, `ring`, and
`nccl`.
timeout: If set to a non zero, set a completion timeout to detect staleness.
If the timer goes off, a DeadlineExceededError is raised.
The timeout value in seconds. This feature is experimental.
Returns:
An Op implementing the distributed reduction.
Raises:
ValueError: if any of the input parameter constraints are not met."
8213,all_gather,tensorflow/tensorflow/python/ops/collective_ops.py,74,function,"Accumulates tensors collectively, across devices, along first dimension.
Args:
t: the tensor to participate in the accumulation.
group_size: the total number of tensors to be collectively accumulated.
Each must reside on a different device. Should be a positive integer.
group_key: an integer identifying the group of devices.
instance_key: an integer identifying the participating group of Ops.
communication_hint: preferred collective communication. The implementation
may fall back to another mechanism. Options include `auto`, `ring`, and
`nccl`.
timeout: If set to a non zero, set a completion timeout to detect staleness.
If the timer goes off, a DeadlineExceededError is raised.
The timeout value in seconds. This feature is experimental.
Returns:
An Op implementing the distributed operation.
Raises:
ValueError: if any of the input parameter constraints are not met."
8214,broadcast_send,tensorflow/tensorflow/python/ops/collective_ops.py,113,function,"Broadcasts one tensor to a group of others, across devices.
Args:
t: the tensor to be sent.
shape: the shape of the tensor being sent, which must agree with t.
dtype: the type of the tensor being sent, which must agree with t.
group_size: one plus the number of receiving tensors, i.e. the total
number of devices participating. Each tensor must reside on a
different device.
group_key: an integer identifying the group of devices.
instance_key: an integer identifying the participating group of Ops.
communication_hint: preferred collective communication. The implementation
may fall back to another mechanism. Options include `auto`, `ring`, and
`nccl`.
timeout: If set to a non zero, set a completion timeout to detect staleness.
If the timer goes off, a DeadlineExceededError is raised.
The timeout value in seconds. This feature is experimental.
Returns:
An Op implementing the distributed broadcast send.
Raises:
ValueError: if any of the input parameter constraints are not met.
Note that the shape and dtype arguments appear redundant since they
should be obtainable from t. The are two reasons for including
them. First, the shape and type of tensors passed via broadcast must
be known ahead of time in their most specific form so that the receive
side can allocate memory for the operation and shape/type inference can
carry forward from there. Including the same declarations on the
send side clarifies a commitment already made. Secondly, having nearly
identical use syntax for send and receive sides may simplify tool-driven
generation of broadcast."
8215,broadcast_recv,tensorflow/tensorflow/python/ops/collective_ops.py,174,function,"Receives a broadcasts tensor, across devices.
Args:
shape: Shape of the tensor to be received.
dtype: Type of the tensor to be received.
group_size: one plus the number of receiving tensors, i.e. the total
number of devices participating. Each tensor must reside on a
different device.
group_key: an integer identifying the group of devices.
instance_key: an integer identifying the participating group of Ops.
communication_hint: preferred collective communication. The implementation
may fall back to another mechanism. Options include `auto`, `ring`, and
`nccl`.
timeout: If set to a non zero, set a completion timeout to detect staleness.
If the timer goes off, a DeadlineExceededError is raised.
The timeout value in seconds. This feature is experimental.
Returns:
An Op implementing the broadcast receive.
Raises:
ValueError: if any of the input parameter constraints are not met."
8216,CollectiveOpBenchmark,tensorflow/tensorflow/python/ops/collective_ops_benchmark.py,33,class,Benchmarks for local CPU collective op execution.
8217,benchmark_collective,tensorflow/tensorflow/python/ops/collective_ops_benchmark.py,36,method,Measures the performance of local CPU collective execution.
8218,build_graph,tensorflow/tensorflow/python/ops/concat_benchmark.py,35,function,"Build a graph containing a sequence of concat operations.
Args:
device: string, the device to run on.
input_shape: shape of the input tensors.
variable: whether or not to randomize the input shape
num_inputs: the number of inputs to concat
axis: axis to be concat'ed
grad: if True compute the gradient
Returns:
An array of tensors to run()"
8219,ConcatBenchmark,tensorflow/tensorflow/python/ops/concat_benchmark.py,78,class,Benchmark concat.
8220,benchmark_concat,tensorflow/tensorflow/python/ops/concat_benchmark.py,132,method,
8221,cond_v2,tensorflow/tensorflow/python/ops/cond_v2.py,59,function,"Like tf.cond, except emits a single If op."
8222,get_func_graphs,tensorflow/tensorflow/python/ops/cond_v2.py,301,function,"Returns `FuncGraph`s for the input op branches.
Args:
op: The If or Case Operation.
Returns:
A tuple of the `FuncGraph`s of the then_branch and else_branch (all branches
for Case)."
8223,verify_captures,tensorflow/tensorflow/python/ops/cond_v2.py,815,function,Verify that a branch's tensor is not accessed in another branch fn.
8224,indexed_case,tensorflow/tensorflow/python/ops/cond_v2.py,946,function,"Like conv_v2, except emits a Case op instead of an If."
8225,remove_squeezable_dimensions,tensorflow/tensorflow/python/ops/confusion_matrix.py,34,function,"Squeeze last dim if ranks differ from expected by exactly 1.
In the common case where we expect shapes to match, `expected_rank_diff`
defaults to 0, and we squeeze the last dimension of the larger rank if they
differ by 1.
But, for example, if `labels` contains class IDs and `predictions` contains 1
probability per class, we expect `predictions` to have 1 more dimension than
`labels`, so `expected_rank_diff` would be 1. In this case, we'd squeeze
`labels` if `rank(predictions) - rank(labels) == 0`, and
`predictions` if `rank(predictions) - rank(labels) == 2`.
This will use static shape if available. Otherwise, it will add graph
operations, which could result in a performance hit.
Args:
labels: Label values, a `Tensor` whose dimensions match `predictions`.
predictions: Predicted values, a `Tensor` of arbitrary dimensions.
expected_rank_diff: Expected result of `rank(predictions) - rank(labels)`.
name: Name of the op.
Returns:
Tuple of `labels` and `predictions`, possibly with last dim squeezed."
8226,confusion_matrix,tensorflow/tensorflow/python/ops/confusion_matrix.py,98,function,"Computes the confusion matrix from predictions and labels.
The matrix columns represent the prediction labels and the rows represent the
real labels. The confusion matrix is always a 2-D array of shape `[n, n]`,
where `n` is the number of valid labels for a given classification task. Both
prediction and labels must be 1-D arrays of the same shape in order for this
function to work.
If `num_classes` is `None`, then `num_classes` will be set to one plus the
maximum value in either predictions or labels. Class labels are expected to
start at 0. For example, if `num_classes` is 3, then the possible labels
would be `[0, 1, 2]`.
If `weights` is not `None`, then each prediction contributes its
corresponding weight to the total value of the confusion matrix cell.
For example:
```python
tf.math.confusion_matrix([1, 2, 4], [2, 2, 4]) ==>
[[0 0 0 0 0]
[0 0 1 0 0]
[0 0 1 0 0]
[0 0 0 0 0]
[0 0 0 0 1]]
```
Note that the possible labels are assumed to be `[0, 1, 2, 3, 4]`,
resulting in a 5x5 confusion matrix.
Args:
labels: 1-D `Tensor` of real labels for the classification task.
predictions: 1-D `Tensor` of predictions for a given classification.
num_classes: The possible number of labels the classification task can
have. If this value is not provided, it will be calculated
using both predictions and labels array.
weights: An optional `Tensor` whose shape matches `predictions`.
dtype: Data type of the confusion matrix.
name: Scope name.
Returns:
A `Tensor` of type `dtype` with shape `[n, n]` representing the confusion
matrix, where `n` is the number of possible labels in the classification
task.
Raises:
ValueError: If both predictions and labels are not 1-D vectors and have
mismatched shapes, or if `weights` is not `None` and its shape doesn't
match `predictions`."
8227,confusion_matrix_v1,tensorflow/tensorflow/python/ops/confusion_matrix.py,209,function,"Computes the confusion matrix from predictions and labels.
The matrix columns represent the prediction labels and the rows represent the
real labels. The confusion matrix is always a 2-D array of shape `[n, n]`,
where `n` is the number of valid labels for a given classification task. Both
prediction and labels must be 1-D arrays of the same shape in order for this
function to work.
If `num_classes` is `None`, then `num_classes` will be set to one plus the
maximum value in either predictions or labels. Class labels are expected to
start at 0. For example, if `num_classes` is 3, then the possible labels
would be `[0, 1, 2]`.
If `weights` is not `None`, then each prediction contributes its
corresponding weight to the total value of the confusion matrix cell.
For example:
```python
tf.math.confusion_matrix([1, 2, 4], [2, 2, 4]) ==>
[[0 0 0 0 0]
[0 0 1 0 0]
[0 0 1 0 0]
[0 0 0 0 0]
[0 0 0 0 1]]
```
Note that the possible labels are assumed to be `[0, 1, 2, 3, 4]`,
resulting in a 5x5 confusion matrix.
Args:
labels: 1-D `Tensor` of real labels for the classification task.
predictions: 1-D `Tensor` of predictions for a given classification.
num_classes: The possible number of labels the classification task can have.
If this value is not provided, it will be calculated using both
predictions and labels array.
dtype: Data type of the confusion matrix.
name: Scope name.
weights: An optional `Tensor` whose shape matches `predictions`.
Returns:
A `Tensor` of type `dtype` with shape `[n, n]` representing the confusion
matrix, where `n` is the number of possible labels in the classification
task.
Raises:
ValueError: If both predictions and labels are not 1-D vectors and have
mismatched shapes, or if `weights` is not `None` and its shape doesn't
match `predictions`."
8228,Assert,tensorflow/tensorflow/python/ops/control_flow_ops.py,117,function,"Asserts that the given condition is true.
If `condition` evaluates to false, print the list of tensors in `data`.
`summarize` determines how many entries of the tensors to print.
Args:
condition: The condition to evaluate.
data: The tensors to print out when condition is false.
summarize: Print this many entries of each tensor.
name: A name for this operation (optional).
Returns:
assert_op: An `Operation` that, when executed, raises a
`tf.errors.InvalidArgumentError` if `condition` is not true.
@compatibility(eager)
returns None
@end_compatibility
Raises:
@compatibility(TF1)
When in TF V1 mode (that is, outside `tf.function`) Assert needs a control
dependency on the output to ensure the assertion executes:
```python
# Ensure maximum element of x is smaller or equal to 1
assert_op = tf.Assert(tf.less_equal(tf.reduce_max(x), 1.), [x])
with tf.control_dependencies([assert_op]):
... code using x ...
```
@end_compatibility"
8229,exit,tensorflow/tensorflow/python/ops/control_flow_ops.py,264,function,"Exits the current frame to its parent frame.
Exit makes its input `data` available to the parent frame.
Args:
data: The tensor to be made available to the parent frame.
name: A name for this operation (optional).
Returns:
The same tensor as `data`."
8230,switch,tensorflow/tensorflow/python/ops/control_flow_ops.py,288,function,"Forwards `data` to an output determined by `pred`.
If `pred` is false, the `data` input is forwarded to the first output.
Otherwise, the data goes to the second output.
This op handles `Tensor`s and `IndexedSlices`.
Args:
data: The tensor to be forwarded to the appropriate output.
pred: A scalar that specifies which output port will receive data.
dtype: Optional element type for the returned tensor. If missing, the type
is inferred from the type of `value`.
name: A name for this operation (optional).
Returns:
`(output_false, output_true)`: If `pred` is true, data will be forwarded
to `output_true`, otherwise it goes to `output_false`."
8231,merge,tensorflow/tensorflow/python/ops/control_flow_ops.py,367,function,"Returns the value of an available element of `inputs`.
This op tests each of the tensors in `inputs` in turn to determine if any of
them is available. If it finds an available tensor, it returns it and its
index in `inputs`.
It is an error if more than one tensor in `inputs` is available. If no tensor
in `inputs` is available, the returned tensor and index are not set.
This op handles both `Tensor`s and `IndexedSlices`. If inputs has a mix of
`Tensor`s and `IndexedSlices`, all inputs are converted to IndexedSlices
before merging.
Args:
inputs: The input tensors, at most one of which is available.
name: A name for this operation (optional).
Returns:
A tuple containing the chosen input tensor and its index in `inputs`.
Raises:
ValueError: If any of the inputs is None, or inputs are IndexedSlices and
some but not all have a dense_shape property."
8232,ControlFlowContext,tensorflow/tensorflow/python/ops/control_flow_ops.py,624,class,"The base class for control flow context.
The usage pattern is a sequence of (Enter, Exit) followed by a final
ExitResult.
We maintain the following state for control flow contexts during graph
construction:
1. graph has _control_flow_context: the current context used to
construct new nodes. Changed by ctxt.Enter() and ctxt.Exit()
2. op has _control_flow_context: the context to which the op belongs.
Set at the time the op is created. Immutable.
3. A ControlFlowContext has _outer_context: the context in which this
context is created. Set at the time a context is created. Immutable.
4. A ControlFlowContext has _context_stack.
Pushed and popped by ctxt.Enter() and ctxt.Exit()"
8233,name,tensorflow/tensorflow/python/ops/control_flow_ops.py,685,method,
8234,outer_context,tensorflow/tensorflow/python/ops/control_flow_ops.py,689,method,Return the context containing this context.
8235,grad_state,tensorflow/tensorflow/python/ops/control_flow_ops.py,694,method,
8236,back_prop,tensorflow/tensorflow/python/ops/control_flow_ops.py,698,method,
8237,to_control_flow_context_def,tensorflow/tensorflow/python/ops/control_flow_ops.py,702,method,"Serializes this into `context_def`.
Args:
context_def: a `ControlFlowContextDef` protocol buffer.
export_scope: Optional `string`. Name scope to remove."
8238,AddName,tensorflow/tensorflow/python/ops/control_flow_ops.py,728,method,
8239,Enter,tensorflow/tensorflow/python/ops/control_flow_ops.py,732,method,Enter this control flow context.
8240,Exit,tensorflow/tensorflow/python/ops/control_flow_ops.py,738,method,Exit this control flow context.
8241,EnterGradientColocation,tensorflow/tensorflow/python/ops/control_flow_ops.py,744,method,Start building a gradient colocated with an op.
8242,ExitGradientColocation,tensorflow/tensorflow/python/ops/control_flow_ops.py,749,method,Start building a gradient colocated with an op.
8243,ExitResult,tensorflow/tensorflow/python/ops/control_flow_ops.py,754,method,Make a list of tensors available in the outer context.
8244,GetWhileContext,tensorflow/tensorflow/python/ops/control_flow_ops.py,762,method,Return the while context containing this context.
8245,AddInnerOp,tensorflow/tensorflow/python/ops/control_flow_ops.py,791,method,Notifies a scope about an operator added to an inner scope.
8246,GetControlPivot,tensorflow/tensorflow/python/ops/control_flow_ops.py,796,method,"Returns the pivot node for this context, or None."
8247,IsWhileContext,tensorflow/tensorflow/python/ops/control_flow_ops.py,800,method,
8248,IsCondContext,tensorflow/tensorflow/python/ops/control_flow_ops.py,803,method,
8249,IsXLAContext,tensorflow/tensorflow/python/ops/control_flow_ops.py,806,method,
8250,fn,tensorflow/tensorflow/python/ops/control_flow_ops.py,757,method,
8251,CondContext,tensorflow/tensorflow/python/ops/control_flow_ops.py,813,class,The context for the conditional construct.
8252,pred,tensorflow/tensorflow/python/ops/control_flow_ops.py,873,method,
8253,pivot,tensorflow/tensorflow/python/ops/control_flow_ops.py,877,method,
8254,branch,tensorflow/tensorflow/python/ops/control_flow_ops.py,881,method,
8255,grad_state,tensorflow/tensorflow/python/ops/control_flow_ops.py,885,method,
8256,back_prop,tensorflow/tensorflow/python/ops/control_flow_ops.py,891,method,
8257,GetControlPivot,tensorflow/tensorflow/python/ops/control_flow_ops.py,896,method,
8258,to_proto,tensorflow/tensorflow/python/ops/control_flow_ops.py,899,method,"Converts a `CondContext` to a `CondContextDef` protocol buffer.
Args:
export_scope: Optional `string`. Name scope to remove.
Returns:
A `CondContextDef` protocol buffer."
8259,from_proto,tensorflow/tensorflow/python/ops/control_flow_ops.py,927,method,Returns a `CondContext` object created from `context_def`.
8260,to_control_flow_context_def,tensorflow/tensorflow/python/ops/control_flow_ops.py,937,method,
8261,AddValue,tensorflow/tensorflow/python/ops/control_flow_ops.py,940,method,Add `val` to the current context and its outer context recursively.
8262,AddOp,tensorflow/tensorflow/python/ops/control_flow_ops.py,976,method,
8263,BuildCondBranch,tensorflow/tensorflow/python/ops/control_flow_ops.py,1063,method,Add the subgraph defined by fn() to the graph.
8264,IsCondContext,tensorflow/tensorflow/python/ops/control_flow_ops.py,1087,method,
8265,cond,tensorflow/tensorflow/python/ops/control_flow_ops.py,1105,function,"Return `true_fn()` if the predicate `pred` is true else `false_fn()`.
`true_fn` and `false_fn` both return lists of output tensors. `true_fn` and
`false_fn` must have the same non-zero number and type of outputs.
**WARNING**: Any Tensors or Operations created outside of `true_fn` and
`false_fn` will be executed regardless of which branch is selected at runtime.
Although this behavior is consistent with the dataflow model of TensorFlow,
it has frequently surprised users who expected a lazier semantics.
Consider the following simple program:
```python
z = tf.multiply(a, b)
result = tf.cond(x < y, lambda: tf.add(x, z), lambda: tf.square(y))
```
If `x < y`, the `tf.add` operation will be executed and `tf.square`
operation will not be executed. Since `z` is needed for at least one
branch of the `cond`, the `tf.multiply` operation is always executed,
unconditionally.
Note that `cond` calls `true_fn` and `false_fn` *exactly once* (inside the
call to `cond`, and not at all during `Session.run()`). `cond`
stitches together the graph fragments created during the `true_fn` and
`false_fn` calls with some additional graph nodes to ensure that the right
branch gets executed depending on the value of `pred`.
`tf.cond` supports nested structures as implemented in
`tensorflow.python.util.nest`. Both `true_fn` and `false_fn` must return the
same (possibly nested) value structure of lists, tuples, and/or named tuples.
Singleton lists and tuples form the only exceptions to this: when returned by
`true_fn` and/or `false_fn`, they are implicitly unpacked to single values.
This behavior is disabled by passing `strict=True`.
Args:
pred: A scalar determining whether to return the result of `true_fn` or
`false_fn`.
true_fn: The callable to be performed if pred is true.
false_fn: The callable to be performed if pred is false.
strict: A boolean that enables/disables 'strict' mode; see above.
name: Optional name prefix for the returned tensors.
Returns:
Tensors returned by the call to either `true_fn` or `false_fn`. If the
callables return a singleton list, the element is extracted from the list.
Raises:
TypeError: if `true_fn` or `false_fn` is not callable.
ValueError: if `true_fn` and `false_fn` do not return the same number of
tensors, or return tensors of different types.
Example:
```python
x = tf.constant(2)
y = tf.constant(5)
def f1(): return tf.multiply(x, 17)
def f2(): return tf.add(y, 23)
r = tf.cond(tf.less(x, y), f1, f2)
# r is set to f1().
# Operations in f2 (e.g., tf.add) are not executed.
```"
8266,cond_for_tf_v2,tensorflow/tensorflow/python/ops/control_flow_ops.py,1326,function,"Return `true_fn()` if the predicate `pred` is true else `false_fn()`.
`true_fn` and `false_fn` both return lists of output tensors. `true_fn` and
`false_fn` must have the same non-zero number and type of outputs.
**WARNING**: Any Tensors or Operations created outside of `true_fn` and
`false_fn` will be executed regardless of which branch is selected at runtime.
Although this behavior is consistent with the dataflow model of TensorFlow,
it has frequently surprised users who expected a lazier semantics.
Consider the following simple program:
```python
z = tf.multiply(a, b)
result = tf.cond(x < y, lambda: tf.add(x, z), lambda: tf.square(y))
```
If `x < y`, the `tf.add` operation will be executed and `tf.square`
operation will not be executed. Since `z` is needed for at least one
branch of the `cond`, the `tf.multiply` operation is always executed,
unconditionally.
Note that `cond` calls `true_fn` and `false_fn` *exactly once* (inside the
call to `cond`, and not at all during `Session.run()`). `cond`
stitches together the graph fragments created during the `true_fn` and
`false_fn` calls with some additional graph nodes to ensure that the right
branch gets executed depending on the value of `pred`.
`tf.cond` supports nested structures as implemented in
`tensorflow.python.util.nest`. Both `true_fn` and `false_fn` must return the
same (possibly nested) value structure of lists, tuples, and/or named tuples.
Singleton lists and tuples form the only exceptions to this: when returned by
`true_fn` and/or `false_fn`, they are implicitly unpacked to single values.
Note: It is illegal to ""directly"" use tensors created inside a cond branch
outside it, e.g. by storing a reference to a branch tensor in the python
state. If you need to use a tensor created in a branch function you should
return it as an output of the branch function and use the output from
`tf.cond` instead.
Args:
pred: A scalar determining whether to return the result of `true_fn` or
`false_fn`.
true_fn: The callable to be performed if pred is true.
false_fn: The callable to be performed if pred is false.
name: Optional name prefix for the returned tensors.
Returns:
Tensors returned by the call to either `true_fn` or `false_fn`. If the
callables return a singleton list, the element is extracted from the list.
Raises:
TypeError: if `true_fn` or `false_fn` is not callable.
ValueError: if `true_fn` and `false_fn` do not return the same number of
tensors, or return tensors of different types.
Example:
```python
x = tf.constant(2)
y = tf.constant(5)
def f1(): return tf.multiply(x, 17)
def f2(): return tf.add(y, 23)
r = tf.cond(tf.less(x, y), f1, f2)
# r is set to f1().
# Operations in f2 (e.g., tf.add) are not executed.
```"
8267,WhileContext,tensorflow/tensorflow/python/ops/control_flow_ops.py,1411,class,The context for the loop construct.
8268,maximum_iterations,tensorflow/tensorflow/python/ops/control_flow_ops.py,1536,method,The maximum number of iterations that will be executed.
8269,parallel_iterations,tensorflow/tensorflow/python/ops/control_flow_ops.py,1541,method,The number of iterations allowed to run in parallel.
8270,back_prop,tensorflow/tensorflow/python/ops/control_flow_ops.py,1546,method,True iff backprop is enabled for this while loop.
8271,swap_memory,tensorflow/tensorflow/python/ops/control_flow_ops.py,1551,method,True iff GPU-CPU memory swap is enabled for this while loop.
8272,pivot,tensorflow/tensorflow/python/ops/control_flow_ops.py,1556,method,The boolean tensor representing the loop termination condition.
8273,loop_enters,tensorflow/tensorflow/python/ops/control_flow_ops.py,1561,method,The list of enter tensors for loop variables.
8274,loop_exits,tensorflow/tensorflow/python/ops/control_flow_ops.py,1566,method,The list of exit tensors for loop variables.
8275,grad_state,tensorflow/tensorflow/python/ops/control_flow_ops.py,1571,method,The gradient loop state.
8276,to_proto,tensorflow/tensorflow/python/ops/control_flow_ops.py,1575,method,"Converts a `WhileContext` to a `WhileContextDef` protocol buffer.
Args:
export_scope: Optional `string`. Name scope to remove.
Returns:
A `WhileContextDef` protocol buffer."
8277,to_control_flow_context_def,tensorflow/tensorflow/python/ops/control_flow_ops.py,1615,method,
8278,from_proto,tensorflow/tensorflow/python/ops/control_flow_ops.py,1619,method,"Returns a `WhileContext` object created from `context_def`.
Args:
context_def: A `WhileContextDef` protocol buffer.
import_scope: Optional `string`. Name scope to add.
Returns:
A `WhileContext` Python object."
8279,GetWhileContext,tensorflow/tensorflow/python/ops/control_flow_ops.py,1636,method,
8280,GetControlPivot,tensorflow/tensorflow/python/ops/control_flow_ops.py,1639,method,
8281,AddValue,tensorflow/tensorflow/python/ops/control_flow_ops.py,1644,method,Add `val` to the current context and its outer context recursively.
8282,AddOp,tensorflow/tensorflow/python/ops/control_flow_ops.py,1700,method,Add `op` to the current context.
8283,AddForwardLoopCounter,tensorflow/tensorflow/python/ops/control_flow_ops.py,1799,method,"Adds a loop that counts the number of iterations.
This is added to the forward loop at the time when we start to
create the loop for backprop gradient computation. Called in
the outer context of this forward context.
The pseudocode is:
`n = 0; while (_pivot) { n++; }`
Note that a control dependency is added to `n` to ensure the correct
execution order of stack push ops.
Args:
outer_grad_state: The outer grad state. None if not nested.
Returns:
The number of iterations taken by the forward loop and the loop index."
8284,AddBackpropLoopCounter,tensorflow/tensorflow/python/ops/control_flow_ops.py,1848,method,"Add the backprop loop that controls the iterations.
This is added to the backprop loop. It is used to control the loop
termination of the backprop loop. Called in the outer context of
this grad context.
The pseudocode is:
`n = count; while (n >= 1) { n--; }`
Note that a control dependency is added to `final_zero` to ensure the
correct execution order of stack pop ops.
Args:
count: The number of iterations for backprop.
outer_grad_state: The outer grad state. None if not nested.
Returns:
The loop index."
8285,AddBackpropAccumulator,tensorflow/tensorflow/python/ops/control_flow_ops.py,1914,method,"Add an accumulation loop for every loop invariant.
This is added to the backprop loop. It is used to accumulate partial
gradients within each loop iteration. Called when in the gradient while
context.
The pseudocode is:
```
acc = 0.0;
while (_pivot) {
acc += grad;
}
```
Args:
op: The Enter op for a loop invariant.
grad: The partial gradient of an iteration for a loop invariant.
Returns:
The gradient for a loop invariant."
8286,AddBackpropIndexedSlicesAccumulator,tensorflow/tensorflow/python/ops/control_flow_ops.py,1996,method,"This is used for accumulating gradients that are IndexedSlices.
This is essentially the equivalent of AddBackpropAccumulator but optimized
for things like updating embeddings from within a while loop.
Args:
op: The Enter op for a loop invariant.
grad: The partial gradients represented as an IndexedSlices.
Returns:
The accumulated IndexedSlices gradient of the loop invariant."
8287,BuildLoop,tensorflow/tensorflow/python/ops/control_flow_ops.py,2233,method,Add the loop termination condition and body to the graph.
8288,IsWhileContext,tensorflow/tensorflow/python/ops/control_flow_ops.py,2307,method,
8289,map_fn,tensorflow/tensorflow/python/ops/control_flow_ops.py,2191,method,
8290,while_loop_v2,tensorflow/tensorflow/python/ops/control_flow_ops.py,2323,function,"Repeat `body` while the condition `cond` is true.
`cond` is a callable returning a boolean scalar tensor. `body` is a callable
returning a (possibly nested) tuple, namedtuple or list of tensors of the same
arity (length and structure) and types as `loop_vars`. `loop_vars` is a
(possibly nested) tuple, namedtuple or list of tensors that is passed to both
`cond` and `body`. `cond` and `body` both take as many arguments as there are
`loop_vars`.
In addition to regular Tensors or IndexedSlices, the body may accept and
return TensorArray objects. The flows of the TensorArray objects will
be appropriately forwarded between loops and during gradient calculations.
Note that `while_loop` calls `cond` and `body` *exactly once* (inside the
call to `while_loop`, and not at all during `Session.run()`). `while_loop`
stitches together the graph fragments created during the `cond` and `body`
calls with some additional graph nodes to create the graph flow that
repeats `body` until `cond` returns false.
For correctness, `tf.while_loop()` strictly enforces shape invariants for
the loop variables. A shape invariant is a (possibly partial) shape that
is unchanged across the iterations of the loop. An error will be raised
if the shape of a loop variable after an iteration is determined to be more
general than or incompatible with its shape invariant. For example, a shape
of [11, None] is more general than a shape of [11, 17], and [11, 21] is not
compatible with [11, 17]. By default (if the argument `shape_invariants` is
not specified), it is assumed that the initial shape of each tensor in
`loop_vars` is the same in every iteration. The `shape_invariants` argument
allows the caller to specify a less specific shape invariant for each loop
variable, which is needed if the shape varies between iterations. The
`tf.Tensor.set_shape`
function may also be used in the `body` function to indicate that
the output loop variable has a particular shape. The shape invariant for
SparseTensor and IndexedSlices are treated specially as follows:
a) If a loop variable is a SparseTensor, the shape invariant must be
TensorShape([r]) where r is the rank of the dense tensor represented
by the sparse tensor. It means the shapes of the three tensors of the
SparseTensor are ([None], [None, r], [r]). NOTE: The shape invariant here
is the shape of the SparseTensor.dense_shape property. It must be the shape of
a vector.
b) If a loop variable is an IndexedSlices, the shape invariant must be
a shape invariant of the values tensor of the IndexedSlices. It means
the shapes of the three tensors of the IndexedSlices are (shape, [shape[0]],
[shape.ndims]).
`while_loop` implements non-strict semantics, enabling multiple iterations
to run in parallel. The maximum number of parallel iterations can be
controlled by `parallel_iterations`, which gives users some control over
memory consumption and execution order. For correct programs, `while_loop`
should return the same result for any parallel_iterations > 0.
For training, TensorFlow stores the tensors that are produced in the
forward inference and are needed in back propagation. These tensors are a
main source of memory consumption and often cause OOM errors when training
on GPUs. When the flag swap_memory is true, we swap out these tensors from
GPU to CPU. This for example allows us to train RNN models with very long
sequences and large batches.
Args:
cond: A callable that represents the termination condition of the loop.
body: A callable that represents the loop body.
loop_vars: A (possibly nested) tuple, namedtuple or list of numpy array,
`Tensor`, and `TensorArray` objects.
shape_invariants: The shape invariants for the loop variables.
parallel_iterations: The number of iterations allowed to run in parallel. It
must be a positive integer.
back_prop: (optional) Deprecated. False disables support for back
propagation. Prefer using `tf.stop_gradient` instead.
swap_memory: Whether GPU-CPU memory swap is enabled for this loop.
maximum_iterations: Optional maximum number of iterations of the while loop
to run. If provided, the `cond` output is AND-ed with an additional
condition ensuring the number of iterations executed is no greater than
`maximum_iterations`.
name: Optional name prefix for the returned tensors.
Returns:
The output tensors for the loop variables after the loop. The return value
has the same structure as `loop_vars`.
Raises:
TypeError: if `cond` or `body` is not callable.
ValueError: if `loop_vars` is empty.
Example:
```python
i = tf.constant(0)
c = lambda i: tf.less(i, 10)
b = lambda i: (tf.add(i, 1), )
r = tf.while_loop(c, b, [i])
```
Example with nesting and a namedtuple:
```python
import collections
Pair = collections.namedtuple('Pair', 'j, k')
ijk_0 = (tf.constant(0), Pair(tf.constant(1), tf.constant(2)))
c = lambda i, p: i < 10
b = lambda i, p: (i + 1, Pair((p.j + p.k), (p.j - p.k)))
ijk_final = tf.while_loop(c, b, ijk_0)
```
Example using shape_invariants:
```python
i0 = tf.constant(0)
m0 = tf.ones([2, 2])
c = lambda i, m: i < 10
b = lambda i, m: [i+1, tf.concat([m, m], axis=0)]
tf.while_loop(
c, b, loop_vars=[i0, m0],
shape_invariants=[i0.get_shape(), tf.TensorShape([None, 2])])
```
Example which demonstrates non-strict semantics: In the following
example, the final value of the counter `i` does not depend on `x`. So
the `while_loop` can increment the counter parallel to updates of `x`.
However, because the loop counter at one loop iteration depends
on the value at the previous iteration, the loop counter itself cannot
be incremented in parallel. Hence if we just want the final value of the
counter (which we print on the line `print(sess.run(i))`), then
`x` will never be incremented, but the counter will be updated on a
single thread. Conversely, if we want the value of the output (which we
print on the line `print(sess.run(out).shape)`), then the counter may be
incremented on its own thread, while `x` can be incremented in
parallel on a separate thread. In the extreme case, it is conceivable
that the thread incrementing the counter runs until completion before
`x` is incremented even a single time. The only thing that can never
happen is that the thread updating `x` can never get ahead of the
counter thread because the thread incrementing `x` depends on the value
of the counter.
```python
import tensorflow as tf
n = 10000
x = tf.constant(list(range(n)))
c = lambda i, x: i < n
b = lambda i, x: (tf.compat.v1.Print(i + 1, [i]), tf.compat.v1.Print(x + 1,
[i], ""x:""))
i, out = tf.while_loop(c, b, (0, x))
with tf.compat.v1.Session() as sess:
print(sess.run(i)) # prints [0] ... [9999]
# The following line may increment the counter and x in parallel.
# The counter thread may get ahead of the other thread, but not the
# other way around. So you may see things like
# [9996] x:[9987]
# meaning that the counter thread is on iteration 9996,
# while the other thread is on iteration 9987
print(sess.run(out).shape)
```"
8291,while_loop,tensorflow/tensorflow/python/ops/control_flow_ops.py,2504,function,"Repeat `body` while the condition `cond` is true.
`cond` is a callable returning a boolean scalar tensor. `body` is a callable
returning a (possibly nested) tuple, namedtuple or list of tensors of the same
arity (length and structure) and types as `loop_vars`. `loop_vars` is a
(possibly nested) tuple, namedtuple or list of tensors that is passed to both
`cond` and `body`. `cond` and `body` both take as many arguments as there are
`loop_vars`.
In addition to regular Tensors or IndexedSlices, the body may accept and
return TensorArray objects. The flows of the TensorArray objects will
be appropriately forwarded between loops and during gradient calculations.
Note that `while_loop` calls `cond` and `body` *exactly once* (inside the
call to `while_loop`, and not at all during `Session.run()`). `while_loop`
stitches together the graph fragments created during the `cond` and `body`
calls with some additional graph nodes to create the graph flow that
repeats `body` until `cond` returns false.
For correctness, `tf.while_loop()` strictly enforces shape invariants for
the loop variables. A shape invariant is a (possibly partial) shape that
is unchanged across the iterations of the loop. An error will be raised
if the shape of a loop variable after an iteration is determined to be more
general than or incompatible with its shape invariant. For example, a shape
of [11, None] is more general than a shape of [11, 17], and [11, 21] is not
compatible with [11, 17]. By default (if the argument `shape_invariants` is
not specified), it is assumed that the initial shape of each tensor in
`loop_vars` is the same in every iteration. The `shape_invariants` argument
allows the caller to specify a less specific shape invariant for each loop
variable, which is needed if the shape varies between iterations. The
`tf.Tensor.set_shape`
function may also be used in the `body` function to indicate that
the output loop variable has a particular shape. The shape invariant for
SparseTensor and IndexedSlices are treated specially as follows:
a) If a loop variable is a SparseTensor, the shape invariant must be
TensorShape([r]) where r is the rank of the dense tensor represented
by the sparse tensor. It means the shapes of the three tensors of the
SparseTensor are ([None], [None, r], [r]). NOTE: The shape invariant here
is the shape of the SparseTensor.dense_shape property. It must be the shape of
a vector.
b) If a loop variable is an IndexedSlices, the shape invariant must be
a shape invariant of the values tensor of the IndexedSlices. It means
the shapes of the three tensors of the IndexedSlices are (shape, [shape[0]],
[shape.ndims]).
`while_loop` implements non-strict semantics, enabling multiple iterations
to run in parallel. The maximum number of parallel iterations can be
controlled by `parallel_iterations`, which gives users some control over
memory consumption and execution order. For correct programs, `while_loop`
should return the same result for any parallel_iterations > 0.
For training, TensorFlow stores the tensors that are produced in the
forward inference and are needed in back propagation. These tensors are a
main source of memory consumption and often cause OOM errors when training
on GPUs. When the flag swap_memory is true, we swap out these tensors from
GPU to CPU. This for example allows us to train RNN models with very long
sequences and large batches.
Args:
cond: A callable that represents the termination condition of the loop.
body: A callable that represents the loop body.
loop_vars: A (possibly nested) tuple, namedtuple or list of numpy array,
`Tensor`, and `TensorArray` objects.
shape_invariants: The shape invariants for the loop variables.
parallel_iterations: The number of iterations allowed to run in parallel. It
must be a positive integer.
back_prop: Whether backprop is enabled for this while loop.
swap_memory: Whether GPU-CPU memory swap is enabled for this loop.
name: Optional name prefix for the returned tensors.
maximum_iterations: Optional maximum number of iterations of the while loop
to run. If provided, the `cond` output is AND-ed with an additional
condition ensuring the number of iterations executed is no greater than
`maximum_iterations`.
return_same_structure: If True, output has same structure as `loop_vars`. If
eager execution is enabled, this is ignored (and always treated as True).
Returns:
The output tensors for the loop variables after the loop.
If `return_same_structure` is True, the return value has the same
structure as `loop_vars`.
If `return_same_structure` is False, the return value is a Tensor,
TensorArray or IndexedSlice if the length of `loop_vars` is 1, or a list
otherwise.
Raises:
TypeError: if `cond` or `body` is not callable.
ValueError: if `loop_vars` is empty.
Example:
```python
i = tf.constant(0)
c = lambda i: tf.less(i, 10)
b = lambda i: tf.add(i, 1)
r = tf.while_loop(c, b, [i])
```
Example with nesting and a namedtuple:
```python
import collections
Pair = collections.namedtuple('Pair', 'j, k')
ijk_0 = (tf.constant(0), Pair(tf.constant(1), tf.constant(2)))
c = lambda i, p: i < 10
b = lambda i, p: (i + 1, Pair((p.j + p.k), (p.j - p.k)))
ijk_final = tf.while_loop(c, b, ijk_0)
```
Example using shape_invariants:
```python
i0 = tf.constant(0)
m0 = tf.ones([2, 2])
c = lambda i, m: i < 10
b = lambda i, m: [i+1, tf.concat([m, m], axis=0)]
tf.while_loop(
c, b, loop_vars=[i0, m0],
shape_invariants=[i0.get_shape(), tf.TensorShape([None, 2])])
```
Example which demonstrates non-strict semantics: In the following
example, the final value of the counter `i` does not depend on `x`. So
the `while_loop` can increment the counter parallel to updates of `x`.
However, because the loop counter at one loop iteration depends
on the value at the previous iteration, the loop counter itself cannot
be incremented in parallel. Hence if we just want the final value of the
counter (which we print on the line `print(sess.run(i))`), then
`x` will never be incremented, but the counter will be updated on a
single thread. Conversely, if we want the value of the output (which we
print on the line `print(sess.run(out).shape)`), then the counter may be
incremented on its own thread, while `x` can be incremented in
parallel on a separate thread. In the extreme case, it is conceivable
that the thread incrementing the counter runs until completion before
`x` is incremented even a single time. The only thing that can never
happen is that the thread updating `x` can never get ahead of the
counter thread because the thread incrementing `x` depends on the value
of the counter.
```python
import tensorflow as tf
n = 10000
x = tf.constant(list(range(n)))
c = lambda i, x: i < n
b = lambda i, x: (tf.compat.v1.Print(i + 1, [i]), tf.compat.v1.Print(x + 1,
[i], ""x:""))
i, out = tf.while_loop(c, b, (0, x))
with tf.compat.v1.Session() as sess:
print(sess.run(i)) # prints [0] ... [9999]
# The following line may increment the counter and x in parallel.
# The counter thread may get ahead of the other thread, but not the
# other way around. So you may see things like
# [9996] x:[9987]
# meaning that the counter thread is on iteration 9996,
# while the other thread is on iteration 9987
print(sess.run(out).shape)
```"
8292,with_dependencies,tensorflow/tensorflow/python/ops/control_flow_ops.py,2823,function,"Produces the content of `output_tensor` only after `dependencies`.
In some cases, a user may want the output of an operation to be
consumed externally only after some other dependencies have run
first. This function ensures returns `output_tensor`, but only after all
operations in `dependencies` have run. Note that this means that there is
no guarantee that `output_tensor` will be evaluated after any `dependencies`
have run.
See also `tf.tuple` and `tf.group`.
Args:
dependencies: Iterable of operations to run before this op finishes.
output_tensor: A `Tensor` or `IndexedSlices` that will be returned.
name: (Optional) A name for this operation.
Returns:
Same as `output_tensor`.
Raises:
TypeError: if `output_tensor` is not a `Tensor` or `IndexedSlices`."
8293,group,tensorflow/tensorflow/python/ops/control_flow_ops.py,2872,function,"Create an op that groups multiple operations.
When this op finishes, all ops in `inputs` have finished. This op has no
output.
Note: *In TensorFlow 2 with eager and/or Autograph, you should not require
this method, as code executes in your expected order.* Only use tf.group when
working with v1-style code or in a graph context such as inside `Dataset.map`.
When operating in a v1-style graph context, ops are not executed in the same
order as specified in the code; TensorFlow will attempt to execute ops in
parallel or in an order convienient to the result it is computing. `tf.group`
allows you to request that one or more results finish before execution
continues.
`tf.group` creates a single op (of type `NoOp`), and then adds appropriate
control dependencies. Thus, `c = tf.group(a, b)` will compute the same graph
as this:
with tf.control_dependencies([a, b]):
c = tf.no_op()
See also `tf.tuple` and
`tf.control_dependencies`.
Args:
*inputs: Zero or more tensors to group.
name: A name for this operation (optional).
Returns:
An Operation that executes all its inputs.
Raises:
ValueError: If an unknown keyword argument is provided."
8294,tuple_v2,tensorflow/tensorflow/python/ops/control_flow_ops.py,2951,function,"Group tensors together.
This creates a tuple of tensors with the same values as the `tensors`
argument, except that the value of each tensor is only returned after the
values of all tensors have been computed.
`control_inputs` contains additional ops that have to finish before this op
finishes, but whose outputs are not returned.
This can be used as a ""join"" mechanism for parallel computations: all the
argument tensors can be computed in parallel, but the values of any tensor
returned by `tuple` are only available after all the parallel computations
are done.
See also `tf.group` and
`tf.control_dependencies`.
Args:
tensors: A list of `Tensor`s or `IndexedSlices`, some entries can be `None`.
control_inputs: List of additional ops to finish before returning.
name: (optional) A name to use as a `name_scope` for the operation.
Returns:
Same as `tensors`.
Raises:
ValueError: If `tensors` does not contain any `Tensor` or `IndexedSlices`.
TypeError: If `control_inputs` is not a list of `Operation` or `Tensor`
objects."
8295,tuple,tensorflow/tensorflow/python/ops/control_flow_ops.py,2988,function,"Group tensors together.
This creates a tuple of tensors with the same values as the `tensors`
argument, except that the value of each tensor is only returned after the
values of all tensors have been computed.
`control_inputs` contains additional ops that have to finish before this op
finishes, but whose outputs are not returned.
This can be used as a ""join"" mechanism for parallel computations: all the
argument tensors can be computed in parallel, but the values of any tensor
returned by `tuple` are only available after all the parallel computations
are done.
See also `tf.group` and
`tf.control_dependencies`.
Args:
tensors: A list of `Tensor`s or `IndexedSlices`, some entries can be `None`.
name: (optional) A name to use as a `name_scope` for the operation.
control_inputs: List of additional ops to finish before returning.
Returns:
Same as `tensors`.
Raises:
ValueError: If `tensors` does not contain any `Tensor` or `IndexedSlices`.
TypeError: If `control_inputs` is not a list of `Operation` or `Tensor`
objects."
8296,case_v2,tensorflow/tensorflow/python/ops/control_flow_ops.py,3331,function,"Create a case operation.
See also `tf.switch_case`.
The `pred_fn_pairs` parameter is a list of pairs of size N.
Each pair contains a boolean scalar tensor and a python callable that
creates the tensors to be returned if the boolean evaluates to True.
`default` is a callable generating a list of tensors. All the callables
in `pred_fn_pairs` as well as `default` (if provided) should return the same
number and types of tensors.
If `exclusive==True`, all predicates are evaluated, and an exception is
thrown if more than one of the predicates evaluates to `True`.
If `exclusive==False`, execution stops at the first predicate which
evaluates to True, and the tensors generated by the corresponding function
are returned immediately. If none of the predicates evaluate to True, this
operation returns the tensors generated by `default`.
`tf.case` supports nested structures as implemented in
`tf.contrib.framework.nest`. All of the callables must return the same
(possibly nested) value structure of lists, tuples, and/or named tuples.
Singleton lists and tuples form the only exceptions to this: when returned by
a callable, they are implicitly unpacked to single values. This
behavior is disabled by passing `strict=True`.
@compatibility(v2)
`pred_fn_pairs` could be a dictionary in v1. However, tf.Tensor and
tf.Variable are no longer hashable in v2, so cannot be used as a key for a
dictionary. Please use a list or a tuple instead.
@end_compatibility
**Example 1:**
Pseudocode:
```
if (x < y) return 17;
else return 23;
```
Expressions:
```python
f1 = lambda: tf.constant(17)
f2 = lambda: tf.constant(23)
r = tf.case([(tf.less(x, y), f1)], default=f2)
```
**Example 2:**
Pseudocode:
```
if (x < y && x > z) raise OpError(""Only one predicate may evaluate to True"");
if (x < y) return 17;
else if (x > z) return 23;
else return -1;
```
Expressions:
```python
def f1(): return tf.constant(17)
def f2(): return tf.constant(23)
def f3(): return tf.constant(-1)
r = tf.case([(tf.less(x, y), f1), (tf.greater(x, z), f2)],
default=f3, exclusive=True)
```
Args:
pred_fn_pairs: List of pairs of a boolean scalar tensor and a callable which
returns a list of tensors.
default: Optional callable that returns a list of tensors.
exclusive: True iff at most one predicate is allowed to evaluate to `True`.
strict: A boolean that enables/disables 'strict' mode; see above.
name: A name for this operation (optional).
Returns:
The tensors returned by the first pair whose predicate evaluated to True, or
those returned by `default` if none does.
Raises:
TypeError: If `pred_fn_pairs` is not a list/tuple.
TypeError: If `pred_fn_pairs` is a list but does not contain 2-tuples.
TypeError: If `fns[i]` is not callable for any i, or `default` is not
callable."
8297,case,tensorflow/tensorflow/python/ops/control_flow_ops.py,3436,function,"Create a case operation.
See also `tf.switch_case`.
The `pred_fn_pairs` parameter is a dict or list of pairs of size N.
Each pair contains a boolean scalar tensor and a python callable that
creates the tensors to be returned if the boolean evaluates to True.
`default` is a callable generating a list of tensors. All the callables
in `pred_fn_pairs` as well as `default` (if provided) should return the same
number and types of tensors.
If `exclusive==True`, all predicates are evaluated, and an exception is
thrown if more than one of the predicates evaluates to `True`.
If `exclusive==False`, execution stops at the first predicate which
evaluates to True, and the tensors generated by the corresponding function
are returned immediately. If none of the predicates evaluate to True, this
operation returns the tensors generated by `default`.
`tf.case` supports nested structures as implemented in
`tf.contrib.framework.nest`. All of the callables must return the same
(possibly nested) value structure of lists, tuples, and/or named tuples.
Singleton lists and tuples form the only exceptions to this: when returned by
a callable, they are implicitly unpacked to single values. This
behavior is disabled by passing `strict=True`.
If an unordered dictionary is used for `pred_fn_pairs`, the order of the
conditional tests is not guaranteed. However, the order is guaranteed to be
deterministic, so that variables created in conditional branches are created
in fixed order across runs.
@compatibility(eager)
Unordered dictionaries are not supported in eager mode when `exclusive=False`.
Use a list of tuples instead.
@end_compatibility
**Example 1:**
Pseudocode:
```
if (x < y) return 17;
else return 23;
```
Expressions:
```python
f1 = lambda: tf.constant(17)
f2 = lambda: tf.constant(23)
r = tf.case([(tf.less(x, y), f1)], default=f2)
```
**Example 2:**
Pseudocode:
```
if (x < y && x > z) raise OpError(""Only one predicate may evaluate to True"");
if (x < y) return 17;
else if (x > z) return 23;
else return -1;
```
Expressions:
```python
def f1(): return tf.constant(17)
def f2(): return tf.constant(23)
def f3(): return tf.constant(-1)
r = tf.case({tf.less(x, y): f1, tf.greater(x, z): f2},
default=f3, exclusive=True)
```
Args:
pred_fn_pairs: Dict or list of pairs of a boolean scalar tensor and a
callable which returns a list of tensors.
default: Optional callable that returns a list of tensors.
exclusive: True iff at most one predicate is allowed to evaluate to `True`.
strict: A boolean that enables/disables 'strict' mode; see above.
name: A name for this operation (optional).
Returns:
The tensors returned by the first pair whose predicate evaluated to True, or
those returned by `default` if none does.
Raises:
TypeError: If `pred_fn_pairs` is not a list/dictionary.
TypeError: If `pred_fn_pairs` is a list but does not contain 2-tuples.
TypeError: If `fns[i]` is not callable for any i, or `default` is not
callable."
8298,switch_case,tensorflow/tensorflow/python/ops/control_flow_ops.py,3544,function,"Create a switch/case operation, i.e. an integer-indexed conditional.
See also `tf.case`.
This op can be substantially more efficient than `tf.case` when exactly one
branch will be selected. `tf.switch_case` is more like a C++ switch/case
statement than `tf.case`, which is more like an if/elif/elif/else chain.
The `branch_fns` parameter is either a dict from `int` to callables, or list
of (`int`, callable) pairs, or simply a list of callables (in which case the
index is implicitly the key). The `branch_index` `Tensor` is used to select an
element in `branch_fns` with matching `int` key, falling back to `default`
if none match, or `max(keys)` if no `default` is provided. The keys must form
a contiguous set from `0` to `len(branch_fns) - 1`.
`tf.switch_case` supports nested structures as implemented in `tf.nest`. All
callables must return the same (possibly nested) value structure of lists,
tuples, and/or named tuples.
**Example:**
Pseudocode:
```c++
switch (branch_index) { // c-style switch
case 0: return 17;
case 1: return 31;
default: return -1;
}
```
or
```python
branches = {0: lambda: 17, 1: lambda: 31}
branches.get(branch_index, lambda: -1)()
```
Expressions:
```python
def f1(): return tf.constant(17)
def f2(): return tf.constant(31)
def f3(): return tf.constant(-1)
r = tf.switch_case(branch_index, branch_fns={0: f1, 1: f2}, default=f3)
# Equivalent: tf.switch_case(branch_index, branch_fns={0: f1, 1: f2, 2: f3})
```
Args:
branch_index: An int Tensor specifying which of `branch_fns` should be
executed.
branch_fns: A `dict` mapping `int`s to callables, or a `list` of
(`int`, callable) pairs, or simply a list of callables (in which case the
index serves as the key). Each callable must return a matching structure
of tensors.
default: Optional callable that returns a structure of tensors.
name: A name for this operation (optional).
Returns:
The tensors returned by the callable identified by `branch_index`, or those
returned by `default` if no key matches and `default` was provided, or those
returned by the max-keyed `branch_fn` if no `default` is provided.
Raises:
TypeError: If `branch_fns` is not a list/dictionary.
TypeError: If `branch_fns` is a list but does not contain 2-tuples or
callables.
TypeError: If `fns[i]` is not callable for any i, or `default` is not
callable."
8299,execute_fn_for_device,tensorflow/tensorflow/python/ops/control_flow_ops.py,3619,function,"Executes one of the provided callables based on the device placement.
This API is used when the implementations for high level function depend on
the underlying device placement. It takes a dictionary of device type to
callables. The device type includes ""CPU"", ""GPU"", ""TPU"", etc. When the type of
the device where to run this op matches the key in 'device_branch_fns',
the corresponding callable is executed, falling back to 'default_fn' if none
matches.
**Example:**
```python
def f1(): return tf.constant(1)
def f2(): return tf.constant(2)
r = tf.execute_fn_for_device({""CPU"": f1, ""GPU"": f2}, default_fn=f1)
```
'r' is evaluated as 1 when it runs on CPU, 2 running on GPU, 1 running on
any other device types.
Args:
device_branch_fns: a dictionary of device types to the callables. Each
callable must return a matching structure of tensors.
default_fn: fallback callable when the underlying device does not match any
key in the 'device_branch_fns'.
name: A name for this operation (optional).
Returns:
The tensors returned by the callable identified by device type during
execution, or those returned by 'default_fn' if no key matches."
8300,XLAControlFlowContext,tensorflow/tensorflow/python/ops/control_flow_ops.py,3663,class,Base class for XLA and TPU control flow contexts.
8301,to_control_flow_context_def,tensorflow/tensorflow/python/ops/control_flow_ops.py,3670,method,
8302,IsXLAContext,tensorflow/tensorflow/python/ops/control_flow_ops.py,3676,method,
8303,AddOp,tensorflow/tensorflow/python/ops/control_flow_ops.py,3679,method,
8304,AddValue,tensorflow/tensorflow/python/ops/control_flow_ops.py,3682,method,
8305,RequiresUniqueFunctionRetracing,tensorflow/tensorflow/python/ops/control_flow_ops.py,3685,method,"Returns whether the tf.function should be retraced if the context changes.
"
8306,from_control_flow_context_def,tensorflow/tensorflow/python/ops/control_flow_ops.py,3691,function,"Deserializes `context_def` into the appropriate ControlFlowContext.
Args:
context_def: ControlFlowContextDef proto
import_scope: Optional `string`. Name scope to add.
Returns:
A ControlFlowContext subclass"
8307,CondWithManyIntermediatesBenchmark,tensorflow/tensorflow/python/ops/control_flow_ops_benchmark.py,36,class,Checks the runtime performance of outputting all intermediates.
8308,benchmark_cond_v1_defun,tensorflow/tensorflow/python/ops/control_flow_ops_benchmark.py,97,method,
8309,benchmark_cond_v2_defun,tensorflow/tensorflow/python/ops/control_flow_ops_benchmark.py,103,method,
8310,benchmark_cond_v1_graph,tensorflow/tensorflow/python/ops/control_flow_ops_benchmark.py,109,method,
8311,benchmark_cond_v2_graph,tensorflow/tensorflow/python/ops/control_flow_ops_benchmark.py,115,method,
8312,branch_fn,tensorflow/tensorflow/python/ops/control_flow_ops_benchmark.py,45,method,
8313,cond_fn,tensorflow/tensorflow/python/ops/control_flow_ops_benchmark.py,58,method,
8314,MaybeCreateControlFlowState,tensorflow/tensorflow/python/ops/control_flow_state.py,761,function,"Create the state for all the while loops involved in one gradients().
We create a _ControlFlowState when there are while loops involved in
gradients(). In gradients(), control flow logic is only invoked when
the _ControlFlowState is not None.
Note that this method modifies `between_op_list` and `between_ops`."
8315,ZerosLike,tensorflow/tensorflow/python/ops/control_flow_state.py,834,function,Create zeros_like for the specified output of an op.
8316,enable_control_flow_v2,tensorflow/tensorflow/python/ops/control_flow_util.py,41,function,"Use control flow v2.
Do not use this symbol. This will be removed."
8317,EnableControlFlowV2,tensorflow/tensorflow/python/ops/control_flow_util.py,50,function,Returns whether control flow v2 should be used in `graph`.
8318,IsInXLAContext,tensorflow/tensorflow/python/ops/control_flow_util.py,58,function,
8319,InXlaContext,tensorflow/tensorflow/python/ops/control_flow_util.py,68,function,
8320,GraphOrParentsInXlaContext,tensorflow/tensorflow/python/ops/control_flow_util.py,73,function,
8321,IsInWhileLoop,tensorflow/tensorflow/python/ops/control_flow_util.py,82,function,
8322,IsInCond,tensorflow/tensorflow/python/ops/control_flow_util.py,87,function,
8323,IsSwitch,tensorflow/tensorflow/python/ops/control_flow_util.py,92,function,Return true if `op` is a Switch.
8324,IsMerge,tensorflow/tensorflow/python/ops/control_flow_util.py,97,function,Return true if `op` is a Merge.
8325,IsLoopEnter,tensorflow/tensorflow/python/ops/control_flow_util.py,102,function,Returns true if `op` is an Enter.
8326,IsLoopExit,tensorflow/tensorflow/python/ops/control_flow_util.py,107,function,Return true if `op` is an Exit.
8327,IsCondSwitch,tensorflow/tensorflow/python/ops/control_flow_util.py,112,function,Return true if `op` is the Switch for a conditional.
8328,IsCondMerge,tensorflow/tensorflow/python/ops/control_flow_util.py,133,function,Return true if `op` is the Merge for a conditional.
8329,IsLoopSwitch,tensorflow/tensorflow/python/ops/control_flow_util.py,150,function,Return true if `op` is the Switch for a while loop.
8330,IsLoopMerge,tensorflow/tensorflow/python/ops/control_flow_util.py,158,function,Return true if `op` is the Merge for a while loop.
8331,IsLoopConstantEnter,tensorflow/tensorflow/python/ops/control_flow_util.py,166,function,Return true iff op is a loop invariant.
8332,GetLoopConstantEnter,tensorflow/tensorflow/python/ops/control_flow_util.py,171,function,Return the enter op if we can infer `value` to be a loop invariant.
8333,GetOutputContext,tensorflow/tensorflow/python/ops/control_flow_util.py,180,function,Return the control flow context for the output of an op.
8334,GetContainingWhileContext,tensorflow/tensorflow/python/ops/control_flow_util.py,191,function,"Returns the first ancestor WhileContext of `ctxt`.
Returns `ctxt` if `ctxt` is a WhileContext, or None if `ctxt` is not in a
while loop.
Args:
ctxt: ControlFlowContext
stop_ctxt: ControlFlowContext, optional. If provided, the search will end
if it sees stop_ctxt.
Returns:
`ctxt` if `ctxt` is a WhileContext, the most nested WhileContext containing
`ctxt`, or None if `ctxt` is not in a while loop. If `stop_ctxt` is not
`None`, this returns `ctxt` if it matches `stop_ctxt` in its traversal."
8335,GetContainingXLAContext,tensorflow/tensorflow/python/ops/control_flow_util.py,213,function,"Returns the first ancestor XLAContext of `ctxt`.
Returns `ctxt` if `ctxt` is a XLAContext, or None if `ctxt` is not in a
while loop.
Args:
ctxt: ControlFlowContext
Returns:
`ctxt` if `ctxt` is a XLAContext, the most nested XLAContext containing
`ctxt`, or None if `ctxt` is not in a while loop."
8336,GetContainingCondContext,tensorflow/tensorflow/python/ops/control_flow_util.py,232,function,"Returns the first ancestor CondContext of `ctxt`.
Returns `ctxt` if `ctxt` is a CondContext, or None if `ctxt` is not in a cond.
Args:
ctxt: ControlFlowContext
Returns:
`ctxt` if `ctxt` is a CondContext, the most nested CondContext containing
`ctxt`, or None if `ctxt` is not in a cond."
8337,IsContainingContext,tensorflow/tensorflow/python/ops/control_flow_util.py,250,function,Returns true if `maybe_containing_ctxt` is or contains `ctxt`.
8338,OpInContext,tensorflow/tensorflow/python/ops/control_flow_util.py,258,function,
8339,TensorInContext,tensorflow/tensorflow/python/ops/control_flow_util.py,262,function,
8340,CheckInputFromValidContext,tensorflow/tensorflow/python/ops/control_flow_util.py,266,function,"Returns whether `input_op` can be used from `op`s context.
Conceptually, only inputs from op's while context or any ancestor while
context (including outside of any context) are valid. In practice, there are
many other edge cases as well.
Args:
op: Operation
input_op: Operation
Raises:
ValueError: if input_op is from an invalid context."
8341,GetWhileContext,tensorflow/tensorflow/python/ops/control_flow_util.py,366,function,Get the WhileContext to which this op belongs.
8342,in_defun,tensorflow/tensorflow/python/ops/control_flow_util_v2.py,44,function,"Returns if the current graph is, or is nested in, a defun."
8343,in_while_loop_defun,tensorflow/tensorflow/python/ops/control_flow_util_v2.py,56,function,Returns if the graph is a while loop FuncGraph.
8344,create_new_tf_function,tensorflow/tensorflow/python/ops/control_flow_util_v2.py,63,function,"Converts func_graph to a TF_Function and adds it to the current graph.
Args:
func_graph: FuncGraph
Returns:
The name of the new TF_Function."
8345,unique_fn_name,tensorflow/tensorflow/python/ops/control_flow_util_v2.py,78,function,"Returns a unique name to use for a control flow function.
Args:
scope: A name scope string.
name: An identifier for this function (e.g. ""true"", ""body"").
Returns:
A string, the name to use for the function."
8346,unique_grad_fn_name,tensorflow/tensorflow/python/ops/control_flow_util_v2.py,91,function,
8347,maybe_set_lowering_attr,tensorflow/tensorflow/python/ops/control_flow_util_v2.py,95,function,"Sets the flag to enable lowering on `op` if necessary.
Lowering allows cond_v2 and while_v2 to avoid some of the limitations of
Functions, allowing users to specify devices & colocation inside of cond_v2
and while_v2 input functions, and enabling non-strict evaluation & partial
pruning. This brings v2 control flow closer to feature parity with v1 control
flow.
However, we do not lower in the following cases:
- When the `If` or `While` ops are in the XLA context. Because it is easier
for XLA to apply its own optimizations when dealing with un-lowered
control flow operators than with low-level control flow primitives.
- When the eager execution context specifies the executor of functions to
be the single threaded executor (see context.function_executor_type()).
Because the single threaded executor does not support v1 control flow ops.
- When 'lower_using_switch_merge' is explicitly set to False.
Args:
op: An `If` or `While` Operation.
lower_using_switch_merge: Explicit value to lower or not (optional)."
8348,maybe_propagate_compile_time_consts_in_xla,tensorflow/tensorflow/python/ops/control_flow_util_v2.py,131,function,"Tells XLA whether to propagate compile-time consts in the loop body.
This is needed to make compile time constants available to ops, for example
`max_num_elements` in `EmptyTensorList`, inside the loop body. Ideally this
would always be turned on, but that doesn't work with legacy functionalized
while_loops.
Args:
op: A `While` Operation."
8349,resource_input_index,tensorflow/tensorflow/python/ops/control_flow_util_v2.py,149,function,"Returns the index of the input corresponding to `tensor_name`.
This method is used to find the corresponding index of an arbitrary resource
tensor in a function (the function could be a loop body). We assume that
resource handles are never created in functions, so that every resource
tensor can be traced back to a function input.
The awkward signature of this method is to make it work with both FuncGraphs
and FunctionDefs. This is so we can recurse on function call ops without
building the corresponding FuncGraph (note that even if a FuncGraph for a
FunctionDef already exists, the input/output/node names may have been
changed when the FuncGraph was serialized to the FunctionDef, which makes it
unusable with this algorithm).
Args:
tensor_name: the name of the resource tensor to be resolved to an input.
input_names: a list of the names of all inputs to the function.
node_defs: a dict mapping op name -> NodeDef for every op in the function.
functions: a dict mapping function name -> _EagerDefinedFunction.
Returns:
The index into input_names corresponding to `tensor_name`."
8350,clear_control_inputs,tensorflow/tensorflow/python/ops/control_flow_util_v2.py,221,function,"Clears the control inputs but preserves the ControlFlowContext.
This is needed to preserve the XLAControlFlowControl when clearing
control inputs for the gradient accumulators in while_v2.
`ops.control_dependencies` does not allow that.
Yields:
A context manager in which the ops created will not have any control inputs
by default but the control flow context is the same."
8351,output_all_intermediates,tensorflow/tensorflow/python/ops/control_flow_util_v2.py,260,function,"Whether to output all intermediates of a functional control flow op.
The default behavior is to output intermediates only when building a Keras
Layer in graph mode and that too when certain other conditions are met:
1. We do not output intermediates if the functional control flow op
is being built inside a FuncGraph which is not a If/While graph. This
guards against outputting intermediates in eager mode since keras adds
tensors to a FuncGraph named ""keras_graph"" in that case. Also because we
do not output intermediates of tf.function (since this feature is only for
backwards compatibility) outputting intermediates of functional control
flow ops built inside tf.function is of no value.
2. We do not output intermediates when the compilation is using XLA or for a
TPU.
3. We do not output intermediates when a single threaded executor is used
since that does not perform inlining and pruning.
Returns:
A bool telling whether to output all intermediates."
8352,get_func_graph,tensorflow/tensorflow/python/ops/control_flow_util_v2.py,293,function,Generates and returns a FuncGraph for the given op and input_shapes.
8353,CondBranchFuncGraph,tensorflow/tensorflow/python/ops/control_flow_v2_func_graphs.py,25,class,"FuncGraph for branches of tf.cond().
This is used to distinguish cond branches from other functions."
8354,WhileCondFuncGraph,tensorflow/tensorflow/python/ops/control_flow_v2_func_graphs.py,38,class,"FuncGraph for the condition of tf.while_loop().
This is used to distinguish while conditions from other functions."
8355,WhileBodyFuncGraph,tensorflow/tensorflow/python/ops/control_flow_v2_func_graphs.py,51,class,"FuncGraph for the body of tf.while_loop().
This is used to distinguish while bodies from other functions."
8356,enable_control_flow_v2,tensorflow/tensorflow/python/ops/control_flow_v2_toggles.py,29,function,"Use control flow v2.
control flow v2 (cfv2) is an improved version of control flow in TensorFlow
with support for higher order derivatives. Enabling cfv2 will change the
graph/function representation of control flow, e.g., `tf.while_loop` and
`tf.cond` will generate functional `While` and `If` ops instead of low-level
`Switch`, `Merge` etc. ops. Note: Importing and running graphs exported
with old control flow will still be supported.
Calling tf.enable_control_flow_v2() lets you opt-in to this TensorFlow 2.0
feature.
Note: v2 control flow is always enabled inside of tf.function. Calling this
function is not required."
8357,disable_control_flow_v2,tensorflow/tensorflow/python/ops/control_flow_v2_toggles.py,51,function,"Opts out of control flow v2.
Note: v2 control flow is always enabled inside of tf.function. Calling this
function has no effect in that case.
If your code needs tf.disable_control_flow_v2() to be called to work
properly please file a bug."
8358,control_flow_v2_enabled,tensorflow/tensorflow/python/ops/control_flow_v2_toggles.py,66,function,"Returns `True` if v2 control flow is enabled.
Note: v2 control flow is always enabled inside of tf.function."
8359,output_all_intermediates,tensorflow/tensorflow/python/ops/control_flow_v2_toggles.py,75,function,"Whether to output all intermediates from functional control flow ops.
The ""default"" behavior to is to output all intermediates when using v2 control
flow inside Keras models in graph mode (possibly inside Estimators). This is
needed to support taking gradients of v2 control flow. In graph mode, Keras
can sometimes freeze the forward graph before the gradient computation which
does not work for v2 control flow since it requires updating the forward ops
to output the needed intermediates. We work around this by proactively
outputting the needed intermediates when building the forward pass itself.
Ideally any such extra tensors should be pruned out at runtime. However, if
for any reason this doesn't work for you or if you have an inference-only
model you can turn this behavior off using
`tf.compat.v1.experimental.output_all_intermediates(False)`.
If with the default behavior you are still seeing errors of the form
""Connecting to invalid output X of source node Y which has Z outputs"" try
setting `tf.compat.v1.experimental.output_all_intermediates(True)` and
please file an issue at https://github.com/tensorflow/tensorflow/issues.
Args:
state: True, False or None. None restores the default behavior."
8360,build_graph,tensorflow/tensorflow/python/ops/conv2d_benchmark.py,44,function,"builds a graph containing a sequence of conv2d operations.
Args:
device: String, the device to run on.
dtype: Data type for the convolution.
data_format: A string from: ""NHWC"" or ""NCHW"". Data format for input and
output data.
input_shape: Shape of the input tensor.
filter_shape: Shape of the filter tensor.
strides: A list of ints. 1-D of length 4. The stride of sliding
window for each dimension of input.
padding: A string from: ""SAME"", ""VALID"". The type of padding
algorithm to use.
num_iters: number of iterations to run conv2d.
warmup_iters: number of iterations for warmup runs.
Returns:
An array of tensors to run()"
8361,Conv2DBenchmark,tensorflow/tensorflow/python/ops/conv2d_benchmark.py,94,class,Benchmark conv2d!
8362,benchmark_conv2d,tensorflow/tensorflow/python/ops/conv2d_benchmark.py,173,method,
8363,CriticalSection,tensorflow/tensorflow/python/ops/critical_section_ops.py,126,class,"Critical section.
A `CriticalSection` object is a resource in the graph which executes subgraphs
in **serial** order. A common example of a subgraph one may wish to run
exclusively is the one given by the following function:
```python
v = resource_variable_ops.ResourceVariable(0.0, name=""v"")
def count():
value = v.read_value()
with tf.control_dependencies([value]):
with tf.control_dependencies([v.assign_add(1)]):
return tf.identity(value)
```
Here, a snapshot of `v` is captured in `value`; and then `v` is updated.
The snapshot value is returned.
If multiple workers or threads all execute `count` in parallel, there is no
guarantee that access to the variable `v` is atomic at any point within
any thread's calculation of `count`. In fact, even implementing an atomic
counter that guarantees that the user will see each value `0, 1, ...,` is
currently impossible.
The solution is to ensure any access to the underlying resource `v` is
only processed through a critical section:
```python
cs = CriticalSection()
f1 = cs.execute(count)
f2 = cs.execute(count)
output = f1 + f2
session.run(output)
```
The functions `f1` and `f2` will be executed serially, and updates to `v`
will be atomic.
**NOTES**
All resource objects, including the critical section and any captured
variables of functions executed on that critical section, will be
colocated to the same device (host and cpu/gpu).
When using multiple critical sections on the same resources, there is no
guarantee of exclusive access to those resources. This behavior is disallowed
by default (but see the kwarg `exclusive_resource_access`).
For example, running the same function in two separate critical sections
will not ensure serial execution:
```python
v = tf.compat.v1.get_variable(""v"", initializer=0.0, use_resource=True)
def accumulate(up):
x = v.read_value()
with tf.control_dependencies([x]):
with tf.control_dependencies([v.assign_add(up)]):
return tf.identity(x)
ex1 = CriticalSection().execute(
accumulate, 1.0, exclusive_resource_access=False)
ex2 = CriticalSection().execute(
accumulate, 1.0, exclusive_resource_access=False)
bad_sum = ex1 + ex2
sess.run(v.initializer)
sess.run(bad_sum) # May return 0.0
```"
8364,name,tensorflow/tensorflow/python/ops/critical_section_ops.py,231,method,
8365,execute,tensorflow/tensorflow/python/ops/critical_section_ops.py,234,method,"Execute function `fn()` inside the critical section.
`fn` should not accept any arguments. To add extra arguments to when
calling `fn` in the critical section, create a lambda:
```python
critical_section.execute(lambda: fn(*my_args, **my_kwargs))
```
Args:
fn: The function to execute. Must return at least one tensor.
exclusive_resource_access: Whether the resources required by
`fn` should be exclusive to this `CriticalSection`. Default: `True`.
You may want to set this to `False` if you will be accessing a
resource in read-only mode in two different CriticalSections.
name: The name to use when creating the execute operation.
Returns:
The tensors returned from `fn()`.
Raises:
ValueError: If `fn` attempts to lock this `CriticalSection` in any nested
or lazy way that may cause a deadlock.
ValueError: If `exclusive_resource_access == True` and
another `CriticalSection` has an execution requesting the same
resources as `fn``. Note, even if `exclusive_resource_access` is
`True`, if another execution in another `CriticalSection` was created
without `exclusive_resource_access=True`, a `ValueError` will be raised."
8366,ctc_loss,tensorflow/tensorflow/python/ops/ctc_ops.py,75,function,"Computes the CTC (Connectionist Temporal Classification) Loss.
This op implements the CTC loss as presented in (Graves et al., 2006).
Input requirements:
```
sequence_length(b) <= time for all b
max(labels.indices(labels.indices[:, 1] == b, 2))
<= sequence_length(b) for all b.
```
Notes:
This class performs the softmax operation for you, so inputs should
be e.g. linear projections of outputs by an LSTM.
The `inputs` Tensor's innermost dimension size, `num_classes`, represents
`num_labels + 1` classes, where num_labels is the number of true labels, and
the largest value `(num_classes - 1)` is reserved for the blank label.
For example, for a vocabulary containing 3 labels `[a, b, c]`,
`num_classes = 4` and the labels indexing is `{a: 0, b: 1, c: 2, blank: 3}`.
Regarding the arguments `preprocess_collapse_repeated` and
`ctc_merge_repeated`:
If `preprocess_collapse_repeated` is True, then a preprocessing step runs
before loss calculation, wherein repeated labels passed to the loss
are merged into single labels. This is useful if the training labels come
from, e.g., forced alignments and therefore have unnecessary repetitions.
If `ctc_merge_repeated` is set False, then deep within the CTC calculation,
repeated non-blank labels will not be merged and are interpreted
as individual labels. This is a simplified (non-standard) version of CTC.
Here is a table of the (roughly) expected first order behavior:
* `preprocess_collapse_repeated=False`, `ctc_merge_repeated=True`
Classical CTC behavior: Outputs true repeated classes with blanks in
between, and can also output repeated classes with no blanks in
between that need to be collapsed by the decoder.
* `preprocess_collapse_repeated=True`, `ctc_merge_repeated=False`
Never learns to output repeated classes, as they are collapsed
in the input labels before training.
* `preprocess_collapse_repeated=False`, `ctc_merge_repeated=False`
Outputs repeated classes with blanks in between, but generally does not
require the decoder to collapse/merge repeated classes.
* `preprocess_collapse_repeated=True`, `ctc_merge_repeated=True`
Untested. Very likely will not learn to output repeated classes.
The `ignore_longer_outputs_than_inputs` option allows to specify the behavior
of the CTCLoss when dealing with sequences that have longer outputs than
inputs. If true, the CTCLoss will simply return zero gradient for those
items, otherwise an InvalidArgument error is returned, stopping training.
Args:
labels: An `int32` `SparseTensor`.
`labels.indices[i, :] == [b, t]` means `labels.values[i]` stores the id
for (batch b, time t). `labels.values[i]` must take on values in `[0,
num_labels)`. See `core/ops/ctc_ops.cc` for more details.
inputs: 3-D `float` `Tensor`.
If time_major == False, this will be a `Tensor` shaped: `[batch_size,
max_time, num_classes]`.
If time_major == True (default), this will be a `Tensor` shaped:
`[max_time, batch_size, num_classes]`. The logits.
sequence_length: 1-D `int32` vector, size `[batch_size]`. The sequence
lengths.
preprocess_collapse_repeated: Boolean. Default: False. If True, repeated
labels are collapsed prior to the CTC calculation.
ctc_merge_repeated: Boolean. Default: True.
ignore_longer_outputs_than_inputs: Boolean. Default: False. If True,
sequences with longer outputs than inputs will be ignored.
time_major: The shape format of the `inputs` Tensors. If True, these
`Tensors` must be shaped `[max_time, batch_size, num_classes]`. If False,
these `Tensors` must be shaped `[batch_size, max_time, num_classes]`.
Using `time_major = True` (default) is a bit more efficient because it
avoids transposes at the beginning of the ctc_loss calculation. However,
most TensorFlow data is batch-major, so by this function also accepts
inputs in batch-major form.
logits: Alias for inputs.
Returns:
A 1-D `float` `Tensor`, size `[batch]`, containing the negative log
probabilities.
Raises:
TypeError: if labels is not a `SparseTensor`.
References:
Connectionist Temporal Classification - Labeling Unsegmented Sequence Data
with Recurrent Neural Networks:
[Graves et al., 2006](https://dl.acm.org/citation.cfm?id=1143891)
([pdf](http://www.cs.toronto.edu/~graves/icml_2006.pdf))"
8367,ctc_greedy_decoder,tensorflow/tensorflow/python/ops/ctc_ops.py,290,function,"Performs greedy decoding on the logits given in input (best path).
Note: Regardless of the value of merge_repeated, if the maximum index of a
given time and batch corresponds to the blank index `(num_classes - 1)`, no
new element is emitted.
If `merge_repeated` is `True`, merge repeated classes in output.
This means that if consecutive logits' maximum indices are the same,
only the first of these is emitted. The sequence `A B B * B * B` (where '*'
is the blank label) becomes
* `A B B B` if `merge_repeated=True`.
* `A B B B B` if `merge_repeated=False`.
Args:
inputs: 3-D `float` `Tensor` sized `[max_time, batch_size, num_classes]`.
The logits.
sequence_length: 1-D `int32` vector containing sequence lengths, having size
`[batch_size]`.
merge_repeated: Boolean. Default: True.
Returns:
A tuple `(decoded, neg_sum_logits)` where
decoded: A single-element list. `decoded[0]`
is an `SparseTensor` containing the decoded outputs s.t.:
`decoded.indices`: Indices matrix `(total_decoded_outputs, 2)`.
The rows store: `[batch, time]`.
`decoded.values`: Values vector, size `(total_decoded_outputs)`.
The vector stores the decoded classes.
`decoded.dense_shape`: Shape vector, size `(2)`.
The shape values are: `[batch_size, max_decoded_length]`
neg_sum_logits: A `float` matrix `(batch_size x 1)` containing, for the
sequence found, the negative of the sum of the greatest logit at each
timeframe."
8368,ctc_beam_search_decoder,tensorflow/tensorflow/python/ops/ctc_ops.py,340,function,"Performs beam search decoding on the logits given in input.
**Note** The `ctc_greedy_decoder` is a special case of the
`ctc_beam_search_decoder` with `top_paths=1` and `beam_width=1` (but
that decoder is faster for this special case).
If `merge_repeated` is `True`, merge repeated classes in the output beams.
This means that if consecutive entries in a beam are the same,
only the first of these is emitted. That is, when the sequence is
`A B B * B * B` (where '*' is the blank label), the return value is:
* `A B` if `merge_repeated = True`.
* `A B B B` if `merge_repeated = False`.
Args:
inputs: 3-D `float` `Tensor`, size `[max_time x batch_size x num_classes]`.
The logits.
sequence_length: 1-D `int32` vector containing sequence lengths, having size
`[batch_size]`.
beam_width: An int scalar >= 0 (beam search beam width).
top_paths: An int scalar >= 0, <= beam_width (controls output size).
merge_repeated: Boolean. Default: True.
Returns:
A tuple `(decoded, log_probabilities)` where
decoded: A list of length top_paths, where `decoded[j]`
is a `SparseTensor` containing the decoded outputs:
`decoded[j].indices`: Indices matrix `(total_decoded_outputs[j] x 2)`
The rows store: [batch, time].
`decoded[j].values`: Values vector, size `(total_decoded_outputs[j])`.
The vector stores the decoded classes for beam j.
`decoded[j].dense_shape`: Shape vector, size `(2)`.
The shape values are: `[batch_size, max_decoded_length[j]]`.
log_probability: A `float` matrix `(batch_size x top_paths)` containing
sequence log-probabilities."
8369,ctc_beam_search_decoder_v2,tensorflow/tensorflow/python/ops/ctc_ops.py,403,function,"Performs beam search decoding on the logits given in input.
**Note** The `ctc_greedy_decoder` is a special case of the
`ctc_beam_search_decoder` with `top_paths=1` and `beam_width=1` (but
that decoder is faster for this special case).
Args:
inputs: 3-D `float` `Tensor`, size `[max_time, batch_size, num_classes]`.
The logits.
sequence_length: 1-D `int32` vector containing sequence lengths, having size
`[batch_size]`.
beam_width: An int scalar >= 0 (beam search beam width).
top_paths: An int scalar >= 0, <= beam_width (controls output size).
Returns:
A tuple `(decoded, log_probabilities)` where
decoded: A list of length top_paths, where `decoded[j]`
is a `SparseTensor` containing the decoded outputs:
`decoded[j].indices`: Indices matrix `[total_decoded_outputs[j], 2]`;
The rows store: `[batch, time]`.
`decoded[j].values`: Values vector, size `[total_decoded_outputs[j]]`.
The vector stores the decoded classes for beam `j`.
`decoded[j].dense_shape`: Shape vector, size `(2)`.
The shape values are: `[batch_size, max_decoded_length[j]]`.
log_probability: A `float` matrix `[batch_size, top_paths]` containing
sequence log-probabilities."
8370,ctc_state_log_probs,tensorflow/tensorflow/python/ops/ctc_ops.py,510,function,"Computes CTC alignment initial and final state log probabilities.
Create the initial/final state values directly as log values to avoid
having to take a float64 log on tpu (which does not exist).
Args:
seq_lengths: int tensor of shape [batch_size], seq lengths in the batch.
max_seq_length: int, max sequence length possible.
Returns:
initial_state_log_probs, final_state_log_probs"
8371,ctc_loss_and_grad,tensorflow/tensorflow/python/ops/ctc_ops.py,629,function,"Computes the CTC loss and gradients.
Most users will want fwd_bwd.ctc_loss
This function returns the computed gradient, it does not have a gradient
of its own defined.
Args:
logits: tensor of shape [frames, batch_size, num_labels]
labels: tensor of shape [batch_size, max_label_seq_length]
label_length: tensor of shape [batch_size] Length of reference label
sequence in labels.
logit_length: tensor of shape [batch_size] Length of input sequence in
logits.
unique: (optional) unique label indices as computed by unique(labels) If
supplied, enables an implementation that is faster and more memory
efficient on TPU.
Returns:
loss: tensor of shape [batch_size]
gradient: tensor of shape [frames, batch_size, num_labels]"
8372,ctc_loss_v2,tensorflow/tensorflow/python/ops/ctc_ops.py,740,function,"Computes CTC (Connectionist Temporal Classification) loss.
This op implements the CTC loss as presented in (Graves et al., 2006).
Notes:
- Same as the ""Classic CTC"" in TensorFlow 1.x's tf.compat.v1.nn.ctc_loss
setting of preprocess_collapse_repeated=False, ctc_merge_repeated=True
- Labels may be supplied as either a dense, zero-padded tensor with a
vector of label sequence lengths OR as a SparseTensor.
- On TPU and GPU: Only dense padded labels are supported.
- On CPU: Caller may use SparseTensor or dense padded labels but calling with
a SparseTensor will be significantly faster.
- Default blank label is 0 rather num_classes - 1, unless overridden by
blank_index.
Args:
labels: tensor of shape [batch_size, max_label_seq_length] or SparseTensor
logits: tensor of shape [frames, batch_size, num_labels], if
logits_time_major == False, shape is [batch_size, frames, num_labels].
label_length: tensor of shape [batch_size], None if labels is SparseTensor
Length of reference label sequence in labels.
logit_length: tensor of shape [batch_size] Length of input sequence in
logits.
logits_time_major: (optional) If True (default), logits is shaped [time,
batch, logits]. If False, shape is [batch, time, logits]
unique: (optional) Unique label indices as computed by
ctc_unique_labels(labels). If supplied, enable a faster, memory efficient
implementation on TPU.
blank_index: (optional) Set the class index to use for the blank label.
Negative values will start from num_classes, ie, -1 will reproduce the
ctc_loss behavior of using num_classes - 1 for the blank symbol. There is
some memory/performance overhead to switching from the default of 0 as an
additional shifted copy of the logits may be created.
name: A name for this `Op`. Defaults to ""ctc_loss_dense"".
Returns:
loss: tensor of shape [batch_size], negative log probabilities.
References:
Connectionist Temporal Classification - Labeling Unsegmented Sequence Data
with Recurrent Neural Networks:
[Graves et al., 2006](https://dl.acm.org/citation.cfm?id=1143891)
([pdf](http://www.cs.toronto.edu/~graves/icml_2006.pdf))"
8373,ctc_loss_v3,tensorflow/tensorflow/python/ops/ctc_ops.py,835,function,"Computes CTC (Connectionist Temporal Classification) loss.
This op implements the CTC loss as presented in (Graves et al., 2016).
Notes:
- Same as the ""Classic CTC"" in TensorFlow 1.x's tf.compat.v1.nn.ctc_loss
setting of preprocess_collapse_repeated=False, ctc_merge_repeated=True
- Labels may be supplied as either a dense, zero-padded tensor with a
vector of label sequence lengths OR as a SparseTensor.
- On TPU and GPU: Only dense padded labels are supported.
- On CPU: Caller may use SparseTensor or dense padded labels but calling with
a SparseTensor will be significantly faster.
- Default blank label is 0 rather num_classes - 1, unless overridden by
blank_index.
Args:
labels: tensor of shape [batch_size, max_label_seq_length] or SparseTensor
logits: tensor of shape [frames, batch_size, num_labels], if
logits_time_major == False, shape is [batch_size, frames, num_labels].
label_length: tensor of shape [batch_size], None if labels is SparseTensor
Length of reference label sequence in labels.
logit_length: tensor of shape [batch_size] Length of input sequence in
logits.
logits_time_major: (optional) If True (default), logits is shaped [time,
batch, logits]. If False, shape is [batch, time, logits]
unique: (optional) Unique label indices as computed by
ctc_unique_labels(labels). If supplied, enable a faster, memory efficient
implementation on TPU.
blank_index: (optional) Set the class index to use for the blank label.
Negative values will start from num_classes, ie, -1 will reproduce the
ctc_loss behavior of using num_classes - 1 for the blank symbol. There is
some memory/performance overhead to switching from the default of 0 as an
additional shifted copy of the logits may be created.
name: A name for this `Op`. Defaults to ""ctc_loss_dense"".
Returns:
loss: tensor of shape [batch_size], negative log probabilities.
References:
Connectionist Temporal Classification - Labeling Unsegmented Sequence Data
with Recurrent Neural Networks:
[Graves et al., 2016](https://dl.acm.org/citation.cfm?id=1143891)
([pdf](http://www.cs.toronto.edu/~graves/icml_2006.pdf))"
8374,ctc_loss_dense,tensorflow/tensorflow/python/ops/ctc_ops.py,939,function,"Computes CTC (Connectionist Temporal Classification) loss.
This op implements the CTC loss as presented in (Graves et al., 2006),
using the batched forward backward algorithm described in (Sim et al., 2017).
Notes:
Significant differences from tf.compat.v1.nn.ctc_loss:
Supports GPU and TPU (tf.compat.v1.nn.ctc_loss supports CPU only):
For batched operations, GPU and TPU are significantly faster than using
ctc_loss on CPU.
This implementation runs on CPU, but significantly slower than ctc_loss.
Blank label is 0 rather num_classes - 1, unless overridden by blank_index.
Logits and labels are dense arrays with padding rather than SparseTensor.
The only mode supported is the same as:
preprocess_collapse_repeated=False, ctc_merge_repeated=True
To collapse labels, the caller can preprocess label sequence first.
The dense implementation supports both CPU, GPU and TPU. A fast path is
provided that significantly improves memory use for large vocabulary if the
caller preprocesses label sequences to get unique label indices on the CPU
(eg. in the data input pipeline) using ctc_ops.unique and simplifies this in
the optional ""unique"" kwarg. This is especially useful for TPU and GPU but
also works with if used on CPU.
Args:
labels: tensor of shape [batch_size, max_label_seq_length]
logits: tensor of shape [frames, batch_size, num_labels], if
logits_time_major == False, shape is [batch_size, frames, num_labels].
label_length: tensor of shape [batch_size] Length of reference label
sequence in labels.
logit_length: tensor of shape [batch_size] Length of input sequence in
logits.
logits_time_major: (optional) If True (default), logits is shaped [time,
batch, logits]. If False, shape is [batch, time, logits]
unique: (optional) Unique label indices as computed by unique(labels). If
supplied, enable a faster, memory efficient implementation on TPU.
blank_index: (optional) Set the class index to use for the blank label.
Negative values will start from num_classes, ie, -1 will reproduce the
ctc_loss behavior of using num_classes - 1 for the blank symbol. There is
some memory/performance overhead to switching from the default of 0 as an
additional shifted copy of the logits may be created.
name: A name for this `Op`. Defaults to ""ctc_loss_dense"".
Returns:
loss: tensor of shape [batch_size], negative log probabilities.
References:
Connectionist Temporal Classification - Labeling Unsegmented Sequence Data
with Recurrent Neural Networks:
[Graves et al., 2006](https://dl.acm.org/citation.cfm?id=1143891)
([pdf](http://www.cs.toronto.edu/~graves/icml_2006.pdf))
Improving the efficiency of forward-backward algorithm using batched
computation in TensorFlow:
[Sim et al., 2017](https://ieeexplore.ieee.org/document/8268944)
([pdf](http://bacchiani.net/resume/papers/ASRU2017.pdf))"
8375,collapse_repeated,tensorflow/tensorflow/python/ops/ctc_ops.py,1067,function,"Merge repeated labels into single labels.
Args:
labels: Tensor of shape [batch, max value in seq_length]
seq_length: Tensor of shape [batch], sequence length of each batch element.
name: A name for this `Op`. Defaults to ""collapse_repeated_labels"".
Returns:
A tuple `(collapsed_labels, new_seq_length)` where
collapsed_labels: Tensor of shape [batch, max_seq_length] with repeated
labels collapsed and padded to max_seq_length, eg:
`[[A, A, B, B, A], [A, B, C, D, E]] => [[A, B, A, 0, 0], [A, B, C, D, E]]`
new_seq_length: int tensor of shape [batch] with new sequence lengths."
8376,dense_labels_to_sparse,tensorflow/tensorflow/python/ops/ctc_ops.py,1129,function,"Convert dense labels with sequence lengths to sparse tensor.
Args:
dense: tensor of shape [batch, max_length]
length: int tensor of shape [batch] The length of each sequence in dense.
Returns:
tf.sparse.SparseTensor with values only for the valid elements of sequences."
8377,ctc_unique_labels,tensorflow/tensorflow/python/ops/ctc_ops.py,1165,function,"Get unique labels and indices for batched labels for `tf.nn.ctc_loss`.
For use with `tf.nn.ctc_loss` optional argument `unique`: This op can be
used to preprocess labels in input pipeline to for better speed/memory use
computing the ctc loss on TPU.
Example:
ctc_unique_labels([[3, 4, 4, 3]]) ->
unique labels padded with 0: [[3, 4, 0, 0]]
indices of original labels in unique: [0, 1, 1, 0]
Args:
labels: tensor of shape [batch_size, max_label_length] padded with 0.
name: A name for this `Op`. Defaults to ""ctc_unique_labels"".
Returns:
tuple of
- unique labels, tensor of shape `[batch_size, max_label_length]`
- indices into unique labels, shape `[batch_size, max_label_length]`"
8378,copy_handle_data,tensorflow/tensorflow/python/ops/custom_gradient.py,45,function,"Copies HandleData for variant and resource type tensors if available.
The CppShapeInferenceResult::HandleData proto contains information about the
shapes and types of the element tensors of resource/variant type tensors.
We need to copy this across function boundaries, i.e., when capturing a
placeholder or when returning a function tensor as output. If we don't do this
the element tensors will have unknown shapes, e.g., if a TensorList variant
tensor is captured as a placeholder, elements popped from that list would have
unknown shape.
Args:
source_t: The tensor to copy HandleData from.
target_t: The tensor to copy HandleData to."
8379,custom_gradient,tensorflow/tensorflow/python/ops/custom_gradient.py,89,function,"Decorator to define a function with a custom gradient.
This decorator allows fine grained control over the gradients of a sequence
for operations. This may be useful for multiple reasons, including providing
a more efficient or numerically stable gradient for a sequence of operations.
For example, consider the following function that commonly occurs in the
computation of cross entropy and log likelihoods:
```python
def log1pexp(x):
return tf.math.log(1 + tf.exp(x))
```
Due to numerical instability, the gradient of this function evaluated at x=100
is NaN. For example:
```python
x = tf.constant(100.)
y = log1pexp(x)
dy = tf.gradients(y, x) # Will be NaN when evaluated.
```
The gradient expression can be analytically simplified to provide numerical
stability:
```python
@tf.custom_gradient
def log1pexp(x):
e = tf.exp(x)
def grad(dy):
return dy * (1 - 1 / (1 + e))
return tf.math.log(1 + e), grad
```
With this definition, the gradient at x=100 will be correctly evaluated as
1.0.
Nesting custom gradients can lead to unintuitive results. The default
behavior does not correspond to n-th order derivatives. For example
```python
@tf.custom_gradient
def op(x):
y = op1(x)
@tf.custom_gradient
def grad_fn(dy):
gdy = op2(x, y, dy)
def grad_grad_fn(ddy): # Not the 2nd order gradient of op w.r.t. x.
return op3(x, y, dy, ddy)
return gdy, grad_grad_fn
return y, grad_fn
```
The function `grad_grad_fn` will be calculating the first order gradient
of `grad_fn` with respect to `dy`, which is used to generate forward-mode
gradient graphs from backward-mode gradient graphs, but is not the same as
the second order gradient of `op` with respect to `x`.
Instead, wrap nested `@tf.custom_gradients` in another function:
```python
@tf.custom_gradient
def op_with_fused_backprop(x):
y, x_grad = fused_op(x)
def first_order_gradient(dy):
@tf.custom_gradient
def first_order_custom(unused_x):
def second_order_and_transpose(ddy):
return second_order_for_x(...), gradient_wrt_dy(...)
return x_grad, second_order_and_transpose
return dy * first_order_custom(x)
return y, first_order_gradient
```
Additional arguments to the inner `@tf.custom_gradient`-decorated function
control the expected return values of the innermost function.
See also `tf.RegisterGradient` which registers a gradient function for a
primitive TensorFlow operation. `tf.custom_gradient` on the other hand allows
for fine grained control over the gradient computation of a sequence of
operations.
Note that if the decorated function uses `Variable`s, the enclosing variable
scope must be using `ResourceVariable`s.
Args:
f: function `f(*x)` that returns a tuple `(y, grad_fn)` where:
- `x` is a sequence of (nested structures of) `Tensor` inputs to the
function.
- `y` is a (nested structure of) `Tensor` outputs of applying TensorFlow
operations in `f` to `x`.
- `grad_fn` is a function with the signature `g(*grad_ys)` which returns
a list of `Tensor`s the same size as (flattened) `x` - the derivatives
of `Tensor`s in `y` with respect to the `Tensor`s in `x`. `grad_ys` is
a sequence of `Tensor`s the same size as (flattened) `y` holding the
initial value gradients for each `Tensor` in `y`.
In a pure mathematical sense, a vector-argument vector-valued function
`f`'s derivatives should be its Jacobian matrix `J`. Here we are
expressing the Jacobian `J` as a function `grad_fn` which defines how
`J` will transform a vector `grad_ys` when left-multiplied with it
(`grad_ys * J`, the vector-Jacobian product, or VJP). This functional
representation of a matrix is convenient to use for chain-rule
calculation (in e.g. the back-propagation algorithm).
If `f` uses `Variable`s (that are not part of the
inputs), i.e. through `get_variable`, then `grad_fn` should have
signature `g(*grad_ys, variables=None)`, where `variables` is a list of
the `Variable`s, and return a 2-tuple `(grad_xs, grad_vars)`, where
`grad_xs` is the same as above, and `grad_vars` is a `list<Tensor>`
with the derivatives of `Tensor`s in `y` with respect to the variables
(that is, grad_vars has one Tensor per variable in variables).
Returns:
A function `h(x)` which returns the same value as `f(x)[0]` and whose
gradient (as calculated by `tf.gradients`) is determined by `f(x)[1]`."
8380,Bind,tensorflow/tensorflow/python/ops/custom_gradient.py,225,class,"When called evaluates `d(f, args, kwargs)` but supports binding `f`.
>>> @Bind.decorator
... def my_decorator(f, args, kwargs):
... print(""my_decorator called with"", args, kwargs)
... return f(*args, **kwargs)
>>> class Foo(object):
... @my_decorator
... def bar(self, a, b, c):
... return a * b * c
>>> Foo.bar(None, 1, 2, c=3)
my_decorator called with (None, 1, 2) {'c': 3}
6
>>> foo = Foo()
>>> foo.bar(1, 2, c=3)
my_decorator called with (1, 2) {'c': 3}
6"
8381,decorator,tensorflow/tensorflow/python/ops/custom_gradient.py,249,method,
8382,get_variable_by_name,tensorflow/tensorflow/python/ops/custom_gradient.py,267,function,"Given a variable name, retrieves a handle on the tensorflow Variable."
8383,recompute_grad,tensorflow/tensorflow/python/ops/custom_gradient.py,492,function,"An eager-compatible version of recompute_grad.
For f(*args, **kwargs), this supports gradients with respect to args or
kwargs, but kwargs are currently only supported in eager-mode.
Note that for keras layer and model objects, this is handled automatically.
Warning: If `f` was originally a tf.keras Model or Layer object, `g` will not
be able to access the member variables of that object, because `g` returns
through the wrapper function `inner`. When recomputing gradients through
objects that inherit from keras, we suggest keeping a reference to the
underlying object around for the purpose of accessing these variables.
Args:
f: function `f(*x)` that returns a `Tensor` or sequence of `Tensor` outputs.
Returns:
A function `g` that wraps `f`, but which recomputes `f` on the backwards
pass of a gradient call."
8384,grad_pass_through,tensorflow/tensorflow/python/ops/custom_gradient.py,565,function,"Creates a grad-pass-through op with the forward behavior provided in f.
Use this function to wrap any op, maintaining its behavior in the forward
pass, but replacing the original op in the backward graph with an identity.
For example:
```python
x = tf.Variable(1.0, name=""x"")
z = tf.Variable(3.0, name=""z"")
with tf.GradientTape() as tape:
# y will evaluate to 9.0
y = tf.grad_pass_through(x.assign)(z**2)
# grads will evaluate to 6.0
grads = tape.gradient(y, z)
```
Another example is a 'differentiable' moving average approximation, where
gradients are allowed to flow into the last value fed to the moving average,
but the moving average is still used for the forward pass:
```python
x = ... # Some scalar value
# A moving average object, we don't need to know how this is implemented
moving_average = MovingAverage()
with backprop.GradientTape() as tape:
# mavg_x will evaluate to the current running average value
mavg_x = tf.grad_pass_through(moving_average)(x)
grads = tape.gradient(mavg_x, x) # grads will evaluate to 1.0
```
Args:
f: function `f(*x)` that returns a `Tensor` or nested structure of `Tensor`
outputs.
Returns:
A function `h(x)` which returns the same values as `f(x)` and whose
gradients are the same as those of an identity function."
8385,QueueBase,tensorflow/tensorflow/python/ops/data_flow_ops.py,119,class,"Base class for queue implementations.
A queue is a TensorFlow data structure that stores tensors across
multiple steps, and exposes operations that enqueue and dequeue
tensors.
Each queue element is a tuple of one or more tensors, where each
tuple component has a static dtype, and may have a static shape. The
queue implementations support versions of enqueue and dequeue that
handle single elements, versions that support enqueuing and
dequeuing a batch of elements at once.
See `tf.queue.FIFOQueue` and
`tf.queue.RandomShuffleQueue` for concrete
implementations of this class, and instructions on how to create
them."
8386,from_list,tensorflow/tensorflow/python/ops/data_flow_ops.py,185,method,"Create a queue using the queue reference from `queues[index]`.
Args:
index: An integer scalar tensor that determines the input that gets
selected.
queues: A list of `QueueBase` objects.
Returns:
A `QueueBase` object.
Raises:
TypeError: When `queues` is not a list of `QueueBase` objects,
or when the data types of `queues` are not all the same."
8387,queue_ref,tensorflow/tensorflow/python/ops/data_flow_ops.py,226,method,The underlying queue reference.
8388,name,tensorflow/tensorflow/python/ops/data_flow_ops.py,231,method,The name of the underlying queue.
8389,dtypes,tensorflow/tensorflow/python/ops/data_flow_ops.py,238,method,The list of dtypes for each component of a queue element.
8390,shapes,tensorflow/tensorflow/python/ops/data_flow_ops.py,243,method,The list of shapes for each component of a queue element.
8391,names,tensorflow/tensorflow/python/ops/data_flow_ops.py,248,method,The list of names for each component of a queue element.
8392,enqueue,tensorflow/tensorflow/python/ops/data_flow_ops.py,311,method,"Enqueues one element to this queue.
If the queue is full when this operation executes, it will block
until the element has been enqueued.
At runtime, this operation may raise an error if the queue is
`tf.QueueBase.close` before or during its execution. If the
queue is closed before this operation runs,
`tf.errors.CancelledError` will be raised. If this operation is
blocked, and either (i) the queue is closed by a close operation
with `cancel_pending_enqueues=True`, or (ii) the session is
`tf.Session.close`,
`tf.errors.CancelledError` will be raised.
Args:
vals: A tensor, a list or tuple of tensors, or a dictionary containing
the values to enqueue.
name: A name for the operation (optional).
Returns:
The operation that enqueues a new tuple of tensors to the queue."
8393,enqueue_many,tensorflow/tensorflow/python/ops/data_flow_ops.py,350,method,"Enqueues zero or more elements to this queue.
This operation slices each component tensor along the 0th dimension to
make multiple queue elements. All of the tensors in `vals` must have the
same size in the 0th dimension.
If the queue is full when this operation executes, it will block
until all of the elements have been enqueued.
At runtime, this operation may raise an error if the queue is
`tf.QueueBase.close` before or during its execution. If the
queue is closed before this operation runs,
`tf.errors.CancelledError` will be raised. If this operation is
blocked, and either (i) the queue is closed by a close operation
with `cancel_pending_enqueues=True`, or (ii) the session is
`tf.Session.close`,
`tf.errors.CancelledError` will be raised.
Args:
vals: A tensor, a list or tuple of tensors, or a dictionary
from which the queue elements are taken.
name: A name for the operation (optional).
Returns:
The operation that enqueues a batch of tuples of tensors to the queue."
8394,dequeue,tensorflow/tensorflow/python/ops/data_flow_ops.py,421,method,"Dequeues one element from this queue.
If the queue is empty when this operation executes, it will block
until there is an element to dequeue.
At runtime, this operation may raise an error if the queue is
`tf.QueueBase.close` before or during its execution. If the
queue is closed, the queue is empty, and there are no pending
enqueue operations that can fulfill this request,
`tf.errors.OutOfRangeError` will be raised. If the session is
`tf.Session.close`,
`tf.errors.CancelledError` will be raised.
Args:
name: A name for the operation (optional).
Returns:
The tuple of tensors that was dequeued."
8395,dequeue_many,tensorflow/tensorflow/python/ops/data_flow_ops.py,459,method,"Dequeues and concatenates `n` elements from this queue.
This operation concatenates queue-element component tensors along
the 0th dimension to make a single component tensor. All of the
components in the dequeued tuple will have size `n` in the 0th dimension.
If the queue is closed and there are less than `n` elements left, then an
`OutOfRange` exception is raised.
At runtime, this operation may raise an error if the queue is
`tf.QueueBase.close` before or during its execution. If the
queue is closed, the queue contains fewer than `n` elements, and
there are no pending enqueue operations that can fulfill this
request, `tf.errors.OutOfRangeError` will be raised. If the
session is `tf.Session.close`,
`tf.errors.CancelledError` will be raised.
Args:
n: A scalar `Tensor` containing the number of elements to dequeue.
name: A name for the operation (optional).
Returns:
The list of concatenated tensors that was dequeued."
8396,dequeue_up_to,tensorflow/tensorflow/python/ops/data_flow_ops.py,502,method,"Dequeues and concatenates `n` elements from this queue.
**Note** This operation is not supported by all queues. If a queue does not
support DequeueUpTo, then a `tf.errors.UnimplementedError` is raised.
This operation concatenates queue-element component tensors along
the 0th dimension to make a single component tensor. If the queue
has not been closed, all of the components in the dequeued tuple
will have size `n` in the 0th dimension.
If the queue is closed and there are more than `0` but fewer than
`n` elements remaining, then instead of raising a
`tf.errors.OutOfRangeError` like `tf.QueueBase.dequeue_many`,
less than `n` elements are returned immediately. If the queue is
closed and there are `0` elements left in the queue, then a
`tf.errors.OutOfRangeError` is raised just like in `dequeue_many`.
Otherwise the behavior is identical to `dequeue_many`.
Args:
n: A scalar `Tensor` containing the number of elements to dequeue.
name: A name for the operation (optional).
Returns:
The tuple of concatenated tensors that was dequeued."
8397,close,tensorflow/tensorflow/python/ops/data_flow_ops.py,543,method,"Closes this queue.
This operation signals that no more elements will be enqueued in
the given queue. Subsequent `enqueue` and `enqueue_many`
operations will fail. Subsequent `dequeue` and `dequeue_many`
operations will continue to succeed if sufficient elements remain
in the queue. Subsequently dequeue and dequeue_many operations
that would otherwise block waiting for more elements (if close
hadn't been called) will now fail immediately.
If `cancel_pending_enqueues` is `True`, all pending requests will also
be canceled.
Args:
cancel_pending_enqueues: (Optional.) A boolean, defaulting to
`False` (described above).
name: A name for the operation (optional).
Returns:
The operation that closes the queue."
8398,is_closed,tensorflow/tensorflow/python/ops/data_flow_ops.py,578,method,"Returns true if queue is closed.
This operation returns true if the queue is closed and false if the queue
is open.
Args:
name: A name for the operation (optional).
Returns:
True if the queue is closed and false if the queue is open."
8399,size,tensorflow/tensorflow/python/ops/data_flow_ops.py,597,method,"Compute the number of elements in this queue.
Args:
name: A name for the operation (optional).
Returns:
A scalar tensor containing the number of elements in this queue."
8400,RandomShuffleQueue,tensorflow/tensorflow/python/ops/data_flow_ops.py,625,class,"A queue implementation that dequeues elements in a random order.
See `tf.queue.QueueBase` for a description of the methods on
this class."
8401,FIFOQueue,tensorflow/tensorflow/python/ops/data_flow_ops.py,711,class,"A queue implementation that dequeues elements in first-in first-out order.
See `tf.queue.QueueBase` for a description of the methods on
this class."
8402,GPUCompatibleFIFOQueue,tensorflow/tensorflow/python/ops/data_flow_ops.py,771,class,"A queue implementation that dequeues elements in first-in first-out order.
GPUCompatibleFIFOQueue is like FIFOQueue, but the queue resource may be placed
either on a CPU or on a GPU. It is not cross-device: enqueues and dequeues
will be colocated with the queue resource. GPUCompatibleFIFOQueue only
supports enqueue and dequeue at the moment, not enqueue_many or dequeue_many.
See `tf.queue.QueueBase` for a description of the methods on this class."
8403,enqueue_many,tensorflow/tensorflow/python/ops/data_flow_ops.py,832,method,enqueue_many is not supported on GPUCompatibleFIFOQueue.
8404,dequeue_many,tensorflow/tensorflow/python/ops/data_flow_ops.py,838,method,dequeue_many is not supported on GPUCompatibleFIFOQueue.
8405,PaddingFIFOQueue,tensorflow/tensorflow/python/ops/data_flow_ops.py,849,class,"A FIFOQueue that supports batching variable-sized tensors by padding.
A `PaddingFIFOQueue` may contain components with dynamic shape, while also
supporting `dequeue_many`. See the constructor for more details.
See `tf.queue.QueueBase` for a description of the methods on
this class."
8406,PriorityQueue,tensorflow/tensorflow/python/ops/data_flow_ops.py,925,class,"A queue implementation that dequeues elements in prioritized order.
See `tf.queue.QueueBase` for a description of the methods on
this class."
8407,Barrier,tensorflow/tensorflow/python/ops/data_flow_ops.py,994,class,Represents a key-value map that persists across graph executions.
8408,barrier_ref,tensorflow/tensorflow/python/ops/data_flow_ops.py,1078,method,Get the underlying barrier reference.
8409,name,tensorflow/tensorflow/python/ops/data_flow_ops.py,1083,method,The name of the underlying barrier.
8410,insert_many,tensorflow/tensorflow/python/ops/data_flow_ops.py,1089,method,"For each key, assigns the respective value to the specified component.
This operation updates each element at component_index.
Args:
component_index: The component of the value that is being assigned.
keys: A vector of keys, with length n.
values: An any-dimensional tensor of values, which are associated with the
respective keys. The first dimension must have length n.
name: Optional name for the op.
Returns:
The operation that performs the insertion.
Raises:
InvalidArgumentsError: If inserting keys and values without elements."
8411,take_many,tensorflow/tensorflow/python/ops/data_flow_ops.py,1111,method,"Takes the given number of completed elements from this barrier.
This operation concatenates completed-element component tensors along
the 0th dimension to make a single component tensor.
If barrier has no completed elements, this operation will block
until there are 'num_elements' elements to take.
TODO(b/25743580): the semantics of `allow_small_batch` are experimental
and may be extended to other cases in the future.
TODO(ebrevdo): If a take_many(allow_small_batch=True) is blocking
already when the barrier is closed, it will block for ever. Fix this
by using asynchronous operations.
Args:
num_elements: The number of elements to take.
allow_small_batch: If the barrier is closed, don't block if there are less
completed elements than requested, but instead return all available
completed elements.
timeout: This specifies the number of milliseconds to block
before returning with DEADLINE_EXCEEDED. (This option is not
supported yet.)
name: A name for the operation (optional).
Returns:
A tuple of (index, key, value_list).
""index"" is a int64 tensor of length num_elements containing the
index of the insert_many call for which the very first component of
the given element was inserted into the Barrier, starting with
the value -2**63. Note, this value is different from the
index of the insert_many call for which the element was completed.
""key"" is a string tensor of length num_elements containing the keys.
""value_list"" is a tuple of tensors, each one with size num_elements
in the 0th dimension for each component in the barrier's values."
8412,close,tensorflow/tensorflow/python/ops/data_flow_ops.py,1180,method,"Closes this barrier.
This operation signals that no more new key values will be inserted in the
given barrier. Subsequent InsertMany operations with new keys will fail.
InsertMany operations that just complement already existing keys with other
components, will continue to succeed. Subsequent TakeMany operations will
continue to succeed if sufficient elements remain in the barrier. Subsequent
TakeMany operations that would block will fail immediately.
If `cancel_pending_enqueues` is `True`, all pending requests to the
underlying queue will also be canceled, and completing of already
started values is also not acceptable anymore.
Args:
cancel_pending_enqueues: (Optional.) A boolean, defaulting to
`False` (described above).
name: Optional name for the op.
Returns:
The operation that closes the barrier."
8413,ready_size,tensorflow/tensorflow/python/ops/data_flow_ops.py,1209,method,"Compute the number of complete elements in the given barrier.
Args:
name: A name for the operation (optional).
Returns:
A single-element tensor containing the number of complete elements in the
given barrier."
8414,incomplete_size,tensorflow/tensorflow/python/ops/data_flow_ops.py,1223,method,"Compute the number of incomplete elements in the given barrier.
Args:
name: A name for the operation (optional).
Returns:
A single-element tensor containing the number of incomplete elements in
the given barrier."
8415,ConditionalAccumulatorBase,tensorflow/tensorflow/python/ops/data_flow_ops.py,1240,class,"A conditional accumulator for aggregating gradients.
Up-to-date gradients (i.e., time step at which gradient was computed is
equal to the accumulator's time step) are added to the accumulator.
Extraction of the average gradient is blocked until the required number of
gradients has been accumulated."
8416,accumulator_ref,tensorflow/tensorflow/python/ops/data_flow_ops.py,1271,method,The underlying accumulator reference.
8417,name,tensorflow/tensorflow/python/ops/data_flow_ops.py,1276,method,The name of the underlying accumulator.
8418,dtype,tensorflow/tensorflow/python/ops/data_flow_ops.py,1281,method,The datatype of the gradients accumulated by this accumulator.
8419,num_accumulated,tensorflow/tensorflow/python/ops/data_flow_ops.py,1285,method,"Number of gradients that have currently been aggregated in accumulator.
Args:
name: Optional name for the operation.
Returns:
Number of accumulated gradients currently in accumulator."
8420,set_global_step,tensorflow/tensorflow/python/ops/data_flow_ops.py,1300,method,"Sets the global time step of the accumulator.
The operation logs a warning if we attempt to set to a time step that is
lower than the accumulator's own time step.
Args:
new_global_step: Value of new time step. Can be a variable or a constant
name: Optional name for the operation.
Returns:
Operation that sets the accumulator's time step."
8421,ConditionalAccumulator,tensorflow/tensorflow/python/ops/data_flow_ops.py,1320,class,"A conditional accumulator for aggregating gradients.
Up-to-date gradients (i.e., time step at which gradient was computed is
equal to the accumulator's time step) are added to the accumulator.
Extraction of the average gradient is blocked until the required number of
gradients has been accumulated."
8422,apply_grad,tensorflow/tensorflow/python/ops/data_flow_ops.py,1358,method,"Attempts to apply a gradient to the accumulator.
The attempt is silently dropped if the gradient is stale, i.e., local_step
is less than the accumulator's global time step.
Args:
grad: The gradient tensor to be applied.
local_step: Time step at which the gradient was computed.
name: Optional name for the operation.
Returns:
The operation that (conditionally) applies a gradient to the accumulator.
Raises:
ValueError: If grad is of the wrong shape"
8423,take_grad,tensorflow/tensorflow/python/ops/data_flow_ops.py,1382,method,"Attempts to extract the average gradient from the accumulator.
The operation blocks until sufficient number of gradients have been
successfully applied to the accumulator.
Once successful, the following actions are also triggered:
- Counter of accumulated gradients is reset to 0.
- Aggregated gradient is reset to 0 tensor.
- Accumulator's internal time step is incremented by 1.
Args:
num_required: Number of gradients that needs to have been aggregated
name: Optional name for the operation
Returns:
A tensor holding the value of the average gradient.
Raises:
InvalidArgumentError: If num_required < 1"
8424,SparseConditionalAccumulator,tensorflow/tensorflow/python/ops/data_flow_ops.py,1412,class,"A conditional accumulator for aggregating sparse gradients.
Sparse gradients are represented by `IndexedSlices`.
Up-to-date gradients (i.e., time step at which gradient was computed is
equal to the accumulator's time step) are added to the accumulator.
Extraction of the average gradient is blocked until the required number of
gradients has been accumulated.
Args:
dtype: Datatype of the accumulated gradients.
shape: Shape of the accumulated gradients.
shared_name: Optional. If non-empty, this accumulator will be shared under
the given name across multiple sessions.
name: Optional name for the accumulator.
reduction_type: Reduction type to use when taking the gradient."
8425,apply_indexed_slices_grad,tensorflow/tensorflow/python/ops/data_flow_ops.py,1447,method,"Attempts to apply a gradient to the accumulator.
The attempt is silently dropped if the gradient is stale, i.e., `local_step`
is less than the accumulator's global time step.
Args:
grad: The gradient `IndexedSlices` to be applied.
local_step: Time step at which the gradient was computed.
name: Optional name for the operation.
Returns:
The operation that (conditionally) applies a gradient to the accumulator.
Raises:
InvalidArgumentError: If grad is of the wrong shape"
8426,apply_grad,tensorflow/tensorflow/python/ops/data_flow_ops.py,1471,method,"Attempts to apply a sparse gradient to the accumulator.
The attempt is silently dropped if the gradient is stale, i.e., `local_step`
is less than the accumulator's global time step.
A sparse gradient is represented by its indices, values and possibly empty
or None shape. Indices must be a vector representing the locations of
non-zero entries in the tensor. Values are the non-zero slices of the
gradient, and must have the same first dimension as indices, i.e., the nnz
represented by indices and values must be consistent. Shape, if not empty or
None, must be consistent with the accumulator's shape (if also provided).
Example:
A tensor [[0, 0], [0, 1], [2, 3]] can be represented
indices: [1,2]
values: [[0,1],[2,3]]
shape: [3, 2]
Args:
grad_indices: Indices of the sparse gradient to be applied.
grad_values: Values of the sparse gradient to be applied.
grad_shape: Shape of the sparse gradient to be applied.
local_step: Time step at which the gradient was computed.
name: Optional name for the operation.
Returns:
The operation that (conditionally) applies a gradient to the accumulator.
Raises:
InvalidArgumentError: If grad is of the wrong shape"
8427,take_grad,tensorflow/tensorflow/python/ops/data_flow_ops.py,1519,method,"Attempts to extract the average gradient from the accumulator.
The operation blocks until sufficient number of gradients have been
successfully applied to the accumulator.
Once successful, the following actions are also triggered:
- Counter of accumulated gradients is reset to 0.
- Aggregated gradient is reset to 0 tensor.
- Accumulator's internal time step is incremented by 1.
Args:
num_required: Number of gradients that needs to have been aggregated
name: Optional name for the operation
Returns:
A tuple of indices, values, and shape representing the average gradient.
Raises:
InvalidArgumentError: If `num_required` < 1"
8428,take_indexed_slices_grad,tensorflow/tensorflow/python/ops/data_flow_ops.py,1543,method,"Attempts to extract the average gradient from the accumulator.
The operation blocks until sufficient number of gradients have been
successfully applied to the accumulator.
Once successful, the following actions are also triggered:
- Counter of accumulated gradients is reset to 0.
- Aggregated gradient is reset to 0 tensor.
- Accumulator's internal time step is incremented by 1.
Args:
num_required: Number of gradients that needs to have been aggregated
name: Optional name for the operation
Returns:
An `IndexedSlices` holding the value of the average gradient.
Raises:
InvalidArgumentError: If `num_required` < 1"
8429,num_accumulated,tensorflow/tensorflow/python/ops/data_flow_ops.py,1572,method,"Number of gradients that have currently been aggregated in accumulator.
Args:
name: Optional name for the operation.
Returns:
Number of accumulated gradients currently in accumulator."
8430,set_global_step,tensorflow/tensorflow/python/ops/data_flow_ops.py,1587,method,"Sets the global time step of the accumulator.
The operation logs a warning if we attempt to set to a time step that is
lower than the accumulator's own time step.
Args:
new_global_step: Value of new time step. Can be a variable or a constant
name: Optional name for the operation.
Returns:
Operation that sets the accumulator's time step."
8431,BaseStagingArea,tensorflow/tensorflow/python/ops/data_flow_ops.py,1606,class,Base class for Staging Areas.
8432,name,tensorflow/tensorflow/python/ops/data_flow_ops.py,1650,method,The name of the staging area.
8433,dtypes,tensorflow/tensorflow/python/ops/data_flow_ops.py,1655,method,The list of dtypes for each component of a staging area element.
8434,shapes,tensorflow/tensorflow/python/ops/data_flow_ops.py,1660,method,The list of shapes for each component of a staging area element.
8435,names,tensorflow/tensorflow/python/ops/data_flow_ops.py,1665,method,The list of names for each component of a staging area element.
8436,capacity,tensorflow/tensorflow/python/ops/data_flow_ops.py,1670,method,The maximum number of elements of this staging area.
8437,memory_limit,tensorflow/tensorflow/python/ops/data_flow_ops.py,1675,method,The maximum number of bytes of this staging area.
8438,StagingArea,tensorflow/tensorflow/python/ops/data_flow_ops.py,1819,class,"Class for staging inputs. No ordering guarantees.
A `StagingArea` is a TensorFlow data structure that stores tensors across
multiple steps, and exposes operations that can put and get tensors.
Each `StagingArea` element is a tuple of one or more tensors, where each
tuple component has a static dtype, and may have a static shape.
The capacity of a `StagingArea` may be bounded or unbounded.
It supports multiple concurrent producers and consumers; and
provides exactly-once delivery.
Each element of a `StagingArea` is a fixed-length tuple of tensors whose
dtypes are described by `dtypes`, and whose shapes are optionally described
by the `shapes` argument.
If the `shapes` argument is specified, each component of a staging area
element must have the respective fixed shape. If it is
unspecified, different elements may have different shapes,
It can be configured with a capacity in which case
put(values) will block until space becomes available.
Similarly, it can be configured with a memory limit which
will block put(values) until space is available.
This is mostly useful for limiting the number of tensors on
devices such as GPUs.
All get() and peek() commands block if the requested data
is not present in the Staging Area."
8439,put,tensorflow/tensorflow/python/ops/data_flow_ops.py,1897,method,"Create an op that places a value into the staging area.
This operation will block if the `StagingArea` has reached
its capacity.
Args:
values: A single tensor, a list or tuple of tensors, or a dictionary with
tensor values. The number of elements must match the length of the
list provided to the dtypes argument when creating the StagingArea.
name: A name for the operation (optional).
Returns:
The created op.
Raises:
ValueError: If the number or type of inputs don't match the staging area."
8440,get,tensorflow/tensorflow/python/ops/data_flow_ops.py,1942,method,"Gets one element from this staging area.
If the staging area is empty when this operation executes, it will block
until there is an element to dequeue.
Note that unlike others ops that can block, like the queue Dequeue
operations, this can stop other work from happening. To avoid this, the
intended use is for this to be called only when there will be an element
already available. One method for doing this in a training loop would be to
run a `put()` call during a warmup session.run call, and then call both
`get()` and `put()` in each subsequent step.
The placement of the returned tensor will be determined by the current
device scope when this function is called.
Args:
name: A name for the operation (optional).
Returns:
The tuple of tensors that was gotten."
8441,peek,tensorflow/tensorflow/python/ops/data_flow_ops.py,1976,method,"Peeks at an element in the staging area.
If the staging area is too small to contain the element at
the specified index, it will block until enough elements
are inserted to complete the operation.
The placement of the returned tensor will be determined by
the current device scope when this function is called.
Args:
index: The index of the tensor within the staging area
to look up.
name: A name for the operation (optional).
Returns:
The tuple of tensors that was gotten."
8442,size,tensorflow/tensorflow/python/ops/data_flow_ops.py,2006,method,"Returns the number of elements in the staging area.
Args:
name: A name for the operation (optional)
Returns:
The created op"
8443,clear,tensorflow/tensorflow/python/ops/data_flow_ops.py,2025,method,"Clears the staging area.
Args:
name: A name for the operation (optional)
Returns:
The created op"
8444,MapStagingArea,tensorflow/tensorflow/python/ops/data_flow_ops.py,2045,class,"A `MapStagingArea` is a TensorFlow data structure that stores tensors
across multiple steps, and exposes operations that can put and get tensors.
Each `MapStagingArea` element is a (key, value) pair.
Only int64 keys are supported, other types should be
hashed to produce a key.
Values are a tuple of one or more tensors.
Each tuple component has a static dtype,
and may have a static shape.
The capacity of a `MapStagingArea` may be bounded or unbounded.
It supports multiple concurrent producers and consumers; and
provides exactly-once delivery.
Each value tuple of a `MapStagingArea` is a fixed-length tuple of tensors
whose
dtypes are described by `dtypes`, and whose shapes are optionally described
by the `shapes` argument.
If the `shapes` argument is specified, each component of a staging area
element must have the respective fixed shape. If it is
unspecified, different elements may have different shapes,
It behaves like an associative container with support for:
- put(key, values)
- peek(key) like dict.get(key)
- get(key) like dict.pop(key)
- get(key=None) like dict.popitem()
- size()
- clear()
If ordered a tree structure ordered by key will be used and
get(key=None) will remove (key, value) pairs in increasing key order.
Otherwise a hashtable
It can be configured with a capacity in which case
put(key, values) will block until space becomes available.
Similarly, it can be configured with a memory limit which
will block put(key, values) until space is available.
This is mostly useful for limiting the number of tensors on
devices such as GPUs.
All get() and peek() commands block if the requested
(key, value) pair is not present in the staging area.
Partial puts are supported and will be placed in an incomplete
map until such time as all values associated with the key have
been inserted. Once completed, this (key, value) pair will be
inserted into the map. Data in the incomplete map
counts towards the memory limit, but not towards capacity limit.
Partial gets from the map are also supported.
This removes the partially requested tensors from the entry,
but the entry is only removed from the map once all tensors
associated with it are removed."
8445,put,tensorflow/tensorflow/python/ops/data_flow_ops.py,2163,method,"Create an op that stores the (key, vals) pair in the staging area.
Incomplete puts are possible, preferably using a dictionary for vals
as the appropriate dtypes and shapes can be inferred from the value names
dictionary key values. If vals is a list or tuple, indices must
also be specified so that the op knows at which element position
to perform the insert.
This operation will block if the capacity or memory limit of this
container is reached.
Args:
key: Key associated with the data
vals: Tensor (or a dict/tuple of Tensors) to place
into the staging area.
indices: (Optional) if vals is a tuple/list, this is required.
name: A name for the operation (optional)
Returns:
The created op
Raises:
ValueError: If the number or type of inputs don't match the staging
area."
8446,peek,tensorflow/tensorflow/python/ops/data_flow_ops.py,2237,method,"Peeks at staging area data associated with the key.
If the key is not in the staging area, it will block
until the associated (key, value) is inserted.
Args:
key: Key associated with the required data
indices: Partial list of tensors to retrieve (optional).
A list of integer or string indices.
String indices are only valid if the Staging Area
has names associated with it.
name: A name for the operation (optional)
Returns:
The created op"
8447,get,tensorflow/tensorflow/python/ops/data_flow_ops.py,2272,method,"If the key is provided, the associated (key, value) is returned from the staging area.
If the key is not in the staging area, this method will block until
the associated (key, value) is inserted.
If no key is provided and the staging area is ordered,
the (key, value) with the smallest key will be returned.
Otherwise, a random (key, value) will be returned.
If the staging area is empty when this operation executes,
it will block until there is an element to dequeue.
Args:
key: Key associated with the required data (Optional)
indices: Partial list of tensors to retrieve (optional).
A list of integer or string indices.
String indices are only valid if the Staging Area
has names associated with it.
name: A name for the operation (optional)
Returns:
The created op"
8448,size,tensorflow/tensorflow/python/ops/data_flow_ops.py,2372,method,"Returns the number of elements in the staging area.
Args:
name: A name for the operation (optional)
Returns:
The created op"
8449,incomplete_size,tensorflow/tensorflow/python/ops/data_flow_ops.py,2391,method,"Returns the number of incomplete elements in the staging area.
Args:
name: A name for the operation (optional)
Returns:
The created op"
8450,clear,tensorflow/tensorflow/python/ops/data_flow_ops.py,2410,method,"Clears the staging area.
Args:
name: A name for the operation (optional)
Returns:
The created op"
8451,RecordInput,tensorflow/tensorflow/python/ops/data_flow_ops.py,2430,class,"RecordInput asynchronously reads and randomly yields TFRecords.
A RecordInput Op will continuously read a batch of records asynchronously
into a buffer of some fixed capacity. It can also asynchronously yield
random records from this buffer.
It will not start yielding until at least `buffer_size / 2` elements have been
placed into the buffer so that sufficient randomization can take place.
The order the files are read will be shifted each epoch by `shift_amount` so
that the data is presented in a different order every epoch."
8452,get_yield_op,tensorflow/tensorflow/python/ops/data_flow_ops.py,2491,method,"Adds a node that yields a group of records every time it is executed.
If RecordInput `batches` parameter is not None, it yields a list of
record batches with the specified `batch_size`."
8453,get_zeros_dtype,tensorflow/tensorflow/python/ops/default_gradient.py,26,function,Return the dtype for the default gradient for a Tensor.
8454,shape_and_dtype,tensorflow/tensorflow/python/ops/default_gradient.py,38,function,Return the shape and dtype for the default gradient for a Tensor.
8455,zeros_like,tensorflow/tensorflow/python/ops/default_gradient.py,52,function,"Like array_ops.zeros_like, but respects resource handles."
8456,ones_like,tensorflow/tensorflow/python/ops/default_gradient.py,60,function,"Like array_ops.ones_like, but respects resource handles."
8457,supports_default_grad,tensorflow/tensorflow/python/ops/default_gradient.py,68,function,"Whether tensor `t` supports creating a default gradient.
This function assumes that `t` is of a trainable type.
Args:
t: Tensor
Returns:
Bool"
8458,embedding_lookup,tensorflow/tensorflow/python/ops/embedding_ops.py,255,function,"Looks up embeddings for the given `ids` from a list of tensors.
This function is used to perform parallel lookups on the list of tensors in
`params`. It is a generalization of `tf.gather`, where `params` is
interpreted as a partitioning of a large embedding tensor. `params` may be
a `PartitionedVariable` as returned by using `tf.compat.v1.get_variable()`
with a partitioner.
If `len(params) > 1`, each element `id` of `ids` is partitioned between
the elements of `params` according to the `partition_strategy`.
In all strategies, if the id space does not evenly divide the number of
partitions, each of the first `(max_id + 1) % len(params)` partitions will
be assigned one more id.
If `partition_strategy` is `""mod""`, we assign each id to partition
`p = id % len(params)`. For instance,
13 ids are split across 5 partitions as:
`[[0, 5, 10], [1, 6, 11], [2, 7, 12], [3, 8], [4, 9]]`
If `partition_strategy` is `""div""`, we assign ids to partitions in a
contiguous manner. In this case, 13 ids are split across 5 partitions as:
`[[0, 1, 2], [3, 4, 5], [6, 7, 8], [9, 10], [11, 12]]`
If the input ids are ragged tensors, partition variables are not supported and
the partition strategy and the max_norm are ignored.
The results of the lookup are concatenated into a dense
tensor. The returned tensor has shape `shape(ids) + shape(params)[1:]`.
Args:
params: A single tensor representing the complete embedding tensor, or a
list of P tensors all of same shape except for the first dimension,
representing sharded embedding tensors. Alternatively, a
`PartitionedVariable`, created by partitioning along dimension 0. Each
element must be appropriately sized for the given `partition_strategy`.
ids: A `Tensor` or a 'RaggedTensor' with type `int32` or `int64` containing
the ids to be looked up in `params`.
partition_strategy: A string specifying the partitioning strategy, relevant
if `len(params) > 1`. Currently `""div""` and `""mod""` are supported. Default
is `""mod""`.
name: A name for the operation (optional).
validate_indices: DEPRECATED. If this operation is assigned to CPU, values
in `indices` are always validated to be within range. If assigned to GPU,
out-of-bound indices result in safe but unspecified behavior, which may
include raising an error.
max_norm: If not `None`, each embedding is clipped if its l2-norm is larger
than this value.
Returns:
A `Tensor` or a 'RaggedTensor', depending on the input, with the same type
as the tensors in `params`.
Raises:
ValueError: If `params` is empty."
8459,embedding_lookup_v2,tensorflow/tensorflow/python/ops/embedding_ops.py,333,function,"Looks up embeddings for the given `ids` from a list of tensors.
This function is used to perform parallel lookups on the list of tensors in
`params`. It is a generalization of `tf.gather`, where `params` is
interpreted as a partitioning of a large embedding tensor.
If `len(params) > 1`, each element `id` of `ids` is partitioned between the
elements of `params` according to the ""div"" partition strategy, which means we
assign ids to partitions in a contiguous manner. For instance, 13 ids are
split across 5 partitions as:
`[[0, 1, 2], [3, 4, 5], [6, 7, 8], [9, 10], [11, 12]]`.
If the id space does not evenly divide the number of partitions, each of the
first `(max_id + 1) % len(params)` partitions will be assigned one more id.
The results of the lookup are concatenated into a dense
tensor. The returned tensor has shape `shape(ids) + shape(params)[1:]`.
Args:
params: A single tensor representing the complete embedding tensor, or a
list of tensors all of same shape except for the first dimension,
representing sharded embedding tensors following ""div"" partition strategy.
ids: A `Tensor` with type `int32` or `int64` containing the ids to be looked
up in `params`.
max_norm: If not `None`, each embedding is clipped if its l2-norm is larger
than this value.
name: A name for the operation (optional).
Returns:
A `Tensor` with the same type as the tensors in `params`.
For instance, if `params` is a 5x2 matrix:
```python
[[1, 2], [3, 4], [5, 6], [7, 8], [9, 10]]
```
or a list of matrices:
```python
params[0]: [[1, 2], [3, 4]]
params[1]: [[5, 6], [7, 8]]
params[2]: [[9, 10]]
```
and `ids` is:
```python
[0, 3, 4]
```
The output will be a 3x2 matrix:
```python
[[1, 2], [7, 8], [9, 10]]
```
Raises:
ValueError: If `params` is empty."
8460,embedding_lookup_sparse,tensorflow/tensorflow/python/ops/embedding_ops.py,399,function,"Looks up embeddings for the given ids and weights from a list of tensors.
This op assumes that there is at least one id for each row in the dense tensor
represented by sp_ids (i.e. there are no rows with empty features), and that
all the indices of sp_ids are in canonical row-major order.
`sp_ids` and `sp_weights` (if not None) are `SparseTensor`s with rank of 2.
Embeddings are always aggregated along the last dimension.
It also assumes that all id values lie in the range [0, p0), where p0
is the sum of the size of params along dimension 0.
Args:
params: A single tensor representing the complete embedding tensor, or a
list tensors all of same shape except for the first dimension,
representing sharded embedding tensors. Alternatively, a
`PartitionedVariable`, created by partitioning along dimension 0. Each
element must be appropriately sized for the given `partition_strategy`.
sp_ids: N x M `SparseTensor` of int64 ids where N is typically batch size
and M is arbitrary.
sp_weights: either a `SparseTensor` of float / double weights, or `None` to
indicate all weights should be taken to be 1. If specified, `sp_weights`
must have exactly the same shape and indices as `sp_ids`.
partition_strategy: A string specifying the partitioning strategy, relevant
if `len(params) > 1`. Currently `""div""` and `""mod""` are supported. Default
is `""mod""`. See `tf.nn.embedding_lookup` for more details.
name: Optional name for the op.
combiner: A string specifying the reduction op. Currently ""mean"", ""sqrtn""
and ""sum"" are supported. ""sum"" computes the weighted sum of the embedding
results for each row. ""mean"" is the weighted sum divided by the total
weight. ""sqrtn"" is the weighted sum divided by the square root of the sum
of the squares of the weights. Defaults to `mean`.
max_norm: If not `None`, each embedding is clipped if its l2-norm is larger
than this value, before combining.
Returns:
A dense tensor representing the combined embeddings for the
sparse ids. For each row in the dense tensor represented by `sp_ids`, the op
looks up the embeddings for all ids in that row, multiplies them by the
corresponding weight, and combines these embeddings as specified.
In other words, if
`shape(combined params) = [p0, p1, ..., pm]`
and
`shape(sp_ids) = shape(sp_weights) = [d0, d1]`
then
`shape(output) = [d0, p1, ..., pm]`.
For instance, if params is a 10x20 matrix, and sp_ids / sp_weights are
```python
[0, 0]: id 1, weight 2.0
[0, 1]: id 3, weight 0.5
[1, 0]: id 0, weight 1.0
[2, 3]: id 1, weight 3.0
```
with `combiner`=""mean"", then the output will be a 3x20 matrix where
```python
output[0, :] = (params[1, :] * 2.0 + params[3, :] * 0.5) / (2.0 + 0.5)
output[1, :] = (params[0, :] * 1.0) / 1.0
output[2, :] = (params[1, :] * 3.0) / 3.0
```
Raises:
TypeError: If `sp_ids` is not a `SparseTensor`, or if `sp_weights` is
neither `None` nor `SparseTensor`.
ValueError: If `combiner` is not one of {""mean"", ""sqrtn"", ""sum""}."
8461,embedding_lookup_sparse_v2,tensorflow/tensorflow/python/ops/embedding_ops.py,582,function,"Looks up embeddings for the given ids and weights from a list of tensors.
This op assumes that there is at least one id for each row in the dense tensor
represented by sp_ids (i.e. there are no rows with empty features), and that
all the indices of sp_ids are in canonical row-major order.
`sp_ids` and `sp_weights` (if not None) are `SparseTensor`s with rank of 2.
Embeddings are always aggregated along the last dimension.
It also assumes that all id values lie in the range [0, p0), where p0
is the sum of the size of params along dimension 0.
If `len(params) > 1`, each element of `sp_ids` is partitioned between the
elements of `params` according to the ""div"" partition strategy, which means we
assign ids to partitions in a contiguous manner. For instance, 13 ids are
split across 5 partitions as:
`[[0, 1, 2], [3, 4, 5], [6, 7, 8], [9, 10], [11, 12]]`.
If the id space does not evenly divide the number of partitions, each of the
first `(max_id + 1) % len(params)` partitions will be assigned one more id.
Args:
params: A single tensor representing the complete embedding tensor, or a
list of tensors all of same shape except for the first dimension,
representing sharded embedding tensors following ""div"" partition strategy.
sp_ids: N x M `SparseTensor` of int64 ids where N is typically batch size
and M is arbitrary.
sp_weights: either a `SparseTensor` of float / double weights, or `None` to
indicate all weights should be taken to be 1. If specified, `sp_weights`
must have exactly the same shape and indices as `sp_ids`.
combiner: A string specifying the reduction op. Currently ""mean"", ""sqrtn""
and ""sum"" are supported. ""sum"" computes the weighted sum of the embedding
results for each row. ""mean"" is the weighted sum divided by the total
weight. ""sqrtn"" is the weighted sum divided by the square root of the sum
of the squares of the weights. Defaults to `mean`.
max_norm: If not `None`, each embedding is clipped if its l2-norm is larger
than this value, before combining.
name: Optional name for the op.
Returns:
A dense tensor representing the combined embeddings for the
sparse ids. For each row in the dense tensor represented by `sp_ids`, the op
looks up the embeddings for all ids in that row, multiplies them by the
corresponding weight, and combines these embeddings as specified.
In other words, if
`shape(combined params) = [p0, p1, ..., pm]`
and
`shape(sp_ids) = shape(sp_weights) = [d0, d1]`
then
`shape(output) = [d0, p1, ..., pm]`.
For instance, if params is a 10x20 matrix, and sp_ids / sp_weights are
```python
[0, 0]: id 1, weight 2.0
[0, 1]: id 3, weight 0.5
[1, 0]: id 0, weight 1.0
[2, 3]: id 1, weight 3.0
```
with `combiner`=""mean"", then the output will be a 3x20 matrix where
```python
output[0, :] = (params[1, :] * 2.0 + params[3, :] * 0.5) / (2.0 + 0.5)
output[1, :] = (params[0, :] * 1.0) / 1.0
output[2, :] = (params[1, :] * 3.0) / 3.0
```
Raises:
TypeError: If `sp_ids` is not a `SparseTensor`, or if `sp_weights` is
neither `None` nor `SparseTensor`.
ValueError: If `combiner` is not one of {""mean"", ""sqrtn"", ""sum""}."
8462,safe_embedding_lookup_sparse_v2,tensorflow/tensorflow/python/ops/embedding_ops.py,673,function,"Lookup embedding results, accounting for invalid IDs and empty features.
The partitioned embedding in `embedding_weights` must all be the same shape
except for the first dimension. The first dimension is allowed to vary as the
vocabulary size is not necessarily a multiple of num of shards.
Invalid IDs (< 0) are pruned from input IDs and weights, as well as any IDs
with non-positive weight. For an entry with no features, the embedding vector
for `default_id` is returned, or the 0-vector if `default_id` is not supplied.
The ids and weights may be multi-dimensional. Embeddings are always aggregated
along the last dimension.
If `len(embedding_weights) > 1`, each element `id` of `ids` is partitioned
between the elements of `embedding_weights` according to the ""div"" partition
strategy, which means we assign ids to partitions in a contiguous manner. For
instance, 13 ids are split across 5 partitions as:
`[[0, 1, 2], [3, 4, 5], [6, 7, 8], [9, 10], [11, 12]]`.
If the id space does not evenly divide the number of partitions, each of the
first `(max_id + 1) % len(embedding_weights)` partitions will be assigned one
more id.
Args:
embedding_weights: A single tensor representing the complete embedding
tensor, or a list of tensors all of same shape except for the first
dimension, representing sharded embedding tensors following ""div""
partition strategy.
sparse_ids: `SparseTensor` of shape `[d_0, d_1, ..., d_n]` containing the
ids. `d_0` is typically batch size.
sparse_weights: `SparseTensor` of same shape as `sparse_ids`, containing
float weights corresponding to `sparse_ids`, or `None` if all weights are
be assumed to be 1.0.
combiner: A string specifying how to combine embedding results for each
entry. Currently ""mean"", ""sqrtn"" and ""sum"" are supported, with ""mean"" the
default.
default_id: The id to use for an entry with no features. Defaults to
0-vector.
max_norm: If not `None`, all embeddings are l2-normalized to max_norm before
combining.
name: A name for this operation (optional).
Returns:
A dense tensor representing the combined embeddings for the
sparse ids. For each row in the dense tensor represented by `sparse_ids`,
the op looks up the embeddings for all ids in that row, multiplies them by
the corresponding weight, and combines these embeddings as specified.
In other words, if
`shape(combined embedding_weights) = [p0, p1, ..., pm]`
and
`shape(sparse_ids) = shape(sparse_weights) = [d0, d1, ..., dn]`
then
`shape(output) = [d0, d1, ... dn-1, p1, ..., pm]`.
For instance, if params is a 10x20 matrix, and sp_ids / sp_weights are
```python
[0, 0]: id 1, weight 2.0
[0, 1]: id 3, weight 0.5
[1, 0]: id -1, weight 1.0
[2, 3]: id 1, weight 3.0
```
`default_id` is 0.
with `combiner`=""mean"", then the output will be a 3x20 matrix where
```python
output[0, :] = (params[1, :] * 2.0 + params[3, :] * 0.5) / (2.0 + 0.5)
output[1, :] = (params[0, :] * 1.0) / 1.0
output[2, :] = (params[1, :] * 3.0) / 3.0
```
Raises:
ValueError: if `embedding_weights` is empty."
8463,safe_embedding_lookup_sparse,tensorflow/tensorflow/python/ops/embedding_ops.py,775,function,"Lookup embedding results, accounting for invalid IDs and empty features.
The partitioned embedding in `embedding_weights` must all be the same shape
except for the first dimension. The first dimension is allowed to vary as the
vocabulary size is not necessarily a multiple of `P`. `embedding_weights`
may be a `PartitionedVariable` as returned by using
`tf.compat.v1.get_variable()` with a
partitioner.
Invalid IDs (< 0) are pruned from input IDs and weights, as well as any IDs
with non-positive weight. For an entry with no features, the embedding vector
for `default_id` is returned, or the 0-vector if `default_id` is not supplied.
The ids and weights may be multi-dimensional. Embeddings are always aggregated
along the last dimension.
Args:
embedding_weights: A single tensor representing the complete embedding
tensor, or a list tensors all of same shape except for the first
dimension, representing sharded embedding tensors. Alternatively, a
`PartitionedVariable`, created by partitioning along dimension 0. Each
element must be appropriately sized for the given `partition_strategy`.
sparse_ids: `SparseTensor` of shape `[d_0, d_1, ..., d_n]` containing the
ids. `d_0` is typically batch size.
sparse_weights: `SparseTensor` of same shape as `sparse_ids`, containing
float weights corresponding to `sparse_ids`, or `None` if all weights are
be assumed to be 1.0.
combiner: A string specifying how to combine embedding results for each
entry. Currently ""mean"", ""sqrtn"" and ""sum"" are supported, with ""mean"" the
default.
default_id: The id to use for an entry with no features.
name: A name for this operation (optional).
partition_strategy: A string specifying the partitioning strategy. Currently
`""div""` and `""mod""` are supported. Default is `""div""`.
max_norm: If not `None`, all embeddings are l2-normalized to max_norm before
combining.
Returns:
A dense tensor representing the combined embeddings for the
sparse ids. For each row in the dense tensor represented by `sp_ids`, the op
looks up the embeddings for all ids in that row, multiplies them by the
corresponding weight, and combines these embeddings as specified.
In other words, if
`shape(combined embedding_weights) = [p0, p1, ..., pm]`
and
`shape(sparse_ids) = shape(sparse_weights) = [d0, d1, ..., dn]`
then
`shape(output) = [d0, d1, ... dn-1, p1, ..., pm]`.
For instance, if params is a 10x20 matrix, and sp_ids / sp_weights are
```python
[0, 0]: id 1, weight 2.0
[0, 1]: id 3, weight 0.5
[1, 0]: id -1, weight 1.0
[2, 3]: id 1, weight 3.0
```
`default_id` is 0.
with `combiner`=""mean"", then the output will be a 3x20 matrix where
```python
output[0, :] = (params[1, :] * 2.0 + params[3, :] * 0.5) / (2.0 + 0.5)
output[1, :] = (params[0, :] * 1.0) / 1.0
output[2, :] = (params[1, :] * 3.0) / 3.0
```
Raises:
ValueError: if `embedding_weights` is empty."
8464,embedding_lookup_ragged,tensorflow/tensorflow/python/ops/embedding_ops.py,943,function,"Look up the ragged ids in a list of embedding tensors.
Args:
embedding_weights: A tensor representing the complete embedding tensor
having the shape [e1, ...eM]
ragged_ids: A 'RaggedTensor' with type 'int32' or 'int64' containing the ids
to be looked up in 'embedding_weights' of shape [r0, ..rN]. Values must be
in the range '[0, embedding_weights.shape[0]]'.
partition_strategy: A string specifying the partitioning strategy.
max_norm: If not `None`, each embedding is clipped if its l2-norm is larger
than this value.
name: A name for the operation (optional)
Returns:
A ragged tensor of shape [r0, r1, ...rN, e1, ...eM].
Raises:
ValueError: whether the embedding_weights is empty or the ragged_ids is
not a RaggedTensor."
8465,foldl,tensorflow/tensorflow/python/ops/functional_ops.py,50,function,"foldl on the list of tensors unpacked from `elems` on dimension 0.
This foldl operator repeatedly applies the callable `fn` to a sequence
of elements from first to last. The elements are made of the tensors
unpacked from `elems` on dimension 0. The callable fn takes two tensors as
arguments. The first argument is the accumulated value computed from the
preceding invocation of fn, and the second is the value at the current
position of `elems`. If `initializer` is None, `elems` must contain at least
one element, and its first element is used as the initializer.
Suppose that `elems` is unpacked into `values`, a list of tensors. The shape
of the result tensor is fn(initializer, values[0]).shape`.
This method also allows multi-arity `elems` and output of `fn`. If `elems`
is a (possibly nested) list or tuple of tensors, then each of these tensors
must have a matching first (unpack) dimension. The signature of `fn` may
match the structure of `elems`. That is, if `elems` is
`(t1, [t2, t3, [t4, t5]])`, then an appropriate signature for `fn` is:
`fn = lambda (t1, [t2, t3, [t4, t5]]):`.
Args:
fn: The callable to be performed.
elems: A tensor or (possibly nested) sequence of tensors, each of which will
be unpacked along their first dimension. The nested sequence of the
resulting slices will be the first argument to `fn`.
initializer: (optional) A tensor or (possibly nested) sequence of tensors,
as the initial value for the accumulator.
parallel_iterations: (optional) The number of iterations allowed to run in
parallel.
back_prop: (optional) True enables support for back propagation.
swap_memory: (optional) True enables GPU-CPU memory swapping.
name: (optional) Name prefix for the returned tensors.
Returns:
A tensor or (possibly nested) sequence of tensors, resulting from applying
`fn` consecutively to the list of tensors unpacked from `elems`, from first
to last.
Raises:
TypeError: if `fn` is not callable.
Example:
```python
elems = tf.constant([1, 2, 3, 4, 5, 6])
sum = foldl(lambda a, x: a + x, elems)
# sum == 21
```"
8466,foldl_v2,tensorflow/tensorflow/python/ops/functional_ops.py,177,function,"foldl on the list of tensors unpacked from `elems` on dimension 0.
This foldl operator repeatedly applies the callable `fn` to a sequence
of elements from first to last. The elements are made of the tensors
unpacked from `elems` on dimension 0. The callable fn takes two tensors as
arguments. The first argument is the accumulated value computed from the
preceding invocation of fn, and the second is the value at the current
position of `elems`. If `initializer` is None, `elems` must contain at least
one element, and its first element is used as the initializer.
Suppose that `elems` is unpacked into `values`, a list of tensors. The shape
of the result tensor is fn(initializer, values[0]).shape`.
This method also allows multi-arity `elems` and output of `fn`. If `elems`
is a (possibly nested) list or tuple of tensors, then each of these tensors
must have a matching first (unpack) dimension. The signature of `fn` may
match the structure of `elems`. That is, if `elems` is
`(t1, [t2, t3, [t4, t5]])`, then an appropriate signature for `fn` is:
`fn = lambda (t1, [t2, t3, [t4, t5]]):`.
Args:
fn: The callable to be performed.
elems: A tensor or (possibly nested) sequence of tensors, each of which will
be unpacked along their first dimension. The nested sequence of the
resulting slices will be the first argument to `fn`.
initializer: (optional) A tensor or (possibly nested) sequence of tensors,
as the initial value for the accumulator.
parallel_iterations: (optional) The number of iterations allowed to run in
parallel.
back_prop: (optional) Deprecated. False disables support for back
propagation. Prefer using `tf.stop_gradient` instead.
swap_memory: (optional) True enables GPU-CPU memory swapping.
name: (optional) Name prefix for the returned tensors.
Returns:
A tensor or (possibly nested) sequence of tensors, resulting from applying
`fn` consecutively to the list of tensors unpacked from `elems`, from first
to last.
Raises:
TypeError: if `fn` is not callable.
Example:
```python
elems = tf.constant([1, 2, 3, 4, 5, 6])
sum = foldl(lambda a, x: a + x, elems)
# sum == 21
```"
8467,foldr,tensorflow/tensorflow/python/ops/functional_ops.py,245,function,"foldr on the list of tensors unpacked from `elems` on dimension 0.
This foldr operator repeatedly applies the callable `fn` to a sequence
of elements from last to first. The elements are made of the tensors
unpacked from `elems`. The callable fn takes two tensors as arguments.
The first argument is the accumulated value computed from the preceding
invocation of fn, and the second is the value at the current position of
`elems`. If `initializer` is None, `elems` must contain at least one element,
and its first element is used as the initializer.
Suppose that `elems` is unpacked into `values`, a list of tensors. The shape
of the result tensor is `fn(initializer, values[0]).shape`.
This method also allows multi-arity `elems` and output of `fn`. If `elems`
is a (possibly nested) list or tuple of tensors, then each of these tensors
must have a matching first (unpack) dimension. The signature of `fn` may
match the structure of `elems`. That is, if `elems` is
`(t1, [t2, t3, [t4, t5]])`, then an appropriate signature for `fn` is:
`fn = lambda (t1, [t2, t3, [t4, t5]]):`.
Args:
fn: The callable to be performed.
elems: A tensor or (possibly nested) sequence of tensors, each of which will
be unpacked along their first dimension. The nested sequence of the
resulting slices will be the first argument to `fn`.
initializer: (optional) A tensor or (possibly nested) sequence of tensors,
as the initial value for the accumulator.
parallel_iterations: (optional) The number of iterations allowed to run in
parallel.
back_prop: (optional) True enables support for back propagation.
swap_memory: (optional) True enables GPU-CPU memory swapping.
name: (optional) Name prefix for the returned tensors.
Returns:
A tensor or (possibly nested) sequence of tensors, resulting from applying
`fn` consecutively to the list of tensors unpacked from `elems`, from last
to first.
Raises:
TypeError: if `fn` is not callable.
Example:
```python
elems = [1, 2, 3, 4, 5, 6]
sum = foldr(lambda a, x: a + x, elems)
# sum == 21
```"
8468,foldr_v2,tensorflow/tensorflow/python/ops/functional_ops.py,373,function,"foldr on the list of tensors unpacked from `elems` on dimension 0.
This foldr operator repeatedly applies the callable `fn` to a sequence
of elements from last to first. The elements are made of the tensors
unpacked from `elems`. The callable fn takes two tensors as arguments.
The first argument is the accumulated value computed from the preceding
invocation of fn, and the second is the value at the current position of
`elems`. If `initializer` is None, `elems` must contain at least one element,
and its first element is used as the initializer.
Suppose that `elems` is unpacked into `values`, a list of tensors. The shape
of the result tensor is `fn(initializer, values[0]).shape`.
This method also allows multi-arity `elems` and output of `fn`. If `elems`
is a (possibly nested) list or tuple of tensors, then each of these tensors
must have a matching first (unpack) dimension. The signature of `fn` may
match the structure of `elems`. That is, if `elems` is
`(t1, [t2, t3, [t4, t5]])`, then an appropriate signature for `fn` is:
`fn = lambda (t1, [t2, t3, [t4, t5]]):`.
Args:
fn: The callable to be performed.
elems: A tensor or (possibly nested) sequence of tensors, each of which will
be unpacked along their first dimension. The nested sequence of the
resulting slices will be the first argument to `fn`.
initializer: (optional) A tensor or (possibly nested) sequence of tensors,
as the initial value for the accumulator.
parallel_iterations: (optional) The number of iterations allowed to run in
parallel.
back_prop: (optional) Deprecated. False disables support for back
propagation. Prefer using `tf.stop_gradient` instead.
swap_memory: (optional) True enables GPU-CPU memory swapping.
name: (optional) Name prefix for the returned tensors.
Returns:
A tensor or (possibly nested) sequence of tensors, resulting from applying
`fn` consecutively to the list of tensors unpacked from `elems`, from last
to first.
Raises:
TypeError: if `fn` is not callable.
Example:
```python
elems = [1, 2, 3, 4, 5, 6]
sum = foldr(lambda a, x: a + x, elems)
# sum == 21
```"
8469,scan,tensorflow/tensorflow/python/ops/functional_ops.py,441,function,"scan on the list of tensors unpacked from `elems` on dimension 0.
See also `tf.map_fn`.
The simplest version of `scan` repeatedly applies the callable `fn` to a
sequence of elements from first to last. The elements are made of the tensors
unpacked from `elems` on dimension 0. The callable fn takes two tensors as
arguments. The first argument is the accumulated value computed from the
preceding invocation of fn, and the second is the value at the current
position of `elems`. If `initializer` is None, `elems` must contain at least
one element, and its first element is used as the initializer.
Suppose that `elems` is unpacked into `values`, a list of tensors. The shape
of the result tensor is `[len(values)] + fn(initializer, values[0]).shape`.
If reverse=True, it's fn(initializer, values[-1]).shape.
This method also allows multi-arity `elems` and accumulator. If `elems`
is a (possibly nested) list or tuple of tensors, then each of these tensors
must have a matching first (unpack) dimension. The second argument of
`fn` must match the structure of `elems`.
If no `initializer` is provided, the output structure and dtypes of `fn`
are assumed to be the same as its input; and in this case, the first
argument of `fn` must match the structure of `elems`.
If an `initializer` is provided, then the output of `fn` must have the same
structure as `initializer`; and the first argument of `fn` must match
this structure.
For example, if `elems` is `(t1, [t2, t3])` and `initializer` is
`[i1, i2]` then an appropriate signature for `fn` in `python2` is:
`fn = lambda (acc_p1, acc_p2), (t1, [t2, t3]):` and `fn` must return a list,
`[acc_n1, acc_n2]`. An alternative correct signature for `fn`, and the
one that works in `python3`, is:
`fn = lambda a, t:`, where `a` and `t` correspond to the input tuples.
Args:
fn: The callable to be performed. It accepts two arguments. The first will
have the same structure as `initializer` if one is provided, otherwise it
will have the same structure as `elems`. The second will have the same
(possibly nested) structure as `elems`. Its output must have the same
structure as `initializer` if one is provided, otherwise it must have the
same structure as `elems`.
elems: A tensor or (possibly nested) sequence of tensors, each of which will
be unpacked along their first dimension. The nested sequence of the
resulting slices will be the first argument to `fn`.
initializer: (optional) A tensor or (possibly nested) sequence of tensors,
initial value for the accumulator, and the expected output type of `fn`.
parallel_iterations: (optional) The number of iterations allowed to run in
parallel.
back_prop: (optional) True enables support for back propagation.
swap_memory: (optional) True enables GPU-CPU memory swapping.
infer_shape: (optional) False disables tests for consistent output shapes.
reverse: (optional) True scans the tensor last to first (instead of first to
last).
name: (optional) Name prefix for the returned tensors.
Returns:
A tensor or (possibly nested) sequence of tensors. Each tensor packs the
results of applying `fn` to tensors unpacked from `elems` along the first
dimension, and the previous accumulator value(s), from first to last (or
last to first, if `reverse=True`).
Raises:
TypeError: if `fn` is not callable or the structure of the output of
`fn` and `initializer` do not match.
ValueError: if the lengths of the output of `fn` and `initializer`
do not match.
Examples:
```python
elems = np.array([1, 2, 3, 4, 5, 6])
sum = scan(lambda a, x: a + x, elems)
# sum == [1, 3, 6, 10, 15, 21]
sum = scan(lambda a, x: a + x, elems, reverse=True)
# sum == [21, 20, 18, 15, 11, 6]
```
```python
elems = np.array([1, 2, 3, 4, 5, 6])
initializer = np.array(0)
sum_one = scan(
lambda a, x: x[0] - x[1] + a, (elems + 1, elems), initializer)
# sum_one == [1, 2, 3, 4, 5, 6]
```
```python
elems = np.array([1, 0, 0, 0, 0, 0])
initializer = (np.array(0), np.array(1))
fibonaccis = scan(lambda a, _: (a[1], a[0] + a[1]), elems, initializer)
# fibonaccis == ([1, 1, 2, 3, 5, 8], [1, 2, 3, 5, 8, 13])
```"
8470,scan_v2,tensorflow/tensorflow/python/ops/functional_ops.py,705,function,"scan on the list of tensors unpacked from `elems` on dimension 0.
The simplest version of `scan` repeatedly applies the callable `fn` to a
sequence of elements from first to last. The elements are made of the tensors
unpacked from `elems` on dimension 0. The callable fn takes two tensors as
arguments. The first argument is the accumulated value computed from the
preceding invocation of fn, and the second is the value at the current
position of `elems`. If `initializer` is None, `elems` must contain at least
one element, and its first element is used as the initializer.
Suppose that `elems` is unpacked into `values`, a list of tensors. The shape
of the result tensor is `[len(values)] + fn(initializer, values[0]).shape`.
If reverse=True, it's fn(initializer, values[-1]).shape.
This method also allows multi-arity `elems` and accumulator. If `elems`
is a (possibly nested) list or tuple of tensors, then each of these tensors
must have a matching first (unpack) dimension. The second argument of
`fn` must match the structure of `elems`.
If no `initializer` is provided, the output structure and dtypes of `fn`
are assumed to be the same as its input; and in this case, the first
argument of `fn` must match the structure of `elems`.
If an `initializer` is provided, then the output of `fn` must have the same
structure as `initializer`; and the first argument of `fn` must match
this structure.
For example, if `elems` is `(t1, [t2, t3])` and `initializer` is
`[i1, i2]` then an appropriate signature for `fn` in `python2` is:
`fn = lambda (acc_p1, acc_p2), (t1, [t2, t3]):` and `fn` must return a list,
`[acc_n1, acc_n2]`. An alternative correct signature for `fn`, and the
one that works in `python3`, is:
`fn = lambda a, t:`, where `a` and `t` correspond to the input tuples.
Args:
fn: The callable to be performed. It accepts two arguments. The first will
have the same structure as `initializer` if one is provided, otherwise it
will have the same structure as `elems`. The second will have the same
(possibly nested) structure as `elems`. Its output must have the same
structure as `initializer` if one is provided, otherwise it must have the
same structure as `elems`.
elems: A tensor or (possibly nested) sequence of tensors, each of which will
be unpacked along their first dimension. The nested sequence of the
resulting slices will be the first argument to `fn`.
initializer: (optional) A tensor or (possibly nested) sequence of tensors,
initial value for the accumulator, and the expected output type of `fn`.
parallel_iterations: (optional) The number of iterations allowed to run in
parallel.
back_prop: (optional) Deprecated. False disables support for back
propagation. Prefer using `tf.stop_gradient` instead.
swap_memory: (optional) True enables GPU-CPU memory swapping.
infer_shape: (optional) False disables tests for consistent output shapes.
reverse: (optional) True scans the tensor last to first (instead of first to
last).
name: (optional) Name prefix for the returned tensors.
Returns:
A tensor or (possibly nested) sequence of tensors. Each tensor packs the
results of applying `fn` to tensors unpacked from `elems` along the first
dimension, and the previous accumulator value(s), from first to last (or
last to first, if `reverse=True`).
Raises:
TypeError: if `fn` is not callable or the structure of the output of
`fn` and `initializer` do not match.
ValueError: if the lengths of the output of `fn` and `initializer`
do not match.
Examples:
```python
elems = np.array([1, 2, 3, 4, 5, 6])
sum = scan(lambda a, x: a + x, elems)
# sum == [1, 3, 6, 10, 15, 21]
sum = scan(lambda a, x: a + x, elems, reverse=True)
# sum == [21, 20, 18, 15, 11, 6]
```
```python
elems = np.array([1, 2, 3, 4, 5, 6])
initializer = np.array(0)
sum_one = scan(
lambda a, x: x[0] - x[1] + a, (elems + 1, elems), initializer)
# sum_one == [1, 2, 3, 4, 5, 6]
```
```python
elems = np.array([1, 0, 0, 0, 0, 0])
initializer = (np.array(0), np.array(1))
fibonaccis = scan(lambda a, _: (a[1], a[0] + a[1]), elems, initializer)
# fibonaccis == ([1, 1, 2, 3, 5, 8], [1, 2, 3, 5, 8, 13])
```"
8471,If,tensorflow/tensorflow/python/ops/functional_ops.py,819,function,"output = Cond(inputs) ?
then_branch(inputs) : else_branch(inputs).
Args:
cond: A `Tensor`. A scalar. If the scalar is not a boolean, the scalar is
converted to a boolean according to the following rule: if the scalar is a
numerical value, non-zero means True and zero means False; if the scalar
is a string, non-empty means True and empty means False.
inputs: A list of input tensors.
then_branch: A function takes 'inputs' and returns a list of tensors, whose
types are the same as what else_branch returns.
else_branch: A function takes 'inputs' and returns a list of tensors. whose
types are the same as what then_branch returns.
name: A name for the operation (optional).
Returns:
A list of tensors returned by either then_branch(inputs)
or else_branch(inputs)."
8472,Gradient,tensorflow/tensorflow/python/ops/functional_ops.py,850,function,"Computes the gradient function for function f via backpropagation.
Args:
inputs: A list of tensors of size N + M.
f: The function we want to compute the gradient for. The function 'f' must
be a numerical function which takes N inputs and produces M outputs. Its
gradient function 'g', which is a function taking N + M inputs and
produces N outputs. I.e. if we have (y1, y2, ..., yM) = f(x1, x2, ...,
xN), then, g is (dL/dx1, dL/dx2, ..., dL/dxN) = g(x1, x2, ..., xN, dL/dy1,
dL/dy2, ..., dL/dyM), where L is a scalar-value function of (x1, x2, ...,
xN) (e.g., the loss function). dL/dxi is the partial derivative of L with
respect to xi.
name: A name for the operation (optional).
Returns:
A list of tensors of size N."
8473,While,tensorflow/tensorflow/python/ops/functional_ops.py,911,function,"output = input; While (Cond(output)) { output = Body(output) }.
Args:
input_: A list of `Tensor` objects. A list of input tensors whose types are
T.
cond: . A function takes 'input' and returns a tensor. If the tensor is a
scalar of non-boolean, the scalar is converted to a boolean
according to the following rule: if the scalar is a numerical value,
non-zero means True and zero means False; if the scalar is a string,
non-empty means True and empty means False. If the tensor is not a
scalar, non-emptiness means True and False otherwise.
body: . A function takes a list of tensors and returns another list tensors.
Both lists have the same types as specified by T.
name: A name for the operation (optional).
hostmem: A list of integer. If i is in the list, input[i] is a host memory
tensor.
Raises:
ValueError: if `cond` has implicitly captured inputs or if `cond` and `body`
have different signatures.
Returns:
A list of `Tensor` objects. Has the same type as `input`.
A list of output tensors whose types are T."
8474,For,tensorflow/tensorflow/python/ops/functional_ops.py,1054,function,"out = input; for i in range(start, limit, delta) out = body(i, out).
Args:
start: A `Tensor` of type `int32`.
limit: A `Tensor` of type `int32`.
delta: A `Tensor` of type `int32`.
inputs: A list of `Tensor` objects. A list of input tensors whose types are
T.
body: A function takes a list of tensors and returns another list of
tensors. Both lists have the same types as (int32, T...).
name: A name for the operation (optional).
hostmem: A list of integer. If i is in the list, inputs[i] is a host memory
tensor. In other words, (i+1)-th argument of the body function is
expecting a host memory.
rewrite_with_while: If True, using While op to implement the For.
Returns:
A list of `Tensor` objects. Has the same type as `input`.
A list of output tensors whose types are T."
8475,partitioned_call,tensorflow/tensorflow/python/ops/functional_ops.py,1112,function,"Executes a function while respecting device annotations.
Currently, only those functions that execute within the same address space
can be executed.
Args:
args: The arguments of the function, including captured inputs.
f: The function to execute; an instance of `_DefinedFunction` or
`_EagerDefinedFunction`.
tout: a list containing the output dtypes enums; if `None`, inferred from
the signature of `f`.
executing_eagerly: (Optional) A boolean indicating whether the context is
executing eagerly. If `None`, fetched from the global context.
config: (Optional) A `tensorflow::ConfigProto` proto, serialized. If `None`,
all optimizations are disabled. Currently only handled for eager defined
functions.
executor_type: (Optional) A string for the name of the executor to be used
in the function call. If not set, or set to an empty string, the default
tensorflow executor will be used.
Returns:
The list of `Tensor`s returned by invoking `f(args)`. If the function does
not return anything, then returns `None` if eager execution is enabled, or
the `Operation` if not."
8476,compute_gradient,tensorflow/tensorflow/python/ops/gradient_checker.py,277,function,"Computes and returns the theoretical and numerical Jacobian.
If `x` or `y` is complex, the Jacobian will still be real but the
corresponding Jacobian dimension(s) will be twice as large. This is required
even if both input and output is complex since TensorFlow graphs are not
necessarily holomorphic, and may have gradients not expressible as complex
numbers. For example, if `x` is complex with shape `[m]` and `y` is complex
with shape `[n]`, each Jacobian `J` will have shape `[m * 2, n * 2]` with
J[:m, :n] = d(Re y)/d(Re x)
J[:m, n:] = d(Im y)/d(Re x)
J[m:, :n] = d(Re y)/d(Im x)
J[m:, n:] = d(Im y)/d(Im x)
Args:
x: a tensor or list of tensors
x_shape: the dimensions of x as a tuple or an array of ints. If x is a list,
then this is the list of shapes.
y: a tensor
y_shape: the dimensions of y as a tuple or an array of ints.
x_init_value: (optional) a numpy array of the same shape as ""x""
representing the initial value of x. If x is a list, this should be a list
of numpy arrays. If this is none, the function will pick a random tensor
as the initial value.
delta: (optional) the amount of perturbation.
init_targets: list of targets to run to initialize model params.
extra_feed_dict: dict that allows fixing specified tensor values
during the Jacobian calculation.
Returns:
Two 2-d numpy arrays representing the theoretical and numerical
Jacobian for dy/dx. Each has ""x_size"" rows and ""y_size"" columns
where ""x_size"" is the number of elements in x and ""y_size"" is the
number of elements in y. If x is a list, returns a list of two numpy arrays."
8477,compute_gradient_error,tensorflow/tensorflow/python/ops/gradient_checker.py,354,function,"Computes the gradient error.
Computes the maximum error for dy/dx between the computed Jacobian and the
numerically estimated Jacobian.
This function will modify the tensors passed in as it adds more operations
and hence changing the consumers of the operations of the input tensors.
This function adds operations to the current session. To compute the error
using a particular device, such as a GPU, use the standard methods for
setting a device (e.g. using with sess.graph.device() or setting a device
function in the session constructor).
Args:
x: a tensor or list of tensors
x_shape: the dimensions of x as a tuple or an array of ints. If x is a list,
then this is the list of shapes.
y: a tensor
y_shape: the dimensions of y as a tuple or an array of ints.
x_init_value: (optional) a numpy array of the same shape as ""x""
representing the initial value of x. If x is a list, this should be a list
of numpy arrays. If this is none, the function will pick a random tensor
as the initial value.
delta: (optional) the amount of perturbation.
init_targets: list of targets to run to initialize model params.
extra_feed_dict: dict that allows fixing specified tensor values
during the Jacobian calculation.
Returns:
The maximum error in between the two Jacobians."
8478,compute_gradient,tensorflow/tensorflow/python/ops/gradient_checker_v2.py,295,function,"Computes the theoretical and numeric Jacobian of `f`.
With y = f(x), computes the theoretical and numeric Jacobian dy/dx.
Args:
f: the function.
x: the arguments for the function as a list or tuple of values convertible
to a Tensor.
delta: (optional) perturbation used to compute numeric Jacobian.
Returns:
A pair of lists, where the first is a list of 2-d numpy arrays representing
the theoretical Jacobians for each argument, and the second list is the
numerical ones. Each 2-d array has ""y_size"" rows
and ""x_size"" columns where ""x_size"" is the number of elements in the
corresponding argument and ""y_size"" is the number of elements in f(x).
Raises:
ValueError: If result is empty but the gradient is nonzero.
ValueError: If x is not list, but any other type.
Example:
```python
@tf.function
def test_func(x):
return x*x
theoretical, numerical = tf.test.compute_gradient(test_func, [1.0])
theoretical, numerical
# ((array([[2.]], dtype=float32),), (array([[2.000004]], dtype=float32),))
```"
8479,max_error,tensorflow/tensorflow/python/ops/gradient_checker_v2.py,335,function,"Computes maximum elementwise gap.
Computes the maximum elementwise gap between two lists of tensors of the same
shape.
Args:
grad1: a lists of tensors.
grad2: a lists of tensors with the same shape as grad1.
Returns:
The maximum elementwise gap between the two."
8480,gradients,tensorflow/tensorflow/python/ops/gradients_impl.py,44,function,"Constructs symbolic derivatives of sum of `ys` w.r.t. x in `xs`.
`ys` and `xs` are each a `Tensor` or a list of tensors. `grad_ys`
is a list of `Tensor`, holding the gradients received by the
`ys`. The list must be the same length as `ys`.
`gradients()` adds ops to the graph to output the derivatives of `ys` with
respect to `xs`. It returns a list of `Tensor` of length `len(xs)` where
each tensor is the `sum(dy/dx)` for y in `ys` and for x in `xs`.
`grad_ys` is a list of tensors of the same length as `ys` that holds
the initial gradients for each y in `ys`. When `grad_ys` is None,
we fill in a tensor of '1's of the shape of y for each y in `ys`. A
user can provide their own initial `grad_ys` to compute the
derivatives using a different initial gradient for each y (e.g., if
one wanted to weight the gradient differently for each value in
each y).
`stop_gradients` is a `Tensor` or a list of tensors to be considered constant
with respect to all `xs`. These tensors will not be backpropagated through,
as though they had been explicitly disconnected using `stop_gradient`. Among
other things, this allows computation of partial derivatives as opposed to
total derivatives. For example:
```python
a = tf.constant(0.)
b = 2 * a
g = tf.gradients(a + b, [a, b], stop_gradients=[a, b])
```
Here the partial derivatives `g` evaluate to `[1.0, 1.0]`, compared to the
total derivatives `tf.gradients(a + b, [a, b])`, which take into account the
influence of `a` on `b` and evaluate to `[3.0, 1.0]`. Note that the above is
equivalent to:
```python
a = tf.stop_gradient(tf.constant(0.))
b = tf.stop_gradient(2 * a)
g = tf.gradients(a + b, [a, b])
```
`stop_gradients` provides a way of stopping gradient after the graph has
already been constructed, as compared to `tf.stop_gradient` which is used
during graph construction. When the two approaches are combined,
backpropagation stops at both `tf.stop_gradient` nodes and nodes in
`stop_gradients`, whichever is encountered first.
All integer tensors are considered constant with respect to all `xs`, as if
they were included in `stop_gradients`.
`unconnected_gradients` determines the value returned for each x in xs if it
is unconnected in the graph to ys. By default this is None to safeguard
against errors. Mathematically these gradients are zero which can be requested
using the `'zero'` option. `tf.UnconnectedGradients` provides the
following options and behaviors:
```python
a = tf.ones([1, 2])
b = tf.ones([3, 1])
g1 = tf.gradients([b], [a], unconnected_gradients='none')
sess.run(g1) # [None]
g2 = tf.gradients([b], [a], unconnected_gradients='zero')
sess.run(g2) # [array([[0., 0.]], dtype=float32)]
```
Let us take one practical example which comes during the back propogation
phase. This function is used to evaluate the derivatives of the cost function
with respect to Weights `Ws` and Biases `bs`. Below sample implementation
provides the exaplantion of what it is actually used for :
```python
Ws = tf.constant(0.)
bs = 2 * Ws
cost = Ws + bs # This is just an example. So, please ignore the formulas.
g = tf.gradients(cost, [Ws, bs])
dCost_dW, dCost_db = g
```
Args:
ys: A `Tensor` or list of tensors to be differentiated.
xs: A `Tensor` or list of tensors to be used for differentiation.
grad_ys: Optional. A `Tensor` or list of tensors the same size as
`ys` and holding the gradients computed for each y in `ys`.
name: Optional name to use for grouping all the gradient ops together.
defaults to 'gradients'.
colocate_gradients_with_ops: If True, try colocating gradients with
the corresponding op.
gate_gradients: If True, add a tuple around the gradients returned
for an operations. This avoids some race conditions.
aggregation_method: Specifies the method used to combine gradient terms.
Accepted values are constants defined in the class `AggregationMethod`.
stop_gradients: Optional. A `Tensor` or list of tensors not to differentiate
through.
unconnected_gradients: Optional. Specifies the gradient value returned when
the given input tensors are unconnected. Accepted values are constants
defined in the class `tf.UnconnectedGradients` and the default value is
`none`.
Returns:
A list of `Tensor` of length `len(xs)` where each tensor is the `sum(dy/dx)`
for y in `ys` and for x in `xs`.
Raises:
LookupError: if one of the operations between `x` and `y` does not
have a registered gradient function.
ValueError: if the arguments are invalid.
RuntimeError: if called in Eager mode."
8481,gradients_v2,tensorflow/tensorflow/python/ops/gradients_impl.py,177,function,"Constructs symbolic derivatives of sum of `ys` w.r.t. x in `xs`.
`tf.gradients` is only valid in a graph context. In particular,
it is valid in the context of a `tf.function` wrapper, where code
is executing as a graph.
`ys` and `xs` are each a `Tensor` or a list of tensors. `grad_ys`
is a list of `Tensor`, holding the gradients received by the
`ys`. The list must be the same length as `ys`.
`gradients()` adds ops to the graph to output the derivatives of `ys` with
respect to `xs`. It returns a list of `Tensor` of length `len(xs)` where
each tensor is the `sum(dy/dx)` for y in `ys` and for x in `xs`.
`grad_ys` is a list of tensors of the same length as `ys` that holds
the initial gradients for each y in `ys`. When `grad_ys` is None,
we fill in a tensor of '1's of the shape of y for each y in `ys`. A
user can provide their own initial `grad_ys` to compute the
derivatives using a different initial gradient for each y (e.g., if
one wanted to weight the gradient differently for each value in
each y).
`stop_gradients` is a `Tensor` or a list of tensors to be considered constant
with respect to all `xs`. These tensors will not be backpropagated through,
as though they had been explicitly disconnected using `stop_gradient`. Among
other things, this allows computation of partial derivatives as opposed to
total derivatives. For example:
>>> @tf.function
... def example():
... a = tf.constant(0.)
... b = 2 * a
... return tf.gradients(a + b, [a, b], stop_gradients=[a, b])
>>> example()
[<tf.Tensor: shape=(), dtype=float32, numpy=1.0>,
<tf.Tensor: shape=(), dtype=float32, numpy=1.0>]
Here the partial derivatives `g` evaluate to `[1.0, 1.0]`, compared to the
total derivatives `tf.gradients(a + b, [a, b])`, which take into account the
influence of `a` on `b` and evaluate to `[3.0, 1.0]`. Note that the above is
equivalent to:
>>> @tf.function
... def example():
... a = tf.stop_gradient(tf.constant(0.))
... b = tf.stop_gradient(2 * a)
... return tf.gradients(a + b, [a, b])
>>> example()
[<tf.Tensor: shape=(), dtype=float32, numpy=1.0>,
<tf.Tensor: shape=(), dtype=float32, numpy=1.0>]
`stop_gradients` provides a way of stopping gradient after the graph has
already been constructed, as compared to `tf.stop_gradient` which is used
during graph construction. When the two approaches are combined,
backpropagation stops at both `tf.stop_gradient` nodes and nodes in
`stop_gradients`, whichever is encountered first.
All integer tensors are considered constant with respect to all `xs`, as if
they were included in `stop_gradients`.
`unconnected_gradients` determines the value returned for each x in xs if it
is unconnected in the graph to ys. By default this is None to safeguard
against errors. Mathematically these gradients are zero which can be requested
using the `'zero'` option. `tf.UnconnectedGradients` provides the
following options and behaviors:
>>> @tf.function
... def example(use_zero):
... a = tf.ones([1, 2])
... b = tf.ones([3, 1])
... if use_zero:
... return tf.gradients([b], [a], unconnected_gradients='zero')
... else:
... return tf.gradients([b], [a], unconnected_gradients='none')
>>> example(False)
[None]
>>> example(True)
[<tf.Tensor: shape=(1, 2), dtype=float32, numpy=array([[0., 0.]], ...)>]
Let us take one practical example which comes during the back propogation
phase. This function is used to evaluate the derivatives of the cost function
with respect to Weights `Ws` and Biases `bs`. Below sample implementation
provides the exaplantion of what it is actually used for :
>>> @tf.function
... def example():
... Ws = tf.constant(0.)
... bs = 2 * Ws
... cost = Ws + bs # This is just an example. Please ignore the formulas.
... g = tf.gradients(cost, [Ws, bs])
... dCost_dW, dCost_db = g
... return dCost_dW, dCost_db
>>> example()
(<tf.Tensor: shape=(), dtype=float32, numpy=3.0>,
<tf.Tensor: shape=(), dtype=float32, numpy=1.0>)
Args:
ys: A `Tensor` or list of tensors to be differentiated.
xs: A `Tensor` or list of tensors to be used for differentiation.
grad_ys: Optional. A `Tensor` or list of tensors the same size as
`ys` and holding the gradients computed for each y in `ys`.
name: Optional name to use for grouping all the gradient ops together.
defaults to 'gradients'.
gate_gradients: If True, add a tuple around the gradients returned
for an operations. This avoids some race conditions.
aggregation_method: Specifies the method used to combine gradient terms.
Accepted values are constants defined in the class `AggregationMethod`.
stop_gradients: Optional. A `Tensor` or list of tensors not to differentiate
through.
unconnected_gradients: Optional. Specifies the gradient value returned when
the given input tensors are unconnected. Accepted values are constants
defined in the class `tf.UnconnectedGradients` and the default value is
`none`.
Returns:
A list of `Tensor` of length `len(xs)` where each tensor is the `sum(dy/dx)`
for y in `ys` and for x in `xs`.
Raises:
LookupError: if one of the operations between `x` and `y` does not
have a registered gradient function.
ValueError: if the arguments are invalid.
RuntimeError: if called in Eager mode."
8482,hessians,tensorflow/tensorflow/python/ops/gradients_impl.py,377,function,"Constructs the Hessian of sum of `ys` with respect to `x` in `xs`.
`hessians()` adds ops to the graph to output the Hessian matrix of `ys`
with respect to `xs`. It returns a list of `Tensor` of length `len(xs)`
where each tensor is the Hessian of `sum(ys)`.
The Hessian is a matrix of second-order partial derivatives of a scalar
tensor (see https://en.wikipedia.org/wiki/Hessian_matrix for more details).
Args:
ys: A `Tensor` or list of tensors to be differentiated.
xs: A `Tensor` or list of tensors to be used for differentiation.
name: Optional name to use for grouping all the gradient ops together.
defaults to 'hessians'.
colocate_gradients_with_ops: See `gradients()` documentation for details.
gate_gradients: See `gradients()` documentation for details.
aggregation_method: See `gradients()` documentation for details.
Returns:
A list of Hessian matrices of `sum(ys)` for each `x` in `xs`.
Raises:
LookupError: if one of the operations between `xs` and `ys` does not
have a registered gradient function."
8483,HessiansV2,tensorflow/tensorflow/python/ops/gradients_impl.py,444,function,"Constructs the Hessian of sum of `ys` with respect to `x` in `xs`.
`hessians()` adds ops to the graph to output the Hessian matrix of `ys`
with respect to `xs`. It returns a list of `Tensor` of length `len(xs)`
where each tensor is the Hessian of `sum(ys)`.
The Hessian is a matrix of second-order partial derivatives of a scalar
tensor (see https://en.wikipedia.org/wiki/Hessian_matrix for more details).
Args:
ys: A `Tensor` or list of tensors to be differentiated.
xs: A `Tensor` or list of tensors to be used for differentiation.
gate_gradients: See `gradients()` documentation for details.
aggregation_method: See `gradients()` documentation for details.
name: Optional name to use for grouping all the gradient ops together.
defaults to 'hessians'.
Returns:
A list of Hessian matrices of `sum(ys)` for each `x` in `xs`.
Raises:
LookupError: if one of the operations between `xs` and `ys` does not
have a registered gradient function."
8484,AggregationMethod,tensorflow/tensorflow/python/ops/gradients_util.py,893,class,"A class listing aggregation methods used to combine gradients.
Computing partial derivatives can require aggregating gradient
contributions. This class lists the various methods that can
be used to combine gradients in the graph.
The following aggregation methods are part of the stable API for
aggregating gradients:
* `ADD_N`: All of the gradient terms are summed as part of one
operation using the ""AddN"" op (see `tf.add_n`). This
method has the property that all gradients must be ready and
buffered separately in memory before any aggregation is performed.
* `DEFAULT`: The system-chosen default aggregation method.
The following aggregation methods are experimental and may not
be supported in future releases:
* `EXPERIMENTAL_TREE`: Gradient terms are summed in pairs using
using the ""AddN"" op. This method of summing gradients may reduce
performance, but it can improve memory utilization because the
gradients can be released earlier."
8485,histogram_fixed_width_bins,tensorflow/tensorflow/python/ops/histogram_ops.py,35,function,"Bins the given values for use in a histogram.
Given the tensor `values`, this operation returns a rank 1 `Tensor`
representing the indices of a histogram into which each element
of `values` would be binned. The bins are equal width and
determined by the arguments `value_range` and `nbins`.
Args:
values: Numeric `Tensor`.
value_range: Shape [2] `Tensor` of same `dtype` as `values`.
values <= value_range[0] will be mapped to hist[0],
values >= value_range[1] will be mapped to hist[-1].
nbins: Scalar `int32 Tensor`. Number of histogram bins.
dtype: dtype for returned histogram.
name: A name for this operation (defaults to 'histogram_fixed_width').
Returns:
A `Tensor` holding the indices of the binned values whose shape matches
`values`.
Raises:
TypeError: If any unsupported dtype is provided.
tf.errors.InvalidArgumentError: If value_range does not
satisfy value_range[0] < value_range[1].
Examples:
>>> # Bins will be: (-inf, 1), [1, 2), [2, 3), [3, 4), [4, inf)
...
>>> nbins = 5
>>> value_range = [0.0, 5.0]
>>> new_values = [-1.0, 0.0, 1.5, 2.0, 5.0, 15]
>>> indices = tf.histogram_fixed_width_bins(new_values, value_range, nbins=5)
>>> indices.numpy()
array([0, 0, 1, 2, 4, 4], dtype=int32)"
8486,histogram_fixed_width,tensorflow/tensorflow/python/ops/histogram_ops.py,104,function,"Return histogram of values.
Given the tensor `values`, this operation returns a rank 1 histogram counting
the number of entries in `values` that fell into every bin. The bins are
equal width and determined by the arguments `value_range` and `nbins`.
Args:
values: Numeric `Tensor`.
value_range: Shape [2] `Tensor` of same `dtype` as `values`.
values <= value_range[0] will be mapped to hist[0],
values >= value_range[1] will be mapped to hist[-1].
nbins: Scalar `int32 Tensor`. Number of histogram bins.
dtype: dtype for returned histogram.
name: A name for this operation (defaults to 'histogram_fixed_width').
Returns:
A 1-D `Tensor` holding histogram of values.
Raises:
TypeError: If any unsupported dtype is provided.
tf.errors.InvalidArgumentError: If value_range does not
satisfy value_range[0] < value_range[1].
Examples:
>>> # Bins will be: (-inf, 1), [1, 2), [2, 3), [3, 4), [4, inf)
...
>>> nbins = 5
>>> value_range = [0.0, 5.0]
>>> new_values = [-1.0, 0.0, 1.5, 2.0, 5.0, 15]
>>> hist = tf.histogram_fixed_width(new_values, value_range, nbins=5)
>>> hist.numpy()
array([2, 1, 1, 0, 2], dtype=int32)"
8487,BinValuesFixedWidth,tensorflow/tensorflow/python/ops/histogram_ops_test.py,31,class,
8488,flat_transforms_to_matrices,tensorflow/tensorflow/python/ops/image_ops.py,177,function,"Converts `tf.contrib.image` projective transforms to affine matrices.
Note that the output matrices map output coordinates to input coordinates. For
the forward transformation matrix, call `tf.linalg.inv` on the result.
Args:
transforms: Vector of length 8, or batches of transforms with shape `(N,
8)`.
Returns:
3D tensor of matrices with shape `(N, 3, 3)`. The output matrices map the
*output coordinates* (in homogeneous coordinates) of each transform to the
corresponding *input coordinates*.
Raises:
ValueError: If `transforms` have an invalid shape."
8489,matrices_to_flat_transforms,tensorflow/tensorflow/python/ops/image_ops.py,209,function,"Converts affine matrices to `tf.contrib.image` projective transforms.
Note that we expect matrices that map output coordinates to input coordinates.
To convert forward transformation matrices, call `tf.linalg.inv` on the
matrices and use the result here.
Args:
transform_matrices: One or more affine transformation matrices, for the
reverse transformation in homogeneous coordinates. Shape `(3, 3)` or `(N,
3, 3)`.
Returns:
2D tensor of flat transforms with shape `(N, 8)`, which may be passed into
`tf.contrib.image.transform`.
Raises:
ValueError: If `transform_matrices` have an invalid shape."
8490,fix_image_flip_shape,tensorflow/tensorflow/python/ops/image_ops_impl.py,310,function,"Set the shape to 3 dimensional if we don't know anything else.
Args:
image: original image size
result: flipped or transformed image
Returns:
An image whose shape is at least (None, None, None)."
8491,random_flip_up_down,tensorflow/tensorflow/python/ops/image_ops_impl.py,331,function,"Randomly flips an image vertically (upside down).
With a 1 in 2 chance, outputs the contents of `image` flipped along the first
dimension, which is `height`. Otherwise, output the image as-is.
When passing a batch of images, each image will be randomly flipped
independent of other images.
Example usage:
>>> import numpy as np
>>> image = np.array([[[1], [2]], [[3], [4]]])
>>> tf.image.random_flip_up_down(image, 3).numpy().tolist()
[[[3], [4]], [[1], [2]]]
Randomly flip multiple images.
>>> images = np.array(
... [
... [[[1], [2]], [[3], [4]]],
... [[[5], [6]], [[7], [8]]]
... ])
>>> tf.image.random_flip_up_down(images, 4).numpy().tolist()
[[[[3], [4]], [[1], [2]]], [[[5], [6]], [[7], [8]]]]
Args:
image: 4-D Tensor of shape `[batch, height, width, channels]` or 3-D Tensor
of shape `[height, width, channels]`.
seed: A Python integer. Used to create a random seed. See
`tf.compat.v1.set_random_seed` for behavior.
Returns:
A tensor of the same type and shape as `image`.
Raises:
ValueError: if the shape of `image` not supported."
8492,random_flip_left_right,tensorflow/tensorflow/python/ops/image_ops_impl.py,372,function,"Randomly flip an image horizontally (left to right).
With a 1 in 2 chance, outputs the contents of `image` flipped along the
second dimension, which is `width`. Otherwise output the image as-is.
When passing a batch of images, each image will be randomly flipped
independent of other images.
Example usage:
>>> import numpy as np
>>> image = np.array([[[1], [2]], [[3], [4]]])
>>> tf.image.random_flip_left_right(image, 5).numpy().tolist()
[[[2], [1]], [[4], [3]]]
Randomly flip multiple images.
>>> images = np.array(
... [
... [[[1], [2]], [[3], [4]]],
... [[[5], [6]], [[7], [8]]]
... ])
>>> tf.image.random_flip_left_right(images, 6).numpy().tolist()
[[[[2], [1]], [[4], [3]]], [[[5], [6]], [[7], [8]]]]
Args:
image: 4-D Tensor of shape `[batch, height, width, channels]` or 3-D Tensor
of shape `[height, width, channels]`.
seed: A Python integer. Used to create a random seed. See
`tf.compat.v1.set_random_seed` for behavior.
Returns:
A tensor of the same type and shape as `image`.
Raises:
ValueError: if the shape of `image` not supported."
8493,flip_left_right,tensorflow/tensorflow/python/ops/image_ops_impl.py,472,function,"Flip an image horizontally (left to right).
Outputs the contents of `image` flipped along the width dimension.
See also `reverse()`.
Usage Example:
>>> x = [[[1.0, 2.0, 3.0],
... [4.0, 5.0, 6.0]],
... [[7.0, 8.0, 9.0],
... [10.0, 11.0, 12.0]]]
>>> tf.image.flip_left_right(x)
<tf.Tensor: shape=(2, 2, 3), dtype=float32, numpy=
array([[[ 4., 5., 6.],
[ 1., 2., 3.]],
[[10., 11., 12.],
[ 7., 8., 9.]]], dtype=float32)>
Args:
image: 4-D Tensor of shape `[batch, height, width, channels]` or 3-D Tensor
of shape `[height, width, channels]`.
Returns:
A tensor of the same type and shape as `image`.
Raises:
ValueError: if the shape of `image` not supported."
8494,flip_up_down,tensorflow/tensorflow/python/ops/image_ops_impl.py,507,function,"Flip an image vertically (upside down).
Outputs the contents of `image` flipped along the height dimension.
See also `reverse()`.
Usage Example:
>>> x = [[[1.0, 2.0, 3.0],
... [4.0, 5.0, 6.0]],
... [[7.0, 8.0, 9.0],
... [10.0, 11.0, 12.0]]]
>>> tf.image.flip_up_down(x)
<tf.Tensor: shape=(2, 2, 3), dtype=float32, numpy=
array([[[ 7., 8., 9.],
[10., 11., 12.]],
[[ 1., 2., 3.],
[ 4., 5., 6.]]], dtype=float32)>
Args:
image: 4-D Tensor of shape `[batch, height, width, channels]` or 3-D Tensor
of shape `[height, width, channels]`.
Returns:
A `Tensor` of the same type and shape as `image`.
Raises:
ValueError: if the shape of `image` not supported."
8495,rot90,tensorflow/tensorflow/python/ops/image_ops_impl.py,584,function,"Rotate image(s) counter-clockwise by 90 degrees.
For example:
>>> a=tf.constant([[[1],[2]],
... [[3],[4]]])
>>> # rotating `a` counter clockwise by 90 degrees
>>> a_rot=tf.image.rot90(a)
>>> print(a_rot[...,0].numpy())
[[2 4]
[1 3]]
>>> # rotating `a` counter clockwise by 270 degrees
>>> a_rot=tf.image.rot90(a, k=3)
>>> print(a_rot[...,0].numpy())
[[3 1]
[4 2]]
Args:
image: 4-D Tensor of shape `[batch, height, width, channels]` or 3-D Tensor
of shape `[height, width, channels]`.
k: A scalar integer. The number of times the image is rotated by 90 degrees.
name: A name for this operation (optional).
Returns:
A rotated tensor of the same type and shape as `image`.
Raises:
ValueError: if the shape of `image` not supported."
8496,transpose,tensorflow/tensorflow/python/ops/image_ops_impl.py,707,function,"Transpose image(s) by swapping the height and width dimension.
Usage Example:
>>> x = [[[1.0, 2.0, 3.0],
... [4.0, 5.0, 6.0]],
... [[7.0, 8.0, 9.0],
... [10.0, 11.0, 12.0]]]
>>> tf.image.transpose(x)
<tf.Tensor: shape=(2, 2, 3), dtype=float32, numpy=
array([[[ 1., 2., 3.],
[ 7., 8., 9.]],
[[ 4., 5., 6.],
[10., 11., 12.]]], dtype=float32)>
Args:
image: 4-D Tensor of shape `[batch, height, width, channels]` or 3-D Tensor
of shape `[height, width, channels]`.
name: A name for this operation (optional).
Returns:
If `image` was 4-D, a 4-D float Tensor of shape
`[batch, width, height, channels]`
If `image` was 3-D, a 3-D float Tensor of shape
`[width, height, channels]`
Raises:
ValueError: if the shape of `image` not supported.
Usage Example:
>>> image = [[[1, 2], [3, 4]],
... [[5, 6], [7, 8]],
... [[9, 10], [11, 12]]]
>>> image = tf.constant(image)
>>> tf.image.transpose(image)
<tf.Tensor: shape=(2, 3, 2), dtype=int32, numpy=
array([[[ 1, 2],
[ 5, 6],
[ 9, 10]],
[[ 3, 4],
[ 7, 8],
[11, 12]]], dtype=int32)>"
8497,central_crop,tensorflow/tensorflow/python/ops/image_ops_impl.py,777,function,"Crop the central region of the image(s).
Remove the outer parts of an image but retain the central region of the image
along each dimension. If we specify central_fraction = 0.5, this function
returns the region marked with ""X"" in the below diagram.
--------
| |
| XXXX |
| XXXX |
| | where ""X"" is the central 50% of the image.
--------
This function works on either a single image (`image` is a 3-D Tensor), or a
batch of images (`image` is a 4-D Tensor).
Usage Example:
>>> x = [[[1.0, 2.0, 3.0],
... [4.0, 5.0, 6.0],
... [7.0, 8.0, 9.0],
... [10.0, 11.0, 12.0]],
... [[13.0, 14.0, 15.0],
... [16.0, 17.0, 18.0],
... [19.0, 20.0, 21.0],
... [22.0, 23.0, 24.0]],
... [[25.0, 26.0, 27.0],
... [28.0, 29.0, 30.0],
... [31.0, 32.0, 33.0],
... [34.0, 35.0, 36.0]],
... [[37.0, 38.0, 39.0],
... [40.0, 41.0, 42.0],
... [43.0, 44.0, 45.0],
... [46.0, 47.0, 48.0]]]
>>> tf.image.central_crop(x, 0.5)
<tf.Tensor: shape=(2, 2, 3), dtype=float32, numpy=
array([[[16., 17., 18.],
[19., 20., 21.]],
[[28., 29., 30.],
[31., 32., 33.]]], dtype=float32)>
Args:
image: Either a 3-D float Tensor of shape [height, width, depth], or a 4-D
Tensor of shape [batch_size, height, width, depth].
central_fraction: float (0, 1], fraction of size to crop
Raises:
ValueError: if central_crop_fraction is not within (0, 1].
Returns:
3-D / 4-D float Tensor, as per the input."
8498,pad_to_bounding_box,tensorflow/tensorflow/python/ops/image_ops_impl.py,910,function,"Pad `image` with zeros to the specified `height` and `width`.
Adds `offset_height` rows of zeros on top, `offset_width` columns of
zeros on the left, and then pads the image on the bottom and right
with zeros until it has dimensions `target_height`, `target_width`.
This op does nothing if `offset_*` is zero and the image already has size
`target_height` by `target_width`.
Usage Example:
>>> x = [[[1., 2., 3.],
... [4., 5., 6.]],
... [[7., 8., 9.],
... [10., 11., 12.]]]
>>> padded_image = tf.image.pad_to_bounding_box(x, 1, 1, 4, 4)
>>> padded_image
<tf.Tensor: shape=(4, 4, 3), dtype=float32, numpy=
array([[[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.]],
[[ 0., 0., 0.],
[ 1., 2., 3.],
[ 4., 5., 6.],
[ 0., 0., 0.]],
[[ 0., 0., 0.],
[ 7., 8., 9.],
[10., 11., 12.],
[ 0., 0., 0.]],
[[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.]]], dtype=float32)>
Args:
image: 4-D Tensor of shape `[batch, height, width, channels]` or 3-D Tensor
of shape `[height, width, channels]`.
offset_height: Number of rows of zeros to add on top.
offset_width: Number of columns of zeros to add on the left.
target_height: Height of output image.
target_width: Width of output image.
Returns:
If `image` was 4-D, a 4-D float Tensor of shape
`[batch, target_height, target_width, channels]`
If `image` was 3-D, a 3-D float Tensor of shape
`[target_height, target_width, channels]`
Raises:
ValueError: If the shape of `image` is incompatible with the `offset_*` or
`target_*` arguments, or either `offset_height` or `offset_width` is
negative."
8499,crop_to_bounding_box,tensorflow/tensorflow/python/ops/image_ops_impl.py,1022,function,"Crops an image to a specified bounding box.
This op cuts a rectangular part out of `image`. The top-left corner of the
returned image is at `offset_height, offset_width` in `image`, and its
lower-right corner is at
`offset_height + target_height, offset_width + target_width`.
Args:
image: 4-D Tensor of shape `[batch, height, width, channels]` or 3-D Tensor
of shape `[height, width, channels]`.
offset_height: Vertical coordinate of the top-left corner of the result in
the input.
offset_width: Horizontal coordinate of the top-left corner of the result in
the input.
target_height: Height of the result.
target_width: Width of the result.
Returns:
If `image` was 4-D, a 4-D float Tensor of shape
`[batch, target_height, target_width, channels]`
If `image` was 3-D, a 3-D float Tensor of shape
`[target_height, target_width, channels]`
Raises:
ValueError: If the shape of `image` is incompatible with the `offset_*` or
`target_*` arguments, or either `offset_height` or `offset_width` is
negative, or either `target_height` or `target_width` is not positive."
8500,resize_image_with_crop_or_pad,tensorflow/tensorflow/python/ops/image_ops_impl.py,1107,function,"Crops and/or pads an image to a target width and height.
Resizes an image to a target width and height by either centrally
cropping the image or padding it evenly with zeros.
If `width` or `height` is greater than the specified `target_width` or
`target_height` respectively, this op centrally crops along that dimension.
If `width` or `height` is smaller than the specified `target_width` or
`target_height` respectively, this op centrally pads with 0 along that
dimension.
Args:
image: 4-D Tensor of shape `[batch, height, width, channels]` or 3-D Tensor
of shape `[height, width, channels]`.
target_height: Target height.
target_width: Target width.
Raises:
ValueError: if `target_height` or `target_width` are zero or negative.
Returns:
Cropped and/or padded image.
If `images` was 4-D, a 4-D float Tensor of shape
`[batch, new_height, new_width, channels]`.
If `images` was 3-D, a 3-D float Tensor of shape
`[new_height, new_width, channels]`."
8501,ResizeMethodV1,tensorflow/tensorflow/python/ops/image_ops_impl.py,1226,class,See `v1.image.resize` for details.
8502,ResizeMethod,tensorflow/tensorflow/python/ops/image_ops_impl.py,1235,class,See `tf.image.resize` for details.
8503,resize_images,tensorflow/tensorflow/python/ops/image_ops_impl.py,1327,function,"Resize `images` to `size` using the specified `method`.
Resized images will be distorted if their original aspect ratio is not
the same as `size`. To avoid distortions see
`tf.image.resize_with_pad` or `tf.image.resize_with_crop_or_pad`.
The `method` can be one of:
* <b>`tf.image.ResizeMethod.BILINEAR`</b>: [Bilinear interpolation.](
https://en.wikipedia.org/wiki/Bilinear_interpolation)
* <b>`tf.image.ResizeMethod.NEAREST_NEIGHBOR`</b>: [
Nearest neighbor interpolation.](
https://en.wikipedia.org/wiki/Nearest-neighbor_interpolation)
* <b>`tf.image.ResizeMethod.BICUBIC`</b>: [Bicubic interpolation.](
https://en.wikipedia.org/wiki/Bicubic_interpolation)
* <b>`tf.image.ResizeMethod.AREA`</b>: Area interpolation.
The return value has the same type as `images` if `method` is
`tf.image.ResizeMethod.NEAREST_NEIGHBOR`. It will also have the same type
as `images` if the size of `images` can be statically determined to be the
same as `size`, because `images` is returned in this case. Otherwise, the
return value has type `float32`.
Args:
images: 4-D Tensor of shape `[batch, height, width, channels]` or 3-D Tensor
of shape `[height, width, channels]`.
size: A 1-D int32 Tensor of 2 elements: `new_height, new_width`. The new
size for the images.
method: ResizeMethod. Defaults to `tf.image.ResizeMethod.BILINEAR`.
align_corners: bool. If True, the centers of the 4 corner pixels of the
input and output tensors are aligned, preserving the values at the corner
pixels. Defaults to `False`.
preserve_aspect_ratio: Whether to preserve the aspect ratio. If this is set,
then `images` will be resized to a size that fits in `size` while
preserving the aspect ratio of the original image. Scales up the image if
`size` is bigger than the current size of the `image`. Defaults to False.
name: A name for this operation (optional).
Raises:
ValueError: if the shape of `images` is incompatible with the
shape arguments to this function
ValueError: if `size` has invalid shape or type.
ValueError: if an unsupported resize method is specified.
Returns:
If `images` was 4-D, a 4-D float Tensor of shape
`[batch, new_height, new_width, channels]`.
If `images` was 3-D, a 3-D float Tensor of shape
`[new_height, new_width, channels]`."
8504,resize_images_v2,tensorflow/tensorflow/python/ops/image_ops_impl.py,1413,function,"Resize `images` to `size` using the specified `method`.
Resized images will be distorted if their original aspect ratio is not
the same as `size`. To avoid distortions see
`tf.image.resize_with_pad`.
>>> image = tf.constant([
... [1,0,0,0,0],
... [0,1,0,0,0],
... [0,0,1,0,0],
... [0,0,0,1,0],
... [0,0,0,0,1],
... ])
>>> # Add ""batch"" and ""channels"" dimensions
>>> image = image[tf.newaxis, ..., tf.newaxis]
>>> image.shape.as_list() # [batch, height, width, channels]
[1, 5, 5, 1]
>>> tf.image.resize(image, [3,5])[0,...,0].numpy()
array([[0.6666667, 0.3333333, 0. , 0. , 0. ],
[0. , 0. , 1. , 0. , 0. ],
[0. , 0. , 0. , 0.3333335, 0.6666665]],
dtype=float32)
It works equally well with a single image instead of a batch of images:
>>> tf.image.resize(image[0], [3,5]).shape.as_list()
[3, 5, 1]
When `antialias` is true, the sampling filter will anti-alias the input image
as well as interpolate. When downsampling an image with [anti-aliasing](
https://en.wikipedia.org/wiki/Spatial_anti-aliasing) the sampling filter
kernel is scaled in order to properly anti-alias the input image signal.
`antialias` has no effect when upsampling an image:
>>> a = tf.image.resize(image, [5,10])
>>> b = tf.image.resize(image, [5,10], antialias=True)
>>> tf.reduce_max(abs(a - b)).numpy()
0.0
The `method` argument expects an item from the `image.ResizeMethod` enum, or
the string equivalent. The options are:
* <b>`bilinear`</b>: [Bilinear interpolation.](
https://en.wikipedia.org/wiki/Bilinear_interpolation) If `antialias` is
true, becomes a hat/tent filter function with radius 1 when downsampling.
* <b>`lanczos3`</b>: [Lanczos kernel](
https://en.wikipedia.org/wiki/Lanczos_resampling) with radius 3.
High-quality practical filter but may have some ringing, especially on
synthetic images.
* <b>`lanczos5`</b>: [Lanczos kernel] (
https://en.wikipedia.org/wiki/Lanczos_resampling) with radius 5.
Very-high-quality filter but may have stronger ringing.
* <b>`bicubic`</b>: [Cubic interpolant](
https://en.wikipedia.org/wiki/Bicubic_interpolation) of Keys. Equivalent to
Catmull-Rom kernel. Reasonably good quality and faster than Lanczos3Kernel,
particularly when upsampling.
* <b>`gaussian`</b>: [Gaussian kernel](
https://en.wikipedia.org/wiki/Gaussian_filter) with radius 3,
sigma = 1.5 / 3.0.
* <b>`nearest`</b>: [Nearest neighbor interpolation.](
https://en.wikipedia.org/wiki/Nearest-neighbor_interpolation)
`antialias` has no effect when used with nearest neighbor interpolation.
* <b>`area`</b>: Anti-aliased resampling with area interpolation.
`antialias` has no effect when used with area interpolation; it
always anti-aliases.
* <b>`mitchellcubic`</b>: Mitchell-Netravali Cubic non-interpolating filter.
For synthetic images (especially those lacking proper prefiltering), less
ringing than Keys cubic kernel but less sharp.
Note: Near image edges the filtering kernel may be partially outside the
image boundaries. For these pixels, only input pixels inside the image will be
included in the filter sum, and the output value will be appropriately
normalized.
The return value has type `float32`, unless the `method` is
`ResizeMethod.NEAREST_NEIGHBOR`, then the return dtype is the dtype
of `images`:
>>> nn = tf.image.resize(image, [5,7], method='nearest')
>>> nn[0,...,0].numpy()
array([[1, 0, 0, 0, 0, 0, 0],
[0, 1, 1, 0, 0, 0, 0],
[0, 0, 0, 1, 0, 0, 0],
[0, 0, 0, 0, 1, 1, 0],
[0, 0, 0, 0, 0, 0, 1]], dtype=int32)
With `preserve_aspect_ratio=True`, the aspect ratio is preserved, so `size`
is the maximum for each dimension:
>>> max_10_20 = tf.image.resize(image, [10,20], preserve_aspect_ratio=True)
>>> max_10_20.shape.as_list()
[1, 10, 10, 1]
Args:
images: 4-D Tensor of shape `[batch, height, width, channels]` or 3-D Tensor
of shape `[height, width, channels]`.
size: A 1-D int32 Tensor of 2 elements: `new_height, new_width`. The new
size for the images.
method: An `image.ResizeMethod`, or string equivalent. Defaults to
`bilinear`.
preserve_aspect_ratio: Whether to preserve the aspect ratio. If this is set,
then `images` will be resized to a size that fits in `size` while
preserving the aspect ratio of the original image. Scales up the image if
`size` is bigger than the current size of the `image`. Defaults to False.
antialias: Whether to use an anti-aliasing filter when downsampling an
image.
name: A name for this operation (optional).
Raises:
ValueError: if the shape of `images` is incompatible with the
shape arguments to this function
ValueError: if `size` has an invalid shape or type.
ValueError: if an unsupported resize method is specified.
Returns:
If `images` was 4-D, a 4-D float Tensor of shape
`[batch, new_height, new_width, channels]`.
If `images` was 3-D, a 3-D float Tensor of shape
`[new_height, new_width, channels]`."
8505,resize_image_with_pad_v1,tensorflow/tensorflow/python/ops/image_ops_impl.py,1667,function,"Resizes and pads an image to a target width and height.
Resizes an image to a target width and height by keeping
the aspect ratio the same without distortion. If the target
dimensions don't match the image dimensions, the image
is resized and then padded with zeroes to match requested
dimensions.
Args:
image: 4-D Tensor of shape `[batch, height, width, channels]` or 3-D Tensor
of shape `[height, width, channels]`.
target_height: Target height.
target_width: Target width.
method: Method to use for resizing image. See `resize_images()`
align_corners: bool. If True, the centers of the 4 corner pixels of the
input and output tensors are aligned, preserving the values at the corner
pixels. Defaults to `False`.
Raises:
ValueError: if `target_height` or `target_width` are zero or negative.
Returns:
Resized and padded image.
If `images` was 4-D, a 4-D float Tensor of shape
`[batch, new_height, new_width, channels]`.
If `images` was 3-D, a 3-D float Tensor of shape
`[new_height, new_width, channels]`."
8506,resize_image_with_pad_v2,tensorflow/tensorflow/python/ops/image_ops_impl.py,1710,function,"Resizes and pads an image to a target width and height.
Resizes an image to a target width and height by keeping
the aspect ratio the same without distortion. If the target
dimensions don't match the image dimensions, the image
is resized and then padded with zeroes to match requested
dimensions.
Args:
image: 4-D Tensor of shape `[batch, height, width, channels]` or 3-D Tensor
of shape `[height, width, channels]`.
target_height: Target height.
target_width: Target width.
method: Method to use for resizing image. See `image.resize()`
antialias: Whether to use anti-aliasing when resizing. See 'image.resize()'.
Raises:
ValueError: if `target_height` or `target_width` are zero or negative.
Returns:
Resized and padded image.
If `images` was 4-D, a 4-D float Tensor of shape
`[batch, new_height, new_width, channels]`.
If `images` was 3-D, a 3-D float Tensor of shape
`[new_height, new_width, channels]`."
8507,per_image_standardization,tensorflow/tensorflow/python/ops/image_ops_impl.py,1751,function,"Linearly scales each image in `image` to have mean 0 and variance 1.
For each 3-D image `x` in `image`, computes `(x - mean) / adjusted_stddev`,
where
- `mean` is the average of all values in `x`
- `adjusted_stddev = max(stddev, 1.0/sqrt(N))` is capped away from 0 to
protect against division by 0 when handling uniform images
- `N` is the number of elements in `x`
- `stddev` is the standard deviation of all values in `x`
Args:
image: An n-D Tensor with at least 3 dimensions, the last 3 of which are the
dimensions of each image.
Returns:
A `Tensor` with the same shape and dtype as `image`.
Raises:
ValueError: if the shape of 'image' is incompatible with this function."
8508,random_brightness,tensorflow/tensorflow/python/ops/image_ops_impl.py,1797,function,"Adjust the brightness of images by a random factor.
Equivalent to `adjust_brightness()` using a `delta` randomly picked in the
interval `[-max_delta, max_delta)`.
Args:
image: An image or images to adjust.
max_delta: float, must be non-negative.
seed: A Python integer. Used to create a random seed. See
`tf.compat.v1.set_random_seed` for behavior.
Usage Example:
>>> x = [[[1.0, 2.0, 3.0],
... [4.0, 5.0, 6.0]],
... [[7.0, 8.0, 9.0],
... [10.0, 11.0, 12.0]]]
>>> tf.image.random_brightness(x, 0.2)
<tf.Tensor: shape=(2, 2, 3), dtype=float32, numpy=...>
Returns:
The brightness-adjusted image(s).
Raises:
ValueError: if `max_delta` is negative."
8509,random_contrast,tensorflow/tensorflow/python/ops/image_ops_impl.py,1833,function,"Adjust the contrast of an image or images by a random factor.
Equivalent to `adjust_contrast()` but uses a `contrast_factor` randomly
picked in the interval `[lower, upper)`.
Args:
image: An image tensor with 3 or more dimensions.
lower: float. Lower bound for the random contrast factor.
upper: float. Upper bound for the random contrast factor.
seed: A Python integer. Used to create a random seed. See
`tf.compat.v1.set_random_seed` for behavior.
Usage Example:
>>> x = [[[1.0, 2.0, 3.0],
... [4.0, 5.0, 6.0]],
... [[7.0, 8.0, 9.0],
... [10.0, 11.0, 12.0]]]
>>> tf.image.random_contrast(x, 0.2, 0.5)
<tf.Tensor: shape=(2, 2, 3), dtype=float32, numpy=...>
Returns:
The contrast-adjusted image(s).
Raises:
ValueError: if `upper <= lower` or if `lower < 0`."
8510,adjust_brightness,tensorflow/tensorflow/python/ops/image_ops_impl.py,1873,function,"Adjust the brightness of RGB or Grayscale images.
This is a convenience method that converts RGB images to float
representation, adjusts their brightness, and then converts them back to the
original data type. If several adjustments are chained, it is advisable to
minimize the number of redundant conversions.
The value `delta` is added to all components of the tensor `image`. `image` is
converted to `float` and scaled appropriately if it is in fixed-point
representation, and `delta` is converted to the same data type. For regular
images, `delta` should be in the range `[0,1)`, as it is added to the image in
floating point representation, where pixel values are in the `[0,1)` range.
Usage Example:
>>> x = [[[1.0, 2.0, 3.0],
... [4.0, 5.0, 6.0]],
... [[7.0, 8.0, 9.0],
... [10.0, 11.0, 12.0]]]
>>> tf.image.adjust_brightness(x, delta=0.1)
<tf.Tensor: shape=(2, 2, 3), dtype=float32, numpy=
array([[[ 1.1, 2.1, 3.1],
[ 4.1, 5.1, 6.1]],
[[ 7.1, 8.1, 9.1],
[10.1, 11.1, 12.1]]], dtype=float32)>
Args:
image: RGB image or images to adjust.
delta: A scalar. Amount to add to the pixel values.
Returns:
A brightness-adjusted tensor of the same shape and type as `image`."
8511,adjust_contrast,tensorflow/tensorflow/python/ops/image_ops_impl.py,1925,function,"Adjust contrast of RGB or grayscale images.
This is a convenience method that converts RGB images to float
representation, adjusts their contrast, and then converts them back to the
original data type. If several adjustments are chained, it is advisable to
minimize the number of redundant conversions.
`images` is a tensor of at least 3 dimensions. The last 3 dimensions are
interpreted as `[height, width, channels]`. The other dimensions only
represent a collection of images, such as `[batch, height, width, channels].`
Contrast is adjusted independently for each channel of each image.
For each channel, this Op computes the mean of the image pixels in the
channel and then adjusts each component `x` of each pixel to
`(x - mean) * contrast_factor + mean`.
Usage Example:
>>> x = [[[1.0, 2.0, 3.0],
... [4.0, 5.0, 6.0]],
... [[7.0, 8.0, 9.0],
... [10.0, 11.0, 12.0]]]
>>> tf.image.adjust_contrast(x, 2)
<tf.Tensor: shape=(2, 2, 3), dtype=float32, numpy=
array([[[-3.5, -2.5, -1.5],
[ 2.5, 3.5, 4.5]],
[[ 8.5, 9.5, 10.5],
[14.5, 15.5, 16.5]]], dtype=float32)>
Args:
images: Images to adjust. At least 3-D.
contrast_factor: A float multiplier for adjusting contrast.
Returns:
The contrast-adjusted image or images."
8512,adjust_gamma,tensorflow/tensorflow/python/ops/image_ops_impl.py,1982,function,"Performs [Gamma Correction](http://en.wikipedia.org/wiki/Gamma_correction).
on the input image.
Also known as Power Law Transform. This function converts the
input images at first to float representation, then transforms them
pixelwise according to the equation `Out = gain * In**gamma`,
and then converts the back to the original data type.
Usage Example:
>>> x = [[[1.0, 2.0, 3.0],
... [4.0, 5.0, 6.0]],
... [[7.0, 8.0, 9.0],
... [10.0, 11.0, 12.0]]]
>>> tf.image.adjust_gamma(x, 0.2)
<tf.Tensor: shape=(2, 2, 3), dtype=float32, numpy=
array([[[1. , 1.1486983, 1.2457309],
[1.319508 , 1.3797297, 1.4309691]],
[[1.4757731, 1.5157166, 1.5518456],
[1.5848932, 1.6153942, 1.6437519]]], dtype=float32)>
Args:
image : RGB image or images to adjust.
gamma : A scalar or tensor. Non-negative real number.
gain : A scalar or tensor. The constant multiplier.
Returns:
A Tensor. A Gamma-adjusted tensor of the same shape and type as `image`.
Raises:
ValueError: If gamma is negative.
Notes:
For gamma greater than 1, the histogram will shift towards left and
the output image will be darker than the input image.
For gamma less than 1, the histogram will shift towards right and
the output image will be brighter than the input image.
References:
[Wikipedia](http://en.wikipedia.org/wiki/Gamma_correction)"
8513,convert_image_dtype,tensorflow/tensorflow/python/ops/image_ops_impl.py,2047,function,"Convert `image` to `dtype`, scaling its values if needed.
Images that are represented using floating point values are expected to have
values in the range [0,1). Image data stored in integer data types are
expected to have values in the range `[0,MAX]`, where `MAX` is the largest
positive representable number for the data type.
This op converts between data types, scaling the values appropriately before
casting.
Note that converting from floating point inputs to integer types may lead to
over/underflow problems. Set saturate to `True` to avoid such problem in
problematic conversions. If enabled, saturation will clip the output into the
allowed range before performing a potentially dangerous cast (and only before
performing such a cast, i.e., when casting from a floating point to an integer
type, and when casting from a signed to an unsigned type; `saturate` has no
effect on casts between floats, or on casts that increase the type's range).
Usage Example:
>>> x = [[[1.0, 2.0, 3.0],
... [4.0, 5.0, 6.0]],
... [[7.0, 8.0, 9.0],
... [10.0, 11.0, 12.0]]]
>>> tf.image.convert_image_dtype(x, dtype=tf.float16, saturate=False)
<tf.Tensor: shape=(2, 2, 3), dtype=float16, numpy=
array([[[ 1., 2., 3.],
[ 4., 5., 6.]],
[[ 7., 8., 9.],
[10., 11., 12.]]], dtype=float16)>
Args:
image: An image.
dtype: A `DType` to convert `image` to.
saturate: If `True`, clip the input before casting (if necessary).
name: A name for this operation (optional).
Returns:
`image`, converted to `dtype`.
Raises:
AttributeError: Raises an attribute error when dtype is neither
float nor integer"
8514,rgb_to_grayscale,tensorflow/tensorflow/python/ops/image_ops_impl.py,2147,function,"Converts one or more images from RGB to Grayscale.
Outputs a tensor of the same `DType` and rank as `images`. The size of the
last dimension of the output is 1, containing the Grayscale value of the
pixels.
>>> original = tf.constant([[[1.0, 2.0, 3.0]]])
>>> converted = tf.image.rgb_to_grayscale(original)
>>> print(converted.numpy())
[[[1.81...]]]
Args:
images: The RGB tensor to convert. The last dimension must have size 3 and
should contain RGB values.
name: A name for the operation (optional).
Returns:
The converted grayscale image(s)."
8515,grayscale_to_rgb,tensorflow/tensorflow/python/ops/image_ops_impl.py,2183,function,"Converts one or more images from Grayscale to RGB.
Outputs a tensor of the same `DType` and rank as `images`. The size of the
last dimension of the output is 3, containing the RGB value of the pixels.
The input images' last dimension must be size 1.
>>> original = tf.constant([[[1.0], [2.0], [3.0]]])
>>> converted = tf.image.grayscale_to_rgb(original)
>>> print(converted.numpy())
[[[1. 1. 1.]
[2. 2. 2.]
[3. 3. 3.]]]
Args:
images: The Grayscale tensor to convert. The last dimension must be size 1.
name: A name for the operation (optional).
Returns:
The converted grayscale image(s)."
8516,random_hue,tensorflow/tensorflow/python/ops/image_ops_impl.py,2220,function,"Adjust the hue of RGB images by a random factor.
Equivalent to `adjust_hue()` but uses a `delta` randomly
picked in the interval `[-max_delta, max_delta)`.
`max_delta` must be in the interval `[0, 0.5]`.
Usage Example:
>>> x = [[[1.0, 2.0, 3.0],
... [4.0, 5.0, 6.0]],
... [[7.0, 8.0, 9.0],
... [10.0, 11.0, 12.0]]]
>>> tf.image.random_hue(x, 0.2)
<tf.Tensor: shape=(2, 2, 3), dtype=float32, numpy=...>
Args:
image: RGB image or images. The size of the last dimension must be 3.
max_delta: float. The maximum value for the random delta.
seed: An operation-specific seed. It will be used in conjunction with the
graph-level seed to determine the real seeds that will be used in this
operation. Please see the documentation of set_random_seed for its
interaction with the graph-level random seed.
Returns:
Adjusted image(s), same shape and DType as `image`.
Raises:
ValueError: if `max_delta` is invalid."
8517,adjust_hue,tensorflow/tensorflow/python/ops/image_ops_impl.py,2263,function,"Adjust hue of RGB images.
This is a convenience method that converts an RGB image to float
representation, converts it to HSV, adds an offset to the
hue channel, converts back to RGB and then back to the original
data type. If several adjustments are chained it is advisable to minimize
the number of redundant conversions.
`image` is an RGB image. The image hue is adjusted by converting the
image(s) to HSV and rotating the hue channel (H) by
`delta`. The image is then converted back to RGB.
`delta` must be in the interval `[-1, 1]`.
Usage Example:
>>> x = [[[1.0, 2.0, 3.0],
... [4.0, 5.0, 6.0]],
... [[7.0, 8.0, 9.0],
... [10.0, 11.0, 12.0]]]
>>> tf.image.adjust_hue(x, 0.2)
<tf.Tensor: shape=(2, 2, 3), dtype=float32, numpy=
array([[[ 2.3999996, 1. , 3. ],
[ 5.3999996, 4. , 6. ]],
[[ 8.4 , 7. , 9. ],
[11.4 , 10. , 12. ]]], dtype=float32)>
Args:
image: RGB image or images. The size of the last dimension must be 3.
delta: float. How much to add to the hue channel.
name: A name for this operation (optional).
Returns:
Adjusted image(s), same shape and DType as `image`.
Usage Example:
>>> image = [[[1, 2, 3], [4, 5, 6]],
... [[7, 8, 9], [10, 11, 12]],
... [[13, 14, 15], [16, 17, 18]]]
>>> image = tf.constant(image)
>>> tf.image.adjust_hue(image, 0.2)
<tf.Tensor: shape=(3, 2, 3), dtype=int32, numpy=
array([[[ 2, 1, 3],
[ 5, 4, 6]],
[[ 8, 7, 9],
[11, 10, 12]],
[[14, 13, 15],
[17, 16, 18]]], dtype=int32)>"
8518,random_jpeg_quality,tensorflow/tensorflow/python/ops/image_ops_impl.py,2331,function,"Randomly changes jpeg encoding quality for inducing jpeg noise.
`min_jpeg_quality` must be in the interval `[0, 100]` and less than
`max_jpeg_quality`.
`max_jpeg_quality` must be in the interval `[0, 100]`.
Usage Example:
>>> x = [[[1.0, 2.0, 3.0],
... [4.0, 5.0, 6.0]],
... [[7.0, 8.0, 9.0],
... [10.0, 11.0, 12.0]]]
>>> tf.image.random_jpeg_quality(x, 75, 95)
<tf.Tensor: shape=(2, 2, 3), dtype=float32, numpy=...>
Args:
image: 3D image. Size of the last dimension must be 1 or 3.
min_jpeg_quality: Minimum jpeg encoding quality to use.
max_jpeg_quality: Maximum jpeg encoding quality to use.
seed: An operation-specific seed. It will be used in conjunction with the
graph-level seed to determine the real seeds that will be used in this
operation. Please see the documentation of set_random_seed for its
interaction with the graph-level random seed.
Returns:
Adjusted image(s), same shape and DType as `image`.
Raises:
ValueError: if `min_jpeg_quality` or `max_jpeg_quality` is invalid."
8519,adjust_jpeg_quality,tensorflow/tensorflow/python/ops/image_ops_impl.py,2379,function,"Adjust jpeg encoding quality of an image.
This is a convenience method that converts an image to uint8 representation,
encodes it to jpeg with `jpeg_quality`, decodes it, and then converts back
to the original data type.
`jpeg_quality` must be in the interval `[0, 100]`.
Usage Example:
>>> x = [[[1.0, 2.0, 3.0],
... [4.0, 5.0, 6.0]],
... [[7.0, 8.0, 9.0],
... [10.0, 11.0, 12.0]]]
>>> tf.image.adjust_jpeg_quality(x, 75)
<tf.Tensor: shape=(2, 2, 3), dtype=float32, numpy=
array([[[1., 1., 1.],
[1., 1., 1.]],
[[1., 1., 1.],
[1., 1., 1.]]], dtype=float32)>
Args:
image: 3D image. The size of the last dimension must be None, 1 or 3.
jpeg_quality: Python int or Tensor of type int32. jpeg encoding quality.
name: A name for this operation (optional).
Returns:
Adjusted image, same shape and DType as `image`.
Raises:
InvalidArgumentError: quality must be in [0,100]
InvalidArgumentError: image must have 1 or 3 channels"
8520,random_saturation,tensorflow/tensorflow/python/ops/image_ops_impl.py,2430,function,"Adjust the saturation of RGB images by a random factor.
Equivalent to `adjust_saturation()` but uses a `saturation_factor` randomly
picked in the interval `[lower, upper)`.
Usage Example:
>>> x = [[[1.0, 2.0, 3.0],
... [4.0, 5.0, 6.0]],
... [[7.0, 8.0, 9.0],
... [10.0, 11.0, 12.0]]]
>>> tf.image.random_saturation(x, 5, 10)
<tf.Tensor: shape=(2, 2, 3), dtype=float32, numpy=
array([[[ 0. , 1.5, 3. ],
[ 0. , 3. , 6. ]],
[[ 0. , 4.5, 9. ],
[ 0. , 6. , 12. ]]], dtype=float32)>
Args:
image: RGB image or images. The size of the last dimension must be 3.
lower: float. Lower bound for the random saturation factor.
upper: float. Upper bound for the random saturation factor.
seed: An operation-specific seed. It will be used in conjunction with the
graph-level seed to determine the real seeds that will be used in this
operation. Please see the documentation of set_random_seed for its
interaction with the graph-level random seed.
Returns:
Adjusted image(s), same shape and DType as `image`.
Raises:
ValueError: if `upper <= lower` or if `lower < 0`."
8521,adjust_saturation,tensorflow/tensorflow/python/ops/image_ops_impl.py,2476,function,"Adjust saturation of RGB images.
This is a convenience method that converts RGB images to float
representation, converts them to HSV, adds an offset to the
saturation channel, converts back to RGB and then back to the original
data type. If several adjustments are chained it is advisable to minimize
the number of redundant conversions.
`image` is an RGB image or images. The image saturation is adjusted by
converting the images to HSV and multiplying the saturation (S) channel by
`saturation_factor` and clipping. The images are then converted back to RGB.
Usage Example:
>>> x = [[[1.0, 2.0, 3.0],
... [4.0, 5.0, 6.0]],
... [[7.0, 8.0, 9.0],
... [10.0, 11.0, 12.0]]]
>>> tf.image.adjust_saturation(x, 0.5)
<tf.Tensor: shape=(2, 2, 3), dtype=float32, numpy=
array([[[ 2. , 2.5, 3. ],
[ 5. , 5.5, 6. ]],
[[ 8. , 8.5, 9. ],
[11. , 11.5, 12. ]]], dtype=float32)>
Args:
image: RGB image or images. The size of the last dimension must be 3.
saturation_factor: float. Factor to multiply the saturation by.
name: A name for this operation (optional).
Returns:
Adjusted image(s), same shape and DType as `image`.
Raises:
InvalidArgumentError: input must have 3 channels"
8522,is_jpeg,tensorflow/tensorflow/python/ops/image_ops_impl.py,2528,function,"Convenience function to check if the 'contents' encodes a JPEG image.
Args:
contents: 0-D `string`. The encoded image bytes.
name: A name for the operation (optional)
Returns:
A scalar boolean tensor indicating if 'contents' may be a JPEG image.
is_jpeg is susceptible to false positives."
8523,encode_png,tensorflow/tensorflow/python/ops/image_ops_impl.py,2604,function,"PNG-encode an image.
`image` is a 3-D uint8 or uint16 Tensor of shape `[height, width, channels]`
where `channels` is:
* 1: for grayscale.
* 2: for grayscale + alpha.
* 3: for RGB.
* 4: for RGBA.
The ZLIB compression level, `compression`, can be -1 for the PNG-encoder
default or a value from 0 to 9. 9 is the highest compression level,
generating the smallest output, but is slower.
Args:
image: A `Tensor`. Must be one of the following types: `uint8`, `uint16`.
3-D with shape `[height, width, channels]`.
compression: An optional `int`. Defaults to `-1`. Compression level.
name: A name for the operation (optional).
Returns:
A `Tensor` of type `string`."
8524,decode_image,tensorflow/tensorflow/python/ops/image_ops_impl.py,2637,function,"Function for `decode_bmp`, `decode_gif`, `decode_jpeg`, and `decode_png`.
Detects whether an image is a BMP, GIF, JPEG, or PNG, and performs the
appropriate operation to convert the input bytes `string` into a `Tensor`
of type `dtype`.
Note: `decode_gif` returns a 4-D array `[num_frames, height, width, 3]`, as
opposed to `decode_bmp`, `decode_jpeg` and `decode_png`, which return 3-D
arrays `[height, width, num_channels]`. Make sure to take this into account
when constructing your graph if you are intermixing GIF files with BMP, JPEG,
and/or PNG files. Alternately, set the `expand_animations` argument of this
function to `False`, in which case the op will return 3-dimensional tensors
and will truncate animated GIF files to the first frame.
Args:
contents: 0-D `string`. The encoded image bytes.
channels: An optional `int`. Defaults to `0`. Number of color channels for
the decoded image.
dtype: The desired DType of the returned `Tensor`.
name: A name for the operation (optional)
expand_animations: Controls the shape of the returned op's output. If
`True`, the returned op will produce a 3-D tensor for PNG, JPEG, and BMP
files; and a 4-D tensor for all GIFs, whether animated or not. If,
`False`, the returned op will produce a 3-D tensor for all file types and
will truncate animated GIFs to the first frame.
Returns:
`Tensor` with type `dtype` and a 3- or 4-dimensional shape, depending on
the file type and the value of the `expand_animations` parameter.
Raises:
ValueError: On incorrect number of channels."
8525,total_variation,tensorflow/tensorflow/python/ops/image_ops_impl.py,2770,function,"Calculate and return the total variation for one or more images.
The total variation is the sum of the absolute differences for neighboring
pixel-values in the input images. This measures how much noise is in the
images.
This can be used as a loss-function during optimization so as to suppress
noise in images. If you have a batch of images, then you should calculate
the scalar loss-value as the sum:
`loss = tf.reduce_sum(tf.image.total_variation(images))`
This implements the anisotropic 2-D version of the formula described here:
https://en.wikipedia.org/wiki/Total_variation_denoising
Args:
images: 4-D Tensor of shape `[batch, height, width, channels]` or 3-D Tensor
of shape `[height, width, channels]`.
name: A name for the operation (optional).
Raises:
ValueError: if images.shape is not a 3-D or 4-D vector.
Returns:
The total variation of `images`.
If `images` was 4-D, return a 1-D float Tensor of shape `[batch]` with the
total variation for each image in the batch.
If `images` was 3-D, return a scalar float with the total variation for
that image."
8526,sample_distorted_bounding_box_v2,tensorflow/tensorflow/python/ops/image_ops_impl.py,2842,function,"Generate a single randomly distorted bounding box for an image.
Bounding box annotations are often supplied in addition to ground-truth labels
in image recognition or object localization tasks. A common technique for
training such a system is to randomly distort an image while preserving
its content, i.e. *data augmentation*. This Op outputs a randomly distorted
localization of an object, i.e. bounding box, given an `image_size`,
`bounding_boxes` and a series of constraints.
The output of this Op is a single bounding box that may be used to crop the
original image. The output is returned as 3 tensors: `begin`, `size` and
`bboxes`. The first 2 tensors can be fed directly into `tf.slice` to crop the
image. The latter may be supplied to `tf.image.draw_bounding_boxes` to
visualize what the bounding box looks like.
Bounding boxes are supplied and returned as `[y_min, x_min, y_max, x_max]`.
The bounding box coordinates are floats in `[0.0, 1.0]` relative to the width
and the height of the underlying image.
For example,
```python
# Generate a single distorted bounding box.
begin, size, bbox_for_draw = tf.image.sample_distorted_bounding_box(
tf.shape(image),
bounding_boxes=bounding_boxes,
min_object_covered=0.1)
# Draw the bounding box in an image summary.
image_with_box = tf.image.draw_bounding_boxes(tf.expand_dims(image, 0),
bbox_for_draw)
tf.compat.v1.summary.image('images_with_box', image_with_box)
# Employ the bounding box to distort the image.
distorted_image = tf.slice(image, begin, size)
```
Note that if no bounding box information is available, setting
`use_image_if_no_bounding_boxes = true` will assume there is a single implicit
bounding box covering the whole image. If `use_image_if_no_bounding_boxes` is
false and no bounding boxes are supplied, an error is raised.
Args:
image_size: A `Tensor`. Must be one of the following types: `uint8`, `int8`,
`int16`, `int32`, `int64`. 1-D, containing `[height, width, channels]`.
bounding_boxes: A `Tensor` of type `float32`. 3-D with shape `[batch, N, 4]`
describing the N bounding boxes associated with the image.
seed: An optional `int`. Defaults to `0`. If `seed` is set to non-zero, the
random number generator is seeded by the given `seed`. Otherwise, it is
seeded by a random seed.
min_object_covered: A Tensor of type `float32`. Defaults to `0.1`. The
cropped area of the image must contain at least this fraction of any
bounding box supplied. The value of this parameter should be non-negative.
In the case of 0, the cropped area does not need to overlap any of the
bounding boxes supplied.
aspect_ratio_range: An optional list of `floats`. Defaults to `[0.75,
1.33]`. The cropped area of the image must have an aspect `ratio = width /
height` within this range.
area_range: An optional list of `floats`. Defaults to `[0.05, 1]`. The
cropped area of the image must contain a fraction of the supplied image
within this range.
max_attempts: An optional `int`. Defaults to `100`. Number of attempts at
generating a cropped region of the image of the specified constraints.
After `max_attempts` failures, return the entire image.
use_image_if_no_bounding_boxes: An optional `bool`. Defaults to `False`.
Controls behavior if no bounding boxes supplied. If true, assume an
implicit bounding box covering the whole input. If false, raise an error.
name: A name for the operation (optional).
Returns:
A tuple of `Tensor` objects (begin, size, bboxes).
begin: A `Tensor`. Has the same type as `image_size`. 1-D, containing
`[offset_height, offset_width, 0]`. Provide as input to
`tf.slice`.
size: A `Tensor`. Has the same type as `image_size`. 1-D, containing
`[target_height, target_width, -1]`. Provide as input to
`tf.slice`.
bboxes: A `Tensor` of type `float32`. 3-D with shape `[1, 1, 4]` containing
the distorted bounding box.
Provide as input to `tf.image.draw_bounding_boxes`."
8527,sample_distorted_bounding_box,tensorflow/tensorflow/python/ops/image_ops_impl.py,2946,function,"Generate a single randomly distorted bounding box for an image.
Bounding box annotations are often supplied in addition to ground-truth labels
in image recognition or object localization tasks. A common technique for
training such a system is to randomly distort an image while preserving
its content, i.e. *data augmentation*. This Op outputs a randomly distorted
localization of an object, i.e. bounding box, given an `image_size`,
`bounding_boxes` and a series of constraints.
The output of this Op is a single bounding box that may be used to crop the
original image. The output is returned as 3 tensors: `begin`, `size` and
`bboxes`. The first 2 tensors can be fed directly into `tf.slice` to crop the
image. The latter may be supplied to `tf.image.draw_bounding_boxes` to
visualize what the bounding box looks like.
Bounding boxes are supplied and returned as `[y_min, x_min, y_max, x_max]`.
The
bounding box coordinates are floats in `[0.0, 1.0]` relative to the width and
height of the underlying image.
For example,
```python
# Generate a single distorted bounding box.
begin, size, bbox_for_draw = tf.image.sample_distorted_bounding_box(
tf.shape(image),
bounding_boxes=bounding_boxes,
min_object_covered=0.1)
# Draw the bounding box in an image summary.
image_with_box = tf.image.draw_bounding_boxes(tf.expand_dims(image, 0),
bbox_for_draw)
tf.compat.v1.summary.image('images_with_box', image_with_box)
# Employ the bounding box to distort the image.
distorted_image = tf.slice(image, begin, size)
```
Note that if no bounding box information is available, setting
`use_image_if_no_bounding_boxes = True` will assume there is a single implicit
bounding box covering the whole image. If `use_image_if_no_bounding_boxes` is
false and no bounding boxes are supplied, an error is raised.
Args:
image_size: A `Tensor`. Must be one of the following types: `uint8`, `int8`,
`int16`, `int32`, `int64`. 1-D, containing `[height, width, channels]`.
bounding_boxes: A `Tensor` of type `float32`. 3-D with shape `[batch, N, 4]`
describing the N bounding boxes associated with the image.
seed: An optional `int`. Defaults to `0`. If either `seed` or `seed2` are
set to non-zero, the random number generator is seeded by the given
`seed`. Otherwise, it is seeded by a random seed.
seed2: An optional `int`. Defaults to `0`. A second seed to avoid seed
collision.
min_object_covered: A Tensor of type `float32`. Defaults to `0.1`. The
cropped area of the image must contain at least this fraction of any
bounding box supplied. The value of this parameter should be non-negative.
In the case of 0, the cropped area does not need to overlap any of the
bounding boxes supplied.
aspect_ratio_range: An optional list of `floats`. Defaults to `[0.75,
1.33]`. The cropped area of the image must have an aspect ratio = width /
height within this range.
area_range: An optional list of `floats`. Defaults to `[0.05, 1]`. The
cropped area of the image must contain a fraction of the supplied image
within this range.
max_attempts: An optional `int`. Defaults to `100`. Number of attempts at
generating a cropped region of the image of the specified constraints.
After `max_attempts` failures, return the entire image.
use_image_if_no_bounding_boxes: An optional `bool`. Defaults to `False`.
Controls behavior if no bounding boxes supplied. If true, assume an
implicit bounding box covering the whole input. If false, raise an error.
name: A name for the operation (optional).
Returns:
A tuple of `Tensor` objects (begin, size, bboxes).
begin: A `Tensor`. Has the same type as `image_size`. 1-D, containing
`[offset_height, offset_width, 0]`. Provide as input to
`tf.slice`.
size: A `Tensor`. Has the same type as `image_size`. 1-D, containing
`[target_height, target_width, -1]`. Provide as input to
`tf.slice`.
bboxes: A `Tensor` of type `float32`. 3-D with shape `[1, 1, 4]` containing
the distorted bounding box.
Provide as input to `tf.image.draw_bounding_boxes`."
8528,non_max_suppression,tensorflow/tensorflow/python/ops/image_ops_impl.py,3057,function,"Greedily selects a subset of bounding boxes in descending order of score.
Prunes away boxes that have high intersection-over-union (IOU) overlap
with previously selected boxes. Bounding boxes are supplied as
`[y1, x1, y2, x2]`, where `(y1, x1)` and `(y2, x2)` are the coordinates of any
diagonal pair of box corners and the coordinates can be provided as normalized
(i.e., lying in the interval `[0, 1]`) or absolute. Note that this algorithm
is agnostic to where the origin is in the coordinate system. Note that this
algorithm is invariant to orthogonal transformations and translations
of the coordinate system; thus translating or reflections of the coordinate
system result in the same boxes being selected by the algorithm.
The output of this operation is a set of integers indexing into the input
collection of bounding boxes representing the selected boxes. The bounding
box coordinates corresponding to the selected indices can then be obtained
using the `tf.gather` operation. For example:
```python
selected_indices = tf.image.non_max_suppression(
boxes, scores, max_output_size, iou_threshold)
selected_boxes = tf.gather(boxes, selected_indices)
```
Args:
boxes: A 2-D float `Tensor` of shape `[num_boxes, 4]`.
scores: A 1-D float `Tensor` of shape `[num_boxes]` representing a single
score corresponding to each box (each row of boxes).
max_output_size: A scalar integer `Tensor` representing the maximum number
of boxes to be selected by non-max suppression.
iou_threshold: A float representing the threshold for deciding whether boxes
overlap too much with respect to IOU.
score_threshold: A float representing the threshold for deciding when to
remove boxes based on score.
name: A name for the operation (optional).
Returns:
selected_indices: A 1-D integer `Tensor` of shape `[M]` representing the
selected indices from the boxes tensor, where `M <= max_output_size`."
8529,non_max_suppression_with_scores,tensorflow/tensorflow/python/ops/image_ops_impl.py,3110,function,"Greedily selects a subset of bounding boxes in descending order of score.
Prunes away boxes that have high intersection-over-union (IOU) overlap
with previously selected boxes. Bounding boxes are supplied as
`[y1, x1, y2, x2]`, where `(y1, x1)` and `(y2, x2)` are the coordinates of any
diagonal pair of box corners and the coordinates can be provided as normalized
(i.e., lying in the interval `[0, 1]`) or absolute. Note that this algorithm
is agnostic to where the origin is in the coordinate system. Note that this
algorithm is invariant to orthogonal transformations and translations
of the coordinate system; thus translating or reflections of the coordinate
system result in the same boxes being selected by the algorithm.
The output of this operation is a set of integers indexing into the input
collection of bounding boxes representing the selected boxes. The bounding
box coordinates corresponding to the selected indices can then be obtained
using the `tf.gather` operation. For example:
```python
selected_indices, selected_scores = tf.image.non_max_suppression_padded(
boxes, scores, max_output_size, iou_threshold=1.0, score_threshold=0.1,
soft_nms_sigma=0.5)
selected_boxes = tf.gather(boxes, selected_indices)
```
This function generalizes the `tf.image.non_max_suppression` op by also
supporting a Soft-NMS (with Gaussian weighting) mode (c.f.
Bodla et al, https://arxiv.org/abs/1704.04503) where boxes reduce the score
of other overlapping boxes instead of directly causing them to be pruned.
Consequently, in contrast to `tf.image.non_max_suppression`,
`tf.image.non_max_suppression_padded` returns the new scores of each input box
in the second output, `selected_scores`.
To enable this Soft-NMS mode, set the `soft_nms_sigma` parameter to be
larger than 0. When `soft_nms_sigma` equals 0, the behavior of
`tf.image.non_max_suppression_padded` is identical to that of
`tf.image.non_max_suppression` (except for the extra output) both in function
and in running time.
Args:
boxes: A 2-D float `Tensor` of shape `[num_boxes, 4]`.
scores: A 1-D float `Tensor` of shape `[num_boxes]` representing a single
score corresponding to each box (each row of boxes).
max_output_size: A scalar integer `Tensor` representing the maximum number
of boxes to be selected by non-max suppression.
iou_threshold: A float representing the threshold for deciding whether boxes
overlap too much with respect to IOU.
score_threshold: A float representing the threshold for deciding when to
remove boxes based on score.
soft_nms_sigma: A scalar float representing the Soft NMS sigma parameter;
See Bodla et al, https://arxiv.org/abs/1704.04503). When
`soft_nms_sigma=0.0` (which is default), we fall back to standard (hard)
NMS.
name: A name for the operation (optional).
Returns:
selected_indices: A 1-D integer `Tensor` of shape `[M]` representing the
selected indices from the boxes tensor, where `M <= max_output_size`.
selected_scores: A 1-D float tensor of shape `[M]` representing the
corresponding scores for each selected box, where `M <= max_output_size`.
Scores only differ from corresponding input scores when using Soft NMS
(i.e. when `soft_nms_sigma>0`)"
8530,non_max_suppression_with_overlaps,tensorflow/tensorflow/python/ops/image_ops_impl.py,3197,function,"Greedily selects a subset of bounding boxes in descending order of score.
Prunes away boxes that have high overlap with previously selected boxes.
N-by-n overlap values are supplied as square matrix.
The output of this operation is a set of integers indexing into the input
collection of bounding boxes representing the selected boxes. The bounding
box coordinates corresponding to the selected indices can then be obtained
using the `tf.gather` operation. For example:
```python
selected_indices = tf.image.non_max_suppression_overlaps(
overlaps, scores, max_output_size, iou_threshold)
selected_boxes = tf.gather(boxes, selected_indices)
```
Args:
overlaps: A 2-D float `Tensor` of shape `[num_boxes, num_boxes]`.
scores: A 1-D float `Tensor` of shape `[num_boxes]` representing a single
score corresponding to each box (each row of boxes).
max_output_size: A scalar integer `Tensor` representing the maximum number
of boxes to be selected by non-max suppression.
overlap_threshold: A float representing the threshold for deciding whether
boxes overlap too much with respect to the provided overlap values.
score_threshold: A float representing the threshold for deciding when to
remove boxes based on score.
name: A name for the operation (optional).
Returns:
selected_indices: A 1-D integer `Tensor` of shape `[M]` representing the
selected indices from the overlaps tensor, where `M <= max_output_size`."
8531,rgb_to_yiq,tensorflow/tensorflow/python/ops/image_ops_impl.py,3249,function,"Converts one or more images from RGB to YIQ.
Outputs a tensor of the same shape as the `images` tensor, containing the YIQ
value of the pixels.
The output is only well defined if the value in images are in [0,1].
Usage Example:
>>> x = tf.constant([[[1.0, 2.0, 3.0]]])
>>> tf.image.rgb_to_yiq(x)
<tf.Tensor: shape=(1, 1, 3), dtype=float32,
numpy=array([[[ 1.815 , -0.91724455, 0.09962624]]], dtype=float32)>
Args:
images: 2-D or higher rank. Image data to convert. Last dimension must be
size 3.
Returns:
images: tensor with the same shape as `images`."
8532,yiq_to_rgb,tensorflow/tensorflow/python/ops/image_ops_impl.py,3283,function,"Converts one or more images from YIQ to RGB.
Outputs a tensor of the same shape as the `images` tensor, containing the RGB
value of the pixels.
The output is only well defined if the Y value in images are in [0,1],
I value are in [-0.5957,0.5957] and Q value are in [-0.5226,0.5226].
Args:
images: 2-D or higher rank. Image data to convert. Last dimension must be
size 3.
Returns:
images: tensor with the same shape as `images`."
8533,rgb_to_yuv,tensorflow/tensorflow/python/ops/image_ops_impl.py,3312,function,"Converts one or more images from RGB to YUV.
Outputs a tensor of the same shape as the `images` tensor, containing the YUV
value of the pixels.
The output is only well defined if the value in images are in [0,1].
Args:
images: 2-D or higher rank. Image data to convert. Last dimension must be
size 3.
Returns:
images: tensor with the same shape as `images`."
8534,yuv_to_rgb,tensorflow/tensorflow/python/ops/image_ops_impl.py,3339,function,"Converts one or more images from YUV to RGB.
Outputs a tensor of the same shape as the `images` tensor, containing the RGB
value of the pixels.
The output is only well defined if the Y value in images are in [0,1],
U and V value are in [-0.5,0.5].
As per the above description, you need to scale your YUV images if their
pixel values are not in the required range. Below given example illustrates
preprocessing of each channel of images before feeding them to `yuv_to_rgb`.
```python
yuv_images = tf.random.uniform(shape=[100, 64, 64, 3], maxval=255)
last_dimension_axis = len(yuv_images.shape) - 1
yuv_tensor_images = tf.truediv(
tf.subtract(
yuv_images,
tf.reduce_min(yuv_images)
),
tf.subtract(
tf.reduce_max(yuv_images),
tf.reduce_min(yuv_images)
)
)
y, u, v = tf.split(yuv_tensor_images, 3, axis=last_dimension_axis)
target_uv_min, target_uv_max = -0.5, 0.5
u = u * (target_uv_max - target_uv_min) + target_uv_min
v = v * (target_uv_max - target_uv_min) + target_uv_min
preprocessed_yuv_images = tf.concat([y, u, v], axis=last_dimension_axis)
rgb_tensor_images = tf.image.yuv_to_rgb(preprocessed_yuv_images)
```
Args:
images: 2-D or higher rank. Image data to convert. Last dimension must be
size 3.
Returns:
images: tensor with the same shape as `images`."
8535,psnr,tensorflow/tensorflow/python/ops/image_ops_impl.py,3433,function,"Returns the Peak Signal-to-Noise Ratio between a and b.
This is intended to be used on signals (or images). Produces a PSNR value for
each image in batch.
The last three dimensions of input are expected to be [height, width, depth].
Example:
```python
# Read images from file.
im1 = tf.decode_png('path/to/im1.png')
im2 = tf.decode_png('path/to/im2.png')
# Compute PSNR over tf.uint8 Tensors.
psnr1 = tf.image.psnr(im1, im2, max_val=255)
# Compute PSNR over tf.float32 Tensors.
im1 = tf.image.convert_image_dtype(im1, tf.float32)
im2 = tf.image.convert_image_dtype(im2, tf.float32)
psnr2 = tf.image.psnr(im1, im2, max_val=1.0)
# psnr1 and psnr2 both have type tf.float32 and are almost equal.
```
Arguments:
a: First set of images.
b: Second set of images.
max_val: The dynamic range of the images (i.e., the difference between the
maximum the and minimum allowed values).
name: Namespace to embed the computation in.
Returns:
The scalar PSNR between a and b. The returned tensor has type `tf.float32`
and shape [batch_size, 1]."
8536,ssim,tensorflow/tensorflow/python/ops/image_ops_impl.py,3645,function,"Computes SSIM index between img1 and img2.
This function is based on the standard SSIM implementation from:
Wang, Z., Bovik, A. C., Sheikh, H. R., & Simoncelli, E. P. (2004). Image
quality assessment: from error visibility to structural similarity. IEEE
transactions on image processing.
Note: The true SSIM is only defined on grayscale. This function does not
perform any colorspace transform. (If the input is already YUV, then it will
compute YUV SSIM average.)
Details:
- 11x11 Gaussian filter of width 1.5 is used.
- k1 = 0.01, k2 = 0.03 as in the original paper.
The image sizes must be at least 11x11 because of the filter size.
Example:
```python
# Read images from file.
im1 = tf.decode_png('path/to/im1.png')
im2 = tf.decode_png('path/to/im2.png')
# Compute SSIM over tf.uint8 Tensors.
ssim1 = tf.image.ssim(im1, im2, max_val=255, filter_size=11,
filter_sigma=1.5, k1=0.01, k2=0.03)
# Compute SSIM over tf.float32 Tensors.
im1 = tf.image.convert_image_dtype(im1, tf.float32)
im2 = tf.image.convert_image_dtype(im2, tf.float32)
ssim2 = tf.image.ssim(im1, im2, max_val=1.0, filter_size=11,
filter_sigma=1.5, k1=0.01, k2=0.03)
# ssim1 and ssim2 both have type tf.float32 and are almost equal.
```
Args:
img1: First image batch.
img2: Second image batch.
max_val: The dynamic range of the images (i.e., the difference between the
maximum the and minimum allowed values).
filter_size: Default value 11 (size of gaussian filter).
filter_sigma: Default value 1.5 (width of gaussian filter).
k1: Default value 0.01
k2: Default value 0.03 (SSIM is less sensitivity to K2 for lower values, so
it would be better if we took the values in the range of 0 < K2 < 0.4).
Returns:
A tensor containing an SSIM value for each image in batch. Returned SSIM
values are in range (-1, 1], when pixel values are non-negative. Returns
a tensor with shape: broadcast(img1.shape[:-3], img2.shape[:-3])."
8537,ssim_multiscale,tensorflow/tensorflow/python/ops/image_ops_impl.py,3730,function,"Computes the MS-SSIM between img1 and img2.
This function assumes that `img1` and `img2` are image batches, i.e. the last
three dimensions are [height, width, channels].
Note: The true SSIM is only defined on grayscale. This function does not
perform any colorspace transform. (If the input is already YUV, then it will
compute YUV SSIM average.)
Original paper: Wang, Zhou, Eero P. Simoncelli, and Alan C. Bovik. ""Multiscale
structural similarity for image quality assessment."" Signals, Systems and
Computers, 2004.
Arguments:
img1: First image batch.
img2: Second image batch. Must have the same rank as img1.
max_val: The dynamic range of the images (i.e., the difference between the
maximum the and minimum allowed values).
power_factors: Iterable of weights for each of the scales. The number of
scales used is the length of the list. Index 0 is the unscaled
resolution's weight and each increasing scale corresponds to the image
being downsampled by 2. Defaults to (0.0448, 0.2856, 0.3001, 0.2363,
0.1333), which are the values obtained in the original paper.
filter_size: Default value 11 (size of gaussian filter).
filter_sigma: Default value 1.5 (width of gaussian filter).
k1: Default value 0.01
k2: Default value 0.03 (SSIM is less sensitivity to K2 for lower values, so
it would be better if we took the values in the range of 0 < K2 < 0.4).
Returns:
A tensor containing an MS-SSIM value for each image in batch. The values
are in range [0, 1]. Returns a tensor with shape:
broadcast(img1.shape[:-3], img2.shape[:-3])."
8538,image_gradients,tensorflow/tensorflow/python/ops/image_ops_impl.py,3858,function,"Returns image gradients (dy, dx) for each color channel.
Both output tensors have the same shape as the input: [batch_size, h, w,
d]. The gradient values are organized so that [I(x+1, y) - I(x, y)] is in
location (x, y). That means that dy will always have zeros in the last row,
and dx will always have zeros in the last column.
Usage Example:
```python
BATCH_SIZE = 1
IMAGE_HEIGHT = 5
IMAGE_WIDTH = 5
CHANNELS = 1
image = tf.reshape(tf.range(IMAGE_HEIGHT * IMAGE_WIDTH * CHANNELS,
delta=1, dtype=tf.float32),
shape=(BATCH_SIZE, IMAGE_HEIGHT, IMAGE_WIDTH, CHANNELS))
dx, dy = tf.image.image_gradients(image)
print(image[0, :,:,0])
tf.Tensor(
[[ 0. 1. 2. 3. 4.]
[ 5. 6. 7. 8. 9.]
[10. 11. 12. 13. 14.]
[15. 16. 17. 18. 19.]
[20. 21. 22. 23. 24.]], shape=(5, 5), dtype=float32)
print(dx[0, :,:,0])
tf.Tensor(
[[5. 5. 5. 5. 5.]
[5. 5. 5. 5. 5.]
[5. 5. 5. 5. 5.]
[5. 5. 5. 5. 5.]
[0. 0. 0. 0. 0.]], shape=(5, 5), dtype=float32)
print(dy[0, :,:,0])
tf.Tensor(
[[1. 1. 1. 1. 0.]
[1. 1. 1. 1. 0.]
[1. 1. 1. 1. 0.]
[1. 1. 1. 1. 0.]
[1. 1. 1. 1. 0.]], shape=(5, 5), dtype=float32)
```
Arguments:
image: Tensor with shape [batch_size, h, w, d].
Returns:
Pair of tensors (dy, dx) holding the vertical and horizontal image
gradients (1-step finite difference).
Raises:
ValueError: If `image` is not a 4D tensor."
8539,sobel_edges,tensorflow/tensorflow/python/ops/image_ops_impl.py,3932,function,"Returns a tensor holding Sobel edge maps.
Arguments:
image: Image tensor with shape [batch_size, h, w, d] and type float32 or
float64. The image(s) must be 2x2 or larger.
Returns:
Tensor holding edge maps for each channel. Returns a tensor with shape
[batch_size, h, w, d, 2] where the last two dimensions hold [[dy[0], dx[0]],
[dy[1], dx[1]], ..., [dy[d-1], dx[d-1]]] calculated using the Sobel filter."
8540,resize_bicubic,tensorflow/tensorflow/python/ops/image_ops_impl.py,3972,function,
8541,resize_bilinear,tensorflow/tensorflow/python/ops/image_ops_impl.py,3985,function,
8542,resize_nearest_neighbor,tensorflow/tensorflow/python/ops/image_ops_impl.py,3998,function,
8543,crop_and_resize_v2,tensorflow/tensorflow/python/ops/image_ops_impl.py,4045,function,"Extracts crops from the input image tensor and resizes them.
Extracts crops from the input image tensor and resizes them using bilinear
sampling or nearest neighbor sampling (possibly with aspect ratio change) to a
common output size specified by `crop_size`. This is more general than the
`crop_to_bounding_box` op which extracts a fixed size slice from the input
image and does not allow resizing or aspect ratio change.
Returns a tensor with `crops` from the input `image` at positions defined at
the bounding box locations in `boxes`. The cropped boxes are all resized (with
bilinear or nearest neighbor interpolation) to a fixed
`size = [crop_height, crop_width]`. The result is a 4-D tensor
`[num_boxes, crop_height, crop_width, depth]`. The resizing is corner aligned.
In particular, if `boxes = [[0, 0, 1, 1]]`, the method will give identical
results to using `tf.compat.v1.image.resize_bilinear()` or
`tf.compat.v1.image.resize_nearest_neighbor()`(depends on the `method`
argument) with
`align_corners=True`.
Args:
image: A 4-D tensor of shape `[batch, image_height, image_width, depth]`.
Both `image_height` and `image_width` need to be positive.
boxes: A 2-D tensor of shape `[num_boxes, 4]`. The `i`-th row of the tensor
specifies the coordinates of a box in the `box_ind[i]` image and is
specified in normalized coordinates `[y1, x1, y2, x2]`. A normalized
coordinate value of `y` is mapped to the image coordinate at `y *
(image_height - 1)`, so as the `[0, 1]` interval of normalized image
height is mapped to `[0, image_height - 1]` in image height coordinates.
We do allow `y1` > `y2`, in which case the sampled crop is an up-down
flipped version of the original image. The width dimension is treated
similarly. Normalized coordinates outside the `[0, 1]` range are allowed,
in which case we use `extrapolation_value` to extrapolate the input image
values.
box_indices: A 1-D tensor of shape `[num_boxes]` with int32 values in `[0,
batch)`. The value of `box_ind[i]` specifies the image that the `i`-th box
refers to.
crop_size: A 1-D tensor of 2 elements, `size = [crop_height, crop_width]`.
All cropped image patches are resized to this size. The aspect ratio of
the image content is not preserved. Both `crop_height` and `crop_width`
need to be positive.
method: An optional string specifying the sampling method for resizing. It
can be either `""bilinear""` or `""nearest""` and default to `""bilinear""`.
Currently two sampling methods are supported: Bilinear and Nearest
Neighbor.
extrapolation_value: An optional `float`. Defaults to `0`. Value used for
extrapolation, when applicable.
name: A name for the operation (optional).
Returns:
A 4-D tensor of shape `[num_boxes, crop_height, crop_width, depth]`.
Example:
```python
import tensorflow as tf
BATCH_SIZE = 1
NUM_BOXES = 5
IMAGE_HEIGHT = 256
IMAGE_WIDTH = 256
CHANNELS = 3
CROP_SIZE = (24, 24)
image = tf.random.normal(shape=(BATCH_SIZE, IMAGE_HEIGHT, IMAGE_WIDTH,
CHANNELS) )
boxes = tf.random.uniform(shape=(NUM_BOXES, 4))
box_indices = tf.random.uniform(shape=(NUM_BOXES,), minval=0,
maxval=BATCH_SIZE, dtype=tf.int32)
output = tf.image.crop_and_resize(image, boxes, box_indices, CROP_SIZE)
output.shape #=> (5, 24, 24, 3)
```"
8544,crop_and_resize_v1,tensorflow/tensorflow/python/ops/image_ops_impl.py,4132,function,
8545,extract_glimpse,tensorflow/tensorflow/python/ops/image_ops_impl.py,4152,function,"Extracts a glimpse from the input tensor.
Returns a set of windows called glimpses extracted at location
`offsets` from the input tensor. If the windows only partially
overlaps the inputs, the non-overlapping areas will be filled with
random noise.
The result is a 4-D tensor of shape `[batch_size, glimpse_height,
glimpse_width, channels]`. The channels and batch dimensions are the
same as that of the input tensor. The height and width of the output
windows are specified in the `size` parameter.
The argument `normalized` and `centered` controls how the windows are built:
* If the coordinates are normalized but not centered, 0.0 and 1.0
correspond to the minimum and maximum of each height and width
dimension.
* If the coordinates are both normalized and centered, they range from
-1.0 to 1.0. The coordinates (-1.0, -1.0) correspond to the upper
left corner, the lower right corner is located at (1.0, 1.0) and the
center is at (0, 0).
* If the coordinates are not normalized they are interpreted as
numbers of pixels.
Usage Example:
>>> x = [[[[0.0],
... [1.0],
... [2.0]],
... [[3.0],
... [4.0],
... [5.0]],
... [[6.0],
... [7.0],
... [8.0]]]]
>>> tf.compat.v1.image.extract_glimpse(x, size=(2, 2), offsets=[[1, 1]],
... centered=False, normalized=False)
<tf.Tensor: shape=(1, 2, 2, 1), dtype=float32, numpy=
array([[[[0.],
[1.]],
[[3.],
[4.]]]], dtype=float32)>
Args:
input: A `Tensor` of type `float32`. A 4-D float tensor of shape
`[batch_size, height, width, channels]`.
size: A `Tensor` of type `int32`. A 1-D tensor of 2 elements containing the
size of the glimpses to extract. The glimpse height must be specified
first, following by the glimpse width.
offsets: A `Tensor` of type `float32`. A 2-D integer tensor of shape
`[batch_size, 2]` containing the y, x locations of the center of each
window.
centered: An optional `bool`. Defaults to `True`. indicates if the offset
coordinates are centered relative to the image, in which case the (0, 0)
offset is relative to the center of the input images. If false, the (0,0)
offset corresponds to the upper left corner of the input images.
normalized: An optional `bool`. Defaults to `True`. indicates if the offset
coordinates are normalized.
uniform_noise: An optional `bool`. Defaults to `True`. indicates if the
noise should be generated using a uniform distribution or a Gaussian
distribution.
name: A name for the operation (optional).
Returns:
A `Tensor` of type `float32`."
8546,extract_glimpse_v2,tensorflow/tensorflow/python/ops/image_ops_impl.py,4238,function,"Extracts a glimpse from the input tensor.
Returns a set of windows called glimpses extracted at location
`offsets` from the input tensor. If the windows only partially
overlaps the inputs, the non-overlapping areas will be filled with
random noise.
The result is a 4-D tensor of shape `[batch_size, glimpse_height,
glimpse_width, channels]`. The channels and batch dimensions are the
same as that of the input tensor. The height and width of the output
windows are specified in the `size` parameter.
The argument `normalized` and `centered` controls how the windows are built:
* If the coordinates are normalized but not centered, 0.0 and 1.0
correspond to the minimum and maximum of each height and width
dimension.
* If the coordinates are both normalized and centered, they range from
-1.0 to 1.0. The coordinates (-1.0, -1.0) correspond to the upper
left corner, the lower right corner is located at (1.0, 1.0) and the
center is at (0, 0).
* If the coordinates are not normalized they are interpreted as
numbers of pixels.
Usage Example:
>>> x = [[[[0.0],
... [1.0],
... [2.0]],
... [[3.0],
... [4.0],
... [5.0]],
... [[6.0],
... [7.0],
... [8.0]]]]
>>> tf.image.extract_glimpse(x, size=(2, 2), offsets=[[1, 1]],
... centered=False, normalized=False)
<tf.Tensor: shape=(1, 2, 2, 1), dtype=float32, numpy=
array([[[[4.],
[5.]],
[[7.],
[8.]]]], dtype=float32)>
Args:
input: A `Tensor` of type `float32`. A 4-D float tensor of shape
`[batch_size, height, width, channels]`.
size: A `Tensor` of type `int32`. A 1-D tensor of 2 elements containing the
size of the glimpses to extract. The glimpse height must be specified
first, following by the glimpse width.
offsets: A `Tensor` of type `float32`. A 2-D integer tensor of shape
`[batch_size, 2]` containing the y, x locations of the center of each
window.
centered: An optional `bool`. Defaults to `True`. indicates if the offset
coordinates are centered relative to the image, in which case the (0, 0)
offset is relative to the center of the input images. If false, the (0,0)
offset corresponds to the upper left corner of the input images.
normalized: An optional `bool`. Defaults to `True`. indicates if the offset
coordinates are normalized.
noise: An optional `string`. Defaults to `uniform`. indicates if the noise
should be `uniform` (uniform distribution), `gaussian` (gaussian
distribution), or `zero` (zero padding).
name: A name for the operation (optional).
Returns:
A `Tensor` of type `float32`."
8547,combined_non_max_suppression,tensorflow/tensorflow/python/ops/image_ops_impl.py,4325,function,"Greedily selects a subset of bounding boxes in descending order of score.
This operation performs non_max_suppression on the inputs per batch, across
all classes.
Prunes away boxes that have high intersection-over-union (IOU) overlap
with previously selected boxes. Bounding boxes are supplied as
[y1, x1, y2, x2], where (y1, x1) and (y2, x2) are the coordinates of any
diagonal pair of box corners and the coordinates can be provided as normalized
(i.e., lying in the interval [0, 1]) or absolute. Note that this algorithm
is agnostic to where the origin is in the coordinate system. Also note that
this algorithm is invariant to orthogonal transformations and translations
of the coordinate system; thus translating or reflections of the coordinate
system result in the same boxes being selected by the algorithm.
The output of this operation is the final boxes, scores and classes tensor
returned after performing non_max_suppression.
Args:
boxes: A 4-D float `Tensor` of shape `[batch_size, num_boxes, q, 4]`. If `q`
is 1 then same boxes are used for all classes otherwise, if `q` is equal
to number of classes, class-specific boxes are used.
scores: A 3-D float `Tensor` of shape `[batch_size, num_boxes, num_classes]`
representing a single score corresponding to each box (each row of boxes).
max_output_size_per_class: A scalar integer `Tensor` representing the
maximum number of boxes to be selected by non-max suppression per class
max_total_size: A scalar representing the maximum number of boxes retained
over all classes.
iou_threshold: A float representing the threshold for deciding whether boxes
overlap too much with respect to IOU.
score_threshold: A float representing the threshold for deciding when to
remove boxes based on score.
pad_per_class: If false, the output nmsed boxes, scores and classes are
padded/clipped to `max_total_size`. If true, the output nmsed boxes,
scores and classes are padded to be of length
`max_size_per_class`*`num_classes`, unless it exceeds `max_total_size` in
which case it is clipped to `max_total_size`. Defaults to false.
clip_boxes: If true, the coordinates of output nmsed boxes will be clipped
to [0, 1]. If false, output the box coordinates as it is. Defaults to
true.
name: A name for the operation (optional).
Returns:
'nmsed_boxes': A [batch_size, max_detections, 4] float32 tensor
containing the non-max suppressed boxes.
'nmsed_scores': A [batch_size, max_detections] float32 tensor containing
the scores for the boxes.
'nmsed_classes': A [batch_size, max_detections] float32 tensor
containing the class for boxes.
'valid_detections': A [batch_size] int32 tensor indicating the number of
valid detections per batch item. Only the top valid_detections[i] entries
in nms_boxes[i], nms_scores[i] and nms_class[i] are valid. The rest of the
entries are zero paddings."
8548,non_max_suppression_padded,tensorflow/tensorflow/python/ops/image_ops_impl.py,4578,function,"Greedily selects a subset of bounding boxes in descending order of score.
Performs algorithmically equivalent operation to tf.image.non_max_suppression,
with the addition of an optional parameter which zero-pads the output to
be of size `max_output_size`.
The output of this operation is a tuple containing the set of integers
indexing into the input collection of bounding boxes representing the selected
boxes and the number of valid indices in the index set. The bounding box
coordinates corresponding to the selected indices can then be obtained using
the `tf.slice` and `tf.gather` operations. For example:
```python
selected_indices_padded, num_valid = tf.image.non_max_suppression_padded(
boxes, scores, max_output_size, iou_threshold,
score_threshold, pad_to_max_output_size=True)
selected_indices = tf.slice(
selected_indices_padded, tf.constant([0]), num_valid)
selected_boxes = tf.gather(boxes, selected_indices)
Args:
boxes: a tensor of rank 2 or higher with a shape of [..., num_boxes, 4].
Dimensions except the last two are batch dimensions.
scores: a tensor of rank 1 or higher with a shape of [..., num_boxes].
max_output_size: a scalar integer `Tensor` representing the maximum number
of boxes to be selected by non max suppression.
iou_threshold: a float representing the threshold for deciding whether boxes
overlap too much with respect to IoU (intersection over union).
score_threshold: a float representing the threshold for box scores. Boxes
with a score that is not larger than this threshold will be suppressed.
pad_to_max_output_size: whether to pad the output idx to max_output_size.
Must be set to True when the input is a batch of images.
name: name of operation.
sorted_input: a boolean indicating whether the input boxes and scores
are sorted in descending order by the score.
canonicalized_coordinates: if box coordinates are given as
`[y_min, x_min, y_max, x_max]`, setting to True eliminate redundant
computation to canonicalize box coordinates.
tile_size: an integer representing the number of boxes in a tile, i.e.,
the maximum number of boxes per image that can be used to suppress other
boxes in parallel; larger tile_size means larger parallelism and
potentially more redundant work.
Returns:
idx: a tensor with a shape of [..., num_boxes] representing the
indices selected by non-max suppression. The leading dimensions
are the batch dimensions of the input boxes. All numbers are within
[0, num_boxes). For each image (i.e., idx[i]), only the first num_valid[i]
indices (i.e., idx[i][:num_valid[i]]) are valid.
num_valid: a tensor of rank 0 or higher with a shape of [...]
representing the number of valid indices in idx. Its dimensions are the
batch dimensions of the input boxes.
Raises:
ValueError: When set pad_to_max_output_size to False for batched input."
8549,non_max_suppression_padded_v2,tensorflow/tensorflow/python/ops/image_ops_impl.py,4680,function,"Non-maximum suppression.
Prunes away boxes that have high intersection-over-union (IOU) overlap
with previously selected boxes. Bounding boxes are supplied as
`[y1, x1, y2, x2]`, where `(y1, x1)` and `(y2, x2)` are the coordinates of any
diagonal pair of box corners and the coordinates can be provided as normalized
(i.e., lying in the interval `[0, 1]`) or absolute. The bounding box
coordinates are cannonicalized to `[y_min, x_min, y_max, x_max]`,
where `(y_min, x_min)` and `(y_max, x_mas)` are the coordinates of the lower
left and upper right corner. User may indiciate the input box coordinates are
already canonicalized to eliminate redundant work by setting
canonicalized_coordinates to `True`. Note that this algorithm is agnostic to
where the origin is in the coordinate system. Note that this algorithm is
invariant to orthogonal transformations and translations of the coordinate
system; thus translating or reflections of the coordinate system result in the
same boxes being selected by the algorithm.
Similar to tf.image.non_max_suppression, non_max_suppression_padded
implements hard NMS but can operate on a batch of images and improves
performance by titling the bounding boxes. Non_max_suppression_padded should
be preferred over tf.image_non_max_suppression when running on devices with
abundant parallelsim for higher computation speed. For soft NMS, refer to
tf.image.non_max_suppression_with_scores.
While a serial NMS algorithm iteratively uses the highest-scored unprocessed
box to suppress boxes, this algorithm uses many boxes to suppress other boxes
in parallel. The key idea is to partition boxes into tiles based on their
score and suppresses boxes tile by tile, thus achieving parallelism within a
tile. The tile size determines the degree of parallelism.
In cross suppression (using boxes of tile A to suppress boxes of tile B),
all boxes in A can independently suppress boxes in B.
Self suppression (suppressing boxes of the same tile) needs to be iteratively
applied until there's no more suppression. In each iteration, boxes that
cannot be suppressed are used to suppress boxes in the same tile.
boxes = boxes.pad_to_multiply_of(tile_size)
num_tiles = len(boxes) // tile_size
output_boxes = []
for i in range(num_tiles):
box_tile = boxes[i*tile_size : (i+1)*tile_size]
for j in range(i - 1):
# in parallel suppress boxes in box_tile using boxes from suppressing_tile
suppressing_tile = boxes[j*tile_size : (j+1)*tile_size]
iou = _bbox_overlap(box_tile, suppressing_tile)
# if the box is suppressed in iou, clear it to a dot
box_tile *= _update_boxes(iou)
# Iteratively handle the diagnal tile.
iou = _box_overlap(box_tile, box_tile)
iou_changed = True
while iou_changed:
# boxes that are not suppressed by anything else
suppressing_boxes = _get_suppressing_boxes(iou)
# boxes that are suppressed by suppressing_boxes
suppressed_boxes = _get_suppressed_boxes(iou, suppressing_boxes)
# clear iou to 0 for boxes that are suppressed, as they cannot be used
# to suppress other boxes any more
new_iou = _clear_iou(iou, suppressed_boxes)
iou_changed = (new_iou != iou)
iou = new_iou
# remaining boxes that can still suppress others, are selected boxes.
output_boxes.append(_get_suppressing_boxes(iou))
if len(output_boxes) >= max_output_size:
break
Args:
boxes: a tensor of rank 2 or higher with a shape of [..., num_boxes, 4].
Dimensions except the last two are batch dimensions. The last dimension
represents box coordinates, given as [y_1, x_1, y_2, x_2]. The coordinates
on each dimension can be given in any order
(see also `canonicalized_coordinates`) but must describe a box with
a positive area.
scores: a tensor of rank 1 or higher with a shape of [..., num_boxes].
max_output_size: a scalar integer `Tensor` representing the maximum number
of boxes to be selected by non max suppression.
iou_threshold: a float representing the threshold for deciding whether boxes
overlap too much with respect to IoU (intersection over union).
score_threshold: a float representing the threshold for box scores. Boxes
with a score that is not larger than this threshold will be suppressed.
sorted_input: a boolean indicating whether the input boxes and scores
are sorted in descending order by the score.
canonicalized_coordinates: if box coordinates are given as
`[y_min, x_min, y_max, x_max]`, setting to True eliminate redundant
computation to canonicalize box coordinates.
tile_size: an integer representing the number of boxes in a tile, i.e.,
the maximum number of boxes per image that can be used to suppress other
boxes in parallel; larger tile_size means larger parallelism and
potentially more redundant work.
Returns:
idx: a tensor with a shape of [..., num_boxes] representing the
indices selected by non-max suppression. The leading dimensions
are the batch dimensions of the input boxes. All numbers are within
[0, num_boxes). For each image (i.e., idx[i]), only the first num_valid[i]
indices (i.e., idx[i][:num_valid[i]]) are valid.
num_valid: a tensor of rank 0 or higher with a shape of [...]
representing the number of valid indices in idx. Its dimensions are the
batch dimensions of the input boxes.
Raises:
ValueError: When set pad_to_max_output_size to False for batched input."
8550,non_max_suppression_padded_v1,tensorflow/tensorflow/python/ops/image_ops_impl.py,4917,function,"Greedily selects a subset of bounding boxes in descending order of score.
Performs algorithmically equivalent operation to tf.image.non_max_suppression,
with the addition of an optional parameter which zero-pads the output to
be of size `max_output_size`.
The output of this operation is a tuple containing the set of integers
indexing into the input collection of bounding boxes representing the selected
boxes and the number of valid indices in the index set. The bounding box
coordinates corresponding to the selected indices can then be obtained using
the `tf.slice` and `tf.gather` operations. For example:
```python
selected_indices_padded, num_valid = tf.image.non_max_suppression_padded(
boxes, scores, max_output_size, iou_threshold,
score_threshold, pad_to_max_output_size=True)
selected_indices = tf.slice(
selected_indices_padded, tf.constant([0]), num_valid)
selected_boxes = tf.gather(boxes, selected_indices)
```
Args:
boxes: A 2-D float `Tensor` of shape `[num_boxes, 4]`.
scores: A 1-D float `Tensor` of shape `[num_boxes]` representing a single
score corresponding to each box (each row of boxes).
max_output_size: A scalar integer `Tensor` representing the maximum number
of boxes to be selected by non-max suppression.
iou_threshold: A float representing the threshold for deciding whether boxes
overlap too much with respect to IOU.
score_threshold: A float representing the threshold for deciding when to
remove boxes based on score.
pad_to_max_output_size: bool. If True, size of `selected_indices` output is
padded to `max_output_size`.
name: A name for the operation (optional).
Returns:
selected_indices: A 1-D integer `Tensor` of shape `[M]` representing the
selected indices from the boxes tensor, where `M <= max_output_size`.
valid_outputs: A scalar integer `Tensor` denoting how many elements in
`selected_indices` are valid. Valid elements occur first, then padding."
8551,draw_bounding_boxes_v2,tensorflow/tensorflow/python/ops/image_ops_impl.py,4974,function,"Draw bounding boxes on a batch of images.
Outputs a copy of `images` but draws on top of the pixels zero or more
bounding boxes specified by the locations in `boxes`. The coordinates of the
each bounding box in `boxes` are encoded as `[y_min, x_min, y_max, x_max]`.
The bounding box coordinates are floats in `[0.0, 1.0]` relative to the width
and the height of the underlying image.
For example, if an image is 100 x 200 pixels (height x width) and the bounding
box is `[0.1, 0.2, 0.5, 0.9]`, the upper-left and bottom-right coordinates of
the bounding box will be `(40, 10)` to `(180, 50)` (in (x,y) coordinates).
Parts of the bounding box may fall outside the image.
Args:
images: A `Tensor`. Must be one of the following types: `float32`, `half`.
4-D with shape `[batch, height, width, depth]`. A batch of images.
boxes: A `Tensor` of type `float32`. 3-D with shape `[batch,
num_bounding_boxes, 4]` containing bounding boxes.
colors: A `Tensor` of type `float32`. 2-D. A list of RGBA colors to cycle
through for the boxes.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `images`.
Usage Example:
>>> # create an empty image
>>> img = tf.zeros([1, 3, 3, 3])
>>> # draw a box around the image
>>> box = np.array([0, 0, 1, 1])
>>> boxes = box.reshape([1, 1, 4])
>>> # alternate between red and blue
>>> colors = np.array([[1.0, 0.0, 0.0], [0.0, 0.0, 1.0]])
>>> tf.image.draw_bounding_boxes(img, boxes, colors)
<tf.Tensor: shape=(1, 3, 3, 3), dtype=float32, numpy=
array([[[[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.]],
[[1., 0., 0.],
[0., 0., 0.],
[1., 0., 0.]],
[[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.]]]], dtype=float32)>"
8552,draw_bounding_boxes,tensorflow/tensorflow/python/ops/image_ops_impl.py,5029,function,"Draw bounding boxes on a batch of images.
Outputs a copy of `images` but draws on top of the pixels zero or more
bounding boxes specified by the locations in `boxes`. The coordinates of the
each bounding box in `boxes` are encoded as `[y_min, x_min, y_max, x_max]`.
The bounding box coordinates are floats in `[0.0, 1.0]` relative to the width
and the height of the underlying image.
For example, if an image is 100 x 200 pixels (height x width) and the bounding
box is `[0.1, 0.2, 0.5, 0.9]`, the upper-left and bottom-right coordinates of
the bounding box will be `(40, 10)` to `(180, 50)` (in (x,y) coordinates).
Parts of the bounding box may fall outside the image.
Args:
images: A `Tensor`. Must be one of the following types: `float32`, `half`.
4-D with shape `[batch, height, width, depth]`. A batch of images.
boxes: A `Tensor` of type `float32`. 3-D with shape `[batch,
num_bounding_boxes, 4]` containing bounding boxes.
name: A name for the operation (optional).
colors: A `Tensor` of type `float32`. 2-D. A list of RGBA colors to cycle
through for the boxes.
Returns:
A `Tensor`. Has the same type as `images`.
Usage Example:
>>> # create an empty image
>>> img = tf.zeros([1, 3, 3, 3])
>>> # draw a box around the image
>>> box = np.array([0, 0, 1, 1])
>>> boxes = box.reshape([1, 1, 4])
>>> # alternate between red and blue
>>> colors = np.array([[1.0, 0.0, 0.0], [0.0, 0.0, 1.0]])
>>> tf.image.draw_bounding_boxes(img, boxes, colors)
<tf.Tensor: shape=(1, 3, 3, 3), dtype=float32, numpy=
array([[[[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.]],
[[1., 0., 0.],
[0., 0., 0.],
[1., 0., 0.]],
[[1., 0., 0.],
[1., 0., 0.],
[1., 0., 0.]]]], dtype=float32)>"
8553,generate_bounding_box_proposals,tensorflow/tensorflow/python/ops/image_ops_impl.py,5082,function,"Generate bounding box proposals from encoded bounding boxes.
Returns:
rois: Region of interest boxes sorted by their scores.
roi_probabilities: scores of the ROI boxes in the ROIs' tensor."
8554,AdjustGamma,tensorflow/tensorflow/python/ops/image_ops_test.py,277,class,
8555,FlipImageBenchmark,tensorflow/tensorflow/python/ops/image_ops_test.py,539,class,
8556,benchmarkFlipLeftRightCpu1,tensorflow/tensorflow/python/ops/image_ops_test.py,632,method,
8557,benchmarkFlipLeftRightCpuAll,tensorflow/tensorflow/python/ops/image_ops_test.py,635,method,
8558,benchmarkFlipLeftRightGpu,tensorflow/tensorflow/python/ops/image_ops_test.py,638,method,
8559,benchmarkRandomFlipLeftRightCpu1,tensorflow/tensorflow/python/ops/image_ops_test.py,641,method,
8560,benchmarkRandomFlipLeftRightCpuAll,tensorflow/tensorflow/python/ops/image_ops_test.py,644,method,
8561,benchmarkRandomFlipLeftRightGpu,tensorflow/tensorflow/python/ops/image_ops_test.py,647,method,
8562,benchmarkBatchedRandomFlipLeftRightCpu1,tensorflow/tensorflow/python/ops/image_ops_test.py,650,method,
8563,benchmarkBatchedRandomFlipLeftRightCpuAll,tensorflow/tensorflow/python/ops/image_ops_test.py,653,method,
8564,benchmarkBatchedRandomFlipLeftRightGpu,tensorflow/tensorflow/python/ops/image_ops_test.py,656,method,
8565,AdjustHueBenchmark,tensorflow/tensorflow/python/ops/image_ops_test.py,660,class,
8566,benchmarkAdjustHueCpu1,tensorflow/tensorflow/python/ops/image_ops_test.py,693,method,
8567,benchmarkAdjustHueCpuAll,tensorflow/tensorflow/python/ops/image_ops_test.py,696,method,
8568,benchmarkAdjustHueGpu,tensorflow/tensorflow/python/ops/image_ops_test.py,699,method,
8569,AdjustSaturationBenchmark,tensorflow/tensorflow/python/ops/image_ops_test.py,703,class,
8570,benchmarkAdjustSaturationCpu1,tensorflow/tensorflow/python/ops/image_ops_test.py,737,method,
8571,benchmarkAdjustSaturationCpuAll,tensorflow/tensorflow/python/ops/image_ops_test.py,740,method,
8572,benchmarkAdjustSaturationGpu,tensorflow/tensorflow/python/ops/image_ops_test.py,743,method,
8573,ResizeBilinearBenchmark,tensorflow/tensorflow/python/ops/image_ops_test.py,747,class,
8574,benchmarkSimilar3Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,776,method,
8575,benchmarkScaleUp3Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,779,method,
8576,benchmarkScaleDown3Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,782,method,
8577,benchmarkSimilar1Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,785,method,
8578,benchmarkScaleUp1Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,788,method,
8579,benchmarkScaleDown1Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,791,method,
8580,ResizeBicubicBenchmark,tensorflow/tensorflow/python/ops/image_ops_test.py,795,class,
8581,benchmarkSimilar3Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,825,method,
8582,benchmarkScaleUp3Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,828,method,
8583,benchmarkScaleDown3Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,831,method,
8584,benchmarkSimilar1Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,834,method,
8585,benchmarkScaleUp1Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,837,method,
8586,benchmarkScaleDown1Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,840,method,
8587,benchmarkSimilar4Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,843,method,
8588,benchmarkScaleUp4Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,846,method,
8589,benchmarkScaleDown4Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,849,method,
8590,ResizeAreaBenchmark,tensorflow/tensorflow/python/ops/image_ops_test.py,853,class,
8591,benchmarkSimilar3Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,881,method,
8592,benchmarkScaleUp3Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,884,method,
8593,benchmarkScaleDown3Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,887,method,
8594,benchmarkSimilar1Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,890,method,
8595,benchmarkScaleUp1Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,893,method,
8596,benchmarkScaleDown1Channel,tensorflow/tensorflow/python/ops/image_ops_test.py,896,method,
8597,simple_color_ramp,tensorflow/tensorflow/python/ops/image_ops_test.py,3815,function,Build a simple color ramp RGB image.
8598,Initializer,tensorflow/tensorflow/python/ops/init_ops.py,55,class,Initializer base class: all initializers inherit from this class.
8599,get_config,tensorflow/tensorflow/python/ops/init_ops.py,70,method,"Returns the configuration of the initializer as a JSON-serializable dict.
Returns:
A JSON-serializable Python dict."
8600,from_config,tensorflow/tensorflow/python/ops/init_ops.py,79,method,"Instantiates an initializer from a configuration dictionary.
Example:
```python
initializer = RandomUniform(-1, 1)
config = initializer.get_config()
initializer = RandomUniform.from_config(config)
```
Args:
config: A Python dictionary. It will typically be the output of
`get_config`.
Returns:
An Initializer instance."
8601,Zeros,tensorflow/tensorflow/python/ops/init_ops.py,102,class,Initializer that generates tensors initialized to 0.
8602,get_config,tensorflow/tensorflow/python/ops/init_ops.py,116,method,
8603,Ones,tensorflow/tensorflow/python/ops/init_ops.py,122,class,Initializer that generates tensors initialized to 1.
8604,get_config,tensorflow/tensorflow/python/ops/init_ops.py,136,method,
8605,Constant,tensorflow/tensorflow/python/ops/init_ops.py,142,class,"Initializer that generates tensors with constant values.
The resulting tensor is populated with values of type `dtype`, as
specified by arguments `value` following the desired `shape` of the
new tensor (see examples below).
The argument `value` can be a constant value, or a list of values of type
`dtype`. If `value` is a list, then the length of the list must be less
than or equal to the number of elements implied by the desired shape of the
tensor. In the case where the total number of elements in `value` is less
than the number of elements required by the tensor shape, the last element
in `value` will be used to fill the remaining entries. If the total number of
elements in `value` is greater than the number of elements required by the
tensor shape, the initializer will raise a `ValueError`.
Args:
value: A Python scalar, list or tuple of values, or a N-dimensional numpy
array. All elements of the initialized variable will be set to the
corresponding value in the `value` argument.
dtype: Default data type, used if no `dtype` argument is provided when
calling the initializer.
verify_shape: Boolean that enables verification of the shape of `value`. If
`True`, the initializer will throw an error if the shape of `value` is not
compatible with the shape of the initialized tensor.
Raises:
TypeError: If the input `value` is not one of the expected types.
Examples:
The following example can be rewritten using a numpy.ndarray instead
of the `value` list, even reshaped, as shown in the two commented lines
below the `value` list initialization.
>>> value = [0, 1, 2, 3, 4, 5, 6, 7]
>>> init = tf.compat.v1.constant_initializer(value)
>>> # fitting shape
>>> with tf.compat.v1.Session():
... x = tf.compat.v1.get_variable('x', shape=[2, 4], initializer=init)
... x.initializer.run()
... print(x.eval())
[[0. 1. 2. 3.]
[4. 5. 6. 7.]]
>>> # Larger shape
>>> with tf.compat.v1.Session():
... y = tf.compat.v1.get_variable('y', shape=[3, 4], initializer=init)
... y.initializer.run()
... print(y.eval())
[[0. 1. 2. 3.]
[4. 5. 6. 7.]
[7. 7. 7. 7.]]
>>> # Smaller shape
>>> with tf.compat.v1.Session():
... z = tf.compat.v1.get_variable('z', shape=[2, 3], initializer=init)
Traceback (most recent call last):
...
ValueError: Too many elements provided. Needed at most 6, but received 8
>>> # Shape verification
>>> init_verify = tf.compat.v1.constant_initializer(value, verify_shape=True)
>>> with tf.compat.v1.Session():
... u = tf.compat.v1.get_variable('u', shape=[3, 4],
... initializer=init_verify)
Traceback (most recent call last):
...
TypeError: Expected Tensor's shape: (3, 4), got (8,)."
8606,get_config,tensorflow/tensorflow/python/ops/init_ops.py,232,method,
8607,RandomUniform,tensorflow/tensorflow/python/ops/init_ops.py,242,class,"Initializer that generates tensors with a uniform distribution.
Args:
minval: A python scalar or a scalar tensor. Lower bound of the range of
random values to generate.
maxval: A python scalar or a scalar tensor. Upper bound of the range of
random values to generate. Defaults to 1 for float types.
seed: A Python integer. Used to create random seeds. See
`tf.compat.v1.set_random_seed` for behavior.
dtype: Default data type, used if no `dtype` argument is provided when
calling the initializer."
8608,get_config,tensorflow/tensorflow/python/ops/init_ops.py,271,method,
8609,RandomNormal,tensorflow/tensorflow/python/ops/init_ops.py,282,class,"Initializer that generates tensors with a normal distribution.
Args:
mean: a python scalar or a scalar tensor. Mean of the random values to
generate.
stddev: a python scalar or a scalar tensor. Standard deviation of the random
values to generate.
seed: A Python integer. Used to create random seeds. See
`tf.compat.v1.set_random_seed` for behavior.
dtype: Default data type, used if no `dtype` argument is provided when
calling the initializer. Only floating point types are supported."
8610,get_config,tensorflow/tensorflow/python/ops/init_ops.py,311,method,
8611,TruncatedNormal,tensorflow/tensorflow/python/ops/init_ops.py,323,class,"Initializer that generates a truncated normal distribution.
These values are similar to values from a `random_normal_initializer`
except that values more than two standard deviations from the mean
are discarded and re-drawn. This is the recommended initializer for
neural network weights and filters.
Args:
mean: a python scalar or a scalar tensor. Mean of the random values to
generate.
stddev: a python scalar or a scalar tensor. Standard deviation of the random
values to generate.
seed: A Python integer. Used to create random seeds. See
`tf.compat.v1.set_random_seed` for behavior.
dtype: Default data type, used if no `dtype` argument is provided when
calling the initializer. Only floating point types are supported."
8612,get_config,tensorflow/tensorflow/python/ops/init_ops.py,357,method,
8613,UniformUnitScaling,tensorflow/tensorflow/python/ops/init_ops.py,371,class,"Initializer that generates tensors without scaling variance.
When initializing a deep network, it is in principle advantageous to keep
the scale of the input variance constant, so it does not explode or diminish
by reaching the final layer. If the input is `x` and the operation `x * W`,
and we want to initialize `W` uniformly at random, we need to pick `W` from
[-sqrt(3) / sqrt(dim), sqrt(3) / sqrt(dim)]
to keep the scale intact, where `dim = W.shape[0]` (the size of the input).
A similar calculation for convolutional networks gives an analogous result
with `dim` equal to the product of the first 3 dimensions. When
nonlinearities are present, we need to multiply this by a constant `factor`.
See (Sussillo et al., 2014) for deeper motivation, experiments
and the calculation of constants. In section 2.3 there, the constants were
numerically computed: for a linear layer it's 1.0, relu: ~1.43, tanh: ~1.15.
Args:
factor: Float. A multiplicative factor by which the values will be scaled.
seed: A Python integer. Used to create random seeds. See
`tf.compat.v1.set_random_seed` for behavior.
dtype: Default data type, used if no `dtype` argument is provided when
calling the initializer. Only floating point types are supported.
References:
[Sussillo et al., 2014](https://arxiv.org/abs/1412.6558)
([pdf](http://arxiv.org/pdf/1412.6558.pdf))"
8614,get_config,tensorflow/tensorflow/python/ops/init_ops.py,430,method,
8615,VarianceScaling,tensorflow/tensorflow/python/ops/init_ops.py,437,class,"Initializer capable of adapting its scale to the shape of weights tensors.
With `distribution=""truncated_normal"" or ""untruncated_normal""`,
samples are drawn from a truncated/untruncated normal
distribution with a mean of zero and a standard deviation (after truncation,
if used) `stddev = sqrt(scale / n)`
where n is:
- number of input units in the weight tensor, if mode = ""fan_in""
- number of output units, if mode = ""fan_out""
- average of the numbers of input and output units, if mode = ""fan_avg""
With `distribution=""uniform""`, samples are drawn from a uniform distribution
within [-limit, limit], with `limit = sqrt(3 * scale / n)`.
Args:
scale: Scaling factor (positive float).
mode: One of ""fan_in"", ""fan_out"", ""fan_avg"".
distribution: Random distribution to use. One of ""normal"", ""uniform"".
seed: A Python integer. Used to create random seeds. See
`tf.compat.v1.set_random_seed` for behavior.
dtype: Default data type, used if no `dtype` argument is provided when
calling the initializer. Only floating point types are supported.
Raises:
ValueError: In case of an invalid value for the ""scale"", mode"" or
""distribution"" arguments."
8616,get_config,tensorflow/tensorflow/python/ops/init_ops.py,521,method,
8617,Orthogonal,tensorflow/tensorflow/python/ops/init_ops.py,534,class,"Initializer that generates an orthogonal matrix.
If the shape of the tensor to initialize is two-dimensional, it is initialized
with an orthogonal matrix obtained from the QR decomposition of a matrix of
random numbers drawn from a normal distribution.
If the matrix has fewer rows than columns then the output will have orthogonal
rows. Otherwise, the output will have orthogonal columns.
If the shape of the tensor to initialize is more than two-dimensional,
a matrix of shape `(shape[0] * ... * shape[n - 2], shape[n - 1])`
is initialized, where `n` is the length of the shape vector.
The matrix is subsequently reshaped to give a tensor of the desired shape.
Args:
gain: multiplicative factor to apply to the orthogonal matrix
seed: A Python integer. Used to create random seeds. See
`tf.compat.v1.set_random_seed` for behavior.
dtype: Default data type, used if no `dtype` argument is provided when
calling the initializer. Only floating point types are supported.
References:
[Saxe et al., 2014](https://openreview.net/forum?id=_wzZwKpTDF_9C)
([pdf](https://arxiv.org/pdf/1312.6120.pdf))"
8618,get_config,tensorflow/tensorflow/python/ops/init_ops.py,597,method,
8619,ConvolutionDeltaOrthogonal,tensorflow/tensorflow/python/ops/init_ops.py,603,class,"Initializer that generates a delta orthogonal kernel for ConvNets.
The shape of the tensor must have length 3, 4 or 5. The number of input
filters must not exceed the number of output filters. The center pixels of the
tensor form an orthogonal matrix. Other pixels are set to be zero. See
algorithm 2 in (Xiao et al., 2018).
Args:
gain: Multiplicative factor to apply to the orthogonal matrix. Default is 1.
The 2-norm of an input is multiplied by a factor of `gain` after applying
this convolution.
seed: A Python integer. Used to create random seeds. See
`tf.compat.v1.set_random_seed` for behavior.
dtype: Default data type, used if no `dtype` argument is provided when
calling the initializer. Only floating point types are supported.
References:
[Xiao et al., 2018](http://proceedings.mlr.press/v80/xiao18a.html)
([pdf](http://proceedings.mlr.press/v80/xiao18a/xiao18a.pdf))"
8620,get_config,tensorflow/tensorflow/python/ops/init_ops.py,665,method,
8621,ConvolutionOrthogonal,tensorflow/tensorflow/python/ops/init_ops.py,669,class,"Initializer that generates orthogonal kernel for ConvNets.
Base class used to construct 1D, 2D and 3D orthogonal kernels for convolution.
Args:
gain: multiplicative factor to apply to the orthogonal matrix. Default is 1.
The 2-norm of an input is multiplied by a factor of `gain` after applying
this convolution.
seed: A Python integer. Used to create random seeds. See
`tf.compat.v1.set_random_seed` for behavior.
dtype: Default data type, used if no `dtype` argument is provided when
calling the initializer. Only floating point types are supported.
References:
[Xiao et al., 2018](http://proceedings.mlr.press/v80/xiao18a.html)
([pdf](http://proceedings.mlr.press/v80/xiao18a/xiao18a.pdf))"
8622,get_config,tensorflow/tensorflow/python/ops/init_ops.py,695,method,
8623,ConvolutionOrthogonal2D,tensorflow/tensorflow/python/ops/init_ops.py,736,class,"Initializer that generates a 2D orthogonal kernel for ConvNets.
The shape of the tensor must have length 4. The number of input
filters must not exceed the number of output filters.
The orthogonality(==isometry) is exact when the inputs are circular padded.
There are finite-width effects with non-circular padding (e.g. zero padding).
See algorithm 1 in (Xiao et al., 2018).
Args:
gain: Multiplicative factor to apply to the orthogonal matrix. Default is 1.
This has the effect of scaling the output 2-norm by a factor of `gain`.
seed: A Python integer. Used to create random seeds. See
`tf.compat.v1.set_random_seed` for behavior.
dtype: Default data type, used if no `dtype` argument is provided when
calling the initializer. Only floating point types are supported.
References:
[Xiao et al., 2018](http://proceedings.mlr.press/v80/xiao18a.html)
([pdf](http://proceedings.mlr.press/v80/xiao18a/xiao18a.pdf))"
8624,ConvolutionOrthogonal1D,tensorflow/tensorflow/python/ops/init_ops.py,880,class,"Initializer that generates a 1D orthogonal kernel for ConvNets.
The shape of the tensor must have length 3. The number of input
filters must not exceed the number of output filters.
The orthogonality(==isometry) is exact when the inputs are circular padded.
There are finite-width effects with non-circular padding (e.g. zero padding).
See algorithm 1 in (Xiao et al., 2018).
Args:
gain: Multiplicative factor to apply to the orthogonal matrix. Default is 1.
The 2-norm of an input is multiplied by a factor of `gain` after applying
this convolution.
seed: A Python integer. Used to create random seeds. See
`tf.compat.v1.set_random_seed` for behavior.
dtype: Default data type, used if no `dtype` argument is provided when
calling the initializer. Only floating point types are supported.
References:
[Xiao et al., 2018](http://proceedings.mlr.press/v80/xiao18a.html)
([pdf](http://proceedings.mlr.press/v80/xiao18a/xiao18a.pdf))"
8625,ConvolutionOrthogonal3D,tensorflow/tensorflow/python/ops/init_ops.py,1005,class,"Initializer that generates a 3D orthogonal kernel for ConvNets.
The shape of the tensor must have length 5. The number of input
filters must not exceed the number of output filters.
The orthogonality(==isometry) is exact when the inputs are circular padded.
There are finite-width effects with non-circular padding (e.g. zero padding).
See algorithm 1 (Xiao et al., 2018).
Args:
gain: Multiplicative factor to apply to the orthogonal matrix. Default is 1.
The 2-norm of an input is multiplied by a factor of `gain` after applying
this convolution.
seed: A Python integer. Used to create random seeds. See
`tf.compat.v1.set_random_seed` for behavior.
dtype: Default data type, used if no `dtype` argument is provided when
calling the initializer. Only floating point types are supported.
References:
[Xiao et al., 2018](http://proceedings.mlr.press/v80/xiao18a.html)
([pdf](http://proceedings.mlr.press/v80/xiao18a/xiao18a.pdf))"
8626,matmul,tensorflow/tensorflow/python/ops/init_ops.py,1082,method,
8627,cast,tensorflow/tensorflow/python/ops/init_ops.py,1085,method,Return p or (1-p).
8628,Identity,tensorflow/tensorflow/python/ops/init_ops.py,1170,class,"Initializer that generates the identity matrix.
Only use for 2D matrices.
Args:
gain: Multiplicative factor to apply to the identity matrix.
dtype: Default data type, used if no `dtype` argument is provided when
calling the initializer. Only floating point types are supported."
8629,get_config,tensorflow/tensorflow/python/ops/init_ops.py,1203,method,
8630,GlorotUniform,tensorflow/tensorflow/python/ops/init_ops.py,1210,class,"The Glorot uniform initializer, also called Xavier uniform initializer.
It draws samples from a uniform distribution within [-limit, limit]
where `limit` is `sqrt(6 / (fan_in + fan_out))`
where `fan_in` is the number of input units in the weight tensor
and `fan_out` is the number of output units in the weight tensor.
Args:
seed: A Python integer. Used to create random seeds. See
`tf.compat.v1.set_random_seed` for behavior.
dtype: Default data type, used if no `dtype` argument is provided when
calling the initializer. Only floating point types are supported.
References:
[Glorot et al., 2010](http://proceedings.mlr.press/v9/glorot10a.html)
([pdf](http://jmlr.org/proceedings/papers/v9/glorot10a/glorot10a.pdf))"
8631,get_config,tensorflow/tensorflow/python/ops/init_ops.py,1235,method,
8632,GlorotNormal,tensorflow/tensorflow/python/ops/init_ops.py,1242,class,"The Glorot normal initializer, also called Xavier normal initializer.
It draws samples from a truncated normal distribution centered on 0
with standard deviation (after truncation) given by
`stddev = sqrt(2 / (fan_in + fan_out))` where `fan_in` is the number
of input units in the weight tensor and `fan_out` is the number of
output units in the weight tensor.
Args:
seed: A Python integer. Used to create random seeds. See
`tf.compat.v1.set_random_seed` for behavior.
dtype: Default data type, used if no `dtype` argument is provided when
calling the initializer. Only floating point types are supported.
References:
[Glorot et al., 2010](http://proceedings.mlr.press/v9/glorot10a.html)
([pdf](http://jmlr.org/proceedings/papers/v9/glorot10a/glorot10a.pdf))"
8633,get_config,tensorflow/tensorflow/python/ops/init_ops.py,1268,method,
8634,lecun_normal,tensorflow/tensorflow/python/ops/init_ops.py,1295,function,"LeCun normal initializer.
It draws samples from a truncated normal distribution centered on 0
with standard deviation (after truncation) given by
`stddev = sqrt(1 / fan_in)` where `fan_in` is the number of
input units in the weight tensor.
Arguments:
seed: A Python integer. Used to seed the random generator.
Returns:
An initializer.
References:
- Self-Normalizing Neural Networks,
[Klambauer et al.,
2017](https://papers.nips.cc/paper/6698-self-normalizing-neural-networks)
# pylint: disable=line-too-long
([pdf](https://papers.nips.cc/paper/6698-self-normalizing-neural-networks.pdf))
- Efficient Backprop,
[Lecun et al., 1998](http://yann.lecun.com/exdb/publis/pdf/lecun-98b.pdf)"
8635,lecun_uniform,tensorflow/tensorflow/python/ops/init_ops.py,1323,function,"LeCun uniform initializer.
It draws samples from a uniform distribution within [-limit, limit]
where `limit` is `sqrt(3 / fan_in)`
where `fan_in` is the number of input units in the weight tensor.
Arguments:
seed: A Python integer. Used to seed the random generator.
Returns:
An initializer.
References:
- Self-Normalizing Neural Networks,
[Klambauer et al.,
2017](https://papers.nips.cc/paper/6698-self-normalizing-neural-networks)
# pylint: disable=line-too-long
([pdf](https://papers.nips.cc/paper/6698-self-normalizing-neural-networks.pdf))
- Efficient Backprop,
[Lecun et al., 1998](http://yann.lecun.com/exdb/publis/pdf/lecun-98b.pdf)"
8636,he_normal,tensorflow/tensorflow/python/ops/init_ops.py,1350,function,"He normal initializer.
It draws samples from a truncated normal distribution centered on 0
with standard deviation (after truncation) given by
`stddev = sqrt(2 / fan_in)` where `fan_in` is the number of
input units in the weight tensor.
Arguments:
seed: A Python integer. Used to seed the random generator.
Returns:
An initializer.
References:
[He et al., 2015]
(https://www.cv-foundation.org/openaccess/content_iccv_2015/html/He_Delving_Deep_into_ICCV_2015_paper.html)
# pylint: disable=line-too-long
([pdf](https://www.cv-foundation.org/openaccess/content_iccv_2015/papers/He_Delving_Deep_into_ICCV_2015_paper.pdf))"
8637,he_uniform,tensorflow/tensorflow/python/ops/init_ops.py,1375,function,"He uniform variance scaling initializer.
It draws samples from a uniform distribution within [-limit, limit]
where `limit` is `sqrt(6 / fan_in)`
where `fan_in` is the number of input units in the weight tensor.
Arguments:
seed: A Python integer. Used to seed the random generator.
Returns:
An initializer.
References:
[He et al., 2015]
(https://www.cv-foundation.org/openaccess/content_iccv_2015/html/He_Delving_Deep_into_ICCV_2015_paper.html)
# pylint: disable=line-too-long
([pdf](https://www.cv-foundation.org/openaccess/content_iccv_2015/papers/He_Delving_Deep_into_ICCV_2015_paper.pdf))"
8638,Initializer,tensorflow/tensorflow/python/ops/init_ops_v2.py,48,class,"Initializer base class: all initializers inherit from this class.
"
8639,get_config,tensorflow/tensorflow/python/ops/init_ops_v2.py,62,method,"Returns the configuration of the initializer as a JSON-serializable dict.
Returns:
A JSON-serializable Python dict."
8640,from_config,tensorflow/tensorflow/python/ops/init_ops_v2.py,71,method,"Instantiates an initializer from a configuration dictionary.
Example:
```python
initializer = RandomUniform(-1, 1)
config = initializer.get_config()
initializer = RandomUniform.from_config(config)
```
Args:
config: A Python dictionary.
It will typically be the output of `get_config`.
Returns:
An Initializer instance."
8641,Zeros,tensorflow/tensorflow/python/ops/init_ops_v2.py,94,class,"Initializer that generates tensors initialized to 0.
Initializers allow you to pre-specify an initialization strategy, encoded in
the Initializer object, without knowing the shape and dtype of the variable
being initialized.
Examples:
>>> def make_variables(k, initializer):
... return (tf.Variable(initializer(shape=[k], dtype=tf.float32)),
... tf.Variable(initializer(shape=[k, k], dtype=tf.float32)))
>>> v1, v2 = make_variables(3, tf.zeros_initializer())
>>> v1
<tf.Variable ... shape=(3,) ... numpy=array([0., 0., 0.], dtype=float32)>
>>> v2
<tf.Variable ... shape=(3, 3) ... numpy=
array([[0., 0., 0.],
[0., 0., 0.],
[0., 0., 0.]], dtype=float32)>
>>> make_variables(4, tf.random_uniform_initializer(minval=-1., maxval=1.))
(<tf.Variable...shape=(4,) dtype=float32...>, <tf.Variable...shape=(4, 4) ..."
8642,Ones,tensorflow/tensorflow/python/ops/init_ops_v2.py,136,class,"Initializer that generates tensors initialized to 1.
Initializers allow you to pre-specify an initialization strategy, encoded in
the Initializer object, without knowing the shape and dtype of the variable
being initialized.
Examples:
>>> def make_variables(k, initializer):
... return (tf.Variable(initializer(shape=[k], dtype=tf.float32)),
... tf.Variable(initializer(shape=[k, k], dtype=tf.float32)))
>>> v1, v2 = make_variables(3, tf.ones_initializer())
>>> v1
<tf.Variable ... shape=(3,) ... numpy=array([1., 1., 1.], dtype=float32)>
>>> v2
<tf.Variable ... shape=(3, 3) ... numpy=
array([[1., 1., 1.],
[1., 1., 1.],
[1., 1., 1.]], dtype=float32)>
>>> make_variables(4, tf.random_uniform_initializer(minval=-1., maxval=1.))
(<tf.Variable...shape=(4,) dtype=float32...>, <tf.Variable...shape=(4, 4) ..."
8643,Constant,tensorflow/tensorflow/python/ops/init_ops_v2.py,178,class,"Initializer that generates tensors with constant values.
Initializers allow you to pre-specify an initialization strategy, encoded in
the Initializer object, without knowing the shape and dtype of the variable
being initialized.
`tf.constant_initializer` returns an object which when called returns a tensor
populated with the `value` specified in the constructor. This `value` must be
convertible to the requested `dtype`.
The argument `value` can be a scalar constant value, or a list of
values. Scalars broadcast to whichever shape is requested from the
initializer.
If `value` is a list, then the length of the list must be equal to the number
of elements implied by the desired shape of the tensor. If the total number of
elements in `value` is not equal to the number of elements required by the
tensor shape, the initializer will raise a `TypeError`.
Examples:
>>> def make_variables(k, initializer):
... return (tf.Variable(initializer(shape=[k], dtype=tf.float32)),
... tf.Variable(initializer(shape=[k, k], dtype=tf.float32)))
>>> v1, v2 = make_variables(3, tf.constant_initializer(2.))
>>> v1
<tf.Variable ... shape=(3,) ... numpy=array([2., 2., 2.], dtype=float32)>
>>> v2
<tf.Variable ... shape=(3, 3) ... numpy=
array([[2., 2., 2.],
[2., 2., 2.],
[2., 2., 2.]], dtype=float32)>
>>> make_variables(4, tf.random_uniform_initializer(minval=-1., maxval=1.))
(<tf.Variable...shape=(4,) dtype=float32...>, <tf.Variable...shape=(4, 4) ...
>>> value = [0, 1, 2, 3, 4, 5, 6, 7]
>>> init = tf.constant_initializer(value)
>>> # Fitting shape
>>> tf.Variable(init(shape=[2, 4], dtype=tf.float32))
<tf.Variable ...
array([[0., 1., 2., 3.],
[4., 5., 6., 7.]], dtype=float32)>
>>> # Larger shape
>>> tf.Variable(init(shape=[3, 4], dtype=tf.float32))
Traceback (most recent call last):
...
TypeError: ...value has 8 elements, shape is (3, 4) with 12 elements...
>>> # Smaller shape
>>> tf.Variable(init(shape=[2, 3], dtype=tf.float32))
Traceback (most recent call last):
...
TypeError: ...value has 8 elements, shape is (2, 3) with 6 elements...
Args:
value: A Python scalar, list or tuple of values, or a N-dimensional numpy
array. All elements of the initialized variable will be set to the
corresponding value in the `value` argument.
Raises:
TypeError: If the input `value` is not one of the expected types."
8644,get_config,tensorflow/tensorflow/python/ops/init_ops_v2.py,265,method,
8645,RandomUniform,tensorflow/tensorflow/python/ops/init_ops_v2.py,270,class,"Initializer that generates tensors with a uniform distribution.
Initializers allow you to pre-specify an initialization strategy, encoded in
the Initializer object, without knowing the shape and dtype of the variable
being initialized.
Examples:
>>> def make_variables(k, initializer):
... return (tf.Variable(initializer(shape=[k], dtype=tf.float32)),
... tf.Variable(initializer(shape=[k, k], dtype=tf.float32)))
>>> v1, v2 = make_variables(3, tf.ones_initializer())
>>> v1
<tf.Variable ... shape=(3,) ... numpy=array([1., 1., 1.], dtype=float32)>
>>> v2
<tf.Variable ... shape=(3, 3) ... numpy=
array([[1., 1., 1.],
[1., 1., 1.],
[1., 1., 1.]], dtype=float32)>
>>> make_variables(4, tf.random_uniform_initializer(minval=-1., maxval=1.))
(<tf.Variable...shape=(4,) dtype=float32...>, <tf.Variable...shape=(4, 4) ...
Args:
minval: A python scalar or a scalar tensor. Lower bound of the range of
random values to generate (inclusive).
maxval: A python scalar or a scalar tensor. Upper bound of the range of
random values to generate (exclusive).
seed: A Python integer. Used to create random seeds. See
`tf.random.set_seed` for behavior."
8646,get_config,tensorflow/tensorflow/python/ops/init_ops_v2.py,325,method,
8647,RandomNormal,tensorflow/tensorflow/python/ops/init_ops_v2.py,334,class,"Initializer that generates tensors with a normal distribution.
Initializers allow you to pre-specify an initialization strategy, encoded in
the Initializer object, without knowing the shape and dtype of the variable
being initialized.
Examples:
>>> def make_variables(k, initializer):
... return (tf.Variable(initializer(shape=[k], dtype=tf.float32)),
... tf.Variable(initializer(shape=[k, k], dtype=tf.float32)))
>>> v1, v2 = make_variables(3,
... tf.random_normal_initializer(mean=1., stddev=2.))
>>> v1
<tf.Variable ... shape=(3,) ... numpy=array([...], dtype=float32)>
>>> v2
<tf.Variable ... shape=(3, 3) ... numpy=
...
>>> make_variables(4, tf.random_uniform_initializer(minval=-1., maxval=1.))
(<tf.Variable...shape=(4,) dtype=float32...>, <tf.Variable...shape=(4, 4) ...
Args:
mean: a python scalar or a scalar tensor. Mean of the random values to
generate.
stddev: a python scalar or a scalar tensor. Standard deviation of the random
values to generate.
seed: A Python integer. Used to create random seeds. See
`tf.random.set_seed` for behavior."
8648,get_config,tensorflow/tensorflow/python/ops/init_ops_v2.py,387,method,
8649,TruncatedNormal,tensorflow/tensorflow/python/ops/init_ops_v2.py,395,class,"Initializer that generates a truncated normal distribution.
Initializers allow you to pre-specify an initialization strategy, encoded in
the Initializer object, without knowing the shape and dtype of the variable
being initialized.
These values are similar to values from a `tf.initializers.RandomNormal`
except that values more than two standard deviations from the mean are
discarded and re-drawn. This is the recommended initializer for neural network
weights and filters.
Examples:
>>> def make_variables(k, initializer):
... return (tf.Variable(initializer(shape=[k], dtype=tf.float32)),
... tf.Variable(initializer(shape=[k, k], dtype=tf.float32)))
>>> v1, v2 = make_variables(
... 3, tf.initializers.TruncatedNormal(mean=1., stddev=2.))
>>> v1
<tf.Variable ... shape=(3,) ... numpy=array([...], dtype=float32)>
>>> v2
<tf.Variable ... shape=(3, 3) ... numpy=
...
>>> make_variables(4, tf.initializers.RandomUniform(minval=-1., maxval=1.))
(<tf.Variable...shape=(4,) dtype=float32...>, <tf.Variable...shape=(4, 4) ...
Args:
mean: a python scalar or a scalar tensor. Mean of the random values
to generate.
stddev: a python scalar or a scalar tensor. Standard deviation of the
random values to generate.
seed: A Python integer. Used to create random seeds. See
`tf.random.set_seed` for behavior."
8650,get_config,tensorflow/tensorflow/python/ops/init_ops_v2.py,452,method,
8651,VarianceScaling,tensorflow/tensorflow/python/ops/init_ops_v2.py,460,class,"Initializer capable of adapting its scale to the shape of weights tensors.
Initializers allow you to pre-specify an initialization strategy, encoded in
the Initializer object, without knowing the shape and dtype of the variable
being initialized.
With `distribution=""truncated_normal"" or ""untruncated_normal""`, samples are
drawn from a truncated/untruncated normal distribution with a mean of zero and
a standard deviation (after truncation, if used) `stddev = sqrt(scale / n)`
where n is:
- number of input units in the weight tensor, if mode = ""fan_in""
- number of output units, if mode = ""fan_out""
- average of the numbers of input and output units, if mode = ""fan_avg""
With `distribution=""uniform""`, samples are drawn from a uniform distribution
within [-limit, limit], with `limit = sqrt(3 * scale / n)`.
Examples:
>>> def make_variables(k, initializer):
... return (tf.Variable(initializer(shape=[k], dtype=tf.float32)),
... tf.Variable(initializer(shape=[k, k], dtype=tf.float32)))
>>> v1, v2 = make_variables(3, tf.initializers.VarianceScaling(scale=1.))
>>> v1
<tf.Variable ... shape=(3,) ... numpy=array([...], dtype=float32)>
>>> v2
<tf.Variable ... shape=(3, 3) ... numpy=
...
>>> make_variables(4, tf.initializers.VarianceScaling(distribution='uniform'))
(<tf.Variable...shape=(4,) dtype=float32...>, <tf.Variable...shape=(4, 4) ...
Args:
scale: Scaling factor (positive float).
mode: One of ""fan_in"", ""fan_out"", ""fan_avg"".
distribution: Random distribution to use. One of ""truncated_normal"",
""untruncated_normal"" and ""uniform"".
seed: A Python integer. Used to create random seeds. See
`tf.random.set_seed` for behavior.
Raises:
ValueError: In case of an invalid value for the ""scale"", mode"" or
""distribution"" arguments."
8652,get_config,tensorflow/tensorflow/python/ops/init_ops_v2.py,563,method,
8653,Orthogonal,tensorflow/tensorflow/python/ops/init_ops_v2.py,572,class,"Initializer that generates an orthogonal matrix.
Initializers allow you to pre-specify an initialization strategy, encoded in
the Initializer object, without knowing the shape and dtype of the variable
being initialized.
If the shape of the tensor to initialize is two-dimensional, it is initialized
with an orthogonal matrix obtained from the QR decomposition of a matrix of
random numbers drawn from a normal distribution.
If the matrix has fewer rows than columns then the output will have orthogonal
rows. Otherwise, the output will have orthogonal columns.
If the shape of the tensor to initialize is more than two-dimensional,
a matrix of shape `(shape[0] * ... * shape[n - 2], shape[n - 1])`
is initialized, where `n` is the length of the shape vector.
The matrix is subsequently reshaped to give a tensor of the desired shape.
Examples:
>>> def make_variables(k, initializer):
... return (tf.Variable(initializer(shape=[k, k], dtype=tf.float32)),
... tf.Variable(initializer(shape=[k, k, k], dtype=tf.float32)))
>>> v1, v2 = make_variables(3, tf.initializers.Orthogonal())
>>> v1
<tf.Variable ... shape=(3, 3) ...
>>> v2
<tf.Variable ... shape=(3, 3, 3) ...
>>> make_variables(4, tf.initializers.Orthogonal(gain=0.5))
(<tf.Variable ... shape=(4, 4) dtype=float32...
<tf.Variable ... shape=(4, 4, 4) dtype=float32...
Args:
gain: multiplicative factor to apply to the orthogonal matrix
seed: A Python integer. Used to create random seeds. See
`tf.random.set_seed` for behavior.
References:
[Saxe et al., 2014](https://openreview.net/forum?id=_wzZwKpTDF_9C)
([pdf](https://arxiv.org/pdf/1312.6120.pdf))"
8654,get_config,tensorflow/tensorflow/python/ops/init_ops_v2.py,655,method,
8655,Identity,tensorflow/tensorflow/python/ops/init_ops_v2.py,659,class,"Initializer that generates the identity matrix.
Initializers allow you to pre-specify an initialization strategy, encoded in
the Initializer object, without knowing the shape and dtype of the variable
being initialized.
Only usable for generating 2D matrices.
Examples:
>>> def make_variable(k, initializer):
... return tf.Variable(initializer(shape=[k, k], dtype=tf.float32))
>>> make_variable(2, tf.initializers.Identity())
<tf.Variable ... shape=(2, 2) dtype=float32, numpy=
array([[1., 0.],
[0., 1.]], dtype=float32)>
>>> make_variable(3, tf.initializers.Identity(gain=0.5))
<tf.Variable ... shape=(3, 3) dtype=float32, numpy=
array([[0.5, 0. , 0. ],
[0. , 0.5, 0. ],
[0. , 0. , 0.5]], dtype=float32)>
Args:
gain: Multiplicative factor to apply to the identity matrix."
8656,get_config,tensorflow/tensorflow/python/ops/init_ops_v2.py,713,method,
8657,GlorotUniform,tensorflow/tensorflow/python/ops/init_ops_v2.py,717,class,"The Glorot uniform initializer, also called Xavier uniform initializer.
Initializers allow you to pre-specify an initialization strategy, encoded in
the Initializer object, without knowing the shape and dtype of the variable
being initialized.
Draws samples from a uniform distribution within [-limit, limit] where `limit`
is `sqrt(6 / (fan_in + fan_out))` where `fan_in` is the number of input units
in the weight tensor and `fan_out` is the number of output units in the weight
tensor.
Examples:
>>> def make_variables(k, initializer):
... return (tf.Variable(initializer(shape=[k, k], dtype=tf.float32)),
... tf.Variable(initializer(shape=[k, k, k], dtype=tf.float32)))
>>> v1, v2 = make_variables(3, tf.initializers.GlorotUniform())
>>> v1
<tf.Variable ... shape=(3, 3) ...
>>> v2
<tf.Variable ... shape=(3, 3, 3) ...
>>> make_variables(4, tf.initializers.RandomNormal())
(<tf.Variable ... shape=(4, 4) dtype=float32...
<tf.Variable ... shape=(4, 4, 4) dtype=float32...
Args:
seed: A Python integer. Used to create random seeds. See
`tf.random.set_seed` for behavior.
References:
[Glorot et al., 2010](http://proceedings.mlr.press/v9/glorot10a.html)
([pdf](http://jmlr.org/proceedings/papers/v9/glorot10a/glorot10a.pdf))"
8658,get_config,tensorflow/tensorflow/python/ops/init_ops_v2.py,759,method,
8659,GlorotNormal,tensorflow/tensorflow/python/ops/init_ops_v2.py,763,class,"The Glorot normal initializer, also called Xavier normal initializer.
Initializers allow you to pre-specify an initialization strategy, encoded in
the Initializer object, without knowing the shape and dtype of the variable
being initialized.
Draws samples from a truncated normal distribution centered on 0 with `stddev
= sqrt(2 / (fan_in + fan_out))` where `fan_in` is the number of input units in
the weight tensor and `fan_out` is the number of output units in the weight
tensor.
Examples:
>>> def make_variables(k, initializer):
... return (tf.Variable(initializer(shape=[k, k], dtype=tf.float32)),
... tf.Variable(initializer(shape=[k, k, k], dtype=tf.float32)))
>>> v1, v2 = make_variables(3, tf.initializers.GlorotNormal())
>>> v1
<tf.Variable ... shape=(3, 3) ...
>>> v2
<tf.Variable ... shape=(3, 3, 3) ...
>>> make_variables(4, tf.initializers.RandomNormal())
(<tf.Variable ... shape=(4, 4) dtype=float32...
<tf.Variable ... shape=(4, 4, 4) dtype=float32...
Args:
seed: A Python integer. Used to create random seeds. See
`tf.random.set_seed` for behavior.
References:
[Glorot et al., 2010](http://proceedings.mlr.press/v9/glorot10a.html)
([pdf](http://jmlr.org/proceedings/papers/v9/glorot10a/glorot10a.pdf))"
8660,get_config,tensorflow/tensorflow/python/ops/init_ops_v2.py,805,method,
8661,lecun_normal,tensorflow/tensorflow/python/ops/init_ops_v2.py,826,function,"LeCun normal initializer.
Initializers allow you to pre-specify an initialization strategy, encoded in
the Initializer object, without knowing the shape and dtype of the variable
being initialized.
Draws samples from a truncated normal distribution centered on 0 with `stddev
= sqrt(1 / fan_in)` where `fan_in` is the number of input units in the weight
tensor.
Examples:
>>> def make_variables(k, initializer):
... return (tf.Variable(initializer(shape=[k, k], dtype=tf.float32)),
... tf.Variable(initializer(shape=[k, k, k], dtype=tf.float32)))
>>> v1, v2 = make_variables(3, tf.initializers.lecun_normal())
>>> v1
<tf.Variable ... shape=(3, 3) ...
>>> v2
<tf.Variable ... shape=(3, 3, 3) ...
>>> make_variables(4, tf.initializers.RandomNormal())
(<tf.Variable ... shape=(4, 4) dtype=float32...
<tf.Variable ... shape=(4, 4, 4) dtype=float32...
Arguments:
seed: A Python integer. Used to seed the random generator.
Returns:
A callable Initializer with `shape` and `dtype` arguments which generates a
tensor.
References:
- Self-Normalizing Neural Networks,
[Klambauer et al., 2017]
(https://papers.nips.cc/paper/6698-self-normalizing-neural-networks)
([pdf]
(https://papers.nips.cc/paper/6698-self-normalizing-neural-networks.pdf))
- Efficient Backprop,
[Lecun et al., 1998](http://yann.lecun.com/exdb/publis/pdf/lecun-98b.pdf)"
8662,lecun_uniform,tensorflow/tensorflow/python/ops/init_ops_v2.py,871,function,"LeCun uniform initializer.
Initializers allow you to pre-specify an initialization strategy, encoded in
the Initializer object, without knowing the shape and dtype of the variable
being initialized.
Draws samples from a uniform distribution within [-limit, limit] where `limit`
is `sqrt(3 / fan_in)` where `fan_in` is the number of input units in the
weight tensor.
Examples:
>>> def make_variables(k, initializer):
... return (tf.Variable(initializer(shape=[k, k], dtype=tf.float32)),
... tf.Variable(initializer(shape=[k, k, k], dtype=tf.float32)))
>>> v1, v2 = make_variables(3, tf.initializers.lecun_uniform())
>>> v1
<tf.Variable ... shape=(3, 3) ...
>>> v2
<tf.Variable ... shape=(3, 3, 3) ...
>>> make_variables(4, tf.initializers.RandomNormal())
(<tf.Variable ... shape=(4, 4) dtype=float32...
<tf.Variable ... shape=(4, 4, 4) dtype=float32...
Arguments:
seed: A Python integer. Used to seed the random generator.
Returns:
A callable Initializer with `shape` and `dtype` arguments which generates a
tensor.
References:
- Self-Normalizing Neural Networks,
[Klambauer et al., 2017](https://papers.nips.cc/paper/6698-self-normalizing-neural-networks) # pylint: disable=line-too-long
([pdf](https://papers.nips.cc/paper/6698-self-normalizing-neural-networks.pdf))
- Efficient Backprop,
[Lecun et al., 1998](http://yann.lecun.com/exdb/publis/pdf/lecun-98b.pdf)"
8663,he_normal,tensorflow/tensorflow/python/ops/init_ops_v2.py,914,function,"He normal initializer.
Initializers allow you to pre-specify an initialization strategy, encoded in
the Initializer object, without knowing the shape and dtype of the variable
being initialized.
It draws samples from a truncated normal distribution centered on 0 with
`stddev = sqrt(2 / fan_in)` where `fan_in` is the number of input units in the
weight tensor.
Examples:
>>> def make_variables(k, initializer):
... return (tf.Variable(initializer(shape=[k, k], dtype=tf.float32)),
... tf.Variable(initializer(shape=[k, k, k], dtype=tf.float32)))
>>> v1, v2 = make_variables(3, tf.initializers.he_normal())
>>> v1
<tf.Variable ... shape=(3, 3) ...
>>> v2
<tf.Variable ... shape=(3, 3, 3) ...
>>> make_variables(4, tf.initializers.RandomNormal())
(<tf.Variable ... shape=(4, 4) dtype=float32...
<tf.Variable ... shape=(4, 4, 4) dtype=float32...
Arguments:
seed: A Python integer. Used to seed the random generator.
Returns:
A callable Initializer with `shape` and `dtype` arguments which generates a
tensor.
References:
[He et al., 2015](https://www.cv-foundation.org/openaccess/content_iccv_2015/html/He_Delving_Deep_into_ICCV_2015_paper.html) # pylint: disable=line-too-long
([pdf](https://www.cv-foundation.org/openaccess/content_iccv_2015/papers/He_Delving_Deep_into_ICCV_2015_paper.pdf))"
8664,he_uniform,tensorflow/tensorflow/python/ops/init_ops_v2.py,954,function,"He uniform variance scaling initializer.
Initializers allow you to pre-specify an initialization strategy, encoded in
the Initializer object, without knowing the shape and dtype of the variable
being initialized.
Draws samples from a uniform distribution within [-limit, limit] where `limit`
is `sqrt(6 / fan_in)` where `fan_in` is the number of input units in the
weight tensor.
Examples:
>>> def make_variables(k, initializer):
... return (tf.Variable(initializer(shape=[k, k], dtype=tf.float32)),
... tf.Variable(initializer(shape=[k, k, k], dtype=tf.float32)))
>>> v1, v2 = make_variables(3, tf.initializers.he_uniform())
>>> v1
<tf.Variable ... shape=(3, 3) ...
>>> v2
<tf.Variable ... shape=(3, 3, 3) ...
>>> make_variables(4, tf.initializers.RandomNormal())
(<tf.Variable ... shape=(4, 4) dtype=float32...
<tf.Variable ... shape=(4, 4, 4) dtype=float32...
Arguments:
seed: A Python integer. Used to seed the random generator.
Returns:
A callable Initializer with `shape` and `dtype` arguments which generates a
tensor.
References:
[He et al., 2015](https://www.cv-foundation.org/openaccess/content_iccv_2015/html/He_Delving_Deep_into_ICCV_2015_paper.html) # pylint: disable=line-too-long
([pdf](https://www.cv-foundation.org/openaccess/content_iccv_2015/papers/He_Delving_Deep_into_ICCV_2015_paper.pdf))"
8665,MethodInitializers,tensorflow/tensorflow/python/ops/init_ops_v2_test.py,469,class,
8666,alias_inplace_update,tensorflow/tensorflow/python/ops/inplace_ops.py,71,function,"Applies an inplace update on input x at index i with value v. Aliases x.
If i is None, x and v must be the same shape. Computes
x = v;
If i is a scalar, x has a rank 1 higher than v's. Computes
x[i, :] = v;
Otherwise, x and v must have the same rank. Computes
x[i, :] = v;
Args:
x: A Tensor.
i: None, a scalar or a vector.
v: A Tensor.
Returns:
Returns x."
8667,alias_inplace_add,tensorflow/tensorflow/python/ops/inplace_ops.py,97,function,"Applies an inplace add on input x at index i with value v. Aliases x.
If i is None, x and v must be the same shape. Computes
x += v;
If i is a scalar, x has a rank 1 higher than v's. Computes
x[i, :] += v;
Otherwise, x and v must have the same rank. Computes
x[i, :] += v;
Args:
x: A Tensor.
i: None, a scalar or a vector.
v: A Tensor.
Returns:
Returns x."
8668,alias_inplace_sub,tensorflow/tensorflow/python/ops/inplace_ops.py,123,function,"Applies an inplace sub on input x at index i with value v. Aliases x.
If i is None, x and v must be the same shape. Computes
x -= v;
If i is a scalar, x has a rank 1 higher than v's. Computes
x[i, :] -= v;
Otherwise, x and v must have the same rank. Computes
x[i, :] -= v;
Args:
x: A Tensor.
i: None, a scalar or a vector.
v: A Tensor.
Returns:
Returns x."
8669,empty_like,tensorflow/tensorflow/python/ops/inplace_ops.py,145,function,"Returns a non-initialized tensor with the same shape and dtype as x.
Args:
x: A Tensor.
init: Initialize the returned tensor with the default value of
x.dtype(), if True. Otherwise, do not initialize. Defaults to
None.
Returns:
A tensor y, whose dtype and shape are the same as those of x.
y is guaranteed not to be an alias of x. Upon return, y may contain
arbitrary data."
8670,inplace_update,tensorflow/tensorflow/python/ops/inplace_ops.py,168,function,"Applies an inplace update on input x at index i with value v.
Note that this function is not actually inplace - it allocates
a copy of x. The utility is not avoiding memory copies but rather
specifying a sparse update.
If i is None, x and v must be the same shape. Computes
y = x; y = v;
If i is a scalar, x has a rank 1 higher than v's. Computes
y = x; y[i, :] = v;
Otherwise, x and v must have the same rank. Computes
y = x; y[i, :] = v;
Args:
x: A Tensor.
i: None, a scalar or a vector.
v: A Tensor.
Returns:
Returns y, which is guaranteed not to be an alias of x."
8671,inplace_add,tensorflow/tensorflow/python/ops/inplace_ops.py,198,function,"Applies an inplace add on input x at index i with value v.
Note that this function is not actually inplace - it allocates
a copy of x. The utility is not avoiding memory copies but rather
specifying a sparse update.
If i is None, x and v must be the same shape. Computes
y = x; y += v;
If i is a scalar, x has a rank 1 higher than v's. Computes
y = x; y[i, :] += v;
Otherwise, x and v must have the same rank. Computes
y = x; y[i, :] += v;
Args:
x: A Tensor.
i: None, a scalar or a vector.
v: A Tensor.
Returns:
Returns y, which is guaranteed not to be an alias of x."
8672,inplace_sub,tensorflow/tensorflow/python/ops/inplace_ops.py,228,function,"Applies an inplace sub on input x at index i with value v.
Note that this function is not actually inplace - it allocates
a copy of x. The utility is not avoiding memory copies but rather
specifying a sparse update.
If i is None, x and v must be the same shape. Computes
y = x; y -= v;
If i is a scalar, x has a rank 1 higher than v's. Computes
y = x; y[i, :] -= v;
Otherwise, x and v must have the same rank. Computes
y = x; y[i, :] -= v;
Args:
x: A Tensor.
i: None, a scalar or a vector.
v: A Tensor.
Returns:
Returns y, which is guaranteed not to be an alias of x."
8673,ReaderBase,tensorflow/tensorflow/python/ops/io_ops.py,100,class,"Base class for different Reader types, that produce a record every step.
Conceptually, Readers convert string 'work units' into records (key,
value pairs). Typically the 'work units' are filenames and the
records are extracted from the contents of those files. We want a
single record produced per step, but a work unit can correspond to
many records.
Therefore we introduce some decoupling using a queue. The queue
contains the work units and the Reader dequeues from the queue when
it is asked to produce a record (via Read()) but it has finished the
last work unit.
@compatibility(eager)
Readers are not compatible with eager execution. Instead, please
use `tf.data` to get data into your model.
@end_compatibility"
8674,reader_ref,tensorflow/tensorflow/python/ops/io_ops.py,140,method,Op that implements the reader.
8675,read,tensorflow/tensorflow/python/ops/io_ops.py,144,method,"Returns the next record (key, value) pair produced by a reader.
Will dequeue a work unit from queue if necessary (e.g. when the
Reader needs to start reading from a new file since it has
finished with the previous file).
Args:
queue: A Queue or a mutable string Tensor representing a handle
to a Queue, with string work items.
name: A name for the operation (optional).
Returns:
A tuple of Tensors (key, value).
key: A string scalar Tensor.
value: A string scalar Tensor."
8676,read_up_to,tensorflow/tensorflow/python/ops/io_ops.py,173,method,"Returns up to num_records (key, value) pairs produced by a reader.
Will dequeue a work unit from queue if necessary (e.g., when the
Reader needs to start reading from a new file since it has
finished with the previous file).
It may return less than num_records even before the last batch.
Args:
queue: A Queue or a mutable string Tensor representing a handle
to a Queue, with string work items.
num_records: Number of records to read.
name: A name for the operation (optional).
Returns:
A tuple of Tensors (keys, values).
keys: A 1-D string Tensor.
values: A 1-D string Tensor."
8677,num_records_produced,tensorflow/tensorflow/python/ops/io_ops.py,211,method,"Returns the number of records this reader has produced.
This is the same as the number of Read executions that have
succeeded.
Args:
name: A name for the operation (optional).
Returns:
An int64 Tensor."
8678,num_work_units_completed,tensorflow/tensorflow/python/ops/io_ops.py,231,method,"Returns the number of work units this reader has finished processing.
Args:
name: A name for the operation (optional).
Returns:
An int64 Tensor."
8679,serialize_state,tensorflow/tensorflow/python/ops/io_ops.py,247,method,"Produce a string tensor that encodes the state of a reader.
Not all Readers support being serialized, so this can produce an
Unimplemented error.
Args:
name: A name for the operation (optional).
Returns:
A string Tensor."
8680,restore_state,tensorflow/tensorflow/python/ops/io_ops.py,264,method,"Restore a reader to a previously saved state.
Not all Readers support being restored, so this can produce an
Unimplemented error.
Args:
state: A string Tensor.
Result of a SerializeState of a Reader with matching type.
name: A name for the operation (optional).
Returns:
The created Operation."
8681,supports_serialize,tensorflow/tensorflow/python/ops/io_ops.py,285,method,Whether the Reader implementation can serialize its state.
8682,reset,tensorflow/tensorflow/python/ops/io_ops.py,289,method,"Restore a reader to its initial clean state.
Args:
name: A name for the operation (optional).
Returns:
The created Operation."
8683,WholeFileReader,tensorflow/tensorflow/python/ops/io_ops.py,314,class,"A Reader that outputs the entire contents of a file as a value.
To use, enqueue filenames in a Queue. The output of Read will
be a filename (key) and the contents of that file (value).
See ReaderBase for supported methods.
@compatibility(eager)
Readers are not compatible with eager execution. Instead, please
use `tf.data` to get data into your model.
@end_compatibility"
8684,TextLineReader,tensorflow/tensorflow/python/ops/io_ops.py,345,class,"A Reader that outputs the lines of a file delimited by newlines.
Newlines are stripped from the output.
See ReaderBase for supported methods.
@compatibility(eager)
Readers are not compatible with eager execution. Instead, please
use `tf.data` to get data into your model.
@end_compatibility"
8685,FixedLengthRecordReader,tensorflow/tensorflow/python/ops/io_ops.py,378,class,"A Reader that outputs fixed-length records from a file.
See ReaderBase for supported methods.
@compatibility(eager)
Readers are not compatible with eager execution. Instead, please
use `tf.data` to get data into your model.
@end_compatibility"
8686,TFRecordReader,tensorflow/tensorflow/python/ops/io_ops.py,424,class,"A Reader that outputs the records from a TFRecords file.
See ReaderBase for supported methods.
@compatibility(eager)
Readers are not compatible with eager execution. Instead, please
use `tf.data` to get data into your model.
@end_compatibility"
8687,LMDBReader,tensorflow/tensorflow/python/ops/io_ops.py,458,class,"A Reader that outputs the records from a LMDB file.
See ReaderBase for supported methods.
@compatibility(eager)
Readers are not compatible with eager execution. Instead, please
use `tf.data` to get data into your model.
@end_compatibility"
8688,IdentityReader,tensorflow/tensorflow/python/ops/io_ops.py,488,class,"A Reader that outputs the queued work as both the key and value.
To use, enqueue strings in a Queue. Read will take the front
work string and output (work, work).
See ReaderBase for supported methods.
@compatibility(eager)
Readers are not compatible with eager execution. Instead, please
use `tf.data` to get data into your model.
@end_compatibility"
8689,matrix_triangular_solve,tensorflow/tensorflow/python/ops/linalg_ops.py,87,function,"Solve systems of linear equations with upper or lower triangular matrices.
`matrix` is a tensor of shape `[..., M, M]` whose inner-most 2 dimensions form
square matrices. If `lower` is `True` then the strictly upper triangular part
of each inner-most matrix is assumed to be zero and not accessed. If `lower`
is `False` then the strictly lower triangular part of each inner-most matrix
is assumed to be zero and not accessed. `rhs` is a tensor of shape
`[..., M, N]`.
The output is a tensor of shape `[..., M, N]`. If `adjoint` is `True` then the
innermost matrices in output satisfy matrix equations `
sum_k matrix[..., i, k] * output[..., k, j] = rhs[..., i, j]`.
If `adjoint` is `False` then the
innermost matrices in output satisfy matrix equations
`sum_k adjoint(matrix[..., i, k]) * output[..., k, j] = rhs[..., i, j]`.
Example:
>>> a = tf.constant([[3, 0, 0, 0],
... [2, 1, 0, 0],
... [1, 0, 1, 0],
... [1, 1, 1, 1]], dtype=tf.float32)
>>> b = tf.constant([[4], [2], [4], [2]], dtype=tf.float32)
>>> x = tf.linalg.triangular_solve(a, b, lower=True)
>>> x
<tf.Tensor: shape=(4, 1), dtype=float32, numpy=
array([[ 1.3333334 ],
[-0.66666675],
[ 2.6666665 ],
[-1.3333331 ]], dtype=float32)>
>>> tf.matmul(a, x)
<tf.Tensor: shape=(4, 1), dtype=float32, numpy=
array([[4.],
[2.],
[4.],
[2.]], dtype=float32)>
Args:
matrix: A `Tensor`. Must be one of the following types: `float64`,
`float32`, `half`, `complex64`, `complex128`. Shape is `[..., M, M]`.
rhs: A `Tensor`. Must have the same type as `matrix`. Shape is `[..., M,
N]`.
lower: An optional `bool`. Defaults to `True`. Boolean indicating whether
the innermost matrices in matrix are lower or upper triangular.
adjoint: An optional `bool`. Defaults to `False`. Boolean indicating whether
to solve with matrix or its (block-wise) adjoint.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as matrix, and shape is `[..., M, N]`."
8690,cholesky_solve,tensorflow/tensorflow/python/ops/linalg_ops.py,150,function,"Solves systems of linear eqns `A X = RHS`, given Cholesky factorizations.
```python
# Solve 10 separate 2x2 linear systems:
A = ... # shape 10 x 2 x 2
RHS = ... # shape 10 x 2 x 1
chol = tf.linalg.cholesky(A) # shape 10 x 2 x 2
X = tf.linalg.cholesky_solve(chol, RHS) # shape 10 x 2 x 1
# tf.matmul(A, X) ~ RHS
X[3, :, 0] # Solution to the linear system A[3, :, :] x = RHS[3, :, 0]
# Solve five linear systems (K = 5) for every member of the length 10 batch.
A = ... # shape 10 x 2 x 2
RHS = ... # shape 10 x 2 x 5
...
X[3, :, 2] # Solution to the linear system A[3, :, :] x = RHS[3, :, 2]
```
Args:
chol: A `Tensor`. Must be `float32` or `float64`, shape is `[..., M, M]`.
Cholesky factorization of `A`, e.g. `chol = tf.linalg.cholesky(A)`.
For that reason, only the lower triangular parts (including the diagonal)
of the last two dimensions of `chol` are used. The strictly upper part is
assumed to be zero and not accessed.
rhs: A `Tensor`, same type as `chol`, shape is `[..., M, K]`.
name: A name to give this `Op`. Defaults to `cholesky_solve`.
Returns:
Solution to `A x = rhs`, shape `[..., M, K]`."
8691,eye,tensorflow/tensorflow/python/ops/linalg_ops.py,194,function,"Construct an identity matrix, or a batch of matrices.
See also `tf.ones`, `tf.zeros`, `tf.fill`, `tf.one_hot`.
```python
# Construct one identity matrix.
tf.eye(2)
==> [[1., 0.],
[0., 1.]]
# Construct a batch of 3 identity matrices, each 2 x 2.
# batch_identity[i, :, :] is a 2 x 2 identity matrix, i = 0, 1, 2.
batch_identity = tf.eye(2, batch_shape=[3])
# Construct one 2 x 3 ""identity"" matrix
tf.eye(2, num_columns=3)
==> [[ 1., 0., 0.],
[ 0., 1., 0.]]
```
Args:
num_rows: Non-negative `int32` scalar `Tensor` giving the number of rows
in each batch matrix.
num_columns: Optional non-negative `int32` scalar `Tensor` giving the number
of columns in each batch matrix. Defaults to `num_rows`.
batch_shape: A list or tuple of Python integers or a 1-D `int32` `Tensor`.
If provided, the returned `Tensor` will have leading batch dimensions of
this shape.
dtype: The type of an element in the resulting `Tensor`
name: A name for this `Op`. Defaults to ""eye"".
Returns:
A `Tensor` of shape `batch_shape + [num_rows, num_columns]`"
8692,matrix_solve_ls,tensorflow/tensorflow/python/ops/linalg_ops.py,243,function,"Solves one or more linear least-squares problems.
`matrix` is a tensor of shape `[..., M, N]` whose inner-most 2 dimensions
form `M`-by-`N` matrices. Rhs is a tensor of shape `[..., M, K]` whose
inner-most 2 dimensions form `M`-by-`K` matrices. The computed output is a
`Tensor` of shape `[..., N, K]` whose inner-most 2 dimensions form `M`-by-`K`
matrices that solve the equations
`matrix[..., :, :] * output[..., :, :] = rhs[..., :, :]` in the least squares
sense.
Below we will use the following notation for each pair of matrix and
right-hand sides in the batch:
`matrix`=\\(A \in \Re^{m \times n}\\),
`rhs`=\\(B \in \Re^{m \times k}\\),
`output`=\\(X \in \Re^{n \times k}\\),
`l2_regularizer`=\\(\lambda\\).
If `fast` is `True`, then the solution is computed by solving the normal
equations using Cholesky decomposition. Specifically, if \\(m \ge n\\) then
\\(X = (A^T A + \lambda I)^{-1} A^T B\\), which solves the least-squares
problem \\(X = \mathrm{argmin}_{Z \in \Re^{n \times k}} ||A Z - B||_F^2 +
\lambda ||Z||_F^2\\). If \\(m \lt n\\) then `output` is computed as
\\(X = A^T (A A^T + \lambda I)^{-1} B\\), which (for \\(\lambda = 0\\)) is
the minimum-norm solution to the under-determined linear system, i.e.
\\(X = \mathrm{argmin}_{Z \in \Re^{n \times k}} ||Z||_F^2 \\), subject to
\\(A Z = B\\). Notice that the fast path is only numerically stable when
\\(A\\) is numerically full rank and has a condition number
\\(\mathrm{cond}(A) \lt \frac{1}{\sqrt{\epsilon_{mach}}}\\) or\\(\lambda\\)
is sufficiently large.
If `fast` is `False` an algorithm based on the numerically robust complete
orthogonal decomposition is used. This computes the minimum-norm
least-squares solution, even when \\(A\\) is rank deficient. This path is
typically 6-7 times slower than the fast path. If `fast` is `False` then
`l2_regularizer` is ignored.
Args:
matrix: `Tensor` of shape `[..., M, N]`.
rhs: `Tensor` of shape `[..., M, K]`.
l2_regularizer: 0-D `double` `Tensor`. Ignored if `fast=False`.
fast: bool. Defaults to `True`.
name: string, optional name of the operation.
Returns:
output: `Tensor` of shape `[..., N, K]` whose inner-most 2 dimensions form
`M`-by-`K` matrices that solve the equations
`matrix[..., :, :] * output[..., :, :] = rhs[..., :, :]` in the least
squares sense.
Raises:
NotImplementedError: linalg.lstsq is currently disabled for complex128
and l2_regularizer != 0 due to poor accuracy."
8693,eig,tensorflow/tensorflow/python/ops/linalg_ops.py,380,function,"Computes the eigen decomposition of a batch of matrices.
The eigenvalues
and eigenvectors for a non-Hermitian matrix in general are complex. The
eigenvectors are not guaranteed to be linearly independent.
Computes the eigenvalues and right eigenvectors of the innermost
N-by-N matrices in `tensor` such that
`tensor[...,:,:] * v[..., :,i] = e[..., i] * v[...,:,i]`, for i=0...N-1.
Args:
tensor: `Tensor` of shape `[..., N, N]`. Only the lower triangular part of
each inner inner matrix is referenced.
name: string, optional name of the operation.
Returns:
e: Eigenvalues. Shape is `[..., N]`. Sorted in non-decreasing order.
v: Eigenvectors. Shape is `[..., N, N]`. The columns of the inner most
matrices contain eigenvectors of the corresponding matrices in `tensor`"
8694,eigvals,tensorflow/tensorflow/python/ops/linalg_ops.py,411,function,"Computes the eigenvalues of one or more matrices.
Note: If your program backpropagates through this function, you should replace
it with a call to tf.linalg.eig (possibly ignoring the second output) to
avoid computing the eigen decomposition twice. This is because the
eigenvectors are used to compute the gradient w.r.t. the eigenvalues. See
_SelfAdjointEigV2Grad in linalg_grad.py.
Args:
tensor: `Tensor` of shape `[..., N, N]`.
name: string, optional name of the operation.
Returns:
e: Eigenvalues. Shape is `[..., N]`. The vector `e[..., :]` contains the `N`
eigenvalues of `tensor[..., :, :]`."
8695,self_adjoint_eig,tensorflow/tensorflow/python/ops/linalg_ops.py,439,function,"Computes the eigen decomposition of a batch of self-adjoint matrices.
Computes the eigenvalues and eigenvectors of the innermost N-by-N matrices
in `tensor` such that
`tensor[...,:,:] * v[..., :,i] = e[..., i] * v[...,:,i]`, for i=0...N-1.
Args:
tensor: `Tensor` of shape `[..., N, N]`. Only the lower triangular part of
each inner inner matrix is referenced.
name: string, optional name of the operation.
Returns:
e: Eigenvalues. Shape is `[..., N]`. Sorted in non-decreasing order.
v: Eigenvectors. Shape is `[..., N, N]`. The columns of the inner most
matrices contain eigenvectors of the corresponding matrices in `tensor`"
8696,self_adjoint_eigvals,tensorflow/tensorflow/python/ops/linalg_ops.py,463,function,"Computes the eigenvalues of one or more self-adjoint matrices.
Note: If your program backpropagates through this function, you should replace
it with a call to tf.linalg.eigh (possibly ignoring the second output) to
avoid computing the eigen decomposition twice. This is because the
eigenvectors are used to compute the gradient w.r.t. the eigenvalues. See
_SelfAdjointEigV2Grad in linalg_grad.py.
Args:
tensor: `Tensor` of shape `[..., N, N]`.
name: string, optional name of the operation.
Returns:
e: Eigenvalues. Shape is `[..., N]`. The vector `e[..., :]` contains the `N`
eigenvalues of `tensor[..., :, :]`."
8697,svd,tensorflow/tensorflow/python/ops/linalg_ops.py,487,function,"Computes the singular value decompositions of one or more matrices.
Computes the SVD of each inner matrix in `tensor` such that
`tensor[..., :, :] = u[..., :, :] * diag(s[..., :, :]) *
transpose(conj(v[..., :, :]))`
```python
# a is a tensor.
# s is a tensor of singular values.
# u is a tensor of left singular vectors.
# v is a tensor of right singular vectors.
s, u, v = svd(a)
s = svd(a, compute_uv=False)
```
Args:
tensor: `Tensor` of shape `[..., M, N]`. Let `P` be the minimum of `M` and
`N`.
full_matrices: If true, compute full-sized `u` and `v`. If false
(the default), compute only the leading `P` singular vectors.
Ignored if `compute_uv` is `False`.
compute_uv: If `True` then left and right singular vectors will be
computed and returned in `u` and `v`, respectively. Otherwise, only the
singular values will be computed, which can be significantly faster.
name: string, optional name of the operation.
Returns:
s: Singular values. Shape is `[..., P]`. The values are sorted in reverse
order of magnitude, so s[..., 0] is the largest value, s[..., 1] is the
second largest, etc.
u: Left singular vectors. If `full_matrices` is `False` (default) then
shape is `[..., M, P]`; if `full_matrices` is `True` then shape is
`[..., M, M]`. Not returned if `compute_uv` is `False`.
v: Right singular vectors. If `full_matrices` is `False` (default) then
shape is `[..., N, P]`. If `full_matrices` is `True` then shape is
`[..., N, N]`. Not returned if `compute_uv` is `False`.
@compatibility(numpy)
Mostly equivalent to numpy.linalg.svd, except that
* The order of output arguments here is `s`, `u`, `v` when `compute_uv` is
`True`, as opposed to `u`, `s`, `v` for numpy.linalg.svd.
* full_matrices is `False` by default as opposed to `True` for
numpy.linalg.svd.
* tf.linalg.svd uses the standard definition of the SVD
\\(A = U \Sigma V^H\\), such that the left singular vectors of `a` are
the columns of `u`, while the right singular vectors of `a` are the
columns of `v`. On the other hand, numpy.linalg.svd returns the adjoint
\\(V^H\\) as the third output argument.
```python
import tensorflow as tf
import numpy as np
s, u, v = tf.linalg.svd(a)
tf_a_approx = tf.matmul(u, tf.matmul(tf.linalg.diag(s), v, adjoint_b=True))
u, s, v_adj = np.linalg.svd(a, full_matrices=False)
np_a_approx = np.dot(u, np.dot(np.diag(s), v_adj))
# tf_a_approx and np_a_approx should be numerically close.
```
@end_compatibility"
8698,norm_v2,tensorflow/tensorflow/python/ops/linalg_ops.py,558,function,"Computes the norm of vectors, matrices, and tensors.
This function can compute several different vector norms (the 1-norm, the
Euclidean or 2-norm, the inf-norm, and in general the p-norm for p > 0) and
matrix norms (Frobenius, 1-norm, 2-norm and inf-norm).
Args:
tensor: `Tensor` of types `float32`, `float64`, `complex64`, `complex128`
ord: Order of the norm. Supported values are `'fro'`, `'euclidean'`,
`1`, `2`, `np.inf` and any positive real number yielding the corresponding
p-norm. Default is `'euclidean'` which is equivalent to Frobenius norm if
`tensor` is a matrix and equivalent to 2-norm for vectors.
Some restrictions apply:
a) The Frobenius norm `'fro'` is not defined for vectors,
b) If axis is a 2-tuple (matrix norm), only `'euclidean'`, '`fro'`, `1`,
`2`, `np.inf` are supported.
See the description of `axis` on how to compute norms for a batch of
vectors or matrices stored in a tensor.
axis: If `axis` is `None` (the default), the input is considered a vector
and a single vector norm is computed over the entire set of values in the
tensor, i.e. `norm(tensor, ord=ord)` is equivalent to
`norm(reshape(tensor, [-1]), ord=ord)`.
If `axis` is a Python integer, the input is considered a batch of vectors,
and `axis` determines the axis in `tensor` over which to compute vector
norms.
If `axis` is a 2-tuple of Python integers it is considered a batch of
matrices and `axis` determines the axes in `tensor` over which to compute
a matrix norm.
Negative indices are supported. Example: If you are passing a tensor that
can be either a matrix or a batch of matrices at runtime, pass
`axis=[-2,-1]` instead of `axis=None` to make sure that matrix norms are
computed.
keepdims: If True, the axis indicated in `axis` are kept with size 1.
Otherwise, the dimensions in `axis` are removed from the output shape.
name: The name of the op.
Returns:
output: A `Tensor` of the same type as tensor, containing the vector or
matrix norms. If `keepdims` is True then the rank of output is equal to
the rank of `tensor`. Otherwise, if `axis` is none the output is a scalar,
if `axis` is an integer, the rank of `output` is one less than the rank
of `tensor`, if `axis` is a 2-tuple the rank of `output` is two less
than the rank of `tensor`.
Raises:
ValueError: If `ord` or `axis` is invalid.
@compatibility(numpy)
Mostly equivalent to numpy.linalg.norm.
Not supported: ord <= 0, 2-norm for matrices, nuclear norm.
Other differences:
a) If axis is `None`, treats the flattened `tensor` as a vector
regardless of rank.
b) Explicitly supports 'euclidean' norm as the default, including for
higher order tensors.
@end_compatibility"
8699,norm,tensorflow/tensorflow/python/ops/linalg_ops.py,632,function,"Computes the norm of vectors, matrices, and tensors.
This function can compute several different vector norms (the 1-norm, the
Euclidean or 2-norm, the inf-norm, and in general the p-norm for p > 0) and
matrix norms (Frobenius, 1-norm, 2-norm and inf-norm).
Args:
tensor: `Tensor` of types `float32`, `float64`, `complex64`, `complex128`
ord: Order of the norm. Supported values are 'fro', 'euclidean',
`1`, `2`, `np.inf` and any positive real number yielding the corresponding
p-norm. Default is 'euclidean' which is equivalent to Frobenius norm if
`tensor` is a matrix and equivalent to 2-norm for vectors.
Some restrictions apply:
a) The Frobenius norm `fro` is not defined for vectors,
b) If axis is a 2-tuple (matrix norm), only 'euclidean', 'fro', `1`,
`2`, `np.inf` are supported.
See the description of `axis` on how to compute norms for a batch of
vectors or matrices stored in a tensor.
axis: If `axis` is `None` (the default), the input is considered a vector
and a single vector norm is computed over the entire set of values in the
tensor, i.e. `norm(tensor, ord=ord)` is equivalent to
`norm(reshape(tensor, [-1]), ord=ord)`.
If `axis` is a Python integer, the input is considered a batch of vectors,
and `axis` determines the axis in `tensor` over which to compute vector
norms.
If `axis` is a 2-tuple of Python integers it is considered a batch of
matrices and `axis` determines the axes in `tensor` over which to compute
a matrix norm.
Negative indices are supported. Example: If you are passing a tensor that
can be either a matrix or a batch of matrices at runtime, pass
`axis=[-2,-1]` instead of `axis=None` to make sure that matrix norms are
computed.
keepdims: If True, the axis indicated in `axis` are kept with size 1.
Otherwise, the dimensions in `axis` are removed from the output shape.
name: The name of the op.
keep_dims: Deprecated alias for `keepdims`.
Returns:
output: A `Tensor` of the same type as tensor, containing the vector or
matrix norms. If `keepdims` is True then the rank of output is equal to
the rank of `tensor`. Otherwise, if `axis` is none the output is a scalar,
if `axis` is an integer, the rank of `output` is one less than the rank
of `tensor`, if `axis` is a 2-tuple the rank of `output` is two less
than the rank of `tensor`.
Raises:
ValueError: If `ord` or `axis` is invalid.
@compatibility(numpy)
Mostly equivalent to numpy.linalg.norm.
Not supported: ord <= 0, 2-norm for matrices, nuclear norm.
Other differences:
a) If axis is `None`, treats the flattened `tensor` as a vector
regardless of rank.
b) Explicitly supports 'euclidean' norm as the default, including for
higher order tensors.
@end_compatibility"
8700,eye,tensorflow/tensorflow/python/ops/linalg_ops_impl.py,33,function,"Construct an identity matrix, or a batch of matrices.
See `linalg_ops.eye`."
8701,empty_tensor_list,tensorflow/tensorflow/python/ops/list_ops.py,45,function,
8702,tensor_list_reserve,tensorflow/tensorflow/python/ops/list_ops.py,59,function,
8703,tensor_list_from_tensor,tensorflow/tensorflow/python/ops/list_ops.py,67,function,
8704,tensor_list_get_item,tensorflow/tensorflow/python/ops/list_ops.py,74,function,
8705,tensor_list_pop_back,tensorflow/tensorflow/python/ops/list_ops.py,84,function,
8706,tensor_list_gather,tensorflow/tensorflow/python/ops/list_ops.py,92,function,
8707,tensor_list_scatter,tensorflow/tensorflow/python/ops/list_ops.py,105,function,
8708,tensor_list_stack,tensorflow/tensorflow/python/ops/list_ops.py,122,function,
8709,tensor_list_concat,tensorflow/tensorflow/python/ops/list_ops.py,135,function,
8710,tensor_list_split,tensorflow/tensorflow/python/ops/list_ops.py,147,function,
8711,tensor_list_set_item,tensorflow/tensorflow/python/ops/list_ops.py,155,function,Sets `item` at `index` in input list.
8712,Print,tensorflow/tensorflow/python/ops/logging_ops.py,74,function,"Prints a list of tensors.
This is an identity op (behaves like `tf.identity`) with the side effect
of printing `data` when evaluating.
Note: This op prints to the standard error. It is not currently compatible
with jupyter notebook (printing to the notebook *server's* output, not into
the notebook).
Args:
input_: A tensor passed through this op.
data: A list of tensors to print out when op is evaluated.
message: A string, prefix of the error message.
first_n: Only log `first_n` number of times. Negative numbers log always;
this is the default.
summarize: Only print this many entries of each tensor. If None, then a
maximum of 3 elements are printed per input tensor.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type and contents as `input_`.
```python
sess = tf.compat.v1.Session()
with sess.as_default():
tensor = tf.range(10)
print_op = tf.print(tensor)
with tf.control_dependencies([print_op]):
out = tf.add(tensor, tensor)
sess.run(out)
```"
8713,print_v2,tensorflow/tensorflow/python/ops/logging_ops.py,135,function,"Print the specified inputs.
A TensorFlow operator that prints the specified inputs to a desired
output stream or logging level. The inputs may be dense or sparse Tensors,
primitive python objects, data structures that contain tensors, and printable
Python objects. Printed tensors will recursively show the first and last
elements of each dimension to summarize.
Example:
Single-input usage:
```python
tensor = tf.range(10)
tf.print(tensor, output_stream=sys.stderr)
```
(This prints ""[0 1 2 ... 7 8 9]"" to sys.stderr)
Multi-input usage:
```python
tensor = tf.range(10)
tf.print(""tensors:"", tensor, {2: tensor * 2}, output_stream=sys.stdout)
```
(This prints ""tensors: [0 1 2 ... 7 8 9] {2: [0 2 4 ... 14 16 18]}"" to
sys.stdout)
Changing the input separator:
```python
tensor_a = tf.range(2)
tensor_b = tensor_a * 2
tf.print(tensor_a, tensor_b, output_stream=sys.stderr, sep=',')
```
(This prints ""[0 1],[0 2]"" to sys.stderr)
Usage in a `tf.function`:
```python
@tf.function
def f():
tensor = tf.range(10)
tf.print(tensor, output_stream=sys.stderr)
return tensor
range_tensor = f()
```
(This prints ""[0 1 2 ... 7 8 9]"" to sys.stderr)
@compatibility(TF 1.x Graphs and Sessions)
In graphs manually created outside of `tf.function`, this method returns
the created TF operator that prints the data. To make sure the
operator runs, users need to pass the produced op to
`tf.compat.v1.Session`'s run method, or to use the op as a control
dependency for executed ops by specifying
`with tf.compat.v1.control_dependencies([print_op])`.
@end_compatibility
Compatibility usage in TF 1.x graphs:
```python
sess = tf.compat.v1.Session()
with sess.as_default():
tensor = tf.range(10)
print_op = tf.print(""tensors:"", tensor, {2: tensor * 2},
output_stream=sys.stdout)
with tf.control_dependencies([print_op]):
tripled_tensor = tensor * 3
sess.run(tripled_tensor)
```
(This prints ""tensors: [0 1 2 ... 7 8 9] {2: [0 2 4 ... 14 16 18]}"" to
sys.stdout)
Note: In Jupyter notebooks and colabs, `tf.print` prints to the notebook
cell outputs. It will not write to the notebook kernel's console logs.
Args:
*inputs: Positional arguments that are the inputs to print. Inputs in the
printed output will be separated by spaces. Inputs may be python
primitives, tensors, data structures such as dicts and lists that may
contain tensors (with the data structures possibly nested in arbitrary
ways), and printable python objects.
output_stream: The output stream, logging level, or file to print to.
Defaults to sys.stderr, but sys.stdout, tf.compat.v1.logging.info,
tf.compat.v1.logging.warning, tf.compat.v1.logging.error,
absl.logging.info, absl.logging.warning and absl.logging.error are also
supported. To print to a file, pass a string started with ""file://""
followed by the file path, e.g., ""file:///tmp/foo.out"".
summarize: The first and last `summarize` elements within each dimension are
recursively printed per Tensor. If None, then the first 3 and last 3
elements of each dimension are printed for each tensor. If set to -1, it
will print all elements of every tensor.
sep: The string to use to separate the inputs. Defaults to "" "".
end: End character that is appended at the end the printed string.
Defaults to the newline character.
name: A name for the operation (optional).
Returns:
None when executing eagerly. During graph tracing this returns
a TF operator that prints the specified inputs in the specified output
stream or logging level. This operator will be automatically executed
except inside of `tf.compat.v1` graphs and sessions.
Raises:
ValueError: If an unsupported output stream is specified."
8714,histogram_summary,tensorflow/tensorflow/python/ops/logging_ops.py,389,function,"Outputs a `Summary` protocol buffer with a histogram.
This ops is deprecated. Please switch to tf.summary.histogram.
For an explanation of why this op was deprecated, and information on how to
migrate, look
['here'](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/contrib/deprecated/__init__.py)
The generated
[`Summary`](https://www.tensorflow.org/code/tensorflow/core/framework/summary.proto)
has one summary value containing a histogram for `values`.
This op reports an `InvalidArgument` error if any value is not finite.
Args:
tag: A `string` `Tensor`. 0-D. Tag to use for the summary value.
values: A real numeric `Tensor`. Any shape. Values to use to build the
histogram.
collections: Optional list of graph collections keys. The new summary op is
added to these collections. Defaults to `[GraphKeys.SUMMARIES]`.
name: A name for the operation (optional).
Returns:
A scalar `Tensor` of type `string`. The serialized `Summary` protocol
buffer."
8715,image_summary,tensorflow/tensorflow/python/ops/logging_ops.py,429,function,"Outputs a `Summary` protocol buffer with images.
For an explanation of why this op was deprecated, and information on how to
migrate, look
['here'](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/contrib/deprecated/__init__.py)
The summary has up to `max_images` summary values containing images. The
images are built from `tensor` which must be 4-D with shape `[batch_size,
height, width, channels]` and where `channels` can be:
* 1: `tensor` is interpreted as Grayscale.
* 3: `tensor` is interpreted as RGB.
* 4: `tensor` is interpreted as RGBA.
The images have the same number of channels as the input tensor. For float
input, the values are normalized one image at a time to fit in the range
`[0, 255]`. `uint8` values are unchanged. The op uses two different
normalization algorithms:
* If the input values are all positive, they are rescaled so the largest one
is 255.
* If any input value is negative, the values are shifted so input value 0.0
is at 127. They are then rescaled so that either the smallest value is 0,
or the largest one is 255.
The `tag` argument is a scalar `Tensor` of type `string`. It is used to
build the `tag` of the summary values:
* If `max_images` is 1, the summary value tag is '*tag*/image'.
* If `max_images` is greater than 1, the summary value tags are
generated sequentially as '*tag*/image/0', '*tag*/image/1', etc.
Args:
tag: A scalar `Tensor` of type `string`. Used to build the `tag` of the
summary values.
tensor: A 4-D `uint8` or `float32` `Tensor` of shape `[batch_size, height,
width, channels]` where `channels` is 1, 3, or 4.
max_images: Max number of batch elements to generate images for.
collections: Optional list of ops.GraphKeys. The collections to add the
summary to. Defaults to [ops.GraphKeys.SUMMARIES]
name: A name for the operation (optional).
Returns:
A scalar `Tensor` of type `string`. The serialized `Summary` protocol
buffer."
8716,audio_summary,tensorflow/tensorflow/python/ops/logging_ops.py,490,function,"Outputs a `Summary` protocol buffer with audio.
This op is deprecated. Please switch to tf.summary.audio.
For an explanation of why this op was deprecated, and information on how to
migrate, look
['here'](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/contrib/deprecated/__init__.py)
The summary has up to `max_outputs` summary values containing audio. The
audio is built from `tensor` which must be 3-D with shape `[batch_size,
frames, channels]` or 2-D with shape `[batch_size, frames]`. The values are
assumed to be in the range of `[-1.0, 1.0]` with a sample rate of
`sample_rate`.
The `tag` argument is a scalar `Tensor` of type `string`. It is used to
build the `tag` of the summary values:
* If `max_outputs` is 1, the summary value tag is '*tag*/audio'.
* If `max_outputs` is greater than 1, the summary value tags are
generated sequentially as '*tag*/audio/0', '*tag*/audio/1', etc.
Args:
tag: A scalar `Tensor` of type `string`. Used to build the `tag` of the
summary values.
tensor: A 3-D `float32` `Tensor` of shape `[batch_size, frames, channels]`
or a 2-D `float32` `Tensor` of shape `[batch_size, frames]`.
sample_rate: A Scalar `float32` `Tensor` indicating the sample rate of the
signal in hertz.
max_outputs: Max number of batch elements to generate audio for.
collections: Optional list of ops.GraphKeys. The collections to add the
summary to. Defaults to [ops.GraphKeys.SUMMARIES]
name: A name for the operation (optional).
Returns:
A scalar `Tensor` of type `string`. The serialized `Summary` protocol
buffer."
8717,merge_summary,tensorflow/tensorflow/python/ops/logging_ops.py,547,function,"Merges summaries.
This op is deprecated. Please switch to tf.compat.v1.summary.merge, which has
identical
behavior.
This op creates a
[`Summary`](https://www.tensorflow.org/code/tensorflow/core/framework/summary.proto)
protocol buffer that contains the union of all the values in the input
summaries.
When the Op is run, it reports an `InvalidArgument` error if multiple values
in the summaries to merge use the same tag.
Args:
inputs: A list of `string` `Tensor` objects containing serialized `Summary`
protocol buffers.
collections: Optional list of graph collections keys. The new summary op is
added to these collections. Defaults to `[GraphKeys.SUMMARIES]`.
name: A name for the operation (optional).
Returns:
A scalar `Tensor` of type `string`. The serialized `Summary` protocol
buffer resulting from the merging."
8718,merge_all_summaries,tensorflow/tensorflow/python/ops/logging_ops.py,581,function,"Merges all summaries collected in the default graph.
This op is deprecated. Please switch to tf.compat.v1.summary.merge_all, which
has
identical behavior.
Args:
key: `GraphKey` used to collect the summaries. Defaults to
`GraphKeys.SUMMARIES`.
Returns:
If no summaries were collected, returns None. Otherwise returns a scalar
`Tensor` of type `string` containing the serialized `Summary` protocol
buffer resulting from the merging."
8719,get_summary_op,tensorflow/tensorflow/python/ops/logging_ops.py,604,function,"Returns a single Summary op that would run all summaries.
Either existing one from `SUMMARY_OP` collection or merges all existing
summaries.
Returns:
If no summaries were collected, returns None. Otherwise returns a scalar
`Tensor` of type `string` containing the serialized `Summary` protocol
buffer resulting from the merging."
8720,scalar_summary,tensorflow/tensorflow/python/ops/logging_ops.py,635,function,"Outputs a `Summary` protocol buffer with scalar values.
This ops is deprecated. Please switch to tf.summary.scalar.
For an explanation of why this op was deprecated, and information on how to
migrate, look
['here'](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/contrib/deprecated/__init__.py)
The input `tags` and `values` must have the same shape. The generated
summary has a summary value for each tag-value pair in `tags` and `values`.
Args:
tags: A `string` `Tensor`. Tags for the summaries.
values: A real numeric Tensor. Values for the summaries.
collections: Optional list of graph collections keys. The new summary op is
added to these collections. Defaults to `[GraphKeys.SUMMARIES]`.
name: A name for the operation (optional).
Returns:
A scalar `Tensor` of type `string`. The serialized `Summary` protocol
buffer."
8721,initialize_all_tables,tensorflow/tensorflow/python/ops/lookup_ops.py,54,function,"Returns an Op that initializes all tables of the default graph.
Args:
name: Optional name for the initialization op.
Returns:
An Op that initializes all tables. Note that if there are
not tables the returned Op is a NoOp."
8722,tables_initializer,tensorflow/tensorflow/python/ops/lookup_ops.py,68,function,"Returns an Op that initializes all tables of the default graph.
See the [Low Level
Intro](https://www.tensorflow.org/guide/low_level_intro#feature_columns)
guide, for an example of usage.
Args:
name: Optional name for the initialization op.
Returns:
An Op that initializes all tables. Note that if there are
not tables the returned Op is a NoOp."
8723,LookupInterface,tensorflow/tensorflow/python/ops/lookup_ops.py,108,class,Represent a lookup table that persists across different steps.
8724,key_dtype,tensorflow/tensorflow/python/ops/lookup_ops.py,126,method,The table key dtype.
8725,value_dtype,tensorflow/tensorflow/python/ops/lookup_ops.py,131,method,The table value dtype.
8726,name,tensorflow/tensorflow/python/ops/lookup_ops.py,136,method,The name of the table.
8727,size,tensorflow/tensorflow/python/ops/lookup_ops.py,140,method,Compute the number of elements in this table.
8728,lookup,tensorflow/tensorflow/python/ops/lookup_ops.py,144,method,"Looks up `keys` in a table, outputs the corresponding values."
8729,InitializableLookupTableBase,tensorflow/tensorflow/python/ops/lookup_ops.py,149,class,"Initializable lookup table interface.
An initializable lookup tables persist across different steps."
8730,default_value,tensorflow/tensorflow/python/ops/lookup_ops.py,186,method,The default value of the table.
8731,size,tensorflow/tensorflow/python/ops/lookup_ops.py,190,method,"Compute the number of elements in this table.
Args:
name: A name for the operation (optional).
Returns:
A scalar tensor containing the number of elements in this table."
8732,lookup,tensorflow/tensorflow/python/ops/lookup_ops.py,202,method,"Looks up `keys` in a table, outputs the corresponding values.
The `default_value` is used for keys not present in the table.
Args:
keys: Keys to look up. May be either a `SparseTensor` or dense `Tensor`.
name: A name for the operation (optional).
Returns:
A `SparseTensor` if keys are sparse, otherwise a dense `Tensor`.
Raises:
TypeError: when `keys` or `default_value` doesn't match the table data
types."
8733,InitializableLookupTableBaseV1,tensorflow/tensorflow/python/ops/lookup_ops.py,240,class,
8734,initializer,tensorflow/tensorflow/python/ops/lookup_ops.py,243,method,
8735,StaticHashTable,tensorflow/tensorflow/python/ops/lookup_ops.py,248,class,"A generic hash table that is immutable once initialized.
Example usage:
```python
keys_tensor = tf.constant([1, 2])
vals_tensor = tf.constant([3, 4])
input_tensor = tf.constant([1, 5])
table = tf.lookup.StaticHashTable(
tf.lookup.KeyValueTensorInitializer(keys_tensor, vals_tensor), -1)
print(table.lookup(input_tensor))
```"
8736,name,tensorflow/tensorflow/python/ops/lookup_ops.py,307,method,
8737,export,tensorflow/tensorflow/python/ops/lookup_ops.py,310,method,"Returns tensors of all keys and values in the table.
Args:
name: A name for the operation (optional).
Returns:
A pair of tensors with the first tensor containing all keys and the
second tensors containing all values in the table."
8738,StaticHashTableV1,tensorflow/tensorflow/python/ops/lookup_ops.py,330,class,"A generic hash table that is immutable once initialized.
When running in graph mode, you must evaluate the tensor returned by
`tf.tables_initializer()` before evaluating the tensor returned by
this class's `lookup()` method. Example usage in graph mode:
```python
keys_tensor = tf.constant([1, 2])
vals_tensor = tf.constant([3, 4])
input_tensor = tf.constant([1, 5])
table = tf.lookup.StaticHashTable(
tf.lookup.KeyValueTensorInitializer(keys_tensor, vals_tensor), -1)
out = table.lookup(input_tensor)
with tf.Session() as sess:
sess.run(tf.tables_initializer())
print(sess.run(out))
```
In eager mode, no special code is needed to initialize the table.
Example usage in eager mode:
```python
tf.enable_eager_execution()
keys_tensor = tf.constant([1, 2])
vals_tensor = tf.constant([3, 4])
input_tensor = tf.constant([1, 5])
table = tf.lookup.StaticHashTable(
tf.lookup.KeyValueTensorInitializer(keys_tensor, vals_tensor), -1)
print(table.lookup(input_tensor))
```"
8739,initializer,tensorflow/tensorflow/python/ops/lookup_ops.py,364,method,
8740,HashTable,tensorflow/tensorflow/python/ops/lookup_ops.py,369,class,
8741,init,tensorflow/tensorflow/python/ops/lookup_ops.py,372,method,
8742,TableInitializerBase,tensorflow/tensorflow/python/ops/lookup_ops.py,376,class,Base class for lookup table initializers.
8743,key_dtype,tensorflow/tensorflow/python/ops/lookup_ops.py,390,method,The expected table key dtype.
8744,value_dtype,tensorflow/tensorflow/python/ops/lookup_ops.py,395,method,The expected table value dtype.
8745,initialize,tensorflow/tensorflow/python/ops/lookup_ops.py,399,method,Returns the table initialization op.
8746,DatasetInitializer,tensorflow/tensorflow/python/ops/lookup_ops.py,416,class,"Creates a table initializer from a `tf.data.Dataset`.
Sample usage:
```python
keys = tf.data.Dataset.range(100)
values = tf.data.Dataset.range(100).map(
lambda x: string_ops.as_string(x * 2))
ds = tf.data.Dataset.zip((keys, values))
init = tf.lookup.experimental.DatasetInitializer(ds)
table = tf.lookup.StaticHashTable(init, """")
output = table.lookup([0, 1, 2])
assertEquals(outputs, [""0"", ""2"", ""4""])
```
Attributes:
dataset: A `tf.data.Dataset` object that produces tuples of scalars. The
first scalar is treated as a key and the second as value.
Raises: ValueError if `dataset` doesn't conform to specifications."
8747,initialize,tensorflow/tensorflow/python/ops/lookup_ops.py,467,method,
8748,KeyValueTensorInitializer,tensorflow/tensorflow/python/ops/lookup_ops.py,476,class,Table initializers given `keys` and `values` tensors.
8749,initialize,tensorflow/tensorflow/python/ops/lookup_ops.py,509,method,"Initializes the given `table` with `keys` and `values` tensors.
Args:
table: The table to initialize.
Returns:
The operation that initializes the table.
Raises:
TypeError: when the keys and values data types do not match the table
key and value data types."
8750,TextFileIndex,tensorflow/tensorflow/python/ops/lookup_ops.py,532,class,"The key and value content to get from each line.
This class defines the key and value used for tf.lookup.TextFileInitializer.
The key and value content to get from each line is specified either
by the following, or a value `>=0`.
* `TextFileIndex.LINE_NUMBER` means use the line number starting from zero,
expects data type int64.
* `TextFileIndex.WHOLE_LINE` means use the whole line content, expects data
type string.
A value `>=0` means use the index (starting at zero) of the split line based
on `delimiter`."
8751,TextFileInitializer,tensorflow/tensorflow/python/ops/lookup_ops.py,552,class,"Table initializers from a text file.
This initializer assigns one entry in the table for each line in the file.
The key and value type of the table to initialize is given by `key_dtype` and
`value_dtype`.
The key and value content to get from each line is specified by
the `key_index` and `value_index`.
* `TextFileIndex.LINE_NUMBER` means use the line number starting from zero,
expects data type int64.
* `TextFileIndex.WHOLE_LINE` means use the whole line content, expects data
type string.
* A value `>=0` means use the index (starting at zero) of the split line based
on `delimiter`.
For example if we have a file with the following content:
```
emerson 10
lake 20
palmer 30
```
The following snippet initializes a table with the first column as keys and
second column as values:
* `emerson -> 10`
* `lake -> 20`
* `palmer -> 30`
```python
table = tf.lookup.StaticHashTable(tf.lookup.TextFileInitializer(
""test.txt"", tf.string, 0, tf.int64, 1, delimiter="" ""), -1)
...
table.init.run()
```
Similarly to initialize the whole line as keys and the line number as values.
* `emerson 10 -> 0`
* `lake 20 -> 1`
* `palmer 30 -> 2`
```python
table = tf.lookup.StaticHashTable(tf.lookup.TextFileInitializer(
""test.txt"", tf.string, tf.lookup.TextFileIndex.WHOLE_LINE,
tf.int64, tf.lookup.TextFileIndex.LINE_NUMBER, delimiter="" ""), -1)
...
table.init.run()
```"
8752,initialize,tensorflow/tensorflow/python/ops/lookup_ops.py,689,method,"Initializes the table from a text file.
Args:
table: The table to be initialized.
Returns:
The operation that initializes the table.
Raises:
TypeError: when the keys and values data types do not match the table
key and value data types."
8753,TextFileStringTableInitializer,tensorflow/tensorflow/python/ops/lookup_ops.py,733,class,Table initializer for `int64` IDs to string tables from a text file.
8754,TextFileIdTableInitializer,tensorflow/tensorflow/python/ops/lookup_ops.py,784,class,Table initializer for string to `int64` IDs tables from a text file.
8755,HasherSpec,tensorflow/tensorflow/python/ops/lookup_ops.py,837,class,"A structure for the spec of the hashing function to use for hash buckets.
`hasher` is the name of the hashing function to use (eg. ""fasthash"",
""stronghash"").
`key` is optional and specify the key to use for the hash function if
supported, currently only used by a strong hash.
Fields:
hasher: The hasher name to use.
key: The key to be used by the hashing function, if required."
8756,StrongHashSpec,tensorflow/tensorflow/python/ops/lookup_ops.py,855,class,"A structure to specify a key of the strong keyed hash spec.
The strong hash requires a `key`, which is a list of 2 unsigned integer
numbers. These should be non-zero; random numbers generated from random.org
would be a fine choice.
Fields:
key: The key to be used by the keyed hashing function."
8757,IdTableWithHashBuckets,tensorflow/tensorflow/python/ops/lookup_ops.py,884,class,"String to Id table wrapper that assigns out-of-vocabulary keys to buckets.
For example, if an instance of `IdTableWithHashBuckets` is initialized with a
string-to-id table that maps:
* `emerson -> 0`
* `lake -> 1`
* `palmer -> 2`
The `IdTableWithHashBuckets` object will performs the following mapping:
* `emerson -> 0`
* `lake -> 1`
* `palmer -> 2`
* `<other term> -> bucket_id`, where bucket_id will be between `3` and
`3 + num_oov_buckets - 1`, calculated by:
`hash(<term>) % num_oov_buckets + vocab_size`
If input_tensor is `[""emerson"", ""lake"", ""palmer"", ""king"", ""crimson""]`,
the lookup result is `[0, 1, 2, 4, 7]`.
If `table` is None, only out-of-vocabulary buckets are used.
Example usage:
```python
num_oov_buckets = 3
input_tensor = tf.constant([""emerson"", ""lake"", ""palmer"", ""king"", ""crimnson""])
table = tf.IdTableWithHashBuckets(
tf.StaticHashTable(
tf.lookup.TextFileInitializer(
filename,
key_dtype=tf.string,
key_index=tf.lookup.TextFileIndex.WHOLE_LINE,
value_dtype=tf.int64,
value_index=tf.lookup.TextFileIndex.LINE_NUMBER,
delimiter=""\t""),
default_value),
num_oov_buckets)
out = table.lookup(input_tensor).
table.init.run()
print(out.eval())
```
The hash function used for generating out-of-vocabulary buckets ID is handled
by `hasher_spec`."
8758,initializer,tensorflow/tensorflow/python/ops/lookup_ops.py,1008,method,
8759,init,tensorflow/tensorflow/python/ops/lookup_ops.py,1016,method,
8760,resource_handle,tensorflow/tensorflow/python/ops/lookup_ops.py,1020,method,
8761,name,tensorflow/tensorflow/python/ops/lookup_ops.py,1026,method,
8762,size,tensorflow/tensorflow/python/ops/lookup_ops.py,1029,method,Compute the number of elements in this table.
8763,lookup,tensorflow/tensorflow/python/ops/lookup_ops.py,1051,method,"Looks up `keys` in the table, outputs the corresponding values.
It assigns out-of-vocabulary keys to buckets based in their hashes.
Args:
keys: Keys to look up. May be either a `SparseTensor` or dense `Tensor`.
name: Optional name for the op.
Returns:
A `SparseTensor` if keys are sparse, otherwise a dense `Tensor`.
Raises:
TypeError: when `keys` doesn't match the table key data type."
8764,StaticVocabularyTable,tensorflow/tensorflow/python/ops/lookup_ops.py,1099,class,"String to Id table wrapper that assigns out-of-vocabulary keys to buckets.
For example, if an instance of `StaticVocabularyTable` is initialized with a
string-to-id initializer that maps:
* `emerson -> 0`
* `lake -> 1`
* `palmer -> 2`
The `Vocabulary` object will performs the following mapping:
* `emerson -> 0`
* `lake -> 1`
* `palmer -> 2`
* `<other term> -> bucket_id`, where bucket_id will be between `3` and
`3 + num_oov_buckets - 1`, calculated by:
`hash(<term>) % num_oov_buckets + vocab_size`
If input_tensor is `[""emerson"", ""lake"", ""palmer"", ""king"", ""crimson""]`,
the lookup result is `[0, 1, 2, 4, 7]`.
If `initializer` is None, only out-of-vocabulary buckets are used.
Example usage:
```python
num_oov_buckets = 3
input_tensor = tf.constant([""emerson"", ""lake"", ""palmer"", ""king"", ""crimnson""])
table = tf.lookup.StaticVocabularyTable(
tf.lookup.TextFileInitializer(
filename,
key_dtype=tf.string, key_index=tf.lookup.TextFileIndex.WHOLE_LINE,
value_dtype=tf.int64, value_index=tf.lookup.TextFileIndex.LINE_NUMBER,
delimiter=""\t""),
num_oov_buckets)
out = table.lookup(input_tensor).
table.init.run()
print(out.eval())
```
The hash function used for generating out-of-vocabulary buckets ID is
Fingerprint64."
8765,resource_handle,tensorflow/tensorflow/python/ops/lookup_ops.py,1219,method,
8766,name,tensorflow/tensorflow/python/ops/lookup_ops.py,1225,method,
8767,size,tensorflow/tensorflow/python/ops/lookup_ops.py,1228,method,Compute the number of elements in this table.
8768,lookup,tensorflow/tensorflow/python/ops/lookup_ops.py,1237,method,"Looks up `keys` in the table, outputs the corresponding values.
It assigns out-of-vocabulary keys to buckets based in their hashes.
Args:
keys: Keys to look up. May be either a `SparseTensor` or dense `Tensor`.
name: Optional name for the op.
Returns:
A `SparseTensor` if keys are sparse, otherwise a dense `Tensor`.
Raises:
TypeError: when `keys` doesn't match the table key data type."
8769,StaticVocabularyTableV1,tensorflow/tensorflow/python/ops/lookup_ops.py,1280,class,
8770,initializer,tensorflow/tensorflow/python/ops/lookup_ops.py,1283,method,
8771,index_table_from_file,tensorflow/tensorflow/python/ops/lookup_ops.py,1290,function,"Returns a lookup table that converts a string tensor into int64 IDs.
This operation constructs a lookup table to convert tensor of strings into
int64 IDs. The mapping can be initialized from a vocabulary file specified in
`vocabulary_file`, where the whole line is the key and the zero-based line
number is the ID.
Any lookup of an out-of-vocabulary token will return a bucket ID based on its
hash if `num_oov_buckets` is greater than zero. Otherwise it is assigned the
`default_value`.
The bucket ID range is
`[vocabulary size, vocabulary size + num_oov_buckets - 1]`.
The underlying table must be initialized by calling
`session.run(tf.compat.v1.tables_initializer())` or
`session.run(table.init())` once.
To specify multi-column vocabulary files, use key_column_index and
value_column_index and delimiter.
- TextFileIndex.LINE_NUMBER means use the line number starting from zero,
expects data type int64.
- TextFileIndex.WHOLE_LINE means use the whole line content, expects data
type string.
- A value >=0 means use the index (starting at zero) of the split line based
on `delimiter`.
Sample Usages:
If we have a vocabulary file ""test.txt"" with the following content:
```
emerson
lake
palmer
```
```python
features = tf.constant([""emerson"", ""lake"", ""and"", ""palmer""])
table = tf.lookup.index_table_from_file(
vocabulary_file=""test.txt"", num_oov_buckets=1)
ids = table.lookup(features)
...
tf.compat.v1.tables_initializer().run()
ids.eval() ==> [0, 1, 3, 2] # where 3 is the out-of-vocabulary bucket
```
Args:
vocabulary_file: The vocabulary filename, may be a constant scalar `Tensor`.
num_oov_buckets: The number of out-of-vocabulary buckets.
vocab_size: Number of the elements in the vocabulary, if known.
default_value: The value to use for out-of-vocabulary feature values.
Defaults to -1.
hasher_spec: A `HasherSpec` to specify the hash function to use for
assignation of out-of-vocabulary buckets.
key_dtype: The `key` data type.
name: A name for this op (optional).
key_column_index: The column index from the text file to get the `key`
values from. The default is to use the whole line content.
value_column_index: The column index from the text file to get the `value`
values from. The default is to use the line number, starting from zero.
delimiter: The delimiter to separate fields in a line.
Returns:
The lookup table to map a `key_dtype` `Tensor` to index `int64` `Tensor`.
Raises:
ValueError: If `vocabulary_file` is not set.
ValueError: If `num_oov_buckets` is negative or `vocab_size` is not greater
than zero."
8772,index_table_from_tensor,tensorflow/tensorflow/python/ops/lookup_ops.py,1411,function,"Returns a lookup table that converts a string tensor into int64 IDs.
This operation constructs a lookup table to convert tensor of strings into
int64 IDs. The mapping can be initialized from a string `vocabulary_list` 1-D
tensor where each element is a key and corresponding index within the tensor
is the value.
Any lookup of an out-of-vocabulary token will return a bucket ID based on its
hash if `num_oov_buckets` is greater than zero. Otherwise it is assigned the
`default_value`. The bucket ID range is
`[vocabulary list size, vocabulary list size + num_oov_buckets - 1]`.
The underlying table must be initialized by calling
`session.run(tf.compat.v1.tables_initializer())` or
`session.run(table.init())` once.
Elements in `vocabulary_list` cannot have duplicates, otherwise when executing
the table initializer op, it will throw a `FailedPreconditionError`.
Sample Usages:
```python
vocabulary_list = tf.constant([""emerson"", ""lake"", ""palmer""])
table = tf.lookup.index_table_from_tensor(
vocabulary_list=vocabulary_list, num_oov_buckets=1, default_value=-1)
features = tf.constant([""emerson"", ""lake"", ""and"", ""palmer""])
ids = table.lookup(features)
...
tf.compat.v1.tables_initializer().run()
ids.eval() ==> [0, 1, 4, 2]
```
Args:
vocabulary_list: A 1-D `Tensor` that specifies the mapping of keys to
indices. The type of this object must be castable to `dtype`.
num_oov_buckets: The number of out-of-vocabulary buckets.
default_value: The value to use for out-of-vocabulary feature values.
Defaults to -1.
hasher_spec: A `HasherSpec` to specify the hash function to use for
assignment of out-of-vocabulary buckets.
dtype: The type of values passed to `lookup`. Only string and integers are
supported.
name: A name for this op (optional).
Returns:
The lookup table to map an input `Tensor` to index `int64` `Tensor`.
Raises:
ValueError: If `vocabulary_list` is invalid.
ValueError: If `num_oov_buckets` is negative."
8773,index_to_string_table_from_file,tensorflow/tensorflow/python/ops/lookup_ops.py,1510,function,"Returns a lookup table that maps a `Tensor` of indices into strings.
This operation constructs a lookup table to map int64 indices into string
values. The table is initialized from a vocabulary file specified in
`vocabulary_file`, where the whole line is the value and the
zero-based line number is the index.
Any input which does not have a corresponding index in the vocabulary file
(an out-of-vocabulary entry) is assigned the `default_value`
The underlying table must be initialized by calling
`session.run(tf.compat.v1.tables_initializer())` or
`session.run(table.init())` once.
To specify multi-column vocabulary files, use key_column_index and
value_column_index and delimiter.
- TextFileIndex.LINE_NUMBER means use the line number starting from zero,
expects data type int64.
- TextFileIndex.WHOLE_LINE means use the whole line content, expects data
type string.
- A value >=0 means use the index (starting at zero) of the split line based
on `delimiter`.
Sample Usages:
If we have a vocabulary file ""test.txt"" with the following content:
```
emerson
lake
palmer
```
```python
indices = tf.constant([1, 5], tf.int64)
table = tf.lookup.index_to_string_table_from_file(
vocabulary_file=""test.txt"", default_value=""UNKNOWN"")
values = table.lookup(indices)
...
tf.compat.v1.tables_initializer().run()
values.eval() ==> [""lake"", ""UNKNOWN""]
```
Args:
vocabulary_file: The vocabulary filename, may be a constant scalar `Tensor`.
vocab_size: Number of the elements in the vocabulary, if known.
default_value: The value to use for out-of-vocabulary indices.
name: A name for this op (optional).
key_column_index: The column index from the text file to get the `key`
values from. The default is to use the line number, starting from zero.
value_column_index: The column index from the text file to get the `value`
values from. The default is to use the whole line content.
delimiter: The delimiter to separate fields in a line.
Returns:
The lookup table to map a string values associated to a given index `int64`
`Tensors`.
Raises:
ValueError: when `vocabulary_file` is empty.
ValueError: when `vocab_size` is invalid."
8774,index_to_string_table_from_tensor,tensorflow/tensorflow/python/ops/lookup_ops.py,1601,function,"Returns a lookup table that maps a `Tensor` of indices into strings.
This operation constructs a lookup table to map int64 indices into string
values. The mapping is initialized from a string `vocabulary_list` 1-D
`Tensor` where each element is a value and the corresponding index within the
tensor is the key.
Any input which does not have a corresponding index in 'vocabulary_list'
(an out-of-vocabulary entry) is assigned the `default_value`
The underlying table must be initialized by calling
`session.run(tf.compat.v1.tables_initializer())` or
`session.run(table.init())` once.
Elements in `vocabulary_list` cannot have duplicates, otherwise when executing
the table initializer op, it will throw a `FailedPreconditionError`.
Sample Usages:
```python
vocabulary_list = tf.constant([""emerson"", ""lake"", ""palmer""])
indices = tf.constant([1, 5], tf.int64)
table = tf.lookup.index_to_string_table_from_tensor(
vocabulary_list, default_value=""UNKNOWN"")
values = table.lookup(indices)
...
tf.compat.v1.tables_initializer().run()
values.eval() ==> [""lake"", ""UNKNOWN""]
```
Args:
vocabulary_list: A 1-D string `Tensor` that specifies the strings to map
from indices.
default_value: The value to use for out-of-vocabulary indices.
name: A name for this op (optional).
Returns:
The lookup table to map a string values associated to a given index `int64`
`Tensors`.
Raises:
ValueError: when `vocabulary_list` is not set."
8775,MutableHashTable,tensorflow/tensorflow/python/ops/lookup_ops.py,1663,class,"A generic mutable hash table implementation.
Data can be inserted by calling the insert method and removed by calling the
remove method. It does not support initialization via the init method.
Example usage:
```python
table = tf.lookup.MutableHashTable(key_dtype=tf.string, value_dtype=tf.int64,
default_value=-1)
sess.run(table.insert(keys, values))
out = table.lookup(query_keys)
print(out.eval())
```"
8776,name,tensorflow/tensorflow/python/ops/lookup_ops.py,1758,method,
8777,size,tensorflow/tensorflow/python/ops/lookup_ops.py,1761,method,"Compute the number of elements in this table.
Args:
name: A name for the operation (optional).
Returns:
A scalar tensor containing the number of elements in this table."
8778,remove,tensorflow/tensorflow/python/ops/lookup_ops.py,1774,method,"Removes `keys` and its associated values from the table.
If a key is not present in the table, it is silently ignored.
Args:
keys: Keys to remove. Can be a tensor of any shape. Must match the table's
key type.
name: A name for the operation (optional).
Returns:
The created Operation.
Raises:
TypeError: when `keys` do not match the table data types."
8779,lookup,tensorflow/tensorflow/python/ops/lookup_ops.py,1800,method,"Looks up `keys` in a table, outputs the corresponding values.
The `default_value` is used for keys not present in the table.
Args:
keys: Keys to look up. Can be a tensor of any shape. Must match the
table's key_dtype.
name: A name for the operation (optional).
Returns:
A tensor containing the values in the same shape as `keys` using the
table's value type.
Raises:
TypeError: when `keys` do not match the table data types."
8780,insert,tensorflow/tensorflow/python/ops/lookup_ops.py,1825,method,"Associates `keys` with `values`.
Args:
keys: Keys to insert. Can be a tensor of any shape. Must match the table's
key type.
values: Values to be associated with keys. Must be a tensor of the same
shape as `keys` and match the table's value type.
name: A name for the operation (optional).
Returns:
The created Operation.
Raises:
TypeError: when `keys` or `values` doesn't match the table data
types."
8781,export,tensorflow/tensorflow/python/ops/lookup_ops.py,1852,method,"Returns tensors of all keys and values in the table.
Args:
name: A name for the operation (optional).
Returns:
A pair of tensors with the first tensor containing all keys and the
second tensors containing all values in the table."
8782,restore,tensorflow/tensorflow/python/ops/lookup_ops.py,1891,method,
8783,DenseHashTable,tensorflow/tensorflow/python/ops/lookup_ops.py,1902,class,"A generic mutable hash table implementation using tensors as backing store.
Data can be inserted by calling the insert method and removed by calling the
remove method. It does not support initialization via the init method.
It uses ""open addressing"" with quadratic reprobing to resolve collisions.
Compared to `MutableHashTable` the insert, remove and lookup operations in a
`DenseHashTable` are typically faster, but memory usage can be higher.
However, `DenseHashTable` does not require additional memory for
temporary tensors created during checkpointing and restore operations.
Example usage:
```python
table = tf.lookup.DenseHashTable(key_dtype=tf.int64,
value_dtype=tf.int64,
default_value=-1,
empty_key=0,
deleted_key=-1)
sess.run(table.insert(keys, values))
out = table.lookup(query_keys)
print(out.eval())
```"
8784,name,tensorflow/tensorflow/python/ops/lookup_ops.py,2016,method,
8785,size,tensorflow/tensorflow/python/ops/lookup_ops.py,2019,method,"Compute the number of elements in this table.
Args:
name: A name for the operation (optional).
Returns:
A scalar tensor containing the number of elements in this table."
8786,lookup,tensorflow/tensorflow/python/ops/lookup_ops.py,2032,method,"Looks up `keys` in a table, outputs the corresponding values.
The `default_value` is used for keys not present in the table.
Args:
keys: Keys to look up. Can be a tensor of any shape. Must match the
table's key_dtype.
name: A name for the operation (optional).
Returns:
A tensor containing the values in the same shape as `keys` using the
table's value type.
Raises:
TypeError: when `keys` do not match the table data types."
8787,insert_or_assign,tensorflow/tensorflow/python/ops/lookup_ops.py,2058,method,"Associates `keys` with `values`.
Args:
keys: Keys to insert. Can be a tensor of any shape. Must match the table's
key type.
values: Values to be associated with keys. Must be a tensor of the same
shape as `keys` and match the table's value type.
name: A name for the operation (optional).
Returns:
The created Operation.
Raises:
TypeError: when `keys` or `values` doesn't match the table data
types."
8788,insert,tensorflow/tensorflow/python/ops/lookup_ops.py,2085,method,"Associates `keys` with `values`.
Args:
keys: Keys to insert. Can be a tensor of any shape. Must match the table's
key type.
values: Values to be associated with keys. Must be a tensor of the same
shape as `keys` and match the table's value type.
name: A name for the operation (optional).
Returns:
The created Operation.
Raises:
TypeError: when `keys` or `values` doesn't match the table data
types."
8789,erase,tensorflow/tensorflow/python/ops/lookup_ops.py,2104,method,"Removes `keys` and its associated values from the table.
If a key is not present in the table, it is silently ignored.
Args:
keys: Keys to remove. Can be a tensor of any shape. Must match the table's
key type.
name: A name for the operation (optional).
Returns:
The created Operation.
Raises:
TypeError: when `keys` do not match the table data types."
8790,remove,tensorflow/tensorflow/python/ops/lookup_ops.py,2131,method,"Removes `keys` and its associated values from the table.
If a key is not present in the table, it is silently ignored.
Args:
keys: Keys to remove. Can be a tensor of any shape. Must match the table's
key type.
name: A name for the operation (optional).
Returns:
The created Operation.
Raises:
TypeError: when `keys` do not match the table data types."
8791,export,tensorflow/tensorflow/python/ops/lookup_ops.py,2149,method,"Returns tensors of all keys and values in the table.
Args:
name: A name for the operation (optional).
Returns:
A pair of tensors with the first tensor containing all keys and the
second tensors containing all values in the table."
8792,restore,tensorflow/tensorflow/python/ops/lookup_ops.py,2189,method,
8793,roll,tensorflow/tensorflow/python/ops/manip_ops.py,31,function,
8794,map_fn,tensorflow/tensorflow/python/ops/map_fn.py,47,function,"Transforms `elems` by applying `fn` to each element unstacked on axis 0.
See also `tf.scan`.
`map_fn` unstacks `elems` on axis 0 to obtain a sequence of elements;
calls `fn` to transform each element; and then stacks the transformed
values back together.
#### Mapping functions with single-Tensor inputs and outputs
If `elems` is a single tensor and `fn`'s signature is `tf.Tensor->tf.Tensor`,
then `map_fn(fn, elems)` is equivalent to
`tf.stack([fn(elem) for elem in tf.unstack(elems)])`. E.g.:
>>> tf.map_fn(fn=lambda t: tf.range(t, t + 3), elems=tf.constant([3, 5, 2]))
<tf.Tensor: shape=(3, 3), dtype=int32, numpy=
array([[3, 4, 5],
[5, 6, 7],
[2, 3, 4]], dtype=int32)>
`map_fn(fn, elems).shape = [elems.shape[0]] + fn(elems[0]).shape`.
#### Mapping functions with multi-arity inputs and outputs
`map_fn` also supports functions with multi-arity inputs and outputs:
* If `elems` is a tuple (or nested structure) of tensors, then those tensors
must all have the same outer-dimension size (`num_elems`); and `fn` is
used to transform each tuple (or structure) of corresponding slices from
`elems`. E.g., if `elems` is a tuple `(t1, t2, t3)`, then `fn` is used to
transform each tuple of slices `(t1[i], t2[i], t3[i])`
(where `0 <= i < num_elems`).
* If `fn` returns a tuple (or nested structure) of tensors, then the
result is formed by stacking corresponding elements from those structures.
#### Specifying `fn`'s output signature
If `fn`'s input and output signatures are different, then the output
signature must be specified using `fn_output_signature`. (The input and
output signatures are differ if their structures, dtypes, or tensor types do
not match). E.g.:
>>> tf.map_fn(fn=tf.strings.length, # input & output have different dtypes
... elems=tf.constant([""hello"", ""moon""]),
... fn_output_signature=tf.int32)
<tf.Tensor: shape=(2,), dtype=int32, numpy=array([5, 4], dtype=int32)>
>>> tf.map_fn(fn=tf.strings.join, # input & output have different structures
... elems=[tf.constant(['The', 'A']), tf.constant(['Dog', 'Cat'])],
... fn_output_signature=tf.string)
<tf.Tensor: shape=(2,), dtype=string,
numpy=array([b'TheDog', b'ACat'], dtype=object)>
`fn_output_signature` can be specified using any of the following:
* A `tf.DType` or `tf.TensorSpec` (to describe a `tf.Tensor`)
* A `tf.RaggedTensorSpec` (to describe a `tf.RaggedTensor`)
* A `tf.SparseTensorSpec` (to describe a `tf.sparse.SparseTensor`)
* A (possibly nested) tuple, list, or dict containing the above types.
#### RaggedTensors
`map_fn` supports `tf.RaggedTensor` inputs and outputs. In particular:
* If `elems` is a `RaggedTensor`, then `fn` will be called with each
row of that ragged tensor.
* If `elems` has only one ragged dimension, then the values passed to
`fn` will be `tf.Tensor`s.
* If `elems` has multiple ragged dimensions, then the values passed to
`fn` will be `tf.RaggedTensor`s with one fewer ragged dimension.
* If the result of `map_fn` should be a `RaggedTensor`, then use a
`tf.RaggedTensorSpec` to specify `fn_output_signature`.
* If `fn` returns `tf.Tensor`s with varying sizes, then use a
`tf.RaggedTensorSpec` with `ragged_rank=0` to combine them into a
single ragged tensor (which will have ragged_rank=1).
* If `fn` returns `tf.RaggedTensor`s, then use a `tf.RaggedTensorSpec`
with the same `ragged_rank`.
>>> # Example: RaggedTensor input
>>> rt = tf.ragged.constant([[1, 2, 3], [], [4, 5], [6]])
>>> tf.map_fn(tf.reduce_sum, rt, fn_output_signature=tf.int32)
<tf.Tensor: shape=(4,), dtype=int32, numpy=array([6, 0, 9, 6], dtype=int32)>
>>> # Example: RaggedTensor output
>>> elems = tf.constant([3, 5, 0, 2])
>>> tf.map_fn(tf.range, elems,
... fn_output_signature=tf.RaggedTensorSpec(shape=[None],
... dtype=tf.int32))
<tf.RaggedTensor [[0, 1, 2], [0, 1, 2, 3, 4], [], [0, 1]]>
Note: `map_fn` should only be used if you need to map a function over the
*rows* of a `RaggedTensor`. If you wish to map a function over the
individual values, then you should use:
* `tf.ragged.map_flat_values(fn, rt)`
(if fn is expressible as TensorFlow ops)
* `rt.with_flat_values(map_fn(fn, rt.flat_values))`
(otherwise)
E.g.:
>>> rt = tf.ragged.constant([[1, 2, 3], [], [4, 5], [6]])
>>> tf.ragged.map_flat_values(lambda x: x + 2, rt)
<tf.RaggedTensor [[3, 4, 5], [], [6, 7], [8]]>
#### SparseTensors
`map_fn` supports `tf.sparse.SparseTensor` inputs and outputs. In particular:
* If `elems` is a `SparseTensor`, then `fn` will be called with each row
of that sparse tensor. In particular, the value passed to `fn` will be a
`tf.sparse.SparseTensor` with one fewer dimension than `elems`.
* If the result of `map_fn` should be a `SparseTensor`, then use a
`tf.SparseTensorSpec` to specify `fn_output_signature`. The individual
`SparseTensor`s returned by `fn` will be stacked into a single
`SparseTensor` with one more dimension.
>>> # Example: SparseTensor input
>>> st = tf.sparse.SparseTensor([[0, 0], [2, 0], [2, 1]], [2, 3, 4], [4, 4])
>>> tf.map_fn(tf.sparse.reduce_sum, st, fn_output_signature=tf.int32)
<tf.Tensor: shape=(4,), dtype=int32, numpy=array([2, 0, 7, 0], dtype=int32)>
>>> # Example: SparseTensor output
>>> tf.sparse.to_dense(
... tf.map_fn(tf.sparse.eye, tf.constant([2, 3]),
... fn_output_signature=tf.SparseTensorSpec(None, tf.float32)))
<tf.Tensor: shape=(2, 3, 3), dtype=float32, numpy=
array([[[1., 0., 0.],
[0., 1., 0.],
[0., 0., 0.]],
[[1., 0., 0.],
[0., 1., 0.],
[0., 0., 1.]]], dtype=float32)>
Note: `map_fn` should only be used if you need to map a function over the
*rows* of a `SparseTensor`. If you wish to map a function over the nonzero
values, then you should use:
* If the function is expressible as TensorFlow ops, use:
```python
tf.sparse.SparseTensor(st.indices, fn(st.values), st.dense_shape)
```
* Otherwise, use:
```python
tf.sparse.SparseTensor(st.indices, tf.map_fn(fn, st.values),
st.dense_shape)
```
#### `map_fn` vs. vectorized operations
`map_fn` will apply the operations used by `fn` to each element of `elems`,
resulting in `O(elems.shape[0])` total operations. This is somewhat
mitigated by the fact that `map_fn` can process elements in parallel.
However, a transform expressed using `map_fn` is still typically less
efficient than an equivalent transform expressed using vectorized operations.
`map_fn` should typically only be used if one of the following is true:
* It is difficult or expensive to express the desired transform with
vectorized operations.
* `fn` creates large intermediate values, so an equivalent vectorized
transform would take too much memory.
* Processing elements in parallel is more efficient than an equivalent
vectorized transform.
* Efficiency of the transform is not critical, and using `map_fn` is
more readable.
E.g., the example given above that maps `fn=lambda t: tf.range(t, t + 3)`
across `elems` could be rewritten more efficiently using vectorized ops:
>>> elems = tf.constant([3, 5, 2])
>>> tf.range(3) + tf.expand_dims(elems, 1)
<tf.Tensor: shape=(3, 3), dtype=int32, numpy=
array([[3, 4, 5],
[5, 6, 7],
[2, 3, 4]], dtype=int32)>
In some cases, `tf.vectorized_map` can be used to automatically convert a
function to a vectorized eqivalent.
#### Eager execution
When executing eagerly, `map_fn` does not execute in parallel even if
`parallel_iterations` is set to a value > 1. You can still get the
performance benefits of running a function in parallel by using the
`tf.function` decorator:
>>> fn=lambda t: tf.range(t, t + 3)
>>> @tf.function
... def func(elems):
... return tf.map_fn(fn, elems, parallel_iterations=3)
>>> func(tf.constant([3, 5, 2]))
<tf.Tensor: shape=(3, 3), dtype=int32, numpy=
array([[3, 4, 5],
[5, 6, 7],
[2, 3, 4]], dtype=int32)>
Note: if you use the `tf.function` decorator, any non-TensorFlow Python
code that you may have written in your function won't get executed. See
`tf.function` for more details. The recommendation would be to debug without
`tf.function` but switch to it to get performance benefits of running `map_fn`
in parallel.
Args:
fn: The callable to be performed. It accepts one argument, which will have
the same (possibly nested) structure as `elems`. Its output must have the
same structure as `fn_output_signature` if one is provided; otherwise it
must have the same structure as `elems`.
elems: A tensor or (possibly nested) sequence of tensors, each of which will
be unstacked along their first dimension. `fn` will be applied to the
nested sequence of the resulting slices. `elems` may include ragged and
sparse tensors. `elems` must consist of at least one tensor.
dtype: Deprecated: Equivalent to `fn_output_signature`.
parallel_iterations: (optional) The number of iterations allowed to run in
parallel. When graph building, the default value is 10. While executing
eagerly, the default value is set to 1.
back_prop: (optional) False disables support for back propagation.
swap_memory: (optional) True enables GPU-CPU memory swapping.
infer_shape: (optional) False disables tests for consistent output shapes.
name: (optional) Name prefix for the returned tensors.
fn_output_signature: The output signature of `fn`. Must be specified if
`fn`'s input and output signatures are different (i.e., if their
structures, dtypes, or tensor types do not match).
`fn_output_signature` can be specified using any of the following:
* A `tf.DType` or `tf.TensorSpec` (to describe a `tf.Tensor`)
* A `tf.RaggedTensorSpec` (to describe a `tf.RaggedTensor`)
* A `tf.SparseTensorSpec` (to describe a `tf.sparse.SparseTensor`)
* A (possibly nested) tuple, list, or dict containing the above types.
Returns:
A tensor or (possibly nested) sequence of tensors. Each tensor stacks the
results of applying `fn` to tensors unstacked from `elems` along the first
dimension, from first to last. The result may include ragged and sparse
tensors.
Raises:
TypeError: if `fn` is not callable or the structure of the output of
`fn` and `fn_output_signature` do not match.
ValueError: if the lengths of the output of `fn` and `fn_output_signature`
do not match, or if the `elems` does not contain any tensor.
Examples:
>>> elems = np.array([1, 2, 3, 4, 5, 6])
>>> tf.map_fn(lambda x: x * x, elems)
<tf.Tensor: shape=(6,), dtype=int64, numpy=array([ 1, 4, 9, 16, 25, 36])>
>>> elems = (np.array([1, 2, 3]), np.array([-1, 1, -1]))
>>> tf.map_fn(lambda x: x[0] * x[1], elems, fn_output_signature=tf.int64)
<tf.Tensor: shape=(3,), dtype=int64, numpy=array([-1, 2, -3])>
>>> elems = np.array([1, 2, 3])
>>> tf.map_fn(lambda x: (x, -x), elems,
... fn_output_signature=(tf.int64, tf.int64))
(<tf.Tensor: shape=(3,), dtype=int64, numpy=array([1, 2, 3])>,
<tf.Tensor: shape=(3,), dtype=int64, numpy=array([-1, -2, -3])>)"
8795,map_fn_v2,tensorflow/tensorflow/python/ops/map_fn.py,628,function,Transform `elems` by applying `fn` to each element unstacked on axis 0.
8796,SmartBroadcastGradientArgs,tensorflow/tensorflow/python/ops/math_grad.py,67,function,"Optimized version of `broadcast_gradient_args` that caches results.
This implementation avoids creating `broadcast_gradient_args` ops in the case
that the input shapes are fully defined, and provides hints to the calling
code that can be used to avoid creating reduction and reshaping ops.
Args:
x: The left input tensor to a broadcasting binary op.
y: The right input tensor to a broadcasting binary op.
grad: The incoming gradient tensor for a broadcasting binary op.
Returns:
A pair of tuples, containing:
* A 3-tuple of broadcast information for x, containing:
* The shape of x (as a tuple or Tensor).
* The reduction indices for x (as a tuple or Tensor).
* A boolean, which if True, indicates that x's shape differs from grad's
shape (and so x's gradient must be reduced and/or reshaped).
* A 3-tuple of broadcast information for y, containing the respective
details for y."
8797,linspace_nd,tensorflow/tensorflow/python/ops/math_ops.py,112,function,"Generates evenly-spaced values in an interval along a given axis.
A sequence of `num` evenly-spaced values are generated beginning at `start`
along a given `axis`.
If `num > 1`, the values in the sequence increase by `stop - start / num - 1`,
so that the last one is exactly `stop`. If `num <= 0`, `ValueError` is raised.
Matches
[np.linspace](https://docs.scipy.org/doc/numpy/reference/generated/numpy.linspace.html)'s
behaviour
except when `num == 0`.
For example:
```
tf.linspace(10.0, 12.0, 3, name=""linspace"") => [ 10.0 11.0 12.0]
```
`Start` and `stop` can be tensors of arbitrary size:
>>> tf.linspace([0., 5.], [10., 40.], 5, axis=0)
<tf.Tensor: shape=(5, 2), dtype=float32, numpy=
array([[ 0. , 5. ],
[ 2.5 , 13.75],
[ 5. , 22.5 ],
[ 7.5 , 31.25],
[10. , 40. ]], dtype=float32)>
`Axis` is where the values will be generated (the dimension in the
returned tensor which corresponds to the axis will be equal to `num`)
>>> tf.linspace([0., 5.], [10., 40.], 5, axis=-1)
<tf.Tensor: shape=(2, 5), dtype=float32, numpy=
array([[ 0. , 2.5 , 5. , 7.5 , 10. ],
[ 5. , 13.75, 22.5 , 31.25, 40. ]], dtype=float32)>
Args:
start: A `Tensor`. Must be one of the following types: `bfloat16`,
`float32`, `float64`. N-D tensor. First entry in the range.
stop: A `Tensor`. Must have the same type and shape as `start`. N-D tensor.
Last entry in the range.
num: A `Tensor`. Must be one of the following types: `int32`, `int64`. 0-D
tensor. Number of values to generate.
name: A name for the operation (optional).
axis: Axis along which the operation is performed (used only when N-D
tensors are provided).
Returns:
A `Tensor`. Has the same type as `start`."
8798,argmax,tensorflow/tensorflow/python/ops/math_ops.py,251,function,
8799,argmax_v2,tensorflow/tensorflow/python/ops/math_ops.py,263,function,"Returns the index with the largest value across axes of a tensor.
In case of identity returns the smallest index.
For example:
>>> A = tf.constant([2, 20, 30, 3, 6])
>>> tf.math.argmax(A) # A[2] is maximum in tensor A
<tf.Tensor: shape=(), dtype=int64, numpy=2>
>>> B = tf.constant([[2, 20, 30, 3, 6], [3, 11, 16, 1, 8],
... [14, 45, 23, 5, 27]])
>>> tf.math.argmax(B, 0)
<tf.Tensor: shape=(5,), dtype=int64, numpy=array([2, 2, 0, 2, 2])>
>>> tf.math.argmax(B, 1)
<tf.Tensor: shape=(3,), dtype=int64, numpy=array([2, 2, 1])>
>>> C = tf.constant([0, 0, 0, 0])
>>> tf.math.argmax(C) # Returns smallest index in case of ties
<tf.Tensor: shape=(), dtype=int64, numpy=0>
Args:
input: A `Tensor`.
axis: An integer, the axis to reduce across. Default to 0.
output_type: An optional output dtype (`tf.int32` or `tf.int64`). Defaults
to `tf.int64`.
name: An optional name for the operation.
Returns:
A `Tensor` of type `output_type`."
8800,argmin,tensorflow/tensorflow/python/ops/math_ops.py,305,function,
8801,argmin_v2,tensorflow/tensorflow/python/ops/math_ops.py,317,function,"Returns the index with the smallest value across axes of a tensor.
Returns the smallest index in case of ties.
Args:
input: A `Tensor`. Must be one of the following types: `float32`, `float64`,
`int32`, `uint8`, `int16`, `int8`, `complex64`, `int64`, `qint8`,
`quint8`, `qint32`, `bfloat16`, `uint16`, `complex128`, `half`, `uint32`,
`uint64`.
axis: A `Tensor`. Must be one of the following types: `int32`, `int64`.
int32 or int64, must be in the range `-rank(input), rank(input))`.
Describes which axis of the input Tensor to reduce across. For vectors,
use axis = 0.
output_type: An optional `tf.DType` from: `tf.int32, tf.int64`. Defaults to
`tf.int64`.
name: A name for the operation (optional).
Returns:
A `Tensor` of type `output_type`.
Usage:
```python
import tensorflow as tf
a = [1, 10, 26.9, 2.8, 166.32, 62.3]
b = tf.math.argmin(input = a)
c = tf.keras.backend.eval(b)
# c = 0
# here a[0] = 1 which is the smallest element of a across axis 0
```"
8802,abs,tensorflow/tensorflow/python/ops/math_ops.py,360,function,"Computes the absolute value of a tensor.
Given a tensor of integer or floating-point values, this operation returns a
tensor of the same type, where each element contains the absolute value of the
corresponding element in the input.
Given a tensor `x` of complex numbers, this operation returns a tensor of type
`float32` or `float64` that is the absolute value of each element in `x`. For
a complex number \\(a + bj\\), its absolute value is computed as
\\(\sqrt{a^2 + b^2}\\). For example:
>>> x = tf.constant([[-2.25 + 4.75j], [-3.25 + 5.75j]])
>>> tf.abs(x)
<tf.Tensor: shape=(2, 1), dtype=float64, numpy=
array([[5.25594901],
[6.60492241]])>
Args:
x: A `Tensor` or `SparseTensor` of type `float16`, `float32`, `float64`,
`int32`, `int64`, `complex64` or `complex128`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor` of the same size, type and sparsity as `x`,
with absolute values. Note, for `complex64` or `complex128` input, the
returned `Tensor` will be of type `float32` or `float64`, respectively."
8803,DivideDelegateWithName,tensorflow/tensorflow/python/ops/math_ops.py,406,class,Use Python2/Python3 division delegation to implement divide for tensors.
8804,divide,tensorflow/tensorflow/python/ops/math_ops.py,431,function,"Computes Python style division of `x` by `y`.
For example:
>>> x = tf.constant([16, 12, 11])
>>> y = tf.constant([4, 6, 2])
>>> tf.divide(x,y)
<tf.Tensor: shape=(3,), dtype=float64,
numpy=array([4. , 2. , 5.5])>
Args:
x: A `Tensor`
y: A `Tensor`
name: A name for the operation (optional).
Returns:
A `Tensor` with same shape as input"
8805,multiply,tensorflow/tensorflow/python/ops/math_ops.py,465,function,"Returns an element-wise x * y.
For example:
>>> x = tf.constant(([1, 2, 3, 4]))
>>> tf.math.multiply(x, x)
<tf.Tensor: shape=(4,), dtype=..., numpy=array([ 1, 4, 9, 16], dtype=int32)>
Since `tf.math.multiply` will convert its arguments to `Tensor`s, you can also
pass in non-`Tensor` arguments:
>>> tf.math.multiply(7,6)
<tf.Tensor: shape=(), dtype=int32, numpy=42>
If `x.shape` is not thes same as `y.shape`, they will be broadcast to a
compatible shape. (More about broadcasting
[here](https://docs.scipy.org/doc/numpy/user/basics.broadcasting.html).)
For example:
>>> x = tf.ones([1, 2]);
>>> y = tf.ones([2, 1]);
>>> x * y # Taking advantage of operator overriding
<tf.Tensor: shape=(2, 2), dtype=float32, numpy=
array([[1., 1.],
[1., 1.]], dtype=float32)>
Args:
x: A Tensor. Must be one of the following types: `bfloat16`,
`half`, `float32`, `float64`, `uint8`, `int8`, `uint16`,
`int16`, `int32`, `int64`, `complex64`, `complex128`.
y: A `Tensor`. Must have the same type as `x`.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `x`.
Raises:
* InvalidArgumentError: When `x` and `y` have incomptatible shapes or types."
8806,subtract,tensorflow/tensorflow/python/ops/math_ops.py,526,function,
8807,scalar_mul,tensorflow/tensorflow/python/ops/math_ops.py,572,function,"Multiplies a scalar times a `Tensor` or `IndexedSlices` object.
Intended for use in gradient code which might deal with `IndexedSlices`
objects, which are easy to multiply by a scalar but more expensive to
multiply with arbitrary tensors.
Args:
scalar: A 0-D scalar `Tensor`. Must have known shape.
x: A `Tensor` or `IndexedSlices` to be scaled.
name: A name for the operation (optional).
Returns:
`scalar * x` of the same type (`Tensor` or `IndexedSlices`) as `x`.
Raises:
ValueError: if scalar is not a 0-D `scalar`."
8808,scalar_mul_v2,tensorflow/tensorflow/python/ops/math_ops.py,606,function,
8809,pow,tensorflow/tensorflow/python/ops/math_ops.py,613,function,"Computes the power of one value to another.
Given a tensor `x` and a tensor `y`, this operation computes \\(x^y\\) for
corresponding elements in `x` and `y`. For example:
```python
x = tf.constant([[2, 2], [3, 3]])
y = tf.constant([[8, 16], [2, 3]])
tf.pow(x, y) # [[256, 65536], [9, 27]]
```
Args:
x: A `Tensor` of type `float16`, `float32`, `float64`, `int32`, `int64`,
`complex64`, or `complex128`.
y: A `Tensor` of type `float16`, `float32`, `float64`, `int32`, `int64`,
`complex64`, or `complex128`.
name: A name for the operation (optional).
Returns:
A `Tensor`."
8810,complex,tensorflow/tensorflow/python/ops/math_ops.py,642,function,"Converts two real numbers to a complex number.
Given a tensor `real` representing the real part of a complex number, and a
tensor `imag` representing the imaginary part of a complex number, this
operation returns complex numbers elementwise of the form \\(a + bj\\), where
*a* represents the `real` part and *b* represents the `imag` part.
The input tensors `real` and `imag` must have the same shape.
For example:
```python
real = tf.constant([2.25, 3.25])
imag = tf.constant([4.75, 5.75])
tf.complex(real, imag) # [[2.25 + 4.75j], [3.25 + 5.75j]]
```
Args:
real: A `Tensor`. Must be one of the following types: `float32`, `float64`.
imag: A `Tensor`. Must have the same type as `real`.
name: A name for the operation (optional).
Returns:
A `Tensor` of type `complex64` or `complex128`.
Raises:
TypeError: Real and imag must be correct types"
8811,sign,tensorflow/tensorflow/python/ops/math_ops.py,687,function,"Returns an element-wise indication of the sign of a number.
y = sign(x) = -1 if x < 0; 0 if x == 0; 1 if x > 0.
For complex numbers, y = sign(x) = x / |x| if x != 0, otherwise y = 0.
Example usage:
>>> tf.math.sign([0., 2., -3.])
<tf.Tensor: ... numpy=array([ 0., 1., -1.], dtype=float32)>
Args:
x: A Tensor. Must be one of the following types: bfloat16, half, float32,
float64, int32, int64, complex64, complex128.
name: A name for the operation (optional).
Returns:
A Tensor. Has the same type as x.
If x is a SparseTensor, returns SparseTensor(x.indices,
tf.math.sign(x.values, ...), x.dense_shape)."
8812,real,tensorflow/tensorflow/python/ops/math_ops.py,728,function,"Returns the real part of a complex (or real) tensor.
Given a tensor `input`, this operation returns a tensor of type `float` that
is the real part of each element in `input` considered as a complex number.
For example:
```python
x = tf.constant([-2.25 + 4.75j, 3.25 + 5.75j])
tf.math.real(x) # [-2.25, 3.25]
```
If `input` is already real, it is returned unchanged.
Args:
input: A `Tensor`. Must have numeric type.
name: A name for the operation (optional).
Returns:
A `Tensor` of type `float32` or `float64`."
8813,imag,tensorflow/tensorflow/python/ops/math_ops.py,763,function,"Returns the imaginary part of a complex (or real) tensor.
Given a tensor `input`, this operation returns a tensor of type `float` that
is the imaginary part of each element in `input` considered as a complex
number. If `input` is real, a tensor of all zeros is returned.
For example:
```python
x = tf.constant([-2.25 + 4.75j, 3.25 + 5.75j])
tf.math.imag(x) # [4.75, 5.75]
```
Args:
input: A `Tensor`. Must be one of the following types: `float`, `double`,
`complex64`, `complex128`.
name: A name for the operation (optional).
Returns:
A `Tensor` of type `float32` or `float64`."
8814,angle,tensorflow/tensorflow/python/ops/math_ops.py,797,function,"Returns the element-wise argument of a complex (or real) tensor.
Given a tensor `input`, this operation returns a tensor of type `float` that
is the argument of each element in `input` considered as a complex number.
The elements in `input` are considered to be complex numbers of the form
\\(a + bj\\), where *a* is the real part and *b* is the imaginary part.
If `input` is real then *b* is zero by definition.
The argument returned by this function is of the form \\(atan2(b, a)\\).
If `input` is real, a tensor of all zeros is returned.
For example:
```
input = tf.constant([-2.25 + 4.75j, 3.25 + 5.75j], dtype=tf.complex64)
tf.math.angle(input).numpy()
# ==> array([2.0131705, 1.056345 ], dtype=float32)
```
Args:
input: A `Tensor`. Must be one of the following types: `float`, `double`,
`complex64`, `complex128`.
name: A name for the operation (optional).
Returns:
A `Tensor` of type `float32` or `float64`."
8815,round,tensorflow/tensorflow/python/ops/math_ops.py,840,function,"Rounds the values of a tensor to the nearest integer, element-wise.
Rounds half to even. Also known as bankers rounding. If you want to round
according to the current system rounding mode use tf::cint.
For example:
```python
x = tf.constant([0.9, 2.5, 2.3, 1.5, -4.5])
tf.round(x) # [ 1.0, 2.0, 2.0, 2.0, -4.0 ]
```
Args:
x: A `Tensor` of type `float16`, `float32`, `float64`, `int32`, or `int64`.
name: A name for the operation (optional).
Returns:
A `Tensor` of same shape and type as `x`."
8816,cast,tensorflow/tensorflow/python/ops/math_ops.py,868,function,"Casts a tensor to a new type.
The operation casts `x` (in case of `Tensor`) or `x.values`
(in case of `SparseTensor` or `IndexedSlices`) to `dtype`.
For example:
>>> x = tf.constant([1.8, 2.2], dtype=tf.float32)
>>> tf.dtypes.cast(x, tf.int32)
<tf.Tensor: shape=(2,), dtype=int32, numpy=array([1, 2], dtype=int32)>
The operation supports data types (for `x` and `dtype`) of
`uint8`, `uint16`, `uint32`, `uint64`, `int8`, `int16`, `int32`, `int64`,
`float16`, `float32`, `float64`, `complex64`, `complex128`, `bfloat16`.
In case of casting from complex types (`complex64`, `complex128`) to real
types, only the real part of `x` is returned. In case of casting from real
types to complex types (`complex64`, `complex128`), the imaginary part of the
returned value is set to `0`. The handling of complex types here matches the
behavior of numpy.
Note casting nan and inf values to integral types has undefined behavior.
Args:
x: A `Tensor` or `SparseTensor` or `IndexedSlices` of numeric type. It could
be `uint8`, `uint16`, `uint32`, `uint64`, `int8`, `int16`, `int32`,
`int64`, `float16`, `float32`, `float64`, `complex64`, `complex128`,
`bfloat16`.
dtype: The destination type. The list of supported dtypes is the same as
`x`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor` or `IndexedSlices` with same shape as `x` and
same type as `dtype`.
Raises:
TypeError: If `x` cannot be cast to the `dtype`."
8817,saturate_cast,tensorflow/tensorflow/python/ops/math_ops.py,933,function,"Performs a safe saturating cast of `value` to `dtype`.
This function casts the input to `dtype` without applying any scaling. If
there is a danger that values would over or underflow in the cast, this op
applies the appropriate clamping before the cast.
Args:
value: A `Tensor`.
dtype: The desired output `DType`.
name: A name for the operation (optional).
Returns:
`value` safely cast to `dtype`."
8818,to_float,tensorflow/tensorflow/python/ops/math_ops.py,967,function,"Casts a tensor to type `float32`.
Args:
x: A `Tensor` or `SparseTensor` or `IndexedSlices`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor` or `IndexedSlices` with same shape as `x` with
type `float32`.
Raises:
TypeError: If `x` cannot be cast to the `float32`."
8819,to_double,tensorflow/tensorflow/python/ops/math_ops.py,987,function,"Casts a tensor to type `float64`.
Args:
x: A `Tensor` or `SparseTensor` or `IndexedSlices`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor` or `IndexedSlices` with same shape as `x` with
type `float64`.
Raises:
TypeError: If `x` cannot be cast to the `float64`."
8820,to_int32,tensorflow/tensorflow/python/ops/math_ops.py,1007,function,"Casts a tensor to type `int32`.
Args:
x: A `Tensor` or `SparseTensor` or `IndexedSlices`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor` or `IndexedSlices` with same shape as `x` with
type `int32`.
Raises:
TypeError: If `x` cannot be cast to the `int32`."
8821,to_int64,tensorflow/tensorflow/python/ops/math_ops.py,1027,function,"Casts a tensor to type `int64`.
Args:
x: A `Tensor` or `SparseTensor` or `IndexedSlices`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor` or `IndexedSlices` with same shape as `x` with
type `int64`.
Raises:
TypeError: If `x` cannot be cast to the `int64`."
8822,to_bfloat16,tensorflow/tensorflow/python/ops/math_ops.py,1047,function,"Casts a tensor to type `bfloat16`.
Args:
x: A `Tensor` or `SparseTensor` or `IndexedSlices`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor` or `IndexedSlices` with same shape as `x` with
type `bfloat16`.
Raises:
TypeError: If `x` cannot be cast to the `bfloat16`."
8823,to_complex64,tensorflow/tensorflow/python/ops/math_ops.py,1067,function,"Casts a tensor to type `complex64`.
Args:
x: A `Tensor` or `SparseTensor` or `IndexedSlices`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor` or `IndexedSlices` with same shape as `x` with
type `complex64`.
Raises:
TypeError: If `x` cannot be cast to the `complex64`."
8824,to_complex128,tensorflow/tensorflow/python/ops/math_ops.py,1087,function,"Casts a tensor to type `complex128`.
Args:
x: A `Tensor` or `SparseTensor` or `IndexedSlices`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor` or `IndexedSlices` with same shape as `x` with
type `complex128`.
Raises:
TypeError: If `x` cannot be cast to the `complex128`."
8825,truediv,tensorflow/tensorflow/python/ops/math_ops.py,1267,function,"Divides x / y elementwise (using Python 3 division operator semantics).
NOTE: Prefer using the Tensor operator or tf.divide which obey Python
division operator semantics.
This function forces Python 3 division operator semantics where all integer
arguments are cast to floating types first. This op is generated by normal
`x / y` division in Python 3 and in Python 2.7 with
`from __future__ import division`. If you want integer division that rounds
down, use `x // y` or `tf.math.floordiv`.
`x` and `y` must have the same numeric type. If the inputs are floating
point, the output will have the same type. If the inputs are integral, the
inputs are cast to `float32` for `int8` and `int16` and `float64` for `int32`
and `int64` (matching the behavior of Numpy).
Args:
x: `Tensor` numerator of numeric type.
y: `Tensor` denominator of numeric type.
name: A name for the operation (optional).
Returns:
`x / y` evaluated in floating point.
Raises:
TypeError: If `x` and `y` have different dtypes."
8826,div,tensorflow/tensorflow/python/ops/math_ops.py,1303,function,"Divides x / y elementwise (using Python 2 division operator semantics).
NOTE: Prefer using the Tensor division operator or tf.divide which obey Python
3 division operator semantics.
This function divides `x` and `y`, forcing Python 2 semantics. That is, if `x`
and `y` are both integers then the result will be an integer. This is in
contrast to Python 3, where division with `/` is always a float while division
with `//` is always an integer.
Args:
x: `Tensor` numerator of real numeric type.
y: `Tensor` denominator of real numeric type.
name: A name for the operation (optional).
Returns:
`x / y` returns the quotient of x and y."
8827,div_no_nan,tensorflow/tensorflow/python/ops/math_ops.py,1329,function,"Computes a safe divide which returns 0 if the y is zero.
Args:
x: A `Tensor`. Must be one of the following types: `float32`, `float64`.
y: A `Tensor` whose dtype is compatible with `x`.
name: A name for the operation (optional).
Returns:
The element-wise value of the x divided by y."
8828,multiply_no_nan,tensorflow/tensorflow/python/ops/math_ops.py,1349,function,"Computes the product of x and y and returns 0 if the y is zero, even if x is NaN or infinite.
Args:
x: A `Tensor`. Must be one of the following types: `float32`, `float64`.
y: A `Tensor` whose dtype is compatible with `x`.
name: A name for the operation (optional).
Returns:
The element-wise value of the x times y."
8829,floordiv,tensorflow/tensorflow/python/ops/math_ops.py,1381,function,"Divides `x / y` elementwise, rounding toward the most negative integer.
The same as `tf.compat.v1.div(x,y)` for integers, but uses
`tf.floor(tf.compat.v1.div(x,y))` for
floating point arguments so that the result is always an integer (though
possibly an integer represented as floating point). This op is generated by
`x // y` floor division in Python 3 and in Python 2.7 with
`from __future__ import division`.
`x` and `y` must have the same type, and the result will have the same type
as well.
Args:
x: `Tensor` numerator of real numeric type.
y: `Tensor` denominator of real numeric type.
name: A name for the operation (optional).
Returns:
`x / y` rounded down.
Raises:
TypeError: If the inputs are complex."
8830,logical_xor,tensorflow/tensorflow/python/ops/math_ops.py,1481,function,"Logical XOR function.
x ^ y = (x | y) & ~(x & y)
The operation works for the following input types:
- Two single elements of type `bool`
- One `tf.Tensor` of type `bool` and one single `bool`, where the result will
be calculated by applying logical XOR with the single element to each
element in the larger Tensor.
- Two `tf.Tensor` objects of type `bool` of the same shape. In this case,
the result will be the element-wise logical XOR of the two input tensors.
Usage:
>>> a = tf.constant([True])
>>> b = tf.constant([False])
>>> tf.math.logical_xor(a, b)
<tf.Tensor: shape=(1,), dtype=bool, numpy=array([ True])>
>>> c = tf.constant([True])
>>> x = tf.constant([False, True, True, False])
>>> tf.math.logical_xor(c, x)
<tf.Tensor: shape=(4,), dtype=bool, numpy=array([ True, False, False, True])>
>>> y = tf.constant([False, False, True, True])
>>> z = tf.constant([False, True, False, True])
>>> tf.math.logical_xor(y, z)
<tf.Tensor: shape=(4,), dtype=bool, numpy=array([False, True, True, False])>
Args:
x: A `tf.Tensor` type bool.
y: A `tf.Tensor` of type bool.
name: A name for the operation (optional).
Returns:
A `tf.Tensor` of type bool with the same size as that of x or y."
8831,logical_and,tensorflow/tensorflow/python/ops/math_ops.py,1529,function,"Logical AND function.
The operation works for the following input types:
- Two single elements of type `bool`
- One `tf.Tensor` of type `bool` and one single `bool`, where the result will
be calculated by applying logical AND with the single element to each
element in the larger Tensor.
- Two `tf.Tensor` objects of type `bool` of the same shape. In this case,
the result will be the element-wise logical AND of the two input tensors.
Usage:
>>> a = tf.constant([True])
>>> b = tf.constant([False])
>>> tf.math.logical_and(a, b)
<tf.Tensor: shape=(1,), dtype=bool, numpy=array([False])>
>>> c = tf.constant([True])
>>> x = tf.constant([False, True, True, False])
>>> tf.math.logical_and(c, x)
<tf.Tensor: shape=(4,), dtype=bool, numpy=array([False, True, True, False])>
>>> y = tf.constant([False, False, True, True])
>>> z = tf.constant([False, True, False, True])
>>> tf.math.logical_and(y, z)
<tf.Tensor: shape=(4,), dtype=bool, numpy=array([False, False, False, True])>
Args:
x: A `tf.Tensor` type bool.
y: A `tf.Tensor` of type bool.
name: A name for the operation (optional).
Returns:
A `tf.Tensor` of type bool with the same size as that of x or y."
8832,and_,tensorflow/tensorflow/python/ops/math_ops.py,1569,function,
8833,or_,tensorflow/tensorflow/python/ops/math_ops.py,1575,function,
8834,xor_,tensorflow/tensorflow/python/ops/math_ops.py,1581,function,
8835,invert_,tensorflow/tensorflow/python/ops/math_ops.py,1587,function,
8836,equal,tensorflow/tensorflow/python/ops/math_ops.py,1607,function,"Returns the truth value of (x == y) element-wise.
Performs a [broadcast](
https://docs.scipy.org/doc/numpy/user/basics.broadcasting.html) with the
arguments and then an element-wise equality comparison, returning a Tensor of
boolean values.
For example:
>>> x = tf.constant([2, 4])
>>> y = tf.constant(2)
>>> tf.math.equal(x, y)
<tf.Tensor: shape=(2,), dtype=bool, numpy=array([ True, False])>
>>> x = tf.constant([2, 4])
>>> y = tf.constant([2, 4])
>>> tf.math.equal(x, y)
<tf.Tensor: shape=(2,), dtype=bool, numpy=array([ True, True])>
Args:
x: A `tf.Tensor` or `tf.sparse.SparseTensor` or `tf.IndexedSlices`.
y: A `tf.Tensor` or `tf.sparse.SparseTensor` or `tf.IndexedSlices`.
name: A name for the operation (optional).
Returns:
A `tf.Tensor` of type bool with the same size as that of x or y.
Raises:
`tf.errors.InvalidArgumentError`: If shapes of arguments are incompatible"
8837,not_equal,tensorflow/tensorflow/python/ops/math_ops.py,1643,function,"Returns the truth value of (x != y) element-wise.
Performs a [broadcast](
https://docs.scipy.org/doc/numpy/user/basics.broadcasting.html) with the
arguments and then an element-wise inequality comparison, returning a Tensor
of boolean values.
For example:
>>> x = tf.constant([2, 4])
>>> y = tf.constant(2)
>>> tf.math.not_equal(x, y)
<tf.Tensor: shape=(2,), dtype=bool, numpy=array([False, True])>
>>> x = tf.constant([2, 4])
>>> y = tf.constant([2, 4])
>>> tf.math.not_equal(x, y)
<tf.Tensor: shape=(2,), dtype=bool, numpy=array([False, False])>
Args:
x: A `tf.Tensor` or `tf.sparse.SparseTensor` or `tf.IndexedSlices`.
y: A `tf.Tensor` or `tf.sparse.SparseTensor` or `tf.IndexedSlices`.
name: A name for the operation (optional).
Returns:
A `tf.Tensor` of type bool with the same size as that of x or y.
Raises:
`tf.errors.InvalidArgumentError`: If shapes of arguments are incompatible"
8838,tensor_equals,tensorflow/tensorflow/python/ops/math_ops.py,1679,function,"The operation invoked by the `Tensor.__eq__` operator.
Compares two tensors element-wise for equality if they are
broadcast-compatible; or returns False if they are not broadcast-compatible.
(Note that this behavior differs from `tf.math.equal`, which raises an
exception if the two tensors are not broadcast-compatible.)
Purpose in the API:
This method is exposed in TensorFlow's API so that library developers
can register dispatching for `Tensor.__eq__` to allow it to handle
custom composite tensors & other custom objects.
The API symbol is not intended to be called by users directly and does
appear in TensorFlow's generated documentation.
Args:
self: The left-hand side of the `==` operator.
other: The right-hand side of the `==` operator.
Returns:
The result of the elementwise `==` operation, or `False` if the arguments
are not broadcast-compatible."
8839,tensor_not_equals,tensorflow/tensorflow/python/ops/math_ops.py,1717,function,"The operation invoked by the `Tensor.__ne__` operator.
Compares two tensors element-wise for inequality if they are
broadcast-compatible; or returns True if they are not broadcast-compatible.
(Note that this behavior differs from `tf.math.not_equal`, which raises an
exception if the two tensors are not broadcast-compatible.)
Purpose in the API:
This method is exposed in TensorFlow's API so that library developers
can register dispatching for `Tensor.__ne__` to allow it to handle
custom composite tensors & other custom objects.
The API symbol is not intended to be called by users directly and does
appear in TensorFlow's generated documentation.
Args:
self: The left-hand side of the `!=` operator.
other: The right-hand side of the `!=` operator.
Returns:
The result of the elementwise `!=` operation, or `True` if the arguments
are not broadcast-compatible."
8840,range,tensorflow/tensorflow/python/ops/math_ops.py,1757,function,"Creates a sequence of numbers.
Creates a sequence of numbers that begins at `start` and extends by
increments of `delta` up to but not including `limit`.
The dtype of the resulting tensor is inferred from the inputs unless
it is provided explicitly.
Like the Python builtin `range`, `start` defaults to 0, so that
`range(n) = range(0, n)`.
For example:
>>> start = 3
>>> limit = 18
>>> delta = 3
>>> tf.range(start, limit, delta)
<tf.Tensor: shape=(5,), dtype=int32,
numpy=array([ 3, 6, 9, 12, 15], dtype=int32)>
>>> start = 3
>>> limit = 1
>>> delta = -0.5
>>> tf.range(start, limit, delta)
<tf.Tensor: shape=(4,), dtype=float32,
numpy=array([3. , 2.5, 2. , 1.5], dtype=float32)>
>>> limit = 5
>>> tf.range(limit)
<tf.Tensor: shape=(5,), dtype=int32,
numpy=array([0, 1, 2, 3, 4], dtype=int32)>
Args:
start: A 0-D `Tensor` (scalar). Acts as first entry in the range if `limit`
is not None; otherwise, acts as range limit and first entry defaults to 0.
limit: A 0-D `Tensor` (scalar). Upper limit of sequence, exclusive. If None,
defaults to the value of `start` while the first entry of the range
defaults to 0.
delta: A 0-D `Tensor` (scalar). Number that increments `start`. Defaults to
1.
dtype: The type of the elements of the resulting tensor.
name: A name for the operation. Defaults to ""range"".
Returns:
An 1-D `Tensor` of type `dtype`.
@compatibility(numpy)
Equivalent to np.arange
@end_compatibility"
8841,reduce_sum_v1,tensorflow/tensorflow/python/ops/math_ops.py,1887,function,"Computes the sum of elements across dimensions of a tensor.
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
of the entries in `axis`, which must be unique. If `keepdims` is true, the
reduced dimensions are retained with length 1.
If `axis` is None, all dimensions are reduced, and a
tensor with a single element is returned.
For example:
```python
x = tf.constant([[1, 1, 1], [1, 1, 1]])
tf.reduce_sum(x) # 6
tf.reduce_sum(x, 0) # [2, 2, 2]
tf.reduce_sum(x, 1) # [3, 3]
tf.reduce_sum(x, 1, keepdims=True) # [[3], [3]]
tf.reduce_sum(x, [0, 1]) # 6
```
Args:
input_tensor: The tensor to reduce. Should have numeric type.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name for the operation (optional).
reduction_indices: The old (deprecated) name for axis.
keep_dims: Deprecated alias for `keepdims`.
Returns:
The reduced tensor, of the same dtype as the input_tensor.
@compatibility(numpy)
Equivalent to np.sum apart the fact that numpy upcast uint8 and int32 to
int64 while tensorflow returns the same dtype as the input.
@end_compatibility"
8842,reduce_sum,tensorflow/tensorflow/python/ops/math_ops.py,1942,function,"Computes the sum of elements across dimensions of a tensor.
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
of the entries in `axis`, which must be unique. If `keepdims` is true, the
reduced dimensions are retained with length 1.
If `axis` is None, all dimensions are reduced, and a
tensor with a single element is returned.
For example:
>>> # x has a shape of (2, 3) (two rows and three columns):
>>> x = tf.constant([[1, 1, 1], [1, 1, 1]])
>>> x.numpy()
array([[1, 1, 1],
[1, 1, 1]], dtype=int32)
>>> # sum all the elements
>>> # 1 + 1 + 1 + 1 + 1+ 1 = 6
>>> tf.reduce_sum(x).numpy()
6
>>> # reduce along the first dimension
>>> # the result is [1, 1, 1] + [1, 1, 1] = [2, 2, 2]
>>> tf.reduce_sum(x, 0).numpy()
array([2, 2, 2], dtype=int32)
>>> # reduce along the second dimension
>>> # the result is [1, 1] + [1, 1] + [1, 1] = [3, 3]
>>> tf.reduce_sum(x, 1).numpy()
array([3, 3], dtype=int32)
>>> # keep the original dimensions
>>> tf.reduce_sum(x, 1, keepdims=True).numpy()
array([[3],
[3]], dtype=int32)
>>> # reduce along both dimensions
>>> # the result is 1 + 1 + 1 + 1 + 1 + 1 = 6
>>> # or, equivalently, reduce along rows, then reduce the resultant array
>>> # [1, 1, 1] + [1, 1, 1] = [2, 2, 2]
>>> # 2 + 2 + 2 = 6
>>> tf.reduce_sum(x, [0, 1]).numpy()
6
Args:
input_tensor: The tensor to reduce. Should have numeric type.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor)]`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name for the operation (optional).
Returns:
The reduced tensor, of the same dtype as the input_tensor.
@compatibility(numpy)
Equivalent to np.sum apart the fact that numpy upcast uint8 and int32 to
int64 while tensorflow returns the same dtype as the input.
@end_compatibility"
8843,reduce_sum_with_dims,tensorflow/tensorflow/python/ops/math_ops.py,2006,function,
8844,reduce_euclidean_norm,tensorflow/tensorflow/python/ops/math_ops.py,2019,function,"Computes the Euclidean norm of elements across dimensions of a tensor.
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
of the entries in `axis`, which must be unique. If `keepdims` is true, the
reduced dimensions are retained with length 1.
If `axis` is None, all dimensions are reduced, and a
tensor with a single element is returned.
For example:
```python
x = tf.constant([[1, 2, 3], [1, 1, 1]]) # x.dtype is tf.int32
tf.math.reduce_euclidean_norm(x) # returns 4 as dtype is tf.int32
y = tf.constant([[1, 2, 3], [1, 1, 1]], dtype = tf.float32)
tf.math.reduce_euclidean_norm(y) # returns 4.1231055 which is sqrt(17)
tf.math.reduce_euclidean_norm(y, 0) # [sqrt(2), sqrt(5), sqrt(10)]
tf.math.reduce_euclidean_norm(y, 1) # [sqrt(14), sqrt(3)]
tf.math.reduce_euclidean_norm(y, 1, keepdims=True) # [[sqrt(14)], [sqrt(3)]]
tf.math.reduce_euclidean_norm(y, [0, 1]) # sqrt(17)
```
Args:
input_tensor: The tensor to reduce. Should have numeric type.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name for the operation (optional).
Returns:
The reduced tensor, of the same dtype as the input_tensor."
8845,count_nonzero,tensorflow/tensorflow/python/ops/math_ops.py,2069,function,"Computes number of nonzero elements across dimensions of a tensor.
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
entry in `axis`. If `keepdims` is true, the reduced dimensions
are retained with length 1.
If `axis` has no entries, all dimensions are reduced, and a
tensor with a single element is returned.
**NOTE** Floating point comparison to zero is done by exact floating point
equality check. Small values are **not** rounded to zero for purposes of
the nonzero check.
For example:
```python
x = tf.constant([[0, 1, 0], [1, 1, 0]])
tf.math.count_nonzero(x) # 3
tf.math.count_nonzero(x, 0) # [1, 2, 0]
tf.math.count_nonzero(x, 1) # [1, 2]
tf.math.count_nonzero(x, 1, keepdims=True) # [[1], [2]]
tf.math.count_nonzero(x, [0, 1]) # 3
```
**NOTE** Strings are compared against zero-length empty string `""""`. Any
string with a size greater than zero is already considered as nonzero.
For example:
```python
x = tf.constant(["""", ""a"", "" "", ""b"", """"])
tf.math.count_nonzero(x) # 3, with ""a"", "" "", and ""b"" as nonzero strings.
```
Args:
input_tensor: The tensor to reduce. Should be of numeric type, `bool`, or
`string`.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
dtype: The output dtype; defaults to `tf.int64`.
name: A name for the operation (optional).
reduction_indices: The old (deprecated) name for axis.
keep_dims: Deprecated alias for `keepdims`.
input: Overrides input_tensor. For compatibility.
Returns:
The reduced tensor (number of nonzero values)."
8846,count_nonzero_v2,tensorflow/tensorflow/python/ops/math_ops.py,2141,function,"Computes number of nonzero elements across dimensions of a tensor.
Reduces `input` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
entry in `axis`. If `keepdims` is true, the reduced dimensions
are retained with length 1.
If `axis` has no entries, all dimensions are reduced, and a
tensor with a single element is returned.
**NOTE** Floating point comparison to zero is done by exact floating point
equality check. Small values are **not** rounded to zero for purposes of
the nonzero check.
For example:
```python
x = tf.constant([[0, 1, 0], [1, 1, 0]])
tf.math.count_nonzero(x) # 3
tf.math.count_nonzero(x, 0) # [1, 2, 0]
tf.math.count_nonzero(x, 1) # [1, 2]
tf.math.count_nonzero(x, 1, keepdims=True) # [[1], [2]]
tf.math.count_nonzero(x, [0, 1]) # 3
```
**NOTE** Strings are compared against zero-length empty string `""""`. Any
string with a size greater than zero is already considered as nonzero.
For example:
```python
x = tf.constant(["""", ""a"", "" "", ""b"", """"])
tf.math.count_nonzero(x) # 3, with ""a"", "" "", and ""b"" as nonzero strings.
```
Args:
input: The tensor to reduce. Should be of numeric type, `bool`, or `string`.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input), rank(input))`.
keepdims: If true, retains reduced dimensions with length 1.
dtype: The output dtype; defaults to `tf.int64`.
name: A name for the operation (optional).
Returns:
The reduced tensor (number of nonzero values)."
8847,reduce_mean_v1,tensorflow/tensorflow/python/ops/math_ops.py,2209,function,"Computes the mean of elements across dimensions of a tensor.
Reduces `input_tensor` along the dimensions given in `axis` by computing the
mean of elements across the dimensions in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
the entries in `axis`, which must be unique. If `keepdims` is true, the
reduced dimensions are retained with length 1.
If `axis` is None, all dimensions are reduced, and a tensor with a single
element is returned.
For example:
>>> x = tf.constant([[1., 1.], [2., 2.]])
>>> tf.reduce_mean(x)
<tf.Tensor: shape=(), dtype=float32, numpy=1.5>
>>> tf.reduce_mean(x, 0)
<tf.Tensor: shape=(2,), dtype=float32, numpy=array([1.5, 1.5], dtype=float32)>
>>> tf.reduce_mean(x, 1)
<tf.Tensor: shape=(2,), dtype=float32, numpy=array([1., 2.], dtype=float32)>
Args:
input_tensor: The tensor to reduce. Should have numeric type.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name for the operation (optional).
reduction_indices: The old (deprecated) name for axis.
keep_dims: Deprecated alias for `keepdims`.
Returns:
The reduced tensor.
@compatibility(numpy)
Equivalent to np.mean
Please note that `np.mean` has a `dtype` parameter that could be used to
specify the output type. By default this is `dtype=float64`. On the other
hand, `tf.reduce_mean` has an aggressive type inference from `input_tensor`,
for example:
>>> x = tf.constant([1, 0, 1, 0])
>>> tf.reduce_mean(x)
<tf.Tensor: shape=(), dtype=int32, numpy=0>
>>> y = tf.constant([1., 0., 1., 0.])
>>> tf.reduce_mean(y)
<tf.Tensor: shape=(), dtype=float32, numpy=0.5>
@end_compatibility"
8848,reduce_mean,tensorflow/tensorflow/python/ops/math_ops.py,2276,function,"Computes the mean of elements across dimensions of a tensor.
Reduces `input_tensor` along the dimensions given in `axis` by computing the
mean of elements across the dimensions in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
of the entries in `axis`, which must be unique. If `keepdims` is true, the
reduced dimensions are retained with length 1.
If `axis` is None, all dimensions are reduced, and a tensor with a single
element is returned.
For example:
>>> x = tf.constant([[1., 1.], [2., 2.]])
>>> tf.reduce_mean(x)
<tf.Tensor: shape=(), dtype=float32, numpy=1.5>
>>> tf.reduce_mean(x, 0)
<tf.Tensor: shape=(2,), dtype=float32, numpy=array([1.5, 1.5], dtype=float32)>
>>> tf.reduce_mean(x, 1)
<tf.Tensor: shape=(2,), dtype=float32, numpy=array([1., 2.], dtype=float32)>
Args:
input_tensor: The tensor to reduce. Should have numeric type.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name for the operation (optional).
Returns:
The reduced tensor.
@compatibility(numpy)
Equivalent to np.mean
Please note that `np.mean` has a `dtype` parameter that could be used to
specify the output type. By default this is `dtype=float64`. On the other
hand, `tf.reduce_mean` has an aggressive type inference from `input_tensor`,
for example:
>>> x = tf.constant([1, 0, 1, 0])
>>> tf.reduce_mean(x)
<tf.Tensor: shape=(), dtype=int32, numpy=0>
>>> y = tf.constant([1., 0., 1., 0.])
>>> tf.reduce_mean(y)
<tf.Tensor: shape=(), dtype=float32, numpy=0.5>
@end_compatibility"
8849,reduce_variance,tensorflow/tensorflow/python/ops/math_ops.py,2336,function,"Computes the variance of elements across dimensions of a tensor.
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
of the entries in `axis`, which must be unique. If `keepdims` is true, the
reduced dimensions are retained with length 1.
If `axis` is None, all dimensions are reduced, and a
tensor with a single element is returned.
For example:
>>> x = tf.constant([[1., 2.], [3., 4.]])
>>> tf.math.reduce_variance(x)
<tf.Tensor: shape=(), dtype=float32, numpy=1.25>
>>> tf.math.reduce_variance(x, 0)
<tf.Tensor: shape=(2,), dtype=float32, numpy=array([1., 1.], ...)>
>>> tf.math.reduce_variance(x, 1)
<tf.Tensor: shape=(2,), dtype=float32, numpy=array([0.25, 0.25], ...)>
Args:
input_tensor: The tensor to reduce. Should have real or complex type.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name scope for the associated operations (optional).
Returns:
The reduced tensor, of the same dtype as the input_tensor. Note, for
`complex64` or `complex128` input, the returned `Tensor` will be of type
`float32` or `float64`, respectively.
@compatibility(numpy)
Equivalent to np.var
Please note `np.var` has a `dtype` parameter that could be used to specify the
output type. By default this is `dtype=float64`. On the other hand,
`tf.math.reduce_variance` has aggressive type inference from `input_tensor`.
@end_compatibility"
8850,reduce_std,tensorflow/tensorflow/python/ops/math_ops.py,2397,function,"Computes the standard deviation of elements across dimensions of a tensor.
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
of the entries in `axis`, which must be unique. If `keepdims` is true, the
reduced dimensions are retained with length 1.
If `axis` is None, all dimensions are reduced, and a
tensor with a single element is returned.
For example:
>>> x = tf.constant([[1., 2.], [3., 4.]])
>>> tf.math.reduce_std(x)
<tf.Tensor: shape=(), dtype=float32, numpy=1.118034>
>>> tf.math.reduce_std(x, 0)
<tf.Tensor: shape=(2,), dtype=float32, numpy=array([1., 1.], dtype=float32)>
>>> tf.math.reduce_std(x, 1)
<tf.Tensor: shape=(2,), dtype=float32, numpy=array([0.5, 0.5], dtype=float32)>
Args:
input_tensor: The tensor to reduce. Should have real or complex type.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name scope for the associated operations (optional).
Returns:
The reduced tensor, of the same dtype as the input_tensor. Note, for
`complex64` or `complex128` input, the returned `Tensor` will be of type
`float32` or `float64`, respectively.
@compatibility(numpy)
Equivalent to np.std
Please note `np.std` has a `dtype` parameter that could be used to specify the
output type. By default this is `dtype=float64`. On the other hand,
`tf.math.reduce_std` has aggressive type inference from `input_tensor`.
@end_compatibility"
8851,reduce_prod,tensorflow/tensorflow/python/ops/math_ops.py,2447,function,"Computes the product of elements across dimensions of a tensor.
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
entry in `axis`. If `keepdims` is true, the reduced dimensions
are retained with length 1.
If `axis` is None, all dimensions are reduced, and a
tensor with a single element is returned.
Args:
input_tensor: The tensor to reduce. Should have numeric type.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name for the operation (optional).
Returns:
The reduced tensor.
@compatibility(numpy)
Equivalent to np.prod
@end_compatibility"
8852,reduce_prod_v1,tensorflow/tensorflow/python/ops/math_ops.py,2486,function,"Computes the product of elements across dimensions of a tensor.
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
of the entries in `axis`, which must be unique. If `keepdims` is true, the
reduced dimensions are retained with length 1.
If `axis` is None, all dimensions are reduced, and a
tensor with a single element is returned.
Args:
input_tensor: The tensor to reduce. Should have numeric type.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name for the operation (optional).
reduction_indices: The old (deprecated) name for axis.
keep_dims: Deprecated alias for `keepdims`.
Returns:
The reduced tensor.
@compatibility(numpy)
Equivalent to np.prod
@end_compatibility"
8853,reduce_min_v1,tensorflow/tensorflow/python/ops/math_ops.py,2532,function,"Computes the minimum of elements across dimensions of a tensor.
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
of the entries in `axis`, which must be unique. If `keepdims` is true, the
reduced dimensions are retained with length 1.
If `axis` is None, all dimensions are reduced, and a
tensor with a single element is returned.
Args:
input_tensor: The tensor to reduce. Should have real numeric type.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name for the operation (optional).
reduction_indices: The old (deprecated) name for axis.
keep_dims: Deprecated alias for `keepdims`.
Returns:
The reduced tensor.
@compatibility(numpy)
Equivalent to np.min
@end_compatibility"
8854,reduce_min,tensorflow/tensorflow/python/ops/math_ops.py,2575,function,"Computes the minimum of elements across dimensions of a tensor.
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
of the entries in `axis`, which must be unique. If `keepdims` is true, the
reduced dimensions are retained with length 1.
If `axis` is None, all dimensions are reduced, and a
tensor with a single element is returned.
Args:
input_tensor: The tensor to reduce. Should have real numeric type.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name for the operation (optional).
Returns:
The reduced tensor.
For example:
>>> a = tf.constant([[1, 2], [3, 4]])
>>> tf.reduce_min(a)
<tf.Tensor: shape=(), dtype=int32, numpy=1>
@compatibility(numpy)
Equivalent to np.min
@end_compatibility"
8855,reduce_max_v1,tensorflow/tensorflow/python/ops/math_ops.py,2619,function,"Computes the maximum of elements across dimensions of a tensor.
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
of the entries in `axis`, which must be unique. If `keepdims` is true, the
reduced dimensions are retained with length 1.
If `axis` is None, all dimensions are reduced, and a
tensor with a single element is returned.
Args:
input_tensor: The tensor to reduce. Should have real numeric type.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name for the operation (optional).
reduction_indices: The old (deprecated) name for axis.
keep_dims: Deprecated alias for `keepdims`.
Returns:
The reduced tensor.
@compatibility(numpy)
Equivalent to np.max
@end_compatibility"
8856,reduce_max,tensorflow/tensorflow/python/ops/math_ops.py,2662,function,"Computes the maximum of elements across dimensions of a tensor.
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
of the entries in `axis`, which must be unique. If `keepdims` is true, the
reduced dimensions are retained with length 1.
If `axis` is None, all dimensions are reduced, and a
tensor with a single element is returned.
Usage example:
>>> x = tf.constant([5, 1, 2, 4])
>>> print(tf.reduce_max(x))
tf.Tensor(5, shape=(), dtype=int32)
>>> x = tf.constant([-5, -1, -2, -4])
>>> print(tf.reduce_max(x))
tf.Tensor(-1, shape=(), dtype=int32)
>>> x = tf.constant([4, float('nan')])
>>> print(tf.reduce_max(x))
tf.Tensor(4.0, shape=(), dtype=float32)
>>> x = tf.constant([float('nan'), float('nan')])
>>> print(tf.reduce_max(x))
tf.Tensor(-inf, shape=(), dtype=float32)
>>> x = tf.constant([float('-inf'), float('inf')])
>>> print(tf.reduce_max(x))
tf.Tensor(inf, shape=(), dtype=float32)
See the numpy docs for `np.amax` and `np.nanmax` behavior.
Args:
input_tensor: The tensor to reduce. Should have real numeric type.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name for the operation (optional).
Returns:
The reduced tensor."
8857,reduce_max_with_dims,tensorflow/tensorflow/python/ops/math_ops.py,2708,function,
8858,reduce_all_v1,tensorflow/tensorflow/python/ops/math_ops.py,2724,function,"Computes the ""logical and"" of elements across dimensions of a tensor.
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
of the entries in `axis`, which must be unique. If `keepdims` is true, the
reduced dimensions are retained with length 1.
If `axis` is None, all dimensions are reduced, and a
tensor with a single element is returned.
For example:
```python
x = tf.constant([[True, True], [False, False]])
tf.reduce_all(x) # False
tf.reduce_all(x, 0) # [False, False]
tf.reduce_all(x, 1) # [True, False]
```
Args:
input_tensor: The boolean tensor to reduce.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name for the operation (optional).
reduction_indices: The old (deprecated) name for axis.
keep_dims: Deprecated alias for `keepdims`.
Returns:
The reduced tensor.
@compatibility(numpy)
Equivalent to np.all
@end_compatibility"
8859,reduce_all,tensorflow/tensorflow/python/ops/math_ops.py,2776,function,"Computes the ""logical and"" of elements across dimensions of a tensor.
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
of the entries in `axis`, which must be unique. If `keepdims` is true, the
reduced dimensions are retained with length 1.
If `axis` is None, all dimensions are reduced, and a
tensor with a single element is returned.
For example:
```python
x = tf.constant([[True, True], [False, False]])
tf.reduce_all(x) # False
tf.reduce_all(x, 0) # [False, False]
tf.reduce_all(x, 1) # [True, False]
```
Args:
input_tensor: The boolean tensor to reduce.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name for the operation (optional).
Returns:
The reduced tensor.
@compatibility(numpy)
Equivalent to np.all
@end_compatibility"
8860,reduce_any_v1,tensorflow/tensorflow/python/ops/math_ops.py,2824,function,"Computes the ""logical or"" of elements across dimensions of a tensor.
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
of the entries in `axis`, which must be unique. If `keepdims` is true, the
reduced dimensions are retained with length 1.
If `axis` is None, all dimensions are reduced, and a
tensor with a single element is returned.
For example:
```python
x = tf.constant([[True, True], [False, False]])
tf.reduce_any(x) # True
tf.reduce_any(x, 0) # [True, True]
tf.reduce_any(x, 1) # [True, False]
```
Args:
input_tensor: The boolean tensor to reduce.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name for the operation (optional).
reduction_indices: The old (deprecated) name for axis.
keep_dims: Deprecated alias for `keepdims`.
Returns:
The reduced tensor.
@compatibility(numpy)
Equivalent to np.any
@end_compatibility"
8861,reduce_any,tensorflow/tensorflow/python/ops/math_ops.py,2876,function,"Computes the ""logical or"" of elements across dimensions of a tensor.
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
of the entries in `axis`, which must be unique. If `keepdims` is true, the
reduced dimensions are retained with length 1.
If `axis` is None, all dimensions are reduced, and a
tensor with a single element is returned.
For example:
```python
x = tf.constant([[True, True], [False, False]])
tf.reduce_any(x) # True
tf.reduce_any(x, 0) # [True, True]
tf.reduce_any(x, 1) # [True, False]
```
Args:
input_tensor: The boolean tensor to reduce.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name for the operation (optional).
Returns:
The reduced tensor.
@compatibility(numpy)
Equivalent to np.any
@end_compatibility"
8862,reduce_logsumexp_v1,tensorflow/tensorflow/python/ops/math_ops.py,2924,function,"Computes log(sum(exp(elements across dimensions of a tensor))).
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
of the entries in `axis`, which must be unique. If `keepdims` is true, the
reduced dimensions are retained with length 1.
If `axis` has no entries, all dimensions are reduced, and a
tensor with a single element is returned.
This function is more numerically stable than log(sum(exp(input))). It avoids
overflows caused by taking the exp of large inputs and underflows caused by
taking the log of small inputs.
For example:
```python
x = tf.constant([[0., 0., 0.], [0., 0., 0.]])
tf.reduce_logsumexp(x) # log(6)
tf.reduce_logsumexp(x, 0) # [log(2), log(2), log(2)]
tf.reduce_logsumexp(x, 1) # [log(3), log(3)]
tf.reduce_logsumexp(x, 1, keepdims=True) # [[log(3)], [log(3)]]
tf.reduce_logsumexp(x, [0, 1]) # log(6)
```
Args:
input_tensor: The tensor to reduce. Should have numeric type.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name for the operation (optional).
reduction_indices: The old (deprecated) name for axis.
keep_dims: Deprecated alias for `keepdims`.
Returns:
The reduced tensor."
8863,reduce_logsumexp,tensorflow/tensorflow/python/ops/math_ops.py,2978,function,"Computes log(sum(exp(elements across dimensions of a tensor))).
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keepdims` is true, the rank of the tensor is reduced by 1 for each
of the entries in `axis`, which must be unique. If `keepdims` is true, the
reduced dimensions are retained with length 1.
If `axis` has no entries, all dimensions are reduced, and a
tensor with a single element is returned.
This function is more numerically stable than log(sum(exp(input))). It avoids
overflows caused by taking the exp of large inputs and underflows caused by
taking the log of small inputs.
For example:
```python
x = tf.constant([[0., 0., 0.], [0., 0., 0.]])
tf.reduce_logsumexp(x) # log(6)
tf.reduce_logsumexp(x, 0) # [log(2), log(2), log(2)]
tf.reduce_logsumexp(x, 1) # [log(3), log(3)]
tf.reduce_logsumexp(x, 1, keepdims=True) # [[log(3)], [log(3)]]
tf.reduce_logsumexp(x, [0, 1]) # log(6)
```
Args:
input_tensor: The tensor to reduce. Should have numeric type.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name for the operation (optional).
Returns:
The reduced tensor."
8864,trace,tensorflow/tensorflow/python/ops/math_ops.py,3041,function,"Compute the trace of a tensor `x`.
`trace(x)` returns the sum along the main diagonal of each inner-most matrix
in x. If x is of rank `k` with shape `[I, J, K, ..., L, M, N]`, then output
is a tensor of rank `k-2` with dimensions `[I, J, K, ..., L]` where
`output[i, j, k, ..., l] = trace(x[i, j, k, ..., l, :, :])`
For example:
```python
x = tf.constant([[1, 2], [3, 4]])
tf.linalg.trace(x) # 5
x = tf.constant([[1, 2, 3],
[4, 5, 6],
[7, 8, 9]])
tf.linalg.trace(x) # 15
x = tf.constant([[[1, 2, 3],
[4, 5, 6],
[7, 8, 9]],
[[-1, -2, -3],
[-4, -5, -6],
[-7, -8, -9]]])
tf.linalg.trace(x) # [15, -15]
```
Args:
x: tensor.
name: A name for the operation (optional).
Returns:
The trace of input tensor."
8865,matmul,tensorflow/tensorflow/python/ops/math_ops.py,3084,function,"Multiplies matrix `a` by matrix `b`, producing `a` * `b`.
The inputs must, following any transpositions, be tensors of rank >= 2
where the inner 2 dimensions specify valid matrix multiplication dimensions,
and any further outer dimensions specify matching batch size.
Both matrices must be of the same type. The supported types are:
`float16`, `float32`, `float64`, `int32`, `complex64`, `complex128`.
Either matrix can be transposed or adjointed (conjugated and transposed) on
the fly by setting one of the corresponding flag to `True`. These are `False`
by default.
If one or both of the matrices contain a lot of zeros, a more efficient
multiplication algorithm can be used by setting the corresponding
`a_is_sparse` or `b_is_sparse` flag to `True`. These are `False` by default.
This optimization is only available for plain matrices (rank-2 tensors) with
datatypes `bfloat16` or `float32`.
A simple 2-D tensor matrix multiplication:
>>> a = tf.constant([1, 2, 3, 4, 5, 6], shape=[2, 3])
>>> a # 2-D tensor
<tf.Tensor: shape=(2, 3), dtype=int32, numpy=
array([[1, 2, 3],
[4, 5, 6]], dtype=int32)>
>>> b = tf.constant([7, 8, 9, 10, 11, 12], shape=[3, 2])
>>> b # 2-D tensor
<tf.Tensor: shape=(3, 2), dtype=int32, numpy=
array([[ 7, 8],
[ 9, 10],
[11, 12]], dtype=int32)>
>>> c = tf.matmul(a, b)
>>> c # `a` * `b`
<tf.Tensor: shape=(2, 2), dtype=int32, numpy=
array([[ 58, 64],
[139, 154]], dtype=int32)>
A batch matrix multiplication with batch shape [2]:
>>> a = tf.constant(np.arange(1, 13, dtype=np.int32), shape=[2, 2, 3])
>>> a # 3-D tensor
<tf.Tensor: shape=(2, 2, 3), dtype=int32, numpy=
array([[[ 1, 2, 3],
[ 4, 5, 6]],
[[ 7, 8, 9],
[10, 11, 12]]], dtype=int32)>
>>> b = tf.constant(np.arange(13, 25, dtype=np.int32), shape=[2, 3, 2])
>>> b # 3-D tensor
<tf.Tensor: shape=(2, 3, 2), dtype=int32, numpy=
array([[[13, 14],
[15, 16],
[17, 18]],
[[19, 20],
[21, 22],
[23, 24]]], dtype=int32)>
>>> c = tf.matmul(a, b)
>>> c # `a` * `b`
<tf.Tensor: shape=(2, 2, 2), dtype=int32, numpy=
array([[[ 94, 100],
[229, 244]],
[[508, 532],
[697, 730]]], dtype=int32)>
Since python >= 3.5 the @ operator is supported
(see [PEP 465](https://www.python.org/dev/peps/pep-0465/)). In TensorFlow,
it simply calls the `tf.matmul()` function, so the following lines are
equivalent:
>>> d = a @ b @ [[10], [11]]
>>> d = tf.matmul(tf.matmul(a, b), [[10], [11]])
Args:
a: `tf.Tensor` of type `float16`, `float32`, `float64`, `int32`,
`complex64`, `complex128` and rank > 1.
b: `tf.Tensor` with same type and rank as `a`.
transpose_a: If `True`, `a` is transposed before multiplication.
transpose_b: If `True`, `b` is transposed before multiplication.
adjoint_a: If `True`, `a` is conjugated and transposed before
multiplication.
adjoint_b: If `True`, `b` is conjugated and transposed before
multiplication.
a_is_sparse: If `True`, `a` is treated as a sparse matrix. Notice, this
**does not support `tf.sparse.SparseTensor`**, it just makes optimizations
that assume most values in `a` are zero.
See `tf.sparse.sparse_dense_matmul`
for some support for `tf.sparse.SparseTensor` multiplication.
b_is_sparse: If `True`, `b` is treated as a sparse matrix. Notice, this
**does not support `tf.sparse.SparseTensor`**, it just makes optimizations
that assume most values in `a` are zero.
See `tf.sparse.sparse_dense_matmul`
for some support for `tf.sparse.SparseTensor` multiplication.
name: Name for the operation (optional).
Returns:
A `tf.Tensor` of the same type as `a` and `b` where each inner-most matrix
is the product of the corresponding matrices in `a` and `b`, e.g. if all
transpose or adjoint attributes are `False`:
`output[..., i, j] = sum_k (a[..., i, k] * b[..., k, j])`,
for all indices `i`, `j`.
Note: This is matrix product, not element-wise product.
Raises:
ValueError: If `transpose_a` and `adjoint_a`, or `transpose_b` and
`adjoint_b` are both set to `True`."
8866,matvec,tensorflow/tensorflow/python/ops/math_ops.py,3279,function,"Multiplies matrix `a` by vector `b`, producing `a` * `b`.
The matrix `a` must, following any transpositions, be a tensor of rank >= 2,
with `shape(a)[-1] == shape(b)[-1]`, and `shape(a)[:-2]` able to broadcast
with `shape(b)[:-1]`.
Both `a` and `b` must be of the same type. The supported types are:
`float16`, `float32`, `float64`, `int32`, `complex64`, `complex128`.
Matrix `a` can be transposed or adjointed (conjugated and transposed) on
the fly by setting one of the corresponding flag to `True`. These are `False`
by default.
If one or both of the inputs contain a lot of zeros, a more efficient
multiplication algorithm can be used by setting the corresponding
`a_is_sparse` or `b_is_sparse` flag to `True`. These are `False` by default.
This optimization is only available for plain matrices/vectors (rank-2/1
tensors) with datatypes `bfloat16` or `float32`.
For example:
```python
# 2-D tensor `a`
# [[1, 2, 3],
# [4, 5, 6]]
a = tf.constant([1, 2, 3, 4, 5, 6], shape=[2, 3])
# 1-D tensor `b`
# [7, 9, 11]
b = tf.constant([7, 9, 11], shape=[3])
# `a` * `b`
# [ 58, 64]
c = tf.linalg.matvec(a, b)
# 3-D tensor `a`
# [[[ 1, 2, 3],
# [ 4, 5, 6]],
# [[ 7, 8, 9],
# [10, 11, 12]]]
a = tf.constant(np.arange(1, 13, dtype=np.int32),
shape=[2, 2, 3])
# 2-D tensor `b`
# [[13, 14, 15],
# [16, 17, 18]]
b = tf.constant(np.arange(13, 19, dtype=np.int32),
shape=[2, 3])
# `a` * `b`
# [[ 86, 212],
# [410, 563]]
c = tf.linalg.matvec(a, b)
```
Args:
a: `Tensor` of type `float16`, `float32`, `float64`, `int32`, `complex64`,
`complex128` and rank > 1.
b: `Tensor` with same type as `a` and compatible dimensions.
transpose_a: If `True`, `a` is transposed before multiplication.
adjoint_a: If `True`, `a` is conjugated and transposed before
multiplication.
a_is_sparse: If `True`, `a` is treated as a sparse matrix.
b_is_sparse: If `True`, `b` is treated as a sparse matrix.
name: Name for the operation (optional).
Returns:
A `Tensor` of the same type as `a` and `b` where each inner-most vector is
the product of the corresponding matrices in `a` and vectors in `b`, e.g. if
all transpose or adjoint attributes are `False`:
`output`[..., i] = sum_k (`a`[..., i, k] * `b`[..., k]), for all indices i.
Note: This is matrix-vector product, not element-wise product.
Raises:
ValueError: If transpose_a and adjoint_a are both set to True."
8867,add_n,tensorflow/tensorflow/python/ops/math_ops.py,3479,function,"Adds all input tensors element-wise.
`tf.math.add_n` performs the same operation as `tf.math.accumulate_n`, but it
waits for all of its inputs to be ready before beginning to sum.
This buffering can result in higher memory consumption when inputs are ready
at different times, since the minimum temporary storage required is
proportional to the input size rather than the output size.
This op does not [broadcast](
https://docs.scipy.org/doc/numpy-1.13.0/user/basics.broadcasting.html)
its inputs. If you need broadcasting, use `tf.math.add` (or the `+` operator)
instead.
For example:
>>> a = tf.constant([[3, 5], [4, 8]])
>>> b = tf.constant([[1, 6], [2, 9]])
>>> tf.math.add_n([a, b, a])
<tf.Tensor: shape=(2, 2), dtype=int32, numpy=
array([[ 7, 16],
[10, 25]], dtype=int32)>
Args:
inputs: A list of `tf.Tensor` or `tf.IndexedSlices` objects, each with the
same shape and type. `tf.IndexedSlices` objects will be converted into
dense tensors prior to adding.
name: A name for the operation (optional).
Returns:
A `tf.Tensor` of the same shape and type as the elements of `inputs`.
Raises:
ValueError: If `inputs` don't all have same shape and dtype or the shape
cannot be inferred."
8868,accumulate_n,tensorflow/tensorflow/python/ops/math_ops.py,3537,function,"Returns the element-wise sum of a list of tensors.
Optionally, pass `shape` and `tensor_dtype` for shape and type checking,
otherwise, these are inferred.
`accumulate_n` performs the same operation as `tf.math.add_n`.
For example:
```python
a = tf.constant([[1, 2], [3, 4]])
b = tf.constant([[5, 0], [0, 6]])
tf.math.accumulate_n([a, b, a]) # [[7, 4], [6, 14]]
# Explicitly pass shape and type
tf.math.accumulate_n([a, b, a], shape=[2, 2], tensor_dtype=tf.int32)
# [[7, 4],
# [6, 14]]
```
Args:
inputs: A list of `Tensor` objects, each with same shape and type.
shape: Expected shape of elements of `inputs` (optional). Also controls the
output shape of this op, which may affect type inference in other ops. A
value of `None` means ""infer the input shape from the shapes in `inputs`"".
tensor_dtype: Expected data type of `inputs` (optional). A value of `None`
means ""infer the input dtype from `inputs[0]`"".
name: A name for the operation (optional).
Returns:
A `Tensor` of same shape and type as the elements of `inputs`.
Raises:
ValueError: If `inputs` don't all have same shape and dtype or the shape
cannot be inferred."
8869,sigmoid,tensorflow/tensorflow/python/ops/math_ops.py,3615,function,"Computes sigmoid of `x` element-wise.
Formula for calculating sigmoid(x): `y = 1 / (1 + exp(-x))`.
For x \in (-inf, inf) => sigmoid(x) \in (0, 1)
Example Usage:
If a positive number is large, then its sigmoid will approach to 1 since the
formula will be `y = <large_num> / (1 + <large_num>)`
>>> x = tf.constant([0.0, 1.0, 50.0, 100.0])
>>> tf.math.sigmoid(x)
<tf.Tensor: shape=(4,), dtype=float32,
numpy=array([0.5 , 0.7310586, 1. , 1. ], dtype=float32)>
If a negative number is large, its sigmoid will approach to 0 since the
formula will be `y = 1 / (1 + <large_num>)`
>>> x = tf.constant([-100.0, -50.0, -1.0, 0.0])
>>> tf.math.sigmoid(x)
<tf.Tensor: shape=(4,), dtype=float32, numpy=
array([0.0000000e+00, 1.9287499e-22, 2.6894143e-01, 0.5],
dtype=float32)>
Args:
x: A Tensor with type `float16`, `float32`, `float64`, `complex64`, or
`complex128`.
name: A name for the operation (optional).
Returns:
A Tensor with the same type as `x`.
Usage Example:
>>> x = tf.constant([-128.0, 0.0, 128.0], dtype=tf.float32)
>>> tf.sigmoid(x)
<tf.Tensor: shape=(3,), dtype=float32,
numpy=array([0. , 0.5, 1. ], dtype=float32)>
@compatibility(scipy)
Equivalent to scipy.special.expit
@end_compatibility"
8870,log_sigmoid,tensorflow/tensorflow/python/ops/math_ops.py,3668,function,"Computes log sigmoid of `x` element-wise.
Specifically, `y = log(1 / (1 + exp(-x)))`. For numerical stability,
we use `y = -tf.nn.softplus(-x)`.
Args:
x: A Tensor with type `float32` or `float64`.
name: A name for the operation (optional).
Returns:
A Tensor with the same type as `x`."
8871,cumsum,tensorflow/tensorflow/python/ops/math_ops.py,3688,function,"Compute the cumulative sum of the tensor `x` along `axis`.
By default, this op performs an inclusive cumsum, which means that the first
element of the input is identical to the first element of the output:
For example:
>>> # tf.cumsum([a, b, c]) # [a, a + b, a + b + c]
>>> x = tf.constant([2, 4, 6, 8])
>>> tf.cumsum(x)
<tf.Tensor: shape=(4,), dtype=int32,
numpy=array([ 2, 6, 12, 20], dtype=int32)>
>>> # using varying `axis` values
>>> y = tf.constant([[2, 4, 6, 8], [1,3,5,7]])
>>> tf.cumsum(y, axis=0)
<tf.Tensor: shape=(2, 4), dtype=int32, numpy=
array([[ 2, 4, 6, 8],
[ 3, 7, 11, 15]], dtype=int32)>
>>> tf.cumsum(y, axis=1)
<tf.Tensor: shape=(2, 4), dtype=int32, numpy=
array([[ 2, 6, 12, 20],
[ 1, 4, 9, 16]], dtype=int32)>
By setting the `exclusive` kwarg to `True`, an exclusive cumsum is performed
instead:
>>> # tf.cumsum([a, b, c], exclusive=True) => [0, a, a + b]
>>> x = tf.constant([2, 4, 6, 8])
>>> tf.cumsum(x, exclusive=True)
<tf.Tensor: shape=(4,), dtype=int32,
numpy=array([ 0, 2, 6, 12], dtype=int32)>
By setting the `reverse` kwarg to `True`, the cumsum is performed in the
opposite direction:
>>> # tf.cumsum([a, b, c], reverse=True) # [a + b + c, b + c, c]
>>> x = tf.constant([2, 4, 6, 8])
>>> tf.cumsum(x, reverse=True)
<tf.Tensor: shape=(4,), dtype=int32,
numpy=array([20, 18, 14, 8], dtype=int32)>
This is more efficient than using separate `tf.reverse` ops.
The `reverse` and `exclusive` kwargs can also be combined:
>>> # tf.cumsum([a, b, c], exclusive=True, reverse=True) # [b + c, c, 0]
>>> x = tf.constant([2, 4, 6, 8])
>>> tf.cumsum(x, exclusive=True, reverse=True)
<tf.Tensor: shape=(4,), dtype=int32,
numpy=array([18, 14, 8, 0], dtype=int32)>
Args:
x: A `Tensor`. Must be one of the following types: `float32`, `float64`,
`int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`,
`complex128`, `qint8`, `quint8`, `qint32`, `half`.
axis: A `Tensor` of type `int32` (default: 0). Must be in the range
`[-rank(x), rank(x))`.
exclusive: If `True`, perform exclusive cumsum.
reverse: A `bool` (default: False).
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `x`."
8872,cumprod,tensorflow/tensorflow/python/ops/math_ops.py,3761,function,"Compute the cumulative product of the tensor `x` along `axis`.
By default, this op performs an inclusive cumprod, which means that the
first element of the input is identical to the first element of the output:
```python
tf.math.cumprod([a, b, c]) # [a, a * b, a * b * c]
```
By setting the `exclusive` kwarg to `True`, an exclusive cumprod is
performed
instead:
```python
tf.math.cumprod([a, b, c], exclusive=True) # [1, a, a * b]
```
By setting the `reverse` kwarg to `True`, the cumprod is performed in the
opposite direction:
```python
tf.math.cumprod([a, b, c], reverse=True) # [a * b * c, b * c, c]
```
This is more efficient than using separate `tf.reverse` ops.
The `reverse` and `exclusive` kwargs can also be combined:
```python
tf.math.cumprod([a, b, c], exclusive=True, reverse=True) # [b * c, c, 1]
```
Args:
x: A `Tensor`. Must be one of the following types: `float32`, `float64`,
`int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`,
`complex128`, `qint8`, `quint8`, `qint32`, `half`.
axis: A `Tensor` of type `int32` (default: 0). Must be in the range
`[-rank(x), rank(x))`.
exclusive: If `True`, perform exclusive cumprod.
reverse: A `bool` (default: False).
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `x`."
8873,cumulative_logsumexp,tensorflow/tensorflow/python/ops/math_ops.py,3814,function,"Compute the cumulative log-sum-exp of the tensor `x` along `axis`.
By default, this op performs an inclusive cumulative log-sum-exp, which means
that the first element of the input is identical to the first element of
the output.
This operation is significantly more numerically stable than the equivalent
tensorflow operation `tf.math.log(tf.math.cumsum(tf.math.exp(x)))`, although
computes the same result given infinite numerical precision. However, note
that in some cases, it may be less stable than `tf.math.reduce_logsumexp`
for a given element, as it applies the ""log-sum-exp trick"" in a different
way.
More precisely, where `tf.math.reduce_logsumexp` uses the following trick:
```
log(sum(exp(x))) == log(sum(exp(x - max(x)))) + max(x)
```
it cannot be directly used here as there is no fast way of applying it
to each prefix `x[:i]`. Instead, this function implements a prefix
scan using pairwise log-add-exp, which is a commutative and associative
(up to floating point precision) operator:
```
log_add_exp(x, y) = log(exp(x) + exp(y))
= log(1 + exp(min(x, y) - max(x, y))) + max(x, y)
```
However, reducing using the above operator leads to a different computation
tree (logs are taken repeatedly instead of only at the end), and the maximum
is only computed pairwise instead of over the entire prefix. In general, this
leads to a different and slightly less precise computation.
Args:
x: A `Tensor`. Must be one of the following types: `float16`, `float32`,
`float64`.
axis: A `Tensor` of type `int32` or `int64` (default: 0). Must be in the
range `[-rank(x), rank(x))`.
exclusive: If `True`, perform exclusive cumulative log-sum-exp.
reverse: If `True`, performs the cumulative log-sum-exp in the reverse
direction.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same shape and type as `x`."
8874,conj,tensorflow/tensorflow/python/ops/math_ops.py,3871,function,"Returns the complex conjugate of a complex number.
Given a tensor `input` of complex numbers, this operation returns a tensor of
complex numbers that are the complex conjugate of each element in `input`. The
complex numbers in `input` must be of the form \\(a + bj\\), where *a* is the
real part and *b* is the imaginary part.
The complex conjugate returned by this operation is of the form \\(a - bj\\).
For example:
# tensor 'input' is [-2.25 + 4.75j, 3.25 + 5.75j]
tf.math.conj(input) ==> [-2.25 - 4.75j, 3.25 - 5.75j]
If `x` is real, it is returned unchanged.
Args:
x: `Tensor` to conjugate. Must have numeric or variant type.
name: A name for the operation (optional).
Returns:
A `Tensor` that is the conjugate of `x` (with the same type).
Raises:
TypeError: If `x` is not a numeric tensor."
8875,reduced_shape,tensorflow/tensorflow/python/ops/math_ops.py,3913,function,"Helper function for reduction ops.
Args:
input_shape: 1-D Tensor, the shape of the Tensor being reduced.
axes: 1-D Tensor, the reduction axes.
Returns:
A 1-D Tensor, the output shape as if keepdims were set to True."
8876,unsorted_segment_mean,tensorflow/tensorflow/python/ops/math_ops.py,3983,function,"Computes the mean along segments of a tensor.
Read [the section on
segmentation](https://www.tensorflow.org/versions/r2.0/api_docs/python/tf/math#about_segmentation)
for an explanation of segments.
This operator is similar to the unsorted segment sum operator found
[here](../../../api_docs/python/math_ops.md#UnsortedSegmentSum).
Instead of computing the sum over segments, it computes the mean of all
entries belonging to a segment such that:
\\(output_i = 1/N_i \sum_{j...} data[j...]\\) where the sum is over tuples
`j...` such that `segment_ids[j...] == i` with \\N_i\\ being the number of
occurrences of id \\i\\.
If there is no entry for a given segment ID `i`, it outputs 0.
If the given segment ID `i` is negative, the value is dropped and will not
be added to the sum of the segment.
Args:
data: A `Tensor` with floating point or complex dtype.
segment_ids: An integer tensor whose shape is a prefix of `data.shape`.
num_segments: An integer scalar `Tensor`. The number of distinct segment
IDs.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has same shape as data, except for the first `segment_ids.rank`
dimensions, which are replaced with a single dimension which has size
`num_segments`."
8877,unsorted_segment_sqrt_n,tensorflow/tensorflow/python/ops/math_ops.py,4030,function,"Computes the sum along segments of a tensor divided by the sqrt(N).
Read [the section on
segmentation](https://www.tensorflow.org/versions/r2.0/api_docs/python/tf/math#about_segmentation)
for an explanation of segments.
This operator is similar to the unsorted segment sum operator found
[here](../../../api_docs/python/math_ops.md#UnsortedSegmentSum).
Additionally to computing the sum over segments, it divides the results by
sqrt(N).
\\(output_i = 1/sqrt(N_i) \sum_{j...} data[j...]\\) where the sum is over
tuples `j...` such that `segment_ids[j...] == i` with \\N_i\\ being the
number of occurrences of id \\i\\.
If there is no entry for a given segment ID `i`, it outputs 0.
Note that this op only supports floating point and complex dtypes,
due to tf.sqrt only supporting these types.
If the given segment ID `i` is negative, the value is dropped and will not
be added to the sum of the segment.
Args:
data: A `Tensor` with floating point or complex dtype.
segment_ids: An integer tensor whose shape is a prefix of `data.shape`.
num_segments: An integer scalar `Tensor`. The number of distinct segment
IDs.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has same shape as data, except for the first `segment_ids.rank`
dimensions, which are replaced with a single dimension which has size
`num_segments`."
8878,sparse_segment_sum,tensorflow/tensorflow/python/ops/math_ops.py,4076,function,"Computes the sum along sparse segments of a tensor.
Read [the section on
segmentation](https://www.tensorflow.org/versions/r2.0/api_docs/python/tf/math#about_segmentation)
for an explanation of segments.
Like `tf.math.segment_sum`, but `segment_ids` can have rank less than `data`'s
first dimension, selecting a subset of dimension 0, specified by `indices`.
`segment_ids` is allowed to have missing ids, in which case the output will
be zeros at those indices. In those cases `num_segments` is used to determine
the size of the output.
For example:
```python
c = tf.constant([[1,2,3,4], [-1,-2,-3,-4], [5,6,7,8]])
# Select two rows, one segment.
tf.sparse.segment_sum(c, tf.constant([0, 1]), tf.constant([0, 0]))
# => [[0 0 0 0]]
# Select two rows, two segment.
tf.sparse.segment_sum(c, tf.constant([0, 1]), tf.constant([0, 1]))
# => [[ 1 2 3 4]
# [-1 -2 -3 -4]]
# With missing segment ids.
tf.sparse.segment_sum(c, tf.constant([0, 1]), tf.constant([0, 2]),
num_segments=4)
# => [[ 1 2 3 4]
# [ 0 0 0 0]
# [-1 -2 -3 -4]
# [ 0 0 0 0]]
# Select all rows, two segments.
tf.sparse.segment_sum(c, tf.constant([0, 1, 2]), tf.constant([0, 0, 1]))
# => [[0 0 0 0]
# [5 6 7 8]]
# Which is equivalent to:
tf.math.segment_sum(c, tf.constant([0, 0, 1]))
```
Args:
data: A `Tensor` with data that will be assembled in the output.
indices: A 1-D `Tensor` with indices into `data`. Has same rank as
`segment_ids`.
segment_ids: A 1-D `Tensor` with indices into the output `Tensor`. Values
should be sorted and can be repeated.
name: A name for the operation (optional).
num_segments: An optional int32 scalar. Indicates the size of the output
`Tensor`.
Returns:
A `tensor` of the shape as data, except for dimension 0 which
has size `k`, the number of segments specified via `num_segments` or
inferred for the last element in `segments_ids`."
8879,sparse_segment_sum_v2,tensorflow/tensorflow/python/ops/math_ops.py,4152,function,"Computes the sum along sparse segments of a tensor.
Read [the section on
segmentation](https://www.tensorflow.org/versions/r2.0/api_docs/python/tf/math#about_segmentation)
for an explanation of segments.
Like `tf.math.segment_sum`, but `segment_ids` can have rank less than `data`'s
first dimension, selecting a subset of dimension 0, specified by `indices`.
`segment_ids` is allowed to have missing ids, in which case the output will
be zeros at those indices. In those cases `num_segments` is used to determine
the size of the output.
For example:
```python
c = tf.constant([[1,2,3,4], [-1,-2,-3,-4], [5,6,7,8]])
# Select two rows, one segment.
tf.sparse.segment_sum(c, tf.constant([0, 1]), tf.constant([0, 0]))
# => [[0 0 0 0]]
# Select two rows, two segment.
tf.sparse.segment_sum(c, tf.constant([0, 1]), tf.constant([0, 1]))
# => [[ 1 2 3 4]
# [-1 -2 -3 -4]]
# With missing segment ids.
tf.sparse.segment_sum(c, tf.constant([0, 1]), tf.constant([0, 2]),
num_segments=4)
# => [[ 1 2 3 4]
# [ 0 0 0 0]
# [-1 -2 -3 -4]
# [ 0 0 0 0]]
# Select all rows, two segments.
tf.sparse.segment_sum(c, tf.constant([0, 1, 2]), tf.constant([0, 0, 1]))
# => [[0 0 0 0]
# [5 6 7 8]]
# Which is equivalent to:
tf.math.segment_sum(c, tf.constant([0, 0, 1]))
```
Args:
data: A `Tensor` with data that will be assembled in the output.
indices: A 1-D `Tensor` with indices into `data`. Has same rank as
`segment_ids`.
segment_ids: A 1-D `Tensor` with indices into the output `Tensor`. Values
should be sorted and can be repeated.
num_segments: An optional int32 scalar. Indicates the size of the output
`Tensor`.
name: A name for the operation (optional).
Returns:
A `tensor` of the shape as data, except for dimension 0 which
has size `k`, the number of segments specified via `num_segments` or
inferred for the last element in `segments_ids`."
8880,sparse_segment_mean,tensorflow/tensorflow/python/ops/math_ops.py,4221,function,"Computes the mean along sparse segments of a tensor.
Read [the section on
segmentation](https://www.tensorflow.org/versions/r2.0/api_docs/python/tf/math#about_segmentation)
for an explanation of segments.
Like `tf.math.segment_mean`, but `segment_ids` can have rank less than
`data`'s first dimension, selecting a subset of dimension 0, specified by
`indices`.
`segment_ids` is allowed to have missing ids, in which case the output will
be zeros at those indices. In those cases `num_segments` is used to determine
the size of the output.
Args:
data: A `Tensor` with data that will be assembled in the output.
indices: A 1-D `Tensor` with indices into `data`. Has same rank as
`segment_ids`.
segment_ids: A 1-D `Tensor` with indices into the output `Tensor`. Values
should be sorted and can be repeated.
name: A name for the operation (optional).
num_segments: An optional int32 scalar. Indicates the size of the output
`Tensor`.
Returns:
A `tensor` of the shape as data, except for dimension 0 which
has size `k`, the number of segments specified via `num_segments` or
inferred for the last element in `segments_ids`."
8881,sparse_segment_mean_v2,tensorflow/tensorflow/python/ops/math_ops.py,4267,function,"Computes the mean along sparse segments of a tensor.
Read [the section on
segmentation](https://www.tensorflow.org/versions/r2.0/api_docs/python/tf/math#about_segmentation)
for an explanation of segments.
Like `tf.math.segment_mean`, but `segment_ids` can have rank less than
`data`'s first dimension, selecting a subset of dimension 0, specified by
`indices`.
`segment_ids` is allowed to have missing ids, in which case the output will
be zeros at those indices. In those cases `num_segments` is used to determine
the size of the output.
Args:
data: A `Tensor` with data that will be assembled in the output.
indices: A 1-D `Tensor` with indices into `data`. Has same rank as
`segment_ids`.
segment_ids: A 1-D `Tensor` with indices into the output `Tensor`. Values
should be sorted and can be repeated.
num_segments: An optional int32 scalar. Indicates the size of the output
`Tensor`.
name: A name for the operation (optional).
Returns:
A `tensor` of the shape as data, except for dimension 0 which
has size `k`, the number of segments specified via `num_segments` or
inferred for the last element in `segments_ids`."
8882,sparse_segment_sqrt_n,tensorflow/tensorflow/python/ops/math_ops.py,4306,function,"Computes the sum along sparse segments of a tensor divided by the sqrt(N).
`N` is the size of the segment being reduced.
Args:
data: A `Tensor` with data that will be assembled in the output.
indices: A 1-D `Tensor` with indices into `data`. Has same rank as
`segment_ids`.
segment_ids: A 1-D `Tensor` with indices into the output `Tensor`. Values
should be sorted and can be repeated.
name: A name for the operation (optional).
num_segments: An optional int32 scalar. Indicates the size of the output
`Tensor`.
Returns:
A `tensor` of the shape as data, except for dimension 0 which
has size `k`, the number of segments specified via `num_segments` or
inferred for the last element in `segments_ids`."
8883,sparse_segment_sqrt_n_v2,tensorflow/tensorflow/python/ops/math_ops.py,4343,function,"Computes the sum along sparse segments of a tensor divided by the sqrt(N).
Read [the section on
segmentation](https://www.tensorflow.org/versions/r2.0/api_docs/python/tf/math#about_segmentation)
for an explanation of segments.
Like `tf.sparse.segment_mean`, but instead of dividing by the size of the
segment, `N`, divide by `sqrt(N)` instead.
Args:
data: A `Tensor` with data that will be assembled in the output.
indices: A 1-D `Tensor` with indices into `data`. Has same rank as
`segment_ids`.
segment_ids: A 1-D `Tensor` with indices into the output `Tensor`. Values
should be sorted and can be repeated.
num_segments: An optional int32 scalar. Indicates the size of the output
`Tensor`.
name: A name for the operation (optional).
Returns:
A `tensor` of the shape as data, except for dimension 0 which
has size `k`, the number of segments specified via `num_segments` or
inferred for the last element in `segments_ids`."
8884,tensordot,tensorflow/tensorflow/python/ops/math_ops.py,4378,function,"Tensor contraction of a and b along specified axes and outer product.
Tensordot (also known as tensor contraction) sums the product of elements
from `a` and `b` over the indices specified by `a_axes` and `b_axes`.
The lists `a_axes` and `b_axes` specify those pairs of axes along which to
contract the tensors. The axis `a_axes[i]` of `a` must have the same dimension
as axis `b_axes[i]` of `b` for all `i` in `range(0, len(a_axes))`. The lists
`a_axes` and `b_axes` must have identical length and consist of unique
integers that specify valid axes for each of the tensors. Additionally
outer product is supported by passing `axes=0`.
This operation corresponds to `numpy.tensordot(a, b, axes)`.
Example 1: When `a` and `b` are matrices (order 2), the case `axes = 1`
is equivalent to matrix multiplication.
Example 2: When `a` and `b` are matrices (order 2), the case
`axes = [[1], [0]]` is equivalent to matrix multiplication.
Example 3: When `a` and `b` are matrices (order 2), the case `axes=0` gives
the outer product, a tensor of order 4.
Example 4: Suppose that \\(a_{ijk}\\) and \\(b_{lmn}\\) represent two
tensors of order 3. Then, `contract(a, b, [[0], [2]])` is the order 4 tensor
\\(c_{jklm}\\) whose entry
corresponding to the indices \\((j,k,l,m)\\) is given by:
\\( c_{jklm} = \sum_i a_{ijk} b_{lmi} \\).
In general, `order(c) = order(a) + order(b) - 2*len(axes[0])`.
Args:
a: `Tensor` of type `float32` or `float64`.
b: `Tensor` with the same type as `a`.
axes: Either a scalar `N`, or a list or an `int32` `Tensor` of shape [2, k].
If axes is a scalar, sum over the last N axes of a and the first N axes of
b in order. If axes is a list or `Tensor` the first and second row contain
the set of unique integers specifying axes along which the contraction is
computed, for `a` and `b`, respectively. The number of axes for `a` and
`b` must be equal. If `axes=0`, computes the outer product between `a` and
`b`.
name: A name for the operation (optional).
Returns:
A `Tensor` with the same type as `a`.
Raises:
ValueError: If the shapes of `a`, `b`, and `axes` are incompatible.
IndexError: If the values in axes exceed the rank of the corresponding
tensor."
8885,polyval,tensorflow/tensorflow/python/ops/math_ops.py,4565,function,"Computes the elementwise value of a polynomial.
If `x` is a tensor and `coeffs` is a list n + 1 tensors,
this function returns the value of the n-th order polynomial
p(x) = coeffs[n-1] + coeffs[n-2] * x + ... + coeffs[0] * x**(n-1)
evaluated using Horner's method, i.e.
p(x) = coeffs[n-1] + x * (coeffs[n-2] + ... + x * (coeffs[1] +
x * coeffs[0]))
Usage Example:
>>> coefficients = [1.0, 2.5, -4.2]
>>> x = 5.0
>>> y = tf.math.polyval(coefficients, x)
>>> y
<tf.Tensor: shape=(), dtype=float32, numpy=33.3>
Usage Example:
>>> tf.math.polyval([2, 1, 0], 3) # evaluates 2 * (3**2) + 1 * (3**1) + 0 * (3**0)
<tf.Tensor: shape=(), dtype=int32, numpy=21>
`tf.math.polyval` can also be used in polynomial regression. Taking
advantage of this function can facilitate writing a polynomial equation
as compared to explicitly writing it out, especially for higher degree
polynomials.
>>> x = tf.constant(3)
>>> theta1 = tf.Variable(2)
>>> theta2 = tf.Variable(1)
>>> theta3 = tf.Variable(0)
>>> tf.math.polyval([theta1, theta2, theta3], x)
<tf.Tensor: shape=(), dtype=int32, numpy=21>
Args:
coeffs: A list of `Tensor` representing the coefficients of the polynomial.
x: A `Tensor` representing the variable of the polynomial.
name: A name for the operation (optional).
Returns:
A `tensor` of the shape as the expression p(x) with usual broadcasting
rules for element-wise addition and multiplication applied.
@compatibility(numpy)
Equivalent to numpy.polyval.
@end_compatibility"
8886,reciprocal_no_nan,tensorflow/tensorflow/python/ops/math_ops.py,4636,function,"Performs a safe reciprocal operation, element wise.
If a particular element is zero, the reciprocal for that element is
also set to zero.
For example:
```python
x = tf.constant([2.0, 0.5, 0, 1], dtype=tf.float32)
tf.math.reciprocal_no_nan(x) # [ 0.5, 2, 0.0, 1.0 ]
```
Args:
x: A `Tensor` of type `float16`, `float32`, `float64` `complex64` or
`complex128`.
name: A name for the operation (optional).
Returns:
A `Tensor` of same shape and type as `x`.
Raises:
TypeError: x must be of a valid dtype."
8887,xlog1py,tensorflow/tensorflow/python/ops/math_ops.py,4669,function,"Compute x * log1p(y).
Given `x` and `y`, compute `x * log1p(y)`. This function safely returns
zero when `x = 0`, no matter what the value of `y` is.
Example:
>>> tf.math.xlog1py(0., 1.)
<tf.Tensor: shape=(), dtype=float32, numpy=0.>
>>> tf.math.xlog1py(1., 1.)
<tf.Tensor: shape=(), dtype=float32, numpy=0.6931472>
>>> tf.math.xlog1py(2., 2.)
<tf.Tensor: shape=(), dtype=float32, numpy=2.1972246>
>>> tf.math.xlog1py(0., -1.)
<tf.Tensor: shape=(), dtype=float32, numpy=0.>
Args:
x: A `tf.Tensor` of type `bfloat16`, `half`, `float32`, `float64`,
`complex64`, `complex128`
y: A `tf.Tensor` of type `bfloat16`, `half`, `float32`, `float64`,
`complex64`, `complex128`
name: A name for the operation (optional).
Returns:
`x * log1p(y)`.
@compatibility(scipy)
Equivalent to scipy.special.xlog1py
@end_compatibility"
8888,erfinv,tensorflow/tensorflow/python/ops/math_ops.py,4706,function,"Compute inverse error function.
Given `x`, compute the inverse error function of `x`. This function
is the inverse of `tf.math.erf`.
Args:
x: `Tensor` with type `float` or `double`.
name: A name for the operation (optional).
Returns:
Inverse error function of `x`."
8889,ndtri,tensorflow/tensorflow/python/ops/math_ops.py,4724,function,"Compute quantile of Standard Normal.
Args:
x: `Tensor` with type `float` or `double`.
name: A name for the operation (optional).
Returns:
Inverse error function of `x`."
8890,ceil,tensorflow/tensorflow/python/ops/math_ops.py,4741,function,"Return the ceiling of the input, element-wise.
For example:
>>> tf.math.ceil([-1.7, -1.5, -0.2, 0.2, 1.5, 1.7, 2.0])
<tf.Tensor: shape=(7,), dtype=float32,
numpy=array([-1., -1., -0., 1., 2., 2., 2.], dtype=float32)>
Args:
x: A `tf.Tensor`. Must be one of the following types: `bfloat16`, `half`,
`float32`, `float64`. `int32`
name: A name for the operation (optional).
Returns:
A `tf.Tensor`. Has the same type as `x`.
@compatibility(numpy)
Equivalent to np.ceil
@end_compatibility"
8891,sqrt,tensorflow/tensorflow/python/ops/math_ops.py,4767,function,"Computes element-wise square root of the input tensor.
Note: This operation does not support integer types.
>>> x = tf.constant([[4.0], [16.0]])
>>> tf.sqrt(x)
<tf.Tensor: shape=(2, 1), dtype=float32, numpy=
array([[2.],
[4.]], dtype=float32)>
>>> y = tf.constant([[-4.0], [16.0]])
>>> tf.sqrt(y)
<tf.Tensor: shape=(2, 1), dtype=float32, numpy=
array([[nan],
[ 4.]], dtype=float32)>
>>> z = tf.constant([[-1.0], [16.0]], dtype=tf.complex128)
>>> tf.sqrt(z)
<tf.Tensor: shape=(2, 1), dtype=complex128, numpy=
array([[0.0+1.j],
[4.0+0.j]])>
Note: In order to support complex complex, please provide an input tensor
of `complex64` or `complex128`.
Args:
x: A `tf.Tensor` of type `bfloat16`, `half`, `float32`, `float64`,
`complex64`, `complex128`
name: A name for the operation (optional).
Returns:
A `tf.Tensor` of same size, type and sparsity as `x`."
8892,exp,tensorflow/tensorflow/python/ops/math_ops.py,4805,function,"Computes exponential of x element-wise. \\(y = e^x\\).
This function computes the exponential of the input tensor element-wise.
i.e. `math.exp(x)` or \\(e^x\\), where `x` is the input tensor.
\\(e\\) denotes Euler's number and is approximately equal to 2.718281.
Output is positive for any real input.
>>> x = tf.constant(2.0)
>>> tf.math.exp(x)
<tf.Tensor: shape=(), dtype=float32, numpy=7.389056>
>>> x = tf.constant([2.0, 8.0])
>>> tf.math.exp(x)
<tf.Tensor: shape=(2,), dtype=float32,
numpy=array([ 7.389056, 2980.958 ], dtype=float32)>
For complex numbers, the exponential value is calculated as
\\(e^{x+iy}={e^x}{e^{iy}}={e^x}{\\cos(y)+i\\sin(y)}\\)
For `1+1j` the value would be computed as:
\\(e^1{\\cos(1)+i\\sin(1)} = 2.7182817 \\times (0.5403023+0.84147096j)\\)
>>> x = tf.constant(1 + 1j)
>>> tf.math.exp(x)
<tf.Tensor: shape=(), dtype=complex128,
numpy=(1.4686939399158851+2.2873552871788423j)>
Args:
x: A `tf.Tensor`. Must be one of the following types: `bfloat16`, `half`,
`float32`, `float64`, `complex64`, `complex128`.
name: A name for the operation (optional).
Returns:
A `tf.Tensor`. Has the same type as `x`.
@compatibility(numpy)
Equivalent to np.exp
@end_compatibility"
8893,sobol_sample,tensorflow/tensorflow/python/ops/math_ops.py,4853,function,"Generates points from the Sobol sequence.
Creates a Sobol sequence with `num_results` samples. Each sample has dimension
`dim`. Skips the first `skip` samples.
Args:
dim: Positive scalar `Tensor` representing each sample's dimension.
num_results: Positive scalar `Tensor` of dtype int32. The number of Sobol
points to return in the output.
skip: (Optional) Positive scalar `Tensor` of dtype int32. The number of
initial points of the Sobol sequence to skip. Default value is 0.
dtype: (Optional) The `tf.Dtype` of the sample. One of: `tf.float32` or
`tf.float64`. Defaults to `tf.float32`.
name: (Optional) Python `str` name prefixed to ops created by this function.
Returns:
`Tensor` of samples from Sobol sequence with `shape` [num_results, dim]."
8894,rsqrt,tensorflow/tensorflow/python/ops/math_ops.py,4880,function,"Computes reciprocal of square root of x element-wise.
For example:
>>> x = tf.constant([2., 0., -2.])
>>> tf.math.rsqrt(x)
<tf.Tensor: shape=(3,), dtype=float32,
numpy=array([0.707, inf, nan], dtype=float32)>
Args:
x: A `tf.Tensor`. Must be one of the following types: `bfloat16`, `half`,
`float32`, `float64`. `int32`
name: A name for the operation (optional).
Returns:
A `tf.Tensor`. Has the same type as `x`."
8895,build_graph,tensorflow/tensorflow/python/ops/matmul_benchmark.py,35,function,"Build a graph containing a sequence of matmul operations.
Args:
device: String, the device to run on.
n: tensor A's first dimension size.
m: tensor A's second dimension size.
k: tensor B's second dimension size.
transpose_a: boolean value to show if tensor A is transposed.
transpose_b: boolean value to show if tensor B is transposed.
dtype: numpy data type of the input tensor.
Returns:
A matmul operation to run()"
8896,MatmulBenchmark,tensorflow/tensorflow/python/ops/matmul_benchmark.py,68,class,Benchmark matmul!
8897,run_graph,tensorflow/tensorflow/python/ops/matmul_benchmark.py,71,method,"Run the graph and print its execution time.
Args:
device: String, the device to run on.
n: tensor A's first dimension size.
m: tensor A's second dimension size.
k: tensor B's second dimension size.
transpose_a: boolean value to show if tensor A is transposed.
transpose_b: boolean value to show if tensor B is transposed.
num_iters: number of iterations to run the benchmark.
dtype: numpy data type of the input tensor.
Returns:
The duration of the run in seconds."
8898,benchmark_matmul,tensorflow/tensorflow/python/ops/matmul_benchmark.py,141,method,
8899,metric_variable,tensorflow/tensorflow/python/ops/metrics_impl.py,41,function,"Create variable in `GraphKeys.(LOCAL|METRIC_VARIABLES)` collections.
If running in a `DistributionStrategy` context, the variable will be
""sync on read"". This means:
* The returned object will be a container with separate variables
per replica of the model.
* When writing to the variable, e.g. using `assign_add` in a metric
update, the update will be applied to the variable local to the
replica.
* To get a metric's result value, we need to sum the variable values
across the replicas before computing the final answer. Furthermore,
the final answer should be computed once instead of in every
replica. Both of these are accomplished by running the computation
of the final result value inside
`distribution_strategy_context.get_replica_context().merge_call(fn)`.
Inside the `merge_call()`, ops are only added to the graph once
and access to a sync on read variable in a computation returns
the sum across all replicas.
Args:
shape: Shape of the created variable.
dtype: Type of the created variable.
validate_shape: (Optional) Whether shape validation is enabled for
the created variable.
name: (Optional) String name of the created variable.
Returns:
A (non-trainable) variable initialized to zero, or if inside a
`DistributionStrategy` scope a sync on read variable container."
8900,mean,tensorflow/tensorflow/python/ops/metrics_impl.py,316,function,"Computes the (weighted) mean of the given values.
The `mean` function creates two local variables, `total` and `count`
that are used to compute the average of `values`. This average is ultimately
returned as `mean` which is an idempotent operation that simply divides
`total` by `count`.
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the `mean`.
`update_op` increments `total` with the reduced sum of the product of `values`
and `weights`, and it increments `count` with the reduced sum of `weights`.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
values: A `Tensor` of arbitrary dimensions.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`values`, and must be broadcastable to `values` (i.e., all dimensions must
be either `1`, or the same as the corresponding `values` dimension).
metrics_collections: An optional list of collections that `mean`
should be added to.
updates_collections: An optional list of collections that `update_op`
should be added to.
name: An optional variable_scope name.
Returns:
mean: A `Tensor` representing the current mean, the value of `total` divided
by `count`.
update_op: An operation that increments the `total` and `count` variables
appropriately and whose value matches `mean_value`.
Raises:
ValueError: If `weights` is not `None` and its shape doesn't match `values`,
or if either `metrics_collections` or `updates_collections` are not a list
or tuple.
RuntimeError: If eager execution is enabled."
8901,accuracy,tensorflow/tensorflow/python/ops/metrics_impl.py,397,function,"Calculates how often `predictions` matches `labels`.
The `accuracy` function creates two local variables, `total` and
`count` that are used to compute the frequency with which `predictions`
matches `labels`. This frequency is ultimately returned as `accuracy`: an
idempotent operation that simply divides `total` by `count`.
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the `accuracy`.
Internally, an `is_correct` operation computes a `Tensor` with elements 1.0
where the corresponding elements of `predictions` and `labels` match and 0.0
otherwise. Then `update_op` increments `total` with the reduced sum of the
product of `weights` and `is_correct`, and it increments `count` with the
reduced sum of `weights`.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: The ground truth values, a `Tensor` whose shape matches
`predictions`.
predictions: The predicted values, a `Tensor` of any shape.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that `accuracy` should
be added to.
updates_collections: An optional list of collections that `update_op` should
be added to.
name: An optional variable_scope name.
Returns:
accuracy: A `Tensor` representing the accuracy, the value of `total` divided
by `count`.
update_op: An operation that increments the `total` and `count` variables
appropriately and whose value matches `accuracy`.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8902,auc,tensorflow/tensorflow/python/ops/metrics_impl.py,633,function,"Computes the approximate AUC via a Riemann sum.
The `auc` function creates four local variables, `true_positives`,
`true_negatives`, `false_positives` and `false_negatives` that are used to
compute the AUC. To discretize the AUC curve, a linearly spaced set of
thresholds is used to compute pairs of recall and precision values. The area
under the ROC-curve is therefore computed using the height of the recall
values by the false positive rate, while the area under the PR-curve is the
computed using the height of the precision values by the recall.
This value is ultimately returned as `auc`, an idempotent operation that
computes the area under a discretized curve of precision versus recall values
(computed using the aforementioned variables). The `num_thresholds` variable
controls the degree of discretization with larger numbers of thresholds more
closely approximating the true AUC. The quality of the approximation may vary
dramatically depending on `num_thresholds`.
For best results, `predictions` should be distributed approximately uniformly
in the range [0, 1] and not peaked around 0 or 1. The quality of the AUC
approximation may be poor if this is not the case. Setting `summation_method`
to 'minoring' or 'majoring' can help quantify the error in the approximation
by providing lower or upper bound estimate of the AUC. The `thresholds`
parameter can be used to manually specify thresholds which split the
predictions more evenly.
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the `auc`.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: A `Tensor` whose shape matches `predictions`. Will be cast to
`bool`.
predictions: A floating point `Tensor` of arbitrary shape and whose values
are in the range `[0, 1]`.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
num_thresholds: The number of thresholds to use when discretizing the roc
curve.
metrics_collections: An optional list of collections that `auc` should be
added to.
updates_collections: An optional list of collections that `update_op` should
be added to.
curve: Specifies the name of the curve to be computed, 'ROC' [default] or
'PR' for the Precision-Recall-curve.
name: An optional variable_scope name.
summation_method: Specifies the Riemann summation method used
(https://en.wikipedia.org/wiki/Riemann_sum): 'trapezoidal' [default] that
applies the trapezoidal rule; 'careful_interpolation', a variant of it
differing only by a more correct interpolation scheme for PR-AUC -
interpolating (true/false) positives but not the ratio that is precision;
'minoring' that applies left summation for increasing intervals and right
summation for decreasing intervals; 'majoring' that does the opposite.
Note that 'careful_interpolation' is strictly preferred to 'trapezoidal'
(to be deprecated soon) as it applies the same method for ROC, and a
better one (see Davis & Goadrich 2006 for details) for the PR curve.
thresholds: An optional list of floating point values to use as the
thresholds for discretizing the curve. If set, the `num_thresholds`
parameter is ignored. Values should be in [0, 1]. Endpoint thresholds
equal to {-epsilon, 1+epsilon} for a small positive epsilon value will be
automatically included with these to correctly handle predictions equal to
exactly 0 or 1.
Returns:
auc: A scalar `Tensor` representing the current area-under-curve.
update_op: An operation that increments the `true_positives`,
`true_negatives`, `false_positives` and `false_negatives` variables
appropriately and whose value matches `auc`.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8903,mean_absolute_error,tensorflow/tensorflow/python/ops/metrics_impl.py,863,function,"Computes the mean absolute error between the labels and predictions.
The `mean_absolute_error` function creates two local variables,
`total` and `count` that are used to compute the mean absolute error. This
average is weighted by `weights`, and it is ultimately returned as
`mean_absolute_error`: an idempotent operation that simply divides `total` by
`count`.
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the
`mean_absolute_error`. Internally, an `absolute_errors` operation computes the
absolute value of the differences between `predictions` and `labels`. Then
`update_op` increments `total` with the reduced sum of the product of
`weights` and `absolute_errors`, and it increments `count` with the reduced
sum of `weights`
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: A `Tensor` of the same shape as `predictions`.
predictions: A `Tensor` of arbitrary shape.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that
`mean_absolute_error` should be added to.
updates_collections: An optional list of collections that `update_op` should
be added to.
name: An optional variable_scope name.
Returns:
mean_absolute_error: A `Tensor` representing the current mean, the value of
`total` divided by `count`.
update_op: An operation that increments the `total` and `count` variables
appropriately and whose value matches `mean_absolute_error`.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8904,mean_cosine_distance,tensorflow/tensorflow/python/ops/metrics_impl.py,924,function,"Computes the cosine distance between the labels and predictions.
The `mean_cosine_distance` function creates two local variables,
`total` and `count` that are used to compute the average cosine distance
between `predictions` and `labels`. This average is weighted by `weights`,
and it is ultimately returned as `mean_distance`, which is an idempotent
operation that simply divides `total` by `count`.
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the
`mean_distance`.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: A `Tensor` of arbitrary shape.
predictions: A `Tensor` of the same shape as `labels`.
dim: The dimension along which the cosine distance is computed.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension). Also,
dimension `dim` must be `1`.
metrics_collections: An optional list of collections that the metric
value variable should be added to.
updates_collections: An optional list of collections that the metric update
ops should be added to.
name: An optional variable_scope name.
Returns:
mean_distance: A `Tensor` representing the current mean, the value of
`total` divided by `count`.
update_op: An operation that increments the `total` and `count` variables
appropriately.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8905,mean_per_class_accuracy,tensorflow/tensorflow/python/ops/metrics_impl.py,998,function,"Calculates the mean of the per-class accuracies.
Calculates the accuracy for each class, then takes the mean of that.
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates the accuracy of each class and returns
them.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: A `Tensor` of ground truth labels with shape [batch size] and of
type `int32` or `int64`. The tensor will be flattened if its rank > 1.
predictions: A `Tensor` of prediction results for semantic labels, whose
shape is [batch size] and type `int32` or `int64`. The tensor will be
flattened if its rank > 1.
num_classes: The possible number of labels the prediction task can
have. This value must be provided, since two variables with shape =
[num_classes] will be allocated.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that
`mean_per_class_accuracy'
should be added to.
updates_collections: An optional list of collections `update_op` should be
added to.
name: An optional variable_scope name.
Returns:
mean_accuracy: A `Tensor` representing the mean per class accuracy.
update_op: An operation that updates the accuracy tensor.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8906,mean_iou,tensorflow/tensorflow/python/ops/metrics_impl.py,1103,function,"Calculate per-step mean Intersection-Over-Union (mIOU).
Mean Intersection-Over-Union is a common evaluation metric for
semantic image segmentation, which first computes the IOU for each
semantic class and then computes the average over classes.
IOU is defined as follows:
IOU = true_positive / (true_positive + false_positive + false_negative).
The predictions are accumulated in a confusion matrix, weighted by `weights`,
and mIOU is then calculated from it.
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the `mean_iou`.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: A `Tensor` of ground truth labels with shape [batch size] and of
type `int32` or `int64`. The tensor will be flattened if its rank > 1.
predictions: A `Tensor` of prediction results for semantic labels, whose
shape is [batch size] and type `int32` or `int64`. The tensor will be
flattened if its rank > 1.
num_classes: The possible number of labels the prediction task can
have. This value must be provided, since a confusion matrix of
dimension = [num_classes, num_classes] will be allocated.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that `mean_iou`
should be added to.
updates_collections: An optional list of collections `update_op` should be
added to.
name: An optional variable_scope name.
Returns:
mean_iou: A `Tensor` representing the mean intersection-over-union.
update_op: An operation that increments the confusion matrix.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8907,mean_relative_error,tensorflow/tensorflow/python/ops/metrics_impl.py,1206,function,"Computes the mean relative error by normalizing with the given values.
The `mean_relative_error` function creates two local variables,
`total` and `count` that are used to compute the mean relative absolute error.
This average is weighted by `weights`, and it is ultimately returned as
`mean_relative_error`: an idempotent operation that simply divides `total` by
`count`.
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the
`mean_reative_error`. Internally, a `relative_errors` operation divides the
absolute value of the differences between `predictions` and `labels` by the
`normalizer`. Then `update_op` increments `total` with the reduced sum of the
product of `weights` and `relative_errors`, and it increments `count` with the
reduced sum of `weights`.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: A `Tensor` of the same shape as `predictions`.
predictions: A `Tensor` of arbitrary shape.
normalizer: A `Tensor` of the same shape as `predictions`.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that
`mean_relative_error` should be added to.
updates_collections: An optional list of collections that `update_op` should
be added to.
name: An optional variable_scope name.
Returns:
mean_relative_error: A `Tensor` representing the current mean, the value of
`total` divided by `count`.
update_op: An operation that increments the `total` and `count` variables
appropriately and whose value matches `mean_relative_error`.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8908,mean_squared_error,tensorflow/tensorflow/python/ops/metrics_impl.py,1275,function,"Computes the mean squared error between the labels and predictions.
The `mean_squared_error` function creates two local variables,
`total` and `count` that are used to compute the mean squared error.
This average is weighted by `weights`, and it is ultimately returned as
`mean_squared_error`: an idempotent operation that simply divides `total` by
`count`.
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the
`mean_squared_error`. Internally, a `squared_error` operation computes the
element-wise square of the difference between `predictions` and `labels`. Then
`update_op` increments `total` with the reduced sum of the product of
`weights` and `squared_error`, and it increments `count` with the reduced sum
of `weights`.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: A `Tensor` of the same shape as `predictions`.
predictions: A `Tensor` of arbitrary shape.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that
`mean_squared_error` should be added to.
updates_collections: An optional list of collections that `update_op` should
be added to.
name: An optional variable_scope name.
Returns:
mean_squared_error: A `Tensor` representing the current mean, the value of
`total` divided by `count`.
update_op: An operation that increments the `total` and `count` variables
appropriately and whose value matches `mean_squared_error`.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8909,mean_tensor,tensorflow/tensorflow/python/ops/metrics_impl.py,1336,function,"Computes the element-wise (weighted) mean of the given tensors.
In contrast to the `mean` function which returns a scalar with the
mean, this function returns an average tensor with the same shape as the
input tensors.
The `mean_tensor` function creates two local variables,
`total_tensor` and `count_tensor` that are used to compute the average of
`values`. This average is ultimately returned as `mean` which is an idempotent
operation that simply divides `total` by `count`.
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the `mean`.
`update_op` increments `total` with the reduced sum of the product of `values`
and `weights`, and it increments `count` with the reduced sum of `weights`.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
values: A `Tensor` of arbitrary dimensions.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`values`, and must be broadcastable to `values` (i.e., all dimensions must
be either `1`, or the same as the corresponding `values` dimension).
metrics_collections: An optional list of collections that `mean`
should be added to.
updates_collections: An optional list of collections that `update_op`
should be added to.
name: An optional variable_scope name.
Returns:
mean: A float `Tensor` representing the current mean, the value of `total`
divided by `count`.
update_op: An operation that increments the `total` and `count` variables
appropriately and whose value matches `mean_value`.
Raises:
ValueError: If `weights` is not `None` and its shape doesn't match `values`,
or if either `metrics_collections` or `updates_collections` are not a list
or tuple.
RuntimeError: If eager execution is enabled."
8910,percentage_below,tensorflow/tensorflow/python/ops/metrics_impl.py,1421,function,"Computes the percentage of values less than the given threshold.
The `percentage_below` function creates two local variables,
`total` and `count` that are used to compute the percentage of `values` that
fall below `threshold`. This rate is weighted by `weights`, and it is
ultimately returned as `percentage` which is an idempotent operation that
simply divides `total` by `count`.
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the
`percentage`.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
values: A numeric `Tensor` of arbitrary size.
threshold: A scalar threshold.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`values`, and must be broadcastable to `values` (i.e., all dimensions must
be either `1`, or the same as the corresponding `values` dimension).
metrics_collections: An optional list of collections that the metric
value variable should be added to.
updates_collections: An optional list of collections that the metric update
ops should be added to.
name: An optional variable_scope name.
Returns:
percentage: A `Tensor` representing the current mean, the value of `total`
divided by `count`.
update_op: An operation that increments the `total` and `count` variables
appropriately.
Raises:
ValueError: If `weights` is not `None` and its shape doesn't match `values`,
or if either `metrics_collections` or `updates_collections` are not a list
or tuple.
RuntimeError: If eager execution is enabled."
8911,false_negatives,tensorflow/tensorflow/python/ops/metrics_impl.py,1522,function,"Computes the total number of false negatives.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: The ground truth values, a `Tensor` whose dimensions must match
`predictions`. Will be cast to `bool`.
predictions: The predicted values, a `Tensor` of arbitrary dimensions. Will
be cast to `bool`.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that the metric
value variable should be added to.
updates_collections: An optional list of collections that the metric update
ops should be added to.
name: An optional variable_scope name.
Returns:
value_tensor: A `Tensor` representing the current value of the metric.
update_op: An operation that accumulates the error from a batch of data.
Raises:
ValueError: If `weights` is not `None` and its shape doesn't match `values`,
or if either `metrics_collections` or `updates_collections` are not a list
or tuple.
RuntimeError: If eager execution is enabled."
8912,false_negatives_at_thresholds,tensorflow/tensorflow/python/ops/metrics_impl.py,1574,function,"Computes false negatives at provided threshold values.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: A `Tensor` whose shape matches `predictions`. Will be cast to
`bool`.
predictions: A floating point `Tensor` of arbitrary shape and whose values
are in the range `[0, 1]`.
thresholds: A python list or tuple of float thresholds in `[0, 1]`.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that `false_negatives`
should be added to.
updates_collections: An optional list of collections that `update_op` should
be added to.
name: An optional variable_scope name.
Returns:
false_negatives: A float `Tensor` of shape `[len(thresholds)]`.
update_op: An operation that updates the `false_negatives` variable and
returns its current value.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8913,false_positives,tensorflow/tensorflow/python/ops/metrics_impl.py,1630,function,"Sum the weights of false positives.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: The ground truth values, a `Tensor` whose dimensions must match
`predictions`. Will be cast to `bool`.
predictions: The predicted values, a `Tensor` of arbitrary dimensions. Will
be cast to `bool`.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that the metric
value variable should be added to.
updates_collections: An optional list of collections that the metric update
ops should be added to.
name: An optional variable_scope name.
Returns:
value_tensor: A `Tensor` representing the current value of the metric.
update_op: An operation that accumulates the error from a batch of data.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8914,false_positives_at_thresholds,tensorflow/tensorflow/python/ops/metrics_impl.py,1683,function,"Computes false positives at provided threshold values.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: A `Tensor` whose shape matches `predictions`. Will be cast to
`bool`.
predictions: A floating point `Tensor` of arbitrary shape and whose values
are in the range `[0, 1]`.
thresholds: A python list or tuple of float thresholds in `[0, 1]`.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that `false_positives`
should be added to.
updates_collections: An optional list of collections that `update_op` should
be added to.
name: An optional variable_scope name.
Returns:
false_positives: A float `Tensor` of shape `[len(thresholds)]`.
update_op: An operation that updates the `false_positives` variable and
returns its current value.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8915,true_negatives,tensorflow/tensorflow/python/ops/metrics_impl.py,1739,function,"Sum the weights of true_negatives.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: The ground truth values, a `Tensor` whose dimensions must match
`predictions`. Will be cast to `bool`.
predictions: The predicted values, a `Tensor` of arbitrary dimensions. Will
be cast to `bool`.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that the metric
value variable should be added to.
updates_collections: An optional list of collections that the metric update
ops should be added to.
name: An optional variable_scope name.
Returns:
value_tensor: A `Tensor` representing the current value of the metric.
update_op: An operation that accumulates the error from a batch of data.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8916,true_negatives_at_thresholds,tensorflow/tensorflow/python/ops/metrics_impl.py,1792,function,"Computes true negatives at provided threshold values.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: A `Tensor` whose shape matches `predictions`. Will be cast to
`bool`.
predictions: A floating point `Tensor` of arbitrary shape and whose values
are in the range `[0, 1]`.
thresholds: A python list or tuple of float thresholds in `[0, 1]`.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that `true_negatives`
should be added to.
updates_collections: An optional list of collections that `update_op` should
be added to.
name: An optional variable_scope name.
Returns:
true_negatives: A float `Tensor` of shape `[len(thresholds)]`.
update_op: An operation that updates the `true_negatives` variable and
returns its current value.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8917,true_positives,tensorflow/tensorflow/python/ops/metrics_impl.py,1848,function,"Sum the weights of true_positives.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: The ground truth values, a `Tensor` whose dimensions must match
`predictions`. Will be cast to `bool`.
predictions: The predicted values, a `Tensor` of arbitrary dimensions. Will
be cast to `bool`.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that the metric
value variable should be added to.
updates_collections: An optional list of collections that the metric update
ops should be added to.
name: An optional variable_scope name.
Returns:
value_tensor: A `Tensor` representing the current value of the metric.
update_op: An operation that accumulates the error from a batch of data.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8918,true_positives_at_thresholds,tensorflow/tensorflow/python/ops/metrics_impl.py,1901,function,"Computes true positives at provided threshold values.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: A `Tensor` whose shape matches `predictions`. Will be cast to
`bool`.
predictions: A floating point `Tensor` of arbitrary shape and whose values
are in the range `[0, 1]`.
thresholds: A python list or tuple of float thresholds in `[0, 1]`.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that `true_positives`
should be added to.
updates_collections: An optional list of collections that `update_op` should
be added to.
name: An optional variable_scope name.
Returns:
true_positives: A float `Tensor` of shape `[len(thresholds)]`.
update_op: An operation that updates the `true_positives` variable and
returns its current value.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8919,precision,tensorflow/tensorflow/python/ops/metrics_impl.py,1957,function,"Computes the precision of the predictions with respect to the labels.
The `precision` function creates two local variables,
`true_positives` and `false_positives`, that are used to compute the
precision. This value is ultimately returned as `precision`, an idempotent
operation that simply divides `true_positives` by the sum of `true_positives`
and `false_positives`.
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the
`precision`. `update_op` weights each prediction by the corresponding value in
`weights`.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: The ground truth values, a `Tensor` whose dimensions must match
`predictions`. Will be cast to `bool`.
predictions: The predicted values, a `Tensor` of arbitrary dimensions. Will
be cast to `bool`.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that `precision` should
be added to.
updates_collections: An optional list of collections that `update_op` should
be added to.
name: An optional variable_scope name.
Returns:
precision: Scalar float `Tensor` with the value of `true_positives`
divided by the sum of `true_positives` and `false_positives`.
update_op: `Operation` that increments `true_positives` and
`false_positives` variables appropriately and whose value matches
`precision`.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8920,precision_at_thresholds,tensorflow/tensorflow/python/ops/metrics_impl.py,2052,function,"Computes precision values for different `thresholds` on `predictions`.
The `precision_at_thresholds` function creates four local variables,
`true_positives`, `true_negatives`, `false_positives` and `false_negatives`
for various values of thresholds. `precision[i]` is defined as the total
weight of values in `predictions` above `thresholds[i]` whose corresponding
entry in `labels` is `True`, divided by the total weight of values in
`predictions` above `thresholds[i]` (`true_positives[i] / (true_positives[i] +
false_positives[i])`).
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the
`precision`.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: The ground truth values, a `Tensor` whose dimensions must match
`predictions`. Will be cast to `bool`.
predictions: A floating point `Tensor` of arbitrary shape and whose values
are in the range `[0, 1]`.
thresholds: A python list or tuple of float thresholds in `[0, 1]`.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that `auc` should be
added to.
updates_collections: An optional list of collections that `update_op` should
be added to.
name: An optional variable_scope name.
Returns:
precision: A float `Tensor` of shape `[len(thresholds)]`.
update_op: An operation that increments the `true_positives`,
`true_negatives`, `false_positives` and `false_negatives` variables that
are used in the computation of `precision`.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8921,recall,tensorflow/tensorflow/python/ops/metrics_impl.py,2133,function,"Computes the recall of the predictions with respect to the labels.
The `recall` function creates two local variables, `true_positives`
and `false_negatives`, that are used to compute the recall. This value is
ultimately returned as `recall`, an idempotent operation that simply divides
`true_positives` by the sum of `true_positives` and `false_negatives`.
For estimation of the metric over a stream of data, the function creates an
`update_op` that updates these variables and returns the `recall`. `update_op`
weights each prediction by the corresponding value in `weights`.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: The ground truth values, a `Tensor` whose dimensions must match
`predictions`. Will be cast to `bool`.
predictions: The predicted values, a `Tensor` of arbitrary dimensions. Will
be cast to `bool`.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that `recall` should
be added to.
updates_collections: An optional list of collections that `update_op` should
be added to.
name: An optional variable_scope name.
Returns:
recall: Scalar float `Tensor` with the value of `true_positives` divided
by the sum of `true_positives` and `false_negatives`.
update_op: `Operation` that increments `true_positives` and
`false_negatives` variables appropriately and whose value matches
`recall`.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8922,recall_at_k,tensorflow/tensorflow/python/ops/metrics_impl.py,2484,function,"Computes recall@k of the predictions with respect to sparse labels.
If `class_id` is specified, we calculate recall by considering only the
entries in the batch for which `class_id` is in the label, and computing
the fraction of them for which `class_id` is in the top-k `predictions`.
If `class_id` is not specified, we'll calculate recall as how often on
average a class among the labels of a batch entry is in the top-k
`predictions`.
`sparse_recall_at_k` creates two local variables,
`true_positive_at_<k>` and `false_negative_at_<k>`, that are used to compute
the recall_at_k frequency. This frequency is ultimately returned as
`recall_at_<k>`: an idempotent operation that simply divides
`true_positive_at_<k>` by total (`true_positive_at_<k>` +
`false_negative_at_<k>`).
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the
`recall_at_<k>`. Internally, a `top_k` operation computes a `Tensor`
indicating the top `k` `predictions`. Set operations applied to `top_k` and
`labels` calculate the true positives and false negatives weighted by
`weights`. Then `update_op` increments `true_positive_at_<k>` and
`false_negative_at_<k>` using these values.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: `int64` `Tensor` or `SparseTensor` with shape
[D1, ... DN, num_labels] or [D1, ... DN], where the latter implies
num_labels=1. N >= 1 and num_labels is the number of target classes for
the associated prediction. Commonly, N=1 and `labels` has shape
[batch_size, num_labels]. [D1, ... DN] must match `predictions`. Values
should be in range [0, num_classes), where num_classes is the last
dimension of `predictions`. Values outside this range always count
towards `false_negative_at_<k>`.
predictions: Float `Tensor` with shape [D1, ... DN, num_classes] where
N >= 1. Commonly, N=1 and predictions has shape [batch size, num_classes].
The final dimension contains the logit values for each class. [D1, ... DN]
must match `labels`.
k: Integer, k for @k metric.
class_id: Integer class ID for which we want binary metrics. This should be
in range [0, num_classes), where num_classes is the last dimension of
`predictions`. If class_id is outside this range, the method returns NAN.
weights: `Tensor` whose rank is either 0, or n-1, where n is the rank of
`labels`. If the latter, it must be broadcastable to `labels` (i.e., all
dimensions must be either `1`, or the same as the corresponding `labels`
dimension).
metrics_collections: An optional list of collections that values should
be added to.
updates_collections: An optional list of collections that updates should
be added to.
name: Name of new update operation, and namespace for other dependent ops.
Returns:
recall: Scalar `float64` `Tensor` with the value of `true_positives` divided
by the sum of `true_positives` and `false_negatives`.
update_op: `Operation` that increments `true_positives` and
`false_negatives` variables appropriately, and whose value matches
`recall`.
Raises:
ValueError: If `weights` is not `None` and its shape doesn't match
`predictions`, or if either `metrics_collections` or `updates_collections`
are not a list or tuple.
RuntimeError: If eager execution is enabled."
8923,recall_at_top_k,tensorflow/tensorflow/python/ops/metrics_impl.py,2577,function,"Computes recall@k of top-k predictions with respect to sparse labels.
Differs from `recall_at_k` in that predictions must be in the form of top `k`
class indices, whereas `recall_at_k` expects logits. Refer to `recall_at_k`
for more details.
Args:
labels: `int64` `Tensor` or `SparseTensor` with shape
[D1, ... DN, num_labels] or [D1, ... DN], where the latter implies
num_labels=1. N >= 1 and num_labels is the number of target classes for
the associated prediction. Commonly, N=1 and `labels` has shape
[batch_size, num_labels]. [D1, ... DN] must match `predictions`. Values
should be in range [0, num_classes), where num_classes is the last
dimension of `predictions`. Values outside this range always count
towards `false_negative_at_<k>`.
predictions_idx: Integer `Tensor` with shape [D1, ... DN, k] where N >= 1.
Commonly, N=1 and predictions has shape [batch size, k]. The final
dimension contains the top `k` predicted class indices. [D1, ... DN] must
match `labels`.
k: Integer, k for @k metric. Only used for the default op name.
class_id: Integer class ID for which we want binary metrics. This should be
in range [0, num_classes), where num_classes is the last dimension of
`predictions`. If class_id is outside this range, the method returns NAN.
weights: `Tensor` whose rank is either 0, or n-1, where n is the rank of
`labels`. If the latter, it must be broadcastable to `labels` (i.e., all
dimensions must be either `1`, or the same as the corresponding `labels`
dimension).
metrics_collections: An optional list of collections that values should
be added to.
updates_collections: An optional list of collections that updates should
be added to.
name: Name of new update operation, and namespace for other dependent ops.
Returns:
recall: Scalar `float64` `Tensor` with the value of `true_positives` divided
by the sum of `true_positives` and `false_negatives`.
update_op: `Operation` that increments `true_positives` and
`false_negatives` variables appropriately, and whose value matches
`recall`.
Raises:
ValueError: If `weights` is not `None` and its shape doesn't match
`predictions`, or if either `metrics_collections` or `updates_collections`
are not a list or tuple."
8924,recall_at_thresholds,tensorflow/tensorflow/python/ops/metrics_impl.py,2661,function,"Computes various recall values for different `thresholds` on `predictions`.
The `recall_at_thresholds` function creates four local variables,
`true_positives`, `true_negatives`, `false_positives` and `false_negatives`
for various values of thresholds. `recall[i]` is defined as the total weight
of values in `predictions` above `thresholds[i]` whose corresponding entry in
`labels` is `True`, divided by the total weight of `True` values in `labels`
(`true_positives[i] / (true_positives[i] + false_negatives[i])`).
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the `recall`.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: The ground truth values, a `Tensor` whose dimensions must match
`predictions`. Will be cast to `bool`.
predictions: A floating point `Tensor` of arbitrary shape and whose values
are in the range `[0, 1]`.
thresholds: A python list or tuple of float thresholds in `[0, 1]`.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that `recall` should be
added to.
updates_collections: An optional list of collections that `update_op` should
be added to.
name: An optional variable_scope name.
Returns:
recall: A float `Tensor` of shape `[len(thresholds)]`.
update_op: An operation that increments the `true_positives`,
`true_negatives`, `false_positives` and `false_negatives` variables that
are used in the computation of `recall`.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8925,root_mean_squared_error,tensorflow/tensorflow/python/ops/metrics_impl.py,2739,function,"Computes the root mean squared error between the labels and predictions.
The `root_mean_squared_error` function creates two local variables,
`total` and `count` that are used to compute the root mean squared error.
This average is weighted by `weights`, and it is ultimately returned as
`root_mean_squared_error`: an idempotent operation that takes the square root
of the division of `total` by `count`.
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the
`root_mean_squared_error`. Internally, a `squared_error` operation computes
the element-wise square of the difference between `predictions` and `labels`.
Then `update_op` increments `total` with the reduced sum of the product of
`weights` and `squared_error`, and it increments `count` with the reduced sum
of `weights`.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: A `Tensor` of the same shape as `predictions`.
predictions: A `Tensor` of arbitrary shape.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
metrics_collections: An optional list of collections that
`root_mean_squared_error` should be added to.
updates_collections: An optional list of collections that `update_op` should
be added to.
name: An optional variable_scope name.
Returns:
root_mean_squared_error: A `Tensor` representing the current mean, the value
of `total` divided by `count`.
update_op: An operation that increments the `total` and `count` variables
appropriately and whose value matches `root_mean_squared_error`.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, or if
`weights` is not `None` and its shape doesn't match `predictions`, or if
either `metrics_collections` or `updates_collections` are not a list or
tuple.
RuntimeError: If eager execution is enabled."
8926,sensitivity_at_specificity,tensorflow/tensorflow/python/ops/metrics_impl.py,2810,function,"Computes the specificity at a given sensitivity.
The `sensitivity_at_specificity` function creates four local
variables, `true_positives`, `true_negatives`, `false_positives` and
`false_negatives` that are used to compute the sensitivity at the given
specificity value. The threshold for the given specificity value is computed
and used to evaluate the corresponding sensitivity.
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the
`sensitivity`. `update_op` increments the `true_positives`, `true_negatives`,
`false_positives` and `false_negatives` counts with the weight of each case
found in the `predictions` and `labels`.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
For additional information about specificity and sensitivity, see the
following: https://en.wikipedia.org/wiki/Sensitivity_and_specificity
Args:
labels: The ground truth values, a `Tensor` whose dimensions must match
`predictions`. Will be cast to `bool`.
predictions: A floating point `Tensor` of arbitrary shape and whose values
are in the range `[0, 1]`.
specificity: A scalar value in range `[0, 1]`.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
num_thresholds: The number of thresholds to use for matching the given
specificity.
metrics_collections: An optional list of collections that `sensitivity`
should be added to.
updates_collections: An optional list of collections that `update_op` should
be added to.
name: An optional variable_scope name.
Returns:
sensitivity: A scalar `Tensor` representing the sensitivity at the given
`specificity` value.
update_op: An operation that increments the `true_positives`,
`true_negatives`, `false_positives` and `false_negatives` variables
appropriately and whose value matches `sensitivity`.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, if
`weights` is not `None` and its shape doesn't match `predictions`, or if
`specificity` is not between 0 and 1, or if either `metrics_collections`
or `updates_collections` are not a list or tuple.
RuntimeError: If eager execution is enabled."
8927,sparse_average_precision_at_k,tensorflow/tensorflow/python/ops/metrics_impl.py,3228,function,"Renamed to `average_precision_at_k`, please use that method instead."
8928,average_precision_at_k,tensorflow/tensorflow/python/ops/metrics_impl.py,3247,function,"Computes average precision@k of predictions with respect to sparse labels.
`average_precision_at_k` creates two local variables,
`average_precision_at_<k>/total` and `average_precision_at_<k>/max`, that
are used to compute the frequency. This frequency is ultimately returned as
`average_precision_at_<k>`: an idempotent operation that simply divides
`average_precision_at_<k>/total` by `average_precision_at_<k>/max`.
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the
`precision_at_<k>`. Internally, a `top_k` operation computes a `Tensor`
indicating the top `k` `predictions`. Set operations applied to `top_k` and
`labels` calculate the true positives and false positives weighted by
`weights`. Then `update_op` increments `true_positive_at_<k>` and
`false_positive_at_<k>` using these values.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: `int64` `Tensor` or `SparseTensor` with shape
[D1, ... DN, num_labels] or [D1, ... DN], where the latter implies
num_labels=1. N >= 1 and num_labels is the number of target classes for
the associated prediction. Commonly, N=1 and `labels` has shape
[batch_size, num_labels]. [D1, ... DN] must match `predictions`. Values
should be in range [0, num_classes), where num_classes is the last
dimension of `predictions`. Values outside this range are ignored.
predictions: Float `Tensor` with shape [D1, ... DN, num_classes] where
N >= 1. Commonly, N=1 and `predictions` has shape
[batch size, num_classes]. The final dimension contains the logit values
for each class. [D1, ... DN] must match `labels`.
k: Integer, k for @k metric. This will calculate an average precision for
range `[1,k]`, as documented above.
weights: `Tensor` whose rank is either 0, or n-1, where n is the rank of
`labels`. If the latter, it must be broadcastable to `labels` (i.e., all
dimensions must be either `1`, or the same as the corresponding `labels`
dimension).
metrics_collections: An optional list of collections that values should
be added to.
updates_collections: An optional list of collections that updates should
be added to.
name: Name of new update operation, and namespace for other dependent ops.
Returns:
mean_average_precision: Scalar `float64` `Tensor` with the mean average
precision values.
update: `Operation` that increments variables appropriately, and whose
value matches `metric`.
Raises:
ValueError: if k is invalid.
RuntimeError: If eager execution is enabled."
8929,precision_at_top_k,tensorflow/tensorflow/python/ops/metrics_impl.py,3427,function,"Computes precision@k of the predictions with respect to sparse labels.
Differs from `sparse_precision_at_k` in that predictions must be in the form
of top `k` class indices, whereas `sparse_precision_at_k` expects logits.
Refer to `sparse_precision_at_k` for more details.
Args:
labels: `int64` `Tensor` or `SparseTensor` with shape
[D1, ... DN, num_labels] or [D1, ... DN], where the latter implies
num_labels=1. N >= 1 and num_labels is the number of target classes for
the associated prediction. Commonly, N=1 and `labels` has shape
[batch_size, num_labels]. [D1, ... DN] must match `predictions`. Values
should be in range [0, num_classes), where num_classes is the last
dimension of `predictions`. Values outside this range are ignored.
predictions_idx: Integer `Tensor` with shape [D1, ... DN, k] where
N >= 1. Commonly, N=1 and predictions has shape [batch size, k].
The final dimension contains the top `k` predicted class indices.
[D1, ... DN] must match `labels`.
k: Integer, k for @k metric. Only used for the default op name.
class_id: Integer class ID for which we want binary metrics. This should be
in range [0, num_classes], where num_classes is the last dimension of
`predictions`. If `class_id` is outside this range, the method returns
NAN.
weights: `Tensor` whose rank is either 0, or n-1, where n is the rank of
`labels`. If the latter, it must be broadcastable to `labels` (i.e., all
dimensions must be either `1`, or the same as the corresponding `labels`
dimension).
metrics_collections: An optional list of collections that values should
be added to.
updates_collections: An optional list of collections that updates should
be added to.
name: Name of new update operation, and namespace for other dependent ops.
Returns:
precision: Scalar `float64` `Tensor` with the value of `true_positives`
divided by the sum of `true_positives` and `false_positives`.
update_op: `Operation` that increments `true_positives` and
`false_positives` variables appropriately, and whose value matches
`precision`.
Raises:
ValueError: If `weights` is not `None` and its shape doesn't match
`predictions`, or if either `metrics_collections` or `updates_collections`
are not a list or tuple.
RuntimeError: If eager execution is enabled."
8930,sparse_precision_at_k,tensorflow/tensorflow/python/ops/metrics_impl.py,3517,function,"Renamed to `precision_at_k`, please use that method instead."
8931,precision_at_k,tensorflow/tensorflow/python/ops/metrics_impl.py,3538,function,"Computes precision@k of the predictions with respect to sparse labels.
If `class_id` is specified, we calculate precision by considering only the
entries in the batch for which `class_id` is in the top-k highest
`predictions`, and computing the fraction of them for which `class_id` is
indeed a correct label.
If `class_id` is not specified, we'll calculate precision as how often on
average a class among the top-k classes with the highest predicted values
of a batch entry is correct and can be found in the label for that entry.
`precision_at_k` creates two local variables,
`true_positive_at_<k>` and `false_positive_at_<k>`, that are used to compute
the precision@k frequency. This frequency is ultimately returned as
`precision_at_<k>`: an idempotent operation that simply divides
`true_positive_at_<k>` by total (`true_positive_at_<k>` +
`false_positive_at_<k>`).
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the
`precision_at_<k>`. Internally, a `top_k` operation computes a `Tensor`
indicating the top `k` `predictions`. Set operations applied to `top_k` and
`labels` calculate the true positives and false positives weighted by
`weights`. Then `update_op` increments `true_positive_at_<k>` and
`false_positive_at_<k>` using these values.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
Args:
labels: `int64` `Tensor` or `SparseTensor` with shape
[D1, ... DN, num_labels] or [D1, ... DN], where the latter implies
num_labels=1. N >= 1 and num_labels is the number of target classes for
the associated prediction. Commonly, N=1 and `labels` has shape
[batch_size, num_labels]. [D1, ... DN] must match `predictions`. Values
should be in range [0, num_classes), where num_classes is the last
dimension of `predictions`. Values outside this range are ignored.
predictions: Float `Tensor` with shape [D1, ... DN, num_classes] where
N >= 1. Commonly, N=1 and predictions has shape [batch size, num_classes].
The final dimension contains the logit values for each class. [D1, ... DN]
must match `labels`.
k: Integer, k for @k metric.
class_id: Integer class ID for which we want binary metrics. This should be
in range [0, num_classes], where num_classes is the last dimension of
`predictions`. If `class_id` is outside this range, the method returns
NAN.
weights: `Tensor` whose rank is either 0, or n-1, where n is the rank of
`labels`. If the latter, it must be broadcastable to `labels` (i.e., all
dimensions must be either `1`, or the same as the corresponding `labels`
dimension).
metrics_collections: An optional list of collections that values should
be added to.
updates_collections: An optional list of collections that updates should
be added to.
name: Name of new update operation, and namespace for other dependent ops.
Returns:
precision: Scalar `float64` `Tensor` with the value of `true_positives`
divided by the sum of `true_positives` and `false_positives`.
update_op: `Operation` that increments `true_positives` and
`false_positives` variables appropriately, and whose value matches
`precision`.
Raises:
ValueError: If `weights` is not `None` and its shape doesn't match
`predictions`, or if either `metrics_collections` or `updates_collections`
are not a list or tuple.
RuntimeError: If eager execution is enabled."
8932,specificity_at_sensitivity,tensorflow/tensorflow/python/ops/metrics_impl.py,3632,function,"Computes the specificity at a given sensitivity.
The `specificity_at_sensitivity` function creates four local
variables, `true_positives`, `true_negatives`, `false_positives` and
`false_negatives` that are used to compute the specificity at the given
sensitivity value. The threshold for the given sensitivity value is computed
and used to evaluate the corresponding specificity.
For estimation of the metric over a stream of data, the function creates an
`update_op` operation that updates these variables and returns the
`specificity`. `update_op` increments the `true_positives`, `true_negatives`,
`false_positives` and `false_negatives` counts with the weight of each case
found in the `predictions` and `labels`.
If `weights` is `None`, weights default to 1. Use weights of 0 to mask values.
For additional information about specificity and sensitivity, see the
following: https://en.wikipedia.org/wiki/Sensitivity_and_specificity
Args:
labels: The ground truth values, a `Tensor` whose dimensions must match
`predictions`. Will be cast to `bool`.
predictions: A floating point `Tensor` of arbitrary shape and whose values
are in the range `[0, 1]`.
sensitivity: A scalar value in range `[0, 1]`.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `labels` dimension).
num_thresholds: The number of thresholds to use for matching the given
sensitivity.
metrics_collections: An optional list of collections that `specificity`
should be added to.
updates_collections: An optional list of collections that `update_op` should
be added to.
name: An optional variable_scope name.
Returns:
specificity: A scalar `Tensor` representing the specificity at the given
`sensitivity` value.
update_op: An operation that increments the `true_positives`,
`true_negatives`, `false_positives` and `false_negatives` variables
appropriately and whose value matches `specificity`.
Raises:
ValueError: If `predictions` and `labels` have mismatched shapes, if
`weights` is not `None` and its shape doesn't match `predictions`, or if
`sensitivity` is not between 0 and 1, or if either `metrics_collections`
or `updates_collections` are not a list or tuple.
RuntimeError: If eager execution is enabled."
8933,all_sum,tensorflow/tensorflow/python/ops/nccl_ops.py,33,function,"Returns a list of tensors with the all-reduce sum across `tensors`.
The computation is done with an all-reduce operation, so if only some of the
returned tensors are evaluated then the computation will hang.
Args:
tensors: The input tensors across which to sum; must be assigned
to GPU devices.
Returns:
List of tensors, each with the sum of the input tensors, where tensor i has
the same device as `tensors[i]`."
8934,all_prod,tensorflow/tensorflow/python/ops/nccl_ops.py,79,function,"Returns a list of tensors with the all-reduce product across `tensors`.
The computation is done with an all-reduce operation, so if only some of the
returned tensors are evaluated then the computation will hang.
Args:
tensors: The input tensors across which to multiply; must be assigned
to GPU devices.
Returns:
List of tensors, each with the product of the input tensors, where tensor i
has the same device as `tensors[i]`."
8935,all_min,tensorflow/tensorflow/python/ops/nccl_ops.py,96,function,"Returns a list of tensors with the all-reduce min across `tensors`.
The computation is done with an all-reduce operation, so if only some of the
returned tensors are evaluated then the computation will hang.
Args:
tensors: The input tensors across which to reduce; must be assigned
to GPU devices.
Returns:
List of tensors, each with the minimum of the input tensors, where tensor i
has the same device as `tensors[i]`."
8936,all_max,tensorflow/tensorflow/python/ops/nccl_ops.py,113,function,"Returns a list of tensors with the all-reduce max across `tensors`.
The computation is done with an all-reduce operation, so if only some of the
returned tensors are evaluated then the computation will hang.
Args:
tensors: The input tensors across which to reduce; must be assigned
to GPU devices.
Returns:
List of tensors, each with the maximum of the input tensors, where tensor i
has the same device as `tensors[i]`."
8937,reduce_sum,tensorflow/tensorflow/python/ops/nccl_ops.py,130,function,"Returns a tensor with the reduce sum across `tensors`.
The computation is done with a reduce operation, so only one tensor is
returned.
Args:
tensors: The input tensors across which to sum; must be assigned
to GPU devices.
Returns:
A tensor containing the sum of the input tensors.
Raises:
LookupError: If context is not currently using a GPU device."
8938,broadcast,tensorflow/tensorflow/python/ops/nccl_ops.py,173,function,"Returns a tensor that can be efficiently transferred to other devices.
Args:
tensor: The tensor to send; must be assigned to a GPU device.
Returns:
A tensor with the value of `src_tensor`, which can be used as input to
ops on other GPU devices."
8939,log_poisson_loss,tensorflow/tensorflow/python/ops/nn_impl.py,49,function,"Computes log Poisson loss given `log_input`.
Gives the log-likelihood loss between the prediction and the target under the
assumption that the target has a Poisson distribution.
Caveat: By default, this is not the exact loss, but the loss minus a
constant term [log(z!)]. That has no effect for optimization, but
does not play well with relative loss comparisons. To compute an
approximation of the log factorial term, specify
compute_full_loss=True to enable Stirling's Approximation.
For brevity, let `c = log(x) = log_input`, `z = targets`. The log Poisson
loss is
-log(exp(-x) * (x^z) / z!)
= -log(exp(-x) * (x^z)) + log(z!)
~ -log(exp(-x)) - log(x^z) [+ z * log(z) - z + 0.5 * log(2 * pi * z)]
[ Note the second term is the Stirling's Approximation for log(z!).
It is invariant to x and does not affect optimization, though
important for correct relative loss comparisons. It is only
computed when compute_full_loss == True. ]
= x - z * log(x) [+ z * log(z) - z + 0.5 * log(2 * pi * z)]
= exp(c) - z * c [+ z * log(z) - z + 0.5 * log(2 * pi * z)]
Args:
targets: A `Tensor` of the same type and shape as `log_input`.
log_input: A `Tensor` of type `float32` or `float64`.
compute_full_loss: whether to compute the full loss. If false, a constant
term is dropped in favor of more efficient optimization.
name: A name for the operation (optional).
Returns:
A `Tensor` of the same shape as `log_input` with the componentwise
logistic losses.
Raises:
ValueError: If `log_input` and `targets` do not have the same shape."
8940,sigmoid_cross_entropy_with_logits,tensorflow/tensorflow/python/ops/nn_impl.py,115,function,"Computes sigmoid cross entropy given `logits`.
Measures the probability error in discrete classification tasks in which each
class is independent and not mutually exclusive. For instance, one could
perform multilabel classification where a picture can contain both an elephant
and a dog at the same time.
For brevity, let `x = logits`, `z = labels`. The logistic loss is
z * -log(sigmoid(x)) + (1 - z) * -log(1 - sigmoid(x))
= z * -log(1 / (1 + exp(-x))) + (1 - z) * -log(exp(-x) / (1 + exp(-x)))
= z * log(1 + exp(-x)) + (1 - z) * (-log(exp(-x)) + log(1 + exp(-x)))
= z * log(1 + exp(-x)) + (1 - z) * (x + log(1 + exp(-x))
= (1 - z) * x + log(1 + exp(-x))
= x - x * z + log(1 + exp(-x))
For x < 0, to avoid overflow in exp(-x), we reformulate the above
x - x * z + log(1 + exp(-x))
= log(exp(x)) - x * z + log(1 + exp(-x))
= - x * z + log(1 + exp(x))
Hence, to ensure stability and avoid overflow, the implementation uses this
equivalent formulation
max(x, 0) - x * z + log(1 + exp(-abs(x)))
`logits` and `labels` must have the same type and shape.
Args:
_sentinel: Used to prevent positional parameters. Internal, do not use.
labels: A `Tensor` of the same type and shape as `logits`.
logits: A `Tensor` of type `float32` or `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` of the same shape as `logits` with the componentwise
logistic losses.
Raises:
ValueError: If `logits` and `labels` do not have the same shape."
8941,sigmoid_cross_entropy_with_logits_v2,tensorflow/tensorflow/python/ops/nn_impl.py,198,function,"Computes sigmoid cross entropy given `logits`.
Measures the probability error in discrete classification tasks in which each
class is independent and not mutually exclusive. For instance, one could
perform multilabel classification where a picture can contain both an elephant
and a dog at the same time.
For brevity, let `x = logits`, `z = labels`. The logistic loss is
z * -log(sigmoid(x)) + (1 - z) * -log(1 - sigmoid(x))
= z * -log(1 / (1 + exp(-x))) + (1 - z) * -log(exp(-x) / (1 + exp(-x)))
= z * log(1 + exp(-x)) + (1 - z) * (-log(exp(-x)) + log(1 + exp(-x)))
= z * log(1 + exp(-x)) + (1 - z) * (x + log(1 + exp(-x))
= (1 - z) * x + log(1 + exp(-x))
= x - x * z + log(1 + exp(-x))
For x < 0, to avoid overflow in exp(-x), we reformulate the above
x - x * z + log(1 + exp(-x))
= log(exp(x)) - x * z + log(1 + exp(-x))
= - x * z + log(1 + exp(x))
Hence, to ensure stability and avoid overflow, the implementation uses this
equivalent formulation
max(x, 0) - x * z + log(1 + exp(-abs(x)))
`logits` and `labels` must have the same type and shape.
Args:
labels: A `Tensor` of the same type and shape as `logits`.
logits: A `Tensor` of type `float32` or `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` of the same shape as `logits` with the componentwise
logistic losses.
Raises:
ValueError: If `logits` and `labels` do not have the same shape."
8942,weighted_cross_entropy_with_logits_v2,tensorflow/tensorflow/python/ops/nn_impl.py,249,function,"Computes a weighted cross entropy.
This is like `sigmoid_cross_entropy_with_logits()` except that `pos_weight`,
allows one to trade off recall and precision by up- or down-weighting the
cost of a positive error relative to a negative error.
The usual cross-entropy cost is defined as:
labels * -log(sigmoid(logits)) +
(1 - labels) * -log(1 - sigmoid(logits))
A value `pos_weight > 1` decreases the false negative count, hence increasing
the recall.
Conversely setting `pos_weight < 1` decreases the false positive count and
increases the precision.
This can be seen from the fact that `pos_weight` is introduced as a
multiplicative coefficient for the positive labels term
in the loss expression:
labels * -log(sigmoid(logits)) * pos_weight +
(1 - labels) * -log(1 - sigmoid(logits))
For brevity, let `x = logits`, `z = labels`, `q = pos_weight`.
The loss is:
qz * -log(sigmoid(x)) + (1 - z) * -log(1 - sigmoid(x))
= qz * -log(1 / (1 + exp(-x))) + (1 - z) * -log(exp(-x) / (1 + exp(-x)))
= qz * log(1 + exp(-x)) + (1 - z) * (-log(exp(-x)) + log(1 + exp(-x)))
= qz * log(1 + exp(-x)) + (1 - z) * (x + log(1 + exp(-x))
= (1 - z) * x + (qz + 1 - z) * log(1 + exp(-x))
= (1 - z) * x + (1 + (q - 1) * z) * log(1 + exp(-x))
Setting `l = (1 + (q - 1) * z)`, to ensure stability and avoid overflow,
the implementation uses
(1 - z) * x + l * (log(1 + exp(-abs(x))) + max(-x, 0))
`logits` and `labels` must have the same type and shape.
Args:
labels: A `Tensor` of the same type and shape as `logits`.
logits: A `Tensor` of type `float32` or `float64`.
pos_weight: A coefficient to use on the positive examples.
name: A name for the operation (optional).
Returns:
A `Tensor` of the same shape as `logits` with the componentwise
weighted logistic losses.
Raises:
ValueError: If `logits` and `labels` do not have the same shape."
8943,weighted_cross_entropy_with_logits,tensorflow/tensorflow/python/ops/nn_impl.py,329,function,"Computes a weighted cross entropy.
This is like `sigmoid_cross_entropy_with_logits()` except that `pos_weight`,
allows one to trade off recall and precision by up- or down-weighting the
cost of a positive error relative to a negative error.
The usual cross-entropy cost is defined as:
labels * -log(sigmoid(logits)) +
(1 - labels) * -log(1 - sigmoid(logits))
A value `pos_weight > 1` decreases the false negative count, hence increasing
the recall.
Conversely setting `pos_weight < 1` decreases the false positive count and
increases the precision.
This can be seen from the fact that `pos_weight` is introduced as a
multiplicative coefficient for the positive labels term
in the loss expression:
labels * -log(sigmoid(logits)) * pos_weight +
(1 - labels) * -log(1 - sigmoid(logits))
For brevity, let `x = logits`, `z = labels`, `q = pos_weight`.
The loss is:
qz * -log(sigmoid(x)) + (1 - z) * -log(1 - sigmoid(x))
= qz * -log(1 / (1 + exp(-x))) + (1 - z) * -log(exp(-x) / (1 + exp(-x)))
= qz * log(1 + exp(-x)) + (1 - z) * (-log(exp(-x)) + log(1 + exp(-x)))
= qz * log(1 + exp(-x)) + (1 - z) * (x + log(1 + exp(-x))
= (1 - z) * x + (qz + 1 - z) * log(1 + exp(-x))
= (1 - z) * x + (1 + (q - 1) * z) * log(1 + exp(-x))
Setting `l = (1 + (q - 1) * z)`, to ensure stability and avoid overflow,
the implementation uses
(1 - z) * x + l * (log(1 + exp(-abs(x))) + max(-x, 0))
`logits` and `labels` must have the same type and shape.
Args:
labels: A `Tensor` of the same type and shape as `logits`.
logits: A `Tensor` of type `float32` or `float64`.
pos_weight: A coefficient to use on the positive examples.
name: A name for the operation (optional).
targets: Deprecated alias for labels.
Returns:
A `Tensor` of the same shape as `logits` with the componentwise
weighted logistic losses.
Raises:
ValueError: If `logits` and `labels` do not have the same shape."
8944,compute_average_loss,tensorflow/tensorflow/python/ops/nn_impl.py,393,function,"Scales per-example losses with sample_weights and computes their average.
Usage with distribution strategy and custom training loop:
```python
with strategy.scope():
def compute_loss(labels, predictions, sample_weight=None):
# If you are using a `Loss` class instead, set reduction to `NONE` so that
# we can do the reduction afterwards and divide by global batch size.
per_example_loss = tf.keras.losses.sparse_categorical_crossentropy(
labels, predictions)
# Compute loss that is scaled by sample_weight and by global batch size.
return tf.nn.compute_average_loss(
per_example_loss,
sample_weight=sample_weight,
global_batch_size=GLOBAL_BATCH_SIZE)
```
Args:
per_example_loss: Per-example loss.
sample_weight: Optional weighting for each example.
global_batch_size: Optional global batch size value. Defaults to (size of
first dimension of `losses`) * (number of replicas).
Returns:
Scalar loss value."
8945,scale_regularization_loss,tensorflow/tensorflow/python/ops/nn_impl.py,451,function,"Scales the sum of the given regularization losses by number of replicas.
Usage with distribution strategy and custom training loop:
```python
with strategy.scope():
def compute_loss(self, label, predictions):
per_example_loss = tf.keras.losses.sparse_categorical_crossentropy(
labels, predictions)
# Compute loss that is scaled by sample_weight and by global batch size.
loss = tf.nn.compute_average_loss(
per_example_loss,
sample_weight=sample_weight,
global_batch_size=GLOBAL_BATCH_SIZE)
# Add scaled regularization losses.
loss += tf.nn.scale_regularization_loss(tf.nn.l2_loss(weights))
return loss
```
Args:
regularization_loss: Regularization loss.
Returns:
Scalar loss value."
8946,relu_layer,tensorflow/tensorflow/python/ops/nn_impl.py,490,function,"Computes Relu(x * weight + biases).
Args:
x: a 2D tensor. Dimensions typically: batch, in_units
weights: a 2D tensor. Dimensions typically: in_units, out_units
biases: a 1D tensor. Dimensions: out_units
name: A name for the operation (optional). If not specified
""nn_relu_layer"" is used.
Returns:
A 2-D Tensor computing relu(matmul(x, weights) + biases).
Dimensions typically: batch, out_units."
8947,swish,tensorflow/tensorflow/python/ops/nn_impl.py,515,function,"Computes the SiLU or Swish activation function: `x * sigmoid(x)`.
The SiLU activation function was introduced in ""Gaussian Error Linear Units
(GELUs)"" [Hendrycks et al. 2016](https://arxiv.org/abs/1606.08415) and
""Sigmoid-Weighted Linear Units for Neural Network Function Approximation in
Reinforcement Learning""
[Elfwing et al. 2017](https://arxiv.org/abs/1702.03118) and was independently
discovered (and called swish) in ""Searching for Activation Functions""
[Ramachandran et al. 2017](https://arxiv.org/abs/1710.05941)
Args:
features: A `Tensor` representing preactivation values.
name: A name for the operation (optional).
Returns:
The activation value."
8948,normalize,tensorflow/tensorflow/python/ops/nn_impl.py,557,function,"Normalizes `tensor` along dimension `axis` using specified norm.
This uses `tf.linalg.norm` to compute the norm along `axis`.
This function can compute several different vector norms (the 1-norm, the
Euclidean or 2-norm, the inf-norm, and in general the p-norm for p > 0) and
matrix norms (Frobenius, 1-norm, 2-norm and inf-norm).
Args:
tensor: `Tensor` of types `float32`, `float64`, `complex64`, `complex128`
ord: Order of the norm. Supported values are `'fro'`, `'euclidean'`, `1`,
`2`, `np.inf` and any positive real number yielding the corresponding
p-norm. Default is `'euclidean'` which is equivalent to Frobenius norm if
`tensor` is a matrix and equivalent to 2-norm for vectors.
Some restrictions apply: a) The Frobenius norm `'fro'` is not defined for
vectors, b) If axis is a 2-tuple (matrix norm), only `'euclidean'`,
'`fro'`, `1`, `2`, `np.inf` are supported. See the description of `axis`
on how to compute norms for a batch of vectors or matrices stored in a
tensor.
axis: If `axis` is `None` (the default), the input is considered a vector
and a single vector norm is computed over the entire set of values in the
tensor, i.e. `norm(tensor, ord=ord)` is equivalent to
`norm(reshape(tensor, [-1]), ord=ord)`. If `axis` is a Python integer, the
input is considered a batch of vectors, and `axis` determines the axis in
`tensor` over which to compute vector norms. If `axis` is a 2-tuple of
Python integers it is considered a batch of matrices and `axis` determines
the axes in `tensor` over which to compute a matrix norm.
Negative indices are supported. Example: If you are passing a tensor that
can be either a matrix or a batch of matrices at runtime, pass
`axis=[-2,-1]` instead of `axis=None` to make sure that matrix norms are
computed.
name: The name of the op.
Returns:
normalized: A normalized `Tensor` with the same shape as `tensor`.
norm: The computed norms with the same shape and dtype `tensor` but the
final axis is 1 instead. Same as running
`tf.cast(tf.linalg.norm(tensor, ord, axis keepdims=True), tensor.dtype)`.
Raises:
ValueError: If `ord` or `axis` is invalid."
8949,l2_normalize,tensorflow/tensorflow/python/ops/nn_impl.py,611,function,"Normalizes along dimension `axis` using an L2 norm.
For a 1-D tensor with `axis = 0`, computes
output = x / sqrt(max(sum(x**2), epsilon))
For `x` with more dimensions, independently normalizes each 1-D slice along
dimension `axis`.
Args:
x: A `Tensor`.
axis: Dimension along which to normalize. A scalar or a vector of
integers.
epsilon: A lower bound value for the norm. Will use `sqrt(epsilon)` as the
divisor if `norm < sqrt(epsilon)`.
name: A name for this operation (optional).
dim: Deprecated alias for axis.
Returns:
A `Tensor` with the same shape as `x`."
8950,l2_normalize_v2,tensorflow/tensorflow/python/ops/nn_impl.py,639,function,"Normalizes along dimension `axis` using an L2 norm.
For a 1-D tensor with `axis = 0`, computes
output = x / sqrt(max(sum(x**2), epsilon))
For `x` with more dimensions, independently normalizes each 1-D slice along
dimension `axis`.
Args:
x: A `Tensor`.
axis: Dimension along which to normalize. A scalar or a vector of
integers.
epsilon: A lower bound value for the norm. Will use `sqrt(epsilon)` as the
divisor if `norm < sqrt(epsilon)`.
name: A name for this operation (optional).
Returns:
A `Tensor` with the same shape as `x`."
8951,zero_fraction,tensorflow/tensorflow/python/ops/nn_impl.py,699,function,"Returns the fraction of zeros in `value`.
If `value` is empty, the result is `nan`.
This is useful in summaries to measure and report sparsity. For example,
```python
z = tf.nn.relu(...)
summ = tf.compat.v1.summary.scalar('sparsity', tf.nn.zero_fraction(z))
```
Args:
value: A tensor of numeric type.
name: A name for the operation (optional).
Returns:
The fraction of zeros in `value`, with type `float32`."
8952,depthwise_conv2d,tensorflow/tensorflow/python/ops/nn_impl.py,742,function,"Depthwise 2-D convolution.
Given a 4D input tensor ('NHWC' or 'NCHW' data formats)
and a filter tensor of shape
`[filter_height, filter_width, in_channels, channel_multiplier]`
containing `in_channels` convolutional filters of depth 1, `depthwise_conv2d`
applies a different filter to each input channel (expanding from 1 channel
to `channel_multiplier` channels for each), then concatenates the results
together. The output has `in_channels * channel_multiplier` channels.
In detail, with the default NHWC format,
output[b, i, j, k * channel_multiplier + q] = sum_{di, dj}
filter[di, dj, k, q] * input[b, strides[1] * i + rate[0] * di,
strides[2] * j + rate[1] * dj, k]
Must have `strides[0] = strides[3] = 1`. For the most common case of the
same horizontal and vertical strides, `strides = [1, stride, stride, 1]`.
If any value in `rate` is greater than 1, we perform atrous depthwise
convolution, in which case all values in the `strides` tensor must be equal
to 1.
Usage Example:
>>> x = np.array([
... [1., 2.],
... [3., 4.],
... [5., 6.]
... ], dtype=np.float32).reshape((1, 3, 2, 1))
>>> kernel = np.array([
... [1., 2.],
... [3., 4]
... ], dtype=np.float32).reshape((2, 1, 1, 2))
>>> tf.compat.v1.nn.depthwise_conv2d(x, kernel, strides=[1, 1, 1, 1],
... padding='VALID').numpy()
array([[[[10., 14.],
[14., 20.]],
[[18., 26.],
[22., 32.]]]], dtype=float32)
>>> tf.compat.v1.nn.depthwise_conv2d(x, kernel, strides=[1, 1, 1, 1],
... padding=[[0, 0], [1, 0], [1, 0], [0, 0]]
... ).numpy()
array([[[[ 0., 0.],
[ 3., 4.],
[ 6., 8.]],
[[ 0., 0.],
[10., 14.],
[14., 20.]],
[[ 0., 0.],
[18., 26.],
[22., 32.]]]], dtype=float32)
Args:
input: 4-D with shape according to `data_format`.
filter: 4-D with shape
`[filter_height, filter_width, in_channels, channel_multiplier]`.
strides: 1-D of size 4. The stride of the sliding window for each
dimension of `input`.
padding: Controls how to pad the image before applying the convolution. Can
be the string `""SAME""` or `""VALID""` indicating the type of padding
algorithm to use, or a list indicating the explicit paddings at the start
and end of each dimension. When explicit padding is used and data_format
is `""NHWC""`, this should be in the form `[[0, 0], [pad_top, pad_bottom],
[pad_left, pad_right], [0, 0]]`. When explicit padding used and
data_format is `""NCHW""`, this should be in the form `[[0, 0], [0, 0],
[pad_top, pad_bottom], [pad_left, pad_right]]`.
rate: 1-D of size 2. The dilation rate in which we sample input values
across the `height` and `width` dimensions in atrous convolution. If it is
greater than 1, then all values of strides must be 1.
name: A name for this operation (optional).
data_format: The data format for input. Either ""NHWC"" (default) or ""NCHW"".
dilations: Alias of rate.
Returns:
A 4-D `Tensor` with shape according to `data_format`. E.g., for
""NHWC"" format, shape is
`[batch, out_height, out_width, in_channels * channel_multiplier].`"
8953,depthwise_conv2d_v2,tensorflow/tensorflow/python/ops/nn_impl.py,871,function,"Depthwise 2-D convolution.
Given a 4D input tensor ('NHWC' or 'NCHW' data formats)
and a filter tensor of shape
`[filter_height, filter_width, in_channels, channel_multiplier]`
containing `in_channels` convolutional filters of depth 1, `depthwise_conv2d`
applies a different filter to each input channel (expanding from 1 channel
to `channel_multiplier` channels for each), then concatenates the results
together. The output has `in_channels * channel_multiplier` channels.
In detail, with the default NHWC format,
output[b, i, j, k * channel_multiplier + q] = sum_{di, dj}
filter[di, dj, k, q] * input[b, strides[1] * i + rate[0] * di,
strides[2] * j + rate[1] * dj, k]
Must have `strides[0] = strides[3] = 1`. For the most common case of the
same horizontal and vertical strides, `strides = [1, stride, stride, 1]`.
If any value in `rate` is greater than 1, we perform atrous depthwise
convolution, in which case all values in the `strides` tensor must be equal
to 1.
Usage Example:
>>> x = np.array([
... [1., 2.],
... [3., 4.],
... [5., 6.]
... ], dtype=np.float32).reshape((1, 3, 2, 1))
>>> kernel = np.array([
... [1., 2.],
... [3., 4]
... ], dtype=np.float32).reshape((2, 1, 1, 2))
>>> tf.nn.depthwise_conv2d(x, kernel, strides=[1, 1, 1, 1],
... padding='VALID').numpy()
array([[[[10., 14.],
[14., 20.]],
[[18., 26.],
[22., 32.]]]], dtype=float32)
>>> tf.nn.depthwise_conv2d(x, kernel, strides=[1, 1, 1, 1],
... padding=[[0, 0], [1, 0], [1, 0], [0, 0]]).numpy()
array([[[[ 0., 0.],
[ 3., 4.],
[ 6., 8.]],
[[ 0., 0.],
[10., 14.],
[14., 20.]],
[[ 0., 0.],
[18., 26.],
[22., 32.]]]], dtype=float32)
Args:
input: 4-D with shape according to `data_format`.
filter: 4-D with shape
`[filter_height, filter_width, in_channels, channel_multiplier]`.
strides: 1-D of size 4. The stride of the sliding window for each
dimension of `input`.
padding: Controls how to pad the image before applying the convolution. Can
be the string `""SAME""` or `""VALID""` indicating the type of padding
algorithm to use, or a list indicating the explicit paddings at the start
and end of each dimension. When explicit padding is used and data_format
is `""NHWC""`, this should be in the form `[[0, 0], [pad_top, pad_bottom],
[pad_left, pad_right], [0, 0]]`. When explicit padding used and
data_format is `""NCHW""`, this should be in the form `[[0, 0], [0, 0],
[pad_top, pad_bottom], [pad_left, pad_right]]`.
data_format: The data format for input. Either ""NHWC"" (default) or ""NCHW"".
dilations: 1-D of size 2. The dilation rate in which we sample input values
across the `height` and `width` dimensions in atrous convolution. If it is
greater than 1, then all values of strides must be 1.
name: A name for this operation (optional).
Returns:
A 4-D `Tensor` with shape according to `data_format`. E.g., for
""NHWC"" format, shape is
`[batch, out_height, out_width, in_channels * channel_multiplier].`"
8954,separable_conv2d,tensorflow/tensorflow/python/ops/nn_impl.py,969,function,"2-D convolution with separable filters.
Performs a depthwise convolution that acts separately on channels followed by
a pointwise convolution that mixes channels. Note that this is separability
between dimensions `[1, 2]` and `3`, not spatial separability between
dimensions `1` and `2`.
In detail, with the default NHWC format,
output[b, i, j, k] = sum_{di, dj, q, r}
input[b, strides[1] * i + di, strides[2] * j + dj, q] *
depthwise_filter[di, dj, q, r] *
pointwise_filter[0, 0, q * channel_multiplier + r, k]
`strides` controls the strides for the depthwise convolution only, since
the pointwise convolution has implicit strides of `[1, 1, 1, 1]`. Must have
`strides[0] = strides[3] = 1`. For the most common case of the same
horizontal and vertical strides, `strides = [1, stride, stride, 1]`.
If any value in `rate` is greater than 1, we perform atrous depthwise
convolution, in which case all values in the `strides` tensor must be equal
to 1.
Args:
input: 4-D `Tensor` with shape according to `data_format`.
depthwise_filter: 4-D `Tensor` with shape
`[filter_height, filter_width, in_channels, channel_multiplier]`.
Contains `in_channels` convolutional filters of depth 1.
pointwise_filter: 4-D `Tensor` with shape
`[1, 1, channel_multiplier * in_channels, out_channels]`. Pointwise
filter to mix channels after `depthwise_filter` has convolved spatially.
strides: 1-D of size 4. The strides for the depthwise convolution for
each dimension of `input`.
padding: Controls how to pad the image before applying the depthwise
convolution. Can be the string `""SAME""` or `""VALID""` indicating the type
of padding algorithm to use, or a Python list indicating the explicit
paddings at the start and end of each dimension. When explicit padding is
used and data_format is `""NHWC""`, this should be in the form `[[0, 0],
[pad_top, pad_bottom], [pad_left, pad_right], [0, 0]]`. When explicit
padding used and data_format is `""NCHW""`, this should be in the form
`[[0, 0], [0, 0], [pad_top, pad_bottom], [pad_left, pad_right]]`.
rate: 1-D of size 2. The dilation rate in which we sample input values
across the `height` and `width` dimensions in atrous convolution. If it is
greater than 1, then all values of strides must be 1.
name: A name for this operation (optional).
data_format: The data format for input. Either ""NHWC"" (default) or ""NCHW"".
dilations: Alias of rate.
Returns:
A 4-D `Tensor` with shape according to 'data_format'. For
example, with data_format=""NHWC"", shape is [batch, out_height,
out_width, out_channels]."
8955,separable_conv2d_v2,tensorflow/tensorflow/python/ops/nn_impl.py,1077,function,"2-D convolution with separable filters.
Performs a depthwise convolution that acts separately on channels followed by
a pointwise convolution that mixes channels. Note that this is separability
between dimensions `[1, 2]` and `3`, not spatial separability between
dimensions `1` and `2`.
In detail, with the default NHWC format,
output[b, i, j, k] = sum_{di, dj, q, r}
input[b, strides[1] * i + di, strides[2] * j + dj, q] *
depthwise_filter[di, dj, q, r] *
pointwise_filter[0, 0, q * channel_multiplier + r, k]
`strides` controls the strides for the depthwise convolution only, since
the pointwise convolution has implicit strides of `[1, 1, 1, 1]`. Must have
`strides[0] = strides[3] = 1`. For the most common case of the same
horizontal and vertical strides, `strides = [1, stride, stride, 1]`.
If any value in `rate` is greater than 1, we perform atrous depthwise
convolution, in which case all values in the `strides` tensor must be equal
to 1.
Args:
input: 4-D `Tensor` with shape according to `data_format`.
depthwise_filter: 4-D `Tensor` with shape `[filter_height, filter_width,
in_channels, channel_multiplier]`. Contains `in_channels` convolutional
filters of depth 1.
pointwise_filter: 4-D `Tensor` with shape `[1, 1, channel_multiplier *
in_channels, out_channels]`. Pointwise filter to mix channels after
`depthwise_filter` has convolved spatially.
strides: 1-D of size 4. The strides for the depthwise convolution for each
dimension of `input`.
padding: Controls how to pad the image before applying the depthwise
convolution. Can be the string `""SAME""` or `""VALID""` indicating the type
of padding algorithm to use, or a Python list indicating the explicit
paddings at the start and end of each dimension. When explicit padding is
used and data_format is `""NHWC""`, this should be in the form `[[0, 0],
[pad_top, pad_bottom], [pad_left, pad_right], [0, 0]]`. When explicit
padding used and data_format is `""NCHW""`, this should be in the form
`[[0, 0], [0, 0], [pad_top, pad_bottom], [pad_left, pad_right]]`.
data_format: The data format for input. Either ""NHWC"" (default) or ""NCHW"".
dilations: 1-D of size 2. The dilation rate in which we sample input values
across the `height` and `width` dimensions in atrous convolution. If it is
greater than 1, then all values of strides must be 1.
name: A name for this operation (optional).
Returns:
A 4-D `Tensor` with shape according to 'data_format'. For
example, with data_format=""NHWC"", shape is [batch, out_height,
out_width, out_channels]."
8956,sufficient_statistics,tensorflow/tensorflow/python/ops/nn_impl.py,1153,function,"Calculate the sufficient statistics for the mean and variance of `x`.
These sufficient statistics are computed using the one pass algorithm on
an input that's optionally shifted. See:
https://en.wikipedia.org/wiki/Algorithms_for_calculating_variance#Computing_shifted_data
For example:
>>> t = [[1, 2, 3], [4, 5, 6]]
>>> sufficient_statistics(t, [1])
(<tf.Tensor: shape=(), dtype=int32, numpy=3>, <tf.Tensor: shape=(2,),
dtype=int32, numpy=array([ 6, 15], dtype=int32)>, <tf.Tensor: shape=(2,),
dtype=int32, numpy=array([14, 77], dtype=int32)>, None)
>>> sufficient_statistics(t, [-1])
(<tf.Tensor: shape=(), dtype=int32, numpy=3>, <tf.Tensor: shape=(2,),
dtype=int32, numpy=array([ 6, 15], dtype=int32)>, <tf.Tensor: shape=(2,),
dtype=int32, numpy=array([14, 77], dtype=int32)>, None)
Args:
x: A `Tensor`.
axes: Array of ints. Axes along which to compute mean and variance. As in
Python, the axes can also be negative numbers. A negative axis is
interpreted as counting from the end of the rank, i.e., axis +
rank(values)-th dimension.
shift: A `Tensor` containing the value by which to shift the data for
numerical stability, or `None` if no shift is to be performed. A shift
close to the true mean provides the most numerically stable results.
keep_dims: produce statistics with the same dimensionality as the input.
name: Name used to scope the operations that compute the sufficient stats.
keepdims: Alias for keep_dims.
Returns:
Four `Tensor` objects of the same type as `x`:
* the count (number of elements to average over).
* the (possibly shifted) sum of the elements in the array.
* the (possibly shifted) sum of squares of the elements in the array.
* the shift by which the mean must be corrected or None if `shift` is None."
8957,sufficient_statistics_v2,tensorflow/tensorflow/python/ops/nn_impl.py,1228,function,"Calculate the sufficient statistics for the mean and variance of `x`.
These sufficient statistics are computed using the one pass algorithm on
an input that's optionally shifted. See:
https://en.wikipedia.org/wiki/Algorithms_for_calculating_variance#Computing_shifted_data
Args:
x: A `Tensor`.
axes: Array of ints. Axes along which to compute mean and variance.
shift: A `Tensor` containing the value by which to shift the data for
numerical stability, or `None` if no shift is to be performed. A shift
close to the true mean provides the most numerically stable results.
keepdims: produce statistics with the same dimensionality as the input.
name: Name used to scope the operations that compute the sufficient stats.
Returns:
Four `Tensor` objects of the same type as `x`:
* the count (number of elements to average over).
* the (possibly shifted) sum of the elements in the array.
* the (possibly shifted) sum of squares of the elements in the array.
* the shift by which the mean must be corrected or None if `shift` is None."
8958,normalize_moments,tensorflow/tensorflow/python/ops/nn_impl.py,1258,function,"Calculate the mean and variance of based on the sufficient statistics.
Args:
counts: A `Tensor` containing the total count of the data (one value).
mean_ss: A `Tensor` containing the mean sufficient statistics: the (possibly
shifted) sum of the elements to average over.
variance_ss: A `Tensor` containing the variance sufficient statistics: the
(possibly shifted) squared sum of the data to compute the variance over.
shift: A `Tensor` containing the value by which the data is shifted for
numerical stability, or `None` if no shift was performed.
name: Name used to scope the operations that compute the moments.
Returns:
Two `Tensor` objects: `mean` and `variance`."
8959,moments,tensorflow/tensorflow/python/ops/nn_impl.py,1291,function,"Calculate the mean and variance of `x`.
The mean and variance are calculated by aggregating the contents of `x`
across `axes`. If `x` is 1-D and `axes = [0]` this is just the mean
and variance of a vector.
Note: shift is currently not used; the true mean is computed and used.
When using these moments for batch normalization (see
`tf.nn.batch_normalization`):
* for so-called ""global normalization"", used with convolutional filters with
shape `[batch, height, width, depth]`, pass `axes=[0, 1, 2]`.
* for simple batch normalization pass `axes=[0]` (batch only).
Args:
x: A `Tensor`.
axes: Array of ints. Axes along which to compute mean and
variance.
shift: Not used in the current implementation
name: Name used to scope the operations that compute the moments.
keep_dims: produce moments with the same dimensionality as the input.
keepdims: Alias to keep_dims.
Returns:
Two `Tensor` objects: `mean` and `variance`."
8960,moments_v2,tensorflow/tensorflow/python/ops/nn_impl.py,1357,function,"Calculates the mean and variance of `x`.
The mean and variance are calculated by aggregating the contents of `x`
across `axes`. If `x` is 1-D and `axes = [0]` this is just the mean
and variance of a vector.
Note: shift is currently not used; the true mean is computed and used.
When using these moments for batch normalization (see
`tf.nn.batch_normalization`):
* for so-called ""global normalization"", used with convolutional filters with
shape `[batch, height, width, depth]`, pass `axes=[0, 1, 2]`.
* for simple batch normalization pass `axes=[0]` (batch only).
Args:
x: A `Tensor`.
axes: Array of ints. Axes along which to compute mean and
variance.
shift: Not used in the current implementation.
keepdims: produce moments with the same dimensionality as the input.
name: Name used to scope the operations that compute the moments.
Returns:
Two `Tensor` objects: `mean` and `variance`."
8961,weighted_moments,tensorflow/tensorflow/python/ops/nn_impl.py,1394,function,"Returns the frequency-weighted mean and variance of `x`.
Args:
x: A tensor.
axes: 1-d tensor of int32 values; these are the axes along which
to compute mean and variance.
frequency_weights: A tensor of positive weights which can be
broadcast with x.
name: Name used to scope the operation.
keep_dims: Produce moments with the same dimensionality as the input.
keepdims: Alias of keep_dims.
Returns:
Two tensors: `weighted_mean` and `weighted_variance`."
8962,weighted_moments_v2,tensorflow/tensorflow/python/ops/nn_impl.py,1473,function,"Returns the frequency-weighted mean and variance of `x`.
Args:
x: A tensor.
axes: 1-d tensor of int32 values; these are the axes along which
to compute mean and variance.
frequency_weights: A tensor of positive weights which can be
broadcast with x.
keepdims: Produce moments with the same dimensionality as the input.
name: Name used to scope the operation.
Returns:
Two tensors: `weighted_mean` and `weighted_variance`."
8963,batch_normalization,tensorflow/tensorflow/python/ops/nn_impl.py,1498,function,"Batch normalization.
Normalizes a tensor by `mean` and `variance`, and applies (optionally) a
`scale` \\(\gamma\\) to it, as well as an `offset` \\(\beta\\):
\\(\frac{\gamma(x-\mu)}{\sigma}+\beta\\)
`mean`, `variance`, `offset` and `scale` are all expected to be of one of two
shapes:
* In all generality, they can have the same number of dimensions as the
input `x`, with identical sizes as `x` for the dimensions that are not
normalized over (the 'depth' dimension(s)), and dimension 1 for the
others which are being normalized over.
`mean` and `variance` in this case would typically be the outputs of
`tf.nn.moments(..., keepdims=True)` during training, or running averages
thereof during inference.
* In the common case where the 'depth' dimension is the last dimension in
the input tensor `x`, they may be one dimensional tensors of the same
size as the 'depth' dimension.
This is the case for example for the common `[batch, depth]` layout of
fully-connected layers, and `[batch, height, width, depth]` for
convolutions.
`mean` and `variance` in this case would typically be the outputs of
`tf.nn.moments(..., keepdims=False)` during training, or running averages
thereof during inference.
See equation 11 in Algorithm 2 of source:
[Batch Normalization: Accelerating Deep Network Training by
Reducing Internal Covariate Shift; S. Ioffe, C. Szegedy]
(http://arxiv.org/abs/1502.03167).
Args:
x: Input `Tensor` of arbitrary dimensionality.
mean: A mean `Tensor`.
variance: A variance `Tensor`.
offset: An offset `Tensor`, often denoted \\(\beta\\) in equations, or
None. If present, will be added to the normalized tensor.
scale: A scale `Tensor`, often denoted \\(\gamma\\) in equations, or
`None`. If present, the scale is applied to the normalized tensor.
variance_epsilon: A small float number to avoid dividing by 0.
name: A name for this operation (optional).
Returns:
the normalized, scaled, offset tensor.
References:
Batch Normalization - Accelerating Deep Network Training by Reducing
Internal Covariate Shift:
[Ioffe et al., 2015](http://arxiv.org/abs/1502.03167)
([pdf](http://proceedings.mlr.press/v37/ioffe15.pdf))"
8964,fused_batch_norm,tensorflow/tensorflow/python/ops/nn_impl.py,1569,function,"Batch normalization.
See Source: [Batch Normalization: Accelerating Deep Network Training by
Reducing Internal Covariate Shift; S. Ioffe, C. Szegedy]
(http://arxiv.org/abs/1502.03167).
Args:
x: Input `Tensor` of 4 dimensions.
scale: A `Tensor` of 1 dimension for scaling.
offset: A `Tensor` of 1 dimension for bias.
mean: A `Tensor` of 1 dimension for population mean. The shape and meaning
of this argument depends on the value of is_training and
exponential_avg_factor as follows:
is_training==False (inference):
Mean must be a `Tensor` of the same shape as scale containing the
estimated population mean computed during training.
is_training==True and exponential_avg_factor == 1.0:
Mean must be None.
is_training==True and exponential_avg_factor != 1.0:
Mean must be a `Tensor` of the same shape as scale containing the
exponential running mean.
variance: A `Tensor` of 1 dimension for population variance. The shape and
meaning of this argument depends on the value of is_training and
exponential_avg_factor as follows:
is_training==False (inference):
Variance must be a `Tensor` of the same shape as scale containing
the estimated population variance computed during training.
is_training==True and exponential_avg_factor == 1.0:
Variance must be None.
is_training==True and exponential_avg_factor != 1.0:
Variance must be a `Tensor` of the same shape as scale containing
the exponential running variance.
epsilon: A small float number added to the variance of x.
data_format: The data format for x. Either ""NHWC"" (default) or ""NCHW"".
is_training: A bool value to specify if the operation is used for
training or inference.
name: A name for this operation (optional).
exponential_avg_factor: A float number (usually between 0 and 1) used
for controlling the decay of the running
population average of mean and variance.
If set to 1.0, the current batch average is
returned.
Returns:
y: A 4D Tensor for the normalized, scaled, offsetted x.
running_mean: A 1D Tensor for the exponential running mean of x.
The output value is (1 - exponential_avg_factor) * mean +
exponential_avg_factor * batch_mean), where batch_mean
is the mean of the current batch in x.
running_var: A 1D Tensor for the exponential running variance
The output value is (1 - exponential_avg_factor) * variance +
exponential_avg_factor * batch_variance), where batch_variance
is the variance of the current batch in x.
References:
Batch Normalization - Accelerating Deep Network Training by Reducing
Internal Covariate Shift:
[Ioffe et al., 2015](http://proceedings.mlr.press/v37/ioffe15.html)
([pdf](http://proceedings.mlr.press/v37/ioffe15.pdf))"
8965,batch_norm_with_global_normalization,tensorflow/tensorflow/python/ops/nn_impl.py,1675,function,"Batch normalization.
This op is deprecated. See `tf.nn.batch_normalization`.
Args:
t: A 4D input Tensor.
m: A 1D mean Tensor with size matching the last dimension of t.
This is the first output from tf.nn.moments,
or a saved moving average thereof.
v: A 1D variance Tensor with size matching the last dimension of t.
This is the second output from tf.nn.moments,
or a saved moving average thereof.
beta: A 1D beta Tensor with size matching the last dimension of t.
An offset to be added to the normalized tensor.
gamma: A 1D gamma Tensor with size matching the last dimension of t.
If ""scale_after_normalization"" is true, this tensor will be multiplied
with the normalized tensor.
variance_epsilon: A small float number to avoid dividing by 0.
scale_after_normalization: A bool indicating whether the resulted tensor
needs to be multiplied with gamma.
name: A name for this operation (optional).
input: Alias for t.
mean: Alias for m.
variance: Alias for v.
Returns:
A batch-normalized `t`.
References:
Batch Normalization - Accelerating Deep Network Training by Reducing
Internal Covariate Shift:
[Ioffe et al., 2015](http://proceedings.mlr.press/v37/ioffe15.html)
([pdf](http://proceedings.mlr.press/v37/ioffe15.pdf))"
8966,batch_norm_with_global_normalization_v2,tensorflow/tensorflow/python/ops/nn_impl.py,1730,function,"Batch normalization.
This op is deprecated. See `tf.nn.batch_normalization`.
Args:
input: A 4D input Tensor.
mean: A 1D mean Tensor with size matching the last dimension of t.
This is the first output from tf.nn.moments,
or a saved moving average thereof.
variance: A 1D variance Tensor with size matching the last dimension of t.
This is the second output from tf.nn.moments,
or a saved moving average thereof.
beta: A 1D beta Tensor with size matching the last dimension of t.
An offset to be added to the normalized tensor.
gamma: A 1D gamma Tensor with size matching the last dimension of t.
If ""scale_after_normalization"" is true, this tensor will be multiplied
with the normalized tensor.
variance_epsilon: A small float number to avoid dividing by 0.
scale_after_normalization: A bool indicating whether the resulted tensor
needs to be multiplied with gamma.
name: A name for this operation (optional).
Returns:
A batch-normalized `t`.
References:
Batch Normalization - Accelerating Deep Network Training by Reducing Internal Covariate Shift:
[Ioffe et al., 2015](http://proceedings.mlr.press/v37/ioffe15.html)
([pdf](http://proceedings.mlr.press/v37/ioffe15.pdf))"
8967,nce_loss_v2,tensorflow/tensorflow/python/ops/nn_impl.py,1980,function,"Computes and returns the noise-contrastive estimation training loss.
See [Noise-contrastive estimation: A new estimation principle for
unnormalized statistical
models](http://www.jmlr.org/proceedings/papers/v9/gutmann10a/gutmann10a.pdf).
Also see our [Candidate Sampling Algorithms
Reference](https://www.tensorflow.org/extras/candidate_sampling.pdf)
A common use case is to use this method for training, and calculate the full
sigmoid loss for evaluation or inference as in the following example:
```python
if mode == ""train"":
loss = tf.nn.nce_loss(
weights=weights,
biases=biases,
labels=labels,
inputs=inputs,
...)
elif mode == ""eval"":
logits = tf.matmul(inputs, tf.transpose(weights))
logits = tf.nn.bias_add(logits, biases)
labels_one_hot = tf.one_hot(labels, n_classes)
loss = tf.nn.sigmoid_cross_entropy_with_logits(
labels=labels_one_hot,
logits=logits)
loss = tf.reduce_sum(loss, axis=1)
```
Note: when doing embedding lookup on `weights` and `bias`, ""div"" partition
strategy will be used. Support for other partition strategy will be added
later.
Note: By default this uses a log-uniform (Zipfian) distribution for sampling,
so your labels must be sorted in order of decreasing frequency to achieve
good results. For more details, see
`tf.random.log_uniform_candidate_sampler`.
Note: In the case where `num_true` > 1, we assign to each target class
the target probability 1 / `num_true` so that the target probabilities
sum to 1 per-example.
Note: It would be useful to allow a variable number of target classes per
example. We hope to provide this functionality in a future release.
For now, if you have a variable number of target classes, you can pad them
out to a constant number by either repeating them or by padding
with an otherwise unused class.
Args:
weights: A `Tensor` of shape `[num_classes, dim]`, or a list of `Tensor`
objects whose concatenation along dimension 0 has shape [num_classes,
dim]. The (possibly-partitioned) class embeddings.
biases: A `Tensor` of shape `[num_classes]`. The class biases.
labels: A `Tensor` of type `int64` and shape `[batch_size, num_true]`. The
target classes.
inputs: A `Tensor` of shape `[batch_size, dim]`. The forward activations of
the input network.
num_sampled: An `int`. The number of negative classes to randomly sample
per batch. This single sample of negative classes is evaluated for each
element in the batch.
num_classes: An `int`. The number of possible classes.
num_true: An `int`. The number of target classes per training example.
sampled_values: a tuple of (`sampled_candidates`, `true_expected_count`,
`sampled_expected_count`) returned by a `*_candidate_sampler` function.
(if None, we default to `log_uniform_candidate_sampler`)
remove_accidental_hits: A `bool`. Whether to remove ""accidental hits""
where a sampled class equals one of the target classes. If set to `True`,
this is a ""Sampled Logistic"" loss instead of NCE, and we are learning to
generate log-odds instead of log probabilities. See our [Candidate
Sampling Algorithms Reference]
(https://www.tensorflow.org/extras/candidate_sampling.pdf). Default is
False.
name: A name for the operation (optional).
Returns:
A `batch_size` 1-D tensor of per-example NCE losses."
8968,nce_loss,tensorflow/tensorflow/python/ops/nn_impl.py,2085,function,"Computes and returns the noise-contrastive estimation training loss.
A common use case is to use this method for training, and calculate the full
sigmoid loss for evaluation or inference. In this case, you must set
`partition_strategy=""div""` for the two losses to be consistent, as in the
following example:
```python
if mode == ""train"":
loss = tf.nn.nce_loss(
weights=weights,
biases=biases,
labels=labels,
inputs=inputs,
...,
partition_strategy=""div"")
elif mode == ""eval"":
logits = tf.matmul(inputs, tf.transpose(weights))
logits = tf.nn.bias_add(logits, biases)
labels_one_hot = tf.one_hot(labels, n_classes)
loss = tf.nn.sigmoid_cross_entropy_with_logits(
labels=labels_one_hot,
logits=logits)
loss = tf.reduce_sum(loss, axis=1)
```
Note: By default this uses a log-uniform (Zipfian) distribution for sampling,
so your labels must be sorted in order of decreasing frequency to achieve
good results. For more details, see
`tf.random.log_uniform_candidate_sampler`.
Note: In the case where `num_true` > 1, we assign to each target class
the target probability 1 / `num_true` so that the target probabilities
sum to 1 per-example.
Note: It would be useful to allow a variable number of target classes per
example. We hope to provide this functionality in a future release.
For now, if you have a variable number of target classes, you can pad them
out to a constant number by either repeating them or by padding
with an otherwise unused class.
Args:
weights: A `Tensor` of shape `[num_classes, dim]`, or a list of `Tensor`
objects whose concatenation along dimension 0 has shape
[num_classes, dim]. The (possibly-partitioned) class embeddings.
biases: A `Tensor` of shape `[num_classes]`. The class biases.
labels: A `Tensor` of type `int64` and shape `[batch_size,
num_true]`. The target classes.
inputs: A `Tensor` of shape `[batch_size, dim]`. The forward
activations of the input network.
num_sampled: An `int`. The number of negative classes to randomly sample
per batch. This single sample of negative classes is evaluated for each
element in the batch.
num_classes: An `int`. The number of possible classes.
num_true: An `int`. The number of target classes per training example.
sampled_values: a tuple of (`sampled_candidates`, `true_expected_count`,
`sampled_expected_count`) returned by a `*_candidate_sampler` function.
(if None, we default to `log_uniform_candidate_sampler`)
remove_accidental_hits: A `bool`. Whether to remove ""accidental hits""
where a sampled class equals one of the target classes. If set to
`True`, this is a ""Sampled Logistic"" loss instead of NCE, and we are
learning to generate log-odds instead of log probabilities. See
our Candidate Sampling Algorithms Reference
([pdf](https://www.tensorflow.org/extras/candidate_sampling.pdf)).
Default is False.
partition_strategy: A string specifying the partitioning strategy, relevant
if `len(weights) > 1`. Currently `""div""` and `""mod""` are supported.
Default is `""mod""`. See `tf.nn.embedding_lookup` for more details.
name: A name for the operation (optional).
Returns:
A `batch_size` 1-D tensor of per-example NCE losses.
References:
Noise-contrastive estimation - A new estimation principle for unnormalized
statistical models:
[Gutmann et al., 2010](http://proceedings.mlr.press/v9/gutmann10a)
([pdf](http://proceedings.mlr.press/v9/gutmann10a/gutmann10a.pdf))"
8969,sampled_softmax_loss_v2,tensorflow/tensorflow/python/ops/nn_impl.py,2197,function,"Computes and returns the sampled softmax training loss.
This is a faster way to train a softmax classifier over a huge number of
classes.
This operation is for training only. It is generally an underestimate of
the full softmax loss.
A common use case is to use this method for training, and calculate the full
sigmoid loss for evaluation or inference as in the following example:
```python
if mode == ""train"":
loss = tf.nn.sampled_softmax_loss(
weights=weights,
biases=biases,
labels=labels,
inputs=inputs,
...)
elif mode == ""eval"":
logits = tf.matmul(inputs, tf.transpose(weights))
logits = tf.nn.bias_add(logits, biases)
labels_one_hot = tf.one_hot(labels, n_classes)
loss = tf.nn.softmax_cross_entropy_with_logits(
labels=labels_one_hot,
logits=logits)
```
See our [Candidate Sampling Algorithms Reference]
(https://www.tensorflow.org/extras/candidate_sampling.pdf)
Also see Section 3 of [Jean et al., 2014](http://arxiv.org/abs/1412.2007)
([pdf](http://arxiv.org/pdf/1412.2007.pdf)) for the math.
Note: when doing embedding lookup on `weights` and `bias`, ""div"" partition
strategy will be used. Support for other partition strategy will be added
later.
Args:
weights: A `Tensor` of shape `[num_classes, dim]`, or a list of `Tensor`
objects whose concatenation along dimension 0 has shape [num_classes,
dim]. The (possibly-sharded) class embeddings.
biases: A `Tensor` of shape `[num_classes]`. The class biases.
labels: A `Tensor` of type `int64` and shape `[batch_size, num_true]`. The
target classes. Note that this format differs from the `labels` argument
of `nn.softmax_cross_entropy_with_logits`.
inputs: A `Tensor` of shape `[batch_size, dim]`. The forward activations of
the input network.
num_sampled: An `int`. The number of classes to randomly sample per batch.
num_classes: An `int`. The number of possible classes.
num_true: An `int`. The number of target classes per training example.
sampled_values: a tuple of (`sampled_candidates`, `true_expected_count`,
`sampled_expected_count`) returned by a `*_candidate_sampler` function.
(if None, we default to `log_uniform_candidate_sampler`)
remove_accidental_hits: A `bool`. whether to remove ""accidental hits""
where a sampled class equals one of the target classes. Default is True.
seed: random seed for candidate sampling. Default to None, which doesn't set
the op-level random seed for candidate sampling.
name: A name for the operation (optional).
Returns:
A `batch_size` 1-D tensor of per-example sampled softmax losses."
8970,sampled_softmax_loss,tensorflow/tensorflow/python/ops/nn_impl.py,2289,function,"Computes and returns the sampled softmax training loss.
This is a faster way to train a softmax classifier over a huge number of
classes.
This operation is for training only. It is generally an underestimate of
the full softmax loss.
A common use case is to use this method for training, and calculate the full
softmax loss for evaluation or inference. In this case, you must set
`partition_strategy=""div""` for the two losses to be consistent, as in the
following example:
```python
if mode == ""train"":
loss = tf.nn.sampled_softmax_loss(
weights=weights,
biases=biases,
labels=labels,
inputs=inputs,
...,
partition_strategy=""div"")
elif mode == ""eval"":
logits = tf.matmul(inputs, tf.transpose(weights))
logits = tf.nn.bias_add(logits, biases)
labels_one_hot = tf.one_hot(labels, n_classes)
loss = tf.nn.softmax_cross_entropy_with_logits(
labels=labels_one_hot,
logits=logits)
```
See our Candidate Sampling Algorithms Reference
([pdf](https://www.tensorflow.org/extras/candidate_sampling.pdf)).
Also see Section 3 of (Jean et al., 2014) for the math.
Args:
weights: A `Tensor` of shape `[num_classes, dim]`, or a list of `Tensor`
objects whose concatenation along dimension 0 has shape
[num_classes, dim]. The (possibly-sharded) class embeddings.
biases: A `Tensor` of shape `[num_classes]`. The class biases.
labels: A `Tensor` of type `int64` and shape `[batch_size,
num_true]`. The target classes. Note that this format differs from
the `labels` argument of `nn.softmax_cross_entropy_with_logits`.
inputs: A `Tensor` of shape `[batch_size, dim]`. The forward
activations of the input network.
num_sampled: An `int`. The number of classes to randomly sample per batch.
num_classes: An `int`. The number of possible classes.
num_true: An `int`. The number of target classes per training example.
sampled_values: a tuple of (`sampled_candidates`, `true_expected_count`,
`sampled_expected_count`) returned by a `*_candidate_sampler` function.
(if None, we default to `log_uniform_candidate_sampler`)
remove_accidental_hits: A `bool`. whether to remove ""accidental hits""
where a sampled class equals one of the target classes. Default is
True.
partition_strategy: A string specifying the partitioning strategy, relevant
if `len(weights) > 1`. Currently `""div""` and `""mod""` are supported.
Default is `""mod""`. See `tf.nn.embedding_lookup` for more details.
name: A name for the operation (optional).
seed: random seed for candidate sampling. Default to None, which doesn't set
the op-level random seed for candidate sampling.
Returns:
A `batch_size` 1-D tensor of per-example sampled softmax losses.
References:
On Using Very Large Target Vocabulary for Neural Machine Translation:
[Jean et al., 2014]
(https://aclanthology.coli.uni-saarland.de/papers/P15-1001/p15-1001)
([pdf](http://aclweb.org/anthology/P15-1001))"
8971,squeeze_batch_dims,tensorflow/tensorflow/python/ops/nn_ops.py,277,function,"Returns `unsqueeze_batch(op(squeeze_batch(inp)))`.
Where `squeeze_batch` reshapes `inp` to shape
`[prod(inp.shape[:-inner_rank])] + inp.shape[-inner_rank:]`
and `unsqueeze_batch` does the reverse reshape but on the output.
Args:
inp: A tensor with dims `batch_shape + inner_shape` where `inner_shape`
is length `inner_rank`.
op: A callable that takes a single input tensor and returns a single.
output tensor.
inner_rank: A python integer.
name: A string.
Returns:
`unsqueeze_batch_op(squeeze_batch(inp))`."
8972,dilation2d_v2,tensorflow/tensorflow/python/ops/nn_ops.py,328,function,"Computes the grayscale dilation of 4-D `input` and 3-D `filters` tensors.
The `input` tensor has shape `[batch, in_height, in_width, depth]` and the
`filters` tensor has shape `[filter_height, filter_width, depth]`, i.e., each
input channel is processed independently of the others with its own
structuring function. The `output` tensor has shape
`[batch, out_height, out_width, depth]`. The spatial dimensions of the output
tensor depend on the `padding` algorithm. We currently only support the
default ""NHWC"" `data_format`.
In detail, the grayscale morphological 2-D dilation is the max-sum correlation
(for consistency with `conv2d`, we use unmirrored filters):
output[b, y, x, c] =
max_{dy, dx} input[b,
strides[1] * y + rates[1] * dy,
strides[2] * x + rates[2] * dx,
c] +
filters[dy, dx, c]
Max-pooling is a special case when the filter has size equal to the pooling
kernel size and contains all zeros.
Note on duality: The dilation of `input` by the `filters` is equal to the
negation of the erosion of `-input` by the reflected `filters`.
Args:
input: A `Tensor`. Must be one of the following types: `float32`, `float64`,
`int32`, `uint8`, `int16`, `int8`, `int64`, `bfloat16`, `uint16`, `half`,
`uint32`, `uint64`.
4-D with shape `[batch, in_height, in_width, depth]`.
filters: A `Tensor`. Must have the same type as `input`.
3-D with shape `[filter_height, filter_width, depth]`.
strides: A list of `ints` that has length `>= 4`.
The stride of the sliding window for each dimension of the input
tensor. Must be: `[1, stride_height, stride_width, 1]`.
padding: A `string` from: `""SAME"", ""VALID""`.
The type of padding algorithm to use.
data_format: A `string`, only `""NHWC""` is currently supported.
dilations: A list of `ints` that has length `>= 4`.
The input stride for atrous morphological dilation. Must be:
`[1, rate_height, rate_width, 1]`.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `input`."
8973,dilation2d_v1,tensorflow/tensorflow/python/ops/nn_ops.py,396,function,
8974,with_space_to_batch,tensorflow/tensorflow/python/ops/nn_ops.py,415,function,"Performs `op` on the space-to-batch representation of `input`.
This has the effect of transforming sliding window operations into the
corresponding ""atrous"" operation in which the input is sampled at the
specified `dilation_rate`.
In the special case that `dilation_rate` is uniformly 1, this simply returns:
op(input, num_spatial_dims, padding)
Otherwise, it returns:
batch_to_space_nd(
op(space_to_batch_nd(input, adjusted_dilation_rate, adjusted_paddings),
num_spatial_dims,
""VALID"")
adjusted_dilation_rate,
adjusted_crops),
where:
adjusted_dilation_rate is an int64 tensor of shape [max(spatial_dims)],
adjusted_{paddings,crops} are int64 tensors of shape [max(spatial_dims), 2]
defined as follows:
We first define two int64 tensors `paddings` and `crops` of shape
`[num_spatial_dims, 2]` based on the value of `padding` and the spatial
dimensions of the `input`:
If `padding = ""VALID""`, then:
paddings, crops = required_space_to_batch_paddings(
input_shape[spatial_dims],
dilation_rate)
If `padding = ""SAME""`, then:
dilated_filter_shape =
filter_shape + (filter_shape - 1) * (dilation_rate - 1)
paddings, crops = required_space_to_batch_paddings(
input_shape[spatial_dims],
dilation_rate,
[(dilated_filter_shape - 1) // 2,
dilated_filter_shape - 1 - (dilated_filter_shape - 1) // 2])
Because `space_to_batch_nd` and `batch_to_space_nd` assume that the spatial
dimensions are contiguous starting at the second dimension, but the specified
`spatial_dims` may not be, we must adjust `dilation_rate`, `paddings` and
`crops` in order to be usable with these operations. For a given dimension,
if the block size is 1, and both the starting and ending padding and crop
amounts are 0, then space_to_batch_nd effectively leaves that dimension alone,
which is what is needed for dimensions not part of `spatial_dims`.
Furthermore, `space_to_batch_nd` and `batch_to_space_nd` handle this case
efficiently for any number of leading and trailing dimensions.
For 0 <= i < len(spatial_dims), we assign:
adjusted_dilation_rate[spatial_dims[i] - 1] = dilation_rate[i]
adjusted_paddings[spatial_dims[i] - 1, :] = paddings[i, :]
adjusted_crops[spatial_dims[i] - 1, :] = crops[i, :]
All unassigned values of `adjusted_dilation_rate` default to 1, while all
unassigned values of `adjusted_paddings` and `adjusted_crops` default to 0.
Note in the case that `dilation_rate` is not uniformly 1, specifying ""VALID""
padding is equivalent to specifying `padding = ""SAME""` with a filter_shape of
`[1]*N`.
Advanced usage. Note the following optimization: A sequence of
`with_space_to_batch` operations with identical (not uniformly 1)
`dilation_rate` parameters and ""VALID"" padding
net = with_space_to_batch(net, dilation_rate, ""VALID"", op_1)
...
net = with_space_to_batch(net, dilation_rate, ""VALID"", op_k)
can be combined into a single `with_space_to_batch` operation as follows:
def combined_op(converted_input, num_spatial_dims, _):
result = op_1(converted_input, num_spatial_dims, ""VALID"")
...
result = op_k(result, num_spatial_dims, ""VALID"")
net = with_space_to_batch(net, dilation_rate, ""VALID"", combined_op)
This eliminates the overhead of `k-1` calls to `space_to_batch_nd` and
`batch_to_space_nd`.
Similarly, a sequence of `with_space_to_batch` operations with identical (not
uniformly 1) `dilation_rate` parameters, ""SAME"" padding, and odd filter
dimensions
net = with_space_to_batch(net, dilation_rate, ""SAME"", op_1, filter_shape_1)
...
net = with_space_to_batch(net, dilation_rate, ""SAME"", op_k, filter_shape_k)
can be combined into a single `with_space_to_batch` operation as follows:
def combined_op(converted_input, num_spatial_dims, _):
result = op_1(converted_input, num_spatial_dims, ""SAME"")
...
result = op_k(result, num_spatial_dims, ""SAME"")
net = with_space_to_batch(net, dilation_rate, ""VALID"", combined_op)
Args:
input: Tensor of rank > max(spatial_dims).
dilation_rate: int32 Tensor of *known* shape [num_spatial_dims].
padding: str constant equal to ""VALID"" or ""SAME""
op: Function that maps (input, num_spatial_dims, padding) -> output
filter_shape: If padding = ""SAME"", specifies the shape of the convolution
kernel/pooling window as an integer Tensor of shape [>=num_spatial_dims].
If padding = ""VALID"", filter_shape is ignored and need not be specified.
spatial_dims: Monotonically increasing sequence of `num_spatial_dims`
integers (which are >= 1) specifying the spatial dimensions of `input`
and output. Defaults to: `range(1, num_spatial_dims+1)`.
data_format: A string or None. Specifies whether the channel dimension of
the `input` and output is the last dimension (default, or if `data_format`
does not start with ""NC""), or the second dimension (if `data_format`
starts with ""NC""). For N=1, the valid values are ""NWC"" (default) and
""NCW"". For N=2, the valid values are ""NHWC"" (default) and ""NCHW"".
For N=3, the valid values are ""NDHWC"" (default) and ""NCDHW"".
Returns:
The output Tensor as described above, dimensions will vary based on the op
provided.
Raises:
ValueError: if `padding` is invalid or the arguments are incompatible.
ValueError: if `spatial_dims` are invalid."
8975,convolution,tensorflow/tensorflow/python/ops/nn_ops.py,866,function,"Computes sums of N-D convolutions (actually cross-correlation).
This also supports either output striding via the optional `strides` parameter
or atrous convolution (also known as convolution with holes or dilated
convolution, based on the French word ""trous"" meaning holes in English) via
the optional `dilation_rate` parameter. Currently, however, output striding
is not supported for atrous convolutions.
Specifically, in the case that `data_format` does not start with ""NC"", given
a rank (N+2) `input` Tensor of shape
[num_batches,
input_spatial_shape[0],
...,
input_spatial_shape[N-1],
num_input_channels],
a rank (N+2) `filter` Tensor of shape
[spatial_filter_shape[0],
...,
spatial_filter_shape[N-1],
num_input_channels,
num_output_channels],
an optional `dilation_rate` tensor of shape [N] (defaulting to [1]*N)
specifying the filter upsampling/input downsampling rate, and an optional list
of N `strides` (defaulting [1]*N), this computes for each N-D spatial output
position (x[0], ..., x[N-1]):
```
output[b, x[0], ..., x[N-1], k] =
sum_{z[0], ..., z[N-1], q}
filter[z[0], ..., z[N-1], q, k] *
padded_input[b,
x[0]*strides[0] + dilation_rate[0]*z[0],
...,
x[N-1]*strides[N-1] + dilation_rate[N-1]*z[N-1],
q]
```
where b is the index into the batch, k is the output channel number, q is the
input channel number, and z is the N-D spatial offset within the filter. Here,
`padded_input` is obtained by zero padding the input using an effective
spatial filter shape of `(spatial_filter_shape-1) * dilation_rate + 1` and
output striding `strides` as described in the
[comment here](https://tensorflow.org/api_guides/python/nn#Convolution).
In the case that `data_format` does start with `""NC""`, the `input` and output
(but not the `filter`) are simply transposed as follows:
convolution(input, data_format, **kwargs) =
tf.transpose(convolution(tf.transpose(input, [0] + range(2,N+2) + [1]),
**kwargs),
[0, N+1] + range(1, N+1))
It is required that 1 <= N <= 3.
Args:
input: An (N+2)-D `Tensor` of type `T`, of shape
`[batch_size] + input_spatial_shape + [in_channels]` if data_format does
not start with ""NC"" (default), or
`[batch_size, in_channels] + input_spatial_shape` if data_format starts
with ""NC"".
filter: An (N+2)-D `Tensor` with the same type as `input` and shape
`spatial_filter_shape + [in_channels, out_channels]`.
padding: A string, either `""VALID""` or `""SAME""`. The padding algorithm.
`""valid""` means no padding. `""same""` results in padding evenly to
the left/right or up/down of the input such that output has the same
height/width dimension as the input.
strides: Optional. Sequence of N ints >= 1. Specifies the output stride.
Defaults to [1]*N. If any value of strides is > 1, then all values of
dilation_rate must be 1.
dilation_rate: Optional. Sequence of N ints >= 1. Specifies the filter
upsampling/input downsampling rate. In the literature, the same parameter
is sometimes called `input stride` or `dilation`. The effective filter
size used for the convolution will be `spatial_filter_shape +
(spatial_filter_shape - 1) * (rate - 1)`, obtained by inserting
(dilation_rate[i]-1) zeros between consecutive elements of the original
filter in each spatial dimension i. If any value of dilation_rate is > 1,
then all values of strides must be 1.
name: Optional name for the returned tensor.
data_format: A string or None. Specifies whether the channel dimension of
the `input` and output is the last dimension (default, or if `data_format`
does not start with ""NC""), or the second dimension (if `data_format`
starts with ""NC""). For N=1, the valid values are ""NWC"" (default) and
""NCW"". For N=2, the valid values are ""NHWC"" (default) and ""NCHW"".
For N=3, the valid values are ""NDHWC"" (default) and ""NCDHW"".
Returns:
A `Tensor` with the same type as `input` of shape
`[batch_size] + output_spatial_shape + [out_channels]`
if data_format is None or does not start with ""NC"", or
`[batch_size, out_channels] + output_spatial_shape`
if data_format starts with ""NC"",
where `output_spatial_shape` depends on the value of `padding`.
If padding == ""SAME"":
output_spatial_shape[i] = ceil(input_spatial_shape[i] / strides[i])
If padding == ""VALID"":
output_spatial_shape[i] =
ceil((input_spatial_shape[i] -
(spatial_filter_shape[i]-1) * dilation_rate[i])
/ strides[i]).
Raises:
ValueError: If input/output depth does not match `filter` shape, if padding
is other than `""VALID""` or `""SAME""`, or if data_format is invalid."
8976,convolution_v2,tensorflow/tensorflow/python/ops/nn_ops.py,1005,function,
8977,convolution_internal,tensorflow/tensorflow/python/ops/nn_ops.py,1029,function,"Internal function which performs rank agnostic convolution.
Args:
input: See `convolution`.
filters: See `convolution`.
strides: See `convolution`.
padding: See `convolution`.
data_format: See `convolution`.
dilations: See `convolution`.
name: See `convolution`.
call_from_convolution: See `convolution`.
num_spatial_dims: (Optional.). It is a integer describing the
rank of the spatial dimensions. For `1-D`, `2-D` and `3-D` convolutions,
the value of `num_spatial_dims` is `1`, `2`, and `3`, respectively.
This argument is only required to disambiguate the rank of `batch_shape`
when `filter_shape.ndims is None` and `len(batch_shape) > 1`. For
backwards compatibility, if `num_spatial_dims is None` and
`filter_shape.ndims is None`, then `len(batch_shape)` is assumed to be
`1` (i.e., the input is expected to be
`[batch_size, num_channels] + input_spatial_shape`
or `[batch_size] + input_spatial_shape + [num_channels]`.
Returns:
A tensor of shape and dtype matching that of `input`.
Raises:
ValueError: If input and filter both have unknown shapes, or if
`num_spatial_dims` is provided and incompatible with the value
estimated from `filters.shape`."
8978,Convolution,tensorflow/tensorflow/python/ops/nn_ops.py,1171,class,"Helper class for convolution.
Note that this class assumes that shapes of input and filter passed to
`__call__` are compatible with `input_shape`, `filter_shape`, and
`num_spatial_dims` passed to the constructor.
Arguments
input_shape: static shape of input. i.e. input.shape. Its length is
`batch_shape + input_spatial_shape + [num_channels]` if `data_format`
does not start with `NC`, or
`batch_shape + [num_channels] + input_spatial_shape` if `data_format`
starts with `NC`.
filter_shape: static shape of the filter. i.e. filter.shape.
padding: The padding algorithm, must be ""SAME"" or ""VALID"".
strides: see convolution.
dilation_rate: see convolution.
name: see convolution.
data_format: A string or `None`. Specifies whether the channel dimension of
the `input` and output is the last dimension (if `data_format` is `None`
or does not start with `NC`), or the first post-batch dimension (i.e. if
`data_format` starts with `NC`).
num_spatial_dims: (Usually optional.) Python integer, the rank of the
spatial and channel dimensions. For `1-D`, `2-D` and `3-D` convolutions,
the value of `num_spatial_dims` is `1`, `2`, and `3`, respectively.
This argument is only required to disambiguate the rank of `batch_shape`
when `filter_shape.ndims is None` and `len(batch_shape) > 1`. For
backwards compatibility, if `num_spatial_dims is None` and
`filter_shape.ndims is None`, then `len(batch_shape)` is assumed to be
`1` (i.e., the input is expected to be
`[batch_size, num_channels] + input_spatial_shape`
or `[batch_size] + input_spatial_shape + [num_channels]`."
8979,pool,tensorflow/tensorflow/python/ops/nn_ops.py,1332,function,"Performs an N-D pooling operation.
In the case that `data_format` does not start with ""NC"", computes for
0 <= b < batch_size,
0 <= x[i] < output_spatial_shape[i],
0 <= c < num_channels:
```
output[b, x[0], ..., x[N-1], c] =
REDUCE_{z[0], ..., z[N-1]}
input[b,
x[0] * strides[0] - pad_before[0] + dilation_rate[0]*z[0],
...
x[N-1]*strides[N-1] - pad_before[N-1] + dilation_rate[N-1]*z[N-1],
c],
```
where the reduction function REDUCE depends on the value of `pooling_type`,
and pad_before is defined based on the value of `padding` as described in
the ""returns"" section of `tf.nn.convolution` for details.
The reduction never includes out-of-bounds positions.
In the case that `data_format` starts with `""NC""`, the `input` and output are
simply transposed as follows:
```
pool(input, data_format, **kwargs) =
tf.transpose(pool(tf.transpose(input, [0] + range(2,N+2) + [1]),
**kwargs),
[0, N+1] + range(1, N+1))
```
Args:
input: Tensor of rank N+2, of shape
`[batch_size] + input_spatial_shape + [num_channels]` if data_format does
not start with ""NC"" (default), or
`[batch_size, num_channels] + input_spatial_shape` if data_format starts
with ""NC"". Pooling happens over the spatial dimensions only.
window_shape: Sequence of N ints >= 1.
pooling_type: Specifies pooling operation, must be ""AVG"" or ""MAX"".
padding: The padding algorithm, must be ""SAME"" or ""VALID"".
See the ""returns"" section of `tf.nn.convolution` for details.
dilation_rate: Optional. Dilation rate. List of N ints >= 1.
Defaults to [1]*N. If any value of dilation_rate is > 1, then all values
of strides must be 1.
strides: Optional. Sequence of N ints >= 1. Defaults to [1]*N.
If any value of strides is > 1, then all values of dilation_rate must be
1.
name: Optional. Name of the op.
data_format: A string or None. Specifies whether the channel dimension of
the `input` and output is the last dimension (default, or if `data_format`
does not start with ""NC""), or the second dimension (if `data_format`
starts with ""NC""). For N=1, the valid values are ""NWC"" (default) and
""NCW"". For N=2, the valid values are ""NHWC"" (default) and ""NCHW"".
For N=3, the valid values are ""NDHWC"" (default) and ""NCDHW"".
dilations: Alias for dilation_rate
Returns:
Tensor of rank N+2, of shape
[batch_size] + output_spatial_shape + [num_channels]
if data_format is None or does not start with ""NC"", or
[batch_size, num_channels] + output_spatial_shape
if data_format starts with ""NC"",
where `output_spatial_shape` depends on the value of padding:
If padding = ""SAME"":
output_spatial_shape[i] = ceil(input_spatial_shape[i] / strides[i])
If padding = ""VALID"":
output_spatial_shape[i] =
ceil((input_spatial_shape[i] - (window_shape[i] - 1) * dilation_rate[i])
/ strides[i]).
Raises:
ValueError: if arguments are invalid."
8980,pool_v2,tensorflow/tensorflow/python/ops/nn_ops.py,1507,function,"Performs an N-D pooling operation.
In the case that `data_format` does not start with ""NC"", computes for
0 <= b < batch_size,
0 <= x[i] < output_spatial_shape[i],
0 <= c < num_channels:
```
output[b, x[0], ..., x[N-1], c] =
REDUCE_{z[0], ..., z[N-1]}
input[b,
x[0] * strides[0] - pad_before[0] + dilation_rate[0]*z[0],
...
x[N-1]*strides[N-1] - pad_before[N-1] + dilation_rate[N-1]*z[N-1],
c],
```
where the reduction function REDUCE depends on the value of `pooling_type`,
and pad_before is defined based on the value of `padding` as described in
the ""returns"" section of `tf.nn.convolution` for details.
The reduction never includes out-of-bounds positions.
In the case that `data_format` starts with `""NC""`, the `input` and output are
simply transposed as follows:
```
pool(input, data_format, **kwargs) =
tf.transpose(pool(tf.transpose(input, [0] + range(2,N+2) + [1]),
**kwargs),
[0, N+1] + range(1, N+1))
```
Args:
input: Tensor of rank N+2, of shape `[batch_size] + input_spatial_shape +
[num_channels]` if data_format does not start with ""NC"" (default), or
`[batch_size, num_channels] + input_spatial_shape` if data_format starts
with ""NC"". Pooling happens over the spatial dimensions only.
window_shape: Sequence of N ints >= 1.
pooling_type: Specifies pooling operation, must be ""AVG"" or ""MAX"".
strides: Optional. Sequence of N ints >= 1. Defaults to [1]*N. If any value of
strides is > 1, then all values of dilation_rate must be 1.
padding: The padding algorithm, must be ""SAME"" or ""VALID"". Defaults to ""SAME"".
See the ""returns"" section of `tf.nn.convolution` for details.
data_format: A string or None. Specifies whether the channel dimension of
the `input` and output is the last dimension (default, or if `data_format`
does not start with ""NC""), or the second dimension (if `data_format`
starts with ""NC""). For N=1, the valid values are ""NWC"" (default) and
""NCW"". For N=2, the valid values are ""NHWC"" (default) and ""NCHW"". For
N=3, the valid values are ""NDHWC"" (default) and ""NCDHW"".
dilations: Optional. Dilation rate. List of N ints >= 1. Defaults to
[1]*N. If any value of dilation_rate is > 1, then all values of strides
must be 1.
name: Optional. Name of the op.
Returns:
Tensor of rank N+2, of shape
[batch_size] + output_spatial_shape + [num_channels]
if data_format is None or does not start with ""NC"", or
[batch_size, num_channels] + output_spatial_shape
if data_format starts with ""NC"",
where `output_spatial_shape` depends on the value of padding:
If padding = ""SAME"":
output_spatial_shape[i] = ceil(input_spatial_shape[i] / strides[i])
If padding = ""VALID"":
output_spatial_shape[i] =
ceil((input_spatial_shape[i] - (window_shape[i] - 1) * dilation_rate[i])
/ strides[i]).
Raises:
ValueError: if arguments are invalid."
8981,atrous_conv2d,tensorflow/tensorflow/python/ops/nn_ops.py,1607,function,"Atrous convolution (a.k.a. convolution with holes or dilated convolution).
This function is a simpler wrapper around the more general
`tf.nn.convolution`, and exists only for backwards compatibility. You can
use `tf.nn.convolution` to perform 1-D, 2-D, or 3-D atrous convolution.
Computes a 2-D atrous convolution, also known as convolution with holes or
dilated convolution, given 4-D `value` and `filters` tensors. If the `rate`
parameter is equal to one, it performs regular 2-D convolution. If the `rate`
parameter is greater than one, it performs convolution with holes, sampling
the input values every `rate` pixels in the `height` and `width` dimensions.
This is equivalent to convolving the input with a set of upsampled filters,
produced by inserting `rate - 1` zeros between two consecutive values of the
filters along the `height` and `width` dimensions, hence the name atrous
convolution or convolution with holes (the French word trous means holes in
English).
More specifically:
```
output[batch, height, width, out_channel] =
sum_{dheight, dwidth, in_channel} (
filters[dheight, dwidth, in_channel, out_channel] *
value[batch, height + rate*dheight, width + rate*dwidth, in_channel]
)
```
Atrous convolution allows us to explicitly control how densely to compute
feature responses in fully convolutional networks. Used in conjunction with
bilinear interpolation, it offers an alternative to `conv2d_transpose` in
dense prediction tasks such as semantic image segmentation, optical flow
computation, or depth estimation. It also allows us to effectively enlarge
the field of view of filters without increasing the number of parameters or
the amount of computation.
For a description of atrous convolution and how it can be used for dense
feature extraction, please see: (Chen et al., 2015). The same operation is
investigated further in (Yu et al., 2016). Previous works that effectively
use atrous convolution in different ways are, among others,
(Sermanet et al., 2014) and (Giusti et al., 2013).
Atrous convolution is also closely related to the so-called noble identities
in multi-rate signal processing.
There are many different ways to implement atrous convolution (see the refs
above). The implementation here reduces
```python
atrous_conv2d(value, filters, rate, padding=padding)
```
to the following three operations:
```python
paddings = ...
net = space_to_batch(value, paddings, block_size=rate)
net = conv2d(net, filters, strides=[1, 1, 1, 1], padding=""VALID"")
crops = ...
net = batch_to_space(net, crops, block_size=rate)
```
Advanced usage. Note the following optimization: A sequence of `atrous_conv2d`
operations with identical `rate` parameters, 'SAME' `padding`, and filters
with odd heights/ widths:
```python
net = atrous_conv2d(net, filters1, rate, padding=""SAME"")
net = atrous_conv2d(net, filters2, rate, padding=""SAME"")
...
net = atrous_conv2d(net, filtersK, rate, padding=""SAME"")
```
can be equivalently performed cheaper in terms of computation and memory as:
```python
pad = ... # padding so that the input dims are multiples of rate
net = space_to_batch(net, paddings=pad, block_size=rate)
net = conv2d(net, filters1, strides=[1, 1, 1, 1], padding=""SAME"")
net = conv2d(net, filters2, strides=[1, 1, 1, 1], padding=""SAME"")
...
net = conv2d(net, filtersK, strides=[1, 1, 1, 1], padding=""SAME"")
net = batch_to_space(net, crops=pad, block_size=rate)
```
because a pair of consecutive `space_to_batch` and `batch_to_space` ops with
the same `block_size` cancel out when their respective `paddings` and `crops`
inputs are identical.
Args:
value: A 4-D `Tensor` of type `float`. It needs to be in the default ""NHWC""
format. Its shape is `[batch, in_height, in_width, in_channels]`.
filters: A 4-D `Tensor` with the same type as `value` and shape
`[filter_height, filter_width, in_channels, out_channels]`. `filters`'
`in_channels` dimension must match that of `value`. Atrous convolution is
equivalent to standard convolution with upsampled filters with effective
height `filter_height + (filter_height - 1) * (rate - 1)` and effective
width `filter_width + (filter_width - 1) * (rate - 1)`, produced by
inserting `rate - 1` zeros along consecutive elements across the
`filters`' spatial dimensions.
rate: A positive int32. The stride with which we sample input values across
the `height` and `width` dimensions. Equivalently, the rate by which we
upsample the filter values by inserting zeros across the `height` and
`width` dimensions. In the literature, the same parameter is sometimes
called `input stride` or `dilation`.
padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm.
name: Optional name for the returned tensor.
Returns:
A `Tensor` with the same type as `value`.
Output shape with `'VALID'` padding is:
[batch, height - 2 * (filter_width - 1),
width - 2 * (filter_height - 1), out_channels].
Output shape with `'SAME'` padding is:
[batch, height, width, out_channels].
Raises:
ValueError: If input/output depth does not match `filters`' shape, or if
padding is other than `'VALID'` or `'SAME'`.
References:
Multi-Scale Context Aggregation by Dilated Convolutions:
[Yu et al., 2016](https://arxiv.org/abs/1511.07122)
([pdf](https://arxiv.org/pdf/1511.07122.pdf))
Semantic Image Segmentation with Deep Convolutional Nets and Fully
Connected CRFs:
[Chen et al., 2015](http://arxiv.org/abs/1412.7062)
([pdf](https://arxiv.org/pdf/1412.7062))
OverFeat - Integrated Recognition, Localization and Detection using
Convolutional Networks:
[Sermanet et al., 2014](https://arxiv.org/abs/1312.6229)
([pdf](https://arxiv.org/pdf/1312.6229.pdf))
Fast Image Scanning with Deep Max-Pooling Convolutional Neural Networks:
[Giusti et al., 2013]
(https://ieeexplore.ieee.org/abstract/document/6738831)
([pdf](https://arxiv.org/pdf/1302.1700.pdf))"
8982,convert_padding,tensorflow/tensorflow/python/ops/nn_ops.py,1755,function,"Converts Python padding to C++ padding for ops which take EXPLICIT padding.
Args:
padding: the `padding` argument for a Python op which supports EXPLICIT
padding.
Returns:
(padding, explicit_paddings) pair, which should be passed as attributes to a
C++ op.
Raises:
ValueError: If padding is invalid."
8983,conv1d,tensorflow/tensorflow/python/ops/nn_ops.py,1805,function,"Computes a 1-D convolution of input with rank `>=3` and a `3-D` filter.
Given an input tensor of shape
`batch_shape + [in_width, in_channels]`
if `data_format` is `""NWC""`, or
`batch_shape + [in_channels, in_width]`
if `data_format` is `""NCW""`,
and a filter / kernel tensor of shape
`[filter_width, in_channels, out_channels]`, this op reshapes
the arguments to pass them to `conv2d` to perform the equivalent
convolution operation.
Internally, this op reshapes the input tensors and invokes `tf.nn.conv2d`.
For example, if `data_format` does not start with ""NC"", a tensor of shape
`batch_shape + [in_width, in_channels]`
is reshaped to
`batch_shape + [1, in_width, in_channels]`,
and the filter is reshaped to
`[1, filter_width, in_channels, out_channels]`.
The result is then reshaped back to
`batch_shape + [out_width, out_channels]`
\(where out_width is a function of the stride and padding as in conv2d\) and
returned to the caller.
Args:
value: A Tensor of rank at least 3. Must be of type `float16`, `float32`, or
`float64`.
filters: A Tensor of rank at least 3. Must have the same type as `value`.
stride: An int or list of `ints` that has length `1` or `3`. The number of
entries by which the filter is moved right at each step.
padding: 'SAME' or 'VALID'
use_cudnn_on_gpu: An optional `bool`. Defaults to `True`.
data_format: An optional `string` from `""NWC"", ""NCW""`. Defaults to `""NWC""`,
the data is stored in the order of `batch_shape + [in_width,
in_channels]`. The `""NCW""` format stores data as `batch_shape +
[in_channels, in_width]`.
name: A name for the operation (optional).
input: Alias for value.
dilations: An int or list of `ints` that has length `1` or `3` which
defaults to 1. The dilation factor for each dimension of input. If set to
k > 1, there will be k-1 skipped cells between each filter element on that
dimension. Dilations in the batch and depth dimensions must be 1.
Returns:
A `Tensor`. Has the same type as input.
Raises:
ValueError: if `data_format` is invalid."
8984,conv1d_v2,tensorflow/tensorflow/python/ops/nn_ops.py,1911,function,"Computes a 1-D convolution given 3-D input and filter tensors.
Given an input tensor of shape
`batch_shape + [in_width, in_channels]`
if `data_format` is `""NWC""`, or
`batch_shape + [in_channels, in_width]`
if `data_format` is `""NCW""`,
and a filter / kernel tensor of shape
`[filter_width, in_channels, out_channels]`, this op reshapes
the arguments to pass them to `conv2d` to perform the equivalent
convolution operation.
Internally, this op reshapes the input tensors and invokes `tf.nn.conv2d`.
For example, if `data_format` does not start with `""NC""`, a tensor of shape
`batch_shape + [in_width, in_channels]`
is reshaped to
`batch_shape + [1, in_width, in_channels]`,
and the filter is reshaped to
`[1, filter_width, in_channels, out_channels]`.
The result is then reshaped back to
`batch_shape + [out_width, out_channels]`
\(where out_width is a function of the stride and padding as in conv2d\) and
returned to the caller.
Args:
input: A Tensor of rank at least 3. Must be of type `float16`, `float32`, or
`float64`.
filters: A Tensor of rank at least 3. Must have the same type as `input`.
stride: An int or list of `ints` that has length `1` or `3`. The number of
entries by which the filter is moved right at each step.
padding: 'SAME' or 'VALID'
data_format: An optional `string` from `""NWC"", ""NCW""`. Defaults to `""NWC""`,
the data is stored in the order of
`batch_shape + [in_width, in_channels]`. The `""NCW""` format stores data
as `batch_shape + [in_channels, in_width]`.
dilations: An int or list of `ints` that has length `1` or `3` which
defaults to 1. The dilation factor for each dimension of input. If set to
k > 1, there will be k-1 skipped cells between each filter element on that
dimension. Dilations in the batch and depth dimensions must be 1.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as input.
Raises:
ValueError: if `data_format` is invalid."
8985,conv1d_transpose,tensorflow/tensorflow/python/ops/nn_ops.py,1979,function,"The transpose of `conv1d`.
This operation is sometimes called ""deconvolution"" after
(Zeiler et al., 2010), but is actually the transpose (gradient) of `conv1d`
rather than an actual deconvolution.
Args:
input: A 3-D `Tensor` of type `float` and shape
`[batch, in_width, in_channels]` for `NWC` data format or
`[batch, in_channels, in_width]` for `NCW` data format.
filters: A 3-D `Tensor` with the same type as `input` and shape
`[filter_width, output_channels, in_channels]`. `filter`'s
`in_channels` dimension must match that of `input`.
output_shape: A 1-D `Tensor`, containing three elements, representing the
output shape of the deconvolution op.
strides: An int or list of `ints` that has length `1` or `3`. The number of
entries by which the filter is moved right at each step.
padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm.
See the ""returns"" section of `tf.nn.convolution` for details.
data_format: A string. `'NWC'` and `'NCW'` are supported.
dilations: An int or list of `ints` that has length `1` or `3` which
defaults to 1. The dilation factor for each dimension of input. If set to
k > 1, there will be k-1 skipped cells between each filter element on that
dimension. Dilations in the batch and depth dimensions must be 1.
name: Optional name for the returned tensor.
Returns:
A `Tensor` with the same type as `input`.
Raises:
ValueError: If input/output depth does not match `filter`'s shape, if
`output_shape` is not at 3-element vector, if `padding` is other than
`'VALID'` or `'SAME'`, or if `data_format` is invalid.
References:
Deconvolutional Networks:
[Zeiler et al., 2010]
(https://ieeexplore.ieee.org/abstract/document/5539957)
([pdf]
(http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.232.4023&rep=rep1&type=pdf))"
8986,conv2d_v2,tensorflow/tensorflow/python/ops/nn_ops.py,2068,function,"Computes a 2-D convolution given `input` and 4-D `filters` tensors.
The `input` tensor may have rank `4` or higher, where shape dimensions `[:-3]`
are considered batch dimensions (`batch_shape`).
Given an input tensor of shape
`batch_shape + [in_height, in_width, in_channels]` and a filter / kernel
tensor of shape `[filter_height, filter_width, in_channels, out_channels]`,
this op performs the following:
1. Flattens the filter to a 2-D matrix with shape
`[filter_height * filter_width * in_channels, output_channels]`.
2. Extracts image patches from the input tensor to form a *virtual*
tensor of shape `[batch, out_height, out_width,
filter_height * filter_width * in_channels]`.
3. For each patch, right-multiplies the filter matrix and the image patch
vector.
In detail, with the default NHWC format,
output[b, i, j, k] =
sum_{di, dj, q} input[b, strides[1] * i + di, strides[2] * j + dj, q] *
filter[di, dj, q, k]
Must have `strides[0] = strides[3] = 1`. For the most common case of the same
horizontal and vertical strides, `strides = [1, stride, stride, 1]`.
Usage Example:
>>> x_in = np.array([[
... [[2], [1], [2], [0], [1]],
... [[1], [3], [2], [2], [3]],
... [[1], [1], [3], [3], [0]],
... [[2], [2], [0], [1], [1]],
... [[0], [0], [3], [1], [2]], ]])
>>> kernel_in = np.array([
... [ [[2, 0.1]], [[3, 0.2]] ],
... [ [[0, 0.3]],[[1, 0.4]] ], ])
>>> x = tf.constant(x_in, dtype=tf.float32)
>>> kernel = tf.constant(kernel_in, dtype=tf.float32)
>>> tf.nn.conv2d(x, kernel, strides=[1, 1, 1, 1], padding='VALID')
<tf.Tensor: shape=(1, 4, 4, 2), dtype=float32, numpy=..., dtype=float32)>
Args:
input: A `Tensor`. Must be one of the following types:
`half`, `bfloat16`, `float32`, `float64`.
A Tensor of rank at least 4. The dimension order is interpreted according
to the value of `data_format`; with the all-but-inner-3 dimensions acting
as batch dimensions. See below for details.
filters: A `Tensor`. Must have the same type as `input`.
A 4-D tensor of shape
`[filter_height, filter_width, in_channels, out_channels]`
strides: An int or list of `ints` that has length `1`, `2` or `4`. The
stride of the sliding window for each dimension of `input`. If a single
value is given it is replicated in the `H` and `W` dimension. By default
the `N` and `C` dimensions are set to 1. The dimension order is determined
by the value of `data_format`, see below for details.
padding: Either the `string` `""SAME""` or `""VALID""` indicating the type of
padding algorithm to use, or a list indicating the explicit paddings at
the start and end of each dimension. When explicit padding is used and
data_format is `""NHWC""`, this should be in the form `[[0, 0], [pad_top,
pad_bottom], [pad_left, pad_right], [0, 0]]`. When explicit padding used
and data_format is `""NCHW""`, this should be in the form `[[0, 0], [0, 0],
[pad_top, pad_bottom], [pad_left, pad_right]]`.
data_format: An optional `string` from: `""NHWC"", ""NCHW""`.
Defaults to `""NHWC""`.
Specify the data format of the input and output data. With the
default format ""NHWC"", the data is stored in the order of:
`batch_shape + [height, width, channels]`.
Alternatively, the format could be ""NCHW"", the data storage order of:
`batch_shape + [channels, height, width]`.
dilations: An int or list of `ints` that has length `1`, `2` or `4`,
defaults to 1. The dilation factor for each dimension of`input`. If a
single value is given it is replicated in the `H` and `W` dimension. By
default the `N` and `C` dimensions are set to 1. If set to k > 1, there
will be k-1 skipped cells between each filter element on that dimension.
The dimension order is determined by the value of `data_format`, see above
for details. Dilations in the batch and depth dimensions if a 4-d tensor
must be 1.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `input` and the same outer batch shape."
8987,conv2d,tensorflow/tensorflow/python/ops/nn_ops.py,2173,function,"Computes a 2-D convolution given 4-D `input` and `filter` tensors.
Given an input tensor of shape `[batch, in_height, in_width, in_channels]`
and a filter / kernel tensor of shape
`[filter_height, filter_width, in_channels, out_channels]`, this op
performs the following:
1. Flattens the filter to a 2-D matrix with shape
`[filter_height * filter_width * in_channels, output_channels]`.
2. Extracts image patches from the input tensor to form a *virtual*
tensor of shape `[batch, out_height, out_width,
filter_height * filter_width * in_channels]`.
3. For each patch, right-multiplies the filter matrix and the image patch
vector.
In detail, with the default NHWC format,
output[b, i, j, k] =
sum_{di, dj, q} input[b, strides[1] * i + di, strides[2] * j + dj, q]
* filter[di, dj, q, k]
Must have `strides[0] = strides[3] = 1`. For the most common case of the same
horizontal and vertical strides, `strides = [1, stride, stride, 1]`.
Args:
input: A `Tensor`. Must be one of the following types:
`half`, `bfloat16`, `float32`, `float64`.
A 4-D tensor. The dimension order is interpreted according to the value
of `data_format`, see below for details.
filter: A `Tensor`. Must have the same type as `input`.
A 4-D tensor of shape
`[filter_height, filter_width, in_channels, out_channels]`
strides: An int or list of `ints` that has length `1`, `2` or `4`. The
stride of the sliding window for each dimension of `input`. If a single
value is given it is replicated in the `H` and `W` dimension. By default
the `N` and `C` dimensions are set to 1. The dimension order is determined
by the value of `data_format`, see below for details.
padding: Either the `string` `""SAME""` or `""VALID""` indicating the type of
padding algorithm to use, or a list indicating the explicit paddings at
the start and end of each dimension. When explicit padding is used and
data_format is `""NHWC""`, this should be in the form `[[0, 0], [pad_top,
pad_bottom], [pad_left, pad_right], [0, 0]]`. When explicit padding used
and data_format is `""NCHW""`, this should be in the form `[[0, 0], [0, 0],
[pad_top, pad_bottom], [pad_left, pad_right]]`.
use_cudnn_on_gpu: An optional `bool`. Defaults to `True`.
data_format: An optional `string` from: `""NHWC"", ""NCHW""`.
Defaults to `""NHWC""`.
Specify the data format of the input and output data. With the
default format ""NHWC"", the data is stored in the order of:
[batch, height, width, channels].
Alternatively, the format could be ""NCHW"", the data storage order of:
[batch, channels, height, width].
dilations: An int or list of `ints` that has length `1`, `2` or `4`,
defaults to 1. The dilation factor for each dimension of`input`. If a
single value is given it is replicated in the `H` and `W` dimension. By
default the `N` and `C` dimensions are set to 1. If set to k > 1, there
will be k-1 skipped cells between each filter element on that dimension.
The dimension order is determined by the value of `data_format`, see above
for details. Dilations in the batch and depth dimensions if a 4-d tensor
must be 1.
name: A name for the operation (optional).
filters: Alias for filter.
Returns:
A `Tensor`. Has the same type as `input`."
8988,conv2d_backprop_filter,tensorflow/tensorflow/python/ops/nn_ops.py,2295,function,"Computes the gradients of convolution with respect to the filter.
Args:
input: A `Tensor`. Must be one of the following types:
`half`, `bfloat16`, `float32`, `float64`.
4-D with shape `[batch, in_height, in_width, in_channels]`.
filter_sizes: A `Tensor` of type `int32`.
An integer vector representing the tensor shape of `filter`,
where `filter` is a 4-D
`[filter_height, filter_width, in_channels, out_channels]` tensor.
out_backprop: A `Tensor`. Must have the same type as `input`.
4-D with shape `[batch, out_height, out_width, out_channels]`.
Gradients w.r.t. the output of the convolution.
strides: A list of `ints`.
The stride of the sliding window for each dimension of the input
of the convolution. Must be in the same order as the dimension specified
with format.
padding: Either the `string `""SAME""` or `""VALID""` indicating the type of
padding algorithm to use, or a list indicating the explicit paddings at
the start and end of each dimension. When explicit padding is used and
data_format is `""NHWC""`, this should be in the form `[[0, 0], [pad_top,
pad_bottom], [pad_left, pad_right], [0, 0]]`. When explicit padding used
and data_format is `""NCHW""`, this should be in the form `[[0, 0], [0, 0],
[pad_top, pad_bottom], [pad_left, pad_right]]`.
use_cudnn_on_gpu: An optional `bool`. Defaults to `True`.
data_format: An optional `string` from: `""NHWC"", ""NCHW""`.
Defaults to `""NHWC""`.
Specify the data format of the input and output data. With the
default format ""NHWC"", the data is stored in the order of:
[batch, in_height, in_width, in_channels].
Alternatively, the format could be ""NCHW"", the data storage order of:
[batch, in_channels, in_height, in_width].
dilations: An optional list of `ints`. Defaults to `[1, 1, 1, 1]`.
1-D tensor of length 4. The dilation factor for each dimension of
`input`. If set to k > 1, there will be k-1 skipped cells between each
filter element on that dimension. The dimension order is determined by
the value of `data_format`, see above for details. Dilations in the batch
and depth dimensions must be 1.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `input`."
8989,conv2d_backprop_input,tensorflow/tensorflow/python/ops/nn_ops.py,2356,function,"Computes the gradients of convolution with respect to the input.
Args:
input_sizes: A `Tensor` of type `int32`.
An integer vector representing the shape of `input`,
where `input` is a 4-D `[batch, height, width, channels]` tensor.
filter: A `Tensor`. Must be one of the following types:
`half`, `bfloat16`, `float32`, `float64`.
4-D with shape
`[filter_height, filter_width, in_channels, out_channels]`.
out_backprop: A `Tensor`. Must have the same type as `filter`.
4-D with shape `[batch, out_height, out_width, out_channels]`.
Gradients w.r.t. the output of the convolution.
strides: A list of `ints`.
The stride of the sliding window for each dimension of the input
of the convolution. Must be in the same order as the dimension specified
with format.
padding: Either the `string `""SAME""` or `""VALID""` indicating the type of
padding algorithm to use, or a list indicating the explicit paddings at
the start and end of each dimension. When explicit padding is used and
data_format is `""NHWC""`, this should be in the form `[[0, 0], [pad_top,
pad_bottom], [pad_left, pad_right], [0, 0]]`. When explicit padding used
and data_format is `""NCHW""`, this should be in the form `[[0, 0], [0, 0],
[pad_top, pad_bottom], [pad_left, pad_right]]`.
use_cudnn_on_gpu: An optional `bool`. Defaults to `True`.
data_format: An optional `string` from: `""NHWC"", ""NCHW""`.
Defaults to `""NHWC""`.
Specify the data format of the input and output data. With the
default format ""NHWC"", the data is stored in the order of:
[batch, in_height, in_width, in_channels].
Alternatively, the format could be ""NCHW"", the data storage order of:
[batch, in_channels, in_height, in_width].
dilations: An optional list of `ints`. Defaults to `[1, 1, 1, 1]`.
1-D tensor of length 4. The dilation factor for each dimension of
`input`. If set to k > 1, there will be k-1 skipped cells between each
filter element on that dimension. The dimension order is determined by
the value of `data_format`, see above for details. Dilations in the batch
and depth dimensions must be 1.
name: A name for the operation (optional).
filters: Alias for filter.
Returns:
A `Tensor`. Has the same type as `filter`."
8990,conv2d_transpose,tensorflow/tensorflow/python/ops/nn_ops.py,2421,function,"The transpose of `conv2d`.
This operation is sometimes called ""deconvolution"" after
(Zeiler et al., 2010), but is really the transpose (gradient) of `conv2d`
rather than an actual deconvolution.
Args:
value: A 4-D `Tensor` of type `float` and shape
`[batch, height, width, in_channels]` for `NHWC` data format or
`[batch, in_channels, height, width]` for `NCHW` data format.
filter: A 4-D `Tensor` with the same type as `value` and shape
`[height, width, output_channels, in_channels]`. `filter`'s
`in_channels` dimension must match that of `value`.
output_shape: A 1-D `Tensor` representing the output shape of the
deconvolution op.
strides: An int or list of `ints` that has length `1`, `2` or `4`. The
stride of the sliding window for each dimension of `input`. If a single
value is given it is replicated in the `H` and `W` dimension. By default
the `N` and `C` dimensions are set to 0. The dimension order is determined
by the value of `data_format`, see below for details.
padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm.
See the ""returns"" section of `tf.nn.convolution` for details.
data_format: A string. 'NHWC' and 'NCHW' are supported.
name: Optional name for the returned tensor.
input: Alias for value.
filters: Alias for filter.
dilations: An int or list of `ints` that has length `1`, `2` or `4`,
defaults to 1. The dilation factor for each dimension of`input`. If a
single value is given it is replicated in the `H` and `W` dimension. By
default the `N` and `C` dimensions are set to 1. If set to k > 1, there
will be k-1 skipped cells between each filter element on that dimension.
The dimension order is determined by the value of `data_format`, see above
for details. Dilations in the batch and depth dimensions if a 4-d tensor
must be 1.
Returns:
A `Tensor` with the same type as `value`.
Raises:
ValueError: If input/output depth does not match `filter`'s shape, or if
padding is other than `'VALID'` or `'SAME'`.
References:
Deconvolutional Networks:
[Zeiler et al., 2010]
(https://ieeexplore.ieee.org/abstract/document/5539957)
([pdf]
(http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.232.4023&rep=rep1&type=pdf))"
8991,conv2d_transpose_v2,tensorflow/tensorflow/python/ops/nn_ops.py,2498,function,"The transpose of `conv2d`.
This operation is sometimes called ""deconvolution"" after
(Zeiler et al., 2010), but is really the transpose (gradient) of
`atrous_conv2d` rather than an actual deconvolution.
Args:
input: A 4-D `Tensor` of type `float` and shape `[batch, height, width,
in_channels]` for `NHWC` data format or `[batch, in_channels, height,
width]` for `NCHW` data format.
filters: A 4-D `Tensor` with the same type as `input` and shape `[height,
width, output_channels, in_channels]`. `filter`'s `in_channels` dimension
must match that of `input`.
output_shape: A 1-D `Tensor` representing the output shape of the
deconvolution op.
strides: An int or list of `ints` that has length `1`, `2` or `4`. The
stride of the sliding window for each dimension of `input`. If a single
value is given it is replicated in the `H` and `W` dimension. By default
the `N` and `C` dimensions are set to 0. The dimension order is determined
by the value of `data_format`, see below for details.
padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm. See
the ""returns"" section of `tf.nn.convolution` for details.
data_format: A string. 'NHWC' and 'NCHW' are supported.
dilations: An int or list of `ints` that has length `1`, `2` or `4`,
defaults to 1. The dilation factor for each dimension of`input`. If a
single value is given it is replicated in the `H` and `W` dimension. By
default the `N` and `C` dimensions are set to 1. If set to k > 1, there
will be k-1 skipped cells between each filter element on that dimension.
The dimension order is determined by the value of `data_format`, see above
for details. Dilations in the batch and depth dimensions if a 4-d tensor
must be 1.
name: Optional name for the returned tensor.
Returns:
A `Tensor` with the same type as `input`.
Raises:
ValueError: If input/output depth does not match `filter`'s shape, or if
padding is other than `'VALID'` or `'SAME'`.
References:
Deconvolutional Networks:
[Zeiler et al., 2010]
(https://ieeexplore.ieee.org/abstract/document/5539957)
([pdf]
(http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.232.4023&rep=rep1&type=pdf))"
8992,atrous_conv2d_transpose,tensorflow/tensorflow/python/ops/nn_ops.py,2611,function,"The transpose of `atrous_conv2d`.
This operation is sometimes called ""deconvolution"" after
(Zeiler et al., 2010), but is really the transpose (gradient) of
`atrous_conv2d` rather than an actual deconvolution.
Args:
value: A 4-D `Tensor` of type `float`. It needs to be in the default `NHWC`
format. Its shape is `[batch, in_height, in_width, in_channels]`.
filters: A 4-D `Tensor` with the same type as `value` and shape
`[filter_height, filter_width, out_channels, in_channels]`. `filters`'
`in_channels` dimension must match that of `value`. Atrous convolution is
equivalent to standard convolution with upsampled filters with effective
height `filter_height + (filter_height - 1) * (rate - 1)` and effective
width `filter_width + (filter_width - 1) * (rate - 1)`, produced by
inserting `rate - 1` zeros along consecutive elements across the
`filters`' spatial dimensions.
output_shape: A 1-D `Tensor` of shape representing the output shape of the
deconvolution op.
rate: A positive int32. The stride with which we sample input values across
the `height` and `width` dimensions. Equivalently, the rate by which we
upsample the filter values by inserting zeros across the `height` and
`width` dimensions. In the literature, the same parameter is sometimes
called `input stride` or `dilation`.
padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm.
name: Optional name for the returned tensor.
Returns:
A `Tensor` with the same type as `value`.
Raises:
ValueError: If input/output depth does not match `filters`' shape, or if
padding is other than `'VALID'` or `'SAME'`, or if the `rate` is less
than one, or if the output_shape is not a tensor with 4 elements.
References:
Deconvolutional Networks:
[Zeiler et al., 2010]
(https://ieeexplore.ieee.org/abstract/document/5539957)
([pdf]
(http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.232.4023&rep=rep1&type=pdf))"
8993,depthwise_conv2d_native,tensorflow/tensorflow/python/ops/nn_ops.py,2771,function,"Computes a 2-D depthwise convolution.
Given an input tensor of shape `[batch, in_height, in_width, in_channels]`
and a filter / kernel tensor of shape
`[filter_height, filter_width, in_channels, channel_multiplier]`, containing
`in_channels` convolutional filters of depth 1, `depthwise_conv2d` applies
a different filter to each input channel (expanding from 1 channel to
`channel_multiplier` channels for each), then concatenates the results
together. Thus, the output has `in_channels * channel_multiplier` channels.
```
for k in 0..in_channels-1
for q in 0..channel_multiplier-1
output[b, i, j, k * channel_multiplier + q] =
sum_{di, dj} input[b, strides[1] * i + di, strides[2] * j + dj, k] *
filter[di, dj, k, q]
```
Must have `strides[0] = strides[3] = 1`. For the most common case of the same
horizontal and vertices strides, `strides = [1, stride, stride, 1]`.
Args:
input: A `Tensor`. Must be one of the following types: `half`, `bfloat16`,
`float32`, `float64`.
filter: A `Tensor`. Must have the same type as `input`.
strides: A list of `ints`. 1-D of length 4. The stride of the sliding
window for each dimension of `input`.
padding: Controls how to pad the image before applying the convolution. Can
be the string `""SAME""` or `""VALID""` indicating the type of padding
algorithm to use, or a list indicating the explicit paddings at the start
and end of each dimension. When explicit padding is used and data_format
is `""NHWC""`, this should be in the form `[[0, 0], [pad_top, pad_bottom],
[pad_left, pad_right], [0, 0]]`. When explicit padding used and
data_format is `""NCHW""`, this should be in the form `[[0, 0], [0, 0],
[pad_top, pad_bottom], [pad_left, pad_right]]`.
data_format: An optional `string` from: `""NHWC"", ""NCHW""`. Defaults to
`""NHWC""`. Specify the data format of the input and output data. With the
default format ""NHWC"", the data is stored in the order of: [batch, height,
width, channels].
Alternatively, the format could be ""NCHW"", the data storage order of:
[batch, channels, height, width].
dilations: An optional list of `ints`. Defaults to `[1, 1, 1, 1]`. 1-D
tensor of length 4. The dilation factor for each dimension of `input`. If
set to k > 1, there will be k-1 skipped cells between each filter element
on that dimension. The dimension order is determined by the value of
`data_format`, see above for details. Dilations in the batch and depth
dimensions must be 1.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `input`."
8994,depthwise_conv2d_native_backprop_input,tensorflow/tensorflow/python/ops/nn_ops.py,2851,function,"Computes the gradients of depthwise convolution with respect to the input.
Args:
input_sizes: A `Tensor` of type `int32`. An integer vector representing the
shape of `input`, based on `data_format`. For example, if `data_format`
is 'NHWC' then `input` is a 4-D `[batch, height, width, channels]` tensor.
filter: A `Tensor`. Must be one of the following types: `half`, `bfloat16`,
`float32`, `float64`. 4-D with shape `[filter_height, filter_width,
in_channels, depthwise_multiplier]`.
out_backprop: A `Tensor`. Must have the same type as `filter`. 4-D with
shape based on `data_format`. For example, if `data_format` is 'NHWC'
then out_backprop shape is `[batch, out_height, out_width, out_channels]`.
Gradients w.r.t. the output of the convolution.
strides: A list of `ints`. The stride of the sliding window for each
dimension of the input of the convolution.
padding: Controls how to pad the image before applying the convolution. Can
be the string `""SAME""` or `""VALID""` indicating the type of padding
algorithm to use, or a list indicating the explicit paddings at the start
and end of each dimension. When explicit padding is used and data_format
is `""NHWC""`, this should be in the form `[[0, 0], [pad_top, pad_bottom],
[pad_left, pad_right], [0, 0]]`. When explicit padding used and
data_format is `""NCHW""`, this should be in the form `[[0, 0], [0, 0],
[pad_top, pad_bottom], [pad_left, pad_right]]`.
data_format: An optional `string` from: `""NHWC"", ""NCHW""`. Defaults to
`""NHWC""`. Specify the data format of the input and output data. With the
default format ""NHWC"", the data is stored in the order of: [batch, height,
width, channels].
Alternatively, the format could be ""NCHW"", the data storage order of:
[batch, channels, height, width].
dilations: An optional list of `ints`. Defaults to `[1, 1, 1, 1]`. 1-D
tensor of length 4. The dilation factor for each dimension of `input`. If
set to k > 1, there will be k-1 skipped cells between each filter element
on that dimension. The dimension order is determined by the value of
`data_format`, see above for details. Dilations in the batch and depth
dimensions must be 1.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `filter`."
8995,depthwise_conv2d_native_backprop_filter,tensorflow/tensorflow/python/ops/nn_ops.py,2921,function,"Computes the gradients of depthwise convolution with respect to the filter.
Args:
input: A `Tensor`. Must be one of the following types: `half`, `bfloat16`,
`float32`, `float64`. 4-D with shape based on `data_format`. For example,
if `data_format` is 'NHWC' then `input` is a 4-D `[batch, in_height,
in_width, in_channels]` tensor.
filter_sizes: A `Tensor` of type `int32`. An integer vector representing the
tensor shape of `filter`, where `filter` is a 4-D `[filter_height,
filter_width, in_channels, depthwise_multiplier]` tensor.
out_backprop: A `Tensor`. Must have the same type as `input`. 4-D with shape
based on `data_format`. For example, if `data_format` is 'NHWC' then
out_backprop shape is `[batch, out_height, out_width, out_channels]`.
Gradients w.r.t. the output of the convolution.
strides: A list of `ints`. The stride of the sliding window for each
dimension of the input of the convolution.
padding: Controls how to pad the image before applying the convolution. Can
be the string `""SAME""` or `""VALID""` indicating the type of padding
algorithm to use, or a list indicating the explicit paddings at the start
and end of each dimension. When explicit padding is used and data_format
is `""NHWC""`, this should be in the form `[[0, 0], [pad_top, pad_bottom],
[pad_left, pad_right], [0, 0]]`. When explicit padding used and
data_format is `""NCHW""`, this should be in the form `[[0, 0], [0, 0],
[pad_top, pad_bottom], [pad_left, pad_right]]`.
data_format: An optional `string` from: `""NHWC"", ""NCHW""`. Defaults to
`""NHWC""`. Specify the data format of the input and output data. With the
default format ""NHWC"", the data is stored in the order of: [batch, height,
width, channels].
Alternatively, the format could be ""NCHW"", the data storage order of:
[batch, channels, height, width].
dilations: An optional list of `ints`. Defaults to `[1, 1, 1, 1]`. 1-D
tensor of length 4. The dilation factor for each dimension of `input`. If
set to k > 1, there will be k-1 skipped cells between each filter element
on that dimension. The dimension order is determined by the value of
`data_format`, see above for details. Dilations in the batch and depth
dimensions must be 1.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `input`."
8996,conv3d_v2,tensorflow/tensorflow/python/ops/nn_ops.py,3026,function,
8997,conv3d_v1,tensorflow/tensorflow/python/ops/nn_ops.py,3041,function,
8998,conv3d_transpose,tensorflow/tensorflow/python/ops/nn_ops.py,3062,function,"The transpose of `conv3d`.
This operation is sometimes called ""deconvolution"" after
(Zeiler et al., 2010), but is really the transpose (gradient) of `conv3d`
rather than an actual deconvolution.
Args:
value: A 5-D `Tensor` of type `float` and shape
`[batch, depth, height, width, in_channels]`.
filter: A 5-D `Tensor` with the same type as `value` and shape
`[depth, height, width, output_channels, in_channels]`. `filter`'s
`in_channels` dimension must match that of `value`.
output_shape: A 1-D `Tensor` representing the output shape of the
deconvolution op.
strides: A list of ints. The stride of the sliding window for each
dimension of the input tensor.
padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm.
See the ""returns"" section of `tf.nn.convolution` for details.
data_format: A string, either `'NDHWC'` or `'NCDHW`' specifying the layout
of the input and output tensors. Defaults to `'NDHWC'`.
name: Optional name for the returned tensor.
input: Alias of value.
filters: Alias of filter.
dilations: An int or list of `ints` that has length `1`, `3` or `5`,
defaults to 1. The dilation factor for each dimension of`input`. If a
single value is given it is replicated in the `D`, `H` and `W` dimension.
By default the `N` and `C` dimensions are set to 1. If set to k > 1, there
will be k-1 skipped cells between each filter element on that dimension.
The dimension order is determined by the value of `data_format`, see above
for details. Dilations in the batch and depth dimensions if a 5-d tensor
must be 1.
Returns:
A `Tensor` with the same type as `value`.
Raises:
ValueError: If input/output depth does not match `filter`'s shape, or if
padding is other than `'VALID'` or `'SAME'`.
References:
Deconvolutional Networks:
[Zeiler et al., 2010]
(https://ieeexplore.ieee.org/abstract/document/5539957)
([pdf]
(http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.232.4023&rep=rep1&type=pdf))"
8999,conv3d_transpose_v2,tensorflow/tensorflow/python/ops/nn_ops.py,3134,function,"The transpose of `conv3d`.
This operation is sometimes called ""deconvolution"" after
(Zeiler et al., 2010), but is really the transpose (gradient) of `conv3d`
rather than an actual deconvolution.
Args:
input: A 5-D `Tensor` of type `float` and shape `[batch, depth, height,
width, in_channels]` for `NDHWC` data format or `[batch, in_channels,
depth, height, width]` for `NCDHW` data format.
filters: A 5-D `Tensor` with the same type as `input` and shape `[depth,
height, width, output_channels, in_channels]`. `filter`'s `in_channels`
dimension must match that of `input`.
output_shape: A 1-D `Tensor` representing the output shape of the
deconvolution op.
strides: An int or list of `ints` that has length `1`, `3` or `5`. The
stride of the sliding window for each dimension of `input`. If a single
value is given it is replicated in the `D`, `H` and `W` dimension. By
default the `N` and `C` dimensions are set to 0. The dimension order is
determined by the value of `data_format`, see below for details.
padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm. See
the ""returns"" section of `tf.nn.convolution` for details.
data_format: A string. 'NDHWC' and 'NCDHW' are supported.
dilations: An int or list of `ints` that has length `1`, `3` or `5`,
defaults to 1. The dilation factor for each dimension of`input`. If a
single value is given it is replicated in the `D`, `H` and `W` dimension.
By default the `N` and `C` dimensions are set to 1. If set to k > 1, there
will be k-1 skipped cells between each filter element on that dimension.
The dimension order is determined by the value of `data_format`, see above
for details. Dilations in the batch and depth dimensions if a 5-d tensor
must be 1.
name: Optional name for the returned tensor.
Returns:
A `Tensor` with the same type as `input`.
References:
Deconvolutional Networks:
[Zeiler et al., 2010]
(https://ieeexplore.ieee.org/abstract/document/5539957)
([pdf]
(http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.232.4023&rep=rep1&type=pdf))"
9000,conv_transpose,tensorflow/tensorflow/python/ops/nn_ops.py,3214,function,"The transpose of `convolution`.
This operation is sometimes called ""deconvolution"" after
(Zeiler et al., 2010), but is really the transpose (gradient) of `conv3d`
rather than an actual deconvolution.
Args:
input: An N+2 dimensional `Tensor` of shape
`[batch_size] + input_spatial_shape + [in_channels]` if data_format does
not start with ""NC"" (default), or
`[batch_size, in_channels] + input_spatial_shape` if data_format starts
with ""NC"". It must be one of the following types:
`half`, `bfloat16`, `float32`, `float64`.
filters: An N+2 dimensional `Tensor` with the same type as `input` and
shape `spatial_filter_shape + [in_channels, out_channels]`.
output_shape: A 1-D `Tensor` representing the output shape of the
deconvolution op.
strides: An int or list of `ints` that has length `1`, `N` or `N+2`. The
stride of the sliding window for each dimension of `input`. If a single
value is given it is replicated in the spatial dimensions. By default
the `N` and `C` dimensions are set to 0. The dimension order is determined
by the value of `data_format`, see below for details.
padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm. See
the ""returns"" section of `tf.nn.convolution` for details.
data_format: A string or None. Specifies whether the channel dimension of
the `input` and output is the last dimension (default, or if `data_format`
does not start with ""NC""), or the second dimension (if `data_format`
starts with ""NC""). For N=1, the valid values are ""NWC"" (default) and
""NCW"". For N=2, the valid values are ""NHWC"" (default) and ""NCHW"".
For N=3, the valid values are ""NDHWC"" (default) and ""NCDHW"".
dilations: An int or list of `ints` that has length `1`, `N` or `N+2`,
defaults to 1. The dilation factor for each dimension of`input`. If a
single value is given it is replicated in the spatial dimensions. By
default the `N` and `C` dimensions are set to 1. If set to k > 1, there
will be k-1 skipped cells between each filter element on that dimension.
The dimension order is determined by the value of `data_format`, see above
for details.
name: A name for the operation (optional). If not specified ""conv_transpose""
is used.
Returns:
A `Tensor` with the same type as `value`.
References:
Deconvolutional Networks:
[Zeiler et al., 2010]
(https://ieeexplore.ieee.org/abstract/document/5539957)
([pdf]
(http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.232.4023&rep=rep1&type=pdf))"
9001,bias_add,tensorflow/tensorflow/python/ops/nn_ops.py,3312,function,"Adds `bias` to `value`.
This is (mostly) a special case of `tf.add` where `bias` is restricted to 1-D.
Broadcasting is supported, so `value` may have any number of dimensions.
Unlike `tf.add`, the type of `bias` is allowed to differ from `value` in the
case where both types are quantized.
Args:
value: A `Tensor` with type `float`, `double`, `int64`, `int32`, `uint8`,
`int16`, `int8`, `complex64`, or `complex128`.
bias: A 1-D `Tensor` with size matching the channel dimension of `value`.
Must be the same type as `value` unless `value` is a quantized type,
in which case a different quantized type may be used.
data_format: A string. 'N...C' and 'NC...' are supported. If `None` (the
default) is specified then 'N..C' is assumed.
name: A name for the operation (optional).
Returns:
A `Tensor` with the same type as `value`.
Raises:
ValueError if data format is unrecognized, if `value` has less than two
dimensions when `data_format` is 'N..C'/`None` or `value` has less
then three dimensions when `data_format` is `NC..`, if `bias` does not
have exactly one dimension (is a vector), or if the size of `bias`
does not match the size of the channel dimension of `value`."
9002,bias_add_v1,tensorflow/tensorflow/python/ops/nn_ops.py,3373,function,"Adds `bias` to `value`.
This is a deprecated version of bias_add and will soon to be removed.
This is (mostly) a special case of `tf.add` where `bias` is restricted to 1-D.
Broadcasting is supported, so `value` may have any number of dimensions.
Unlike `tf.add`, the type of `bias` is allowed to differ from `value` in the
case where both types are quantized.
Args:
value: A `Tensor` with type `float`, `double`, `int64`, `int32`, `uint8`,
`int16`, `int8`, `complex64`, or `complex128`.
bias: A 1-D `Tensor` with size matching the last dimension of `value`.
Must be the same type as `value` unless `value` is a quantized type,
in which case a different quantized type may be used.
name: A name for the operation (optional).
Returns:
A `Tensor` with the same type as `value`."
9003,crelu,tensorflow/tensorflow/python/ops/nn_ops.py,3402,function,"Computes Concatenated ReLU.
Concatenates a ReLU which selects only the positive part of the activation
with a ReLU which selects only the *negative* part of the activation.
Note that as a result this non-linearity doubles the depth of the activations.
Source: [Understanding and Improving Convolutional Neural Networks via
Concatenated Rectified Linear Units. W. Shang, et
al.](https://arxiv.org/abs/1603.05201)
Args:
features: A `Tensor` with type `float`, `double`, `int32`, `int64`, `uint8`,
`int16`, or `int8`.
name: A name for the operation (optional).
axis: The axis that the output values are concatenated along. Default is -1.
Returns:
A `Tensor` with the same type as `features`.
References:
Understanding and Improving Convolutional Neural Networks via Concatenated
Rectified Linear Units:
[Shang et al., 2016](http://proceedings.mlr.press/v48/shang16)
([pdf](http://proceedings.mlr.press/v48/shang16.pdf))"
9004,crelu_v2,tensorflow/tensorflow/python/ops/nn_ops.py,3435,function,
9005,relu6,tensorflow/tensorflow/python/ops/nn_ops.py,3442,function,"Computes Rectified Linear 6: `min(max(features, 0), 6)`.
Args:
features: A `Tensor` with type `float`, `double`, `int32`, `int64`, `uint8`,
`int16`, or `int8`.
name: A name for the operation (optional).
Returns:
A `Tensor` with the same type as `features`.
References:
Convolutional Deep Belief Networks on CIFAR-10:
Krizhevsky et al., 2010
([pdf](http://www.cs.utoronto.ca/~kriz/conv-cifar10-aug2010.pdf))"
9006,leaky_relu,tensorflow/tensorflow/python/ops/nn_ops.py,3465,function,"Compute the Leaky ReLU activation function.
Source: [Rectifier Nonlinearities Improve Neural Network Acoustic Models.
AL Maas, AY Hannun, AY Ng - Proc. ICML, 2013]
(https://ai.stanford.edu/~amaas/papers/relu_hybrid_icml2013_final.pdf).
Args:
features: A `Tensor` representing preactivation values. Must be one of
the following types: `float16`, `float32`, `float64`, `int32`, `int64`.
alpha: Slope of the activation function at x < 0.
name: A name for the operation (optional).
Returns:
The activation value.
References:
Rectifier Nonlinearities Improve Neural Network Acoustic Models:
[Maas et al., 2013]
(http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.693.1422)
([pdf]
(http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.693.1422&rep=rep1&type=pdf))"
9007,softmax,tensorflow/tensorflow/python/ops/nn_ops.py,3605,function,"Computes softmax activations.
This function performs the equivalent of
softmax = tf.exp(logits) / tf.reduce_sum(tf.exp(logits), axis)
See: https://en.wikipedia.org/wiki/Softmax_function
Example usage:
>>> tf.nn.softmax([-1, 0., 1.])
<tf.Tensor: shape=(3,), dtype=float32,
numpy=array([0.09003057, 0.24472848, 0.66524094], dtype=float32)>
Args:
logits: A non-empty `Tensor`, or an object whose type has a registered
`Tensor` conversion function. Must be one of the following types:
`half`,`float32`, `float64`. See also `convert_to_tensor`
axis: The dimension softmax would be performed on. The default is -1 which
indicates the last dimension.
name: A name for the operation (optional).
dim: Deprecated alias for `axis`.
Returns:
A `Tensor`. Has the same type and shape as `logits`.
Raises:
InvalidArgumentError: if `logits` is empty or `axis` is beyond the last
dimension of `logits`.
TypeError: If no conversion function is registered for `logits` to
Tensor.
RuntimeError: If a registered conversion function returns an invalid
value."
9008,softmax_v2,tensorflow/tensorflow/python/ops/nn_ops.py,3649,function,"Computes softmax activations.
This function performs the equivalent of
softmax = tf.exp(logits) / tf.reduce_sum(tf.exp(logits), axis)
Args:
logits: A non-empty `Tensor`. Must be one of the following types: `half`,
`float32`, `float64`.
axis: The dimension softmax would be performed on. The default is -1 which
indicates the last dimension.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type and shape as `logits`.
Raises:
InvalidArgumentError: if `logits` is empty or `axis` is beyond the last
dimension of `logits`."
9009,log_softmax,tensorflow/tensorflow/python/ops/nn_ops.py,3678,function,"Computes log softmax activations.
For each batch `i` and class `j` we have
logsoftmax = logits - log(reduce_sum(exp(logits), axis))
Args:
logits: A non-empty `Tensor`. Must be one of the following types: `half`,
`float32`, `float64`.
axis: The dimension softmax would be performed on. The default is -1 which
indicates the last dimension.
name: A name for the operation (optional).
dim: Deprecated alias for `axis`.
Returns:
A `Tensor`. Has the same type as `logits`. Same shape as `logits`.
Raises:
InvalidArgumentError: if `logits` is empty or `axis` is beyond the last
dimension of `logits`."
9010,log_softmax_v2,tensorflow/tensorflow/python/ops/nn_ops.py,3708,function,"Computes log softmax activations.
For each batch `i` and class `j` we have
logsoftmax = logits - log(reduce_sum(exp(logits), axis))
Args:
logits: A non-empty `Tensor`. Must be one of the following types: `half`,
`float32`, `float64`.
axis: The dimension softmax would be performed on. The default is -1 which
indicates the last dimension.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `logits`. Same shape as `logits`.
Raises:
InvalidArgumentError: if `logits` is empty or `axis` is beyond the last
dimension of `logits`."
9011,softmax_cross_entropy_with_logits_v2,tensorflow/tensorflow/python/ops/nn_ops.py,3745,function,"Computes softmax cross entropy between `logits` and `labels`.
Measures the probability error in discrete classification tasks in which the
classes are mutually exclusive (each entry is in exactly one class). For
example, each CIFAR-10 image is labeled with one and only one label: an image
can be a dog or a truck, but not both.
**NOTE:** While the classes are mutually exclusive, their probabilities
need not be. All that is required is that each row of `labels` is
a valid probability distribution. If they are not, the computation of the
gradient will be incorrect.
If using exclusive `labels` (wherein one and only
one class is true at a time), see `sparse_softmax_cross_entropy_with_logits`.
Usage:
>>> logits = [[4.0, 2.0, 1.0], [0.0, 5.0, 1.0]]
>>> labels = [[1.0, 0.0, 0.0], [0.0, 0.8, 0.2]]
>>> tf.nn.softmax_cross_entropy_with_logits(labels=labels, logits=logits)
<tf.Tensor: shape=(2,), dtype=float32,
numpy=array([0.16984604, 0.82474494], dtype=float32)>
**WARNING:** This op expects unscaled logits, since it performs a `softmax`
on `logits` internally for efficiency. Do not call this op with the
output of `softmax`, as it will produce incorrect results.
A common use case is to have logits and labels of shape
`[batch_size, num_classes]`, but higher dimensions are supported, with
the `axis` argument specifying the class dimension.
`logits` and `labels` must have the same dtype (either `float16`, `float32`,
or `float64`).
Backpropagation will happen into both `logits` and `labels`. To disallow
backpropagation into `labels`, pass label tensors through `tf.stop_gradient`
before feeding it to this function.
**Note that to avoid confusion, it is required to pass only named arguments to
this function.**
Args:
labels: Each vector along the class dimension should hold a valid
probability distribution e.g. for the case in which labels are of shape
`[batch_size, num_classes]`, each row of `labels[i]` must be a valid
probability distribution.
logits: Per-label activations, typically a linear output. These activation
energies are interpreted as unnormalized log probabilities.
axis: The class dimension. Defaulted to -1 which is the last dimension.
name: A name for the operation (optional).
Returns:
A `Tensor` that contains the softmax cross entropy loss. Its type is the
same as `logits` and its shape is the same as `labels` except that it does
not have the last dimension of `labels`."
9012,softmax_cross_entropy_with_logits_v2_helper,tensorflow/tensorflow/python/ops/nn_ops.py,3809,function,"Computes softmax cross entropy between `logits` and `labels`.
Measures the probability error in discrete classification tasks in which the
classes are mutually exclusive (each entry is in exactly one class). For
example, each CIFAR-10 image is labeled with one and only one label: an image
can be a dog or a truck, but not both.
**NOTE:** While the classes are mutually exclusive, their probabilities
need not be. All that is required is that each row of `labels` is
a valid probability distribution. If they are not, the computation of the
gradient will be incorrect.
If using exclusive `labels` (wherein one and only
one class is true at a time), see `sparse_softmax_cross_entropy_with_logits`.
**WARNING:** This op expects unscaled logits, since it performs a `softmax`
on `logits` internally for efficiency. Do not call this op with the
output of `softmax`, as it will produce incorrect results.
A common use case is to have logits and labels of shape
`[batch_size, num_classes]`, but higher dimensions are supported, with
the `axis` argument specifying the class dimension.
`logits` and `labels` must have the same dtype (either `float16`, `float32`,
or `float64`).
Backpropagation will happen into both `logits` and `labels`. To disallow
backpropagation into `labels`, pass label tensors through `tf.stop_gradient`
before feeding it to this function.
**Note that to avoid confusion, it is required to pass only named arguments to
this function.**
Args:
labels: Each vector along the class dimension should hold a valid
probability distribution e.g. for the case in which labels are of shape
`[batch_size, num_classes]`, each row of `labels[i]` must be a valid
probability distribution.
logits: Unscaled log probabilities.
axis: The class dimension. Defaulted to -1 which is the last dimension.
name: A name for the operation (optional).
dim: Deprecated alias for axis.
Returns:
A `Tensor` that contains the softmax cross entropy loss. Its type is the
same as `logits` and its shape is the same as `labels` except that it does
not have the last dimension of `labels`."
9013,softmax_cross_entropy_with_logits,tensorflow/tensorflow/python/ops/nn_ops.py,3937,function,"Computes softmax cross entropy between `logits` and `labels`.
Measures the probability error in discrete classification tasks in which the
classes are mutually exclusive (each entry is in exactly one class). For
example, each CIFAR-10 image is labeled with one and only one label: an image
can be a dog or a truck, but not both.
**NOTE:** While the classes are mutually exclusive, their probabilities
need not be. All that is required is that each row of `labels` is
a valid probability distribution. If they are not, the computation of the
gradient will be incorrect.
If using exclusive `labels` (wherein one and only
one class is true at a time), see `sparse_softmax_cross_entropy_with_logits`.
**WARNING:** This op expects unscaled logits, since it performs a `softmax`
on `logits` internally for efficiency. Do not call this op with the
output of `softmax`, as it will produce incorrect results.
A common use case is to have logits and labels of shape
`[batch_size, num_classes]`, but higher dimensions are supported, with
the `dim` argument specifying the class dimension.
Backpropagation will happen only into `logits`. To calculate a cross entropy
loss that allows backpropagation into both `logits` and `labels`, see
`tf.nn.softmax_cross_entropy_with_logits_v2`.
**Note that to avoid confusion, it is required to pass only named arguments to
this function.**
Args:
_sentinel: Used to prevent positional parameters. Internal, do not use.
labels: Each vector along the class dimension should hold a valid
probability distribution e.g. for the case in which labels are of shape
`[batch_size, num_classes]`, each row of `labels[i]` must be a valid
probability distribution.
logits: Per-label activations, typically a linear output. These activation
energies are interpreted as unnormalized log probabilities.
dim: The class dimension. Defaulted to -1 which is the last dimension.
name: A name for the operation (optional).
axis: Alias for dim.
Returns:
A `Tensor` that contains the softmax cross entropy loss. Its type is the
same as `logits` and its shape is the same as `labels` except that it does
not have the last dimension of `labels`."
9014,sparse_softmax_cross_entropy_with_logits,tensorflow/tensorflow/python/ops/nn_ops.py,4005,function,"Computes sparse softmax cross entropy between `logits` and `labels`.
Measures the probability error in discrete classification tasks in which the
classes are mutually exclusive (each entry is in exactly one class). For
example, each CIFAR-10 image is labeled with one and only one label: an image
can be a dog or a truck, but not both.
**NOTE:** For this operation, the probability of a given label is considered
exclusive. That is, soft classes are not allowed, and the `labels` vector
must provide a single specific index for the true class for each row of
`logits` (each minibatch entry). For soft softmax classification with
a probability distribution for each entry, see
`softmax_cross_entropy_with_logits_v2`.
**WARNING:** This op expects unscaled logits, since it performs a `softmax`
on `logits` internally for efficiency. Do not call this op with the
output of `softmax`, as it will produce incorrect results.
A common use case is to have logits of shape
`[batch_size, num_classes]` and have labels of shape
`[batch_size]`, but higher dimensions are supported, in which
case the `dim`-th dimension is assumed to be of size `num_classes`.
`logits` must have the dtype of `float16`, `float32`, or `float64`, and
`labels` must have the dtype of `int32` or `int64`.
**Note that to avoid confusion, it is required to pass only named arguments to
this function.**
Args:
_sentinel: Used to prevent positional parameters. Internal, do not use.
labels: `Tensor` of shape `[d_0, d_1, ..., d_{r-1}]` (where `r` is rank of
`labels` and result) and dtype `int32` or `int64`. Each entry in `labels`
must be an index in `[0, num_classes)`. Other values will raise an
exception when this op is run on CPU, and return `NaN` for corresponding
loss and gradient rows on GPU.
logits: Per-label activations (typically a linear output) of shape
`[d_0, d_1, ..., d_{r-1}, num_classes]` and dtype `float16`, `float32`, or
`float64`. These activation energies are interpreted as unnormalized log
probabilities.
name: A name for the operation (optional).
Returns:
A `Tensor` of the same shape as `labels` and of the same type as `logits`
with the softmax cross entropy loss.
Raises:
ValueError: If logits are scalars (need to have rank >= 1) or if the rank
of the labels is not equal to the rank of the logits minus one."
9015,sparse_softmax_cross_entropy_with_logits_v2,tensorflow/tensorflow/python/ops/nn_ops.py,4131,function,"Computes sparse softmax cross entropy between `logits` and `labels`.
Measures the probability error in discrete classification tasks in which the
classes are mutually exclusive (each entry is in exactly one class). For
example, each CIFAR-10 image is labeled with one and only one label: an image
can be a dog or a truck, but not both.
**NOTE:** For this operation, the probability of a given label is considered
exclusive. That is, soft classes are not allowed, and the `labels` vector
must provide a single specific index for the true class for each row of
`logits` (each minibatch entry). For soft softmax classification with
a probability distribution for each entry, see
`softmax_cross_entropy_with_logits_v2`.
**WARNING:** This op expects unscaled logits, since it performs a `softmax`
on `logits` internally for efficiency. Do not call this op with the
output of `softmax`, as it will produce incorrect results.
A common use case is to have logits of shape
`[batch_size, num_classes]` and have labels of shape
`[batch_size]`, but higher dimensions are supported, in which
case the `dim`-th dimension is assumed to be of size `num_classes`.
`logits` must have the dtype of `float16`, `float32`, or `float64`, and
`labels` must have the dtype of `int32` or `int64`.
**Note that to avoid confusion, it is required to pass only named arguments to
this function.**
Args:
labels: `Tensor` of shape `[d_0, d_1, ..., d_{r-1}]` (where `r` is rank of
`labels` and result) and dtype `int32` or `int64`. Each entry in `labels`
must be an index in `[0, num_classes)`. Other values will raise an
exception when this op is run on CPU, and return `NaN` for corresponding
loss and gradient rows on GPU.
logits: Unscaled log probabilities of shape `[d_0, d_1, ..., d_{r-1},
num_classes]` and dtype `float16`, `float32`, or `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` of the same shape as `labels` and of the same type as `logits`
with the softmax cross entropy loss.
Raises:
ValueError: If logits are scalars (need to have rank >= 1) or if the rank
of the labels is not equal to the rank of the logits minus one."
9016,avg_pool_v2,tensorflow/tensorflow/python/ops/nn_ops.py,4184,function,"Performs the avg pooling on the input.
Each entry in `output` is the mean of the corresponding size `ksize`
window in `value`.
Args:
input: Tensor of rank N+2, of shape `[batch_size] + input_spatial_shape +
[num_channels]` if `data_format` does not start with ""NC"" (default), or
`[batch_size, num_channels] + input_spatial_shape` if data_format starts
with ""NC"". Pooling happens over the spatial dimensions only.
ksize: An int or list of `ints` that has length `1`, `N` or `N+2`. The size
of the window for each dimension of the input tensor.
strides: An int or list of `ints` that has length `1`, `N` or `N+2`. The
stride of the sliding window for each dimension of the input tensor.
padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm. See
the ""returns"" section of `tf.nn.convolution` for details.
data_format: A string. Specifies the channel dimension. For N=1 it can be
either ""NWC"" (default) or ""NCW"", for N=2 it can be either ""NHWC"" (default)
or ""NCHW"" and for N=3 either ""NDHWC"" (default) or ""NCDHW"".
name: Optional name for the operation.
Returns:
A `Tensor` of format specified by `data_format`.
The average pooled output tensor."
9017,avg_pool,tensorflow/tensorflow/python/ops/nn_ops.py,4247,function,"Performs the average pooling on the input.
Each entry in `output` is the mean of the corresponding size `ksize`
window in `value`.
Args:
value: A 4-D `Tensor` of shape `[batch, height, width, channels]` and type
`float32`, `float64`, `qint8`, `quint8`, or `qint32`.
ksize: An int or list of `ints` that has length `1`, `2` or `4`. The size of
the window for each dimension of the input tensor.
strides: An int or list of `ints` that has length `1`, `2` or `4`. The
stride of the sliding window for each dimension of the input tensor.
padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm.
See the ""returns"" section of `tf.nn.convolution` for details.
data_format: A string. 'NHWC' and 'NCHW' are supported.
name: Optional name for the operation.
input: Alias for value.
Returns:
A `Tensor` with the same type as `value`. The average pooled output tensor."
9018,avg_pool2d,tensorflow/tensorflow/python/ops/nn_ops.py,4292,function,"Performs the average pooling on the input.
Each entry in `output` is the mean of the corresponding size `ksize`
window in `value`.
Args:
input: A 4-D `Tensor` of shape `[batch, height, width, channels]` and type
`float32`, `float64`, `qint8`, `quint8`, or `qint32`.
ksize: An int or list of `ints` that has length `1`, `2` or `4`. The size of
the window for each dimension of the input tensor.
strides: An int or list of `ints` that has length `1`, `2` or `4`. The
stride of the sliding window for each dimension of the input tensor.
padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm.
See the ""returns"" section of `tf.nn.convolution` for details.
data_format: A string. 'NHWC' and 'NCHW' are supported.
name: Optional name for the operation.
Returns:
A `Tensor` with the same type as `value`. The average pooled output tensor."
9019,avg_pool1d,tensorflow/tensorflow/python/ops/nn_ops.py,4332,function,"Performs the average pooling on the input.
Each entry in `output` is the mean of the corresponding size `ksize`
window in `value`.
Note internally this op reshapes and uses the underlying 2d operation.
Args:
input: A 3-D `Tensor` of the format specified by `data_format`.
ksize: An int or list of `ints` that has length `1` or `3`. The size of the
window for each dimension of the input tensor.
strides: An int or list of `ints` that has length `1` or `3`. The stride of
the sliding window for each dimension of the input tensor.
padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm. See
the ""returns"" section of `tf.nn.convolution` for details.
data_format: An optional string from: ""NWC"", ""NCW"". Defaults to ""NWC"".
name: A name for the operation (optional).
Returns:
A `Tensor` of format specified by `data_format`.
The max pooled output tensor."
9020,avg_pool3d,tensorflow/tensorflow/python/ops/nn_ops.py,4378,function,"Performs the average pooling on the input.
Each entry in `output` is the mean of the corresponding size `ksize`
window in `value`.
Args:
input: A 5-D `Tensor` of shape `[batch, height, width, channels]` and type
`float32`, `float64`, `qint8`, `quint8`, or `qint32`.
ksize: An int or list of `ints` that has length `1`, `3` or `5`. The size of
the window for each dimension of the input tensor.
strides: An int or list of `ints` that has length `1`, `3` or `5`. The
stride of the sliding window for each dimension of the input tensor.
padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm.
See the ""returns"" section of `tf.nn.convolution` for details.
data_format: A string. 'NDHWC' and 'NCDHW' are supported.
name: Optional name for the operation.
Returns:
A `Tensor` with the same type as `value`. The average pooled output tensor."
9021,max_pool_v2,tensorflow/tensorflow/python/ops/nn_ops.py,4419,function,"Performs the max pooling on the input.
Args:
input: Tensor of rank N+2, of shape `[batch_size] + input_spatial_shape +
[num_channels]` if `data_format` does not start with ""NC"" (default), or
`[batch_size, num_channels] + input_spatial_shape` if data_format starts
with ""NC"". Pooling happens over the spatial dimensions only.
ksize: An int or list of `ints` that has length `1`, `N` or `N+2`. The size
of the window for each dimension of the input tensor.
strides: An int or list of `ints` that has length `1`, `N` or `N+2`. The
stride of the sliding window for each dimension of the input tensor.
padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm. See
the ""returns"" section of `tf.nn.convolution` for details.
data_format: A string. Specifies the channel dimension. For N=1 it can be
either ""NWC"" (default) or ""NCW"", for N=2 it can be either ""NHWC"" (default)
or ""NCHW"" and for N=3 either ""NDHWC"" (default) or ""NCDHW"".
name: Optional name for the operation.
Returns:
A `Tensor` of format specified by `data_format`.
The max pooled output tensor."
9022,max_pool,tensorflow/tensorflow/python/ops/nn_ops.py,4480,function,"Performs the max pooling on the input.
Args:
value: A 4-D `Tensor` of the format specified by `data_format`.
ksize: An int or list of `ints` that has length `1`, `2` or `4`.
The size of the window for each dimension of the input tensor.
strides: An int or list of `ints` that has length `1`, `2` or `4`.
The stride of the sliding window for each dimension of the input tensor.
padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm.
See the ""returns"" section of `tf.nn.convolution` for details.
data_format: A string. 'NHWC', 'NCHW' and 'NCHW_VECT_C' are supported.
name: Optional name for the operation.
input: Alias for value.
Returns:
A `Tensor` of format specified by `data_format`.
The max pooled output tensor."
9023,max_pool1d,tensorflow/tensorflow/python/ops/nn_ops.py,4530,function,"Performs the max pooling on the input.
Note internally this op reshapes and uses the underlying 2d operation.
Args:
input: A 3-D `Tensor` of the format specified by `data_format`.
ksize: An int or list of `ints` that has length `1` or `3`. The size of the
window for each dimension of the input tensor.
strides: An int or list of `ints` that has length `1` or `3`. The stride of
the sliding window for each dimension of the input tensor.
padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm. See
the ""returns"" section of `tf.nn.convolution` for details.
data_format: An optional string from: ""NWC"", ""NCW"". Defaults to ""NWC"".
name: A name for the operation (optional).
Returns:
A `Tensor` of format specified by `data_format`.
The max pooled output tensor."
9024,max_pool2d,tensorflow/tensorflow/python/ops/nn_ops.py,4575,function,"Performs the max pooling on the input.
Args:
input: A 4-D `Tensor` of the format specified by `data_format`.
ksize: An int or list of `ints` that has length `1`, `2` or `4`. The size of
the window for each dimension of the input tensor.
strides: An int or list of `ints` that has length `1`, `2` or `4`. The
stride of the sliding window for each dimension of the input tensor.
padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm. See
the ""returns"" section of `tf.nn.convolution` for details.
data_format: A string. 'NHWC', 'NCHW' and 'NCHW_VECT_C' are supported.
name: Optional name for the operation.
Returns:
A `Tensor` of format specified by `data_format`.
The max pooled output tensor."
9025,max_pool3d,tensorflow/tensorflow/python/ops/nn_ops.py,4614,function,"Performs the max pooling on the input.
Args:
input: A 5-D `Tensor` of the format specified by `data_format`.
ksize: An int or list of `ints` that has length `1`, `3` or `5`. The size of
the window for each dimension of the input tensor.
strides: An int or list of `ints` that has length `1`, `3` or `5`. The
stride of the sliding window for each dimension of the input tensor.
padding: A string, either `'VALID'` or `'SAME'`. The padding algorithm. See
the ""returns"" section of `tf.nn.convolution` for details.
data_format: An optional string from: ""NDHWC"", ""NCDHW"". Defaults to ""NDHWC"".
The data format of the input and output data. With the default format
""NDHWC"", the data is stored in the order of: [batch, in_depth, in_height,
in_width, in_channels]. Alternatively, the format could be ""NCDHW"", the
data storage order is: [batch, in_channels, in_depth, in_height,
in_width].
name: A name for the operation (optional).
Returns:
A `Tensor` of format specified by `data_format`.
The max pooled output tensor."
9026,max_pool_with_argmax_v2,tensorflow/tensorflow/python/ops/nn_ops.py,4657,function,"Performs max pooling on the input and outputs both max values and indices.
The indices in `argmax` are flattened, so that a maximum value at position
`[b, y, x, c]` becomes flattened index: `(y * width + x) * channels + c` if
`include_batch_in_index` is False;
`((b * height + y) * width + x) * channels + c`
if `include_batch_in_index` is True.
The indices returned are always in `[0, height) x [0, width)` before
flattening, even if padding is involved and the mathematically correct answer
is outside (either negative or too large). This is a bug, but fixing it is
difficult to do in a safe backwards compatible way, especially due to
flattening.
Args:
input: A `Tensor`. Must be one of the following types: `float32`, `float64`,
`int32`, `uint8`, `int16`, `int8`, `int64`, `bfloat16`, `uint16`, `half`,
`uint32`, `uint64`.
4-D with shape `[batch, height, width, channels]`. Input to pool over.
ksize: An int or list of `ints` that has length `1`, `2` or `4`.
The size of the window for each dimension of the input tensor.
strides: An int or list of `ints` that has length `1`, `2` or `4`.
The stride of the sliding window for each dimension of the
input tensor.
padding: A `string` from: `""SAME"", ""VALID""`.
The type of padding algorithm to use.
data_format: An optional `string`, must be set to `""NHWC""`. Defaults to
`""NHWC""`.
Specify the data format of the input and output data.
output_dtype: An optional `tf.DType` from: `tf.int32, tf.int64`.
Defaults to `tf.int64`.
The dtype of the returned argmax tensor.
include_batch_in_index: An optional `boolean`. Defaults to `False`.
Whether to include batch dimension in flattened index of `argmax`.
name: A name for the operation (optional).
Returns:
A tuple of `Tensor` objects (output, argmax).
output: A `Tensor`. Has the same type as `input`.
argmax: A `Tensor` of type `output_dtype`."
9027,max_pool_with_argmax_v1,tensorflow/tensorflow/python/ops/nn_ops.py,4727,function,
9028,xw_plus_b,tensorflow/tensorflow/python/ops/nn_ops.py,4822,function,"Computes matmul(x, weights) + biases.
Args:
x: a 2D tensor. Dimensions typically: batch, in_units
weights: a 2D tensor. Dimensions typically: in_units, out_units
biases: a 1D tensor. Dimensions: out_units
name: A name for the operation (optional). If not specified
""xw_plus_b"" is used.
Returns:
A 2-D Tensor computing matmul(x, weights) + biases.
Dimensions typically: batch, out_units."
9029,xw_plus_b_v1,tensorflow/tensorflow/python/ops/nn_ops.py,4844,function,"Computes matmul(x, weights) + biases.
This is a deprecated version of that will soon be removed.
Args:
x: a 2D tensor. Dimensions typically: batch, in_units
weights: a 2D tensor. Dimensions typically: in_units, out_units
biases: a 1D tensor. Dimensions: out_units
name: A name for the operation (optional). If not specified
""xw_plus_b_v1"" is used.
Returns:
A 2-D Tensor computing matmul(x, weights) + biases.
Dimensions typically: batch, out_units."
9030,dropout,tensorflow/tensorflow/python/ops/nn_ops.py,4898,function,"Computes dropout.
For each element of `x`, with probability `rate`, outputs `0`, and otherwise
scales up the input by `1 / (1-rate)`. The scaling is such that the expected
sum is unchanged.
By default, each element is kept or dropped independently. If `noise_shape`
is specified, it must be
[broadcastable](http://docs.scipy.org/doc/numpy/user/basics.broadcasting.html)
to the shape of `x`, and only dimensions with `noise_shape[i] == shape(x)[i]`
will make independent decisions. For example, if `shape(x) = [k, l, m, n]`
and `noise_shape = [k, 1, 1, n]`, each batch and channel component will be
kept independently and each row and column will be kept or not kept together.
Args:
x: A floating point tensor.
keep_prob: (deprecated) A deprecated alias for `(1-rate)`.
noise_shape: A 1-D `Tensor` of type `int32`, representing the
shape for randomly generated keep/drop flags.
seed: A Python integer. Used to create random seeds. See
`tf.random.set_seed` for behavior.
name: A name for this operation (optional).
rate: A scalar `Tensor` with the same type as `x`. The probability that each
element of `x` is discarded.
Returns:
A Tensor of the same shape of `x`.
Raises:
ValueError: If `rate` is not in `[0, 1)` or if `x` is not a floating
point tensor."
9031,dropout_v2,tensorflow/tensorflow/python/ops/nn_ops.py,4950,function,"Computes dropout: randomly sets elements to zero to prevent overfitting.
Note: The behavior of dropout has changed between TensorFlow 1.x and 2.x.
When converting 1.x code, please use named arguments to ensure behavior stays
consistent.
See also: `tf.keras.layers.Dropout` for a dropout layer.
[Dropout](https://arxiv.org/abs/1207.0580) is useful for regularizing DNN
models. Inputs elements are randomly set to zero (and the other elements are
rescaled). This encourages each node to be independently useful, as it cannot
rely on the output of other nodes.
More precisely: With probability `rate` elements of `x` are set to `0`.
The remaining elements are scaled up by `1.0 / (1 - rate)`, so that the
expected value is preserved.
>>> tf.random.set_seed(0)
>>> x = tf.ones([3,5])
>>> tf.nn.dropout(x, rate = 0.5, seed = 1).numpy()
array([[2., 0., 0., 2., 2.],
[2., 2., 2., 2., 2.],
[2., 0., 2., 0., 2.]], dtype=float32)
>>> tf.random.set_seed(0)
>>> x = tf.ones([3,5])
>>> tf.nn.dropout(x, rate = 0.8, seed = 1).numpy()
array([[0., 0., 0., 5., 5.],
[0., 5., 0., 5., 0.],
[5., 0., 5., 0., 5.]], dtype=float32)
>>> tf.nn.dropout(x, rate = 0.0) == x
<tf.Tensor: shape=(3, 5), dtype=bool, numpy=
array([[ True, True, True, True, True],
[ True, True, True, True, True],
[ True, True, True, True, True]])>
By default, each element is kept or dropped independently. If `noise_shape`
is specified, it must be
[broadcastable](http://docs.scipy.org/doc/numpy/user/basics.broadcasting.html)
to the shape of `x`, and only dimensions with `noise_shape[i] == shape(x)[i]`
will make independent decisions. This is useful for dropping whole
channels from an image or sequence. For example:
>>> tf.random.set_seed(0)
>>> x = tf.ones([3,10])
>>> tf.nn.dropout(x, rate = 2/3, noise_shape=[1,10], seed=1).numpy()
array([[0., 0., 0., 3., 3., 0., 3., 3., 3., 0.],
[0., 0., 0., 3., 3., 0., 3., 3., 3., 0.],
[0., 0., 0., 3., 3., 0., 3., 3., 3., 0.]], dtype=float32)
Args:
x: A floating point tensor.
rate: A scalar `Tensor` with the same type as x. The probability
that each element is dropped. For example, setting rate=0.1 would drop
10% of input elements.
noise_shape: A 1-D `Tensor` of type `int32`, representing the
shape for randomly generated keep/drop flags.
seed: A Python integer. Used to create random seeds. See
`tf.random.set_seed` for behavior.
name: A name for this operation (optional).
Returns:
A Tensor of the same shape of `x`.
Raises:
ValueError: If `rate` is not in `[0, 1)` or if `x` is not a floating point
tensor. `rate=1` is disallowed, because the output would be all zeros,
which is likely not what was intended."
9032,top_k,tensorflow/tensorflow/python/ops/nn_ops.py,5072,function,"Finds values and indices of the `k` largest entries for the last dimension.
If the input is a vector (rank=1), finds the `k` largest entries in the vector
and outputs their values and indices as vectors. Thus `values[j]` is the
`j`-th largest entry in `input`, and its index is `indices[j]`.
For matrices (resp. higher rank input), computes the top `k` entries in each
row (resp. vector along the last dimension). Thus,
values.shape = indices.shape = input.shape[:-1] + [k]
If two elements are equal, the lower-index element appears first.
Args:
input: 1-D or higher `Tensor` with last dimension at least `k`.
k: 0-D `int32` `Tensor`. Number of top elements to look for along the last
dimension (along each row for matrices).
sorted: If true the resulting `k` elements will be sorted by the values in
descending order.
name: Optional name for the operation.
Returns:
values: The `k` largest elements along each last dimensional slice.
indices: The indices of `values` within the last dimension of `input`."
9033,nth_element,tensorflow/tensorflow/python/ops/nn_ops.py,5101,function,"Finds values of the `n`-th smallest value for the last dimension.
Note that n is zero-indexed.
If the input is a vector (rank-1), finds the entries which is the nth-smallest
value in the vector and outputs their values as scalar tensor.
For matrices (resp. higher rank input), computes the entries which is the
nth-smallest value in each row (resp. vector along the last dimension). Thus,
values.shape = input.shape[:-1]
Args:
input: 1-D or higher `Tensor` with last dimension at least `n+1`.
n: A `Tensor` of type `int32`.
0-D. Position of sorted vector to select along the last dimension (along
each row for matrices). Valid range of n is `[0, input.shape[:-1])`
reverse: An optional `bool`. Defaults to `False`.
When set to True, find the nth-largest value in the vector and vice
versa.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `input`.
The `n`-th order statistic along each last dimensional slice."
9034,fractional_max_pool,tensorflow/tensorflow/python/ops/nn_ops.py,5135,function,"Performs fractional max pooling on the input.
This is a deprecated version of `fractional_max_pool`.
Fractional max pooling is slightly different than regular max pooling. In
regular max pooling, you downsize an input set by taking the maximum value of
smaller N x N subsections of the set (often 2x2), and try to reduce the set by
a factor of N, where N is an integer. Fractional max pooling, as you might
expect from the word ""fractional"", means that the overall reduction ratio N
does not have to be an integer.
The sizes of the pooling regions are generated randomly but are fairly
uniform. For example, let's look at the height dimension, and the constraints
on the list of rows that will be pool boundaries.
First we define the following:
1. input_row_length : the number of rows from the input set
2. output_row_length : which will be smaller than the input
3. alpha = input_row_length / output_row_length : our reduction ratio
4. K = floor(alpha)
5. row_pooling_sequence : this is the result list of pool boundary rows
Then, row_pooling_sequence should satisfy:
1. a[0] = 0 : the first value of the sequence is 0
2. a[end] = input_row_length : the last value of the sequence is the size
3. K <= (a[i+1] - a[i]) <= K+1 : all intervals are K or K+1 size
4. length(row_pooling_sequence) = output_row_length+1
Args:
value: A `Tensor`. 4-D with shape `[batch, height, width, channels]`.
pooling_ratio: A list of `floats` that has length >= 4. Pooling ratio for
each dimension of `value`, currently only supports row and col dimension
and should be >= 1.0. For example, a valid pooling ratio looks like [1.0,
1.44, 1.73, 1.0]. The first and last elements must be 1.0 because we don't
allow pooling on batch and channels dimensions. 1.44 and 1.73 are pooling
ratio on height and width dimensions respectively.
pseudo_random: An optional `bool`. Defaults to `False`. When set to `True`,
generates the pooling sequence in a pseudorandom fashion, otherwise, in a
random fashion. Check (Graham, 2015) for difference between
pseudorandom and random.
overlapping: An optional `bool`. Defaults to `False`. When set to `True`,
it means when pooling, the values at the boundary of adjacent pooling
cells are used by both cells. For example:
`index 0 1 2 3 4`
`value 20 5 16 3 7`
If the pooling sequence is [0, 2, 4], then 16, at index 2 will be used
twice. The result would be [20, 16] for fractional max pooling.
deterministic: An optional `bool`. Deprecated; use `fractional_max_pool_v2`
instead.
seed: An optional `int`. Defaults to `0`. If set to be non-zero, the
random number generator is seeded by the given seed. Otherwise it is
seeded by a random seed.
seed2: An optional `int`. Deprecated; use `fractional_max_pool_v2` instead.
name: A name for the operation (optional).
Returns:
A tuple of `Tensor` objects (`output`, `row_pooling_sequence`,
`col_pooling_sequence`).
output: Output `Tensor` after fractional max pooling. Has the same type as
`value`.
row_pooling_sequence: A `Tensor` of type `int64`.
col_pooling_sequence: A `Tensor` of type `int64`.
References:
Fractional Max-Pooling:
[Graham, 2015](https://arxiv.org/abs/1412.6071)
([pdf](https://arxiv.org/pdf/1412.6071.pdf))"
9035,fractional_max_pool_v2,tensorflow/tensorflow/python/ops/nn_ops.py,5220,function,"Performs fractional max pooling on the input.
Fractional max pooling is slightly different than regular max pooling. In
regular max pooling, you downsize an input set by taking the maximum value of
smaller N x N subsections of the set (often 2x2), and try to reduce the set by
a factor of N, where N is an integer. Fractional max pooling, as you might
expect from the word ""fractional"", means that the overall reduction ratio N
does not have to be an integer.
The sizes of the pooling regions are generated randomly but are fairly
uniform. For example, let's look at the height dimension, and the constraints
on the list of rows that will be pool boundaries.
First we define the following:
1. input_row_length : the number of rows from the input set
2. output_row_length : which will be smaller than the input
3. alpha = input_row_length / output_row_length : our reduction ratio
4. K = floor(alpha)
5. row_pooling_sequence : this is the result list of pool boundary rows
Then, row_pooling_sequence should satisfy:
1. a[0] = 0 : the first value of the sequence is 0
2. a[end] = input_row_length : the last value of the sequence is the size
3. K <= (a[i+1] - a[i]) <= K+1 : all intervals are K or K+1 size
4. length(row_pooling_sequence) = output_row_length+1
Args:
value: A `Tensor`. 4-D with shape `[batch, height, width, channels]`.
pooling_ratio: An int or list of `ints` that has length `1`, `2` or `4`.
Pooling ratio for each dimension of `value`, currently only supports row
and col dimension and should be >= 1.0. For example, a valid pooling ratio
looks like [1.0, 1.44, 1.73, 1.0]. The first and last elements must be 1.0
because we don't allow pooling on batch and channels dimensions. 1.44 and
1.73 are pooling ratio on height and width dimensions respectively.
pseudo_random: An optional `bool`. Defaults to `False`. When set to `True`,
generates the pooling sequence in a pseudorandom fashion, otherwise, in a
random fashion. Check paper (Graham, 2015) for difference between
pseudorandom and random.
overlapping: An optional `bool`. Defaults to `False`. When set to `True`,
it means when pooling, the values at the boundary of adjacent pooling
cells are used by both cells. For example:
`index 0 1 2 3 4`
`value 20 5 16 3 7`
If the pooling sequence is [0, 2, 4], then 16, at index 2 will be used
twice. The result would be [20, 16] for fractional max pooling.
seed: An optional `int`. Defaults to `0`. If set to be non-zero, the
random number generator is seeded by the given seed. Otherwise it is
seeded by a random seed.
name: A name for the operation (optional).
Returns:
A tuple of `Tensor` objects (`output`, `row_pooling_sequence`,
`col_pooling_sequence`).
output: Output `Tensor` after fractional max pooling. Has the same type as
`value`.
row_pooling_sequence: A `Tensor` of type `int64`.
col_pooling_sequence: A `Tensor` of type `int64`.
References:
Fractional Max-Pooling:
[Graham, 2015](https://arxiv.org/abs/1412.6071)
([pdf](https://arxiv.org/pdf/1412.6071.pdf))"
9036,fractional_avg_pool,tensorflow/tensorflow/python/ops/nn_ops.py,5308,function,"Performs fractional average pooling on the input.
This is a deprecated version of `fractional_avg_pool`.
Fractional average pooling is similar to Fractional max pooling in the pooling
region generation step. The only difference is that after pooling regions are
generated, a mean operation is performed instead of a max operation in each
pooling region.
Args:
value: A `Tensor`. 4-D with shape `[batch, height, width, channels]`.
pooling_ratio: A list of `floats` that has length >= 4. Pooling ratio for
each dimension of `value`, currently only supports row and col dimension
and should be >= 1.0. For example, a valid pooling ratio looks like [1.0,
1.44, 1.73, 1.0]. The first and last elements must be 1.0 because we don't
allow pooling on batch and channels dimensions. 1.44 and 1.73 are pooling
ratio on height and width dimensions respectively.
pseudo_random: An optional `bool`. Defaults to `False`. When set to `True`,
generates the pooling sequence in a pseudorandom fashion, otherwise, in a
random fashion. Check paper (Graham, 2015) for difference between
pseudorandom and random.
overlapping: An optional `bool`. Defaults to `False`. When set to `True`,
it means when pooling, the values at the boundary of adjacent pooling
cells are used by both cells. For example:
`index 0 1 2 3 4`
`value 20 5 16 3 7`
If the pooling sequence is [0, 2, 4], then 16, at index 2 will be used
twice. The result would be [20, 16] for fractional avg pooling.
deterministic: An optional `bool`. Deprecated; use `fractional_avg_pool_v2`
instead.
seed: An optional `int`. Defaults to `0`. If set to be non-zero, the
random number generator is seeded by the given seed. Otherwise it is
seeded by a random seed.
seed2: An optional `int`. Deprecated; use `fractional_avg_pool_v2` instead.
name: A name for the operation (optional).
Returns:
A tuple of `Tensor` objects (`output`, `row_pooling_sequence`,
`col_pooling_sequence`).
output: Output `Tensor` after fractional avg pooling. Has the same type as
`value`.
row_pooling_sequence: A `Tensor` of type `int64`.
col_pooling_sequence: A `Tensor` of type `int64`.
References:
Fractional Max-Pooling:
[Graham, 2015](https://arxiv.org/abs/1412.6071)
([pdf](https://arxiv.org/pdf/1412.6071.pdf))"
9037,fractional_avg_pool_v2,tensorflow/tensorflow/python/ops/nn_ops.py,5372,function,"Performs fractional average pooling on the input.
Fractional average pooling is similar to Fractional max pooling in the pooling
region generation step. The only difference is that after pooling regions are
generated, a mean operation is performed instead of a max operation in each
pooling region.
Args:
value: A `Tensor`. 4-D with shape `[batch, height, width, channels]`.
pooling_ratio: A list of `floats` that has length >= 4. Pooling ratio for
each dimension of `value`, currently only supports row and col dimension
and should be >= 1.0. For example, a valid pooling ratio looks like [1.0,
1.44, 1.73, 1.0]. The first and last elements must be 1.0 because we don't
allow pooling on batch and channels dimensions. 1.44 and 1.73 are pooling
ratio on height and width dimensions respectively.
pseudo_random: An optional `bool`. Defaults to `False`. When set to `True`,
generates the pooling sequence in a pseudorandom fashion, otherwise, in a
random fashion. Check paper (Graham, 2015) for difference between
pseudorandom and random.
overlapping: An optional `bool`. Defaults to `False`. When set to `True`,
it means when pooling, the values at the boundary of adjacent pooling
cells are used by both cells. For example:
`index 0 1 2 3 4`
`value 20 5 16 3 7`
If the pooling sequence is [0, 2, 4], then 16, at index 2 will be used
twice. The result would be [20, 16] for fractional avg pooling.
seed: An optional `int`. Defaults to `0`. If set to be non-zero, the
random number generator is seeded by the given seed. Otherwise it is
seeded by a random seed.
name: A name for the operation (optional).
Returns:
A tuple of `Tensor` objects (`output`, `row_pooling_sequence`,
`col_pooling_sequence`).
output: Output `Tensor` after fractional avg pooling. Has the same type as
`value`.
row_pooling_sequence: A `Tensor` of type `int64`.
col_pooling_sequence: A `Tensor` of type `int64`.
References:
Fractional Max-Pooling:
[Graham, 2015](https://arxiv.org/abs/1412.6071)
([pdf](https://arxiv.org/pdf/1412.6071.pdf))"
9038,erosion2d,tensorflow/tensorflow/python/ops/nn_ops.py,5451,function,"Computes the grayscale erosion of 4-D `value` and 3-D `kernel` tensors.
The `value` tensor has shape `[batch, in_height, in_width, depth]` and the
`kernel` tensor has shape `[kernel_height, kernel_width, depth]`, i.e.,
each input channel is processed independently of the others with its own
structuring function. The `output` tensor has shape
`[batch, out_height, out_width, depth]`. The spatial dimensions of the
output tensor depend on the `padding` algorithm. We currently only support the
default ""NHWC"" `data_format`.
In detail, the grayscale morphological 2-D erosion is given by:
output[b, y, x, c] =
min_{dy, dx} value[b,
strides[1] * y - rates[1] * dy,
strides[2] * x - rates[2] * dx,
c] -
kernel[dy, dx, c]
Duality: The erosion of `value` by the `kernel` is equal to the negation of
the dilation of `-value` by the reflected `kernel`.
Args:
value: A `Tensor`. 4-D with shape `[batch, in_height, in_width, depth]`.
kernel: A `Tensor`. Must have the same type as `value`.
3-D with shape `[kernel_height, kernel_width, depth]`.
strides: A list of `ints` that has length `>= 4`.
1-D of length 4. The stride of the sliding window for each dimension of
the input tensor. Must be: `[1, stride_height, stride_width, 1]`.
rates: A list of `ints` that has length `>= 4`.
1-D of length 4. The input stride for atrous morphological dilation.
Must be: `[1, rate_height, rate_width, 1]`.
padding: A `string` from: `""SAME"", ""VALID""`.
The type of padding algorithm to use.
name: A name for the operation (optional). If not specified ""erosion2d""
is used.
Returns:
A `Tensor`. Has the same type as `value`.
4-D with shape `[batch, out_height, out_width, depth]`.
Raises:
ValueError: If the `value` depth does not match `kernel`' shape, or if
padding is other than `'VALID'` or `'SAME'`."
9039,erosion2d_v2,tensorflow/tensorflow/python/ops/nn_ops.py,5511,function,"Computes the grayscale erosion of 4-D `value` and 3-D `filters` tensors.
The `value` tensor has shape `[batch, in_height, in_width, depth]` and the
`filters` tensor has shape `[filters_height, filters_width, depth]`, i.e.,
each input channel is processed independently of the others with its own
structuring function. The `output` tensor has shape
`[batch, out_height, out_width, depth]`. The spatial dimensions of the
output tensor depend on the `padding` algorithm. We currently only support the
default ""NHWC"" `data_format`.
In detail, the grayscale morphological 2-D erosion is given by:
output[b, y, x, c] =
min_{dy, dx} value[b,
strides[1] * y - dilations[1] * dy,
strides[2] * x - dilations[2] * dx,
c] -
filters[dy, dx, c]
Duality: The erosion of `value` by the `filters` is equal to the negation of
the dilation of `-value` by the reflected `filters`.
Args:
value: A `Tensor`. 4-D with shape `[batch, in_height, in_width, depth]`.
filters: A `Tensor`. Must have the same type as `value`.
3-D with shape `[filters_height, filters_width, depth]`.
strides: A list of `ints` that has length `>= 4`.
1-D of length 4. The stride of the sliding window for each dimension of
the input tensor. Must be: `[1, stride_height, stride_width, 1]`.
padding: A `string` from: `""SAME"", ""VALID""`.
The type of padding algorithm to use.
data_format: A `string`, only `""NHWC""` is currently supported.
dilations: A list of `ints` that has length `>= 4`.
1-D of length 4. The input stride for atrous morphological dilation.
Must be: `[1, rate_height, rate_width, 1]`.
name: A name for the operation (optional). If not specified ""erosion2d""
is used.
Returns:
A `Tensor`. Has the same type as `value`.
4-D with shape `[batch, out_height, out_width, depth]`.
Raises:
ValueError: If the `value` depth does not match `filters`' shape, or if
padding is other than `'VALID'` or `'SAME'`."
9040,in_top_k,tensorflow/tensorflow/python/ops/nn_ops.py,5581,function,"Says whether the targets are in the top `K` predictions.
This outputs a `batch_size` bool array, an entry `out[i]` is `true` if the
prediction for the target class is finite (not inf, -inf, or nan) and among
the top `k` predictions among all predictions for example `i`. Note that the
behavior of `InTopK` differs from the `TopK` op in its handling of ties; if
multiple classes have the same prediction value and straddle the top-`k`
boundary, all of those classes are considered to be in the top `k`.
More formally, let
\\(predictions_i\\) be the predictions for all classes for example `i`,
\\(targets_i\\) be the target class for example `i`,
\\(out_i\\) be the output for example `i`,
$$out_i = predictions_{i, targets_i} \in TopKIncludingTies(predictions_i)$$
Args:
predictions: A `Tensor` of type `float32`.
A `batch_size` x `classes` tensor.
targets: A `Tensor`. Must be one of the following types: `int32`, `int64`.
A `batch_size` vector of class ids.
k: An `int`. Number of top elements to look at for computing precision.
name: A name for the operation (optional).
Returns:
A `Tensor` of type `bool`. Computed Precision at `k` as a `bool Tensor`."
9041,in_top_k_v2,tensorflow/tensorflow/python/ops/nn_ops.py,5616,function,
9042,verify_tensor_all_finite,tensorflow/tensorflow/python/ops/numerics.py,35,function,"Assert that the tensor does not contain any NaN's or Inf's.
Args:
t: Tensor to check.
msg: Message to log on failure.
name: A name for this operation (optional).
x: Alias for t.
message: Alias for msg.
Returns:
Same tensor as `t`."
9043,verify_tensor_all_finite_v2,tensorflow/tensorflow/python/ops/numerics.py,56,function,"Assert that the tensor does not contain any NaN's or Inf's.
Args:
x: Tensor to check.
message: Message to log on failure.
name: A name for this operation (optional).
Returns:
Same tensor as `x`."
9044,add_check_numerics_ops,tensorflow/tensorflow/python/ops/numerics.py,76,function,"Connect a `tf.debugging.check_numerics` to every floating point tensor.
`check_numerics` operations themselves are added for each `half`, `float`,
or `double` tensor in the current default graph. For all ops in the graph, the
`check_numerics` op for all of its (`half`, `float`, or `double`) inputs
is guaranteed to run before the `check_numerics` op on any of its outputs.
Note: This API is not compatible with the use of `tf.cond` or
`tf.while_loop`, and will raise a `ValueError` if you attempt to call it
in such a graph.
Returns:
A `group` op depending on all `check_numerics` ops added.
Raises:
ValueError: If the graph contains any numeric operations in a control flow
structure.
RuntimeError: If called with eager execution enabled.
@compatibility(eager)
Not compatible with eager execution. To check for `Inf`s and `NaN`s under
eager execution, call `tf.debugging.enable_check_numerics()` once before
executing the checked operations.
@end_compatibility"
9045,is_differentiable,tensorflow/tensorflow/python/ops/op_selector.py,25,function,
9046,is_iterable,tensorflow/tensorflow/python/ops/op_selector.py,32,function,Return true if the object is iterable.
9047,concatenate_unique,tensorflow/tensorflow/python/ops/op_selector.py,43,function,"Add all the elements of `lb` to `la` if they are not there already.
The elements added to `la` maintain ordering with respect to `lb`.
Args:
la: List of Python objects.
lb: List of Python objects.
Returns:
`la`: The list `la` with missing elements from `lb`."
9048,get_tensors,tensorflow/tensorflow/python/ops/op_selector.py,62,function,"get all the tensors which are input or output of an op in the graph.
Args:
graph: a `tf.Graph`.
Returns:
A list of `tf.Tensor`.
Raises:
TypeError: if graph is not a `tf.Graph`."
9049,get_unique_graph,tensorflow/tensorflow/python/ops/op_selector.py,80,function,"Return the unique graph used by the all the elements in tops.
Args:
tops: list of elements to check (usually a list of tf.Operation and/or
tf.Tensor). Or a tf.Graph.
check_types: check that the element in tops are of given type(s). If None,
the types (tf.Operation, tf.Tensor) are used.
none_if_empty: don't raise an error if tops is an empty list, just return
None.
Returns:
The unique graph used by all the tops.
Raises:
TypeError: if tops is not a iterable of tf.Operation.
ValueError: if the graph is not unique."
9050,check_graphs,tensorflow/tensorflow/python/ops/op_selector.py,118,function,"Check that all the element in args belong to the same graph.
Args:
*args: a list of object with a obj.graph property.
Raises:
ValueError: if all the elements do not belong to the same graph."
9051,make_list_of_t,tensorflow/tensorflow/python/ops/op_selector.py,134,function,"Convert ts to a list of `tf.Tensor`.
Args:
ts: can be an iterable of `tf.Tensor`, a `tf.Graph` or a single tensor.
check_graph: if `True` check if all the tensors belong to the same graph.
allow_graph: if `False` a `tf.Graph` cannot be converted.
ignore_ops: if `True`, silently ignore `tf.Operation`.
Returns:
A newly created list of `tf.Tensor`.
Raises:
TypeError: if `ts` cannot be converted to a list of `tf.Tensor` or,
if `check_graph` is `True`, if all the ops do not belong to the same graph."
9052,get_generating_ops,tensorflow/tensorflow/python/ops/op_selector.py,164,function,"Return all the generating ops of the tensors in `ts`.
Args:
ts: a list of `tf.Tensor`
Returns:
A list of all the generating `tf.Operation` of the tensors in `ts`.
Raises:
TypeError: if `ts` cannot be converted to a list of `tf.Tensor`."
9053,get_consuming_ops,tensorflow/tensorflow/python/ops/op_selector.py,178,function,"Return all the consuming ops of the tensors in ts.
Args:
ts: a list of `tf.Tensor`
Returns:
A list of all the consuming `tf.Operation` of the tensors in `ts`.
Raises:
TypeError: if ts cannot be converted to a list of `tf.Tensor`."
9054,make_list_of_op,tensorflow/tensorflow/python/ops/op_selector.py,197,function,"Convert ops to a list of `tf.Operation`.
Args:
tops: can be an iterable of `tf.Operation`, a `tf.Graph` or a single
operation.
check_graph: if `True` check if all the operations belong to the same graph.
allow_graph: if `False` a `tf.Graph` cannot be converted.
ignore_ts: if True, silently ignore `tf.Tensor`.
Returns:
A newly created list of `tf.Operation`.
Raises:
TypeError: if tops cannot be converted to a list of `tf.Operation` or,
if `check_graph` is `True`, if all the ops do not belong to the
same graph."
9055,get_backward_walk_ops,tensorflow/tensorflow/python/ops/op_selector.py,237,function,"Do a backward graph walk and return all the visited ops.
Args:
seed_ops: an iterable of operations from which the backward graph
walk starts. If a list of tensors is given instead, the seed_ops are set
to be the generators of those tensors.
inclusive: if True the given seed_ops are also part of the resulting set.
within_ops: an iterable of `tf.Operation` within which the search is
restricted. If `within_ops` is `None`, the search is performed within
the whole graph.
within_ops_fn: if provided, a function on ops that should return True iff
the op is within the graph traversal. This can be used along within_ops,
in which case an op is within if it is also in within_ops.
stop_at_ts: an iterable of tensors at which the graph walk stops.
control_inputs: if True, control inputs will be used while moving backward.
only_differentiable: if True, only traverse ops which are differentiable.
This includes natively differentiable ops, or ops with custom gradients.
Returns:
A Python set of all the `tf.Operation` behind `seed_ops`.
Raises:
TypeError: if `seed_ops` or `within_ops` cannot be converted to a list of
`tf.Operation`."
9056,UnliftableError,tensorflow/tensorflow/python/ops/op_selector.py,311,class,Raised if a Tensor cannot be lifted from the graph.
9057,graph_inputs,tensorflow/tensorflow/python/ops/op_selector.py,324,function,
9058,map_subgraph,tensorflow/tensorflow/python/ops/op_selector.py,368,function,"Walk a Graph and capture the subgraph between init_tensor and sources.
Note: This function mutates visited_ops and op_outputs.
Arguments:
init_tensor: A Tensor or Operation where the subgraph terminates.
sources: A set of Tensors where subgraph extraction should stop.
disallowed_placeholders: An optional set of ops which may not appear in the
lifted graph. Defaults to all placeholders.
visited_ops: A set of operations which were visited in a prior pass.
op_outputs: A defaultdict containing the outputs of an op which are to be
copied into the new subgraph.
add_sources: A boolean indicating whether placeholders which are not in
sources should be allowed.
Returns:
The set of placeholders upon which init_tensor depends and are not in
sources.
Raises:
UnliftableError: if init_tensor depends on a placeholder which is not in
sources and add_sources is False."
9059,VarLenFeature,tensorflow/tensorflow/python/ops/parsing_config.py,48,class,"Configuration for parsing a variable-length input feature.
Fields:
dtype: Data type of input."
9060,RaggedFeature,tensorflow/tensorflow/python/ops/parsing_config.py,58,class,"Configuration for passing a RaggedTensor input feature.
`value_key` specifies the feature key for a variable-length list of values;
and `partitions` specifies zero or more feature keys for partitioning those
values into higher dimensions. Each element of `partitions` must be one of
the following:
* `tf.io.RaggedFeature.RowSplits(key: string)`
* `tf.io.RaggedFeature.RowLengths(key: string)`
* `tf.io.RaggedFeature.RowStarts(key: string)`
* `tf.io.RaggedFeature.RowLimits(key: string)`
* `tf.io.RaggedFeature.ValueRowIds(key: string)`
* `tf.io.RaggedFeature.UniformRowLength(length: int)`.
Where `key` is a feature key whose values are used to partition the values.
Partitions are listed from outermost to innermost.
* If `len(partitions) == 0` (the default), then:
* A feature from a single `tf.Example` is parsed into a 1D `tf.Tensor`.
* A feature from a batch of `tf.Example`s is parsed into a 2D
`tf.RaggedTensor`, where the outer dimension is the batch dimension, and
the inner (ragged) dimension is the feature length in each example.
* If `len(partitions) == 1`, then:
* A feature from a single `tf.Example` is parsed into a 2D
`tf.RaggedTensor`, where the values taken from the `value_key` are
separated into rows using the partition key.
* A feature from a batch of `tf.Example`s is parsed into a 3D
`tf.RaggedTensor`, where the outer dimension is the batch dimension,
the two inner dimensions are formed by separating the `value_key` values
from each example into rows using that example's partition key.
* If `len(partitions) > 1`, then:
* A feature from a single `tf.Example` is parsed into a `tf.RaggedTensor`
whose rank is `len(partitions)+1`, and whose ragged_rank is
`len(partitions)`.
* A feature from a batch of `tf.Example`s is parsed into a `tf.RaggedTensor`
whose rank is `len(partitions)+2` and whose ragged_rank is
`len(partitions)+1`, where the outer dimension is the batch dimension.
There is one exception: if the final (i.e., innermost) element(s) of
`partitions` are `UniformRowLength`s, then the values are simply reshaped (as
a higher-dimensional `tf.Tensor`), rather than being wrapped in a
`tf.RaggedTensor`.
#### Examples
>>> import google.protobuf.text_format as pbtext
>>> example_batch = [
... pbtext.Merge(r'''
... features {
... feature {key: ""v"" value {int64_list {value: [3, 1, 4, 1, 5, 9]}}}
... feature {key: ""s1"" value {int64_list {value: [0, 2, 3, 3, 6]}}}
... feature {key: ""s2"" value {int64_list {value: [0, 2, 3, 4]}}}
... }''', tf.train.Example()).SerializeToString(),
... pbtext.Merge(r'''
... features {
... feature {key: ""v"" value {int64_list {value: [2, 7, 1, 8, 2, 8, 1]}}}
... feature {key: ""s1"" value {int64_list {value: [0, 3, 4, 5, 7]}}}
... feature {key: ""s2"" value {int64_list {value: [0, 1, 1, 4]}}}
... }''', tf.train.Example()).SerializeToString()]
>>> features = {
... # Zero partitions: returns 1D tf.Tensor for each Example.
... 'f1': tf.io.RaggedFeature(value_key=""v"", dtype=tf.int64),
... # One partition: returns 2D tf.RaggedTensor for each Example.
... 'f2': tf.io.RaggedFeature(value_key=""v"", dtype=tf.int64, partitions=[
... tf.io.RaggedFeature.RowSplits(""s1"")]),
... # Two partitions: returns 3D tf.RaggedTensor for each Example.
... 'f3': tf.io.RaggedFeature(value_key=""v"", dtype=tf.int64, partitions=[
... tf.io.RaggedFeature.RowSplits(""s2""),
... tf.io.RaggedFeature.RowSplits(""s1"")])
... }
>>> feature_dict = tf.io.parse_single_example(example_batch[0], features)
>>> for (name, val) in sorted(feature_dict.items()):
... print('%s: %s' % (name, val))
f1: tf.Tensor([3 1 4 1 5 9], shape=(6,), dtype=int64)
f2: <tf.RaggedTensor [[3, 1], [4], [], [1, 5, 9]]>
f3: <tf.RaggedTensor [[[3, 1], [4]], [[]], [[1, 5, 9]]]>
>>> feature_dict = tf.io.parse_example(example_batch, features)
>>> for (name, val) in sorted(feature_dict.items()):
... print('%s: %s' % (name, val))
f1: <tf.RaggedTensor [[3, 1, 4, 1, 5, 9],
[2, 7, 1, 8, 2, 8, 1]]>
f2: <tf.RaggedTensor [[[3, 1], [4], [], [1, 5, 9]],
[[2, 7, 1], [8], [2], [8, 1]]]>
f3: <tf.RaggedTensor [[[[3, 1], [4]], [[]], [[1, 5, 9]]],
[[[2, 7, 1]], [], [[8], [2], [8, 1]]]]>
Fields:
dtype: Data type of the `RaggedTensor`. Must be one of:
`tf.dtypes.int64`, `tf.dtypes.float32`, `tf.dtypes.string`.
value_key: (Optional.) Key for a `Feature` in the input `Example`, whose
parsed `Tensor` will be the resulting `RaggedTensor.flat_values`. If
not specified, then it defaults to the key for this `RaggedFeature`.
partitions: (Optional.) A list of objects specifying the row-partitioning
tensors (from outermost to innermost). Each entry in this list must be
one of:
* `tf.io.RaggedFeature.RowSplits(key: string)`
* `tf.io.RaggedFeature.RowLengths(key: string)`
* `tf.io.RaggedFeature.RowStarts(key: string)`
* `tf.io.RaggedFeature.RowLimits(key: string)`
* `tf.io.RaggedFeature.ValueRowIds(key: string)`
* `tf.io.RaggedFeature.UniformRowLength(length: int)`.
Where `key` is a key for a `Feature` in the input `Example`, whose parsed
`Tensor` will be the resulting row-partitioning tensor.
row_splits_dtype: (Optional.) Data type for the row-partitioning tensor(s).
One of `int32` or `int64`. Defaults to `int32`.
validate: (Optional.) Boolean indicating whether or not to validate that
the input values form a valid RaggedTensor. Defaults to `False`."
9061,SparseFeature,tensorflow/tensorflow/python/ops/parsing_config.py,224,class,"Configuration for parsing a sparse input feature from an `Example`.
Note, preferably use `VarLenFeature` (possibly in combination with a
`SequenceExample`) in order to parse out `SparseTensor`s instead of
`SparseFeature` due to its simplicity.
Closely mimicking the `SparseTensor` that will be obtained by parsing an
`Example` with a `SparseFeature` config, a `SparseFeature` contains a
* `value_key`: The name of key for a `Feature` in the `Example` whose parsed
`Tensor` will be the resulting `SparseTensor.values`.
* `index_key`: A list of names - one for each dimension in the resulting
`SparseTensor` whose `indices[i][dim]` indicating the position of
the `i`-th value in the `dim` dimension will be equal to the `i`-th value in
the Feature with key named `index_key[dim]` in the `Example`.
* `size`: A list of ints for the resulting `SparseTensor.dense_shape`.
For example, we can represent the following 2D `SparseTensor`
```python
SparseTensor(indices=[[3, 1], [20, 0]],
values=[0.5, -1.0]
dense_shape=[100, 3])
```
with an `Example` input proto
```python
features {
feature { key: ""val"" value { float_list { value: [ 0.5, -1.0 ] } } }
feature { key: ""ix0"" value { int64_list { value: [ 3, 20 ] } } }
feature { key: ""ix1"" value { int64_list { value: [ 1, 0 ] } } }
}
```
and `SparseFeature` config with 2 `index_key`s
```python
SparseFeature(index_key=[""ix0"", ""ix1""],
value_key=""val"",
dtype=tf.float32,
size=[100, 3])
```
Fields:
index_key: A single string name or a list of string names of index features.
For each key the underlying feature's type must be `int64` and its length
must always match that of the `value_key` feature.
To represent `SparseTensor`s with a `dense_shape` of `rank` higher than 1
a list of length `rank` should be used.
value_key: Name of value feature. The underlying feature's type must
be `dtype` and its length must always match that of all the `index_key`s'
features.
dtype: Data type of the `value_key` feature.
size: A Python int or list thereof specifying the dense shape. Should be a
list if and only if `index_key` is a list. In that case the list must be
equal to the length of `index_key`. Each for each entry `i` all values in
the `index_key`[i] feature must be in `[0, size[i])`.
already_sorted: A Python boolean to specify whether the values in
`value_key` are already sorted by their index position. If so skip
sorting. False by default (optional)."
9062,FixedLenFeature,tensorflow/tensorflow/python/ops/parsing_config.py,299,class,"Configuration for parsing a fixed-length input feature.
To treat sparse input as dense, provide a `default_value`; otherwise,
the parse functions will fail on any examples missing this feature.
Fields:
shape: Shape of input data.
dtype: Data type of input.
default_value: Value to be used if an example is missing this feature. It
must be compatible with `dtype` and of the specified `shape`."
9063,FixedLenSequenceFeature,tensorflow/tensorflow/python/ops/parsing_config.py,320,class,"Configuration for parsing a variable-length input feature into a `Tensor`.
The resulting `Tensor` of parsing a single `SequenceExample` or `Example` has
a static `shape` of `[None] + shape` and the specified `dtype`.
The resulting `Tensor` of parsing a `batch_size` many `Example`s has
a static `shape` of `[batch_size, None] + shape` and the specified `dtype`.
The entries in the `batch` from different `Examples` will be padded with
`default_value` to the maximum length present in the `batch`.
To treat a sparse input as dense, provide `allow_missing=True`; otherwise,
the parse functions will fail on any examples missing this feature.
Fields:
shape: Shape of input data for dimension 2 and higher. First dimension is
of variable length `None`.
dtype: Data type of input.
allow_missing: Whether to allow this feature to be missing from a feature
list item. Is available only for parsing `SequenceExample` not for
parsing `Examples`.
default_value: Scalar value to be used to pad multiple `Example`s to their
maximum length. Irrelevant for parsing a single `Example` or
`SequenceExample`. Defaults to """" for dtype string and 0 otherwise
(optional)."
9064,parse_example_v2,tensorflow/tensorflow/python/ops/parsing_ops.py,82,function,"Parses `Example` protos into a `dict` of tensors.
Parses a number of serialized [`Example`](https://www.tensorflow.org/code/tensorflow/core/example/example.proto)
protos given in `serialized`. We refer to `serialized` as a batch with
`batch_size` many entries of individual `Example` protos.
`example_names` may contain descriptive names for the corresponding serialized
protos. These may be useful for debugging purposes, but they have no effect on
the output. If not `None`, `example_names` must be the same length as
`serialized`.
This op parses serialized examples into a dictionary mapping keys to `Tensor`
`SparseTensor`, and `RaggedTensor` objects. `features` is a dict from keys to
`VarLenFeature`, `SparseFeature`, `RaggedFeature`, and `FixedLenFeature`
objects. Each `VarLenFeature` and `SparseFeature` is mapped to a
`SparseTensor`; each `FixedLenFeature` is mapped to a `Tensor`; and each
`RaggedFeature` is mapped to a `RaggedTensor`.
Each `VarLenFeature` maps to a `SparseTensor` of the specified type
representing a ragged matrix. Its indices are `[batch, index]` where `batch`
identifies the example in `serialized`, and `index` is the value's index in
the list of values associated with that feature and example.
Each `SparseFeature` maps to a `SparseTensor` of the specified type
representing a Tensor of `dense_shape` `[batch_size] + SparseFeature.size`.
Its `values` come from the feature in the examples with key `value_key`.
A `values[i]` comes from a position `k` in the feature of an example at batch
entry `batch`. This positional information is recorded in `indices[i]` as
`[batch, index_0, index_1, ...]` where `index_j` is the `k-th` value of
the feature in the example at with key `SparseFeature.index_key[j]`.
In other words, we split the indices (except the first index indicating the
batch entry) of a `SparseTensor` by dimension into different features of the
`Example`. Due to its complexity a `VarLenFeature` should be preferred over a
`SparseFeature` whenever possible.
Each `FixedLenFeature` `df` maps to a `Tensor` of the specified type (or
`tf.float32` if not specified) and shape `(serialized.size(),) + df.shape`.
`FixedLenFeature` entries with a `default_value` are optional. With no default
value, we will fail if that `Feature` is missing from any example in
`serialized`.
Each `FixedLenSequenceFeature` `df` maps to a `Tensor` of the specified type
(or `tf.float32` if not specified) and shape
`(serialized.size(), None) + df.shape`.
All examples in `serialized` will be padded with `default_value` along the
second dimension.
Each `RaggedFeature` maps to a `RaggedTensor` of the specified type. It
is formed by stacking the `RaggedTensor` for each example, where the
`RaggedTensor` for each individual example is constructed using the tensors
specified by `RaggedTensor.values_key` and `RaggedTensor.partition`. See
the `tf.io.RaggedFeature` documentation for details and examples.
Examples:
For example, if one expects a `tf.float32` `VarLenFeature` `ft` and three
serialized `Example`s are provided:
```
serialized = [
features
{ feature { key: ""ft"" value { float_list { value: [1.0, 2.0] } } } },
features
{ feature []},
features
{ feature { key: ""ft"" value { float_list { value: [3.0] } } }
]
```
then the output will look like:
```python
{""ft"": SparseTensor(indices=[[0, 0], [0, 1], [2, 0]],
values=[1.0, 2.0, 3.0],
dense_shape=(3, 2)) }
```
If instead a `FixedLenSequenceFeature` with `default_value = -1.0` and
`shape=[]` is used then the output will look like:
```python
{""ft"": [[1.0, 2.0], [3.0, -1.0]]}
```
Given two `Example` input protos in `serialized`:
```
[
features {
feature { key: ""kw"" value { bytes_list { value: [ ""knit"", ""big"" ] } } }
feature { key: ""gps"" value { float_list { value: [] } } }
},
features {
feature { key: ""kw"" value { bytes_list { value: [ ""emmy"" ] } } }
feature { key: ""dank"" value { int64_list { value: [ 42 ] } } }
feature { key: ""gps"" value { } }
}
]
```
And arguments
```
example_names: [""input0"", ""input1""],
features: {
""kw"": VarLenFeature(tf.string),
""dank"": VarLenFeature(tf.int64),
""gps"": VarLenFeature(tf.float32),
}
```
Then the output is a dictionary:
```python
{
""kw"": SparseTensor(
indices=[[0, 0], [0, 1], [1, 0]],
values=[""knit"", ""big"", ""emmy""]
dense_shape=[2, 2]),
""dank"": SparseTensor(
indices=[[1, 0]],
values=[42],
dense_shape=[2, 1]),
""gps"": SparseTensor(
indices=[],
values=[],
dense_shape=[2, 0]),
}
```
For dense results in two serialized `Example`s:
```
[
features {
feature { key: ""age"" value { int64_list { value: [ 0 ] } } }
feature { key: ""gender"" value { bytes_list { value: [ ""f"" ] } } }
},
features {
feature { key: ""age"" value { int64_list { value: [] } } }
feature { key: ""gender"" value { bytes_list { value: [ ""f"" ] } } }
}
]
```
We can use arguments:
```
example_names: [""input0"", ""input1""],
features: {
""age"": FixedLenFeature([], dtype=tf.int64, default_value=-1),
""gender"": FixedLenFeature([], dtype=tf.string),
}
```
And the expected output is:
```python
{
""age"": [[0], [-1]],
""gender"": [[""f""], [""f""]],
}
```
An alternative to `VarLenFeature` to obtain a `SparseTensor` is
`SparseFeature`. For example, given two `Example` input protos in
`serialized`:
```
[
features {
feature { key: ""val"" value { float_list { value: [ 0.5, -1.0 ] } } }
feature { key: ""ix"" value { int64_list { value: [ 3, 20 ] } } }
},
features {
feature { key: ""val"" value { float_list { value: [ 0.0 ] } } }
feature { key: ""ix"" value { int64_list { value: [ 42 ] } } }
}
]
```
And arguments
```
example_names: [""input0"", ""input1""],
features: {
""sparse"": SparseFeature(
index_key=""ix"", value_key=""val"", dtype=tf.float32, size=100),
}
```
Then the output is a dictionary:
```python
{
""sparse"": SparseTensor(
indices=[[0, 3], [0, 20], [1, 42]],
values=[0.5, -1.0, 0.0]
dense_shape=[2, 100]),
}
```
See the `tf.io.RaggedFeature` documentation for examples showing how
`RaggedFeature` can be used to obtain `RaggedTensor`s.
Args:
serialized: A vector (1-D Tensor) of strings, a batch of binary
serialized `Example` protos.
features: A `dict` mapping feature keys to `FixedLenFeature`,
`VarLenFeature`, `SparseFeature`, and `RaggedFeature` values.
example_names: A vector (1-D Tensor) of strings (optional), the names of
the serialized protos in the batch.
name: A name for this operation (optional).
Returns:
A `dict` mapping feature keys to `Tensor`, `SparseTensor`, and
`RaggedTensor` values.
Raises:
ValueError: if any feature is invalid."
9065,parse_example,tensorflow/tensorflow/python/ops/parsing_ops.py,320,function,
9066,parse_single_example,tensorflow/tensorflow/python/ops/parsing_ops.py,380,function,"Parses a single `Example` proto.
Similar to `parse_example`, except:
For dense tensors, the returned `Tensor` is identical to the output of
`parse_example`, except there is no batch dimension, the output shape is the
same as the shape given in `dense_shape`.
For `SparseTensor`s, the first (batch) column of the indices matrix is removed
(the indices matrix is a column vector), the values vector is unchanged, and
the first (`batch_size`) entry of the shape vector is removed (it is now a
single element vector).
One might see performance advantages by batching `Example` protos with
`parse_example` instead of using this function directly.
Args:
serialized: A scalar string Tensor, a single serialized Example.
features: A `dict` mapping feature keys to `FixedLenFeature` or
`VarLenFeature` values.
name: A name for this operation (optional).
example_names: (Optional) A scalar string Tensor, the associated name.
Returns:
A `dict` mapping feature keys to `Tensor` and `SparseTensor` values.
Raises:
ValueError: if any feature is invalid."
9067,parse_single_example_v2,tensorflow/tensorflow/python/ops/parsing_ops.py,415,function,"Parses a single `Example` proto.
Similar to `parse_example`, except:
For dense tensors, the returned `Tensor` is identical to the output of
`parse_example`, except there is no batch dimension, the output shape is the
same as the shape given in `dense_shape`.
For `SparseTensor`s, the first (batch) column of the indices matrix is removed
(the indices matrix is a column vector), the values vector is unchanged, and
the first (`batch_size`) entry of the shape vector is removed (it is now a
single element vector).
One might see performance advantages by batching `Example` protos with
`parse_example` instead of using this function directly.
Args:
serialized: A scalar string Tensor, a single serialized Example.
features: A `dict` mapping feature keys to `FixedLenFeature` or
`VarLenFeature` values.
example_names: (Optional) A scalar string Tensor, the associated name.
name: A name for this operation (optional).
Returns:
A `dict` mapping feature keys to `Tensor` and `SparseTensor` values.
Raises:
ValueError: if any feature is invalid."
9068,parse_sequence_example,tensorflow/tensorflow/python/ops/parsing_ops.py,457,function,"Parses a batch of `SequenceExample` protos.
Parses a vector of serialized
[`SequenceExample`](https://www.tensorflow.org/code/tensorflow/core/example/example.proto)
protos given in `serialized`.
This op parses serialized sequence examples into a tuple of dictionaries,
each mapping keys to `Tensor` and `SparseTensor` objects.
The first dictionary contains mappings for keys appearing in
`context_features`, and the second dictionary contains mappings for keys
appearing in `sequence_features`.
At least one of `context_features` and `sequence_features` must be provided
and non-empty.
The `context_features` keys are associated with a `SequenceExample` as a
whole, independent of time / frame. In contrast, the `sequence_features` keys
provide a way to access variable-length data within the `FeatureList` section
of the `SequenceExample` proto. While the shapes of `context_features` values
are fixed with respect to frame, the frame dimension (the first dimension)
of `sequence_features` values may vary between `SequenceExample` protos,
and even between `feature_list` keys within the same `SequenceExample`.
`context_features` contains `VarLenFeature`, `RaggedFeature`, and
`FixedLenFeature` objects. Each `VarLenFeature` is mapped to a
`SparseTensor`; each `RaggedFeature` is mapped to a `RaggedTensor`; and each
`FixedLenFeature` is mapped to a `Tensor`, of the specified type, shape, and
default value.
`sequence_features` contains `VarLenFeature`, `RaggedFeature`, and
`FixedLenSequenceFeature` objects. Each `VarLenFeature` is mapped to a
`SparseTensor`; each `RaggedFeature` is mapped to a `RaggedTensor; and
each `FixedLenSequenceFeature` is mapped to a `Tensor`, each of the specified
type. The shape will be `(B,T,) + df.dense_shape` for
`FixedLenSequenceFeature` `df`, where `B` is the batch size, and `T` is the
length of the associated `FeatureList` in the `SequenceExample`. For instance,
`FixedLenSequenceFeature([])` yields a scalar 2-D `Tensor` of static shape
`[None, None]` and dynamic shape `[B, T]`, while
`FixedLenSequenceFeature([k])` (for `int k >= 1`) yields a 3-D matrix `Tensor`
of static shape `[None, None, k]` and dynamic shape `[B, T, k]`.
Like the input, the resulting output tensors have a batch dimension. This
means that the original per-example shapes of `VarLenFeature`s and
`FixedLenSequenceFeature`s can be lost. To handle that situation, this op also
provides dicts of shape tensors as part of the output. There is one dict for
the context features, and one for the feature_list features. Context features
of type `FixedLenFeature`s will not be present, since their shapes are already
known by the caller. In situations where the input 'FixedLenFeature`s are of
different lengths across examples, the shorter examples will be padded with
default datatype values: 0 for numeric types, and the empty string for string
types.
Each `SparseTensor` corresponding to `sequence_features` represents a ragged
vector. Its indices are `[time, index]`, where `time` is the `FeatureList`
entry and `index` is the value's index in the list of values associated with
that time.
`FixedLenFeature` entries with a `default_value` and `FixedLenSequenceFeature`
entries with `allow_missing=True` are optional; otherwise, we will fail if
that `Feature` or `FeatureList` is missing from any example in `serialized`.
`example_name` may contain a descriptive name for the corresponding serialized
proto. This may be useful for debugging purposes, but it has no effect on the
output. If not `None`, `example_name` must be a scalar.
Args:
serialized: A vector (1-D Tensor) of type string containing binary
serialized `SequenceExample` protos.
context_features: A `dict` mapping feature keys to `FixedLenFeature` or
`VarLenFeature` or `RaggedFeature` values. These features are associated
with a `SequenceExample` as a whole.
sequence_features: A `dict` mapping feature keys to
`FixedLenSequenceFeature` or `VarLenFeature` or `RaggedFeature` values.
These features are associated with data within the `FeatureList` section
of the `SequenceExample` proto.
example_names: A vector (1-D Tensor) of strings (optional), the name of the
serialized protos.
name: A name for this operation (optional).
Returns:
A tuple of three `dict`s, each mapping keys to `Tensor`s,
`SparseTensor`s, and `RaggedTensor`. The first dict contains the context
key/values, the second dict contains the feature_list key/values, and the
final dict contains the lengths of any dense feature_list features.
Raises:
ValueError: if any feature is invalid."
9069,parse_single_sequence_example,tensorflow/tensorflow/python/ops/parsing_ops.py,702,function,"Parses a single `SequenceExample` proto.
Parses a single serialized [`SequenceExample`](https://www.tensorflow.org/code/tensorflow/core/example/example.proto)
proto given in `serialized`.
This op parses a serialized sequence example into a tuple of dictionaries,
each mapping keys to `Tensor` and `SparseTensor` objects.
The first dictionary contains mappings for keys appearing in
`context_features`, and the second dictionary contains mappings for keys
appearing in `sequence_features`.
At least one of `context_features` and `sequence_features` must be provided
and non-empty.
The `context_features` keys are associated with a `SequenceExample` as a
whole, independent of time / frame. In contrast, the `sequence_features` keys
provide a way to access variable-length data within the `FeatureList` section
of the `SequenceExample` proto. While the shapes of `context_features` values
are fixed with respect to frame, the frame dimension (the first dimension)
of `sequence_features` values may vary between `SequenceExample` protos,
and even between `feature_list` keys within the same `SequenceExample`.
`context_features` contains `VarLenFeature`, `RaggedFeature`, and
`FixedLenFeature` objects. Each `VarLenFeature` is mapped to a `SparseTensor`;
each `RaggedFeature` is mapped to a `RaggedTensor`; and each `FixedLenFeature`
is mapped to a `Tensor`, of the specified type, shape, and default value.
`sequence_features` contains `VarLenFeature`, `RaggedFeature`, and
`FixedLenSequenceFeature` objects. Each `VarLenFeature` is mapped to a
`SparseTensor`; each `RaggedFeature` is mapped to a `RaggedTensor`; and each
`FixedLenSequenceFeature` is mapped to a `Tensor`, each of the specified type.
The shape will be `(T,) + df.dense_shape` for `FixedLenSequenceFeature` `df`,
where `T` is the length of the associated `FeatureList` in the
`SequenceExample`. For instance, `FixedLenSequenceFeature([])` yields a scalar
1-D `Tensor` of static shape `[None]` and dynamic shape `[T]`, while
`FixedLenSequenceFeature([k])` (for `int k >= 1`) yields a 2-D matrix `Tensor`
of static shape `[None, k]` and dynamic shape `[T, k]`.
Each `SparseTensor` corresponding to `sequence_features` represents a ragged
vector. Its indices are `[time, index]`, where `time` is the `FeatureList`
entry and `index` is the value's index in the list of values associated with
that time.
`FixedLenFeature` entries with a `default_value` and `FixedLenSequenceFeature`
entries with `allow_missing=True` are optional; otherwise, we will fail if
that `Feature` or `FeatureList` is missing from any example in `serialized`.
`example_name` may contain a descriptive name for the corresponding serialized
proto. This may be useful for debugging purposes, but it has no effect on the
output. If not `None`, `example_name` must be a scalar.
Note that the batch version of this function, `tf.parse_sequence_example`,
is written for better memory efficiency and will be faster on large
`SequenceExample`s.
Args:
serialized: A scalar (0-D Tensor) of type string, a single binary
serialized `SequenceExample` proto.
context_features: A `dict` mapping feature keys to `FixedLenFeature` or
`VarLenFeature` or `RaggedFeature` values. These features are associated
with a `SequenceExample` as a whole.
sequence_features: A `dict` mapping feature keys to
`FixedLenSequenceFeature` or `VarLenFeature` or `RaggedFeature` values.
These features are associated with data within the `FeatureList` section
of the `SequenceExample` proto.
example_name: A scalar (0-D Tensor) of strings (optional), the name of
the serialized proto.
name: A name for this operation (optional).
Returns:
A tuple of two `dict`s, each mapping keys to `Tensor`s and `SparseTensor`s
and `RaggedTensor`s.
* The first dict contains the context key/values.
* The second dict contains the feature_list key/values.
Raises:
ValueError: if any feature is invalid."
9070,decode_raw,tensorflow/tensorflow/python/ops/parsing_ops.py,846,function,"Convert raw byte strings into tensors.
Args:
input_bytes:
Each element of the input Tensor is converted to an array of bytes.
out_type:
`DType` of the output. Acceptable types are `half`, `float`, `double`,
`int32`, `uint16`, `uint8`, `int16`, `int8`, `int64`.
little_endian:
Whether the `input_bytes` data is in little-endian format. Data will be
converted into host byte order if necessary.
fixed_length:
If set, the first `fixed_length` bytes of each element will be converted.
Data will be zero-padded or truncated to the specified length.
`fixed_length` must be a multiple of the size of `out_type`.
`fixed_length` must be specified if the elements of `input_bytes` are of
variable length.
name: A name for the operation (optional).
Returns:
A `Tensor` object storing the decoded bytes."
9071,decode_raw_v1,tensorflow/tensorflow/python/ops/parsing_ops.py,892,function,"Convert raw byte strings into tensors.
Args:
input_bytes:
Each element of the input Tensor is converted to an array of bytes.
out_type:
`DType` of the output. Acceptable types are `half`, `float`, `double`,
`int32`, `uint16`, `uint8`, `int16`, `int8`, `int64`.
little_endian:
Whether the `input_bytes` data is in little-endian format. Data will be
converted into host byte order if necessary.
name: A name for the operation (optional).
bytes: Deprecated parameter. Use `input_bytes` instead.
Returns:
A `Tensor` object storing the decoded bytes."
9072,decode_csv,tensorflow/tensorflow/python/ops/parsing_ops.py,935,function,"Convert CSV records to tensors. Each column maps to one tensor.
RFC 4180 format is expected for the CSV records.
(https://tools.ietf.org/html/rfc4180)
Note that we allow leading and trailing spaces with int or float field.
Args:
records: A `Tensor` of type `string`.
Each string is a record/row in the csv and all records should have
the same format.
record_defaults: A list of `Tensor` objects with specific types.
Acceptable types are `float32`, `float64`, `int32`, `int64`, `string`.
One tensor per column of the input record, with either a
scalar default value for that column or an empty vector if the column is
required.
field_delim: An optional `string`. Defaults to `"",""`.
char delimiter to separate fields in a record.
use_quote_delim: An optional `bool`. Defaults to `True`.
If false, treats double quotation marks as regular
characters inside of the string fields (ignoring RFC 4180, Section 2,
Bullet 5).
name: A name for the operation (optional).
na_value: Additional string to recognize as NA/NaN.
select_cols: Optional sorted list of column indices to select. If specified,
only this subset of columns will be parsed and returned.
Returns:
A list of `Tensor` objects. Has the same type as `record_defaults`.
Each tensor will have the same shape as records.
Raises:
ValueError: If any of the arguments is malformed."
9073,decode_csv_v2,tensorflow/tensorflow/python/ops/parsing_ops.py,984,function,"Convert CSV records to tensors. Each column maps to one tensor.
RFC 4180 format is expected for the CSV records.
(https://tools.ietf.org/html/rfc4180)
Note that we allow leading and trailing spaces with int or float field.
Args:
records: A `Tensor` of type `string`.
Each string is a record/row in the csv and all records should have
the same format.
record_defaults: A list of `Tensor` objects with specific types.
Acceptable types are `float32`, `float64`, `int32`, `int64`, `string`.
One tensor per column of the input record, with either a
scalar default value for that column or an empty vector if the column is
required.
field_delim: An optional `string`. Defaults to `"",""`.
char delimiter to separate fields in a record.
use_quote_delim: An optional `bool`. Defaults to `True`.
If false, treats double quotation marks as regular
characters inside of the string fields (ignoring RFC 4180, Section 2,
Bullet 5).
na_value: Additional string to recognize as NA/NaN.
select_cols: Optional sorted list of column indices to select. If specified,
only this subset of columns will be parsed and returned.
name: A name for the operation (optional).
Returns:
A list of `Tensor` objects. Has the same type as `record_defaults`.
Each tensor will have the same shape as records.
Raises:
ValueError: If any of the arguments is malformed."
9074,variable_axis_size_partitioner,tensorflow/tensorflow/python/ops/partitioned_variables.py,72,function,"Get a partitioner for VariableScope to keep shards below `max_shard_bytes`.
This partitioner will shard a Variable along one axis, attempting to keep
the maximum shard size below `max_shard_bytes`. In practice, this is not
always possible when sharding along only one axis. When this happens,
this axis is sharded as much as possible (i.e., every dimension becomes
a separate shard).
If the partitioner hits the `max_shards` limit, then each shard may end up
larger than `max_shard_bytes`. By default `max_shards` equals `None` and no
limit on the number of shards is enforced.
One reasonable value for `max_shard_bytes` is `(64 << 20) - 1`, or almost
`64MB`, to keep below the protobuf byte limit.
Args:
max_shard_bytes: The maximum size any given shard is allowed to be.
axis: The axis to partition along. Default: outermost axis.
bytes_per_string_element: If the `Variable` is of type string, this provides
an estimate of how large each scalar in the `Variable` is.
max_shards: The maximum number of shards in int created taking precedence
over `max_shard_bytes`.
Returns:
A partition function usable as the `partitioner` argument to
`variable_scope` and `get_variable`.
Raises:
ValueError: If any of the byte counts are non-positive."
9075,min_max_variable_partitioner,tensorflow/tensorflow/python/ops/partitioned_variables.py,158,function,"Partitioner to allocate minimum size per slice.
Returns a partitioner that partitions the variable of given shape and dtype
such that each partition has a minimum of `min_slice_size` slice of the
variable. The maximum number of such partitions (upper bound) is given by
`max_partitions`.
Args:
max_partitions: Upper bound on the number of partitions. Defaults to 1.
axis: Axis along which to partition the variable. Defaults to 0.
min_slice_size: Minimum size of the variable slice per partition. Defaults
to 256K.
bytes_per_string_element: If the `Variable` is of type string, this provides
an estimate of how large each scalar in the `Variable` is.
Returns:
A partition function usable as the `partitioner` argument to
`variable_scope` and `get_variable`."
9076,fixed_size_partitioner,tensorflow/tensorflow/python/ops/partitioned_variables.py,222,function,"Partitioner to specify a fixed number of shards along given axis.
Args:
num_shards: `int`, number of shards to partition variable.
axis: `int`, axis to partition on.
Returns:
A partition function usable as the `partitioner` argument to
`variable_scope` and `get_variable`."
9077,create_partitioned_variables,tensorflow/tensorflow/python/ops/partitioned_variables.py,244,function,"Create a list of partitioned variables according to the given `slicing`.
Currently only one dimension of the full variable can be sliced, and the
full variable can be reconstructed by the concatenation of the returned
list along that dimension.
Args:
shape: List of integers. The shape of the full variable.
slicing: List of integers. How to partition the variable.
Must be of the same length as `shape`. Each value
indicate how many slices to create in the corresponding
dimension. Presently only one of the values can be more than 1;
that is, the variable can only be sliced along one dimension.
For convenience, The requested number of partitions does not have to
divide the corresponding dimension evenly. If it does not, the
shapes of the partitions are incremented by 1 starting from partition
0 until all slack is absorbed. The adjustment rules may change in the
future, but as you can save/restore these variables with different
slicing specifications this should not be a problem.
initializer: A `Tensor` of shape `shape` or a variable initializer
function. If a function, it will be called once for each slice,
passing the shape and data type of the slice as parameters. The
function must return a tensor with the same shape as the slice.
dtype: Type of the variables. Ignored if `initializer` is a `Tensor`.
trainable: If True also add all the variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES`.
collections: List of graph collections keys to add the variables to.
Defaults to `[GraphKeys.GLOBAL_VARIABLES]`.
name: Optional name for the full variable. Defaults to
`""PartitionedVariable""` and gets uniquified automatically.
reuse: Boolean or `None`; if `True` and name is set, it would reuse
previously created variables. if `False` it will create new variables.
if `None`, it would inherit the parent scope reuse.
Returns:
A list of Variables corresponding to the slicing.
Raises:
ValueError: If any of the arguments is malformed."
9078,add_leading_unit_dimensions,tensorflow/tensorflow/python/ops/random_grad.py,33,function,
9079,random_normal,tensorflow/tensorflow/python/ops/random_ops.py,46,function,"Outputs random values from a normal distribution.
Example that generates a new set of random values every time:
>>> tf.random.set_seed(5);
>>> tf.random.normal([4], 0, 1, tf.float32)
<tf.Tensor: shape=(4,), dtype=float32, numpy=..., dtype=float32)>
Example that outputs a reproducible result:
>>> tf.random.set_seed(5);
>>> tf.random.normal([2,2], 0, 1, tf.float32, seed=1)
<tf.Tensor: shape=(2, 2), dtype=float32, numpy=
array([[-1.3768897 , -0.01258316],
[-0.169515 , 1.0824056 ]], dtype=float32)>
In this case, we are setting both the global and operation-level seed to
ensure this result is reproducible. See `tf.random.set_seed` for more
information.
Args:
shape: A 1-D integer Tensor or Python array. The shape of the output tensor.
mean: A Tensor or Python value of type `dtype`, broadcastable with `stddev`.
The mean of the normal distribution.
stddev: A Tensor or Python value of type `dtype`, broadcastable with `mean`.
The standard deviation of the normal distribution.
dtype: The type of the output.
seed: A Python integer. Used to create a random seed for the distribution.
See
`tf.random.set_seed`
for behavior.
name: A name for the operation (optional).
Returns:
A tensor of the specified shape filled with random normal values."
9080,parameterized_truncated_normal,tensorflow/tensorflow/python/ops/random_ops.py,104,function,"Outputs random values from a truncated normal distribution.
The generated values follow a normal distribution with specified mean and
standard deviation, except that values whose magnitude is more than 2 standard
deviations from the mean are dropped and re-picked.
Args:
shape: A 1-D integer Tensor or Python array. The shape of the output tensor.
means: A 0-D Tensor or Python value of type `dtype`. The mean of the
truncated normal distribution.
stddevs: A 0-D Tensor or Python value of type `dtype`. The standard
deviation of the truncated normal distribution.
minvals: A 0-D Tensor or Python value of type `dtype`. The minimum value of
the truncated normal distribution.
maxvals: A 0-D Tensor or Python value of type `dtype`. The maximum value of
the truncated normal distribution.
dtype: The type of the output.
seed: A Python integer. Used to create a random seed for the distribution.
See
`tf.random.set_seed`
for behavior.
name: A name for the operation (optional).
Returns:
A tensor of the specified shape filled with random truncated normal values."
9081,truncated_normal,tensorflow/tensorflow/python/ops/random_ops.py,162,function,"Outputs random values from a truncated normal distribution.
The generated values follow a normal distribution with specified mean and
standard deviation, except that values whose magnitude is more than 2 standard
deviations from the mean are dropped and re-picked.
Args:
shape: A 1-D integer Tensor or Python array. The shape of the output tensor.
mean: A 0-D Tensor or Python value of type `dtype`. The mean of the
truncated normal distribution.
stddev: A 0-D Tensor or Python value of type `dtype`. The standard deviation
of the normal distribution, before truncation.
dtype: The type of the output.
seed: A Python integer. Used to create a random seed for the distribution.
See
`tf.random.set_seed`
for behavior.
name: A name for the operation (optional).
Returns:
A tensor of the specified shape filled with random truncated normal values."
9082,random_uniform,tensorflow/tensorflow/python/ops/random_ops.py,210,function,"Outputs random values from a uniform distribution.
The generated values follow a uniform distribution in the range
`[minval, maxval)`. The lower bound `minval` is included in the range, while
the upper bound `maxval` is excluded.
For floats, the default range is `[0, 1)`. For ints, at least `maxval` must
be specified explicitly.
In the integer case, the random integers are slightly biased unless
`maxval - minval` is an exact power of two. The bias is small for values of
`maxval - minval` significantly smaller than the range of the output (either
`2**32` or `2**64`).
Examples:
>>> tf.random.uniform(shape=[2])
<tf.Tensor: shape=(2,), dtype=float32, numpy=array([..., ...], dtype=float32)>
>>> tf.random.uniform(shape=[], minval=-1., maxval=0.)
<tf.Tensor: shape=(), dtype=float32, numpy=-...>
>>> tf.random.uniform(shape=[], minval=5, maxval=10, dtype=tf.int64)
<tf.Tensor: shape=(), dtype=int64, numpy=...>
The `seed` argument produces a deterministic sequence of tensors across
multiple calls. To repeat that sequence, use `tf.random.set_seed`:
>>> tf.random.set_seed(5)
>>> tf.random.uniform(shape=[], maxval=3, dtype=tf.int32, seed=10)
<tf.Tensor: shape=(), dtype=int32, numpy=2>
>>> tf.random.uniform(shape=[], maxval=3, dtype=tf.int32, seed=10)
<tf.Tensor: shape=(), dtype=int32, numpy=0>
>>> tf.random.set_seed(5)
>>> tf.random.uniform(shape=[], maxval=3, dtype=tf.int32, seed=10)
<tf.Tensor: shape=(), dtype=int32, numpy=2>
>>> tf.random.uniform(shape=[], maxval=3, dtype=tf.int32, seed=10)
<tf.Tensor: shape=(), dtype=int32, numpy=0>
Without `tf.random.set_seed` but with a `seed` argument is specified, small
changes to function graphs or previously executed operations will change the
returned value. See `tf.random.set_seed` for details.
Args:
shape: A 1-D integer Tensor or Python array. The shape of the output tensor.
minval: A Tensor or Python value of type `dtype`, broadcastable with
`shape` (for integer types, broadcasting is not supported, so it needs to
be a scalar). The lower bound on the range of random values to generate
(inclusive). Defaults to 0.
maxval: A Tensor or Python value of type `dtype`, broadcastable with
`shape` (for integer types, broadcasting is not supported, so it needs to
be a scalar). The upper bound on the range of random values to generate
(exclusive). Defaults to 1 if `dtype` is floating point.
dtype: The type of the output: `float16`, `float32`, `float64`, `int32`,
or `int64`.
seed: A Python integer. Used in combination with `tf.random.set_seed` to
create a reproducible sequence of tensors across multiple calls.
name: A name for the operation (optional).
Returns:
A tensor of the specified shape filled with random uniform values.
Raises:
ValueError: If `dtype` is integral and `maxval` is not specified."
9083,random_shuffle,tensorflow/tensorflow/python/ops/random_ops.py,322,function,"Randomly shuffles a tensor along its first dimension.
The tensor is shuffled along dimension 0, such that each `value[j]` is mapped
to one and only one `output[i]`. For example, a mapping that might occur for a
3x2 tensor is:
```python
[[1, 2], [[5, 6],
[3, 4], ==> [1, 2],
[5, 6]] [3, 4]]
```
Args:
value: A Tensor to be shuffled.
seed: A Python integer. Used to create a random seed for the distribution.
See
`tf.random.set_seed`
for behavior.
name: A name for the operation (optional).
Returns:
A tensor of same shape and type as `value`, shuffled along its first
dimension."
9084,random_crop,tensorflow/tensorflow/python/ops/random_ops.py,355,function,"Randomly crops a tensor to a given size.
Slices a shape `size` portion out of `value` at a uniformly chosen offset.
Requires `value.shape >= size`.
If a dimension should not be cropped, pass the full size of that dimension.
For example, RGB images can be cropped with
`size = [crop_height, crop_width, 3]`.
Args:
value: Input tensor to crop.
size: 1-D tensor with size the rank of `value`.
seed: Python integer. Used to create a random seed. See
`tf.random.set_seed`
for behavior.
name: A name for this operation (optional).
Returns:
A cropped tensor of the same rank as `value` and shape `size`."
9085,multinomial,tensorflow/tensorflow/python/ops/random_ops.py,401,function,"Draws samples from a multinomial distribution.
Example:
```python
# samples has shape [1, 5], where each value is either 0 or 1 with equal
# probability.
samples = tf.random.categorical(tf.math.log([[0.5, 0.5]]), 5)
```
Args:
logits: 2-D Tensor with shape `[batch_size, num_classes]`. Each slice
`[i, :]` represents the unnormalized log-probabilities for all classes.
num_samples: 0-D. Number of independent samples to draw for each row slice.
seed: A Python integer. Used to create a random seed for the distribution.
See `tf.random.set_seed` for behavior.
name: Optional name for the operation.
output_dtype: integer type to use for the output. Defaults to int64.
Returns:
The drawn samples of shape `[batch_size, num_samples]`."
9086,categorical,tensorflow/tensorflow/python/ops/random_ops.py,429,function,"Draws samples from a categorical distribution.
Example:
```python
# samples has shape [1, 5], where each value is either 0 or 1 with equal
# probability.
samples = tf.random.categorical(tf.math.log([[0.5, 0.5]]), 5)
```
Args:
logits: 2-D Tensor with shape `[batch_size, num_classes]`. Each slice
`[i, :]` represents the unnormalized log-probabilities for all classes.
num_samples: 0-D. Number of independent samples to draw for each row slice.
dtype: integer type to use for the output. Defaults to int64.
seed: A Python integer. Used to create a random seed for the distribution.
See `tf.random.set_seed` for behavior.
name: Optional name for the operation.
Returns:
The drawn samples of shape `[batch_size, num_samples]`."
9087,multinomial_categorical_impl,tensorflow/tensorflow/python/ops/random_ops.py,456,function,Implementation for random.categorical (v1) and random.categorical (v2).
9088,random_gamma,tensorflow/tensorflow/python/ops/random_ops.py,480,function,"Draws `shape` samples from each of the given Gamma distribution(s).
`alpha` is the shape parameter describing the distribution(s), and `beta` is
the inverse scale parameter(s).
Note: Because internal calculations are done using `float64` and casting has
`floor` semantics, we must manually map zero outcomes to the smallest
possible positive floating-point value, i.e., `np.finfo(dtype).tiny`. This
means that `np.finfo(dtype).tiny` occurs more frequently than it otherwise
should. This bias can only happen for small values of `alpha`, i.e.,
`alpha << 1` or large values of `beta`, i.e., `beta >> 1`.
The samples are differentiable w.r.t. alpha and beta.
The derivatives are computed using the approach described in
(Figurnov et al., 2018).
Example:
```python
samples = tf.random.gamma([10], [0.5, 1.5])
# samples has shape [10, 2], where each slice [:, 0] and [:, 1] represents
# the samples drawn from each distribution
samples = tf.random.gamma([7, 5], [0.5, 1.5])
# samples has shape [7, 5, 2], where each slice [:, :, 0] and [:, :, 1]
# represents the 7x5 samples drawn from each of the two distributions
alpha = tf.constant([[1.],[3.],[5.]])
beta = tf.constant([[3., 4.]])
samples = tf.random.gamma([30], alpha=alpha, beta=beta)
# samples has shape [30, 3, 2], with 30 samples each of 3x2 distributions.
loss = tf.reduce_mean(tf.square(samples))
dloss_dalpha, dloss_dbeta = tf.gradients(loss, [alpha, beta])
# unbiased stochastic derivatives of the loss function
alpha.shape == dloss_dalpha.shape # True
beta.shape == dloss_dbeta.shape # True
```
Args:
shape: A 1-D integer Tensor or Python array. The shape of the output samples
to be drawn per alpha/beta-parameterized distribution.
alpha: A Tensor or Python value or N-D array of type `dtype`. `alpha`
provides the shape parameter(s) describing the gamma distribution(s) to
sample. Must be broadcastable with `beta`.
beta: A Tensor or Python value or N-D array of type `dtype`. Defaults to 1.
`beta` provides the inverse scale parameter(s) of the gamma
distribution(s) to sample. Must be broadcastable with `alpha`.
dtype: The type of alpha, beta, and the output: `float16`, `float32`, or
`float64`.
seed: A Python integer. Used to create a random seed for the distributions.
See
`tf.random.set_seed`
for behavior.
name: Optional name for the operation.
Returns:
samples: a `Tensor` of shape
`tf.concat([shape, tf.shape(alpha + beta)], axis=0)` with values of type
`dtype`.
References:
Implicit Reparameterization Gradients:
[Figurnov et al., 2018]
(http://papers.nips.cc/paper/7326-implicit-reparameterization-gradients)
([pdf]
(http://papers.nips.cc/paper/7326-implicit-reparameterization-gradients.pdf))"
9089,random_poisson,tensorflow/tensorflow/python/ops/random_ops.py,574,function,"Draws `shape` samples from each of the given Poisson distribution(s).
`lam` is the rate parameter describing the distribution(s).
Example:
```python
samples = tf.random.poisson([0.5, 1.5], [10])
# samples has shape [10, 2], where each slice [:, 0] and [:, 1] represents
# the samples drawn from each distribution
samples = tf.random.poisson([12.2, 3.3], [7, 5])
# samples has shape [7, 5, 2], where each slice [:, :, 0] and [:, :, 1]
# represents the 7x5 samples drawn from each of the two distributions
```
Args:
lam: A Tensor or Python value or N-D array of type `dtype`.
`lam` provides the rate parameter(s) describing the poisson
distribution(s) to sample.
shape: A 1-D integer Tensor or Python array. The shape of the output samples
to be drawn per ""rate""-parameterized distribution.
dtype: The type of the output: `float16`, `float32`, `float64`, `int32` or
`int64`.
seed: A Python integer. Used to create a random seed for the distributions.
See
`tf.random.set_seed`
for behavior.
name: Optional name for the operation.
Returns:
samples: a `Tensor` of shape `tf.concat([shape, tf.shape(lam)], axis=0)`
with values of type `dtype`."
9090,random_poisson_v2,tensorflow/tensorflow/python/ops/random_ops.py,614,function,"Draws `shape` samples from each of the given Poisson distribution(s).
`lam` is the rate parameter describing the distribution(s).
Example:
```python
samples = tf.random.poisson([10], [0.5, 1.5])
# samples has shape [10, 2], where each slice [:, 0] and [:, 1] represents
# the samples drawn from each distribution
samples = tf.random.poisson([7, 5], [12.2, 3.3])
# samples has shape [7, 5, 2], where each slice [:, :, 0] and [:, :, 1]
# represents the 7x5 samples drawn from each of the two distributions
```
Args:
shape: A 1-D integer Tensor or Python array. The shape of the output samples
to be drawn per ""rate""-parameterized distribution.
lam: A Tensor or Python value or N-D array of type `dtype`.
`lam` provides the rate parameter(s) describing the poisson
distribution(s) to sample.
dtype: The type of the output: `float16`, `float32`, `float64`, `int32` or
`int64`.
seed: A Python integer. Used to create a random seed for the distributions.
See
`tf.random.set_seed`
for behavior.
name: Optional name for the operation.
Returns:
samples: a `Tensor` of shape `tf.concat([shape, tf.shape(lam)], axis=0)`
with values of type `dtype`."
9091,get_resource_handle_data,tensorflow/tensorflow/python/ops/resource_variable_ops.py,66,function,
9092,get_eager_safe_handle_data,tensorflow/tensorflow/python/ops/resource_variable_ops.py,76,function,Get the data handle from the Tensor `handle`.
9093,eager_safe_variable_handle,tensorflow/tensorflow/python/ops/resource_variable_ops.py,200,function,"Creates a variable handle with information to do shape inference.
The dtype is read from `initial_value` and stored in the returned
resource tensor's handle data.
If `initial_value.dtype == tf.variant`, we additionally extract the handle
data (if any) from `initial_value` and append it to the `handle_data`.
In this case, the returned tensor's handle data is in the form
```
is_set: true
shape_and_type {
shape {
// initial_value.shape
}
dtype: DT_VARIANT
}
shape_and_type {
// handle_data(initial_value).shape_and_type[0]
}
shape_and_type {
// handle_data(initial_value).shape_and_type[1]
}
...
```
Ops that read from this tensor, such as `ReadVariableOp` and
`AssignVariableOp`, know that `handle_data(handle).shape_and_type[1:]`
correspond to the handle data of the variant(s) stored in the Variable.
Args:
initial_value: A `Tensor`.
shape: The shape of the handle data. Can be `TensorShape(None)`
(i.e. unknown shape).
shared_name: A string.
name: A string.
graph_mode: A python bool.
Returns:
The handle, a `Tensor` of type `resource`."
9094,EagerResourceDeleter,tensorflow/tensorflow/python/ops/resource_variable_ops.py,260,class,"An object which cleans up a resource handle.
An alternative to defining a __del__ method on an object. The intended use is
that ResourceVariables or other objects with resource handles will maintain a
single reference to this object. When the parent object is collected, this
object will be too. Even if the parent object is part of a reference cycle,
the cycle will be collectable."
9095,shape_safe_assign_variable_handle,tensorflow/tensorflow/python/ops/resource_variable_ops.py,311,function,Helper that checks shape compatibility and assigns variable.
9096,variable_accessed,tensorflow/tensorflow/python/ops/resource_variable_ops.py,333,function,Records that `variable` was accessed for the tape and FuncGraph.
9097,BaseResourceVariable,tensorflow/tensorflow/python/ops/resource_variable_ops.py,341,class,A python variable from an existing handle.
9098,dtype,tensorflow/tensorflow/python/ops/resource_variable_ops.py,520,method,The dtype of this variable.
9099,device,tensorflow/tensorflow/python/ops/resource_variable_ops.py,525,method,The device this variable is on.
9100,graph,tensorflow/tensorflow/python/ops/resource_variable_ops.py,530,method,The `Graph` of this variable.
9101,name,tensorflow/tensorflow/python/ops/resource_variable_ops.py,535,method,The name of the handle for this variable.
9102,shape,tensorflow/tensorflow/python/ops/resource_variable_ops.py,540,method,The shape of this variable.
9103,set_shape,tensorflow/tensorflow/python/ops/resource_variable_ops.py,544,method,
9104,create,tensorflow/tensorflow/python/ops/resource_variable_ops.py,559,method,The op responsible for initializing this variable.
9105,handle,tensorflow/tensorflow/python/ops/resource_variable_ops.py,567,method,The handle by which this variable can be accessed.
9106,value,tensorflow/tensorflow/python/ops/resource_variable_ops.py,571,method,A cached operation which reads the value of this variable.
9107,initializer,tensorflow/tensorflow/python/ops/resource_variable_ops.py,583,method,The op responsible for initializing this variable.
9108,initial_value,tensorflow/tensorflow/python/ops/resource_variable_ops.py,588,method,Returns the Tensor used as the initial value for the variable.
9109,constraint,tensorflow/tensorflow/python/ops/resource_variable_ops.py,595,method,"Returns the constraint function associated with this variable.
Returns:
The constraint function that was passed to the variable constructor.
Can be `None` if no constraint was passed."
9110,op,tensorflow/tensorflow/python/ops/resource_variable_ops.py,605,method,The op for this variable.
9111,trainable,tensorflow/tensorflow/python/ops/resource_variable_ops.py,610,method,
9112,synchronization,tensorflow/tensorflow/python/ops/resource_variable_ops.py,614,method,
9113,aggregation,tensorflow/tensorflow/python/ops/resource_variable_ops.py,618,method,
9114,eval,tensorflow/tensorflow/python/ops/resource_variable_ops.py,621,method,Evaluates and returns the value of this variable.
9115,numpy,tensorflow/tensorflow/python/ops/resource_variable_ops.py,627,method,
9116,count_up_to,tensorflow/tensorflow/python/ops/resource_variable_ops.py,634,method,"Increments this variable until it reaches `limit`.
When that Op is run it tries to increment the variable by `1`. If
incrementing the variable would bring it above `limit` then the Op raises
the exception `OutOfRangeError`.
If no error is raised, the Op outputs the value of the variable before
the increment.
This is essentially a shortcut for `count_up_to(self, limit)`.
Args:
limit: value at which incrementing the variable raises an error.
Returns:
A `Tensor` that will hold the variable value before the increment. If no
other Op modifies this variable, the values produced will all be
distinct."
9117,read_value,tensorflow/tensorflow/python/ops/resource_variable_ops.py,694,method,"Constructs an op which reads the value of this variable.
Should be used when there are multiple reads, or when it is desirable to
read the value only after some condition is true.
Returns:
the read operation."
9118,sparse_read,tensorflow/tensorflow/python/ops/resource_variable_ops.py,709,method,"Reads the value of this variable sparsely, using `gather`."
9119,gather_nd,tensorflow/tensorflow/python/ops/resource_variable_ops.py,728,method,"Reads the value of this variable sparsely, using `gather_nd`."
9120,to_proto,tensorflow/tensorflow/python/ops/resource_variable_ops.py,738,method,"Converts a `ResourceVariable` to a `VariableDef` protocol buffer.
Args:
export_scope: Optional `string`. Name scope to remove.
Raises:
RuntimeError: If run in EAGER mode.
Returns:
A `VariableDef` protocol buffer, or `None` if the `Variable` is not
in the specified name scope."
9121,from_proto,tensorflow/tensorflow/python/ops/resource_variable_ops.py,784,method,
9122,is_initialized,tensorflow/tensorflow/python/ops/resource_variable_ops.py,792,method,"Checks whether a resource variable has been initialized.
Outputs boolean scalar indicating whether the tensor has been initialized.
Args:
name: A name for the operation (optional).
Returns:
A `Tensor` of type `bool`."
9123,assign_sub,tensorflow/tensorflow/python/ops/resource_variable_ops.py,805,method,"Subtracts a value from this variable.
Args:
delta: A `Tensor`. The value to subtract from this variable.
use_locking: If `True`, use locking during the operation.
name: The name to use for the operation.
read_value: A `bool`. Whether to read and return the new value of the
variable or not.
Returns:
If `read_value` is `True`, this method will return the new value of the
variable after the assignment has completed. Otherwise, when in graph mode
it will return the `Operation` that does the assignment, and when in eager
mode it will return `None`."
9124,assign_add,tensorflow/tensorflow/python/ops/resource_variable_ops.py,832,method,"Adds a value to this variable.
Args:
delta: A `Tensor`. The value to add to this variable.
use_locking: If `True`, use locking during the operation.
name: The name to use for the operation.
read_value: A `bool`. Whether to read and return the new value of the
variable or not.
Returns:
If `read_value` is `True`, this method will return the new value of the
variable after the assignment has completed. Otherwise, when in graph mode
it will return the `Operation` that does the assignment, and when in eager
mode it will return `None`."
9125,assign,tensorflow/tensorflow/python/ops/resource_variable_ops.py,864,method,"Assigns a new value to this variable.
Args:
value: A `Tensor`. The new value for this variable.
use_locking: If `True`, use locking during the assignment.
name: The name to use for the assignment.
read_value: A `bool`. Whether to read and return the new value of the
variable or not.
Returns:
If `read_value` is `True`, this method will return the new value of the
variable after the assignment has completed. Otherwise, when in graph mode
it will return the `Operation` that does the assignment, and when in eager
mode it will return `None`."
9126,scatter_sub,tensorflow/tensorflow/python/ops/resource_variable_ops.py,902,method,"Subtracts `tf.IndexedSlices` from this variable.
Args:
sparse_delta: `tf.IndexedSlices` to be subtracted from this variable.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
The updated variable.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9127,scatter_add,tensorflow/tensorflow/python/ops/resource_variable_ops.py,922,method,"Adds `tf.IndexedSlices` to this variable.
Args:
sparse_delta: `tf.IndexedSlices` to be added to this variable.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
The updated variable.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9128,scatter_max,tensorflow/tensorflow/python/ops/resource_variable_ops.py,942,method,"Updates this variable with the max of `tf.IndexedSlices` and itself.
Args:
sparse_delta: `tf.IndexedSlices` to use as an argument of max
with this variable.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
The updated variable.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9129,scatter_min,tensorflow/tensorflow/python/ops/resource_variable_ops.py,963,method,"Updates this variable with the min of `tf.IndexedSlices` and itself.
Args:
sparse_delta: `tf.IndexedSlices` to use as an argument of min
with this variable.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
The updated variable.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9130,scatter_mul,tensorflow/tensorflow/python/ops/resource_variable_ops.py,984,method,"Multiply this variable by `tf.IndexedSlices`.
Args:
sparse_delta: `tf.IndexedSlices` to multiply this variable by.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
The updated variable.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9131,scatter_div,tensorflow/tensorflow/python/ops/resource_variable_ops.py,1004,method,"Divide this variable by `tf.IndexedSlices`.
Args:
sparse_delta: `tf.IndexedSlices` to divide this variable by.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
The updated variable.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9132,scatter_update,tensorflow/tensorflow/python/ops/resource_variable_ops.py,1024,method,"Assigns `tf.IndexedSlices` to this variable.
Args:
sparse_delta: `tf.IndexedSlices` to be assigned to this variable.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
The updated variable.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9133,batch_scatter_update,tensorflow/tensorflow/python/ops/resource_variable_ops.py,1044,method,"Assigns `tf.IndexedSlices` to this variable batch-wise.
Analogous to `batch_gather`. This assumes that this variable and the
sparse_delta IndexedSlices have a series of leading dimensions that are the
same for all of them, and the updates are performed on the last dimension of
indices. In other words, the dimensions should be the following:
`num_prefix_dims = sparse_delta.indices.ndims - 1`
`batch_dim = num_prefix_dims + 1`
`sparse_delta.updates.shape = sparse_delta.indices.shape + var.shape[
batch_dim:]`
where
`sparse_delta.updates.shape[:num_prefix_dims]`
`== sparse_delta.indices.shape[:num_prefix_dims]`
`== var.shape[:num_prefix_dims]`
And the operation performed can be expressed as:
`var[i_1, ..., i_n,
sparse_delta.indices[i_1, ..., i_n, j]] = sparse_delta.updates[
i_1, ..., i_n, j]`
When sparse_delta.indices is a 1D tensor, this operation is equivalent to
`scatter_update`.
To avoid this operation one can looping over the first `ndims` of the
variable and using `scatter_update` on the subtensors that result of slicing
the first dimension. This is a valid option for `ndims = 1`, but less
efficient than this implementation.
Args:
sparse_delta: `tf.IndexedSlices` to be assigned to this variable.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
The updated variable.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9134,scatter_nd_sub,tensorflow/tensorflow/python/ops/resource_variable_ops.py,1094,method,"Applies sparse subtraction to individual values or slices in a Variable.
`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`.
`indices` must be integer tensor, containing indices into `ref`.
It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`.
The innermost dimension of `indices` (with length `K`) corresponds to
indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th
dimension of `ref`.
`updates` is `Tensor` of rank `Q-1+P-K` with shape:
```
[d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]].
```
For example, say we want to add 4 scattered elements to a rank-1 tensor to
8 elements. In Python, that update would look like this:
```python
ref = tf.Variable([1, 2, 3, 4, 5, 6, 7, 8])
indices = tf.constant([[4], [3], [1] ,[7]])
updates = tf.constant([9, 10, 11, 12])
op = ref.scatter_nd_sub(indices, updates)
with tf.compat.v1.Session() as sess:
print sess.run(op)
```
The resulting update to ref would look like this:
[1, -9, 3, -6, -6, 6, 7, -4]
See `tf.scatter_nd` for more details about how to make updates to
slices.
Args:
indices: The indices to be used in the operation.
updates: The values to be used in the operation.
name: the name of the operation.
Returns:
The updated variable."
9135,scatter_nd_add,tensorflow/tensorflow/python/ops/resource_variable_ops.py,1143,method,"Applies sparse addition to individual values or slices in a Variable.
`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`.
`indices` must be integer tensor, containing indices into `ref`.
It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`.
The innermost dimension of `indices` (with length `K`) corresponds to
indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th
dimension of `ref`.
`updates` is `Tensor` of rank `Q-1+P-K` with shape:
```
[d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]].
```
For example, say we want to add 4 scattered elements to a rank-1 tensor to
8 elements. In Python, that update would look like this:
```python
ref = tf.Variable([1, 2, 3, 4, 5, 6, 7, 8])
indices = tf.constant([[4], [3], [1] ,[7]])
updates = tf.constant([9, 10, 11, 12])
add = ref.scatter_nd_add(indices, updates)
with tf.compat.v1.Session() as sess:
print sess.run(add)
```
The resulting update to ref would look like this:
[1, 13, 3, 14, 14, 6, 7, 20]
See `tf.scatter_nd` for more details about how to make updates to
slices.
Args:
indices: The indices to be used in the operation.
updates: The values to be used in the operation.
name: the name of the operation.
Returns:
The updated variable."
9136,scatter_nd_update,tensorflow/tensorflow/python/ops/resource_variable_ops.py,1192,method,"Applies sparse assignment to individual values or slices in a Variable.
`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`.
`indices` must be integer tensor, containing indices into `ref`.
It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`.
The innermost dimension of `indices` (with length `K`) corresponds to
indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th
dimension of `ref`.
`updates` is `Tensor` of rank `Q-1+P-K` with shape:
```
[d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]].
```
For example, say we want to add 4 scattered elements to a rank-1 tensor to
8 elements. In Python, that update would look like this:
```python
ref = tf.Variable([1, 2, 3, 4, 5, 6, 7, 8])
indices = tf.constant([[4], [3], [1] ,[7]])
updates = tf.constant([9, 10, 11, 12])
op = ref.scatter_nd_update(indices, updates)
with tf.compat.v1.Session() as sess:
print sess.run(op)
```
The resulting update to ref would look like this:
[1, 11, 3, 10, 9, 6, 7, 12]
See `tf.scatter_nd` for more details about how to make updates to
slices.
Args:
indices: The indices to be used in the operation.
updates: The values to be used in the operation.
name: the name of the operation.
Returns:
The updated variable."
9137,scatter_nd_max,tensorflow/tensorflow/python/ops/resource_variable_ops.py,1241,method,"Updates this variable with the max of `tf.IndexedSlices` and itself.
`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`.
`indices` must be integer tensor, containing indices into `ref`.
It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`.
The innermost dimension of `indices` (with length `K`) corresponds to
indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th
dimension of `ref`.
`updates` is `Tensor` of rank `Q-1+P-K` with shape:
```
[d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]].
```
See `tf.scatter_nd` for more details about how to make updates to
slices.
Args:
indices: The indices to be used in the operation.
updates: The values to be used in the operation.
name: the name of the operation.
Returns:
The updated variable."
9138,scatter_nd_min,tensorflow/tensorflow/python/ops/resource_variable_ops.py,1277,method,"Updates this variable with the min of `tf.IndexedSlices` and itself.
`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`.
`indices` must be integer tensor, containing indices into `ref`.
It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`.
The innermost dimension of `indices` (with length `K`) corresponds to
indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th
dimension of `ref`.
`updates` is `Tensor` of rank `Q-1+P-K` with shape:
```
[d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]].
```
See `tf.scatter_nd` for more details about how to make updates to
slices.
Args:
indices: The indices to be used in the operation.
updates: The values to be used in the operation.
name: the name of the operation.
Returns:
The updated variable."
9139,read_and_set_handle,tensorflow/tensorflow/python/ops/resource_variable_ops.py,672,method,
9140,ResourceVariable,tensorflow/tensorflow/python/ops/resource_variable_ops.py,1392,class,"Variable based on resource handles.
See the [Variables How To](https://tensorflow.org/guide/variables)
for a high level overview.
A `ResourceVariable` allows you to maintain state across subsequent calls to
session.run.
The `ResourceVariable` constructor requires an initial value for the variable,
which can be a `Tensor` of any type and shape. The initial value defines the
type and shape of the variable. After construction, the type and shape of
the variable are fixed. The value can be changed using one of the assign
methods.
Just like any `Tensor`, variables created with
`tf.Variable(use_resource=True)` can be used as inputs for other Ops in the
graph. Additionally, all the operators overloaded for the `Tensor` class are
carried over to variables, so you can also add nodes to the graph by just
doing arithmetic on variables.
Unlike ref-based variable, a ResourceVariable has well-defined semantics. Each
usage of a ResourceVariable in a TensorFlow graph adds a read_value operation
to the graph. The Tensors returned by a read_value operation are guaranteed to
see all modifications to the value of the variable which happen in any
operation on which the read_value depends on (either directly, indirectly, or
via a control dependency) and guaranteed to not see any modification to the
value of the variable from operations that depend on the read_value operation.
Updates from operations that have no dependency relationship to the read_value
operation might or might not be visible to read_value.
For example, if there is more than one assignment to a ResourceVariable in
a single session.run call there is a well-defined value for each operation
which uses the variable's value if the assignments and the read are connected
by edges in the graph. Consider the following example, in which two writes
can cause tf.Variable and tf.ResourceVariable to behave differently:
```python
a = tf.Variable(1.0, use_resource=True)
a.initializer.run()
assign = a.assign(2.0)
with tf.control_dependencies([assign]):
b = a.read_value()
with tf.control_dependencies([b]):
other_assign = a.assign(3.0)
with tf.control_dependencies([other_assign]):
# Will print 2.0 because the value was read before other_assign ran. If
# `a` was a tf.Variable instead, 2.0 or 3.0 could be printed.
tf.compat.v1.Print(b, [b]).eval()
```"
9141,UninitializedVariable,tensorflow/tensorflow/python/ops/resource_variable_ops.py,1844,class,A variable with no initializer.
9142,variable_shape,tensorflow/tensorflow/python/ops/resource_variable_ops.py,2080,function,
9143,is_resource_variable,tensorflow/tensorflow/python/ops/resource_variable_ops.py,2153,function,"""Returns True if `var` is to be considered a ResourceVariable."
9144,copy_to_graph_uninitialized,tensorflow/tensorflow/python/ops/resource_variable_ops.py,2159,function,"Copies an existing variable to a new graph, with no initializer."
9145,VariableSpec,tensorflow/tensorflow/python/ops/resource_variable_ops.py,2182,class,Describes a tf.Variable.
9146,register_resource,tensorflow/tensorflow/python/ops/resources.py,38,function,"Registers a resource into the appropriate collections.
This makes the resource findable in either the shared or local resources
collection.
Args:
handle: op which returns a handle for the resource.
create_op: op which initializes the resource.
is_initialized_op: op which returns a scalar boolean tensor of whether
the resource has been initialized.
is_shared: if True, the resource gets added to the shared resource
collection; otherwise it gets added to the local resource collection."
9147,shared_resources,tensorflow/tensorflow/python/ops/resources.py,60,function,Returns resources visible to all tasks in the cluster.
9148,local_resources,tensorflow/tensorflow/python/ops/resources.py,65,function,Returns resources intended to be local to this session.
9149,report_uninitialized_resources,tensorflow/tensorflow/python/ops/resources.py,70,function,"Returns the names of all uninitialized resources in resource_list.
If the returned tensor is empty then all resources have been initialized.
Args:
resource_list: resources to check. If None, will use shared_resources() +
local_resources().
name: name for the resource-checking op.
Returns:
Tensor containing names of the handles of all resources which have not
yet been initialized."
9150,initialize_resources,tensorflow/tensorflow/python/ops/resources.py,108,function,"Initializes the resources in the given list.
Args:
resource_list: list of resources to initialize.
name: name of the initialization op.
Returns:
op responsible for initializing all resources."
9151,bidirectional_dynamic_rnn,tensorflow/tensorflow/python/ops/rnn.py,347,function,"Creates a dynamic version of bidirectional recurrent neural network.
Takes input and builds independent forward and backward RNNs. The input_size
of forward and backward cell must match. The initial state for both directions
is zero by default (but can be set optionally) and no intermediate states are
ever returned -- the network is fully unrolled for the given (passed in)
length(s) of the sequence(s) or completely unrolled if length(s) is not
given.
Args:
cell_fw: An instance of RNNCell, to be used for forward direction.
cell_bw: An instance of RNNCell, to be used for backward direction.
inputs: The RNN inputs.
If time_major == False (default), this must be a tensor of shape:
`[batch_size, max_time, ...]`, or a nested tuple of such elements.
If time_major == True, this must be a tensor of shape: `[max_time,
batch_size, ...]`, or a nested tuple of such elements.
sequence_length: (optional) An int32/int64 vector, size `[batch_size]`,
containing the actual lengths for each of the sequences in the batch. If
not provided, all batch entries are assumed to be full sequences; and time
reversal is applied from time `0` to `max_time` for each sequence.
initial_state_fw: (optional) An initial state for the forward RNN. This must
be a tensor of appropriate type and shape `[batch_size,
cell_fw.state_size]`. If `cell_fw.state_size` is a tuple, this should be a
tuple of tensors having shapes `[batch_size, s] for s in
cell_fw.state_size`.
initial_state_bw: (optional) Same as for `initial_state_fw`, but using the
corresponding properties of `cell_bw`.
dtype: (optional) The data type for the initial states and expected output.
Required if initial_states are not provided or RNN states have a
heterogeneous dtype.
parallel_iterations: (Default: 32). The number of iterations to run in
parallel. Those operations which do not have any temporal dependency and
can be run in parallel, will be. This parameter trades off time for
space. Values >> 1 use more memory but take less time, while smaller
values use less memory but computations take longer.
swap_memory: Transparently swap the tensors produced in forward inference
but needed for back prop from GPU to CPU. This allows training RNNs which
would typically not fit on a single GPU, with very minimal (or no)
performance penalty.
time_major: The shape format of the `inputs` and `outputs` Tensors. If true,
these `Tensors` must be shaped `[max_time, batch_size, depth]`. If false,
these `Tensors` must be shaped `[batch_size, max_time, depth]`. Using
`time_major = True` is a bit more efficient because it avoids transposes
at the beginning and end of the RNN calculation. However, most TensorFlow
data is batch-major, so by default this function accepts input and emits
output in batch-major form.
scope: VariableScope for the created subgraph; defaults to
""bidirectional_rnn""
Returns:
A tuple (outputs, output_states) where:
outputs: A tuple (output_fw, output_bw) containing the forward and
the backward rnn output `Tensor`.
If time_major == False (default),
output_fw will be a `Tensor` shaped:
`[batch_size, max_time, cell_fw.output_size]`
and output_bw will be a `Tensor` shaped:
`[batch_size, max_time, cell_bw.output_size]`.
If time_major == True,
output_fw will be a `Tensor` shaped:
`[max_time, batch_size, cell_fw.output_size]`
and output_bw will be a `Tensor` shaped:
`[max_time, batch_size, cell_bw.output_size]`.
It returns a tuple instead of a single concatenated `Tensor`, unlike
in the `bidirectional_rnn`. If the concatenated one is preferred,
the forward and backward outputs can be concatenated as
`tf.concat(outputs, 2)`.
output_states: A tuple (output_state_fw, output_state_bw) containing
the forward and the backward final states of bidirectional rnn.
Raises:
TypeError: If `cell_fw` or `cell_bw` is not an instance of `RNNCell`."
9152,dynamic_rnn,tensorflow/tensorflow/python/ops/rnn.py,505,function,"Creates a recurrent neural network specified by RNNCell `cell`.
Performs fully dynamic unrolling of `inputs`.
Example:
```python
# create a BasicRNNCell
rnn_cell = tf.compat.v1.nn.rnn_cell.BasicRNNCell(hidden_size)
# 'outputs' is a tensor of shape [batch_size, max_time, cell_state_size]
# defining initial state
initial_state = rnn_cell.zero_state(batch_size, dtype=tf.float32)
# 'state' is a tensor of shape [batch_size, cell_state_size]
outputs, state = tf.compat.v1.nn.dynamic_rnn(rnn_cell, input_data,
initial_state=initial_state,
dtype=tf.float32)
```
```python
# create 2 LSTMCells
rnn_layers = [tf.compat.v1.nn.rnn_cell.LSTMCell(size) for size in [128, 256]]
# create a RNN cell composed sequentially of a number of RNNCells
multi_rnn_cell = tf.compat.v1.nn.rnn_cell.MultiRNNCell(rnn_layers)
# 'outputs' is a tensor of shape [batch_size, max_time, 256]
# 'state' is a N-tuple where N is the number of LSTMCells containing a
# tf.nn.rnn_cell.LSTMStateTuple for each cell
outputs, state = tf.compat.v1.nn.dynamic_rnn(cell=multi_rnn_cell,
inputs=data,
dtype=tf.float32)
```
Args:
cell: An instance of RNNCell.
inputs: The RNN inputs.
If `time_major == False` (default), this must be a `Tensor` of shape:
`[batch_size, max_time, ...]`, or a nested tuple of such elements.
If `time_major == True`, this must be a `Tensor` of shape: `[max_time,
batch_size, ...]`, or a nested tuple of such elements. This may also be
a (possibly nested) tuple of Tensors satisfying this property. The
first two dimensions must match across all the inputs, but otherwise the
ranks and other shape components may differ. In this case, input to
`cell` at each time-step will replicate the structure of these tuples,
except for the time dimension (from which the time is taken). The input
to `cell` at each time step will be a `Tensor` or (possibly nested)
tuple of Tensors each with dimensions `[batch_size, ...]`.
sequence_length: (optional) An int32/int64 vector sized `[batch_size]`. Used
to copy-through state and zero-out outputs when past a batch element's
sequence length. This parameter enables users to extract the last valid
state and properly padded outputs, so it is provided for correctness.
initial_state: (optional) An initial state for the RNN. If `cell.state_size`
is an integer, this must be a `Tensor` of appropriate type and shape
`[batch_size, cell.state_size]`. If `cell.state_size` is a tuple, this
should be a tuple of tensors having shapes `[batch_size, s] for s in
cell.state_size`.
dtype: (optional) The data type for the initial state and expected output.
Required if initial_state is not provided or RNN state has a heterogeneous
dtype.
parallel_iterations: (Default: 32). The number of iterations to run in
parallel. Those operations which do not have any temporal dependency and
can be run in parallel, will be. This parameter trades off time for
space. Values >> 1 use more memory but take less time, while smaller
values use less memory but computations take longer.
swap_memory: Transparently swap the tensors produced in forward inference
but needed for back prop from GPU to CPU. This allows training RNNs which
would typically not fit on a single GPU, with very minimal (or no)
performance penalty.
time_major: The shape format of the `inputs` and `outputs` Tensors. If true,
these `Tensors` must be shaped `[max_time, batch_size, depth]`. If false,
these `Tensors` must be shaped `[batch_size, max_time, depth]`. Using
`time_major = True` is a bit more efficient because it avoids transposes
at the beginning and end of the RNN calculation. However, most TensorFlow
data is batch-major, so by default this function accepts input and emits
output in batch-major form.
scope: VariableScope for the created subgraph; defaults to ""rnn"".
Returns:
A pair (outputs, state) where:
outputs: The RNN output `Tensor`.
If time_major == False (default), this will be a `Tensor` shaped:
`[batch_size, max_time, cell.output_size]`.
If time_major == True, this will be a `Tensor` shaped:
`[max_time, batch_size, cell.output_size]`.
Note, if `cell.output_size` is a (possibly nested) tuple of integers
or `TensorShape` objects, then `outputs` will be a tuple having the
same structure as `cell.output_size`, containing Tensors having shapes
corresponding to the shape data in `cell.output_size`.
state: The final state. If `cell.state_size` is an int, this
will be shaped `[batch_size, cell.state_size]`. If it is a
`TensorShape`, this will be shaped `[batch_size] + cell.state_size`.
If it is a (possibly nested) tuple of ints or `TensorShape`, this will
be a tuple having the corresponding shapes. If cells are `LSTMCells`
`state` will be a tuple containing a `LSTMStateTuple` for each cell.
Raises:
TypeError: If `cell` is not an instance of RNNCell.
ValueError: If inputs is None or an empty list."
9153,raw_rnn,tensorflow/tensorflow/python/ops/rnn.py,919,function,"Creates an `RNN` specified by RNNCell `cell` and loop function `loop_fn`.
**NOTE: This method is still in testing, and the API may change.**
This function is a more primitive version of `dynamic_rnn` that provides
more direct access to the inputs each iteration. It also provides more
control over when to start and finish reading the sequence, and
what to emit for the output.
For example, it can be used to implement the dynamic decoder of a seq2seq
model.
Instead of working with `Tensor` objects, most operations work with
`TensorArray` objects directly.
The operation of `raw_rnn`, in pseudo-code, is basically the following:
```python
time = tf.constant(0, dtype=tf.int32)
(finished, next_input, initial_state, emit_structure, loop_state) = loop_fn(
time=time, cell_output=None, cell_state=None, loop_state=None)
emit_ta = TensorArray(dynamic_size=True, dtype=initial_state.dtype)
state = initial_state
while not all(finished):
(output, cell_state) = cell(next_input, state)
(next_finished, next_input, next_state, emit, loop_state) = loop_fn(
time=time + 1, cell_output=output, cell_state=cell_state,
loop_state=loop_state)
# Emit zeros and copy forward state for minibatch entries that are finished.
state = tf.where(finished, state, next_state)
emit = tf.where(finished, tf.zeros_like(emit_structure), emit)
emit_ta = emit_ta.write(time, emit)
# If any new minibatch entries are marked as finished, mark these.
finished = tf.logical_or(finished, next_finished)
time += 1
return (emit_ta, state, loop_state)
```
with the additional properties that output and state may be (possibly nested)
tuples, as determined by `cell.output_size` and `cell.state_size`, and
as a result the final `state` and `emit_ta` may themselves be tuples.
A simple implementation of `dynamic_rnn` via `raw_rnn` looks like this:
```python
inputs = tf.compat.v1.placeholder(shape=(max_time, batch_size, input_depth),
dtype=tf.float32)
sequence_length = tf.compat.v1.placeholder(shape=(batch_size,),
dtype=tf.int32)
inputs_ta = tf.TensorArray(dtype=tf.float32, size=max_time)
inputs_ta = inputs_ta.unstack(inputs)
cell = tf.compat.v1.nn.rnn_cell.LSTMCell(num_units)
def loop_fn(time, cell_output, cell_state, loop_state):
emit_output = cell_output # == None for time == 0
if cell_output is None: # time == 0
next_cell_state = cell.zero_state(batch_size, tf.float32)
else:
next_cell_state = cell_state
elements_finished = (time >= sequence_length)
finished = tf.reduce_all(elements_finished)
next_input = tf.cond(
finished,
lambda: tf.zeros([batch_size, input_depth], dtype=tf.float32),
lambda: inputs_ta.read(time))
next_loop_state = None
return (elements_finished, next_input, next_cell_state,
emit_output, next_loop_state)
outputs_ta, final_state, _ = raw_rnn(cell, loop_fn)
outputs = outputs_ta.stack()
```
Args:
cell: An instance of RNNCell.
loop_fn: A callable that takes inputs `(time, cell_output, cell_state,
loop_state)` and returns the tuple `(finished, next_input,
next_cell_state, emit_output, next_loop_state)`. Here `time` is an int32
scalar `Tensor`, `cell_output` is a `Tensor` or (possibly nested) tuple of
tensors as determined by `cell.output_size`, and `cell_state` is a
`Tensor` or (possibly nested) tuple of tensors, as determined by the
`loop_fn` on its first call (and should match `cell.state_size`).
The outputs are: `finished`, a boolean `Tensor` of
shape `[batch_size]`, `next_input`: the next input to feed to `cell`,
`next_cell_state`: the next state to feed to `cell`,
and `emit_output`: the output to store for this iteration. Note that
`emit_output` should be a `Tensor` or (possibly nested) tuple of tensors
which is aggregated in the `emit_ta` inside the `while_loop`. For the
first call to `loop_fn`, the `emit_output` corresponds to the
`emit_structure` which is then used to determine the size of the
`zero_tensor` for the `emit_ta` (defaults to `cell.output_size`). For
the subsequent calls to the `loop_fn`, the `emit_output` corresponds to
the actual output tensor that is to be aggregated in the `emit_ta`. The
parameter `cell_state` and output `next_cell_state` may be either a
single or (possibly nested) tuple of tensors. The parameter
`loop_state` and output `next_loop_state` may be either a single or
(possibly nested) tuple of `Tensor` and `TensorArray` objects. This
last parameter may be ignored by `loop_fn` and the return value may be
`None`. If it is not `None`, then the `loop_state` will be propagated
through the RNN loop, for use purely by `loop_fn` to keep track of its
own state. The `next_loop_state` parameter returned may be `None`. The
first call to `loop_fn` will be `time = 0`, `cell_output = None`,
`cell_state = None`, and `loop_state = None`. For this call: The
`next_cell_state` value should be the value with which to initialize the
cell's state. It may be a final state from a previous RNN or it may be
the output of `cell.zero_state()`. It should be a (possibly nested)
tuple structure of tensors. If `cell.state_size` is an integer, this
must be a `Tensor` of appropriate type and shape `[batch_size,
cell.state_size]`. If `cell.state_size` is a `TensorShape`, this must be
a `Tensor` of appropriate type and shape `[batch_size] +
cell.state_size`. If `cell.state_size` is a (possibly nested) tuple of
ints or `TensorShape`, this will be a tuple having the corresponding
shapes. The `emit_output` value may be either `None` or a (possibly
nested) tuple structure of tensors, e.g., `(tf.zeros(shape_0,
dtype=dtype_0), tf.zeros(shape_1, dtype=dtype_1))`. If this first
`emit_output` return value is `None`, then the `emit_ta` result of
`raw_rnn` will have the same structure and dtypes as `cell.output_size`.
Otherwise `emit_ta` will have the same structure, shapes (prepended with
a `batch_size` dimension), and dtypes as `emit_output`. The actual
values returned for `emit_output` at this initializing call are ignored.
Note, this emit structure must be consistent across all time steps.
parallel_iterations: (Default: 32). The number of iterations to run in
parallel. Those operations which do not have any temporal dependency and
can be run in parallel, will be. This parameter trades off time for
space. Values >> 1 use more memory but take less time, while smaller
values use less memory but computations take longer.
swap_memory: Transparently swap the tensors produced in forward inference
but needed for back prop from GPU to CPU. This allows training RNNs which
would typically not fit on a single GPU, with very minimal (or no)
performance penalty.
scope: VariableScope for the created subgraph; defaults to ""rnn"".
Returns:
A tuple `(emit_ta, final_state, final_loop_state)` where:
`emit_ta`: The RNN output `TensorArray`.
If `loop_fn` returns a (possibly nested) set of Tensors for
`emit_output` during initialization, (inputs `time = 0`,
`cell_output = None`, and `loop_state = None`), then `emit_ta` will
have the same structure, dtypes, and shapes as `emit_output` instead.
If `loop_fn` returns `emit_output = None` during this call,
the structure of `cell.output_size` is used:
If `cell.output_size` is a (possibly nested) tuple of integers
or `TensorShape` objects, then `emit_ta` will be a tuple having the
same structure as `cell.output_size`, containing TensorArrays whose
elements' shapes correspond to the shape data in `cell.output_size`.
`final_state`: The final cell state. If `cell.state_size` is an int, this
will be shaped `[batch_size, cell.state_size]`. If it is a
`TensorShape`, this will be shaped `[batch_size] + cell.state_size`.
If it is a (possibly nested) tuple of ints or `TensorShape`, this will
be a tuple having the corresponding shapes.
`final_loop_state`: The final loop state as returned by `loop_fn`.
Raises:
TypeError: If `cell` is not an instance of RNNCell, or `loop_fn` is not
a `callable`."
9154,static_rnn,tensorflow/tensorflow/python/ops/rnn.py,1246,function,"Creates a recurrent neural network specified by RNNCell `cell`.
The simplest form of RNN network generated is:
```python
state = cell.zero_state(...)
outputs = []
for input_ in inputs:
output, state = cell(input_, state)
outputs.append(output)
return (outputs, state)
```
However, a few other options are available:
An initial state can be provided.
If the sequence_length vector is provided, dynamic calculation is performed.
This method of calculation does not compute the RNN steps past the maximum
sequence length of the minibatch (thus saving computational time),
and properly propagates the state at an example's sequence length
to the final state output.
The dynamic calculation performed is, at time `t` for batch row `b`,
```python
(output, state)(b, t) =
(t >= sequence_length(b))
? (zeros(cell.output_size), states(b, sequence_length(b) - 1))
: cell(input(b, t), state(b, t - 1))
```
Args:
cell: An instance of RNNCell.
inputs: A length T list of inputs, each a `Tensor` of shape `[batch_size,
input_size]`, or a nested tuple of such elements.
initial_state: (optional) An initial state for the RNN. If `cell.state_size`
is an integer, this must be a `Tensor` of appropriate type and shape
`[batch_size, cell.state_size]`. If `cell.state_size` is a tuple, this
should be a tuple of tensors having shapes `[batch_size, s] for s in
cell.state_size`.
dtype: (optional) The data type for the initial state and expected output.
Required if initial_state is not provided or RNN state has a heterogeneous
dtype.
sequence_length: Specifies the length of each sequence in inputs. An int32
or int64 vector (tensor) size `[batch_size]`, values in `[0, T)`.
scope: VariableScope for the created subgraph; defaults to ""rnn"".
Returns:
A pair (outputs, state) where:
- outputs is a length T list of outputs (one for each input), or a nested
tuple of such elements.
- state is the final state
Raises:
TypeError: If `cell` is not an instance of RNNCell.
ValueError: If `inputs` is `None` or an empty list, or if the input depth
(column size) cannot be inferred from inputs via shape inference."
9155,static_state_saving_rnn,tensorflow/tensorflow/python/ops/rnn.py,1425,function,"RNN that accepts a state saver for time-truncated RNN calculation.
Args:
cell: An instance of `RNNCell`.
inputs: A length T list of inputs, each a `Tensor` of shape `[batch_size,
input_size]`.
state_saver: A state saver object with methods `state` and `save_state`.
state_name: Python string or tuple of strings. The name to use with the
state_saver. If the cell returns tuples of states (i.e., `cell.state_size`
is a tuple) then `state_name` should be a tuple of strings having the same
length as `cell.state_size`. Otherwise it should be a single string.
sequence_length: (optional) An int32/int64 vector size [batch_size]. See the
documentation for rnn() for more details about sequence_length.
scope: VariableScope for the created subgraph; defaults to ""rnn"".
Returns:
A pair (outputs, state) where:
outputs is a length T list of outputs (one for each input)
states is the final state
Raises:
TypeError: If `cell` is not an instance of RNNCell.
ValueError: If `inputs` is `None` or an empty list, or if the arity and
type of `state_name` does not match that of `cell.state_size`."
9156,static_bidirectional_rnn,tensorflow/tensorflow/python/ops/rnn.py,1520,function,"Creates a bidirectional recurrent neural network.
Similar to the unidirectional case above (rnn) but takes input and builds
independent forward and backward RNNs with the final forward and backward
outputs depth-concatenated, such that the output will have the format
[time][batch][cell_fw.output_size + cell_bw.output_size]. The input_size of
forward and backward cell must match. The initial state for both directions
is zero by default (but can be set optionally) and no intermediate states are
ever returned -- the network is fully unrolled for the given (passed in)
length(s) of the sequence(s) or completely unrolled if length(s) is not given.
Args:
cell_fw: An instance of RNNCell, to be used for forward direction.
cell_bw: An instance of RNNCell, to be used for backward direction.
inputs: A length T list of inputs, each a tensor of shape [batch_size,
input_size], or a nested tuple of such elements.
initial_state_fw: (optional) An initial state for the forward RNN. This must
be a tensor of appropriate type and shape `[batch_size,
cell_fw.state_size]`. If `cell_fw.state_size` is a tuple, this should be a
tuple of tensors having shapes `[batch_size, s] for s in
cell_fw.state_size`.
initial_state_bw: (optional) Same as for `initial_state_fw`, but using the
corresponding properties of `cell_bw`.
dtype: (optional) The data type for the initial state. Required if either
of the initial states are not provided.
sequence_length: (optional) An int32/int64 vector, size `[batch_size]`,
containing the actual lengths for each of the sequences.
scope: VariableScope for the created subgraph; defaults to
""bidirectional_rnn""
Returns:
A tuple (outputs, output_state_fw, output_state_bw) where:
outputs is a length `T` list of outputs (one for each input), which
are depth-concatenated forward and backward outputs.
output_state_fw is the final state of the forward rnn.
output_state_bw is the final state of the backward rnn.
Raises:
TypeError: If `cell_fw` or `cell_bw` is not an instance of `RNNCell`.
ValueError: If inputs is None or an empty list."
9157,deterministic_random_uniform,tensorflow/tensorflow/python/ops/rnn_grad_test.py,139,function,
9158,icfo_to_ifco,tensorflow/tensorflow/python/ops/rnn_grad_test.py,143,function,Convert gates' weights and biases from ICFO to IFCO layout.
9159,EagerFunc,tensorflow/tensorflow/python/ops/script_ops.py,71,class,A wrapper for a function owned by an EagerPyFunc.
9160,FuncRegistry,tensorflow/tensorflow/python/ops/script_ops.py,154,class,"A helper class to keep track of registered py functions.
FuncRegistry keeps a map from unique tokens (string) to python
functions, which takes numpy arrays and outputs numpy arrays."
9161,insert,tensorflow/tensorflow/python/ops/script_ops.py,175,method,Registers `func` and returns a unique token for this entry.
9162,remove,tensorflow/tensorflow/python/ops/script_ops.py,182,method,Removes the registered function corresponding to `token`.
9163,size,tensorflow/tensorflow/python/ops/script_ops.py,256,method,Returns how many functions are currently registered.
9164,eager_py_func,tensorflow/tensorflow/python/ops/script_ops.py,375,function,"Wraps a python function into a TensorFlow op that executes it eagerly.
This function allows expressing computations in a TensorFlow graph as
Python functions. In particular, it wraps a Python function `func`
in a once-differentiable TensorFlow operation that executes it with eager
execution enabled. As a consequence, `tf.py_function` makes it
possible to express control flow using Python constructs (`if`, `while`,
`for`, etc.), instead of TensorFlow control flow constructs (`tf.cond`,
`tf.while_loop`). For example, you might use `tf.py_function` to
implement the log huber function:
```python
def log_huber(x, m):
if tf.abs(x) <= m:
return x**2
else:
return m**2 * (1 - 2 * tf.math.log(m) + tf.math.log(x**2))
x = tf.compat.v1.placeholder(tf.float32)
m = tf.compat.v1.placeholder(tf.float32)
y = tf.py_function(func=log_huber, inp=[x, m], Tout=tf.float32)
dy_dx = tf.gradients(y, x)[0]
with tf.compat.v1.Session() as sess:
# The session executes `log_huber` eagerly. Given the feed values below,
# it will take the first branch, so `y` evaluates to 1.0 and
# `dy_dx` evaluates to 2.0.
y, dy_dx = sess.run([y, dy_dx], feed_dict={x: 1.0, m: 2.0})
```
You can also use `tf.py_function` to debug your models at runtime
using Python tools, i.e., you can isolate portions of your code that
you want to debug, wrap them in Python functions and insert `pdb` tracepoints
or print statements as desired, and wrap those functions in
`tf.py_function`.
For more information on eager execution, see the
[Eager guide](https://tensorflow.org/guide/eager).
`tf.py_function` is similar in spirit to `tf.compat.v1.py_func`, but unlike
the latter, the former lets you use TensorFlow operations in the wrapped
Python function. In particular, while `tf.compat.v1.py_func` only runs on CPUs
and
wraps functions that take NumPy arrays as inputs and return NumPy arrays as
outputs, `tf.py_function` can be placed on GPUs and wraps functions
that take Tensors as inputs, execute TensorFlow operations in their bodies,
and return Tensors as outputs.
Like `tf.compat.v1.py_func`, `tf.py_function` has the following limitations
with respect to serialization and distribution:
* The body of the function (i.e. `func`) will not be serialized in a
`GraphDef`. Therefore, you should not use this function if you need to
serialize your model and restore it in a different environment.
* The operation must run in the same address space as the Python program
that calls `tf.py_function()`. If you are using distributed
TensorFlow, you must run a `tf.distribute.Server` in the same process as the
program that calls `tf.py_function()` and you must pin the created
operation to a device in that server (e.g. using `with tf.device():`).
Args:
func: A Python function which accepts a list of `Tensor` objects having
element types that match the corresponding `tf.Tensor` objects in `inp`
and returns a list of `Tensor` objects (or a single `Tensor`, or `None`)
having element types that match the corresponding values in `Tout`.
inp: A list of `Tensor` objects.
Tout: A list or tuple of tensorflow data types or a single tensorflow data
type if there is only one, indicating what `func` returns; an empty list
if no value is returned (i.e., if the return value is `None`).
name: A name for the operation (optional).
Returns:
A list of `Tensor` or a single `Tensor` which `func` computes; an empty list
if `func` returns None."
9165,py_func_common,tensorflow/tensorflow/python/ops/script_ops.py,462,function,"Wraps a python function and uses it as a TensorFlow op.
Given a python function `func`, which takes numpy arrays as its
arguments and returns numpy arrays as its outputs, wrap this function as an
operation in a TensorFlow graph. The following snippet constructs a simple
TensorFlow graph that invokes the `np.sinh()` NumPy function as a operation
in the graph:
```python
def my_func(x):
# x will be a numpy array with the contents of the placeholder below
return np.sinh(x)
input = tf.compat.v1.placeholder(tf.float32)
y = tf.compat.v1.py_func(my_func, [input], tf.float32)
```
**N.B.** The `tf.compat.v1.py_func()` operation has the following known
limitations:
* The body of the function (i.e. `func`) will not be serialized in a
`GraphDef`. Therefore, you should not use this function if you need to
serialize your model and restore it in a different environment.
* The operation must run in the same address space as the Python program
that calls `tf.compat.v1.py_func()`. If you are using distributed
TensorFlow, you
must run a `tf.distribute.Server` in the same process as the program that
calls
`tf.compat.v1.py_func()` and you must pin the created operation to a device
in that
server (e.g. using `with tf.device():`).
Args:
func: A Python function, which accepts `ndarray` objects as arguments and
returns a list of `ndarray` objects (or a single `ndarray`). This function
must accept as many arguments as there are tensors in `inp`, and these
argument types will match the corresponding `tf.Tensor` objects in `inp`.
The returns `ndarray`s must match the number and types defined `Tout`.
Important Note: Input and output numpy `ndarray`s of `func` are not
guaranteed to be copies. In some cases their underlying memory will be
shared with the corresponding TensorFlow tensors. In-place modification
or storing `func` input or return values in python datastructures
without explicit (np.)copy can have non-deterministic consequences.
inp: A list of `Tensor` objects.
Tout: A list or tuple of tensorflow data types or a single tensorflow data
type if there is only one, indicating what `func` returns.
stateful: (Boolean.) If True, the function should be considered stateful. If
a function is stateless, when given the same input it will return the same
output and have no observable side effects. Optimizations such as common
subexpression elimination are only performed on stateless operations.
name: A name for the operation (optional).
Returns:
A list of `Tensor` or a single `Tensor` which `func` computes."
9166,py_func,tensorflow/tensorflow/python/ops/script_ops.py,557,function,
9167,numpy_function,tensorflow/tensorflow/python/ops/script_ops.py,566,function,"Wraps a python function and uses it as a TensorFlow op.
Given a python function `func` wrap this function as an operation in a
TensorFlow function. `func` must take numpy arrays as its arguments and
return numpy arrays as its outputs.
The following example creates a TensorFlow graph with `np.sinh()` as an
operation in the graph:
>>> def my_numpy_func(x):
... # x will be a numpy array with the contents of the input to the
... # tf.function
... return np.sinh(x)
>>> @tf.function(input_signature=[tf.TensorSpec(None, tf.float32)])
... def tf_function(input):
... y = tf.numpy_function(my_numpy_func, [input], tf.float32)
... return y * y
>>> tf_function(tf.constant(1.))
<tf.Tensor: shape=(), dtype=float32, numpy=1.3810978>
Comparison to `tf.py_function`:
`tf.py_function` and `tf.numpy_function` are very similar, except that
`tf.numpy_function` takes numpy arrays, and not `tf.Tensor`s. If you want the
function to contain `tf.Tensors`, and have any TensorFlow operations executed
in the function be differentiable, please use `tf.py_function`.
Note: The `tf.numpy_function` operation has the following known
limitations:
* The body of the function (i.e. `func`) will not be serialized in a
`tf.SavedModel`. Therefore, you should not use this function if you need to
serialize your model and restore it in a different environment.
* The operation must run in the same address space as the Python program
that calls `tf.numpy_function()`. If you are using distributed
TensorFlow, you must run a `tf.distribute.Server` in the same process as the
program that calls `tf.numpy_function` you must pin the created
operation to a device in that server (e.g. using `with tf.device():`).
* Since the function takes numpy arrays, you cannot take gradients
through a numpy_function. If you require something that is differentiable,
please consider using tf.py_function.
* The resulting function is assumed stateful and will never be optimized.
Args:
func: A Python function, which accepts `numpy.ndarray` objects as arguments
and returns a list of `numpy.ndarray` objects (or a single
`numpy.ndarray`). This function must accept as many arguments as there are
tensors in `inp`, and these argument types will match the corresponding
`tf.Tensor` objects in `inp`. The returns `numpy.ndarray`s must match the
number and types defined `Tout`.
Important Note: Input and output `numpy.ndarray`s of `func` are not
guaranteed to be copies. In some cases their underlying memory will be
shared with the corresponding TensorFlow tensors. In-place modification
or storing `func` input or return values in python datastructures
without explicit (np.)copy can have non-deterministic consequences.
inp: A list of `tf.Tensor` objects.
Tout: A list or tuple of tensorflow data types or a single tensorflow data
type if there is only one, indicating what `func` returns.
name: (Optional) A name for the operation.
Returns:
Single or list of `tf.Tensor` which `func` computes."
9168,encode_resource_handle,tensorflow/tensorflow/python/ops/session_ops.py,36,function,Encode a ResourceHandle proto as custom numpy struct type.
9169,TensorHandle,tensorflow/tensorflow/python/ops/session_ops.py,42,class,Represents a handle for a live tensor in a session.
9170,to_numpy_array,tensorflow/tensorflow/python/ops/session_ops.py,78,method,"Convert a TensorHandle object to a feedable numpy value.
Returns:
A numpy array of a custom struct type that can be used as a feed value
to run()."
9171,handle,tensorflow/tensorflow/python/ops/session_ops.py,88,method,The string representation of this handle.
9172,eval,tensorflow/tensorflow/python/ops/session_ops.py,92,method,Return the value of the tensor represented by this handle.
9173,delete,tensorflow/tensorflow/python/ops/session_ops.py,101,method,Force the deletion of this persistent tensor.
9174,get_raw_handle,tensorflow/tensorflow/python/ops/session_ops.py,110,method,"Return the raw handle of the tensor.
Note that the method disables the automatic garbage collection of this
persistent tensor. The caller is now responsible for managing the life
time of the tensor."
9175,get_session_handle,tensorflow/tensorflow/python/ops/session_ops.py,139,function,"Return the handle of `data`.
This is EXPERIMENTAL and subject to change.
Keep `data` ""in-place"" in the runtime and create a handle that can be
used to retrieve `data` in a subsequent run().
Combined with `get_session_tensor`, we can keep a tensor produced in
one run call in place, and use it as the input in a future run call.
Args:
data: A tensor to be stored in the session.
name: Optional name prefix for the return tensor.
Returns:
A scalar string tensor representing a unique handle for `data`.
Raises:
TypeError: if `data` is not a Tensor.
Example:
```python
c = tf.multiply(a, b)
h = tf.compat.v1.get_session_handle(c)
h = sess.run(h)
p, a = tf.compat.v1.get_session_tensor(h.handle, tf.float32)
b = tf.multiply(a, 10)
c = sess.run(b, feed_dict={p: h.handle})
```"
9176,get_session_tensor,tensorflow/tensorflow/python/ops/session_ops.py,182,function,"Get the tensor of type `dtype` by feeding a tensor handle.
This is EXPERIMENTAL and subject to change.
Get the value of the tensor from a tensor handle. The tensor
is produced in a previous run() and stored in the state of the
session.
Args:
handle: The string representation of a persistent tensor handle.
dtype: The type of the output tensor.
name: Optional name prefix for the return tensor.
Returns:
A pair of tensors. The first is a placeholder for feeding a
tensor handle and the second is the tensor in the session state
keyed by the tensor handle.
Example:
```python
c = tf.multiply(a, b)
h = tf.compat.v1.get_session_handle(c)
h = sess.run(h)
p, a = tf.compat.v1.get_session_tensor(h.handle, tf.float32)
b = tf.multiply(a, 10)
c = sess.run(b, feed_dict={p: h.handle})
```"
9177,delete_session_tensor,tensorflow/tensorflow/python/ops/session_ops.py,223,function,"Delete the tensor for the given tensor handle.
This is EXPERIMENTAL and subject to change.
Delete the tensor of a given tensor handle. The tensor is produced
in a previous run() and stored in the state of the session.
Args:
handle: The string representation of a persistent tensor handle.
name: Optional name prefix for the return tensor.
Returns:
A pair of graph elements. The first is a placeholder for feeding a
tensor handle and the second is a deletion operation."
9178,set_size,tensorflow/tensorflow/python/ops/sets_impl.py,37,function,"Compute number of unique elements along last dimension of `a`.
Args:
a: `SparseTensor`, with indices sorted in row-major order.
validate_indices: Whether to validate the order and range of sparse indices
in `a`.
Returns:
`int32` `Tensor` of set sizes. For `a` ranked `n`, this is a `Tensor` with
rank `n-1`, and the same 1st `n-1` dimensions as `a`. Each value is the
number of unique elements in the corresponding `[0...n-1]` dimension of `a`.
Raises:
TypeError: If `a` is an invalid types."
9179,set_intersection,tensorflow/tensorflow/python/ops/sets_impl.py,141,function,"Compute set intersection of elements in last dimension of `a` and `b`.
All but the last dimension of `a` and `b` must match.
Example:
```python
import tensorflow as tf
import collections
# Represent the following array of sets as a sparse tensor:
# a = np.array([[{1, 2}, {3}], [{4}, {5, 6}]])
a = collections.OrderedDict([
((0, 0, 0), 1),
((0, 0, 1), 2),
((0, 1, 0), 3),
((1, 0, 0), 4),
((1, 1, 0), 5),
((1, 1, 1), 6),
])
a = tf.sparse.SparseTensor(list(a.keys()), list(a.values()),
dense_shape=[2,2,2])
# b = np.array([[{1}, {}], [{4}, {5, 6, 7, 8}]])
b = collections.OrderedDict([
((0, 0, 0), 1),
((1, 0, 0), 4),
((1, 1, 0), 5),
((1, 1, 1), 6),
((1, 1, 2), 7),
((1, 1, 3), 8),
])
b = tf.sparse.SparseTensor(list(b.keys()), list(b.values()),
dense_shape=[2, 2, 4])
# `tf.sets.intersection` is applied to each aligned pair of sets.
tf.sets.intersection(a, b)
# The result will be equivalent to either of:
#
# np.array([[{1}, {}], [{4}, {5, 6}]])
#
# collections.OrderedDict([
# ((0, 0, 0), 1),
# ((1, 0, 0), 4),
# ((1, 1, 0), 5),
# ((1, 1, 1), 6),
# ])
```
Args:
a: `Tensor` or `SparseTensor` of the same type as `b`. If sparse, indices
must be sorted in row-major order.
b: `Tensor` or `SparseTensor` of the same type as `a`. If sparse, indices
must be sorted in row-major order.
validate_indices: Whether to validate the order and range of sparse indices
in `a` and `b`.
Returns:
A `SparseTensor` whose shape is the same rank as `a` and `b`, and all but
the last dimension the same. Elements along the last dimension contain the
intersections."
9180,set_difference,tensorflow/tensorflow/python/ops/sets_impl.py,212,function,"Compute set difference of elements in last dimension of `a` and `b`.
All but the last dimension of `a` and `b` must match.
Example:
```python
import tensorflow as tf
import collections
# Represent the following array of sets as a sparse tensor:
# a = np.array([[{1, 2}, {3}], [{4}, {5, 6}]])
a = collections.OrderedDict([
((0, 0, 0), 1),
((0, 0, 1), 2),
((0, 1, 0), 3),
((1, 0, 0), 4),
((1, 1, 0), 5),
((1, 1, 1), 6),
])
a = tf.sparse.SparseTensor(list(a.keys()), list(a.values()),
dense_shape=[2, 2, 2])
# np.array([[{1, 3}, {2}], [{4, 5}, {5, 6, 7, 8}]])
b = collections.OrderedDict([
((0, 0, 0), 1),
((0, 0, 1), 3),
((0, 1, 0), 2),
((1, 0, 0), 4),
((1, 0, 1), 5),
((1, 1, 0), 5),
((1, 1, 1), 6),
((1, 1, 2), 7),
((1, 1, 3), 8),
])
b = tf.sparse.SparseTensor(list(b.keys()), list(b.values()),
dense_shape=[2, 2, 4])
# `set_difference` is applied to each aligned pair of sets.
tf.sets.difference(a, b)
# The result will be equivalent to either of:
#
# np.array([[{2}, {3}], [{}, {}]])
#
# collections.OrderedDict([
# ((0, 0, 0), 2),
# ((0, 1, 0), 3),
# ])
```
Args:
a: `Tensor` or `SparseTensor` of the same type as `b`. If sparse, indices
must be sorted in row-major order.
b: `Tensor` or `SparseTensor` of the same type as `a`. If sparse, indices
must be sorted in row-major order.
aminusb: Whether to subtract `b` from `a`, vs vice versa.
validate_indices: Whether to validate the order and range of sparse indices
in `a` and `b`.
Returns:
A `SparseTensor` whose shape is the same rank as `a` and `b`, and all but
the last dimension the same. Elements along the last dimension contain the
differences.
Raises:
TypeError: If inputs are invalid types, or if `a` and `b` have
different types.
ValueError: If `a` is sparse and `b` is dense.
errors_impl.InvalidArgumentError: If the shapes of `a` and `b` do not
match in any dimension other than the last dimension."
9181,set_union,tensorflow/tensorflow/python/ops/sets_impl.py,294,function,"Compute set union of elements in last dimension of `a` and `b`.
All but the last dimension of `a` and `b` must match.
Example:
```python
import tensorflow as tf
import collections
# [[{1, 2}, {3}], [{4}, {5, 6}]]
a = collections.OrderedDict([
((0, 0, 0), 1),
((0, 0, 1), 2),
((0, 1, 0), 3),
((1, 0, 0), 4),
((1, 1, 0), 5),
((1, 1, 1), 6),
])
a = tf.sparse.SparseTensor(list(a.keys()), list(a.values()),
dense_shape=[2, 2, 2])
# [[{1, 3}, {2}], [{4, 5}, {5, 6, 7, 8}]]
b = collections.OrderedDict([
((0, 0, 0), 1),
((0, 0, 1), 3),
((0, 1, 0), 2),
((1, 0, 0), 4),
((1, 0, 1), 5),
((1, 1, 0), 5),
((1, 1, 1), 6),
((1, 1, 2), 7),
((1, 1, 3), 8),
])
b = tf.sparse.SparseTensor(list(b.keys()), list(b.values()),
dense_shape=[2, 2, 4])
# `set_union` is applied to each aligned pair of sets.
tf.sets.union(a, b)
# The result will be a equivalent to either of:
#
# np.array([[{1, 2, 3}, {2, 3}], [{4, 5}, {5, 6, 7, 8}]])
#
# collections.OrderedDict([
# ((0, 0, 0), 1),
# ((0, 0, 1), 2),
# ((0, 0, 2), 3),
# ((0, 1, 0), 2),
# ((0, 1, 1), 3),
# ((1, 0, 0), 4),
# ((1, 0, 1), 5),
# ((1, 1, 0), 5),
# ((1, 1, 1), 6),
# ((1, 1, 2), 7),
# ((1, 1, 3), 8),
# ])
```
Args:
a: `Tensor` or `SparseTensor` of the same type as `b`. If sparse, indices
must be sorted in row-major order.
b: `Tensor` or `SparseTensor` of the same type as `a`. If sparse, indices
must be sorted in row-major order.
validate_indices: Whether to validate the order and range of sparse indices
in `a` and `b`.
Returns:
A `SparseTensor` whose shape is the same rank as `a` and `b`, and all but
the last dimension the same. Elements along the last dimension contain the
unions."
9182,sort,tensorflow/tensorflow/python/ops/sort_ops.py,39,function,"Sorts a tensor.
Usage:
```python
import tensorflow as tf
a = [1, 10, 26.9, 2.8, 166.32, 62.3]
b = tf.sort(a,axis=-1,direction='ASCENDING',name=None)
c = tf.keras.backend.eval(b)
# Here, c = [ 1. 2.8 10. 26.9 62.3 166.32]
```
Args:
values: 1-D or higher numeric `Tensor`.
axis: The axis along which to sort. The default is -1, which sorts the last
axis.
direction: The direction in which to sort the values (`'ASCENDING'` or
`'DESCENDING'`).
name: Optional name for the operation.
Returns:
A `Tensor` with the same dtype and shape as `values`, with the elements
sorted along the given `axis`.
Raises:
ValueError: If axis is not a constant scalar, or the direction is invalid."
9183,argsort,tensorflow/tensorflow/python/ops/sort_ops.py,73,function,"Returns the indices of a tensor that give its sorted order along an axis.
For a 1D tensor, `tf.gather(values, tf.argsort(values))` is equivalent to
`tf.sort(values)`. For higher dimensions, the output has the same shape as
`values`, but along the given axis, values represent the index of the sorted
element in that slice of the tensor at the given position.
Usage:
```python
import tensorflow as tf
a = [1, 10, 26.9, 2.8, 166.32, 62.3]
b = tf.argsort(a,axis=-1,direction='ASCENDING',stable=False,name=None)
c = tf.keras.backend.eval(b)
# Here, c = [0 3 1 2 5 4]
```
Args:
values: 1-D or higher numeric `Tensor`.
axis: The axis along which to sort. The default is -1, which sorts the last
axis.
direction: The direction in which to sort the values (`'ASCENDING'` or
`'DESCENDING'`).
stable: If True, equal elements in the original tensor will not be
re-ordered in the returned order. Unstable sort is not yet implemented,
but will eventually be the default for performance reasons. If you require
a stable order, pass `stable=True` for forwards compatibility.
name: Optional name for the operation.
Returns:
An int32 `Tensor` with the same shape as `values`. The indices that would
sort each slice of the given `values` along the given `axis`.
Raises:
ValueError: If axis is not a constant scalar, or the direction is invalid."
9184,from_dense,tensorflow/tensorflow/python/ops/sparse_ops.py,108,function,"Converts a dense tensor into a sparse tensor.
Only elements not equal to zero will be present in the result. The resulting
`SparseTensor` has the same dtype and shape as the input.
Args:
tensor: A dense `Tensor` to be converted to a `SparseTensor`.
name: Optional name for the op.
Returns:
The `SparseTensor`."
9185,sparse_expand_dims,tensorflow/tensorflow/python/ops/sparse_ops.py,131,function,"Returns a tensor with an length 1 axis inserted at index `axis`.
Given a tensor `input`, this operation inserts a dimension of length 1 at the
dimension index `axis` of `input`'s shape. The dimension index follows python
indexing rules: It's zero-based, a negative index it is counted backward
from the end.
This operation is useful to:
* Add an outer ""batch"" dimension to a single element.
* Align axes for broadcasting.
* To add an inner vector length axis to a tensor of scalars.
For example:
If you have a sparse tensor with shape `[height, width, depth]`:
>>> sp = tf.sparse.SparseTensor(indices=[[3,4,1]], values=[7,],
... dense_shape=[10,10,3])
You can add an outer `batch` axis by passing `axis=0`:
>>> tf.sparse.expand_dims(sp, axis=0).shape.as_list()
[1, 10, 10, 3]
The new axis location matches Python `list.insert(axis, 1)`:
>>> tf.sparse.expand_dims(sp, axis=1).shape.as_list()
[10, 1, 10, 3]
Following standard python indexing rules, a negative `axis` counts from the
end so `axis=-1` adds an inner most dimension:
>>> tf.sparse.expand_dims(sp, axis=-1).shape.as_list()
[10, 10, 3, 1]
Note: Unlike `tf.expand_dims` this function includes a default value for the
`axis`: `-1`. So if `axis is not specified, an inner dimension is added.
>>> sp.shape.as_list()
[10, 10, 3]
>>> tf.sparse.expand_dims(sp).shape.as_list()
[10, 10, 3, 1]
This operation requires that `axis` is a valid index for `input.shape`,
following python indexing rules:
```
-1-tf.rank(input) <= axis <= tf.rank(input)
```
This operation is related to:
* `tf.expand_dims`, which provides this functionality for dense tensors.
* `tf.squeeze`, which removes dimensions of size 1, from dense tensors.
* `tf.sparse.reshape`, which provides more flexible reshaping capability.
Args:
sp_input: A `SparseTensor`.
axis: 0-D (scalar). Specifies the dimension index at which to expand the
shape of `input`. Must be in the range `[-rank(sp_input) - 1,
rank(sp_input)]`. Defaults to `-1`.
name: The name of the output `SparseTensor`.
Returns:
A `SparseTensor` with the same data as `sp_input`, but its shape has an
additional dimension of size 1 added."
9186,sparse_eye,tensorflow/tensorflow/python/ops/sparse_ops.py,237,function,"Creates a two-dimensional sparse tensor with ones along the diagonal.
Args:
num_rows: Non-negative integer or `int32` scalar `tensor` giving the number
of rows in the resulting matrix.
num_columns: Optional non-negative integer or `int32` scalar `tensor` giving
the number of columns in the resulting matrix. Defaults to `num_rows`.
dtype: The type of element in the resulting `Tensor`.
name: A name for this `Op`. Defaults to ""eye"".
Returns:
A `SparseTensor` of shape [num_rows, num_columns] with ones along the
diagonal."
9187,sparse_concat,tensorflow/tensorflow/python/ops/sparse_ops.py,275,function,"Concatenates a list of `SparseTensor` along the specified dimension.
Concatenation is with respect to the dense versions of each sparse input.
It is assumed that each inputs is a `SparseTensor` whose elements are ordered
along increasing dimension number.
If expand_nonconcat_dim is False, all inputs' shapes must match, except for
the concat dimension. If expand_nonconcat_dim is True, then inputs' shapes are
allowed to vary among all inputs.
The `indices`, `values`, and `shapes` lists must have the same length.
If expand_nonconcat_dim is False, then the output shape is identical to the
inputs', except along the concat dimension, where it is the sum of the inputs'
sizes along that dimension.
If expand_nonconcat_dim is True, then the output shape along the non-concat
dimensions will be expand to be the largest among all inputs, and it is the
sum of the inputs sizes along the concat dimension.
The output elements will be resorted to preserve the sort order along
increasing dimension number.
This op runs in `O(M log M)` time, where `M` is the total number of non-empty
values across all inputs. This is due to the need for an internal sort in
order to concatenate efficiently across an arbitrary dimension.
For example, if `axis = 1` and the inputs are
sp_inputs[0]: shape = [2, 3]
[0, 2]: ""a""
[1, 0]: ""b""
[1, 1]: ""c""
sp_inputs[1]: shape = [2, 4]
[0, 1]: ""d""
[0, 2]: ""e""
then the output will be
shape = [2, 7]
[0, 2]: ""a""
[0, 4]: ""d""
[0, 5]: ""e""
[1, 0]: ""b""
[1, 1]: ""c""
Graphically this is equivalent to doing
[ a] concat [ d e ] = [ a d e ]
[b c ] [ ] [b c ]
Another example, if 'axis = 1' and the inputs are
sp_inputs[0]: shape = [3, 3]
[0, 2]: ""a""
[1, 0]: ""b""
[2, 1]: ""c""
sp_inputs[1]: shape = [2, 4]
[0, 1]: ""d""
[0, 2]: ""e""
if expand_nonconcat_dim = False, this will result in an error. But if
expand_nonconcat_dim = True, this will result in:
shape = [3, 7]
[0, 2]: ""a""
[0, 4]: ""d""
[0, 5]: ""e""
[1, 0]: ""b""
[2, 1]: ""c""
Graphically this is equivalent to doing
[ a] concat [ d e ] = [ a d e ]
[b ] [ ] [b ]
[ c ] [ c ]
Args:
axis: Dimension to concatenate along. Must be in range [-rank, rank),
where rank is the number of dimensions in each input `SparseTensor`.
sp_inputs: List of `SparseTensor` to concatenate.
name: A name prefix for the returned tensors (optional).
expand_nonconcat_dim: Whether to allow the expansion in the non-concat
dimensions. Defaulted to False.
concat_dim: The old (deprecated) name for axis.
expand_nonconcat_dims: alias for expand_nonconcat_dim
Returns:
A `SparseTensor` with the concatenated output.
Raises:
TypeError: If `sp_inputs` is not a list of `SparseTensor`."
9188,sparse_concat_v2,tensorflow/tensorflow/python/ops/sparse_ops.py,388,function,
9189,sparse_add,tensorflow/tensorflow/python/ops/sparse_ops.py,430,function,"Adds two tensors, at least one of each is a `SparseTensor`.
If one `SparseTensor` and one `Tensor` are passed in, returns a `Tensor`. If
both arguments are `SparseTensor`s, this returns a `SparseTensor`. The order
of arguments does not matter. Use vanilla `tf.add()` for adding two dense
`Tensor`s.
The shapes of the two operands must match: broadcasting is not supported.
The indices of any input `SparseTensor` are assumed ordered in standard
lexicographic order. If this is not the case, before this step run
`SparseReorder` to restore index ordering.
If both arguments are sparse, we perform ""clipping"" as follows. By default,
if two values sum to zero at some index, the output `SparseTensor` would still
include that particular location in its index, storing a zero in the
corresponding value slot. To override this, callers can specify `thresh`,
indicating that if the sum has a magnitude strictly smaller than `thresh`, its
corresponding value and index would then not be included. In particular,
`thresh == 0.0` (default) means everything is kept and actual thresholding
happens only for a positive value.
For example, suppose the logical sum of two sparse operands is (densified):
[ 2]
[.1 0]
[ 6 -.2]
Then,
* `thresh == 0` (the default): all 5 index/value pairs will be returned.
* `thresh == 0.11`: only .1 and 0 will vanish, and the remaining three
index/value pairs will be returned.
* `thresh == 0.21`: .1, 0, and -.2 will vanish.
Args:
a: The first operand; `SparseTensor` or `Tensor`.
b: The second operand; `SparseTensor` or `Tensor`. At least one operand
must be sparse.
threshold: An optional 0-D `Tensor` (defaults to `0`). The magnitude
threshold that determines if an output value/index pair takes space. Its
dtype should match that of the values if they are real; if the latter are
complex64/complex128, then the dtype should be float32/float64,
correspondingly.
thresh: Deprecated alias for `threshold`.
Returns:
A `SparseTensor` or a `Tensor`, representing the sum.
Raises:
TypeError: If both `a` and `b` are `Tensor`s. Use `tf.add()` instead."
9190,sparse_add_v2,tensorflow/tensorflow/python/ops/sparse_ops.py,491,function,"Adds two tensors, at least one of each is a `SparseTensor`.
If one `SparseTensor` and one `Tensor` are passed in, returns a `Tensor`. If
both arguments are `SparseTensor`s, this returns a `SparseTensor`. The order
of arguments does not matter. Use vanilla `tf.add()` for adding two dense
`Tensor`s.
The shapes of the two operands must match: broadcasting is not supported.
The indices of any input `SparseTensor` are assumed ordered in standard
lexicographic order. If this is not the case, before this step run
`SparseReorder` to restore index ordering.
If both arguments are sparse, we perform ""clipping"" as follows. By default,
if two values sum to zero at some index, the output `SparseTensor` would still
include that particular location in its index, storing a zero in the
corresponding value slot. To override this, callers can specify `threshold`,
indicating that if the sum has a magnitude strictly smaller than `threshold`,
its corresponding value and index would then not be included. In particular,
`threshold == 0.0` (default) means everything is kept and actual thresholding
happens only for a positive value.
For example, suppose the logical sum of two sparse operands is (densified):
[ 2]
[.1 0]
[ 6 -.2]
Then,
* `threshold == 0` (the default): all 5 index/value pairs will be
returned.
* `threshold == 0.11`: only .1 and 0 will vanish, and the remaining three
index/value pairs will be returned.
* `threshold == 0.21`: .1, 0, and -.2 will vanish.
Args:
a: The first operand; `SparseTensor` or `Tensor`.
b: The second operand; `SparseTensor` or `Tensor`. At least one operand
must be sparse.
threshold: A 0-D `Tensor`. The magnitude threshold that determines if an
output value/index pair takes space. Its dtype should match that of the
values if they are real; if the latter are complex64/complex128, then the
dtype should be float32/float64, correspondingly.
Returns:
A `SparseTensor` or a `Tensor`, representing the sum.
Raises:
TypeError: If both `a` and `b` are `Tensor`s. Use `tf.add()` instead."
9191,sparse_cross,tensorflow/tensorflow/python/ops/sparse_ops.py,575,function,"Generates sparse cross from a list of sparse and dense tensors.
For example, if the inputs are
* inputs[0]: SparseTensor with shape = [2, 2]
[0, 0]: ""a""
[1, 0]: ""b""
[1, 1]: ""c""
* inputs[1]: SparseTensor with shape = [2, 1]
[0, 0]: ""d""
[1, 0]: ""e""
* inputs[2]: Tensor [[""f""], [""g""]]
then the output will be:
shape = [2, 2]
[0, 0]: ""a_X_d_X_f""
[1, 0]: ""b_X_e_X_g""
[1, 1]: ""c_X_e_X_g""
Customized separator ""_Y_"":
>>> inp_0 = tf.constant([['a'], ['b']])
>>> inp_1 = tf.constant([['c'], ['d']])
>>> output = tf.sparse.cross([inp_0, inp_1], separator='_Y_')
>>> output.values
<tf.Tensor: shape=(2,), dtype=string, numpy=array([b'a_Y_c', b'b_Y_d'],
dtype=object)>
Args:
inputs: An iterable of `Tensor` or `SparseTensor`.
name: Optional name for the op.
separator: A string added between each string being joined. Defaults to
'_X_'.
Returns:
A `SparseTensor` of type `string`."
9192,sparse_cross_hashed,tensorflow/tensorflow/python/ops/sparse_ops.py,635,function,"Generates hashed sparse cross from a list of sparse and dense tensors.
For example, if the inputs are
* inputs[0]: SparseTensor with shape = [2, 2]
[0, 0]: ""a""
[1, 0]: ""b""
[1, 1]: ""c""
* inputs[1]: SparseTensor with shape = [2, 1]
[0, 0]: ""d""
[1, 0]: ""e""
* inputs[2]: Tensor [[""f""], [""g""]]
then the output will be:
shape = [2, 2]
[0, 0]: FingerprintCat64(
Fingerprint64(""f""), FingerprintCat64(
Fingerprint64(""d""), Fingerprint64(""a"")))
[1, 0]: FingerprintCat64(
Fingerprint64(""g""), FingerprintCat64(
Fingerprint64(""e""), Fingerprint64(""b"")))
[1, 1]: FingerprintCat64(
Fingerprint64(""g""), FingerprintCat64(
Fingerprint64(""e""), Fingerprint64(""c"")))
Args:
inputs: An iterable of `Tensor` or `SparseTensor`.
num_buckets: An `int` that is `>= 0`.
output = hashed_value%num_buckets if num_buckets > 0 else hashed_value.
hash_key: Integer hash_key that will be used by the `FingerprintCat64`
function. If not given, will use a default key.
name: Optional name for the op.
Returns:
A `SparseTensor` of type `int64`."
9193,sparse_dense_cwise_add,tensorflow/tensorflow/python/ops/sparse_ops.py,762,function,"Adds up a SparseTensor and a dense Tensor, using these special rules:
(1) Broadcasts the dense side to have the same shape as the sparse side, if
eligible;
(2) Then, only the dense values pointed to by the indices of the SparseTensor
participate in the cwise addition.
By the rules, the result is a logical SparseTensor with exactly the same
indices and shape, but possibly with different non-zero values. The output of
this Op is the resultant non-zero values.
Args:
sp_t: the SparseTensor operand.
dense_t: the dense Tensor operand; must have the same dtype and a
broadcast-compatible shape as `sp_t`.
Returns:
output: the SparseTensor output."
9194,sparse_reorder,tensorflow/tensorflow/python/ops/sparse_ops.py,789,function,"Reorders a `SparseTensor` into the canonical, row-major ordering.
Note that by convention, all sparse ops preserve the canonical ordering
along increasing dimension number. The only time ordering can be violated
is during manual manipulation of the indices and values to add entries.
Reordering does not affect the shape of the `SparseTensor`.
For example, if `sp_input` has shape `[4, 5]` and `indices` / `values`:
[0, 3]: b
[0, 1]: a
[3, 1]: d
[2, 0]: c
then the output will be a `SparseTensor` of shape `[4, 5]` and
`indices` / `values`:
[0, 1]: a
[0, 3]: b
[2, 0]: c
[3, 1]: d
Args:
sp_input: The input `SparseTensor`.
name: A name prefix for the returned tensors (optional)
Returns:
A `SparseTensor` with the same shape and non-empty values, but in
canonical ordering.
Raises:
TypeError: If `sp_input` is not a `SparseTensor`."
9195,sparse_reshape,tensorflow/tensorflow/python/ops/sparse_ops.py,840,function,"Reshapes a `SparseTensor` to represent values in a new dense shape.
This operation has the same semantics as `reshape` on the represented dense
tensor. The indices of non-empty values in `sp_input` are recomputed based
on the new dense shape, and a new `SparseTensor` is returned containing the
new indices and new shape. The order of non-empty values in `sp_input` is
unchanged.
If one component of `shape` is the special value -1, the size of that
dimension is computed so that the total dense size remains constant. At
most one component of `shape` can be -1. The number of dense elements
implied by `shape` must be the same as the number of dense elements
originally represented by `sp_input`.
For example, if `sp_input` has shape `[2, 3, 6]` and `indices` / `values`:
[0, 0, 0]: a
[0, 0, 1]: b
[0, 1, 0]: c
[1, 0, 0]: d
[1, 2, 3]: e
and `shape` is `[9, -1]`, then the output will be a `SparseTensor` of
shape `[9, 4]` and `indices` / `values`:
[0, 0]: a
[0, 1]: b
[1, 2]: c
[4, 2]: d
[8, 1]: e
Args:
sp_input: The input `SparseTensor`.
shape: A 1-D (vector) int64 `Tensor` specifying the new dense shape of the
represented `SparseTensor`.
name: A name prefix for the returned tensors (optional)
Returns:
A `SparseTensor` with the same non-empty values but with indices calculated
by the new dense shape.
Raises:
TypeError: If `sp_input` is not a `SparseTensor`.
ValueError: If argument `shape` requests a `SparseTensor` with a different
number of elements than `sp_input`.
ValueError: If `shape` has more than one inferred (== -1) dimension."
9196,KeywordRequired,tensorflow/tensorflow/python/ops/sparse_ops.py,945,class,
9197,sparse_split,tensorflow/tensorflow/python/ops/sparse_ops.py,956,function,"Split a `SparseTensor` into `num_split` tensors along `axis`.
If the `sp_input.dense_shape[axis]` is not an integer multiple of `num_split`
each slice starting from 0:`shape[axis] % num_split` gets extra one
dimension. For example, if `axis = 1` and `num_split = 2` and the
input is:
input_tensor = shape = [2, 7]
[ a d e ]
[b c ]
Graphically the output tensors are:
output_tensor[0] =
[ a ]
[b c ]
output_tensor[1] =
[ d e ]
[ ]
Args:
keyword_required: Python 2 standin for * (temporary for argument reorder)
sp_input: The `SparseTensor` to split.
num_split: A Python integer. The number of ways to split.
axis: A 0-D `int32` `Tensor`. The dimension along which to split. Must be in
range [-rank, rank), where rank is the number of dimensions in the input
`SparseTensor`.
name: A name for the operation (optional).
split_dim: Deprecated old name for axis.
Returns:
`num_split` `SparseTensor` objects resulting from splitting `value`.
Raises:
TypeError: If `sp_input` is not a `SparseTensor`.
ValueError: If the deprecated `split_dim` and `axis` are both non None."
9198,sparse_split_v2,tensorflow/tensorflow/python/ops/sparse_ops.py,1029,function,"Split a `SparseTensor` into `num_split` tensors along `axis`.
If the `sp_input.dense_shape[axis]` is not an integer multiple of `num_split`
each slice starting from 0:`shape[axis] % num_split` gets extra one
dimension. For example:
>>> indices = [[0, 2], [0, 4], [0, 5], [1, 0], [1, 1]]
>>> values = [1, 2, 3, 4, 5]
>>> t = tf.SparseTensor(indices=indices, values=values, dense_shape=[2, 7])
>>> tf.sparse.to_dense(t)
<tf.Tensor: shape=(2, 7), dtype=int32, numpy=
array([[0, 0, 1, 0, 2, 3, 0],
[4, 5, 0, 0, 0, 0, 0]], dtype=int32)>
>>> output = tf.sparse.split(sp_input=t, num_split=2, axis=1)
>>> tf.sparse.to_dense(output[0])
<tf.Tensor: shape=(2, 4), dtype=int32, numpy=
array([[0, 0, 1, 0],
[4, 5, 0, 0]], dtype=int32)>
>>> tf.sparse.to_dense(output[1])
<tf.Tensor: shape=(2, 3), dtype=int32, numpy=
array([[2, 3, 0],
[0, 0, 0]], dtype=int32)>
>>> output = tf.sparse.split(sp_input=t, num_split=2, axis=0)
>>> tf.sparse.to_dense(output[0])
<tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[0, 0, 1, 0, 2, 3, 0]],
dtype=int32)>
>>> tf.sparse.to_dense(output[1])
<tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[4, 5, 0, 0, 0, 0, 0]],
dtype=int32)>
>>> output = tf.sparse.split(sp_input=t, num_split=2, axis=-1)
>>> tf.sparse.to_dense(output[0])
<tf.Tensor: shape=(2, 4), dtype=int32, numpy=
array([[0, 0, 1, 0],
[4, 5, 0, 0]], dtype=int32)>
>>> tf.sparse.to_dense(output[1])
<tf.Tensor: shape=(2, 3), dtype=int32, numpy=
array([[2, 3, 0],
[0, 0, 0]], dtype=int32)>
Args:
sp_input: The `SparseTensor` to split.
num_split: A Python integer. The number of ways to split.
axis: A 0-D `int32` `Tensor`. The dimension along which to split. Must be in
range [-rank, rank), where rank is the number of dimensions in the input
`SparseTensor`.
name: A name for the operation (optional).
Returns:
`num_split` `SparseTensor` objects resulting from splitting `value`.
Raises:
TypeError: If `sp_input` is not a `SparseTensor`."
9199,sparse_slice,tensorflow/tensorflow/python/ops/sparse_ops.py,1098,function,"Slice a `SparseTensor` based on the `start` and `size.
For example, if the input is
input_tensor = shape = [2, 7]
[ a d e ]
[b c ]
Graphically the output tensors are:
sparse.slice([0, 0], [2, 4]) = shape = [2, 4]
[ a ]
[b c ]
sparse.slice([0, 4], [2, 3]) = shape = [2, 3]
[ d e ]
[ ]
Args:
sp_input: The `SparseTensor` to split.
start: 1-D. tensor represents the start of the slice.
size: 1-D. tensor represents the size of the slice.
name: A name for the operation (optional).
Returns:
A `SparseTensor` objects resulting from splicing.
Raises:
TypeError: If `sp_input` is not a `SparseTensor`."
9200,sparse_to_dense,tensorflow/tensorflow/python/ops/sparse_ops.py,1151,function,"Converts a sparse representation into a dense tensor.
Builds an array `dense` with shape `output_shape` such that
```python
# If sparse_indices is scalar
dense[i] = (i == sparse_indices ? sparse_values : default_value)
# If sparse_indices is a vector, then for each i
dense[sparse_indices[i]] = sparse_values[i]
# If sparse_indices is an n by d matrix, then for each i in [0, n)
dense[sparse_indices[i][0], ..., sparse_indices[i][d-1]] = sparse_values[i]
```
All other values in `dense` are set to `default_value`. If `sparse_values`
is a scalar, all sparse indices are set to this single value.
Indices should be sorted in lexicographic order, and indices must not
contain any repeats. If `validate_indices` is True, these properties
are checked during execution.
Args:
sparse_indices: A 0-D, 1-D, or 2-D `Tensor` of type `int32` or `int64`.
`sparse_indices[i]` contains the complete index where `sparse_values[i]`
will be placed.
output_shape: A 1-D `Tensor` of the same type as `sparse_indices`. Shape
of the dense output tensor.
sparse_values: A 0-D or 1-D `Tensor`. Values corresponding to each row of
`sparse_indices`, or a scalar value to be used for all sparse indices.
default_value: A 0-D `Tensor` of the same type as `sparse_values`. Value
to set for indices not specified in `sparse_indices`. Defaults to zero.
validate_indices: A boolean value. If True, indices are checked to make
sure they are sorted in lexicographic order and that there are no repeats.
name: A name for the operation (optional).
Returns:
Dense `Tensor` of shape `output_shape`. Has the same type as
`sparse_values`."
9201,sparse_reduce_max_v2,tensorflow/tensorflow/python/ops/sparse_ops.py,1207,function,"Computes the max of elements across dimensions of a SparseTensor.
This Op takes a SparseTensor and is the sparse counterpart to
`tf.reduce_max()`. In particular, this Op also returns a dense `Tensor`
if `output_is_sparse` is `False`, or a `SparseTensor` if `output_is_sparse`
is `True`.
Note: A gradient is not defined for this function, so it can't be used
in training models that need gradient descent.
Reduces `sp_input` along the dimensions given in `axis`. Unless
`keepdims` is true, the rank of the tensor is reduced by 1 for each entry in
`axis`. If `keepdims` is true, the reduced dimensions are retained
with length 1.
If `axis` has no entries, all dimensions are reduced, and a tensor
with a single element is returned. Additionally, the axes can be negative,
similar to the indexing rules in Python.
The values not defined in `sp_input` don't participate in the reduce max,
as opposed to be implicitly assumed 0 -- hence it can return negative values
for sparse `axis`. But, in case there are no values in
`axis`, it will reduce to 0. See second example below.
For example:
```python
# 'x' represents [[1, ?, 2]
# [?, 3, ?]]
# where ? is implicitly-zero.
tf.sparse.reduce_max(x) ==> 3
tf.sparse.reduce_max(x, 0) ==> [1, 3, 2]
tf.sparse.reduce_max(x, 1) ==> [2, 3] # Can also use -1 as the axis.
tf.sparse.reduce_max(x, 1, keepdims=True) ==> [[2], [3]]
tf.sparse.reduce_max(x, [0, 1]) ==> 3
# 'y' represents [[-7, ?]
# [ 4, 3]
# [ ?, ?]
tf.sparse.reduce_max(x, 1) ==> [-7, 4, 0]
```
Args:
sp_input: The SparseTensor to reduce. Should have numeric type.
axis: The dimensions to reduce; list or scalar. If `None` (the
default), reduces all dimensions.
keepdims: If true, retain reduced dimensions with length 1.
output_is_sparse: If true, returns a `SparseTensor` instead of a dense
`Tensor` (the default).
name: A name for the operation (optional).
Returns:
The reduced Tensor or the reduced SparseTensor if `output_is_sparse` is
True."
9202,sparse_reduce_max,tensorflow/tensorflow/python/ops/sparse_ops.py,1295,function,"Computes the max of elements across dimensions of a SparseTensor.
This Op takes a SparseTensor and is the sparse counterpart to
`tf.reduce_max()`. In particular, this Op also returns a dense `Tensor`
instead of a sparse one.
Note: A gradient is not defined for this function, so it can't be used
in training models that need gradient descent.
Reduces `sp_input` along the dimensions given in `reduction_axes`. Unless
`keepdims` is true, the rank of the tensor is reduced by 1 for each entry in
`reduction_axes`. If `keepdims` is true, the reduced dimensions are retained
with length 1.
If `reduction_axes` has no entries, all dimensions are reduced, and a tensor
with a single element is returned. Additionally, the axes can be negative,
similar to the indexing rules in Python.
The values not defined in `sp_input` don't participate in the reduce max,
as opposed to be implicitly assumed 0 -- hence it can return negative values
for sparse `reduction_axes`. But, in case there are no values in
`reduction_axes`, it will reduce to 0. See second example below.
For example:
```python
# 'x' represents [[1, ?, 2]
# [?, 3, ?]]
# where ? is implicitly-zero.
tf.sparse.reduce_max(x) ==> 3
tf.sparse.reduce_max(x, 0) ==> [1, 3, 2]
tf.sparse.reduce_max(x, 1) ==> [2, 3] # Can also use -1 as the axis.
tf.sparse.reduce_max(x, 1, keepdims=True) ==> [[2], [3]]
tf.sparse.reduce_max(x, [0, 1]) ==> 3
# 'y' represents [[-7, ?]
# [ 4, 3]
# [ ?, ?]
tf.sparse.reduce_max(x, 1) ==> [-7, 4, 0]
```
Args:
sp_input: The SparseTensor to reduce. Should have numeric type.
axis: The dimensions to reduce; list or scalar. If `None` (the
default), reduces all dimensions.
keepdims: If true, retain reduced dimensions with length 1.
reduction_axes: Deprecated name of `axis`.
keep_dims: Deprecated alias for `keepdims`.
Returns:
The reduced Tensor."
9203,sparse_reduce_max_sparse,tensorflow/tensorflow/python/ops/sparse_ops.py,1365,function,"Computes the max of elements across dimensions of a SparseTensor.
This Op takes a SparseTensor and is the sparse counterpart to
`tf.reduce_max()`. In contrast to SparseReduceSum, this Op returns a
SparseTensor.
Note: A gradient is not defined for this function, so it can't be used
in training models that need gradient descent.
Reduces `sp_input` along the dimensions given in `reduction_axes`. Unless
`keepdims` is true, the rank of the tensor is reduced by 1 for each entry in
`reduction_axes`. If `keepdims` is true, the reduced dimensions are retained
with length 1.
If `reduction_axes` has no entries, all dimensions are reduced, and a tensor
with a single element is returned. Additionally, the axes can be negative,
which are interpreted according to the indexing rules in Python.
Args:
sp_input: The SparseTensor to reduce. Should have numeric type.
axis: The dimensions to reduce; list or scalar. If `None` (the
default), reduces all dimensions.
keepdims: If true, retain reduced dimensions with length 1.
reduction_axes: Deprecated name of axis.
keep_dims: Deprecated alias for `keepdims`.
Returns:
The reduced SparseTensor."
9204,sparse_reduce_sum_v2,tensorflow/tensorflow/python/ops/sparse_ops.py,1415,function,"Computes the sum of elements across dimensions of a SparseTensor.
This Op takes a SparseTensor and is the sparse counterpart to
`tf.reduce_sum()`. In particular, this Op also returns a dense `Tensor`
if `output_is_sparse` is `False`, or a `SparseTensor` if `output_is_sparse`
is `True`.
Note: if `output_is_sparse` is True, a gradient is not defined for this
function, so it can't be used in training models that need gradient descent.
Reduces `sp_input` along the dimensions given in `axis`. Unless `keepdims` is
true, the rank of the tensor is reduced by 1 for each entry in `axis`. If
`keepdims` is true, the reduced dimensions are retained with length 1.
If `axis` has no entries, all dimensions are reduced, and a tensor
with a single element is returned. Additionally, the axes can be negative,
similar to the indexing rules in Python.
For example:
```python
# 'x' represents [[1, ?, 1]
# [?, 1, ?]]
# where ? is implicitly-zero.
tf.sparse.reduce_sum(x) ==> 3
tf.sparse.reduce_sum(x, 0) ==> [1, 1, 1]
tf.sparse.reduce_sum(x, 1) ==> [2, 1] # Can also use -1 as the axis.
tf.sparse.reduce_sum(x, 1, keepdims=True) ==> [[2], [1]]
tf.sparse.reduce_sum(x, [0, 1]) ==> 3
```
Args:
sp_input: The SparseTensor to reduce. Should have numeric type.
axis: The dimensions to reduce; list or scalar. If `None` (the
default), reduces all dimensions.
keepdims: If true, retain reduced dimensions with length 1.
output_is_sparse: If true, returns a `SparseTensor` instead of a dense
`Tensor` (the default).
name: A name for the operation (optional).
Returns:
The reduced Tensor or the reduced SparseTensor if `output_is_sparse` is
True."
9205,sparse_reduce_sum,tensorflow/tensorflow/python/ops/sparse_ops.py,1491,function,"Computes the sum of elements across dimensions of a SparseTensor.
This Op takes a SparseTensor and is the sparse counterpart to
`tf.reduce_sum()`. In particular, this Op also returns a dense `Tensor`
instead of a sparse one.
Reduces `sp_input` along the dimensions given in `reduction_axes`. Unless
`keepdims` is true, the rank of the tensor is reduced by 1 for each entry in
`reduction_axes`. If `keepdims` is true, the reduced dimensions are retained
with length 1.
If `reduction_axes` has no entries, all dimensions are reduced, and a tensor
with a single element is returned. Additionally, the axes can be negative,
similar to the indexing rules in Python.
For example:
```python
# 'x' represents [[1, ?, 1]
# [?, 1, ?]]
# where ? is implicitly-zero.
tf.sparse.reduce_sum(x) ==> 3
tf.sparse.reduce_sum(x, 0) ==> [1, 1, 1]
tf.sparse.reduce_sum(x, 1) ==> [2, 1] # Can also use -1 as the axis.
tf.sparse.reduce_sum(x, 1, keepdims=True) ==> [[2], [1]]
tf.sparse.reduce_sum(x, [0, 1]) ==> 3
```
Args:
sp_input: The SparseTensor to reduce. Should have numeric type.
axis: The dimensions to reduce; list or scalar. If `None` (the
default), reduces all dimensions.
keepdims: If true, retain reduced dimensions with length 1.
reduction_axes: Deprecated name of `axis`.
keep_dims: Deprecated alias for `keepdims`.
Returns:
The reduced Tensor."
9206,sparse_reduce_sum_sparse,tensorflow/tensorflow/python/ops/sparse_ops.py,1548,function,"Computes the sum of elements across dimensions of a SparseTensor.
This Op takes a SparseTensor and is the sparse counterpart to
`tf.reduce_sum()`. In contrast to SparseReduceSum, this Op returns a
SparseTensor.
Note: A gradient is not defined for this function, so it can't be used
in training models that need gradient descent.
Reduces `sp_input` along the dimensions given in `reduction_axes`. Unless
`keepdims` is true, the rank of the tensor is reduced by 1 for each entry in
`reduction_axes`. If `keepdims` is true, the reduced dimensions are retained
with length 1.
If `reduction_axes` has no entries, all dimensions are reduced, and a tensor
with a single element is returned. Additionally, the axes can be negative,
which are interpreted according to the indexing rules in Python.
Args:
sp_input: The SparseTensor to reduce. Should have numeric type.
axis: The dimensions to reduce; list or scalar. If `None` (the
default), reduces all dimensions.
keepdims: If true, retain reduced dimensions with length 1.
reduction_axes: Deprecated name of axis.
keep_dims: Deprecated alias for `keepdims`.
Returns:
The reduced SparseTensor."
9207,sparse_tensor_to_dense,tensorflow/tensorflow/python/ops/sparse_ops.py,1599,function,"Converts a `SparseTensor` into a dense tensor.
This op is a convenience wrapper around `sparse_to_dense` for `SparseTensor`s.
For example, if `sp_input` has shape `[3, 5]` and non-empty string values:
[0, 1]: a
[0, 3]: b
[2, 0]: c
and `default_value` is `x`, then the output will be a dense `[3, 5]`
string tensor with values:
[[x a x b x]
[x x x x x]
[c x x x x]]
Indices must be without repeats. This is only
tested if `validate_indices` is `True`.
Args:
sp_input: The input `SparseTensor`.
default_value: Scalar value to set for indices not specified in
`sp_input`. Defaults to zero.
validate_indices: A boolean value. If `True`, indices are checked to make
sure they are sorted in lexicographic order and that there are no repeats.
name: A name prefix for the returned tensors (optional).
Returns:
A dense tensor with shape `sp_input.dense_shape` and values specified by
the non-empty values in `sp_input`. Indices not in `sp_input` are assigned
`default_value`.
Raises:
TypeError: If `sp_input` is not a `SparseTensor`."
9208,sparse_to_indicator,tensorflow/tensorflow/python/ops/sparse_ops.py,1655,function,"Converts a `SparseTensor` of ids into a dense bool indicator tensor.
The last dimension of `sp_input.indices` is discarded and replaced with
the values of `sp_input`. If `sp_input.dense_shape = [D0, D1, ..., Dn, K]`,
then `output.shape = [D0, D1, ..., Dn, vocab_size]`, where
output[d_0, d_1, ..., d_n, sp_input[d_0, d_1, ..., d_n, k]] = True
and False elsewhere in `output`.
For example, if `sp_input.dense_shape = [2, 3, 4]` with non-empty values:
[0, 0, 0]: 0
[0, 1, 0]: 10
[1, 0, 3]: 103
[1, 1, 1]: 150
[1, 1, 2]: 149
[1, 1, 3]: 150
[1, 2, 1]: 121
and `vocab_size = 200`, then the output will be a `[2, 3, 200]` dense bool
tensor with False everywhere except at positions
(0, 0, 0), (0, 1, 10), (1, 0, 103), (1, 1, 149), (1, 1, 150),
(1, 2, 121).
Note that repeats are allowed in the input SparseTensor.
This op is useful for converting `SparseTensor`s into dense formats for
compatibility with ops that expect dense tensors.
The input `SparseTensor` must be in row-major order.
Args:
sp_input: A `SparseTensor` with `values` property of type `int32` or
`int64`.
vocab_size: A scalar int64 Tensor (or Python int) containing the new size
of the last dimension, `all(0 <= sp_input.values < vocab_size)`.
name: A name prefix for the returned tensors (optional)
Returns:
A dense bool indicator tensor representing the indices with specified value.
Raises:
TypeError: If `sp_input` is not a `SparseTensor`."
9209,sparse_merge,tensorflow/tensorflow/python/ops/sparse_ops.py,1719,function,"Combines a batch of feature ids and values into a single `SparseTensor`.
The most common use case for this function occurs when feature ids and
their corresponding values are stored in `Example` protos on disk.
`parse_example` will return a batch of ids and a batch of values, and this
function joins them into a single logical `SparseTensor` for use in
functions such as `sparse_tensor_dense_matmul`, `sparse_to_dense`, etc.
The `SparseTensor` returned by this function has the following properties:
- `indices` is equivalent to `sp_ids.indices` with the last
dimension discarded and replaced with `sp_ids.values`.
- `values` is simply `sp_values.values`.
- If `sp_ids.dense_shape = [D0, D1, ..., Dn, K]`, then
`output.shape = [D0, D1, ..., Dn, vocab_size]`.
For example, consider the following feature vectors:
```python
vector1 = [-3, 0, 0, 0, 0, 0]
vector2 = [ 0, 1, 0, 4, 1, 0]
vector3 = [ 5, 0, 0, 9, 0, 0]
```
These might be stored sparsely in the following Example protos by storing
only the feature ids (column number if the vectors are treated as a matrix)
of the non-zero elements and the corresponding values:
```python
examples = [Example(features={
""ids"": Feature(int64_list=Int64List(value=[0])),
""values"": Feature(float_list=FloatList(value=[-3]))}),
Example(features={
""ids"": Feature(int64_list=Int64List(value=[1, 4, 3])),
""values"": Feature(float_list=FloatList(value=[1, 1, 4]))}),
Example(features={
""ids"": Feature(int64_list=Int64List(value=[0, 3])),
""values"": Feature(float_list=FloatList(value=[5, 9]))})]
```
The result of calling parse_example on these examples will produce a
dictionary with entries for ""ids"" and ""values"". Passing those two objects
to this function along with vocab_size=6, will produce a `SparseTensor` that
sparsely represents all three instances. Namely, the `indices` property will
contain the coordinates of the non-zero entries in the feature matrix (the
first dimension is the row number in the matrix, i.e., the index within the
batch, and the second dimension is the column number, i.e., the feature id);
`values` will contain the actual values. `shape` will be the shape of the
original matrix, i.e., (3, 6). For our example above, the output will be
equal to:
```python
SparseTensor(indices=[[0, 0], [1, 1], [1, 3], [1, 4], [2, 0], [2, 3]],
values=[-3, 1, 4, 1, 5, 9],
dense_shape=[3, 6])
```
This method generalizes to higher-dimensions by simply providing a list for
both the sp_ids as well as the vocab_size.
In this case the resulting `SparseTensor` has the following properties:
- `indices` is equivalent to `sp_ids[0].indices` with the last
dimension discarded and concatenated with
`sp_ids[0].values, sp_ids[1].values, ...`.
- `values` is simply `sp_values.values`.
- If `sp_ids.dense_shape = [D0, D1, ..., Dn, K]`, then
`output.shape = [D0, D1, ..., Dn] + vocab_size`.
Args:
sp_ids: A single `SparseTensor` with `values` property of type `int32`
or `int64` or a Python list of such `SparseTensor`s or a list thereof.
sp_values: A `SparseTensor` of any type.
vocab_size: A scalar `int64` Tensor (or Python int) containing the new size
of the last dimension, `all(0 <= sp_ids.values < vocab_size)`.
Or a list thereof with `all(0 <= sp_ids[i].values < vocab_size[i])` for
all `i`.
name: A name prefix for the returned tensors (optional)
already_sorted: A boolean to specify whether the per-batch values in
`sp_values` are already sorted. If so skip sorting, False by default
(optional).
Returns:
A `SparseTensor` compactly representing a batch of feature ids and values,
useful for passing to functions that expect such a `SparseTensor`.
Raises:
TypeError: If `sp_values` is not a `SparseTensor`. Or if `sp_ids` is neither
a `SparseTensor` nor a list thereof. Or if `vocab_size` is not a
`Tensor` or a Python int and `sp_ids` is a `SparseTensor`. Or if
`vocab_size` is not a or list thereof and `sp_ids` is a list.
ValueError: If `sp_ids` and `vocab_size` are lists of different lengths."
9210,sparse_merge_impl,tensorflow/tensorflow/python/ops/sparse_ops.py,1815,function,Internal implementation for sparse_merge to avoid deprecation warnings.
9211,sparse_retain,tensorflow/tensorflow/python/ops/sparse_ops.py,1873,function,"Retains specified non-empty values within a `SparseTensor`.
For example, if `sp_input` has shape `[4, 5]` and 4 non-empty string values:
[0, 1]: a
[0, 3]: b
[2, 0]: c
[3, 1]: d
and `to_retain = [True, False, False, True]`, then the output will
be a `SparseTensor` of shape `[4, 5]` with 2 non-empty values:
[0, 1]: a
[3, 1]: d
Args:
sp_input: The input `SparseTensor` with `N` non-empty elements.
to_retain: A bool vector of length `N` with `M` true values.
Returns:
A `SparseTensor` with the same shape as the input and `M` non-empty
elements corresponding to the true positions in `to_retain`.
Raises:
TypeError: If `sp_input` is not a `SparseTensor`."
9212,sparse_reset_shape,tensorflow/tensorflow/python/ops/sparse_ops.py,1921,function,"Resets the shape of a `SparseTensor` with indices and values unchanged.
If `new_shape` is None, returns a copy of `sp_input` with its shape reset
to the tight bounding box of `sp_input`. This will be a shape consisting of
all zeros if sp_input has no values.
If `new_shape` is provided, then it must be larger or equal in all dimensions
compared to the shape of `sp_input`. When this condition is met, the returned
SparseTensor will have its shape reset to `new_shape` and its indices and
values unchanged from that of `sp_input.`
For example:
Consider a `sp_input` with shape [2, 3, 5]:
[0, 0, 1]: a
[0, 1, 0]: b
[0, 2, 2]: c
[1, 0, 3]: d
- It is an error to set `new_shape` as [3, 7] since this represents a
rank-2 tensor while `sp_input` is rank-3. This is either a ValueError
during graph construction (if both shapes are known) or an OpError during
run time.
- Setting `new_shape` as [2, 3, 6] will be fine as this shape is larger or
equal in every dimension compared to the original shape [2, 3, 5].
- On the other hand, setting new_shape as [2, 3, 4] is also an error: The
third dimension is smaller than the original shape [2, 3, 5] (and an
`InvalidArgumentError` will be raised).
- If `new_shape` is None, the returned SparseTensor will have a shape
[2, 3, 4], which is the tight bounding box of `sp_input`.
Args:
sp_input: The input `SparseTensor`.
new_shape: None or a vector representing the new shape for the returned
`SparseTensor`.
Returns:
A `SparseTensor` indices and values unchanged from `input_sp`. Its shape is
`new_shape` if that is set. Otherwise it is the tight bounding box of
`input_sp`
Raises:
TypeError: If `sp_input` is not a `SparseTensor`.
ValueError: If `new_shape` represents a tensor with a different rank from
that of `sp_input` (if shapes are known when graph is constructed).
ValueError: If `new_shape` is determined during graph build to have
dimension sizes that are too small.
OpError:
- If `new_shape` has dimension sizes that are too small.
- If shapes are not known during graph construction time, and during run
time it is found out that the ranks do not match."
9213,sparse_fill_empty_rows,tensorflow/tensorflow/python/ops/sparse_ops.py,2027,function,"Fills empty rows in the input 2-D `SparseTensor` with a default value.
This op adds entries with the specified `default_value` at index
`[row, 0]` for any row in the input that does not already have a value.
For example, suppose `sp_input` has shape `[5, 6]` and non-empty values:
[0, 1]: a
[0, 3]: b
[2, 0]: c
[3, 1]: d
Rows 1 and 4 are empty, so the output will be of shape `[5, 6]` with values:
[0, 1]: a
[0, 3]: b
[1, 0]: default_value
[2, 0]: c
[3, 1]: d
[4, 0]: default_value
Note that the input may have empty columns at the end, with no effect on
this op.
The output `SparseTensor` will be in row-major order and will have the
same shape as the input.
This op also returns an indicator vector such that
empty_row_indicator[i] = True iff row i was an empty row.
Args:
sp_input: A `SparseTensor` with shape `[N, M]`.
default_value: The value to fill for empty rows, with the same type as
`sp_input.`
name: A name prefix for the returned tensors (optional)
Returns:
sp_ordered_output: A `SparseTensor` with shape `[N, M]`, and with all empty
rows filled in with `default_value`.
empty_row_indicator: A bool vector of length `N` indicating whether each
input row was empty.
Raises:
TypeError: If `sp_input` is not a `SparseTensor`."
9214,serialize_sparse,tensorflow/tensorflow/python/ops/sparse_ops.py,2093,function,"Serialize a `SparseTensor` into a 3-vector (1-D `Tensor`) object.
Args:
sp_input: The input `SparseTensor`.
name: A name prefix for the returned tensors (optional).
out_type: The `dtype` to use for serialization.
Returns:
A 3-vector (1-D `Tensor`), with each column representing the serialized
`SparseTensor`'s indices, values, and shape (respectively).
Raises:
TypeError: If `sp_input` is not a `SparseTensor`."
9215,serialize_sparse_v2,tensorflow/tensorflow/python/ops/sparse_ops.py,2113,function,"Serialize a `SparseTensor` into a 3-vector (1-D `Tensor`) object.
Args:
sp_input: The input `SparseTensor`.
out_type: The `dtype` to use for serialization.
name: A name prefix for the returned tensors (optional).
Returns:
A 3-vector (1-D `Tensor`), with each column representing the serialized
`SparseTensor`'s indices, values, and shape (respectively).
Raises:
TypeError: If `sp_input` is not a `SparseTensor`."
9216,serialize_many_sparse,tensorflow/tensorflow/python/ops/sparse_ops.py,2141,function,"Serialize `N`-minibatch `SparseTensor` into an `[N, 3]` `Tensor`.
The `SparseTensor` must have rank `R` greater than 1, and the first dimension
is treated as the minibatch dimension. Elements of the `SparseTensor`
must be sorted in increasing order of this first dimension. The serialized
`SparseTensor` objects going into each row of the output `Tensor` will have
rank `R-1`.
The minibatch size `N` is extracted from `sparse_shape[0]`.
Args:
sp_input: The input rank `R` `SparseTensor`.
name: A name prefix for the returned tensors (optional).
out_type: The `dtype` to use for serialization.
Returns:
A matrix (2-D `Tensor`) with `N` rows and `3` columns. Each column
represents serialized `SparseTensor`'s indices, values, and shape
(respectively).
Raises:
TypeError: If `sp_input` is not a `SparseTensor`."
9217,serialize_many_sparse_v2,tensorflow/tensorflow/python/ops/sparse_ops.py,2170,function,"Serialize `N`-minibatch `SparseTensor` into an `[N, 3]` `Tensor`.
The `SparseTensor` must have rank `R` greater than 1, and the first dimension
is treated as the minibatch dimension. Elements of the `SparseTensor`
must be sorted in increasing order of this first dimension. The serialized
`SparseTensor` objects going into each row of the output `Tensor` will have
rank `R-1`.
The minibatch size `N` is extracted from `sparse_shape[0]`.
Args:
sp_input: The input rank `R` `SparseTensor`.
out_type: The `dtype` to use for serialization.
name: A name prefix for the returned tensors (optional).
Returns:
A matrix (2-D `Tensor`) with `N` rows and `3` columns. Each column
represents serialized `SparseTensor`'s indices, values, and shape
(respectively).
Raises:
TypeError: If `sp_input` is not a `SparseTensor`."
9218,deserialize_sparse,tensorflow/tensorflow/python/ops/sparse_ops.py,2204,function,"Deserialize `SparseTensor` objects.
The input `serialized_sparse` must have the shape `[?, ?, ..., ?, 3]` where
the last dimension stores serialized `SparseTensor` objects and the other N
dimensions (N >= 0) correspond to a batch. The ranks of the original
`SparseTensor` objects must all match. When the final `SparseTensor` is
created, its rank is the rank of the incoming `SparseTensor` objects plus N;
the sparse tensors have been concatenated along new dimensions, one for each
batch.
The output `SparseTensor` object's shape values for the original dimensions
are the max across the input `SparseTensor` objects' shape values for the
corresponding dimensions. The new dimensions match the size of the batch.
The input `SparseTensor` objects' indices are assumed ordered in
standard lexicographic order. If this is not the case, after this
step run `SparseReorder` to restore index ordering.
For example, if the serialized input is a `[2 x 3]` matrix representing two
original `SparseTensor` objects:
index = [ 0]
[10]
[20]
values = [1, 2, 3]
shape = [50]
and
index = [ 2]
[10]
values = [4, 5]
shape = [30]
then the final deserialized `SparseTensor` will be:
index = [0 0]
[0 10]
[0 20]
[1 2]
[1 10]
values = [1, 2, 3, 4, 5]
shape = [2 50]
Args:
serialized_sparse: The serialized `SparseTensor` objects.
The last dimension must have 3 columns.
dtype: The `dtype` of the serialized `SparseTensor` objects.
rank: (optional) Python int, the rank of the `SparseTensor` objects.
name: A name prefix for the returned tensors (optional).
Returns:
A `SparseTensor` representing the deserialized `SparseTensor` objects."
9219,deserialize_many_sparse,tensorflow/tensorflow/python/ops/sparse_ops.py,2275,function,"Deserialize and concatenate `SparseTensors` from a serialized minibatch.
The input `serialized_sparse` must be a string matrix of shape `[N x 3]` where
`N` is the minibatch size and the rows correspond to packed outputs of
`serialize_sparse`. The ranks of the original `SparseTensor` objects
must all match. When the final `SparseTensor` is created, it has rank one
higher than the ranks of the incoming `SparseTensor` objects (they have been
concatenated along a new row dimension).
The output `SparseTensor` object's shape values for all dimensions but the
first are the max across the input `SparseTensor` objects' shape values
for the corresponding dimensions. Its first shape value is `N`, the minibatch
size.
The input `SparseTensor` objects' indices are assumed ordered in
standard lexicographic order. If this is not the case, after this
step run `sparse.reorder` to restore index ordering.
For example, if the serialized input is a `[2, 3]` matrix representing two
original `SparseTensor` objects:
index = [ 0]
[10]
[20]
values = [1, 2, 3]
shape = [50]
and
index = [ 2]
[10]
values = [4, 5]
shape = [30]
then the final deserialized `SparseTensor` will be:
index = [0 0]
[0 10]
[0 20]
[1 2]
[1 10]
values = [1, 2, 3, 4, 5]
shape = [2 50]
Args:
serialized_sparse: 2-D `Tensor` of type `string` of shape `[N, 3]`.
The serialized and packed `SparseTensor` objects.
dtype: The `dtype` of the serialized `SparseTensor` objects.
rank: (optional) Python int, the rank of the `SparseTensor` objects.
name: A name prefix for the returned tensors (optional)
Returns:
A `SparseTensor` representing the deserialized `SparseTensor`s,
concatenated along the `SparseTensor`s' first dimension.
All of the serialized `SparseTensor`s must have had the same rank and type."
9220,sparse_tensor_dense_matmul,tensorflow/tensorflow/python/ops/sparse_ops.py,2348,function,"Multiply SparseTensor (or dense Matrix) (of rank 2) ""A"" by dense matrix
(or SparseTensor) ""B"". Please note that one and only one of the inputs MUST
be a SparseTensor and the other MUST be a dense matrix.
No validity checking is performed on the indices of `A`. However, the
following input format is recommended for optimal behavior:
* If `adjoint_a == false`: `A` should be sorted in lexicographically
increasing order. Use `sparse.reorder` if you're not sure.
* If `adjoint_a == true`: `A` should be sorted in order of increasing
dimension 1 (i.e., ""column major"" order instead of ""row major"" order).
Using `tf.nn.embedding_lookup_sparse` for sparse multiplication:
It's not obvious but you can consider `embedding_lookup_sparse` as another
sparse and dense multiplication. In some situations, you may prefer to use
`embedding_lookup_sparse` even though you're not dealing with embeddings.
There are two questions to ask in the decision process: Do you need gradients
computed as sparse too? Is your sparse data represented as two
`SparseTensor`s: ids and values? There is more explanation about data format
below. If you answer any of these questions as yes, consider using
`tf.nn.embedding_lookup_sparse`.
Following explains differences between the expected SparseTensors:
For example if dense form of your sparse data has shape `[3, 5]` and values:
[[ a ]
[b c]
[ d ]]
`SparseTensor` format expected by `sparse_tensor_dense_matmul`:
`sp_a` (indices, values):
[0, 1]: a
[1, 0]: b
[1, 4]: c
[2, 2]: d
`SparseTensor` format expected by `embedding_lookup_sparse`:
`sp_ids` `sp_weights`
[0, 0]: 1 [0, 0]: a
[1, 0]: 0 [1, 0]: b
[1, 1]: 4 [1, 1]: c
[2, 0]: 2 [2, 0]: d
Deciding when to use `sparse_tensor_dense_matmul` vs.
`matmul`(a_is_sparse=True):
There are a number of questions to ask in the decision process, including:
* Will the SparseTensor `A` fit in memory if densified?
* Is the column count of the product large (>> 1)?
* Is the density of `A` larger than approximately 15%?
If the answer to several of these questions is yes, consider
converting the `SparseTensor` to a dense one and using `tf.matmul` with
`a_is_sparse=True`.
This operation tends to perform well when `A` is more sparse, if the column
size of the product is small (e.g. matrix-vector multiplication), if
`sp_a.dense_shape` takes on large values.
Below is a rough speed comparison between `sparse_tensor_dense_matmul`,
labeled 'sparse', and `matmul`(a_is_sparse=True), labeled 'dense'. For
purposes of the comparison, the time spent converting from a `SparseTensor` to
a dense `Tensor` is not included, so it is overly conservative with respect to
the time ratio.
Benchmark system:
CPU: Intel Ivybridge with HyperThreading (6 cores) dL1:32KB dL2:256KB dL3:12MB
GPU: NVidia Tesla k40c
Compiled with:
`-c opt --config=cuda --copt=-mavx`
```
tensorflow/python/sparse_tensor_dense_matmul_op_test --benchmarks
A sparse [m, k] with % nonzero values between 1% and 80%
B dense [k, n]
% nnz n gpu m k dt(dense) dt(sparse) dt(sparse)/dt(dense)
0.01 1 True 100 100 0.000221166 0.00010154 0.459112
0.01 1 True 100 1000 0.00033858 0.000109275 0.322745
0.01 1 True 1000 100 0.000310557 9.85661e-05 0.317385
0.01 1 True 1000 1000 0.0008721 0.000100875 0.115669
0.01 1 False 100 100 0.000208085 0.000107603 0.51711
0.01 1 False 100 1000 0.000327112 9.51118e-05 0.290762
0.01 1 False 1000 100 0.000308222 0.00010345 0.335635
0.01 1 False 1000 1000 0.000865721 0.000101397 0.117124
0.01 10 True 100 100 0.000218522 0.000105537 0.482958
0.01 10 True 100 1000 0.000340882 0.000111641 0.327506
0.01 10 True 1000 100 0.000315472 0.000117376 0.372064
0.01 10 True 1000 1000 0.000905493 0.000123263 0.136128
0.01 10 False 100 100 0.000221529 9.82571e-05 0.44354
0.01 10 False 100 1000 0.000330552 0.000112615 0.340687
0.01 10 False 1000 100 0.000341277 0.000114097 0.334324
0.01 10 False 1000 1000 0.000819944 0.000120982 0.147549
0.01 25 True 100 100 0.000207806 0.000105977 0.509981
0.01 25 True 100 1000 0.000322879 0.00012921 0.400181
0.01 25 True 1000 100 0.00038262 0.00014158 0.370035
0.01 25 True 1000 1000 0.000865438 0.000202083 0.233504
0.01 25 False 100 100 0.000209401 0.000104696 0.499979
0.01 25 False 100 1000 0.000321161 0.000130737 0.407076
0.01 25 False 1000 100 0.000377012 0.000136801 0.362856
0.01 25 False 1000 1000 0.000861125 0.00020272 0.235413
0.2 1 True 100 100 0.000206952 9.69219e-05 0.46833
0.2 1 True 100 1000 0.000348674 0.000147475 0.422959
0.2 1 True 1000 100 0.000336908 0.00010122 0.300439
0.2 1 True 1000 1000 0.001022 0.000203274 0.198898
0.2 1 False 100 100 0.000207532 9.5412e-05 0.459746
0.2 1 False 100 1000 0.000356127 0.000146824 0.41228
0.2 1 False 1000 100 0.000322664 0.000100918 0.312764
0.2 1 False 1000 1000 0.000998987 0.000203442 0.203648
0.2 10 True 100 100 0.000211692 0.000109903 0.519165
0.2 10 True 100 1000 0.000372819 0.000164321 0.440753
0.2 10 True 1000 100 0.000338651 0.000144806 0.427596
0.2 10 True 1000 1000 0.00108312 0.000758876 0.70064
0.2 10 False 100 100 0.000215727 0.000110502 0.512231
0.2 10 False 100 1000 0.000375419 0.0001613 0.429653
0.2 10 False 1000 100 0.000336999 0.000145628 0.432132
0.2 10 False 1000 1000 0.00110502 0.000762043 0.689618
0.2 25 True 100 100 0.000218705 0.000129913 0.594009
0.2 25 True 100 1000 0.000394794 0.00029428 0.745402
0.2 25 True 1000 100 0.000404483 0.0002693 0.665788
0.2 25 True 1000 1000 0.0012002 0.00194494 1.62052
0.2 25 False 100 100 0.000221494 0.0001306 0.589632
0.2 25 False 100 1000 0.000396436 0.000297204 0.74969
0.2 25 False 1000 100 0.000409346 0.000270068 0.659754
0.2 25 False 1000 1000 0.00121051 0.00193737 1.60046
0.5 1 True 100 100 0.000214981 9.82111e-05 0.456836
0.5 1 True 100 1000 0.000415328 0.000223073 0.537101
0.5 1 True 1000 100 0.000358324 0.00011269 0.314492
0.5 1 True 1000 1000 0.00137612 0.000437401 0.317851
0.5 1 False 100 100 0.000224196 0.000101423 0.452386
0.5 1 False 100 1000 0.000400987 0.000223286 0.556841
0.5 1 False 1000 100 0.000368825 0.00011224 0.304318
0.5 1 False 1000 1000 0.00136036 0.000429369 0.31563
0.5 10 True 100 100 0.000222125 0.000112308 0.505608
0.5 10 True 100 1000 0.000461088 0.00032357 0.701753
0.5 10 True 1000 100 0.000394624 0.000225497 0.571422
0.5 10 True 1000 1000 0.00158027 0.00190898 1.20801
0.5 10 False 100 100 0.000232083 0.000114978 0.495418
0.5 10 False 100 1000 0.000454574 0.000324632 0.714146
0.5 10 False 1000 100 0.000379097 0.000227768 0.600817
0.5 10 False 1000 1000 0.00160292 0.00190168 1.18638
0.5 25 True 100 100 0.00023429 0.000151703 0.647501
0.5 25 True 100 1000 0.000497462 0.000598873 1.20386
0.5 25 True 1000 100 0.000460778 0.000557038 1.20891
0.5 25 True 1000 1000 0.00170036 0.00467336 2.74845
0.5 25 False 100 100 0.000228981 0.000155334 0.678371
0.5 25 False 100 1000 0.000496139 0.000620789 1.25124
0.5 25 False 1000 100 0.00045473 0.000551528 1.21287
0.5 25 False 1000 1000 0.00171793 0.00467152 2.71927
0.8 1 True 100 100 0.000222037 0.000105301 0.47425
0.8 1 True 100 1000 0.000410804 0.000329327 0.801664
0.8 1 True 1000 100 0.000349735 0.000131225 0.375212
0.8 1 True 1000 1000 0.00139219 0.000677065 0.48633
0.8 1 False 100 100 0.000214079 0.000107486 0.502085
0.8 1 False 100 1000 0.000413746 0.000323244 0.781261
0.8 1 False 1000 100 0.000348983 0.000131983 0.378193
0.8 1 False 1000 1000 0.00136296 0.000685325 0.50282
0.8 10 True 100 100 0.000229159 0.00011825 0.516017
0.8 10 True 100 1000 0.000498845 0.000532618 1.0677
0.8 10 True 1000 100 0.000383126 0.00029935 0.781336
0.8 10 True 1000 1000 0.00162866 0.00307312 1.88689
0.8 10 False 100 100 0.000230783 0.000124958 0.541452
0.8 10 False 100 1000 0.000493393 0.000550654 1.11606
0.8 10 False 1000 100 0.000377167 0.000298581 0.791642
0.8 10 False 1000 1000 0.00165795 0.00305103 1.84024
0.8 25 True 100 100 0.000233496 0.000175241 0.75051
0.8 25 True 100 1000 0.00055654 0.00102658 1.84458
0.8 25 True 1000 100 0.000463814 0.000783267 1.68875
0.8 25 True 1000 1000 0.00186905 0.00755344 4.04132
0.8 25 False 100 100 0.000240243 0.000175047 0.728625
0.8 25 False 100 1000 0.000578102 0.00104499 1.80763
0.8 25 False 1000 100 0.000485113 0.000776849 1.60138
0.8 25 False 1000 1000 0.00211448 0.00752736 3.55992
```
Args:
sp_a: SparseTensor (or dense Matrix) A, of rank 2.
b: dense Matrix (or SparseTensor) B, with the same dtype as sp_a.
adjoint_a: Use the adjoint of A in the matrix multiply. If A is complex,
this is transpose(conj(A)). Otherwise it's transpose(A).
adjoint_b: Use the adjoint of B in the matrix multiply. If B is complex,
this is transpose(conj(B)). Otherwise it's transpose(B).
name: A name prefix for the returned tensors (optional)
Returns:
A dense matrix (pseudo-code in dense np.matrix notation):
`A = A.H if adjoint_a else A`
`B = B.H if adjoint_b else B`
`return A*B`"
9221,sparse_softmax,tensorflow/tensorflow/python/ops/sparse_ops.py,2584,function,"Applies softmax to a batched N-D `SparseTensor`.
The inputs represent an N-D SparseTensor with logical shape `[..., B, C]`
(where `N >= 2`), and with indices sorted in the canonical lexicographic
order.
This op is equivalent to applying the normal `tf.nn.softmax()` to each
innermost logical submatrix with shape `[B, C]`, but with the catch that *the
implicitly zero elements do not participate*. Specifically, the algorithm is
equivalent to:
(1) Applies `tf.nn.softmax()` to a densified view of each innermost
submatrix with shape `[B, C]`, along the size-C dimension;
(2) Masks out the original implicitly-zero locations;
(3) Renormalizes the remaining elements.
Hence, the `SparseTensor` result has exactly the same non-zero indices and
shape.
Example:
```python
# First batch:
# [? e.]
# [1. ? ]
# Second batch:
# [e ? ]
# [e e ]
shape = [2, 2, 2] # 3-D SparseTensor
values = np.asarray([[[0., np.e], [1., 0.]], [[np.e, 0.], [np.e, np.e]]])
indices = np.vstack(np.where(values)).astype(np.int64).T
result = tf.sparse.softmax(tf.sparse.SparseTensor(indices, values, shape))
# ...returning a 3-D SparseTensor, equivalent to:
# [? 1.] [1 ?]
# [1. ? ] and [.5 .5]
# where ? means implicitly zero.
```
Args:
sp_input: N-D `SparseTensor`, where `N >= 2`.
name: optional name of the operation.
Returns:
output: N-D `SparseTensor` representing the results."
9222,sparse_maximum,tensorflow/tensorflow/python/ops/sparse_ops.py,2640,function,"Returns the element-wise max of two SparseTensors.
Assumes the two SparseTensors have the same shape, i.e., no broadcasting.
Example:
```python
sp_zero = sparse_tensor.SparseTensor([[0]], [0], [7])
sp_one = sparse_tensor.SparseTensor([[1]], [1], [7])
res = tf.sparse.maximum(sp_zero, sp_one).eval()
# ""res"" should be equal to SparseTensor([[0], [1]], [0, 1], [7]).
```
Args:
sp_a: a `SparseTensor` operand whose dtype is real, and indices
lexicographically ordered.
sp_b: the other `SparseTensor` operand with the same requirements (and the
same shape).
name: optional name of the operation.
Returns:
output: the output SparseTensor."
9223,sparse_minimum,tensorflow/tensorflow/python/ops/sparse_ops.py,2678,function,"Returns the element-wise min of two SparseTensors.
Assumes the two SparseTensors have the same shape, i.e., no broadcasting.
Example:
```python
sp_zero = sparse_tensor.SparseTensor([[0]], [0], [7])
sp_one = sparse_tensor.SparseTensor([[1]], [1], [7])
res = tf.sparse.minimum(sp_zero, sp_one).eval()
# ""res"" should be equal to SparseTensor([[0], [1]], [0, 0], [7]).
```
Args:
sp_a: a `SparseTensor` operand whose dtype is real, and indices
lexicographically ordered.
sp_b: the other `SparseTensor` operand with the same requirements (and the
same shape).
name: optional name of the operation.
Returns:
output: the output SparseTensor."
9224,sparse_transpose,tensorflow/tensorflow/python/ops/sparse_ops.py,2716,function,"Transposes a `SparseTensor`
The returned tensor's dimension i will correspond to the input dimension
`perm[i]`. If `perm` is not given, it is set to (n-1...0), where n is
the rank of the input tensor. Hence by default, this operation performs a
regular matrix transpose on 2-D input Tensors.
For example, if `sp_input` has shape `[4, 5]` and `indices` / `values`:
[0, 3]: b
[0, 1]: a
[3, 1]: d
[2, 0]: c
then the output will be a `SparseTensor` of shape `[5, 4]` and
`indices` / `values`:
[0, 2]: c
[1, 0]: a
[1, 3]: d
[3, 0]: b
Args:
sp_input: The input `SparseTensor`.
perm: A permutation of the dimensions of `sp_input`.
name: A name prefix for the returned tensors (optional)
Returns:
A transposed `SparseTensor`.
Raises:
TypeError: If `sp_input` is not a `SparseTensor`."
9225,map_values,tensorflow/tensorflow/python/ops/sparse_ops.py,2778,function,"Applies `op` to the `.values` tensor of one or more `SparseTensor`s.
Replaces any `SparseTensor` in `args` or `kwargs` with its `values`
tensor (which contains the non-default values for the SparseTensor),
and then calls `op`. Returns a `SparseTensor` that is constructed
from the input `SparseTensor`s' `indices`, `dense_shape`, and the
value returned by the `op`.
If the input arguments contain multiple `SparseTensor`s, then they must have
equal `indices` and dense shapes.
Examples:
>>> s = tf.sparse.from_dense([[1, 2, 0],
... [0, 4, 0],
... [1, 0, 0]])
>>> tf.sparse.to_dense(tf.sparse.map_values(tf.ones_like, s)).numpy()
array([[1, 1, 0],
[0, 1, 0],
[1, 0, 0]], dtype=int32)
>>> tf.sparse.to_dense(tf.sparse.map_values(tf.multiply, s, s)).numpy()
array([[ 1, 4, 0],
[ 0, 16, 0],
[ 1, 0, 0]], dtype=int32)
>>> tf.sparse.to_dense(tf.sparse.map_values(tf.add, s, 5)).numpy()
array([[6, 7, 0],
[0, 9, 0],
[6, 0, 0]], dtype=int32)
Note: even though `tf.add(0, 5) != 0`, implicit zeros
will remain unchanged. However, if the sparse tensor contains any explict
zeros, these will be affected by the mapping!
Args:
op: The operation that should be applied to the SparseTensor `values`. `op`
is typically an element-wise operation (such as math_ops.add), but any
operation that preserves the shape can be used.
*args: Arguments for `op`.
**kwargs: Keyword arguments for `op`.
Returns:
A `SparseTensor` whose `indices` and `dense_shape` matches the `indices`
and `dense_shape` of all input `SparseTensor`s.
Raises:
ValueError: If args contains no `SparseTensor`, or if the `indices`
or `dense_shape`s of the input `SparseTensor`s are not equal."
9226,lbeta,tensorflow/tensorflow/python/ops/special_math_ops.py,53,function,"Computes \\(ln(|Beta(x)|)\\), reducing along the last dimension.
Given one-dimensional $z = [z_1,...,z_K]$, we define
$$Beta(z) = \frac{\prod_j \Gamma(z_j)}{\Gamma(\sum_j z_j)},$$
where $\Gamma$ is the gamma function.
And for $n + 1$ dimensional $x$ with shape $[N_1, ..., N_n, K]$, we define
$$lbeta(x)[i_1, ..., i_n] = \log{|Beta(x[i_1, ..., i_n, :])|}.$$
In other words, the last dimension is treated as the $z$ vector.
Note that if $z = [u, v]$, then
$$Beta(z) = \frac{\Gamma(u)\Gamma(v)}{\Gamma(u + v)}
= \int_0^1 t^{u-1} (1 - t)^{v-1} \mathrm{d}t,$$
which defines the traditional bivariate beta function.
If the last dimension is empty, we follow the convention that the sum over
the empty set is zero, and the product is one.
Args:
x: A rank `n + 1` `Tensor`, `n >= 0` with type `float`, or `double`.
name: A name for the operation (optional).
Returns:
The logarithm of \\(|Beta(x)|\\) reducing along the last dimension."
9227,dawsn,tensorflow/tensorflow/python/ops/special_math_ops.py,108,function,"Computes Dawson's integral of `x` element-wise.
Dawson's integral is defined as `exp(-x**2)` times the integral of
`exp(t**2)` from `0` to `x`, with the domain of definition all real numbers.
Dawson's function is odd.
>>> tf.math.special.dawsn([-1., -0.5, 0.5, 1.]).numpy()
array([-0.5380795, -0.4244364, 0.4244364, 0.5380795], dtype=float32)
This implementation is based off of the Cephes math library.
Args:
x: A `Tensor` or `SparseTensor`. Must be one of the following types:
`float32`, `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor`, respectively. Has the same type as `x`.
@compatibility(scipy)
Equivalent to scipy.special.dawsn
@end_compatibility"
9228,expint,tensorflow/tensorflow/python/ops/special_math_ops.py,138,function,"Computes the Exponential integral of `x` element-wise.
The Exponential integral is defined as the integral of `exp(t) / t` from
`-inf` to `x`, with the domain of definition all positive real numbers.
>>> tf.math.special.expint([1., 1.1, 2.1, 4.1]).numpy()
array([ 1.8951179, 2.1673784, 5.3332353, 21.048464], dtype=float32)
This implementation is based off of the Cephes math library.
Args:
x: A `Tensor` or `SparseTensor`. Must be one of the following types:
`float32`, `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor`, respectively. Has the same type as `x`.
@compatibility(scipy)
Equivalent to scipy.special.expi
@end_compatibility"
9229,fresnel_cos,tensorflow/tensorflow/python/ops/special_math_ops.py,167,function,"Computes Fresnel's cosine integral of `x` element-wise.
The Fresnel cosine integral is defined as the integral of `cos(t^2)` from
`0` to `x`, with the domain of definition all real numbers.
The Fresnel cosine integral is odd.
>>> tf.math.special.fresnel_cos([-1., -0.1, 0.1, 1.]).numpy()
array([-0.7798934 , -0.09999753, 0.09999753, 0.7798934 ], dtype=float32)
This implementation is based off of the Cephes math library.
Args:
x: A `Tensor` or `SparseTensor`. Must be one of the following types:
`float32`, `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor`, respectively. Has the same type as `x`.
@compatibility(scipy)
Equivalent to scipy.special.fresnel second output.
@end_compatibility"
9230,fresnel_sin,tensorflow/tensorflow/python/ops/special_math_ops.py,197,function,"Computes Fresnel's sine integral of `x` element-wise.
The Fresnel sine integral is defined as the integral of `sin(t^2)` from
`0` to `x`, with the domain of definition all real numbers.
>>> tf.math.special.fresnel_sin([-1., -0.1, 0.1, 1.]).numpy()
array([-0.43825912, -0.00052359, 0.00052359, 0.43825912], dtype=float32)
This implementation is based off of the Cephes math library.
Args:
x: A `Tensor` or `SparseTensor`. Must be one of the following types:
`float32`, `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor`, respectively. Has the same type as `x`.
@compatibility(scipy)
Equivalent to scipy.special.fresnel first output.
@end_compatibility"
9231,spence,tensorflow/tensorflow/python/ops/special_math_ops.py,226,function,"Computes Spence's integral of `x` element-wise.
Spence's integral is defined as the integral of `log(t) / (1 - t)` from
`1` to `x`, with the domain of definition all non-negative real numbers.
>>> tf.math.special.spence([0.5, 1., 2., 3.]).numpy()
array([ 0.58224034, 0. , -0.82246685, -1.4367464], dtype=float32)
This implementation is based off of the Cephes math library.
Args:
x: A `Tensor` or `SparseTensor`. Must be one of the following types:
`float32`, `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor`, respectively. Has the same type as `x`.
@compatibility(scipy)
Equivalent to scipy.special.spence
@end_compatibility"
9232,bessel_i0,tensorflow/tensorflow/python/ops/special_math_ops.py,255,function,"Computes the Bessel i0 function of `x` element-wise.
Modified Bessel function of order 0.
It is preferable to use the numerically stabler function `i0e(x)` instead.
>>> tf.math.special.bessel_i0([-1., -0.5, 0.5, 1.]).numpy()
array([1.26606588, 1.06348337, 1.06348337, 1.26606588], dtype=float32)
Args:
x: A `Tensor` or `SparseTensor`. Must be one of the following types: `half`,
`float32`, `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor`, respectively. Has the same type as `x`.
@compatibility(scipy)
Equivalent to scipy.special.i0
@end_compatibility"
9233,bessel_i0e,tensorflow/tensorflow/python/ops/special_math_ops.py,283,function,"Computes the Bessel i0e function of `x` element-wise.
Modified Bessel function of order 0.
>>> tf.math.special.bessel_i0e([-1., -0.5, 0.5, 1.]).numpy()
array([0.46575961, 0.64503527, 0.64503527, 0.46575961], dtype=float32)
Args:
x: A `Tensor` or `SparseTensor`. Must be one of the following types: `half`,
`float32`, `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor`, respectively. Has the same type as `x`.
@compatibility(scipy)
Equivalent to scipy.special.i0e
@end_compatibility"
9234,bessel_i1,tensorflow/tensorflow/python/ops/special_math_ops.py,309,function,"Computes the Bessel i1 function of `x` element-wise.
Modified Bessel function of order 1.
It is preferable to use the numerically stabler function `i1e(x)` instead.
>>> tf.math.special.bessel_i1([-1., -0.5, 0.5, 1.]).numpy()
array([-0.5651591 , -0.25789431, 0.25789431, 0.5651591 ], dtype=float32)
Args:
x: A `Tensor` or `SparseTensor`. Must be one of the following types: `half`,
`float32`, `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor`, respectively. Has the same type as `x`.
@compatibility(scipy)
Equivalent to scipy.special.i1
@end_compatibility"
9235,bessel_i1e,tensorflow/tensorflow/python/ops/special_math_ops.py,337,function,"Computes the Bessel i1e function of `x` element-wise.
Modified Bessel function of order 1.
>>> tf.math.special.bessel_i1e([-1., -0.5, 0.5, 1.]).numpy()
array([-0.20791042, -0.15642083, 0.15642083, 0.20791042], dtype=float32)
Args:
x: A `Tensor` or `SparseTensor`. Must be one of the following types: `half`,
`float32`, `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor`, respectively. Has the same type as `x`.
@compatibility(scipy)
Equivalent to scipy.special.i1e
@end_compatibility"
9236,bessel_k0,tensorflow/tensorflow/python/ops/special_math_ops.py,363,function,"Computes the Bessel k0 function of `x` element-wise.
Modified Bessel function of order 0.
It is preferable to use the numerically stabler function `k0e(x)` instead.
>>> tf.math.special.bessel_k0([0.5, 1., 2., 4.]).numpy()
array([0.92441907, 0.42102444, 0.11389387, 0.01115968], dtype=float32)
Args:
x: A `Tensor` or `SparseTensor`. Must be one of the following types: `half`,
`float32`, `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor`, respectively. Has the same type as `x`.
@compatibility(scipy)
Equivalent to scipy.special.k0
@end_compatibility"
9237,bessel_k0e,tensorflow/tensorflow/python/ops/special_math_ops.py,391,function,"Computes the Bessel k0e function of `x` element-wise.
Modified Bessel function of order 0.
>>> tf.math.special.bessel_k0e([0.5, 1., 2., 4.]).numpy()
array([1.52410939, 1.14446308, 0.84156822, 0.60929767], dtype=float32)
Args:
x: A `Tensor` or `SparseTensor`. Must be one of the following types: `half`,
`float32`, `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor`, respectively. Has the same type as `x`.
@compatibility(scipy)
Equivalent to scipy.special.k0e
@end_compatibility"
9238,bessel_k1,tensorflow/tensorflow/python/ops/special_math_ops.py,417,function,"Computes the Bessel k1 function of `x` element-wise.
Modified Bessel function of order 1.
It is preferable to use the numerically stabler function `k1e(x)` instead.
>>> tf.math.special.bessel_k1([0.5, 1., 2., 4.]).numpy()
array([1.65644112, 0.60190723, 0.13986588, 0.0124835 ], dtype=float32)
Args:
x: A `Tensor` or `SparseTensor`. Must be one of the following types: `half`,
`float32`, `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor`, respectively. Has the same type as `x`.
@compatibility(scipy)
Equivalent to scipy.special.k1
@end_compatibility"
9239,bessel_k1e,tensorflow/tensorflow/python/ops/special_math_ops.py,445,function,"Computes the Bessel k1e function of `x` element-wise.
Modified Bessel function of order 1.
>>> tf.math.special.bessel_k1e([0.5, 1., 2., 4.]).numpy()
array([2.73100971, 1.63615349, 1.03347685, 0.68157595], dtype=float32)
Args:
x: A `Tensor` or `SparseTensor`. Must be one of the following types: `half`,
`float32`, `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor`, respectively. Has the same type as `x`.
@compatibility(scipy)
Equivalent to scipy.special.k1e
@end_compatibility"
9240,bessel_j0,tensorflow/tensorflow/python/ops/special_math_ops.py,471,function,"Computes the Bessel j0 function of `x` element-wise.
Modified Bessel function of order 0.
>>> tf.math.special.bessel_j0([0.5, 1., 2., 4.]).numpy()
array([ 0.93846981, 0.76519769, 0.22389078, -0.39714981], dtype=float32)
Args:
x: A `Tensor` or `SparseTensor`. Must be one of the following types: `half`,
`float32`, `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor`, respectively. Has the same type as `x`.
@compatibility(scipy)
Equivalent to scipy.special.j0
@end_compatibility"
9241,bessel_j1,tensorflow/tensorflow/python/ops/special_math_ops.py,497,function,"Computes the Bessel j1 function of `x` element-wise.
Modified Bessel function of order 1.
>>> tf.math.special.bessel_j1([0.5, 1., 2., 4.]).numpy()
array([ 0.24226846, 0.44005059, 0.57672481, -0.06604333], dtype=float32)
Args:
x: A `Tensor` or `SparseTensor`. Must be one of the following types: `half`,
`float32`, `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor`, respectively. Has the same type as `x`.
@compatibility(scipy)
Equivalent to scipy.special.j1
@end_compatibility"
9242,bessel_y0,tensorflow/tensorflow/python/ops/special_math_ops.py,523,function,"Computes the Bessel y0 function of `x` element-wise.
Modified Bessel function of order 0.
>>> tf.math.special.bessel_y0([0.5, 1., 2., 4.]).numpy()
array([-0.44451873, 0.08825696, 0.51037567, -0.01694074], dtype=float32)
Args:
x: A `Tensor` or `SparseTensor`. Must be one of the following types: `half`,
`float32`, `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor`, respectively. Has the same type as `x`.
@compatibility(scipy)
Equivalent to scipy.special.y0
@end_compatibility"
9243,bessel_y1,tensorflow/tensorflow/python/ops/special_math_ops.py,549,function,"Computes the Bessel y1 function of `x` element-wise.
Modified Bessel function of order 1.
>>> tf.math.special.bessel_y1([0.5, 1., 2., 4.]).numpy()
array([-1.47147239, -0.78121282, -0.10703243, 0.39792571], dtype=float32)
Args:
x: A `Tensor` or `SparseTensor`. Must be one of the following types: `half`,
`float32`, `float64`.
name: A name for the operation (optional).
Returns:
A `Tensor` or `SparseTensor`, respectively. Has the same type as `x`.
@compatibility(scipy)
Equivalent to scipy.special.y1
@end_compatibility"
9244,einsum,tensorflow/tensorflow/python/ops/special_math_ops.py,608,function,"Tensor contraction over specified indices and outer product.
Einsum allows defining Tensors by defining their element-wise computation.
This computation is defined by `equation`, a shorthand form based on Einstein
summation. As an example, consider multiplying two matrices A and B to form a
matrix C. The elements of C are given by:
```
C[i,k] = sum_j A[i,j] * B[j,k]
```
The corresponding `equation` is:
```
ij,jk->ik
```
In general, to convert the element-wise equation into the `equation` string,
use the following procedure (intermediate strings for matrix multiplication
example provided in parentheses):
1. remove variable names, brackets, and commas, (`ik = sum_j ij * jk`)
2. replace ""*"" with "","", (`ik = sum_j ij , jk`)
3. drop summation signs, and (`ik = ij, jk`)
4. move the output to the right, while replacing ""="" with ""->"". (`ij,jk->ik`)
Many common operations can be expressed in this way. For example:
```python
# Matrix multiplication
einsum('ij,jk->ik', m0, m1) # output[i,k] = sum_j m0[i,j] * m1[j, k]
# Dot product
einsum('i,i->', u, v) # output = sum_i u[i]*v[i]
# Outer product
einsum('i,j->ij', u, v) # output[i,j] = u[i]*v[j]
# Transpose
einsum('ij->ji', m) # output[j,i] = m[i,j]
# Trace
einsum('ii', m) # output[j,i] = trace(m) = sum_i m[i, i]
# Batch matrix multiplication
einsum('aij,ajk->aik', s, t) # out[a,i,k] = sum_j s[a,i,j] * t[a, j, k]
```
To enable and control broadcasting, use an ellipsis. For example, to perform
batch matrix multiplication with NumPy-style broadcasting across the batch
dimensions, use:
```python
einsum('...ij,...jk->...ik', u, v)
```
Args:
equation: a `str` describing the contraction, in the same format as
`numpy.einsum`.
*inputs: the inputs to contract (each one a `Tensor`), whose shapes should
be consistent with `equation`.
**kwargs:
- optimize: Optimization strategy to use to find contraction path using
opt_einsum. Must be 'greedy', 'optimal', 'branch-2', 'branch-all' or
'auto'. (optional, default: 'greedy').
- name: A name for the operation (optional).
Returns:
The contracted `Tensor`, with shape determined by `equation`.
Raises:
ValueError: If
- the format of `equation` is incorrect,
- number of inputs or their shapes are inconsistent with `equation`."
9245,EinsumBenchmark,tensorflow/tensorflow/python/ops/special_math_ops_test.py,1077,class,
9246,benchmark_einsum,tensorflow/tensorflow/python/ops/special_math_ops_test.py,1106,method,
9247,build_graph,tensorflow/tensorflow/python/ops/split_benchmark.py,34,function,"Build a graph containing a sequence of split operations.
Args:
device: string, the device to run on.
input_shape: shape of the input tensor.
output_sizes: size of each output along axis.
axis: axis to be split along.
Returns:
An array of tensors to run()"
9248,SplitBenchmark,tensorflow/tensorflow/python/ops/split_benchmark.py,55,class,Benchmark split!
9249,benchmark_split,tensorflow/tensorflow/python/ops/split_benchmark.py,109,method,
9250,variable_op,tensorflow/tensorflow/python/ops/state_ops.py,41,function,Deprecated. Used variable_op_v2 instead.
9251,variable_op_v2,tensorflow/tensorflow/python/ops/state_ops.py,55,function,"Create a variable Operation.
See also variables.Variable.
Args:
shape: The shape of the tensor managed by this variable
dtype: The underlying type of the tensor values.
name: optional name to use for the variable op.
container: An optional string. Defaults to """".
If non-empty, this variable is placed in the given container.
Otherwise, a default container is used.
shared_name: An optional string. Defaults to """".
If non-empty, this variable is named in the given bucket
with this shared_name. Otherwise, the node name is used instead.
Returns:
A variable tensor."
9252,init_variable,tensorflow/tensorflow/python/ops/state_ops.py,82,function,"Initializes variable with ""init"".
This op does the following:
if init is a Tensor, v = init
if callable(init): v = init(VariableShape(v), v.dtype)
Args:
v: Variable to initialize
init: Tensor to assign to v,
Or an object convertible to Tensor e.g. nparray,
Or an Initializer that generates a tensor given the shape and type of v.
An ""Initializer"" is a callable that returns a tensor that ""v"" should be
set to. It will be called as init(shape, dtype).
name: Optional name for the op.
Returns:
The operation that initializes v."
9253,is_variable_initialized,tensorflow/tensorflow/python/ops/state_ops.py,117,function,"Checks whether a tensor has been initialized.
Outputs boolean scalar indicating whether the tensor has been initialized.
Args:
ref: A mutable `Tensor`.
Should be from a `Variable` node. May be uninitialized.
name: A name for the operation (optional).
Returns:
A `Tensor` of type `bool`."
9254,assign_sub,tensorflow/tensorflow/python/ops/state_ops.py,137,function,"Update `ref` by subtracting `value` from it.
This operation outputs `ref` after the update is done.
This makes it easier to chain operations that need to use the reset value.
Unlike `tf.math.subtract`, this op does not broadcast. `ref` and `value`
must have the same shape.
Args:
ref: A mutable `Tensor`. Must be one of the following types: `float32`,
`float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`,
`complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Should be
from a `Variable` node.
value: A `Tensor`. Must have the same shape and dtype as `ref`. The value to
be subtracted to the variable.
use_locking: An optional `bool`. Defaults to `False`. If True, the
subtraction will be protected by a lock; otherwise the behavior is
undefined, but may exhibit less contention.
name: A name for the operation (optional).
Returns:
Same as ""ref"". Returned as a convenience for operations that want
to use the new value after the variable has been updated."
9255,assign_add,tensorflow/tensorflow/python/ops/state_ops.py,168,function,"Update `ref` by adding `value` to it.
This operation outputs ""ref"" after the update is done.
This makes it easier to chain operations that need to use the reset value.
Unlike `tf.math.add`, this op does not broadcast. `ref` and `value` must have
the same shape.
Args:
ref: A mutable `Tensor`. Must be one of the following types: `float32`,
`float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`,
`complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. Should be
from a `Variable` node.
value: A `Tensor`. Must have the same shape and dtype as `ref`. The value to
be added to the variable.
use_locking: An optional `bool`. Defaults to `False`. If True, the addition
will be protected by a lock; otherwise the behavior is undefined, but may
exhibit less contention.
name: A name for the operation (optional).
Returns:
Same as ""ref"". Returned as a convenience for operations that want
to use the new value after the variable has been updated."
9256,assign,tensorflow/tensorflow/python/ops/state_ops.py,199,function,"Update `ref` by assigning `value` to it.
This operation outputs a Tensor that holds the new value of `ref` after
the value has been assigned. This makes it easier to chain operations that
need to use the reset value.
Args:
ref: A mutable `Tensor`. Should be from a `Variable` node. May be
uninitialized.
value: A `Tensor`. Must have the same shape and dtype as `ref`. The value to
be assigned to the variable.
validate_shape: An optional `bool`. Defaults to `True`. If true, the
operation will validate that the shape of 'value' matches the shape of the
Tensor being assigned to. If false, 'ref' will take on the shape of
'value'.
use_locking: An optional `bool`. Defaults to `True`. If True, the assignment
will be protected by a lock; otherwise the behavior is undefined, but may
exhibit less contention.
name: A name for the operation (optional).
Returns:
A `Tensor` that will hold the new value of `ref` after
the assignment has completed."
9257,count_up_to,tensorflow/tensorflow/python/ops/state_ops.py,233,function,"Increments 'ref' until it reaches 'limit'.
Args:
ref: A Variable. Must be one of the following types: `int32`, `int64`.
Should be from a scalar `Variable` node.
limit: An `int`.
If incrementing ref would bring it above limit, instead generates an
'OutOfRange' error.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `ref`.
A copy of the input before increment. If nothing else modifies the
input, the values produced will all be distinct."
9258,scatter_update,tensorflow/tensorflow/python/ops/state_ops.py,256,function,"Applies sparse updates to a variable reference.
This operation computes
```python
# Scalar indices
ref[indices, ...] = updates[...]
# Vector indices (for each i)
ref[indices[i], ...] = updates[i, ...]
# High rank indices (for each i, ..., j)
ref[indices[i, ..., j], ...] = updates[i, ..., j, ...]
```
This operation outputs `ref` after the update is done.
This makes it easier to chain operations that need to use the reset value.
If values in `ref` is to be updated more than once, because there are
duplicate entries in `indices`, the order at which the updates happen
for each value is undefined.
Requires `updates.shape = indices.shape + ref.shape[1:]`.
<div style=""width:70%; margin:auto; margin-bottom:10px; margin-top:20px;"">
<img style=""width:100%"" src=""https://www.tensorflow.org/images/ScatterUpdate.png"" alt>
</div>
Args:
ref: A `Variable`.
indices: A `Tensor`. Must be one of the following types: `int32`, `int64`.
A tensor of indices into the first dimension of `ref`.
updates: A `Tensor`. Must have the same type as `ref`.
A tensor of updated values to store in `ref`.
use_locking: An optional `bool`. Defaults to `True`.
If True, the assignment will be protected by a lock;
otherwise the behavior is undefined, but may exhibit less contention.
name: A name for the operation (optional).
Returns:
Same as `ref`. Returned as a convenience for operations that want
to use the updated values after the update is done."
9259,scatter_nd_update,tensorflow/tensorflow/python/ops/state_ops.py,310,function,"Applies sparse `updates` to individual values or slices in a Variable.
`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`.
`indices` must be integer tensor, containing indices into `ref`.
It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`.
The innermost dimension of `indices` (with length `K`) corresponds to
indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th
dimension of `ref`.
`updates` is `Tensor` of rank `Q-1+P-K` with shape:
```
[d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]].
```
For example, say we want to update 4 scattered elements to a rank-1 tensor to
8 elements. In Python, that update would look like this:
```python
ref = tf.Variable([1, 2, 3, 4, 5, 6, 7, 8])
indices = tf.constant([[4], [3], [1] ,[7]])
updates = tf.constant([9, 10, 11, 12])
update = tf.compat.v1.scatter_nd_update(ref, indices, updates)
with tf.compat.v1.Session() as sess:
print sess.run(update)
```
The resulting update to ref would look like this:
[1, 11, 3, 10, 9, 6, 7, 12]
See `tf.scatter_nd` for more details about how to make updates to
slices.
Args:
ref: A Variable.
indices: A `Tensor`. Must be one of the following types: `int32`, `int64`.
A tensor of indices into ref.
updates: A `Tensor`. Must have the same type as `ref`.
A Tensor. Must have the same type as ref. A tensor of updated
values to add to ref.
use_locking: An optional `bool`. Defaults to `True`.
An optional bool. Defaults to True. If True, the assignment will
be protected by a lock; otherwise the behavior is undefined,
but may exhibit less contention.
name: A name for the operation (optional).
Returns:
The value of the variable after the update."
9260,scatter_add,tensorflow/tensorflow/python/ops/state_ops.py,372,function,"Adds sparse updates to the variable referenced by `resource`.
This operation computes
```python
# Scalar indices
ref[indices, ...] += updates[...]
# Vector indices (for each i)
ref[indices[i], ...] += updates[i, ...]
# High rank indices (for each i, ..., j)
ref[indices[i, ..., j], ...] += updates[i, ..., j, ...]
```
This operation outputs `ref` after the update is done.
This makes it easier to chain operations that need to use the updated value.
Duplicate entries are handled correctly: if multiple `indices` reference
the same location, their contributions add.
Requires `updates.shape = indices.shape + ref.shape[1:]`.
<div style=""width:70%; margin:auto; margin-bottom:10px; margin-top:20px;"">
<img style=""width:100%"" src='https://www.tensorflow.org/images/ScatterAdd.png' alt>
</div>
Args:
ref: A `Variable`.
indices: A `Tensor`. Must be one of the following types: `int32`, `int64`.
A tensor of indices into the first dimension of `ref`.
updates: A `Tensor`. Must have the same type as `ref`.
A tensor of updated values to store in `ref`.
use_locking: An optional `bool`. Defaults to `False`.
If True, the assignment will be protected by a lock;
otherwise the behavior is undefined, but may exhibit less contention.
name: A name for the operation (optional).
Returns:
Same as `ref`. Returned as a convenience for operations that want
to use the updated values after the update is done."
9261,scatter_nd_add,tensorflow/tensorflow/python/ops/state_ops.py,424,function,"Applies sparse addition to individual values or slices in a Variable.
`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`.
`indices` must be integer tensor, containing indices into `ref`.
It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`.
The innermost dimension of `indices` (with length `K`) corresponds to
indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th
dimension of `ref`.
`updates` is `Tensor` of rank `Q-1+P-K` with shape:
```
[d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]]
```
For example, say we want to add 4 scattered elements to a rank-1 tensor to
8 elements. In Python, that addition would look like this:
```python
ref = tf.Variable([1, 2, 3, 4, 5, 6, 7, 8])
indices = tf.constant([[4], [3], [1], [7]])
updates = tf.constant([9, 10, 11, 12])
add = tf.compat.v1.scatter_nd_add(ref, indices, updates)
with tf.compat.v1.Session() as sess:
print sess.run(add)
```
The resulting update to ref would look like this:
[1, 13, 3, 14, 14, 6, 7, 20]
See `tf.scatter_nd` for more details about how to make updates to
slices.
Args:
ref: A mutable `Tensor`. Must be one of the following types: `float32`,
`float64`, `int32`, `uint8`, `int16`, `int8`, `complex64`, `int64`,
`qint8`, `quint8`, `qint32`, `bfloat16`, `uint16`, `complex128`, `half`,
`uint32`, `uint64`. A mutable Tensor. Should be from a Variable node.
indices: A `Tensor`. Must be one of the following types: `int32`, `int64`.
A tensor of indices into ref.
updates: A `Tensor`. Must have the same type as `ref`.
A tensor of updated values to add to ref.
use_locking: An optional `bool`. Defaults to `False`.
If True, the assignment will be protected by a lock;
otherwise the behavior is undefined, but may exhibit less contention.
name: A name for the operation (optional).
Returns:
A mutable `Tensor`. Has the same type as `ref`."
9262,scatter_sub,tensorflow/tensorflow/python/ops/state_ops.py,487,function,"Subtracts sparse updates to a variable reference.
```python
# Scalar indices
ref[indices, ...] -= updates[...]
# Vector indices (for each i)
ref[indices[i], ...] -= updates[i, ...]
# High rank indices (for each i, ..., j)
ref[indices[i, ..., j], ...] -= updates[i, ..., j, ...]
```
This operation outputs `ref` after the update is done.
This makes it easier to chain operations that need to use the reset value.
Duplicate entries are handled correctly: if multiple `indices` reference
the same location, their (negated) contributions add.
Requires `updates.shape = indices.shape + ref.shape[1:]` or
`updates.shape = []`.
<div style=""width:70%; margin:auto; margin-bottom:10px; margin-top:20px;"">
<img style=""width:100%""
src=""https://www.tensorflow.org/images/ScatterSub.png"" alt>
</div>
Args:
ref: A mutable `Tensor`. Must be one of the following types: `float32`,
`float64`, `int32`, `uint8`, `int16`, `int8`, `complex64`, `int64`,
`qint8`, `quint8`, `qint32`, `bfloat16`, `uint16`, `complex128`, `half`,
`uint32`, `uint64`. Should be from a `Variable` node.
indices: A `Tensor`. Must be one of the following types: `int32`, `int64`.
A tensor of indices into the first dimension of `ref`.
updates: A `Tensor`. Must have the same type as `ref`.
A tensor of updated values to subtract from `ref`.
use_locking: An optional `bool`. Defaults to `False`.
If True, the subtraction will be protected by a lock;
otherwise the behavior is undefined, but may exhibit less contention.
name: A name for the operation (optional).
Returns:
A mutable `Tensor`. Has the same type as `ref`."
9263,scatter_nd_sub,tensorflow/tensorflow/python/ops/state_ops.py,541,function,"Applies sparse subtraction to individual values or slices in a Variable.
`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`.
`indices` must be integer tensor, containing indices into `ref`.
It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`.
The innermost dimension of `indices` (with length `K`) corresponds to
indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th
dimension of `ref`.
`updates` is `Tensor` of rank `Q-1+P-K` with shape:
```
[d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]]
```
For example, say we want to subtract 4 scattered elements from a rank-1 tensor
with 8 elements. In Python, that update would look like this:
```python
ref = tf.Variable([1, 2, 3, 4, 5, 6, 7, 8])
indices = tf.constant([[4], [3], [1] ,[7]])
updates = tf.constant([9, 10, 11, 12])
op = tf.compat.v1.scatter_nd_sub(ref, indices, updates)
with tf.compat.v1.Session() as sess:
print sess.run(op)
```
The resulting update to ref would look like this:
[1, -9, 3, -6, -6, 6, 7, -4]
See `tf.scatter_nd` for more details about how to make updates to
slices.
Args:
ref: A mutable `Tensor`. Must be one of the following types: `float32`,
`float64`, `int32`, `uint8`, `int16`, `int8`, `complex64`, `int64`,
`qint8`, `quint8`, `qint32`, `bfloat16`, `uint16`, `complex128`, `half`,
`uint32`, `uint64`. A mutable Tensor. Should be from a Variable node.
indices: A `Tensor`. Must be one of the following types: `int32`, `int64`.
A tensor of indices into ref.
updates: A `Tensor`. Must have the same type as `ref`.
A tensor of updated values to add to ref.
use_locking: An optional `bool`. Defaults to `False`.
An optional bool. Defaults to True. If True, the assignment will
be protected by a lock; otherwise the behavior is undefined,
but may exhibit less contention.
name: A name for the operation (optional).
Returns:
A mutable `Tensor`. Has the same type as `ref`."
9264,scatter_mul,tensorflow/tensorflow/python/ops/state_ops.py,605,function,"Multiplies sparse updates into a variable reference.
This operation computes
```python
# Scalar indices
ref[indices, ...] *= updates[...]
# Vector indices (for each i)
ref[indices[i], ...] *= updates[i, ...]
# High rank indices (for each i, ..., j)
ref[indices[i, ..., j], ...] *= updates[i, ..., j, ...]
```
This operation outputs `ref` after the update is done.
This makes it easier to chain operations that need to use the reset value.
Duplicate entries are handled correctly: if multiple `indices` reference
the same location, their contributions multiply.
Requires `updates.shape = indices.shape + ref.shape[1:]` or `updates.shape =
[]`.
Args:
ref: A mutable `Tensor`. Must be one of the following types: `float32`,
`float64`, `int32`, `uint8`, `int16`, `int8`, `complex64`, `int64`,
`qint8`, `quint8`, `qint32`, `bfloat16`, `uint16`, `complex128`, `half`,
`uint32`, `uint64`. Should be from a `Variable` node.
indices: A `Tensor`. Must be one of the following types: `int32`, `int64`. A
tensor of indices into the first dimension of `ref`.
updates: A `Tensor`. Must have the same type as `ref`. A tensor of updated
values to multiply to `ref`.
use_locking: An optional `bool`. Defaults to `False`. If True, the operation
will be protected by a lock; otherwise the behavior is undefined, but may
exhibit less contention.
name: A name for the operation (optional).
Returns:
A mutable `Tensor`. Has the same type as `ref`."
9265,scatter_div,tensorflow/tensorflow/python/ops/state_ops.py,657,function,"Divides a variable reference by sparse updates.
This operation computes
```python
# Scalar indices
ref[indices, ...] /= updates[...]
# Vector indices (for each i)
ref[indices[i], ...] /= updates[i, ...]
# High rank indices (for each i, ..., j)
ref[indices[i, ..., j], ...] /= updates[i, ..., j, ...]
```
This operation outputs `ref` after the update is done.
This makes it easier to chain operations that need to use the reset value.
Duplicate entries are handled correctly: if multiple `indices` reference
the same location, their contributions divide.
Requires `updates.shape = indices.shape + ref.shape[1:]` or `updates.shape =
[]`.
Args:
ref: A mutable `Tensor`. Must be one of the following types: `float32`,
`float64`, `int32`, `uint8`, `int16`, `int8`, `complex64`, `int64`,
`qint8`, `quint8`, `qint32`, `bfloat16`, `uint16`, `complex128`, `half`,
`uint32`, `uint64`. Should be from a `Variable` node.
indices: A `Tensor`. Must be one of the following types: `int32`, `int64`. A
tensor of indices into the first dimension of `ref`.
updates: A `Tensor`. Must have the same type as `ref`. A tensor of values
that `ref` is divided by.
use_locking: An optional `bool`. Defaults to `False`. If True, the operation
will be protected by a lock; otherwise the behavior is undefined, but may
exhibit less contention.
name: A name for the operation (optional).
Returns:
A mutable `Tensor`. Has the same type as `ref`."
9266,scatter_max,tensorflow/tensorflow/python/ops/state_ops.py,709,function,"Reduces sparse updates into a variable reference using the `max` operation.
This operation computes
# Scalar indices
ref[indices, ...] = max(ref[indices, ...], updates[...])
# Vector indices (for each i)
ref[indices[i], ...] = max(ref[indices[i], ...], updates[i, ...])
# High rank indices (for each i, ..., j)
ref[indices[i, ..., j], ...] = max(ref[indices[i, ..., j], ...],
updates[i, ..., j, ...])
This operation outputs `ref` after the update is done.
This makes it easier to chain operations that need to use the reset value.
Duplicate entries are handled correctly: if multiple `indices` reference
the same location, their contributions combine.
Requires `updates.shape = indices.shape + ref.shape[1:]` or `updates.shape =
[]`.
<div style=""width:70%; margin:auto; margin-bottom:10px; margin-top:20px;"">
<img style=""width:100%"" src=""https://www.tensorflow.org/images/ScatterAdd.png""
alt>
</div>
Args:
ref: A mutable `Tensor`. Must be one of the following types: `half`,
`bfloat16`, `float32`, `float64`, `int32`, `int64`. Should be from a
`Variable` node.
indices: A `Tensor`. Must be one of the following types: `int32`, `int64`. A
tensor of indices into the first dimension of `ref`.
updates: A `Tensor`. Must have the same type as `ref`. A tensor of updated
values to reduce into `ref`.
use_locking: An optional `bool`. Defaults to `False`. If True, the update
will be protected by a lock; otherwise the behavior is undefined, but may
exhibit less contention.
name: A name for the operation (optional).
Returns:
A mutable `Tensor`. Has the same type as `ref`."
9267,scatter_min,tensorflow/tensorflow/python/ops/state_ops.py,764,function,"Reduces sparse updates into a variable reference using the `min` operation.
This operation computes
# Scalar indices
ref[indices, ...] = min(ref[indices, ...], updates[...])
# Vector indices (for each i)
ref[indices[i], ...] = min(ref[indices[i], ...], updates[i, ...])
# High rank indices (for each i, ..., j)
ref[indices[i, ..., j], ...] = min(ref[indices[i, ..., j], ...],
updates[i, ..., j, ...])
This operation outputs `ref` after the update is done.
This makes it easier to chain operations that need to use the reset value.
Duplicate entries are handled correctly: if multiple `indices` reference
the same location, their contributions combine.
Requires `updates.shape = indices.shape + ref.shape[1:]` or `updates.shape =
[]`.
<div style=""width:70%; margin:auto; margin-bottom:10px; margin-top:20px;"">
<img style=""width:100%"" src=""https://www.tensorflow.org/images/ScatterAdd.png""
alt>
</div>
Args:
ref: A mutable `Tensor`. Must be one of the following types: `half`,
`bfloat16`, `float32`, `float64`, `int32`, `int64`. Should be from a
`Variable` node.
indices: A `Tensor`. Must be one of the following types: `int32`, `int64`. A
tensor of indices into the first dimension of `ref`.
updates: A `Tensor`. Must have the same type as `ref`. A tensor of updated
values to reduce into `ref`.
use_locking: An optional `bool`. Defaults to `False`. If True, the update
will be protected by a lock; otherwise the behavior is undefined, but may
exhibit less contention.
name: A name for the operation (optional).
Returns:
A mutable `Tensor`. Has the same type as `ref`."
9268,batch_scatter_update,tensorflow/tensorflow/python/ops/state_ops.py,821,function,"Generalization of `tf.compat.v1.scatter_update` to axis different than 0.
Analogous to `batch_gather`. This assumes that `ref`, `indices` and `updates`
have a series of leading dimensions that are the same for all of them, and the
updates are performed on the last dimension of indices. In other words, the
dimensions should be the following:
`num_prefix_dims = indices.ndims - 1`
`batch_dim = num_prefix_dims + 1`
`updates.shape = indices.shape + var.shape[batch_dim:]`
where
`updates.shape[:num_prefix_dims]`
`== indices.shape[:num_prefix_dims]`
`== var.shape[:num_prefix_dims]`
And the operation performed can be expressed as:
`var[i_1, ..., i_n, indices[i_1, ..., i_n, j]] = updates[i_1, ..., i_n, j]`
When indices is a 1D tensor, this operation is equivalent to
`tf.compat.v1.scatter_update`.
To avoid this operation there would be 2 alternatives:
1) Reshaping the variable by merging the first `ndims` dimensions. However,
this is not possible because `tf.reshape` returns a Tensor, which we
cannot use `tf.compat.v1.scatter_update` on.
2) Looping over the first `ndims` of the variable and using
`tf.compat.v1.scatter_update` on the subtensors that result of slicing the
first
dimension. This is a valid option for `ndims = 1`, but less efficient than
this implementation.
See also `tf.compat.v1.scatter_update` and `tf.compat.v1.scatter_nd_update`.
Args:
ref: `Variable` to scatter onto.
indices: Tensor containing indices as described above.
updates: Tensor of updates to apply to `ref`.
use_locking: Boolean indicating whether to lock the writing operation.
name: Optional scope name string.
Returns:
Ref to `variable` after it has been modified.
Raises:
ValueError: If the initial `ndims` of `ref`, `indices`, and `updates` are
not the same."
9269,Algorithm,tensorflow/tensorflow/python/ops/stateful_random_ops.py,67,class,
9270,non_deterministic_ints,tensorflow/tensorflow/python/ops/stateful_random_ops.py,77,function,"Non-deterministically generates some integers.
This op may use some OS-provided source of non-determinism (e.g. an RNG), so
each execution will give different results.
Args:
shape: the shape of the result.
dtype: (optional) the dtype of the result.
Returns:
a tensor whose element values are non-deterministically chosen."
9271,create_rng_state,tensorflow/tensorflow/python/ops/stateful_random_ops.py,187,function,"Creates a RNG state from an integer or a vector.
Example:
>>> tf.random.create_rng_state(
... 1234, ""philox"")
array([1234, 0, 0])
>>> tf.random.create_rng_state(
... [12, 34], ""threefry"")
array([12, 34])
Args:
seed: an integer or 1-D numpy array.
alg: the RNG algorithm. Can be a string, an `Algorithm` or an integer.
Returns:
a 1-D numpy array whose size depends on the algorithm."
9272,GeneratorSpec,tensorflow/tensorflow/python/ops/stateful_random_ops.py,226,class,TypeSpec for Generator.
9273,value_type,tensorflow/tensorflow/python/ops/stateful_random_ops.py,251,method,
9274,Generator,tensorflow/tensorflow/python/ops/stateful_random_ops.py,259,class,"Random-number generator.
Example:
Creating a generator from a seed:
>>> g = tf.random.Generator.from_seed(1234)
>>> g.normal(shape=(2, 3))
<tf.Tensor: shape=(2, 3), dtype=float32, numpy=
array([[ 0.9356609 , 1.0854305 , -0.93788373],
[-0.5061547 , 1.3169702 , 0.7137579 ]], dtype=float32)>
Creating a generator from a non-deterministic state:
>>> g = tf.random.Generator.from_non_deterministic_state()
>>> g.normal(shape=(2, 3))
<tf.Tensor: shape=(2, 3), dtype=float32, numpy=...>
All the constructors allow explicitly choosing an Random-Number-Generation
(RNG) algorithm. Supported algorithms are `""philox""` and `""threefry""`. For
example:
>>> g = tf.random.Generator.from_seed(123, alg=""philox"")
>>> g.normal(shape=(2, 3))
<tf.Tensor: shape=(2, 3), dtype=float32, numpy=
array([[ 0.8673864 , -0.29899067, -0.9310337 ],
[-1.5828488 , 1.2481191 , -0.6770643 ]], dtype=float32)>
CPU, GPU and TPU with the same algorithm and seed will generate the same
integer random numbers. Float-point results (such as the output of `normal`)
may have small numerical discrepancies between different devices.
This class uses a `tf.Variable` to manage its internal state. Every time
random numbers are generated, the state of the generator will change. For
example:
>>> g = tf.random.Generator.from_seed(1234)
>>> g.state
<tf.Variable ... numpy=array([1234, 0, 0])>
>>> g.normal(shape=(2, 3))
<...>
>>> g.state
<tf.Variable ... numpy=array([2770, 0, 0])>
The shape of the state is algorithm-specific.
There is also a global generator:
>>> g = tf.random.get_global_generator()
>>> g.normal(shape=(2, 3))
<tf.Tensor: shape=(2, 3), dtype=float32, numpy=...>"
9275,from_state,tensorflow/tensorflow/python/ops/stateful_random_ops.py,387,method,"Creates a generator from a state.
See `__init__` for description of `state` and `alg`.
Args:
state: the new state.
alg: the RNG algorithm.
Returns:
The new generator.
Throws:
ValueError: if the generator is created inside a synchronous
`tf.distribute` strategy such as `MirroredStrategy` or `TPUStrategy`,
because there is ambiguity on how to replicate a generator (e.g. should
it be copied so such each replica will get the same random numbers, or
should it be ""split"" into different generators that generate
different random numbers)."
9276,from_seed,tensorflow/tensorflow/python/ops/stateful_random_ops.py,410,method,"Creates a generator from a seed.
A seed is a 1024-bit unsigned integer represented either as a Python
integer or a vector of integers. Seeds shorter than 1024-bit will be
padded. The padding, the internal structure of a seed and the way a seed
is converted to a state are all opaque (unspecified). The only semantics
specification of seeds is that two different seeds are likely to produce
two independent generators (but no guarantee).
Args:
seed: the seed for the RNG.
alg: (optional) the RNG algorithm. If None, it will be auto-selected. See
`__init__` for its possible values.
Returns:
The new generator.
Throws:
ValueError: if the generator is created inside a synchronous
`tf.distribute` strategy such as `MirroredStrategy` or `TPUStrategy`,
because there is ambiguity on how to replicate a generator (e.g. should
it be copied so such each replica will get the same random numbers, or
should it be ""split"" into different generators that generate
different random numbers)."
9277,from_non_deterministic_state,tensorflow/tensorflow/python/ops/stateful_random_ops.py,444,method,"Creates a generator by non-deterministically initializing its state.
The source of the non-determinism will be platform- and time-dependent.
Args:
alg: (optional) the RNG algorithm. If None, it will be auto-selected. See
`__init__` for its possible values.
Returns:
The new generator.
Throws:
ValueError: if the generator is created inside a synchronous
`tf.distribute` strategy such as `MirroredStrategy` or `TPUStrategy`,
because there is ambiguity on how to replicate a generator (e.g. should
it be copied so such each replica will get the same random numbers, or
should it be ""split"" into different generators that generate
different random numbers)."
9278,from_key_counter,tensorflow/tensorflow/python/ops/stateful_random_ops.py,473,method,"Creates a generator from a key and a counter.
This constructor only applies if the algorithm is a counter-based algorithm.
See method `key` for the meaning of ""key"" and ""counter"".
Args:
key: the key for the RNG, a scalar of type STATE_TYPE.
counter: a vector of dtype STATE_TYPE representing the initial counter for
the RNG, whose length is algorithm-specific.,
alg: the RNG algorithm. If None, it will be auto-selected. See
`__init__` for its possible values.
Returns:
The new generator.
Throws:
ValueError: if the generator is created inside a synchronous
`tf.distribute` strategy such as `MirroredStrategy` or `TPUStrategy`,
because there is ambiguity on how to replicate a generator (e.g. should
it be copied so such each replica will get the same random numbers, or
should it be ""split"" into different generators that generate
different random numbers)."
9279,reset,tensorflow/tensorflow/python/ops/stateful_random_ops.py,506,method,"Resets the generator by a new state.
See `__init__` for the meaning of ""state"".
Args:
state: the new state."
9280,reset_from_seed,tensorflow/tensorflow/python/ops/stateful_random_ops.py,518,method,"Resets the generator by a new seed.
See `from_seed` for the meaning of ""seed"".
Args:
seed: the new seed."
9281,reset_from_key_counter,tensorflow/tensorflow/python/ops/stateful_random_ops.py,529,method,"Resets the generator by a new key-counter pair.
See `from_key_counter` for the meaning of ""key"" and ""counter"".
Args:
key: the new key.
counter: the new counter."
9282,state,tensorflow/tensorflow/python/ops/stateful_random_ops.py,553,method,The internal state of the RNG.
9283,algorithm,tensorflow/tensorflow/python/ops/stateful_random_ops.py,558,method,The RNG algorithm id (a Python integer or scalar integer Tensor).
9284,key,tensorflow/tensorflow/python/ops/stateful_random_ops.py,567,method,"The 'key' part of the state of a counter-based RNG.
For a counter-base RNG algorithm such as Philox and ThreeFry (as
described in paper 'Parallel Random Numbers: As Easy as 1, 2, 3'
[https://www.thesalmons.org/john/random123/papers/random123sc11.pdf]),
the RNG state consists of two parts: counter and key. The output is
generated via the formula: output=hash(key, counter), i.e. a hashing of
the counter parametrized by the key. Two RNGs with two different keys can
be thought as generating two independent random-number streams (a stream
is formed by increasing the counter).
Returns:
A scalar which is the 'key' part of the state, if the RNG algorithm is
counter-based; otherwise it raises a ValueError."
9285,skip,tensorflow/tensorflow/python/ops/stateful_random_ops.py,589,method,"Advance the counter of a counter-based RNG.
Args:
delta: the amount of advancement. The state of the RNG after
`skip(n)` will be the same as that after `normal([n])`
(or any other distribution). The actual increment added to the
counter is an unspecified implementation detail."
9286,normal,tensorflow/tensorflow/python/ops/stateful_random_ops.py,602,method,"Outputs random values from a normal distribution.
Args:
shape: A 1-D integer Tensor or Python array. The shape of the output
tensor.
mean: A 0-D Tensor or Python value of type `dtype`. The mean of the normal
distribution.
stddev: A 0-D Tensor or Python value of type `dtype`. The standard
deviation of the normal distribution.
dtype: The type of the output.
name: A name for the operation (optional).
Returns:
A tensor of the specified shape filled with random normal values."
9287,truncated_normal,tensorflow/tensorflow/python/ops/stateful_random_ops.py,630,method,"Outputs random values from a truncated normal distribution.
The generated values follow a normal distribution with specified mean and
standard deviation, except that values whose magnitude is more than
2 standard deviations from the mean are dropped and re-picked.
Args:
shape: A 1-D integer Tensor or Python array. The shape of the output
tensor.
mean: A 0-D Tensor or Python value of type `dtype`. The mean of the
truncated normal distribution.
stddev: A 0-D Tensor or Python value of type `dtype`. The standard
deviation of the normal distribution, before truncation.
dtype: The type of the output.
name: A name for the operation (optional).
Returns:
A tensor of the specified shape filled with random truncated normal
values."
9288,uniform,tensorflow/tensorflow/python/ops/stateful_random_ops.py,673,method,"Outputs random values from a uniform distribution.
The generated values follow a uniform distribution in the range
`[minval, maxval)`. The lower bound `minval` is included in the range, while
the upper bound `maxval` is excluded. (For float numbers especially
low-precision types like bfloat16, because of
rounding, the result may sometimes include `maxval`.)
For floats, the default range is `[0, 1)`. For ints, at least `maxval` must
be specified explicitly.
In the integer case, the random integers are slightly biased unless
`maxval - minval` is an exact power of two. The bias is small for values of
`maxval - minval` significantly smaller than the range of the output (either
`2**32` or `2**64`).
For full-range random integers, pass `minval=None` and `maxval=None` with an
integer `dtype` (for integer dtypes, `minval` and `maxval` must be both
`None` or both not `None`).
Args:
shape: A 1-D integer Tensor or Python array. The shape of the output
tensor.
minval: A Tensor or Python value of type `dtype`, broadcastable with
`shape` (for integer types, broadcasting is not supported, so it needs
to be a scalar). The lower bound (included) on the range of random
values to generate. Pass `None` for full-range integers. Defaults to 0.
maxval: A Tensor or Python value of type `dtype`, broadcastable with
`shape` (for integer types, broadcasting is not supported, so it needs
to be a scalar). The upper bound (excluded) on the range of random
values to generate. Pass `None` for full-range integers. Defaults to 1
if `dtype` is floating point.
dtype: The type of the output.
name: A name for the operation (optional).
Returns:
A tensor of the specified shape filled with random uniform values.
Raises:
ValueError: If `dtype` is integral and `maxval` is not specified."
9289,uniform_full_int,tensorflow/tensorflow/python/ops/stateful_random_ops.py,739,method,"Uniform distribution on an integer type's entire range.
This method is the same as setting `minval` and `maxval` to `None` in the
`uniform` method.
Args:
shape: the shape of the output.
dtype: (optional) the integer type, default to uint64.
name: (optional) the name of the node.
Returns:
A tensor of random numbers of the required shape."
9290,binomial,tensorflow/tensorflow/python/ops/stateful_random_ops.py,759,method,"Outputs random values from a binomial distribution.
The generated values follow a binomial distribution with specified count and
probability of success parameters.
Example:
```python
counts = [10., 20.]
# Probability of success.
probs = [0.8]
rng = tf.random.Generator.from_seed(seed=234)
binomial_samples = rng.binomial(shape=[2], counts=counts, probs=probs)
counts = ... # Shape [3, 1, 2]
probs = ... # Shape [1, 4, 2]
shape = [3, 4, 3, 4, 2]
rng = tf.random.Generator.from_seed(seed=1717)
# Sample shape will be [3, 4, 3, 4, 2]
binomial_samples = rng.binomial(shape=shape, counts=counts, probs=probs)
```
Args:
shape: A 1-D integer Tensor or Python array. The shape of the output
tensor.
counts: Tensor. The counts of the binomial distribution. Must be
broadcastable with `probs`, and broadcastable with the rightmost
dimensions of `shape`.
probs: Tensor. The probability of success for the
binomial distribution. Must be broadcastable with `counts` and
broadcastable with the rightmost dimensions of `shape`.
dtype: The type of the output. Default: tf.int32
name: A name for the operation (optional).
Returns:
samples: A Tensor of the specified shape filled with random binomial
values. For each i, each samples[i, ...] is an independent draw from
the binomial distribution on counts[i] trials with probability of
success probs[i]."
9291,make_seeds,tensorflow/tensorflow/python/ops/stateful_random_ops.py,826,method,"Generates seeds for stateless random ops.
For example:
```python
seeds = get_global_generator().make_seeds(count=10)
for i in range(10):
seed = seeds[:, i]
numbers = stateless_random_normal(shape=[2, 3], seed=seed)
...
```
Args:
count: the number of seed pairs (note that stateless random ops need a
pair of seeds to invoke).
Returns:
A tensor of shape [2, count] and dtype int64."
9292,split,tensorflow/tensorflow/python/ops/stateful_random_ops.py,856,method,"Returns a list of independent `Generator` objects.
Two generators are independent of each other in the sense that the
random-number streams they generate don't have statistically detectable
correlations. The new generators are also independent of the old one.
The old generator's state will be changed (like other random-number
generating methods), so two calls of `split` will return different
new generators.
For example:
```python
gens = get_global_generator().split(count=10)
for gen in gens:
numbers = gen.normal(shape=[2, 3])
# ...
gens2 = get_global_generator().split(count=10)
# gens2 will be different from gens
```
The new generators will be put on the current device (possible different
from the old generator's), for example:
```python
with tf.device(""/device:CPU:0""):
gen = Generator(seed=1234) # gen is on CPU
with tf.device(""/device:GPU:0""):
gens = gen.split(count=10) # gens are on GPU
```
Args:
count: the number of generators to return.
Returns:
A list (length `count`) of `Generator` objects independent of each other.
The new generators have the same RNG algorithm as the old one."
9293,get_global_generator,tensorflow/tensorflow/python/ops/stateful_random_ops.py,915,function,"Retrieves the global generator.
This function will create the global generator the first time it is called,
and the generator will be placed at the default device at that time, so one
needs to be careful when this function is first called. Using a generator
placed on a less-ideal device will incur performance regression.
Returns:
The global `tf.random.Generator` object."
9294,set_global_generator,tensorflow/tensorflow/python/ops/stateful_random_ops.py,935,function,"Replaces the global generator with another `Generator` object.
This function creates a new Generator object (and the Variable object within),
which does not work well with tf.function because (1) tf.function puts
restrictions on Variable creation thus reset_global_generator can't be freely
used inside tf.function; (2) redirecting a global variable to
a new object is problematic with tf.function because the old object may be
captured by a 'tf.function'ed function and still be used by it.
A 'tf.function'ed function only keeps weak references to variables,
so deleting a variable and then calling that function again may raise an
error, as demonstrated by
random_test.py/RandomTest.testResetGlobalGeneratorBadWithDefun .
Args:
generator: the new `Generator` object."
9295,split,tensorflow/tensorflow/python/ops/stateless_random_ops.py,45,function,"Splits an RNG seed into `num` new seeds by adding a leading axis.
Example:
>>> seed = [1, 2]
>>> new_seeds = tf.random.experimental.stateless_split(seed, num=3)
>>> print(new_seeds)
tf.Tensor(
[[1105988140 1738052849]
[-335576002 370444179]
[ 10670227 -246211131]], shape=(3, 2), dtype=int32)
>>> tf.random.stateless_normal(shape=[3], seed=new_seeds[0, :])
<tf.Tensor: shape=(3,), dtype=float32, numpy=array([-0.59835213, -0.9578608 ,
0.9002807 ], dtype=float32)>
Args:
seed: an RNG seed (a tensor with shape [2] and dtype `int32` or
`int64`). (When using XLA, only `int32` is allowed.)
num: optional, a positive integer or scalar tensor indicating the number of
seeds to produce (default 2).
Returns:
A tensor with shape [num, 2] representing `num` new seeds. It will have the
same dtype as `seed` (if `seed` doesn't have an explict dtype, the dtype
will be determined by `tf.convert_to_tensor`)."
9296,fold_in,tensorflow/tensorflow/python/ops/stateless_random_ops.py,79,function,"Folds in data to an RNG seed to form a new RNG seed.
For example, in a distributed-training setting, suppose we have a master seed
and a replica ID. We want to fold the replica ID into the master seed to
form a ""replica seed"" to be used by that replica later on, so that different
replicas will generate different random numbers but the reproducibility of the
whole system can still be controlled by the master seed:
>>> master_seed = [1, 2]
>>> replica_id = 3
>>> replica_seed = tf.random.experimental.stateless_fold_in(
... master_seed, replica_id)
>>> print(replica_seed)
tf.Tensor([1105988140 3], shape=(2,), dtype=int32)
>>> tf.random.stateless_normal(shape=[3], seed=replica_seed)
<tf.Tensor: shape=(3,), dtype=float32, numpy=array([0.03197195, 0.8979765 ,
0.13253039], dtype=float32)>
Args:
seed: an RNG seed (a tensor with shape [2] and dtype `int32` or
`int64`). (When using XLA, only `int32` is allowed.)
data: an `int32` or `int64` scalar representing data to be folded in to the
seed.
Returns:
A new RNG seed that is a deterministic function of the inputs and is
statistically safe for producing a stream of new pseudo-random values. It
will have the same dtype as `data` (if `data` doesn't have an explict dtype,
the dtype will be determined by `tf.convert_to_tensor`)."
9297,stateless_random_uniform,tensorflow/tensorflow/python/ops/stateless_random_ops.py,118,function,"Outputs deterministic pseudorandom values from a uniform distribution.
This is a stateless version of `tf.random.uniform`: if run twice with the
same seeds and shapes, it will produce the same pseudorandom numbers. The
output is consistent across multiple runs on the same hardware (and between
CPU and GPU), but may change between versions of TensorFlow or on non-CPU/GPU
hardware.
The generated values follow a uniform distribution in the range
`[minval, maxval)`. The lower bound `minval` is included in the range, while
the upper bound `maxval` is excluded.
For floats, the default range is `[0, 1)`. For ints, at least `maxval` must
be specified explicitly.
In the integer case, the random integers are slightly biased unless
`maxval - minval` is an exact power of two. The bias is small for values of
`maxval - minval` significantly smaller than the range of the output (either
`2**32` or `2**64`).
For full-range (i.e. inclusive of both max and min) random integers, pass
`minval=None` and `maxval=None` with an integer `dtype`. For an integer dtype
either both `minval` and `maxval` must be `None` or neither may be `None`. For
example:
```python
ints = tf.random.stateless_uniform(
[10], seed=(2, 3), minval=None, maxval=None, dtype=tf.int32)
```
Args:
shape: A 1-D integer Tensor or Python array. The shape of the output tensor.
seed: A shape [2] Tensor, the seed to the random number generator. Must have
dtype `int32` or `int64`. (When using XLA, only `int32` is allowed.)
minval: A Tensor or Python value of type `dtype`, broadcastable with
`shape` (for integer types, broadcasting is not supported, so it needs to
be a scalar). The lower bound on the range of random values to
generate. Pass `None` for full-range integers. Defaults to 0.
maxval: A Tensor or Python value of type `dtype`, broadcastable with
`shape` (for integer types, broadcasting is not supported, so it needs to
be a scalar). The upper bound on the range of random values to generate.
Defaults to 1 if `dtype` is floating point. Pass `None` for full-range
integers.
dtype: The type of the output: `float16`, `float32`, `float64`, `int32`, or
`int64`. For unbounded uniform ints (`minval`, `maxval` both `None`),
`uint32` and `uint64` may be used.
name: A name for the operation (optional).
Returns:
A tensor of the specified shape filled with random uniform values.
Raises:
ValueError: If `dtype` is integral and only one of `minval` or `maxval` is
specified."
9298,stateless_random_binomial,tensorflow/tensorflow/python/ops/stateless_random_ops.py,213,function,"Outputs deterministic pseudorandom values from a binomial distribution.
The generated values follow a binomial distribution with specified count and
probability of success parameters.
This is a stateless version of `tf.random.Generator.binomial`: if run twice
with the same seeds and shapes, it will produce the same pseudorandom numbers.
The output is consistent across multiple runs on the same hardware (and
between CPU and GPU), but may change between versions of TensorFlow or on
non-CPU/GPU hardware.
Example:
```python
counts = [10., 20.]
# Probability of success.
probs = [0.8]
binomial_samples = tf.random.stateless_binomial(
shape=[2], seed=[123, 456], counts=counts, probs=probs)
counts = ... # Shape [3, 1, 2]
probs = ... # Shape [1, 4, 2]
shape = [3, 4, 3, 4, 2]
# Sample shape will be [3, 4, 3, 4, 2]
binomial_samples = tf.random.stateless_binomial(
shape=shape, seed=[123, 456], counts=counts, probs=probs)
```
Args:
shape: A 1-D integer Tensor or Python array. The shape of the output tensor.
seed: A shape [2] Tensor, the seed to the random number generator. Must have
dtype `int32` or `int64`. (When using XLA, only `int32` is allowed.)
counts: Tensor. The counts of the binomial distribution. Must be
broadcastable with `probs`, and broadcastable with the rightmost
dimensions of `shape`.
probs: Tensor. The probability of success for the binomial distribution.
Must be broadcastable with `counts` and broadcastable with the rightmost
dimensions of `shape`.
output_dtype: The type of the output. Default: tf.int32
name: A name for the operation (optional).
Returns:
samples: A Tensor of the specified shape filled with random binomial
values. For each i, each samples[..., i] is an independent draw from
the binomial distribution on counts[i] trials with probability of
success probs[i]."
9299,stateless_random_gamma,tensorflow/tensorflow/python/ops/stateless_random_ops.py,283,function,"Outputs deterministic pseudorandom values from a gamma distribution.
The generated values follow a gamma distribution with specified concentration
(`alpha`) and inverse scale (`beta`) parameters.
This is a stateless version of `tf.random.gamma`: if run twice with the same
seeds and shapes, it will produce the same pseudorandom numbers. The output is
consistent across multiple runs on the same hardware (and between CPU and
GPU),
but may change between versions of TensorFlow or on non-CPU/GPU hardware.
A slight difference exists in the interpretation of the `shape` parameter
between `stateless_gamma` and `gamma`: in `gamma`, the `shape` is always
prepended to the shape of the broadcast of `alpha` with `beta`; whereas in
`stateless_gamma` the `shape` parameter must always encompass the shapes of
each of `alpha` and `beta` (which must broadcast together to match the
trailing dimensions of `shape`).
Note: Because internal calculations are done using `float64` and casting has
`floor` semantics, we must manually map zero outcomes to the smallest
possible positive floating-point value, i.e., `np.finfo(dtype).tiny`. This
means that `np.finfo(dtype).tiny` occurs more frequently than it otherwise
should. This bias can only happen for small values of `alpha`, i.e.,
`alpha << 1` or large values of `beta`, i.e., `beta >> 1`.
The samples are differentiable w.r.t. alpha and beta.
The derivatives are computed using the approach described in
(Figurnov et al., 2018).
Example:
```python
samples = tf.random.stateless_gamma([10, 2], seed=[12, 34], alpha=[0.5, 1.5])
# samples has shape [10, 2], where each slice [:, 0] and [:, 1] represents
# the samples drawn from each distribution
samples = tf.random.stateless_gamma([7, 5, 2], seed=[12, 34], alpha=[.5, 1.5])
# samples has shape [7, 5, 2], where each slice [:, :, 0] and [:, :, 1]
# represents the 7x5 samples drawn from each of the two distributions
alpha = tf.constant([[1.], [3.], [5.]])
beta = tf.constant([[3., 4.]])
samples = tf.random.stateless_gamma(
[30, 3, 2], seed=[12, 34], alpha=alpha, beta=beta)
# samples has shape [30, 3, 2], with 30 samples each of 3x2 distributions.
with tf.GradientTape() as tape:
tape.watch([alpha, beta])
loss = tf.reduce_mean(tf.square(tf.random.stateless_gamma(
[30, 3, 2], seed=[12, 34], alpha=alpha, beta=beta)))
dloss_dalpha, dloss_dbeta = tape.gradient(loss, [alpha, beta])
# unbiased stochastic derivatives of the loss function
alpha.shape == dloss_dalpha.shape # True
beta.shape == dloss_dbeta.shape # True
```
Args:
shape: A 1-D integer Tensor or Python array. The shape of the output tensor.
seed: A shape [2] Tensor, the seed to the random number generator. Must have
dtype `int32` or `int64`. (When using XLA, only `int32` is allowed.)
alpha: Tensor. The concentration parameter of the gamma distribution. Must
be broadcastable with `beta`, and broadcastable with the rightmost
dimensions of `shape`.
beta: Tensor. The inverse scale parameter of the gamma distribution. Must be
broadcastable with `alpha` and broadcastable with the rightmost dimensions
of `shape`.
dtype: Floating point dtype of `alpha`, `beta`, and the output.
name: A name for the operation (optional).
Returns:
samples: A Tensor of the specified shape filled with random gamma values.
For each i, each `samples[..., i] is an independent draw from the gamma
distribution with concentration alpha[i] and scale beta[i]."
9300,stateless_random_poisson,tensorflow/tensorflow/python/ops/stateless_random_ops.py,383,function,"Outputs deterministic pseudorandom values from a Poisson distribution.
The generated values follow a Poisson distribution with specified rate
parameter.
This is a stateless version of `tf.random.poisson`: if run twice with the same
seeds and shapes, it will produce the same pseudorandom numbers. The output is
consistent across multiple runs on the same hardware, but may change between
versions of TensorFlow or on non-CPU/GPU hardware.
A slight difference exists in the interpretation of the `shape` parameter
between `stateless_poisson` and `poisson`: in `poisson`, the `shape` is always
prepended to the shape of `lam`; whereas in `stateless_poisson` the shape of
`lam` must match the trailing dimensions of `shape`.
Example:
```python
samples = tf.random.stateless_poisson([10, 2], seed=[12, 34], lam=[5, 15])
# samples has shape [10, 2], where each slice [:, 0] and [:, 1] represents
# the samples drawn from each distribution
samples = tf.random.stateless_poisson([7, 5, 2], seed=[12, 34], lam=[5, 15])
# samples has shape [7, 5, 2], where each slice [:, :, 0] and [:, :, 1]
# represents the 7x5 samples drawn from each of the two distributions
rate = tf.constant([[1.], [3.], [5.]])
samples = tf.random.stateless_poisson([30, 3, 1], seed=[12, 34], lam=rate)
# samples has shape [30, 3, 1], with 30 samples each of 3x1 distributions.
```
Args:
shape: A 1-D integer Tensor or Python array. The shape of the output tensor.
seed: A shape [2] Tensor, the seed to the random number generator. Must have
dtype `int32` or `int64`. (When using XLA, only `int32` is allowed.)
lam: Tensor. The rate parameter ""lambda"" of the Poisson distribution. Shape
must match the rightmost dimensions of `shape`.
dtype: Dtype of the samples (int or float dtypes are permissible, as samples
are discrete). Default: int32.
name: A name for the operation (optional).
Returns:
samples: A Tensor of the specified shape filled with random Poisson values.
For each i, each `samples[..., i]` is an independent draw from the Poisson
distribution with rate `lam[i]`."
9301,stateless_random_normal,tensorflow/tensorflow/python/ops/stateless_random_ops.py,446,function,"Outputs deterministic pseudorandom values from a normal distribution.
This is a stateless version of `tf.random.normal`: if run twice with the
same seeds and shapes, it will produce the same pseudorandom numbers. The
output is consistent across multiple runs on the same hardware (and between
CPU and GPU), but may change between versions of TensorFlow or on non-CPU/GPU
hardware.
Args:
shape: A 1-D integer Tensor or Python array. The shape of the output tensor.
seed: A shape [2] Tensor, the seed to the random number generator. Must have
dtype `int32` or `int64`. (When using XLA, only `int32` is allowed.)
mean: A 0-D Tensor or Python value of type `dtype`. The mean of the normal
distribution.
stddev: A 0-D Tensor or Python value of type `dtype`. The standard deviation
of the normal distribution.
dtype: The type of the output.
name: A name for the operation (optional).
Returns:
A tensor of the specified shape filled with random normal values."
9302,stateless_truncated_normal,tensorflow/tensorflow/python/ops/stateless_random_ops.py,487,function,"Outputs deterministic pseudorandom values, truncated normally distributed.
This is a stateless version of `tf.random.truncated_normal`: if run twice with
the same seeds and shapes, it will produce the same pseudorandom numbers. The
output is consistent across multiple runs on the same hardware (and between
CPU and GPU), but may change between versions of TensorFlow or on non-CPU/GPU
hardware.
The generated values follow a normal distribution with specified mean and
standard deviation, except that values whose magnitude is more than 2 standard
deviations from the mean are dropped and re-picked.
Args:
shape: A 1-D integer Tensor or Python array. The shape of the output tensor.
seed: A shape [2] Tensor, the seed to the random number generator. Must have
dtype `int32` or `int64`. (When using XLA, only `int32` is allowed.)
mean: A 0-D Tensor or Python value of type `dtype`. The mean of the
truncated normal distribution.
stddev: A 0-D Tensor or Python value of type `dtype`. The standard deviation
of the normal distribution, before truncation.
dtype: The type of the output.
name: A name for the operation (optional).
Returns:
A tensor of the specified shape filled with random truncated normal values."
9303,stateless_multinomial,tensorflow/tensorflow/python/ops/stateless_random_ops.py,535,function,"Draws deterministic pseudorandom samples from a multinomial distribution.
This is a stateless version of `tf.random.categorical`: if run twice with the
same seeds and shapes, it will produce the same pseudorandom numbers. The
output is consistent across multiple runs on the same hardware (and between
CPU and GPU), but may change between versions of TensorFlow or on non-CPU/GPU
hardware.
Example:
```python
# samples has shape [1, 5], where each value is either 0 or 1 with equal
# probability.
samples = tf.random.stateless_categorical(
tf.math.log([[0.5, 0.5]]), 5, seed=[7, 17])
```
Args:
logits: 2-D Tensor with shape `[batch_size, num_classes]`. Each slice
`[i, :]` represents the unnormalized log-probabilities for all classes.
num_samples: 0-D. Number of independent samples to draw for each row slice.
seed: A shape [2] Tensor, the seed to the random number generator. Must have
dtype `int32` or `int64`. (When using XLA, only `int32` is allowed.)
output_dtype: integer type to use for the output. Defaults to int64.
name: Optional name for the operation.
Returns:
The drawn samples of shape `[batch_size, num_samples]`."
9304,stateless_categorical,tensorflow/tensorflow/python/ops/stateless_random_ops.py,576,function,"Draws deterministic pseudorandom samples from a categorical distribution.
This is a stateless version of `tf.categorical`: if run twice with the
same seeds and shapes, it will produce the same pseudorandom numbers. The
output is consistent across multiple runs on the same hardware (and between
CPU and GPU), but may change between versions of TensorFlow or on non-CPU/GPU
hardware.
Example:
```python
# samples has shape [1, 5], where each value is either 0 or 1 with equal
# probability.
samples = tf.random.stateless_categorical(
tf.math.log([[0.5, 0.5]]), 5, seed=[7, 17])
```
Args:
logits: 2-D Tensor with shape `[batch_size, num_classes]`. Each slice
`[i, :]` represents the unnormalized log-probabilities for all classes.
num_samples: 0-D. Number of independent samples to draw for each row slice.
seed: A shape [2] Tensor, the seed to the random number generator. Must have
dtype `int32` or `int64`. (When using XLA, only `int32` is allowed.)
dtype: integer type to use for the output. Defaults to int64.
name: Optional name for the operation.
Returns:
The drawn samples of shape `[batch_size, num_samples]`."
9305,stateless_multinomial_categorical_impl,tensorflow/tensorflow/python/ops/stateless_random_ops.py,616,function,Implementation for stateless multinomial/categorical ops (v1/v2).
9306,stateless_parameterized_truncated_normal,tensorflow/tensorflow/python/ops/stateless_random_ops.py,625,function,"Outputs random values from a truncated normal distribution.
The generated values follow a normal distribution with specified mean and
standard deviation, except that values whose magnitude is more than 2 standard
deviations from the mean are dropped and re-picked.
Examples:
Sample from a Truncated normal, with deferring shape parameters that
broadcast.
>>> means = 0.
>>> stddevs = tf.math.exp(tf.random.uniform(shape=[2, 3]))
>>> minvals = [-1., -2., -1000.]
>>> maxvals = [[10000.], [1.]]
>>> y = tf.random.stateless_parameterized_truncated_normal(
... shape=[10, 2, 3], seed=[7, 17],
... means=means, stddevs=stddevs, minvals=minvals, maxvals=maxvals)
>>> y.shape
TensorShape([10, 2, 3])
Args:
shape: A 1-D integer `Tensor` or Python array. The shape of the output
tensor.
seed: A shape [2] Tensor, the seed to the random number generator. Must have
dtype `int32` or `int64`. (When using XLA, only `int32` is allowed.)
means: A `Tensor` or Python value of type `dtype`. The mean of the truncated
normal distribution. This must broadcast with `stddevs`, `minvals` and
`maxvals`, and the broadcasted shape must be dominated by `shape`.
stddevs: A `Tensor` or Python value of type `dtype`. The standard deviation
of the truncated normal distribution. This must broadcast with `means`,
`minvals` and `maxvals`, and the broadcasted shape must be dominated by
`shape`.
minvals: A `Tensor` or Python value of type `dtype`. The minimum value of
the truncated normal distribution. This must broadcast with `means`,
`stddevs` and `maxvals`, and the broadcasted shape must be dominated by
`shape`.
maxvals: A `Tensor` or Python value of type `dtype`. The maximum value of
the truncated normal distribution. This must broadcast with `means`,
`stddevs` and `minvals`, and the broadcasted shape must be dominated by
`shape`.
name: A name for the operation (optional).
Returns:
A tensor of the specified shape filled with random truncated normal values."
9307,regex_full_match,tensorflow/tensorflow/python/ops/string_ops.py,50,function,"Match elements of `input` with regex `pattern`.
Args:
input: string `Tensor`, the source strings to process.
pattern: string or scalar string `Tensor`, regular expression to use,
see more details at https://github.com/google/re2/wiki/Syntax
name: Name of the op.
Returns:
bool `Tensor` of the same shape as `input` with match results."
9308,regex_replace,tensorflow/tensorflow/python/ops/string_ops.py,79,function,"Replace elements of `input` matching regex `pattern` with `rewrite`.
>>> tf.strings.regex_replace(""Text with tags.<br /><b>contains html</b>"",
... ""<[^>]+>"", "" "")
<tf.Tensor: shape=(), dtype=string, numpy=b'Text with tags. contains html '>
Args:
input: string `Tensor`, the source strings to process.
pattern: string or scalar string `Tensor`, regular expression to use,
see more details at https://github.com/google/re2/wiki/Syntax
rewrite: string or scalar string `Tensor`, value to use in match
replacement, supports backslash-escaped digits (\1 to \9) can be to insert
text matching corresponding parenthesized group.
replace_global: `bool`, if `True` replace all non-overlapping matches,
else replace only the first match.
name: A name for the operation (optional).
Returns:
string `Tensor` of the same shape as `input` with specified replacements."
9309,string_format,tensorflow/tensorflow/python/ops/string_ops.py,117,function,"Formats a string template using a list of tensors.
Formats a string template using a list of tensors, abbreviating tensors by
only printing the first and last `summarize` elements of each dimension
(recursively). If formatting only one tensor into a template, the tensor does
not have to be wrapped in a list.
Example:
Formatting a single-tensor template:
>>> tensor = tf.range(5)
>>> tf.strings.format(""tensor: {}, suffix"", tensor)
<tf.Tensor: shape=(), dtype=string, numpy=b'tensor: [0 1 2 3 4], suffix'>
Formatting a multi-tensor template:
>>> tensor_a = tf.range(2)
>>> tensor_b = tf.range(1, 4, 2)
>>> tf.strings.format(""a: {}, b: {}, suffix"", (tensor_a, tensor_b))
<tf.Tensor: shape=(), dtype=string, numpy=b'a: [0 1], b: [1 3], suffix'>
Args:
template: A string template to format tensor values into.
inputs: A list of `Tensor` objects, or a single Tensor.
The list of tensors to format into the template string. If a solitary
tensor is passed in, the input tensor will automatically be wrapped as a
list.
placeholder: An optional `string`. Defaults to `{}`.
At each placeholder occurring in the template, a subsequent tensor
will be inserted.
summarize: An optional `int`. Defaults to `3`.
When formatting the tensors, show the first and last `summarize`
entries of each tensor dimension (recursively). If set to -1, all
elements of the tensor will be shown.
name: A name for the operation (optional).
Returns:
A scalar `Tensor` of type `string`.
Raises:
ValueError: if the number of placeholders does not match the number of
inputs."
9310,string_split,tensorflow/tensorflow/python/ops/string_ops.py,180,function,"Split elements of `source` based on `delimiter` into a `SparseTensor`.
Let N be the size of source (typically N will be the batch size). Split each
element of `source` based on `delimiter` and return a `SparseTensor`
containing the split tokens. Empty tokens are ignored.
If `sep` is an empty string, each element of the `source` is split
into individual strings, each containing one byte. (This includes splitting
multibyte sequences of UTF-8.) If delimiter contains multiple bytes, it is
treated as a set of delimiters with each considered a potential split point.
For example:
N = 2, source[0] is 'hello world' and source[1] is 'a b c', then the output
will be
st.indices = [0, 0;
0, 1;
1, 0;
1, 1;
1, 2]
st.shape = [2, 3]
st.values = ['hello', 'world', 'a', 'b', 'c']
Args:
source: `1-D` string `Tensor`, the strings to split.
sep: `0-D` string `Tensor`, the delimiter character, the string should
be length 0 or 1. Default is ' '.
skip_empty: A `bool`. If `True`, skip the empty strings from the result.
delimiter: deprecated alias for `sep`.
Raises:
ValueError: If delimiter is not a string.
Returns:
A `SparseTensor` of rank `2`, the strings split according to the delimiter.
The first column of the indices corresponds to the row in `source` and the
second column corresponds to the index of the split component in this row."
9311,string_split_v2,tensorflow/tensorflow/python/ops/string_ops.py,237,function,"Split elements of `source` based on `sep` into a `SparseTensor`.
Let N be the size of source (typically N will be the batch size). Split each
element of `source` based on `sep` and return a `SparseTensor`
containing the split tokens. Empty tokens are ignored.
For example, N = 2, source[0] is 'hello world' and source[1] is 'a b c',
then the output will be
st.indices = [0, 0;
0, 1;
1, 0;
1, 1;
1, 2]
st.shape = [2, 3]
st.values = ['hello', 'world', 'a', 'b', 'c']
If `sep` is given, consecutive delimiters are not grouped together and are
deemed to delimit empty strings. For example, source of `""1<>2<><>3""` and
sep of `""<>""` returns `[""1"", ""2"", """", ""3""]`. If `sep` is None or an empty
string, consecutive whitespace are regarded as a single separator, and the
result will contain no empty strings at the start or end if the string has
leading or trailing whitespace.
Note that the above mentioned behavior matches python's str.split.
Args:
source: `1-D` string `Tensor`, the strings to split.
sep: `0-D` string `Tensor`, the delimiter character.
maxsplit: An `int`. If `maxsplit > 0`, limit of the split of the result.
Raises:
ValueError: If sep is not a string.
Returns:
A `SparseTensor` of rank `2`, the strings split according to the delimiter.
The first column of the indices corresponds to the row in `source` and the
second column corresponds to the index of the split component in this row."
9312,reduce_join,tensorflow/tensorflow/python/ops/string_ops.py,310,function,
9313,reduce_join_v2,tensorflow/tensorflow/python/ops/string_ops.py,333,function,"Joins all strings into a single string, or joins along an axis.
>>> tf.strings.reduce_join([['abc','123'],
... ['def','456']]).numpy()
b'abc123def456'
>>> tf.strings.reduce_join([['abc','123'],
... ['def','456']], axis=-1).numpy()
array([b'abc123', b'def456'], dtype=object)
>>> tf.strings.reduce_join([['abc','123'],
... ['def','456']],
... axis=-1,
... separator="" "").numpy()
array([b'abc 123', b'def 456'], dtype=object)
Args:
inputs: A `tf.string` tensor.
axis: Which axis to join along. The default behavior is to join all
elements, producing a scalar.
keepdims: If true, retains reduced dimensions with length 1.
separator: a string added between each string being joined.
name: A name for the operation (optional).
Returns:
A `tf.string` tensor."
9314,string_length,tensorflow/tensorflow/python/ops/string_ops.py,381,function,"Computes the length of each string given in the input tensor.
>>> strings = tf.constant(['Hello','TensorFlow', '🙂'])
>>> tf.strings.length(strings).numpy() # default counts bytes
array([ 5, 10, 4], dtype=int32)
>>> tf.strings.length(strings, unit=""UTF8_CHAR"").numpy()
array([ 5, 10, 1], dtype=int32)
Args:
input: A `Tensor` of type `string`. The strings for which to compute the
length for each element.
name: A name for the operation (optional).
unit: An optional `string` from: `""BYTE"", ""UTF8_CHAR""`. Defaults to
`""BYTE""`. The unit that is counted to compute string length. One of:
`""BYTE""` (for the number of bytes in each string) or `""UTF8_CHAR""` (for
the number of UTF-8 encoded Unicode code points in each string). Results
are undefined if `unit=UTF8_CHAR` and the `input` strings do not contain
structurally valid UTF-8.
Returns:
A `Tensor` of type `int32`, containing the length of the input string in
the same element of the input tensor."
9315,string_length_v2,tensorflow/tensorflow/python/ops/string_ops.py,410,function,
9316,substr_deprecated,tensorflow/tensorflow/python/ops/string_ops.py,420,function,
9317,substr,tensorflow/tensorflow/python/ops/string_ops.py,428,function,
9318,substr_v2,tensorflow/tensorflow/python/ops/string_ops.py,436,function,
9319,string_to_number,tensorflow/tensorflow/python/ops/string_ops.py,456,function,"Converts each string in the input Tensor to the specified numeric type.
(Note that int32 overflow results in an error while float overflow
results in a rounded value.)
Examples:
>>> tf.strings.to_number(""1.55"")
<tf.Tensor: shape=(), dtype=float32, numpy=1.55>
>>> tf.strings.to_number(""3"", tf.int32)
<tf.Tensor: shape=(), dtype=int32, numpy=3>
Args:
input: A `Tensor` of type `string`.
out_type: An optional `tf.DType` from: `tf.float32, tf.float64, tf.int32,
tf.int64`. Defaults to `tf.float32`.
The numeric type to interpret each string in `string_tensor` as.
name: A name for the operation (optional).
Returns:
A `Tensor` of type `out_type`."
9320,string_to_number_v1,tensorflow/tensorflow/python/ops/string_ops.py,484,function,
9321,string_to_hash_bucket,tensorflow/tensorflow/python/ops/string_ops.py,498,function,"Converts each string in the input Tensor to its hash mod by a number of buckets.
The hash function is deterministic on the content of the string within the
process.
Note that the hash function may change from time to time.
This functionality will be deprecated and it's recommended to use
`tf.strings.to_hash_bucket_fast()` or `tf.strings.to_hash_bucket_strong()`.
Examples:
>>> tf.strings.to_hash_bucket([""Hello"", ""TensorFlow"", ""2.x""], 3)
<tf.Tensor: shape=(3,), dtype=int64, numpy=array([2, 0, 1])>
Args:
input: A `Tensor` of type `string`.
num_buckets: An `int` that is `>= 1`. The number of buckets.
name: A name for the operation (optional).
Returns:
A `Tensor` of type `int64`."
9322,string_to_hash_bucket_v1,tensorflow/tensorflow/python/ops/string_ops.py,528,function,
9323,string_join,tensorflow/tensorflow/python/ops/string_ops.py,544,function,"Perform element-wise concatenation of a list of string tensors.
Given a list of string tensors of same shape, performs element-wise
concatenation of the strings of the same index in all tensors.
>>> tf.strings.join(['abc','def']).numpy()
b'abcdef'
>>> tf.strings.join([['abc','123'],
... ['def','456'],
... ['ghi','789']]).numpy()
array([b'abcdefghi', b'123456789'], dtype=object)
>>> tf.strings.join([['abc','123'],
... ['def','456']],
... separator="" "").numpy()
array([b'abc def', b'123 456'], dtype=object)
Args:
inputs: A list of `tf.Tensor` objects of same size and `tf.string` dtype.
separator: A string added between each string being joined.
name: A name for the operation (optional).
Returns:
A `tf.string` tensor."
9324,collect,tensorflow/tensorflow/python/ops/summary_op_util.py,28,function,"Adds keys to a collection.
Args:
val: The value to add per each key.
collections: A collection of keys to add.
default_collections: Used if collections is None."
9325,clean_tag,tensorflow/tensorflow/python/ops/summary_op_util.py,45,function,"Cleans a tag. Removes illegal characters for instance.
Args:
name: The original tag name to be processed.
Returns:
The cleaned tag name."
9326,summary_scope,tensorflow/tensorflow/python/ops/summary_op_util.py,72,function,"Enters a scope used for the summary and yields both the name and tag.
To ensure that the summary tag name is always unique, we create a name scope
based on `name` and use the full scope name in the tag.
If `family` is set, then the tag name will be '<family>/<scope_name>', where
`scope_name` is `<outer_scope>/<family>/<name>`. This ensures that `family`
is always the prefix of the tag (and unmodified), while ensuring the scope
respects the outer scope from this summary was created.
Args:
name: A name for the generated summary node.
family: Optional; if provided, used as the prefix of the summary tag name.
default_name: Optional; if provided, used as default name of the summary.
values: Optional; passed as `values` parameter to name_scope.
Yields:
A tuple `(tag, scope)`, both of which are unique and should be used for the
tag and the scope for the summary to output."
9327,should_record_summaries,tensorflow/tensorflow/python/ops/summary_ops_v2.py,119,function,Returns boolean Tensor which is true if summaries should be recorded.
9328,record_if,tensorflow/tensorflow/python/ops/summary_ops_v2.py,126,function,"Sets summary recording on or off per the provided boolean value.
The provided value can be a python boolean, a scalar boolean Tensor, or
or a callable providing such a value; if a callable is passed it will be
invoked on-demand to determine whether summary writing will occur.
Args:
condition: can be True, False, a bool Tensor, or a callable providing such.
Yields:
Returns a context manager that sets this value on enter and restores the
previous value on exit."
9329,record_summaries_every_n_global_steps,tensorflow/tensorflow/python/ops/summary_ops_v2.py,149,function,Sets the should_record_summaries Tensor to true if global_step % n == 0.
9330,always_record_summaries,tensorflow/tensorflow/python/ops/summary_ops_v2.py,160,function,Sets the should_record_summaries Tensor to always true.
9331,never_record_summaries,tensorflow/tensorflow/python/ops/summary_ops_v2.py,165,function,Sets the should_record_summaries Tensor to always false.
9332,get_step,tensorflow/tensorflow/python/ops/summary_ops_v2.py,171,function,"Returns the default summary step for the current thread.
Returns:
The step set by `tf.summary.experimental.set_step()` if one has been set,
otherwise None."
9333,set_step,tensorflow/tensorflow/python/ops/summary_ops_v2.py,182,function,"Sets the default summary step for the current thread.
For convenience, this function sets a default value for the `step` parameter
used in summary-writing functions elsewhere in the API so that it need not
be explicitly passed in every such invocation. The value can be a constant
or a variable, and can be retrieved via `tf.summary.experimental.get_step()`.
Note: when using this with @tf.functions, the step value will be captured at
the time the function is traced, so changes to the step outside the function
will not be reflected inside the function unless using a `tf.Variable` step.
Args:
step: An `int64`-castable default step value, or None to unset."
9334,SummaryWriter,tensorflow/tensorflow/python/ops/summary_ops_v2.py,202,class,Interface representing a stateful summary writer object.
9335,set_as_default,tensorflow/tensorflow/python/ops/summary_ops_v2.py,206,method,"Enables this summary writer for the current thread.
For convenience, if `step` is not None, this function also sets a default
value for the `step` parameter used in summary-writing functions elsewhere
in the API so that it need not be explicitly passed in every such
invocation. The value can be a constant or a variable.
Note: when setting `step` in a @tf.function, the step value will be
captured at the time the function is traced, so changes to the step outside
the function will not be reflected inside the function unless using
a `tf.Variable` step.
Args:
step: An `int64`-castable default step value, or `None`. When not `None`,
the current step is modified to the given value. When `None`, the
current step is not modified."
9336,as_default,tensorflow/tensorflow/python/ops/summary_ops_v2.py,228,method,"Returns a context manager that enables summary writing.
For convenience, if `step` is not None, this function also sets a default
value for the `step` parameter used in summary-writing functions elsewhere
in the API so that it need not be explicitly passed in every such
invocation. The value can be a constant or a variable.
Note: when setting `step` in a @tf.function, the step value will be
captured at the time the function is traced, so changes to the step outside
the function will not be reflected inside the function unless using
a `tf.Variable` step.
For example, `step` can be used as:
```python
with writer_a.as_default(step=10):
tf.summary.scalar(tag, value) # Logged to writer_a with step 10
with writer_b.as_default(step=20):
tf.summary.scalar(tag, value) # Logged to writer_b with step 20
tf.summary.scalar(tag, value) # Logged to writer_a with step 10
```
Args:
step: An `int64`-castable default step value, or `None`. When not `None`,
the current step is captured, replaced by a given one, and the original
one is restored when the context manager exits. When `None`, the current
step is not modified (and not restored when the context manager exits)."
9337,init,tensorflow/tensorflow/python/ops/summary_ops_v2.py,259,method,Initializes the summary writer.
9338,flush,tensorflow/tensorflow/python/ops/summary_ops_v2.py,263,method,Flushes any buffered data.
9339,close,tensorflow/tensorflow/python/ops/summary_ops_v2.py,267,method,Flushes and closes the summary writer.
9340,ResourceSummaryWriter,tensorflow/tensorflow/python/ops/summary_ops_v2.py,272,class,Implementation of SummaryWriter using a SummaryWriterInterface resource.
9341,set_as_default,tensorflow/tensorflow/python/ops/summary_ops_v2.py,295,method,"Enables this summary writer for the current thread.
For convenience, if `step` is not None, this function also sets a default
value for the `step` parameter used in summary-writing functions elsewhere
in the API so that it need not be explicitly passed in every such
invocation. The value can be a constant or a variable.
Note: when setting `step` in a @tf.function, the step value will be
captured at the time the function is traced, so changes to the step outside
the function will not be reflected inside the function unless using
a `tf.Variable` step.
Args:
step: An `int64`-castable default step value, or `None`. When not `None`,
the current step is modified to the given value. When `None`, the
current step is not modified."
9342,as_default,tensorflow/tensorflow/python/ops/summary_ops_v2.py,320,method,"Returns a context manager that enables summary writing.
For convenience, if `step` is not None, this function also sets a default
value for the `step` parameter used in summary-writing functions elsewhere
in the API so that it need not be explicitly passed in every such
invocation. The value can be a constant or a variable.
Note: when setting `step` in a @tf.function, the step value will be
captured at the time the function is traced, so changes to the step outside
the function will not be reflected inside the function unless using
a `tf.Variable` step.
For example, `step` can be used as:
```python
with writer_a.as_default(step=10):
tf.summary.scalar(tag, value) # Logged to writer_a with step 10
with writer_b.as_default(step=20):
tf.summary.scalar(tag, value) # Logged to writer_b with step 20
tf.summary.scalar(tag, value) # Logged to writer_a with step 10
```
Args:
step: An `int64`-castable default step value, or `None`. When not `None`,
the current step is captured, replaced by a given one, and the original
one is restored when the context manager exits. When `None`, the current
step is not modified (and not restored when the context manager exits)."
9343,init,tensorflow/tensorflow/python/ops/summary_ops_v2.py,367,method,Initializes the summary writer.
9344,flush,tensorflow/tensorflow/python/ops/summary_ops_v2.py,376,method,Flushes any buffered data.
9345,close,tensorflow/tensorflow/python/ops/summary_ops_v2.py,382,method,Flushes and closes the summary writer.
9346,NoopSummaryWriter,tensorflow/tensorflow/python/ops/summary_ops_v2.py,395,class,"A summary writer that does nothing, for create_noop_writer()."
9347,set_as_default,tensorflow/tensorflow/python/ops/summary_ops_v2.py,398,method,
9348,as_default,tensorflow/tensorflow/python/ops/summary_ops_v2.py,402,method,
9349,init,tensorflow/tensorflow/python/ops/summary_ops_v2.py,405,method,
9350,flush,tensorflow/tensorflow/python/ops/summary_ops_v2.py,408,method,
9351,close,tensorflow/tensorflow/python/ops/summary_ops_v2.py,411,method,
9352,initialize,tensorflow/tensorflow/python/ops/summary_ops_v2.py,416,function,"Initializes summary writing for graph execution mode.
This operation is a no-op when executing eagerly.
This helper method provides a higher-level alternative to using
`tf.contrib.summary.summary_writer_initializer_op` and
`tf.contrib.summary.graph`.
Most users will also want to call `tf.compat.v1.train.create_global_step`
which can happen before or after this function is called.
Args:
graph: A `tf.Graph` or `tf.compat.v1.GraphDef` to output to the writer.
This function will not write the default graph by default. When
writing to an event log file, the associated step will be zero.
session: So this method can call `tf.Session.run`. This defaults
to `tf.compat.v1.get_default_session`.
Raises:
RuntimeError: If the current thread has no default
`tf.contrib.summary.SummaryWriter`.
ValueError: If session wasn't passed and no default session."
9353,create_file_writer_v2,tensorflow/tensorflow/python/ops/summary_ops_v2.py,458,function,"Creates a summary file writer for the given log directory.
Args:
logdir: a string specifying the directory in which to write an event file.
max_queue: the largest number of summaries to keep in a queue; will
flush once the queue gets bigger than this. Defaults to 10.
flush_millis: the largest interval between flushes. Defaults to 120,000.
filename_suffix: optional suffix for the event file name. Defaults to `.v2`.
name: a name for the op that creates the writer.
Returns:
A SummaryWriter object."
9354,create_file_writer,tensorflow/tensorflow/python/ops/summary_ops_v2.py,519,function,"Creates a summary file writer in the current context under the given name.
Args:
logdir: a string, or None. If a string, creates a summary file writer
which writes to the directory named by the string. If None, returns
a mock object which acts like a summary writer but does nothing,
useful to use as a context manager.
max_queue: the largest number of summaries to keep in a queue; will
flush once the queue gets bigger than this. Defaults to 10.
flush_millis: the largest interval between flushes. Defaults to 120,000.
filename_suffix: optional suffix for the event file name. Defaults to `.v2`.
name: Shared name for this SummaryWriter resource stored to default
Graph. Defaults to the provided logdir prefixed with `logdir:`. Note: if a
summary writer resource with this shared name already exists, the returned
SummaryWriter wraps that resource and the other arguments have no effect.
Returns:
Either a summary writer or an empty object which can be used as a
summary writer."
9355,create_db_writer,tensorflow/tensorflow/python/ops/summary_ops_v2.py,566,function,"Creates a summary database writer in the current context.
This can be used to write tensors from the execution graph directly
to a database. Only SQLite is supported right now. This function
will create the schema if it doesn't exist. Entries in the Users,
Experiments, and Runs tables will be created automatically if they
don't already exist.
Args:
db_uri: For example ""file:/tmp/foo.sqlite"".
experiment_name: Defaults to YYYY-MM-DD in local time if None.
Empty string means the Run will not be associated with an
Experiment. Can't contain ASCII control characters or <>. Case
sensitive.
run_name: Defaults to HH:MM:SS in local time if None. Empty string
means a Tag will not be associated with any Run. Can't contain
ASCII control characters or <>. Case sensitive.
user_name: Defaults to system username if None. Empty means the
Experiment will not be associated with a User. Must be valid as
both a DNS label and Linux username.
name: Shared name for this SummaryWriter resource stored to default
`tf.Graph`.
Returns:
A `tf.summary.SummaryWriter` instance."
9356,create_noop_writer,tensorflow/tensorflow/python/ops/summary_ops_v2.py,619,function,"Returns a summary writer that does nothing.
This is useful as a placeholder in code that expects a context manager."
9357,all_v2_summary_ops,tensorflow/tensorflow/python/ops/summary_ops_v2.py,639,function,"Returns all V2-style summary ops defined in the current default graph.
This includes ops from TF 2.0 tf.summary and TF 1.x tf.contrib.summary (except
for `tf.contrib.summary.graph` and `tf.contrib.summary.import_event`), but
does *not* include TF 1.x tf.summary ops.
Returns:
List of summary ops, or None if called under eager execution."
9358,summary_writer_initializer_op,tensorflow/tensorflow/python/ops/summary_ops_v2.py,654,function,"Graph-mode only. Returns the list of ops to create all summary writers.
Returns:
The initializer ops.
Raises:
RuntimeError: If in Eager mode."
9359,summary_scope,tensorflow/tensorflow/python/ops/summary_ops_v2.py,675,function,"Experimental context manager for use when defining a custom summary op.
This behaves similarly to `tf.name_scope`, except that it returns a generated
summary tag in addition to the scope name. The tag is structurally similar to
the scope name - derived from the user-provided name, prefixed with enclosing
name scopes if any - but we relax the constraint that it be uniquified, as
well as the character set limitation (so the user-provided name can contain
characters not legal for scope names; in the scope name these are removed).
This makes the summary tag more predictable and consistent for the user.
For example, to define a new summary op called `my_op`:
```python
def my_op(name, my_value, step):
with tf.summary.summary_scope(name, ""MyOp"", [my_value]) as (tag, scope):
my_value = tf.convert_to_tensor(my_value)
return tf.summary.write(tag, my_value, step=step)
```
Args:
name: string name for the summary.
default_name: Optional; if provided, used as default name of the summary.
values: Optional; passed as `values` parameter to name_scope.
Yields:
A tuple `(tag, scope)` as described above."
9360,write,tensorflow/tensorflow/python/ops/summary_ops_v2.py,715,function,"Writes a generic summary to the default SummaryWriter if one exists.
This exists primarily to support the definition of type-specific summary ops
like scalar() and image(), and is not intended for direct use unless defining
a new type-specific summary op.
Args:
tag: string tag used to identify the summary (e.g. in TensorBoard), usually
generated with `tf.summary.summary_scope`
tensor: the Tensor holding the summary data to write or a callable that
returns this Tensor. If a callable is passed, it will only be called when
a default SummaryWriter exists and the recording condition specified by
`record_if()` is met.
step: Explicit `int64`-castable monotonic step value for this summary. If
omitted, this defaults to `tf.summary.experimental.get_step()`, which must
not be None.
metadata: Optional SummaryMetadata, as a proto or serialized bytes
name: Optional string name for this op.
Returns:
True on success, or false if no summary was written because no default
summary writer was available.
Raises:
ValueError: if a default writer exists, but no step was provided and
`tf.summary.experimental.get_step()` is None."
9361,write_raw_pb,tensorflow/tensorflow/python/ops/summary_ops_v2.py,782,function,"Writes a summary using raw `tf.compat.v1.Summary` protocol buffers.
Experimental: this exists to support the usage of V1-style manual summary
writing (via the construction of a `tf.compat.v1.Summary` protocol buffer)
with the V2 summary writing API.
Args:
tensor: the string Tensor holding one or more serialized `Summary` protobufs
step: Explicit `int64`-castable monotonic step value for this summary. If
omitted, this defaults to `tf.summary.experimental.get_step()`, which must
not be None.
name: Optional string name for this op.
Returns:
True on success, or false if no summary was written because no default
summary writer was available.
Raises:
ValueError: if a default writer exists, but no step was provided and
`tf.summary.experimental.get_step()` is None."
9362,summary_writer_function,tensorflow/tensorflow/python/ops/summary_ops_v2.py,833,function,"Helper function to write summaries.
Args:
name: name of the summary
tensor: main tensor to form the summary
function: function taking a tag and a scope which writes the summary
family: optional, the summary's family
Returns:
The result of writing the summary."
9363,generic,tensorflow/tensorflow/python/ops/summary_ops_v2.py,865,function,Writes a tensor summary if possible.
9364,scalar,tensorflow/tensorflow/python/ops/summary_ops_v2.py,886,function,"Writes a scalar summary if possible.
Unlike `tf.contrib.summary.generic` this op may change the dtype
depending on the writer, for both practical and efficiency concerns.
Args:
name: An arbitrary name for this summary.
tensor: A `tf.Tensor` Must be one of the following types:
`float32`, `float64`, `int32`, `int64`, `uint8`, `int16`,
`int8`, `uint16`, `half`, `uint32`, `uint64`.
family: Optional, the summary's family.
step: The `int64` monotonic step variable, which defaults
to `tf.compat.v1.train.get_global_step`.
Returns:
The created `tf.Operation` or a `tf.no_op` if summary writing has
not been enabled for this context."
9365,histogram,tensorflow/tensorflow/python/ops/summary_ops_v2.py,918,function,Writes a histogram summary if possible.
9366,image,tensorflow/tensorflow/python/ops/summary_ops_v2.py,933,function,Writes an image summary if possible.
9367,audio,tensorflow/tensorflow/python/ops/summary_ops_v2.py,952,function,Writes an audio summary if possible.
9368,graph,tensorflow/tensorflow/python/ops/summary_ops_v2.py,969,function,"Writes a TensorFlow graph to the summary interface.
The graph summary is, strictly speaking, not a summary. Conditions
like `tf.summary.should_record_summaries` do not apply. Only
a single graph can be associated with a particular run. If multiple
graphs are written, then only the last one will be considered by
TensorBoard.
When not using eager execution mode, the user should consider passing
the `graph` parameter to `tf.compat.v1.summary.initialize` instead of
calling this function. Otherwise special care needs to be taken when
using the graph to record the graph.
Args:
param: A `tf.Tensor` containing a serialized graph proto. When
eager execution is enabled, this function will automatically
coerce `tf.Graph`, `tf.compat.v1.GraphDef`, and string types.
step: The global step variable. This doesn't have useful semantics
for graph summaries, but is used anyway, due to the structure of
event log files. This defaults to the global step.
name: A name for the operation (optional).
Returns:
The created `tf.Operation` or a `tf.no_op` if summary writing has
not been enabled for this context.
Raises:
TypeError: If `param` isn't already a `tf.Tensor` in graph mode."
9369,import_event,tensorflow/tensorflow/python/ops/summary_ops_v2.py,1017,function,"Writes a `tf.compat.v1.Event` binary proto.
This can be used to import existing event logs into a new summary writer sink.
Please note that this is lower level than the other summary functions and
will ignore the `tf.summary.should_record_summaries` setting.
Args:
tensor: A `tf.Tensor` of type `string` containing a serialized
`tf.compat.v1.Event` proto.
name: A name for the operation (optional).
Returns:
The created `tf.Operation`."
9370,flush,tensorflow/tensorflow/python/ops/summary_ops_v2.py,1037,function,"Forces summary writer to send any buffered data to storage.
This operation blocks until that finishes.
Args:
writer: The `tf.summary.SummaryWriter` resource to flush.
The thread default will be used if this parameter is None.
Otherwise a `tf.no_op` is returned.
name: A name for the operation (optional).
Returns:
The created `tf.Operation`."
9371,eval_dir,tensorflow/tensorflow/python/ops/summary_ops_v2.py,1067,function,Construct a logdir for an eval summary writer.
9372,create_summary_file_writer,tensorflow/tensorflow/python/ops/summary_ops_v2.py,1074,function,Please use `tf.contrib.summary.create_file_writer`.
9373,run_metadata,tensorflow/tensorflow/python/ops/summary_ops_v2.py,1120,function,"Writes entire RunMetadata summary.
A RunMetadata can contain DeviceStats, partition graphs, and function graphs.
Please refer to the proto for definition of each field.
Args:
name: A name for this summary. The summary tag used for TensorBoard will be
this name prefixed by any active name scopes.
data: A RunMetadata proto to write.
step: Explicit `int64`-castable monotonic step value for this summary. If
omitted, this defaults to `tf.summary.experimental.get_step()`, which must
not be None.
Returns:
True on success, or false if no summary was written because no default
summary writer was available.
Raises:
ValueError: if a default writer exists, but no step was provided and
`tf.summary.experimental.get_step()` is None."
9374,run_metadata_graphs,tensorflow/tensorflow/python/ops/summary_ops_v2.py,1162,function,"Writes graphs from a RunMetadata summary.
Args:
name: A name for this summary. The summary tag used for TensorBoard will be
this name prefixed by any active name scopes.
data: A RunMetadata proto to write.
step: Explicit `int64`-castable monotonic step value for this summary. If
omitted, this defaults to `tf.summary.experimental.get_step()`, which must
not be None.
Returns:
True on success, or false if no summary was written because no default
summary writer was available.
Raises:
ValueError: if a default writer exists, but no step was provided and
`tf.summary.experimental.get_step()` is None."
9375,keras_model,tensorflow/tensorflow/python/ops/summary_ops_v2.py,1205,function,"Writes a Keras model as JSON to as a Summary.
Writing the Keras model configuration allows the TensorBoard graph plugin to
render a conceptual graph, as opposed to graph of ops. In case the model fails
to serialize as JSON, it ignores and returns False.
Args:
name: A name for this summary. The summary tag used for TensorBoard will be
this name prefixed by any active name scopes.
data: A Keras Model to write.
step: Explicit `int64`-castable monotonic step value for this summary. If
omitted, this defaults to `tf.summary.experimental.get_step()`, which must
not be None.
Returns:
True on success, or False if no summary was written because no default
summary writer was available.
Raises:
ValueError: if a default writer exists, but no step was provided and
`tf.summary.experimental.get_step()` is None."
9376,trace_on,tensorflow/tensorflow/python/ops/summary_ops_v2.py,1258,function,"Starts a trace to record computation graphs and profiling information.
Must be invoked in eager mode.
When enabled, TensorFlow runtime will collection information that can later be
exported and consumed by TensorBoard. The trace is activated across the entire
TensorFlow runtime and affects all threads of execution.
To stop the trace and export the collected information, use
`tf.summary.trace_export`. To stop the trace without exporting, use
`tf.summary.trace_off`.
Args:
graph: If True, enables collection of executed graphs. It includes ones from
tf.function invocation and ones from the legacy graph mode. The default
is True.
profiler: If True, enables the advanced profiler. Enabling profiler
implicitly enables the graph collection. The profiler may incur a high
memory overhead. The default is False."
9377,trace_export,tensorflow/tensorflow/python/ops/summary_ops_v2.py,1303,function,"Stops and exports the active trace as a Summary and/or profile file.
Stops the trace and exports all metadata collected during the trace to the
default SummaryWriter, if one has been set.
Args:
name: A name for the summary to be written.
step: Explicit `int64`-castable monotonic step value for this summary. If
omitted, this defaults to `tf.summary.experimental.get_step()`, which must
not be None.
profiler_outdir: Output directory for profiler. This is only used when the
profiler was enabled when the trace was started. In that case, if there is
a logdir-based default SummaryWriter, this defaults to the same directory,
but otherwise the argument must be passed.
Raises:
ValueError: if a default writer exists, but no step was provided and
`tf.summary.experimental.get_step()` is None."
9378,trace_off,tensorflow/tensorflow/python/ops/summary_ops_v2.py,1359,function,Stops the current trace and discards any collected information.
9379,make_template,tensorflow/tensorflow/python/ops/template.py,40,function,"Given an arbitrary function, wrap it so that it does variable sharing.
This wraps `func_` in a Template and partially evaluates it. Templates are
functions that create variables the first time they are called and reuse them
thereafter. In order for `func_` to be compatible with a `Template` it must
have the following properties:
* The function should create all trainable variables and any variables that
should be reused by calling `tf.compat.v1.get_variable`. If a trainable
variable is
created using `tf.Variable`, then a ValueError will be thrown. Variables
that are intended to be locals can be created by specifying
`tf.Variable(..., trainable=false)`.
* The function may use variable scopes and other templates internally to
create and reuse variables, but it shouldn't use
`tf.compat.v1.global_variables` to
capture variables that are defined outside of the scope of the function.
* Internal scopes and variable names should not depend on any arguments that
are not supplied to `make_template`. In general you will get a ValueError
telling you that you are trying to reuse a variable that doesn't exist
if you make a mistake.
In the following example, both `z` and `w` will be scaled by the same `y`. It
is important to note that if we didn't assign `scalar_name` and used a
different name for z and w that a `ValueError` would be thrown because it
couldn't reuse the variable.
```python
def my_op(x, scalar_name):
var1 = tf.compat.v1.get_variable(scalar_name,
shape=[],
initializer=tf.compat.v1.constant_initializer(1))
return x * var1
scale_by_y = tf.compat.v1.make_template('scale_by_y', my_op, scalar_name='y')
z = scale_by_y(input1)
w = scale_by_y(input2)
```
As a safe-guard, the returned function will raise a `ValueError` after the
first call if trainable variables are created by calling `tf.Variable`.
If all of these are true, then 2 properties are enforced by the template:
1. Calling the same template multiple times will share all non-local
variables.
2. Two different templates are guaranteed to be unique, unless you reenter the
same variable scope as the initial definition of a template and redefine
it. An examples of this exception:
```python
def my_op(x, scalar_name):
var1 = tf.compat.v1.get_variable(scalar_name,
shape=[],
initializer=tf.compat.v1.constant_initializer(1))
return x * var1
with tf.compat.v1.variable_scope('scope') as vs:
scale_by_y = tf.compat.v1.make_template('scale_by_y', my_op,
scalar_name='y')
z = scale_by_y(input1)
w = scale_by_y(input2)
# Creates a template that reuses the variables above.
with tf.compat.v1.variable_scope(vs, reuse=True):
scale_by_y2 = tf.compat.v1.make_template('scale_by_y', my_op,
scalar_name='y')
z2 = scale_by_y2(input1)
w2 = scale_by_y2(input2)
```
Depending on the value of `create_scope_now_`, the full variable scope may be
captured either at the time of first call or at the time of construction. If
this option is set to True, then all Tensors created by repeated calls to the
template will have an extra trailing _N+1 to their name, as the first time the
scope is entered in the Template constructor no Tensors are created.
Note: `name_`, `func_` and `create_scope_now_` have a trailing underscore to
reduce the likelihood of collisions with kwargs.
Args:
name_: A name for the scope created by this template. If necessary, the name
will be made unique by appending `_N` to the name.
func_: The function to wrap.
create_scope_now_: Boolean controlling whether the scope should be created
when the template is constructed or when the template is called. Default
is False, meaning the scope is created when the template is called.
unique_name_: When used, it overrides name_ and is not made unique. If a
template of the same scope/unique_name already exists and reuse is false,
an error is raised. Defaults to None.
custom_getter_: Optional custom getter for variables used in `func_`. See
the `tf.compat.v1.get_variable` `custom_getter` documentation for more
information.
**kwargs: Keyword arguments to apply to `func_`.
Returns:
A function to encapsulate a set of variables which should be created once
and reused. An enclosing scope will be created either when `make_template`
is called or when the result is called, depending on the value of
`create_scope_now_`. Regardless of the value, the first time the template
is called it will enter the scope with no reuse, and call `func_` to create
variables, which are guaranteed to be unique. All subsequent calls will
re-enter the scope and reuse those variables.
Raises:
ValueError: if `name_` is None."
9380,make_template_internal,tensorflow/tensorflow/python/ops/template.py,164,function,"Make a template, optionally compiling func_ into a graph function.
See `make_template` for full documentation.
Args:
name_: A name for the scope created by this template. If necessary, the name
will be made unique by appending `_N` to the name.
func_: The function to wrap.
create_scope_now_: Boolean controlling whether the scope should be created
when the template is constructed or when the template is called. Default
is False, meaning the scope is created when the template is called.
unique_name_: When used, it overrides name_ and is not made unique. If a
template of the same scope/unique_name already exists and reuse is false,
an error is raised. Defaults to None. If executing eagerly, must be None.
custom_getter_: Optional custom getter for variables used in `func_`. See
the `tf.compat.v1.get_variable` `custom_getter` documentation for more
information.
create_graph_function_: When True, `func_` will be executed as a graph
function. This implies that `func_` must satisfy the properties that
`function.defun` requires of functions: See the documentation of
`function.defun` for details. When executing eagerly, setting this flag
to True can improve performance. Regardless of whether eager execution
is enabled, enabling this flag gives the caller access to graph-function
semantics, i.e., accesses to variables are totally ordered and
side-effecting ops are not pruned.
**kwargs: Keyword arguments to apply to `func_`.
Returns:
A function to encapsulate a set of variables which should be created once
and reused. An enclosing scope will be created either when `make_template`
is called or when the result is called, depending on the value of
`create_scope_now_`. Regardless of the value, the first time the template
is called it will enter the scope with no reuse, and call `func_` to create
variables, which are guaranteed to be unique. All subsequent calls will
re-enter the scope and reuse those variables.
Raises:
ValueError: if `name_` is None.
ValueError: if `unique_name_` is not None and eager execution is enabled."
9381,Template,tensorflow/tensorflow/python/ops/template.py,242,class,"Wrap a function to aid in variable sharing.
Templates are functions that create variables the first time they are called
and reuse them thereafter. See `make_template` for full documentation.
Note: By default, the full variable scope is captured at the time of first
call. If `create_scope_now_` is passed as True to the constructor, the full
scope will be captured there, but no variables will created until the first
call."
9382,name,tensorflow/tensorflow/python/ops/template.py,396,method,Returns the name given to this Template.
9383,func,tensorflow/tensorflow/python/ops/template.py,401,method,Returns the func given to this Template.
9384,variable_scope,tensorflow/tensorflow/python/ops/template.py,406,method,Returns the variable scope object created by this Template.
9385,variable_scope_name,tensorflow/tensorflow/python/ops/template.py,411,method,Returns the variable scope name created by this Template.
9386,variables,tensorflow/tensorflow/python/ops/template.py,422,method,Returns the list of global and local variables created by the Template.
9387,trainable_variables,tensorflow/tensorflow/python/ops/template.py,427,method,Returns the list of trainable variables created by the Template.
9388,non_trainable_variables,tensorflow/tensorflow/python/ops/template.py,436,method,Returns the list of non-trainable variables created by the Template.
9389,global_variables,tensorflow/tensorflow/python/ops/template.py,444,method,Returns the list of global variables created by the Template.
9390,local_variables,tensorflow/tensorflow/python/ops/template.py,453,method,Returns the list of global variables created by the Template.
9391,weights,tensorflow/tensorflow/python/ops/template.py,462,method,List of weights/variables created by the Template.
9392,trainable_weights,tensorflow/tensorflow/python/ops/template.py,467,method,List of trainable weights/variables created by the Template.
9393,non_trainable_weights,tensorflow/tensorflow/python/ops/template.py,472,method,List of non-trainable weights/variables created by the Template.
9394,var_scope,tensorflow/tensorflow/python/ops/template.py,480,method,Returns the variable scope object created by this Template.
9395,EagerTemplate,tensorflow/tensorflow/python/ops/template.py,539,class,"Wrap a function to aid in variable sharing in Eager mode.
Templates are functions that create variables the first time they are called
and reuse them thereafter. See `make_template` for full documentation.
Note: By default, the full variable scope is captured at the time of first
call. If `create_scope_now` is passed as True to the constructor, the full
scope will be captured there, but no variables will be created until the first
call."
9396,variables,tensorflow/tensorflow/python/ops/template.py,684,method,Returns the list of variables created by the Template.
9397,trainable_variables,tensorflow/tensorflow/python/ops/template.py,692,method,Returns the list of trainable variables created by the Template.
9398,non_trainable_variables,tensorflow/tensorflow/python/ops/template.py,700,method,Returns the list of non-trainable variables created by the Template.
9399,global_variables,tensorflow/tensorflow/python/ops/template.py,708,method,Returns the list of global variables created by the Template.
9400,local_variables,tensorflow/tensorflow/python/ops/template.py,716,method,Returns the list of global variables created by the Template.
9401,TensorArray,tensorflow/tensorflow/python/ops/tensor_array_ops.py,947,class,"Class wrapping dynamic-sized, per-time-step, write-once Tensor arrays.
This class is meant to be used with dynamic iteration primitives such as
`while_loop` and `map_fn`. It supports gradient back-propagation via special
""flow"" control flow dependencies.
Example 1: Plain reading and writing.
>>> ta = tf.TensorArray(tf.float32, size=0, dynamic_size=True, clear_after_read=False)
>>> ta = ta.write(0, 10)
>>> ta = ta.write(1, 20)
>>> ta = ta.write(2, 30)
>>>
>>> ta.read(0)
<tf.Tensor: shape=(), dtype=float32, numpy=10.0>
>>> ta.read(1)
<tf.Tensor: shape=(), dtype=float32, numpy=20.0>
>>> ta.read(2)
<tf.Tensor: shape=(), dtype=float32, numpy=30.0>
>>> ta.stack()
<tf.Tensor: shape=(3,), dtype=float32, numpy=array([10., 20., 30.],
dtype=float32)>
Example 2: Fibonacci sequence algorithm that writes in a loop then returns.
>>> @tf.function
... def fibonacci(n):
... ta = tf.TensorArray(tf.float32, size=0, dynamic_size=True)
... ta = ta.unstack([0., 1.])
...
... for i in range(2, n):
... ta = ta.write(i, ta.read(i - 1) + ta.read(i - 2))
...
... return ta.stack()
>>>
>>> fibonacci(7)
<tf.Tensor: shape=(7,), dtype=float32,
numpy=array([0., 1., 1., 2., 3., 5., 8.], dtype=float32)>
Example 3: A simple loop interacting with a `tf.Variable`.
# TODO(b/153898334): Convert back to doctest once bug is resolved.
```
v = tf.Variable(1)
@tf.function
def f(x):
ta = tf.TensorArray(tf.int32, size=0, dynamic_size=True)
for i in tf.range(x):
v.assign_add(i)
ta = ta.write(i, v)
return ta.stack()
f(5)
<tf.Tensor: shape=(5,), dtype=int32, numpy=array([ 1, 2, 4, 7, 11],
dtype=int32)>
```"
9402,flow,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1087,method,The flow `Tensor` forcing ops leading to this TensorArray state.
9403,dtype,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1092,method,The data type of this TensorArray.
9404,handle,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1097,method,The reference to the TensorArray.
9405,element_shape,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1102,method,The `tf.TensorShape` of elements in this TensorArray.
9406,dynamic_size,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1107,method,Python bool; if `True` the TensorArray can grow dynamically.
9407,identity,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1118,method,"Returns a TensorArray with the same content and properties.
Returns:
A new TensorArray object with flow that ensures the control dependencies
from the contexts will become control dependencies for writes, reads, etc.
Use this object for all subsequent operations."
9408,grad,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1128,method,
9409,read,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1131,method,"Read the value at location `index` in the TensorArray.
Args:
index: 0-D. int32 tensor with the index to read from.
name: A name for the operation (optional).
Returns:
The tensor at index `index`."
9410,write,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1144,method,"Write `value` into index `index` of the TensorArray.
Args:
index: 0-D. int32 scalar with the index to write to.
value: N-D. Tensor of type `dtype`. The Tensor to write to this index.
name: A name for the operation (optional).
Returns:
A new TensorArray object with flow that ensures the write occurs.
Use this object for all subsequent operations.
Raises:
ValueError: if there are more writers than specified."
9411,stack,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1161,method,"Return the values in the TensorArray as a stacked `Tensor`.
All of the values must have been written and their shapes must all match.
If input shapes have rank-`R`, then output shape will have rank-`(R+1)`.
Args:
name: A name for the operation (optional).
Returns:
All the tensors in the TensorArray stacked into one tensor."
9412,gather,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1175,method,"Return selected values in the TensorArray as a packed `Tensor`.
All of selected values must have been written and their shapes
must all match.
Args:
indices: A `1-D` `Tensor` taking values in `[0, max_value)`. If
the `TensorArray` is not dynamic, `max_value=size()`.
name: A name for the operation (optional).
Returns:
The tensors in the `TensorArray` selected by `indices`, packed into one
tensor."
9413,concat,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1192,method,"Return the values in the TensorArray as a concatenated `Tensor`.
All of the values must have been written, their ranks must match, and
and their shapes must all match for all dimensions except the first.
Args:
name: A name for the operation (optional).
Returns:
All the tensors in the TensorArray concatenated into one tensor."
9414,unstack,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1207,method,"Unstack the values of a `Tensor` in the TensorArray.
If input value shapes have rank-`R`, then the output TensorArray will
contain elements whose shapes are rank-`(R-1)`.
Args:
value: (N+1)-D. Tensor of type `dtype`. The Tensor to unstack.
name: A name for the operation (optional).
Returns:
A new TensorArray object with flow that ensures the unstack occurs.
Use this object for all subsequent operations.
Raises:
ValueError: if the shape inference fails."
9415,scatter,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1227,method,"Scatter the values of a `Tensor` in specific indices of a `TensorArray`.
Args:
indices: A `1-D` `Tensor` taking values in `[0, max_value)`. If
the `TensorArray` is not dynamic, `max_value=size()`.
value: (N+1)-D. Tensor of type `dtype`. The Tensor to unpack.
name: A name for the operation (optional).
Returns:
A new TensorArray object with flow that ensures the scatter occurs.
Use this object for all subsequent operations.
Raises:
ValueError: if the shape inference fails."
9416,split,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1246,method,"Split the values of a `Tensor` into the TensorArray.
Args:
value: (N+1)-D. Tensor of type `dtype`. The Tensor to split.
lengths: 1-D. int32 vector with the lengths to use when splitting
`value` along its first dimension.
name: A name for the operation (optional).
Returns:
A new TensorArray object with flow that ensures the split occurs.
Use this object for all subsequent operations.
Raises:
ValueError: if the shape inference fails."
9417,size,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1264,method,Return the size of the TensorArray.
9418,close,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1269,method,Close the current TensorArray.
9419,build_ta_with_new_flow,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1274,function,Builds a TensorArray with a new `flow` tensor.
9420,TensorArraySpec,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1319,class,Type specification for a `tf.TensorArray`.
9421,is_compatible_with,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1341,method,
9422,most_specific_compatible_type,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1352,method,
9423,from_value,tensorflow/tensorflow/python/ops/tensor_array_ops.py,1397,method,
9424,TreeVariableSaveable,tensorflow/tensorflow/python/ops/tensor_forest_ops.py,26,class,Resource that holds a tree.
9425,restore,tensorflow/tensorflow/python/ops/tensor_forest_ops.py,54,method,"Restores the associated tree from 'restored_tensors'.
Args:
restored_tensors: the tensors that were loaded from a checkpoint.
unused_restored_shapes: the shapes this object should conform to after
restore. Not meaningful for trees.
Returns:
The operation that restores the state of the tree variable."
9426,resource,tensorflow/tensorflow/python/ops/tensor_forest_ops.py,72,method,
9427,tree_variable,tensorflow/tensorflow/python/ops/tensor_forest_ops.py,76,function,
9428,ForestVariables,tensorflow/tensorflow/python/ops/tensor_forest_ops.py,86,class,Resource that holds all trees from a forest.
9429,build_graph,tensorflow/tensorflow/python/ops/transpose_benchmark.py,34,function,"builds a graph containing a sequence of conv2d operations.
Args:
device: String, the device to run on.
input_shape: Shape of the input tensor.
perm: A list of ints with the same length as input tensor's dimension.
datatype: numpy data type of the input tensor.
num_iters: number of iterations to run transpose.
Returns:
An array of tensors to run()"
9430,TransposeBenchmark,tensorflow/tensorflow/python/ops/transpose_benchmark.py,62,class,Benchmark transpose!
9431,benchmark_transpose,tensorflow/tensorflow/python/ops/transpose_benchmark.py,110,method,
9432,UnconnectedGradients,tensorflow/tensorflow/python/ops/unconnected_gradients.py,27,class,"Controls how gradient computation behaves when y does not depend on x.
The gradient of y with respect to x can be zero in two different ways: there
could be no differentiable path in the graph connecting x to y (and so we can
statically prove that the gradient is zero) or it could be that runtime values
of tensors in a particular execution lead to a gradient of zero (say, if a
relu unit happens to not be activated). To allow you to distinguish between
these two cases you can choose what value gets returned for the gradient when
there is no path in the graph from x to y:
* `NONE`: Indicates that [None] will be returned if there is no path from x
to y
* `ZERO`: Indicates that a zero tensor will be returned in the shape of x."
9433,enable_resource_variables,tensorflow/tensorflow/python/ops/variable_scope.py,219,function,"Creates resource variables by default.
Resource variables are improved versions of TensorFlow variables with a
well-defined memory model. Accessing a resource variable reads its value, and
all ops which access a specific read value of the variable are guaranteed to
see the same value for that tensor. Writes which happen after a read (by
having a control or data dependency on the read) are guaranteed not to affect
the value of the read tensor, and similarly writes which happen before a read
are guaranteed to affect the value. No guarantees are made about unordered
read/write pairs.
Calling tf.enable_resource_variables() lets you opt-in to this TensorFlow 2.0
feature."
9434,resource_variables_enabled,tensorflow/tensorflow/python/ops/variable_scope.py,240,function,"Returns `True` if resource variables are enabled.
Resource variables are improved versions of TensorFlow variables with a
well-defined memory model. Accessing a resource variable reads its value, and
all ops which access a specific read value of the variable are guaranteed to
see the same value for that tensor. Writes which happen after a read (by
having a control or data dependency on the read) are guaranteed not to affect
the value of the read tensor, and similarly writes which happen before a read
are guaranteed to affect the value. No guarantees are made about unordered
read/write pairs.
Calling tf.enable_resource_variables() lets you opt-in to this TensorFlow 2.0
feature."
9435,disable_resource_variables,tensorflow/tensorflow/python/ops/variable_scope.py,262,function,"Opts out of resource variables.
If your code needs tf.disable_resource_variables() to be called to work
properly please file a bug."
9436,no_regularizer,tensorflow/tensorflow/python/ops/variable_scope.py,1080,function,Use this function to prevent regularization of variables.
9437,VariableScope,tensorflow/tensorflow/python/ops/variable_scope.py,1087,class,"Variable scope object to carry defaults to provide to `get_variable`.
Many of the arguments we need for `get_variable` in a variable store are most
easily handled with a context. This object is used for the defaults.
Attributes:
name: name of the current scope, used as prefix in get_variable.
initializer: default initializer passed to get_variable.
regularizer: default regularizer passed to get_variable.
reuse: Boolean, None, or tf.compat.v1.AUTO_REUSE, setting the reuse in
get_variable. When eager execution is enabled this argument is always
forced to be False.
caching_device: string, callable, or None: the caching device passed to
get_variable.
partitioner: callable or `None`: the partitioner passed to `get_variable`.
custom_getter: default custom getter passed to get_variable.
name_scope: The name passed to `tf.name_scope`.
dtype: default type passed to get_variable (defaults to DT_FLOAT).
use_resource: if False, create a normal Variable; if True create an
experimental ResourceVariable with well-defined semantics. Defaults to
False (will later change to True). When eager execution is enabled this
argument is always forced to be True.
constraint: An optional projection function to be applied to the variable
after being updated by an `Optimizer` (e.g. used to implement norm
constraints or value constraints for layer weights). The function must
take as input the unprojected Tensor representing the value of the
variable and return the Tensor for the projected value (which must have
the same shape). Constraints are not safe to use when doing asynchronous
distributed training."
9438,name,tensorflow/tensorflow/python/ops/variable_scope.py,1151,method,
9439,original_name_scope,tensorflow/tensorflow/python/ops/variable_scope.py,1155,method,
9440,reuse,tensorflow/tensorflow/python/ops/variable_scope.py,1159,method,
9441,initializer,tensorflow/tensorflow/python/ops/variable_scope.py,1163,method,
9442,dtype,tensorflow/tensorflow/python/ops/variable_scope.py,1167,method,
9443,use_resource,tensorflow/tensorflow/python/ops/variable_scope.py,1171,method,
9444,regularizer,tensorflow/tensorflow/python/ops/variable_scope.py,1175,method,
9445,caching_device,tensorflow/tensorflow/python/ops/variable_scope.py,1179,method,
9446,partitioner,tensorflow/tensorflow/python/ops/variable_scope.py,1183,method,
9447,custom_getter,tensorflow/tensorflow/python/ops/variable_scope.py,1187,method,
9448,constraint,tensorflow/tensorflow/python/ops/variable_scope.py,1191,method,
9449,reuse_variables,tensorflow/tensorflow/python/ops/variable_scope.py,1194,method,Reuse variables in this scope.
9450,set_initializer,tensorflow/tensorflow/python/ops/variable_scope.py,1198,method,Set initializer for this scope.
9451,set_dtype,tensorflow/tensorflow/python/ops/variable_scope.py,1202,method,Set data type for this scope.
9452,set_use_resource,tensorflow/tensorflow/python/ops/variable_scope.py,1206,method,Sets whether to use ResourceVariables for this scope.
9453,set_regularizer,tensorflow/tensorflow/python/ops/variable_scope.py,1213,method,Set regularizer for this scope.
9454,set_caching_device,tensorflow/tensorflow/python/ops/variable_scope.py,1217,method,Set caching_device for this scope.
9455,set_partitioner,tensorflow/tensorflow/python/ops/variable_scope.py,1224,method,Set partitioner for this scope.
9456,set_custom_getter,tensorflow/tensorflow/python/ops/variable_scope.py,1228,method,Set custom getter for this scope.
9457,get_collection,tensorflow/tensorflow/python/ops/variable_scope.py,1232,method,Get this scope's variables.
9458,trainable_variables,tensorflow/tensorflow/python/ops/variable_scope.py,1237,method,Get this scope's trainable variables.
9459,global_variables,tensorflow/tensorflow/python/ops/variable_scope.py,1241,method,Get this scope's global variables.
9460,local_variables,tensorflow/tensorflow/python/ops/variable_scope.py,1245,method,Get this scope's local variables.
9461,get_variable,tensorflow/tensorflow/python/ops/variable_scope.py,1249,method,Gets an existing variable with this name or create a new one.
9462,get_variable_scope_store,tensorflow/tensorflow/python/ops/variable_scope.py,1424,function,Returns the variable scope store for current thread.
9463,get_variable_scope,tensorflow/tensorflow/python/ops/variable_scope.py,1438,function,Returns the current variable scope.
9464,with_variable_store,tensorflow/tensorflow/python/ops/variable_scope.py,1453,function,
9465,EagerVariableStore,tensorflow/tensorflow/python/ops/variable_scope.py,1463,class,"Wrapper allowing functional layers to be used with eager execution.
When eager execution is enabled Variables get deleted when they go out of
scope, and are not stored in global collections by default. A lot of code
(mostly the functional layers in tf.layers) assumes that variables are kept in
a global list.
EagerVariableStore can be used in conjunction with this code to make it
eager-friendly. For example, to create a dense layer, use:
```
container = tfe.EagerVariableStore()
for input in dataset_iterator:
with container.as_default():
x = tf.compat.v1.layers.dense(input, name=""l1"")
print(container.variables) # Should print the variables used in the layer.
```"
9466,as_default,tensorflow/tensorflow/python/ops/variable_scope.py,1494,method,
9467,variables,tensorflow/tensorflow/python/ops/variable_scope.py,1497,method,
9468,trainable_variables,tensorflow/tensorflow/python/ops/variable_scope.py,1500,method,
9469,non_trainable_variables,tensorflow/tensorflow/python/ops/variable_scope.py,1506,method,
9470,copy,tensorflow/tensorflow/python/ops/variable_scope.py,1512,method,"Copy this variable store and all of its contents.
Variables contained in this store will be copied over to the new variable
store, meaning that they can be modified without affecting the variables in
this store.
Returns:
A new EagerVariableStore instance containing copied variables."
9471,get_variable,tensorflow/tensorflow/python/ops/variable_scope.py,1545,function,
9472,get_local_variable,tensorflow/tensorflow/python/ops/variable_scope.py,1688,function,
9473,variable_scope,tensorflow/tensorflow/python/ops/variable_scope.py,2054,class,"A context manager for defining ops that creates variables (layers).
This context manager validates that the (optional) `values` are from the same
graph, ensures that graph is the default graph, and pushes a name scope and a
variable scope.
If `name_or_scope` is not None, it is used as is. If `name_or_scope` is None,
then `default_name` is used. In that case, if the same name has been
previously used in the same scope, it will be made unique by appending `_N`
to it.
Variable scope allows you to create new variables and to share already created
ones while providing checks to not create or share by accident. For details,
see the [Variable Scope How To](https://tensorflow.org/guide/variables), here
we present only a few basic examples.
Simple example of how to create a new variable:
```python
with tf.compat.v1.variable_scope(""foo""):
with tf.compat.v1.variable_scope(""bar""):
v = tf.compat.v1.get_variable(""v"", [1])
assert v.name == ""foo/bar/v:0""
```
Simple example of how to reenter a premade variable scope safely:
```python
with tf.compat.v1.variable_scope(""foo"") as vs:
pass
# Re-enter the variable scope.
with tf.compat.v1.variable_scope(vs,
auxiliary_name_scope=False) as vs1:
# Restore the original name_scope.
with tf.name_scope(vs1.original_name_scope):
v = tf.compat.v1.get_variable(""v"", [1])
assert v.name == ""foo/v:0""
c = tf.constant([1], name=""c"")
assert c.name == ""foo/c:0""
```
Keep in mind that the counters for `default_name` are discarded once the
parent scope is exited. Therefore when the code re-enters the scope (for
instance by saving it), all nested default_name counters will be restarted.
For instance:
```python
with tf.compat.v1.variable_scope(""foo"") as vs:
with tf.compat.v1.variable_scope(None, default_name=""bar""):
v = tf.compat.v1.get_variable(""a"", [1])
assert v.name == ""foo/bar/a:0"", v.name
with tf.compat.v1.variable_scope(None, default_name=""bar""):
v = tf.compat.v1.get_variable(""b"", [1])
assert v.name == ""foo/bar_1/b:0""
with tf.compat.v1.variable_scope(vs):
with tf.compat.v1.variable_scope(None, default_name=""bar""):
v = tf.compat.v1.get_variable(""c"", [1])
assert v.name == ""foo/bar/c:0"" # Uses bar instead of bar_2!
```
Basic example of sharing a variable AUTO_REUSE:
```python
def foo():
with tf.compat.v1.variable_scope(""foo"", reuse=tf.compat.v1.AUTO_REUSE):
v = tf.compat.v1.get_variable(""v"", [1])
return v
v1 = foo() # Creates v.
v2 = foo() # Gets the same, existing v.
assert v1 == v2
```
Basic example of sharing a variable with reuse=True:
```python
with tf.compat.v1.variable_scope(""foo""):
v = tf.compat.v1.get_variable(""v"", [1])
with tf.compat.v1.variable_scope(""foo"", reuse=True):
v1 = tf.compat.v1.get_variable(""v"", [1])
assert v1 == v
```
Sharing a variable by capturing a scope and setting reuse:
```python
with tf.compat.v1.variable_scope(""foo"") as scope:
v = tf.compat.v1.get_variable(""v"", [1])
scope.reuse_variables()
v1 = tf.compat.v1.get_variable(""v"", [1])
assert v1 == v
```
To prevent accidental sharing of variables, we raise an exception when getting
an existing variable in a non-reusing scope.
```python
with tf.compat.v1.variable_scope(""foo""):
v = tf.compat.v1.get_variable(""v"", [1])
v1 = tf.compat.v1.get_variable(""v"", [1])
# Raises ValueError(""... v already exists ..."").
```
Similarly, we raise an exception when trying to get a variable that does not
exist in reuse mode.
```python
with tf.compat.v1.variable_scope(""foo"", reuse=True):
v = tf.compat.v1.get_variable(""v"", [1])
# Raises ValueError(""... v does not exists ..."").
```
Note that the `reuse` flag is inherited: if we open a reusing scope, then all
its sub-scopes become reusing as well.
A note about name scoping: Setting `reuse` does not impact the naming of other
ops such as mult. See related discussion on
[github#6189](https://github.com/tensorflow/tensorflow/issues/6189)
Note that up to and including version 1.0, it was allowed (though explicitly
discouraged) to pass False to the reuse argument, yielding undocumented
behaviour slightly different from None. Starting at 1.1.0 passing None and
False as reuse has exactly the same effect.
A note about using variable scopes in multi-threaded environment: Variable
scopes are thread local, so one thread will not see another thread's current
scope. Also, when using `default_name`, unique scopes names are also generated
only on a per thread basis. If the same name was used within a different
thread, that doesn't prevent a new thread from creating the same scope.
However, the underlying variable store is shared across threads (within the
same graph). As such, if another thread tries to create a new variable with
the same name as a variable created by a previous thread, it will fail unless
reuse is True.
Further, each thread starts with an empty variable scope. So if you wish to
preserve name prefixes from a scope from the main thread, you should capture
the main thread's scope and re-enter it in each thread. For e.g.
```
main_thread_scope = variable_scope.get_variable_scope()
# Thread's target function:
def thread_target_fn(captured_scope):
with variable_scope.variable_scope(captured_scope):
# .... regular code for this thread
thread = threading.Thread(target=thread_target_fn, args=(main_thread_scope,))
```"
9474,variable_op_scope,tensorflow/tensorflow/python/ops/variable_scope.py,2465,function,Deprecated: context manager for defining an op that creates variables.
9475,default_variable_creator,tensorflow/tensorflow/python/ops/variable_scope.py,2561,function,Default variable creator.
9476,default_variable_creator_v2,tensorflow/tensorflow/python/ops/variable_scope.py,2620,function,Default variable creator.
9477,variable_creator_scope_v1,tensorflow/tensorflow/python/ops/variable_scope.py,2668,function,"Scope which defines a variable creation function to be used by variable().
variable_creator is expected to be a function with the following signature:
```
def variable_creator(next_creator, **kwargs)
```
The creator is supposed to eventually call the next_creator to create a
variable if it does want to create a variable and not call Variable or
ResourceVariable directly. This helps make creators composable. A creator may
choose to create multiple variables, return already existing variables, or
simply register that a variable was created and defer to the next creators in
line. Creators can also modify the keyword arguments seen by the next
creators.
Custom getters in the variable scope will eventually resolve down to these
custom creators when they do create variables.
The valid keyword arguments in kwds are:
* initial_value: A `Tensor`, or Python object convertible to a `Tensor`,
which is the initial value for the Variable. The initial value must have
a shape specified unless `validate_shape` is set to False. Can also be a
callable with no argument that returns the initial value when called. In
that case, `dtype` must be specified. (Note that initializer functions
from init_ops.py must first be bound to a shape before being used here.)
* trainable: If `True`, the default, also adds the variable to the graph
collection `GraphKeys.TRAINABLE_VARIABLES`. This collection is used as
the default list of variables to use by the `Optimizer` classes.
`trainable` defaults to `True`, unless `synchronization` is
set to `ON_READ`, in which case it defaults to `False`.
* collections: List of graph collections keys. The new variable is added to
these collections. Defaults to `[GraphKeys.GLOBAL_VARIABLES]`.
* validate_shape: If `False`, allows the variable to be initialized with a
value of unknown shape. If `True`, the default, the shape of
`initial_value` must be known.
* caching_device: Optional device string describing where the Variable
should be cached for reading. Defaults to the Variable's device.
If not `None`, caches on another device. Typical use is to cache
on the device where the Ops using the Variable reside, to deduplicate
copying through `Switch` and other conditional statements.
* name: Optional name for the variable. Defaults to `'Variable'` and gets
uniquified automatically.
* dtype: If set, initial_value will be converted to the given type.
If `None`, either the datatype will be kept (if `initial_value` is
a Tensor), or `convert_to_tensor` will decide.
* constraint: A constraint function to be applied to the variable after
updates by some algorithms.
* use_resource: if True, a ResourceVariable is always created.
* synchronization: Indicates when a distributed a variable will be
aggregated. Accepted values are constants defined in the class
`tf.VariableSynchronization`. By default the synchronization is set to
`AUTO` and the current `DistributionStrategy` chooses
when to synchronize.
* aggregation: Indicates how a distributed variable will be aggregated.
Accepted values are constants defined in the class
`tf.VariableAggregation`.
This set may grow over time, so it's important the signature of creators is as
mentioned above.
Args:
variable_creator: the passed creator
Yields:
A scope in which the creator is active"
9478,variable_creator_scope,tensorflow/tensorflow/python/ops/variable_scope.py,2744,function,"Scope which defines a variable creation function to be used by variable().
variable_creator is expected to be a function with the following signature:
```
def variable_creator(next_creator, **kwargs)
```
The creator is supposed to eventually call the next_creator to create a
variable if it does want to create a variable and not call Variable or
ResourceVariable directly. This helps make creators composable. A creator may
choose to create multiple variables, return already existing variables, or
simply register that a variable was created and defer to the next creators in
line. Creators can also modify the keyword arguments seen by the next
creators.
Custom getters in the variable scope will eventually resolve down to these
custom creators when they do create variables.
The valid keyword arguments in kwds are:
* initial_value: A `Tensor`, or Python object convertible to a `Tensor`,
which is the initial value for the Variable. The initial value must have
a shape specified unless `validate_shape` is set to False. Can also be a
callable with no argument that returns the initial value when called. In
that case, `dtype` must be specified. (Note that initializer functions
from init_ops.py must first be bound to a shape before being used here.)
* trainable: If `True`, the default, GradientTapes automatically watch
uses of this Variable.
* validate_shape: If `False`, allows the variable to be initialized with a
value of unknown shape. If `True`, the default, the shape of
`initial_value` must be known.
* caching_device: Optional device string describing where the Variable
should be cached for reading. Defaults to the Variable's device.
If not `None`, caches on another device. Typical use is to cache
on the device where the Ops using the Variable reside, to deduplicate
copying through `Switch` and other conditional statements.
* name: Optional name for the variable. Defaults to `'Variable'` and gets
uniquified automatically.
dtype: If set, initial_value will be converted to the given type.
If `None`, either the datatype will be kept (if `initial_value` is
a Tensor), or `convert_to_tensor` will decide.
* constraint: A constraint function to be applied to the variable after
updates by some algorithms.
* synchronization: Indicates when a distributed a variable will be
aggregated. Accepted values are constants defined in the class
`tf.VariableSynchronization`. By default the synchronization is set to
`AUTO` and the current `DistributionStrategy` chooses
when to synchronize.
* aggregation: Indicates how a distributed variable will be aggregated.
Accepted values are constants defined in the class
`tf.VariableAggregation`.
This set may grow over time, so it's important the signature of creators is as
mentioned above.
Args:
variable_creator: the passed creator
Yields:
A scope in which the creator is active"
9479,default_variable_creator,tensorflow/tensorflow/python/ops/variables.py,53,function,
9480,default_variable_creator_v2,tensorflow/tensorflow/python/ops/variables.py,58,function,
9481,VariableSynchronization,tensorflow/tensorflow/python/ops/variables.py,73,class,"Indicates when a distributed variable will be synced.
* `AUTO`: Indicates that the synchronization will be determined by the current
`DistributionStrategy` (eg. With `MirroredStrategy` this would be
`ON_WRITE`).
* `NONE`: Indicates that there will only be one copy of the variable, so
there is no need to sync.
* `ON_WRITE`: Indicates that the variable will be updated across devices
every time it is written.
* `ON_READ`: Indicates that the variable will be aggregated across devices
when it is read (eg. when checkpointing or when evaluating an op that uses
the variable)."
9482,VariableAggregationV2,tensorflow/tensorflow/python/ops/variables.py,95,class,"Indicates how a distributed variable will be aggregated.
`tf.distribute.Strategy` distributes a model by making multiple copies
(called ""replicas"") acting data-parallel on different elements of the input
batch. When performing some variable-update operation, say
`var.assign_add(x)`, in a model, we need to resolve how to combine the
different values for `x` computed in the different replicas.
* `NONE`: This is the default, giving an error if you use a
variable-update operation with multiple replicas.
* `SUM`: Add the updates across replicas.
* `MEAN`: Take the arithmetic mean (""average"") of the updates across replicas.
* `ONLY_FIRST_REPLICA`: This is for when every replica is performing the same
update, but we only want to perform the update once. Used, e.g., for the
global step counter."
9483,VariableAggregation,tensorflow/tensorflow/python/ops/variables.py,130,class,
9484,validate_synchronization_aggregation_trainable,tensorflow/tensorflow/python/ops/variables.py,151,function,"Given user-provided variable properties, sets defaults and validates."
9485,VariableMetaclass,tensorflow/tensorflow/python/ops/variables.py,179,class,Metaclass to allow construction of tf.Variable to be overridden.
9486,Variable,tensorflow/tensorflow/python/ops/variables.py,269,class,"See the [variable guide](https://tensorflow.org/guide/variable).
A variable maintains shared, persistent state manipulated by a program.
The `Variable()` constructor requires an initial value for the variable, which
can be a `Tensor` of any type and shape. This initial value defines the type
and shape of the variable. After construction, the type and shape of the
variable are fixed. The value can be changed using one of the assign methods.
>>> v = tf.Variable(1.)
>>> v.assign(2.)
<tf.Variable ... shape=() dtype=float32, numpy=2.0>
>>> v.assign_add(0.5)
<tf.Variable ... shape=() dtype=float32, numpy=2.5>
The `shape` argument to `Variable`'s constructor allows you to construct a
variable with a less defined shape than its `initial_value`:
>>> v = tf.Variable(1., shape=tf.TensorShape(None))
>>> v.assign([[1.]])
<tf.Variable ... shape=<unknown> dtype=float32, numpy=array([[1.]], ...)>
Just like any `Tensor`, variables created with `Variable()` can be used as
inputs to operations. Additionally, all the operators overloaded for the
`Tensor` class are carried over to variables.
>>> w = tf.Variable([[1.], [2.]])
>>> x = tf.constant([[3., 4.]])
>>> tf.matmul(w, x)
<tf.Tensor:... shape=(2, 2), ... numpy=
array([[3., 4.],
[6., 8.]], dtype=float32)>
>>> tf.sigmoid(w + x)
<tf.Tensor:... shape=(2, 2), ...>
When building a machine learning model it is often convenient to distinguish
between variables holding trainable model parameters and other variables such
as a `step` variable used to count training steps. To make this easier, the
variable constructor supports a `trainable=<bool>`
parameter. `tf.GradientTape` watches trainable variables by default:
>>> with tf.GradientTape(persistent=True) as tape:
... trainable = tf.Variable(1.)
... non_trainable = tf.Variable(2., trainable=False)
... x1 = trainable * 2.
... x2 = non_trainable * 3.
>>> tape.gradient(x1, trainable)
<tf.Tensor:... shape=(), dtype=float32, numpy=2.0>
>>> assert tape.gradient(x2, non_trainable) is None # Unwatched
Variables are automatically tracked when assigned to attributes of types
inheriting from `tf.Module`.
>>> m = tf.Module()
>>> m.v = tf.Variable([1.])
>>> m.trainable_variables
(<tf.Variable ... shape=(1,) ... numpy=array([1.], dtype=float32)>,)
This tracking then allows saving variable values to
[training checkpoints](https://www.tensorflow.org/guide/checkpoint), or to
[SavedModels](https://www.tensorflow.org/guide/saved_model) which include
serialized TensorFlow graphs.
Variables are often captured and manipulated by `tf.function`s. This works the
same way the un-decorated function would have:
>>> v = tf.Variable(0.)
>>> read_and_decrement = tf.function(lambda: v.assign_sub(0.1))
>>> read_and_decrement()
<tf.Tensor: shape=(), dtype=float32, numpy=-0.1>
>>> read_and_decrement()
<tf.Tensor: shape=(), dtype=float32, numpy=-0.2>
Variables created inside a `tf.function` must be owned outside the function
and be created only once:
>>> class M(tf.Module):
... @tf.function
... def __call__(self, x):
... if not hasattr(self, ""v""): # Or set self.v to None in __init__
... self.v = tf.Variable(x)
... return self.v * x
>>> m = M()
>>> m(2.)
<tf.Tensor: shape=(), dtype=float32, numpy=4.0>
>>> m(3.)
<tf.Tensor: shape=(), dtype=float32, numpy=6.0>
>>> m.v
<tf.Variable ... shape=() dtype=float32, numpy=2.0>
See the `tf.function` documentation for details."
9487,value,tensorflow/tensorflow/python/ops/variables.py,443,method,"Returns the last snapshot of this variable.
You usually do not need to call this method as all ops that need the value
of the variable call it automatically through a `convert_to_tensor()` call.
Returns a `Tensor` which holds the value of the variable. You can not
assign a new value to this tensor as it is not a reference to the variable.
To avoid copies, if the consumer of the returned value is on the same device
as the variable, this actually returns the live value of the variable, not
a copy. Updates to the variable are seen by the consumer. If the consumer
is on a different device it will get a copy of the variable.
Returns:
A `Tensor` containing the value of the variable."
9488,read_value,tensorflow/tensorflow/python/ops/variables.py,462,method,"Returns the value of this variable, read in the current context.
Can be different from value() if it's on another device, with control
dependencies, etc.
Returns:
A `Tensor` containing the value of the variable."
9489,set_shape,tensorflow/tensorflow/python/ops/variables.py,473,method,"Overrides the shape for this variable.
Args:
shape: the `TensorShape` representing the overridden shape."
9490,trainable,tensorflow/tensorflow/python/ops/variables.py,482,method,
9491,synchronization,tensorflow/tensorflow/python/ops/variables.py,486,method,
9492,aggregation,tensorflow/tensorflow/python/ops/variables.py,490,method,
9493,eval,tensorflow/tensorflow/python/ops/variables.py,493,method,"In a session, computes and returns the value of this variable.
This is not a graph construction method, it does not add ops to the graph.
This convenience method requires a session where the graph
containing this variable has been launched. If no session is
passed, the default session is used. See `tf.compat.v1.Session` for more
information on launching a graph and on sessions.
```python
v = tf.Variable([1, 2])
init = tf.compat.v1.global_variables_initializer()
with tf.compat.v1.Session() as sess:
sess.run(init)
# Usage passing the session explicitly.
print(v.eval(sess))
# Usage with the default session. The 'with' block
# above makes 'sess' the default session.
print(v.eval())
```
Args:
session: The session to use to evaluate this variable. If none, the
default session is used.
Returns:
A numpy `ndarray` with a copy of the value of this variable."
9494,initialized_value,tensorflow/tensorflow/python/ops/variables.py,528,method,"Returns the value of the initialized variable.
You should use this instead of the variable itself to initialize another
variable with a value that depends on the value of this variable.
```python
# Initialize 'v' with a random tensor.
v = tf.Variable(tf.random.truncated_normal([10, 40]))
# Use `initialized_value` to guarantee that `v` has been
# initialized before its value is used to initialize `w`.
# The random values are picked only once.
w = tf.Variable(v.initialized_value() * 2.0)
```
Returns:
A `Tensor` holding the value of this variable after its initializer
has run."
9495,initial_value,tensorflow/tensorflow/python/ops/variables.py,553,method,"Returns the Tensor used as the initial value for the variable.
Note that this is different from `initialized_value()` which runs
the op that initializes the variable before returning its value.
This method returns the tensor that is used by the op that initializes
the variable.
Returns:
A `Tensor`."
9496,constraint,tensorflow/tensorflow/python/ops/variables.py,567,method,"Returns the constraint function associated with this variable.
Returns:
The constraint function that was passed to the variable constructor.
Can be `None` if no constraint was passed."
9497,assign,tensorflow/tensorflow/python/ops/variables.py,576,method,"Assigns a new value to the variable.
This is essentially a shortcut for `assign(self, value)`.
Args:
value: A `Tensor`. The new value for this variable.
use_locking: If `True`, use locking during the assignment.
name: The name of the operation to be created
read_value: if True, will return something which evaluates to the new
value of the variable; if False will return the assign op.
Returns:
The updated variable. If `read_value` is false, instead returns None in
Eager mode and the assign op in graph mode."
9498,assign_add,tensorflow/tensorflow/python/ops/variables.py,594,method,"Adds a value to this variable.
This is essentially a shortcut for `assign_add(self, delta)`.
Args:
delta: A `Tensor`. The value to add to this variable.
use_locking: If `True`, use locking during the operation.
name: The name of the operation to be created
read_value: if True, will return something which evaluates to the new
value of the variable; if False will return the assign op.
Returns:
The updated variable. If `read_value` is false, instead returns None in
Eager mode and the assign op in graph mode."
9499,assign_sub,tensorflow/tensorflow/python/ops/variables.py,612,method,"Subtracts a value from this variable.
This is essentially a shortcut for `assign_sub(self, delta)`.
Args:
delta: A `Tensor`. The value to subtract from this variable.
use_locking: If `True`, use locking during the operation.
name: The name of the operation to be created
read_value: if True, will return something which evaluates to the new
value of the variable; if False will return the assign op.
Returns:
The updated variable. If `read_value` is false, instead returns None in
Eager mode and the assign op in graph mode."
9500,scatter_sub,tensorflow/tensorflow/python/ops/variables.py,630,method,"Subtracts `tf.IndexedSlices` from this variable.
Args:
sparse_delta: `tf.IndexedSlices` to be subtracted from this variable.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
The updated variable.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9501,scatter_add,tensorflow/tensorflow/python/ops/variables.py,646,method,"Adds `tf.IndexedSlices` to this variable.
Args:
sparse_delta: `tf.IndexedSlices` to be added to this variable.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
The updated variable.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9502,scatter_max,tensorflow/tensorflow/python/ops/variables.py,662,method,"Updates this variable with the max of `tf.IndexedSlices` and itself.
Args:
sparse_delta: `tf.IndexedSlices` to use as an argument of max with this
variable.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
The updated variable.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9503,scatter_min,tensorflow/tensorflow/python/ops/variables.py,679,method,"Updates this variable with the min of `tf.IndexedSlices` and itself.
Args:
sparse_delta: `tf.IndexedSlices` to use as an argument of min with this
variable.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
The updated variable.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9504,scatter_mul,tensorflow/tensorflow/python/ops/variables.py,696,method,"Multiply this variable by `tf.IndexedSlices`.
Args:
sparse_delta: `tf.IndexedSlices` to multiply this variable by.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
The updated variable.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9505,scatter_div,tensorflow/tensorflow/python/ops/variables.py,712,method,"Divide this variable by `tf.IndexedSlices`.
Args:
sparse_delta: `tf.IndexedSlices` to divide this variable by.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
The updated variable.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9506,scatter_update,tensorflow/tensorflow/python/ops/variables.py,728,method,"Assigns `tf.IndexedSlices` to this variable.
Args:
sparse_delta: `tf.IndexedSlices` to be assigned to this variable.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
The updated variable.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9507,batch_scatter_update,tensorflow/tensorflow/python/ops/variables.py,744,method,"Assigns `tf.IndexedSlices` to this variable batch-wise.
Analogous to `batch_gather`. This assumes that this variable and the
sparse_delta IndexedSlices have a series of leading dimensions that are the
same for all of them, and the updates are performed on the last dimension of
indices. In other words, the dimensions should be the following:
`num_prefix_dims = sparse_delta.indices.ndims - 1`
`batch_dim = num_prefix_dims + 1`
`sparse_delta.updates.shape = sparse_delta.indices.shape + var.shape[
batch_dim:]`
where
`sparse_delta.updates.shape[:num_prefix_dims]`
`== sparse_delta.indices.shape[:num_prefix_dims]`
`== var.shape[:num_prefix_dims]`
And the operation performed can be expressed as:
`var[i_1, ..., i_n,
sparse_delta.indices[i_1, ..., i_n, j]] = sparse_delta.updates[
i_1, ..., i_n, j]`
When sparse_delta.indices is a 1D tensor, this operation is equivalent to
`scatter_update`.
To avoid this operation one can looping over the first `ndims` of the
variable and using `scatter_update` on the subtensors that result of slicing
the first dimension. This is a valid option for `ndims = 1`, but less
efficient than this implementation.
Args:
sparse_delta: `tf.IndexedSlices` to be assigned to this variable.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
The updated variable.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9508,scatter_nd_sub,tensorflow/tensorflow/python/ops/variables.py,790,method,"Applies sparse subtraction to individual values or slices in a Variable.
Assuming the variable has rank `P` and `indices` is a `Tensor` of rank `Q`.
`indices` must be integer tensor, containing indices into self.
It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`.
The innermost dimension of `indices` (with length `K`) corresponds to
indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th
dimension of self.
`updates` is `Tensor` of rank `Q-1+P-K` with shape:
```
[d_0, ..., d_{Q-2}, self.shape[K], ..., self.shape[P-1]].
```
For example, say we want to add 4 scattered elements to a rank-1 tensor to
8 elements. In Python, that update would look like this:
```python
v = tf.Variable([1, 2, 3, 4, 5, 6, 7, 8])
indices = tf.constant([[4], [3], [1] ,[7]])
updates = tf.constant([9, 10, 11, 12])
op = v.scatter_nd_sub(indices, updates)
with tf.compat.v1.Session() as sess:
print sess.run(op)
```
The resulting update to v would look like this:
[1, -9, 3, -6, -6, 6, 7, -4]
See `tf.scatter_nd` for more details about how to make updates to
slices.
Args:
indices: The indices to be used in the operation.
updates: The values to be used in the operation.
name: the name of the operation.
Returns:
The updated variable."
9509,scatter_nd_add,tensorflow/tensorflow/python/ops/variables.py,837,method,"Applies sparse addition to individual values or slices in a Variable.
The Variable has rank `P` and `indices` is a `Tensor` of rank `Q`.
`indices` must be integer tensor, containing indices into self.
It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`.
The innermost dimension of `indices` (with length `K`) corresponds to
indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th
dimension of self.
`updates` is `Tensor` of rank `Q-1+P-K` with shape:
```
[d_0, ..., d_{Q-2}, self.shape[K], ..., self.shape[P-1]].
```
For example, say we want to add 4 scattered elements to a rank-1 tensor to
8 elements. In Python, that update would look like this:
```python
v = tf.Variable([1, 2, 3, 4, 5, 6, 7, 8])
indices = tf.constant([[4], [3], [1] ,[7]])
updates = tf.constant([9, 10, 11, 12])
add = v.scatter_nd_add(indices, updates)
with tf.compat.v1.Session() as sess:
print sess.run(add)
```
The resulting update to v would look like this:
[1, 13, 3, 14, 14, 6, 7, 20]
See `tf.scatter_nd` for more details about how to make updates to
slices.
Args:
indices: The indices to be used in the operation.
updates: The values to be used in the operation.
name: the name of the operation.
Returns:
The updated variable."
9510,scatter_nd_update,tensorflow/tensorflow/python/ops/variables.py,884,method,"Applies sparse assignment to individual values or slices in a Variable.
The Variable has rank `P` and `indices` is a `Tensor` of rank `Q`.
`indices` must be integer tensor, containing indices into self.
It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`.
The innermost dimension of `indices` (with length `K`) corresponds to
indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th
dimension of self.
`updates` is `Tensor` of rank `Q-1+P-K` with shape:
```
[d_0, ..., d_{Q-2}, self.shape[K], ..., self.shape[P-1]].
```
For example, say we want to add 4 scattered elements to a rank-1 tensor to
8 elements. In Python, that update would look like this:
```python
v = tf.Variable([1, 2, 3, 4, 5, 6, 7, 8])
indices = tf.constant([[4], [3], [1] ,[7]])
updates = tf.constant([9, 10, 11, 12])
op = v.scatter_nd_assign(indices, updates)
with tf.compat.v1.Session() as sess:
print sess.run(op)
```
The resulting update to v would look like this:
[1, 11, 3, 10, 9, 6, 7, 12]
See `tf.scatter_nd` for more details about how to make updates to
slices.
Args:
indices: The indices to be used in the operation.
updates: The values to be used in the operation.
name: the name of the operation.
Returns:
The updated variable."
9511,sparse_read,tensorflow/tensorflow/python/ops/variables.py,931,method,"Gather slices from params axis axis according to indices.
This function supports a subset of tf.gather, see tf.gather for details on
usage.
Args:
indices: The index `Tensor`. Must be one of the following types: `int32`,
`int64`. Must be in range `[0, params.shape[axis])`.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `params`."
9512,gather_nd,tensorflow/tensorflow/python/ops/variables.py,947,method,"Gather slices from `params` into a Tensor with shape specified by `indices`.
See tf.gather_nd for details.
Args:
indices: A `Tensor`. Must be one of the following types: `int32`, `int64`.
Index tensor.
name: A name for the operation (optional).
Returns:
A `Tensor`. Has the same type as `params`."
9513,count_up_to,tensorflow/tensorflow/python/ops/variables.py,963,method,"Increments this variable until it reaches `limit`.
When that Op is run it tries to increment the variable by `1`. If
incrementing the variable would bring it above `limit` then the Op raises
the exception `OutOfRangeError`.
If no error is raised, the Op outputs the value of the variable before
the increment.
This is essentially a shortcut for `count_up_to(self, limit)`.
Args:
limit: value at which incrementing the variable raises an error.
Returns:
A `Tensor` that will hold the variable value before the increment. If no
other Op modifies this variable, the values produced will all be
distinct."
9514,load,tensorflow/tensorflow/python/ops/variables.py,987,method,"Load new value into this variable.
Writes new value to variable's memory. Doesn't add ops to the graph.
This convenience method requires a session where the graph
containing this variable has been launched. If no session is
passed, the default session is used. See `tf.compat.v1.Session` for more
information on launching a graph and on sessions.
```python
v = tf.Variable([1, 2])
init = tf.compat.v1.global_variables_initializer()
with tf.compat.v1.Session() as sess:
sess.run(init)
# Usage passing the session explicitly.
v.load([2, 3], sess)
print(v.eval(sess)) # prints [2 3]
# Usage with the default session. The 'with' block
# above makes 'sess' the default session.
v.load([3, 4], sess)
print(v.eval()) # prints [3 4]
```
Args:
value: New variable value
session: The session to use to evaluate this variable. If none, the
default session is used.
Raises:
ValueError: Session is not passed and no default session"
9515,name,tensorflow/tensorflow/python/ops/variables.py,1128,method,The name of this variable.
9516,initializer,tensorflow/tensorflow/python/ops/variables.py,1145,method,The initializer operation for this variable.
9517,device,tensorflow/tensorflow/python/ops/variables.py,1150,method,The device of this variable.
9518,dtype,tensorflow/tensorflow/python/ops/variables.py,1155,method,The `DType` of this variable.
9519,op,tensorflow/tensorflow/python/ops/variables.py,1160,method,The `Operation` of this variable.
9520,graph,tensorflow/tensorflow/python/ops/variables.py,1165,method,The `Graph` of this variable.
9521,shape,tensorflow/tensorflow/python/ops/variables.py,1170,method,"The `TensorShape` of this variable.
Returns:
A `TensorShape`."
9522,get_shape,tensorflow/tensorflow/python/ops/variables.py,1178,method,Alias of `Variable.shape`.
9523,to_proto,tensorflow/tensorflow/python/ops/variables.py,1186,method,"Converts a `Variable` to a `VariableDef` protocol buffer.
Args:
export_scope: Optional `string`. Name scope to remove.
Returns:
A `VariableDef` protocol buffer, or `None` if the `Variable` is not
in the specified name scope."
9524,from_proto,tensorflow/tensorflow/python/ops/variables.py,1199,method,Returns a `Variable` object created from `variable_def`.
9525,experimental_ref,tensorflow/tensorflow/python/ops/variables.py,1215,method,
9526,ref,tensorflow/tensorflow/python/ops/variables.py,1218,method,"Returns a hashable reference object to this Variable.
The primary use case for this API is to put variables in a set/dictionary.
We can't put variables in a set/dictionary as `variable.__hash__()` is no
longer available starting Tensorflow 2.0.
The following will raise an exception starting 2.0
>>> x = tf.Variable(5)
>>> y = tf.Variable(10)
>>> z = tf.Variable(10)
>>> variable_set = {x, y, z}
Traceback (most recent call last):
...
TypeError: Variable is unhashable. Instead, use tensor.ref() as the key.
>>> variable_dict = {x: 'five', y: 'ten'}
Traceback (most recent call last):
...
TypeError: Variable is unhashable. Instead, use tensor.ref() as the key.
Instead, we can use `variable.ref()`.
>>> variable_set = {x.ref(), y.ref(), z.ref()}
>>> x.ref() in variable_set
True
>>> variable_dict = {x.ref(): 'five', y.ref(): 'ten', z.ref(): 'ten'}
>>> variable_dict[y.ref()]
'ten'
Also, the reference object provides `.deref()` function that returns the
original Variable.
>>> x = tf.Variable(5)
>>> x.ref().deref()
<tf.Variable 'Variable:0' shape=() dtype=int32, numpy=5>"
9527,spec,tensorflow/tensorflow/python/ops/variables.py,1309,method,Computes the spec string used for saving.
9528,to_proto,tensorflow/tensorflow/python/ops/variables.py,1316,method,"Returns a SaveSliceInfoDef() proto.
Args:
export_scope: Optional `string`. Name scope to remove.
Returns:
A `SaveSliceInfoDef` protocol buffer, or None if the `Variable` is not
in the specified name scope."
9529,VariableV1,tensorflow/tensorflow/python/ops/variables.py,1346,class,"See the [Variables Guide](https://tensorflow.org/guide/variables).
A variable maintains state in the graph across calls to `run()`. You add a
variable to the graph by constructing an instance of the class `Variable`.
The `Variable()` constructor requires an initial value for the variable,
which can be a `Tensor` of any type and shape. The initial value defines the
type and shape of the variable. After construction, the type and shape of
the variable are fixed. The value can be changed using one of the assign
methods.
If you want to change the shape of a variable later you have to use an
`assign` Op with `validate_shape=False`.
Just like any `Tensor`, variables created with `Variable()` can be used as
inputs for other Ops in the graph. Additionally, all the operators
overloaded for the `Tensor` class are carried over to variables, so you can
also add nodes to the graph by just doing arithmetic on variables.
```python
import tensorflow as tf
# Create a variable.
w = tf.Variable(<initial-value>, name=<optional-name>)
# Use the variable in the graph like any Tensor.
y = tf.matmul(w, ...another variable or tensor...)
# The overloaded operators are available too.
z = tf.sigmoid(w + y)
# Assign a new value to the variable with `assign()` or a related method.
w.assign(w + 1.0)
w.assign_add(1.0)
```
When you launch the graph, variables have to be explicitly initialized before
you can run Ops that use their value. You can initialize a variable by
running its *initializer op*, restoring the variable from a save file, or
simply running an `assign` Op that assigns a value to the variable. In fact,
the variable *initializer op* is just an `assign` Op that assigns the
variable's initial value to the variable itself.
```python
# Launch the graph in a session.
with tf.compat.v1.Session() as sess:
# Run the variable initializer.
sess.run(w.initializer)
# ...you now can run ops that use the value of 'w'...
```
The most common initialization pattern is to use the convenience function
`global_variables_initializer()` to add an Op to the graph that initializes
all the variables. You then run that Op after launching the graph.
```python
# Add an Op to initialize global variables.
init_op = tf.compat.v1.global_variables_initializer()
# Launch the graph in a session.
with tf.compat.v1.Session() as sess:
# Run the Op that initializes global variables.
sess.run(init_op)
# ...you can now run any Op that uses variable values...
```
If you need to create a variable with an initial value dependent on another
variable, use the other variable's `initialized_value()`. This ensures that
variables are initialized in the right order.
All variables are automatically collected in the graph where they are
created. By default, the constructor adds the new variable to the graph
collection `GraphKeys.GLOBAL_VARIABLES`. The convenience function
`global_variables()` returns the contents of that collection.
When building a machine learning model it is often convenient to distinguish
between variables holding the trainable model parameters and other variables
such as a `global step` variable used to count training steps. To make this
easier, the variable constructor supports a `trainable=<bool>` parameter. If
`True`, the new variable is also added to the graph collection
`GraphKeys.TRAINABLE_VARIABLES`. The convenience function
`trainable_variables()` returns the contents of this collection. The
various `Optimizer` classes use this collection as the default list of
variables to optimize.
WARNING: tf.Variable objects by default have a non-intuitive memory model. A
Variable is represented internally as a mutable Tensor which can
non-deterministically alias other Tensors in a graph. The set of operations
which consume a Variable and can lead to aliasing is undetermined and can
change across TensorFlow versions. Avoid writing code which relies on the
value of a Variable either changing or not changing as other operations
happen. For example, using Variable objects or simple functions thereof as
predicates in a `tf.cond` is dangerous and error-prone:
```
v = tf.Variable(True)
tf.cond(v, lambda: v.assign(False), my_false_fn) # Note: this is broken.
```
Here, adding `use_resource=True` when constructing the variable will
fix any nondeterminism issues:
```
v = tf.Variable(True, use_resource=True)
tf.cond(v, lambda: v.assign(False), my_false_fn)
```
To use the replacement for variables which does
not have these issues:
* Add `use_resource=True` when constructing `tf.Variable`;
* Call `tf.compat.v1.get_variable_scope().set_use_resource(True)` inside a
`tf.compat.v1.variable_scope` before the `tf.compat.v1.get_variable()` call."
9530,RefVariable,tensorflow/tensorflow/python/ops/variables.py,1556,class,Ref-based implementation of variables.
9531,value,tensorflow/tensorflow/python/ops/variables.py,1912,method,"Returns the last snapshot of this variable.
You usually do not need to call this method as all ops that need the value
of the variable call it automatically through a `convert_to_tensor()` call.
Returns a `Tensor` which holds the value of the variable. You can not
assign a new value to this tensor as it is not a reference to the variable.
To avoid copies, if the consumer of the returned value is on the same device
as the variable, this actually returns the live value of the variable, not
a copy. Updates to the variable are seen by the consumer. If the consumer
is on a different device it will get a copy of the variable.
Returns:
A `Tensor` containing the value of the variable."
9532,read_value,tensorflow/tensorflow/python/ops/variables.py,1931,method,"Returns the value of this variable, read in the current context.
Can be different from value() if it's on another device, with control
dependencies, etc.
Returns:
A `Tensor` containing the value of the variable."
9533,set_shape,tensorflow/tensorflow/python/ops/variables.py,1958,method,"Overrides the shape for this variable.
Args:
shape: the `TensorShape` representing the overridden shape."
9534,trainable,tensorflow/tensorflow/python/ops/variables.py,1968,method,
9535,synchronization,tensorflow/tensorflow/python/ops/variables.py,1972,method,
9536,aggregation,tensorflow/tensorflow/python/ops/variables.py,1976,method,
9537,eval,tensorflow/tensorflow/python/ops/variables.py,1979,method,"In a session, computes and returns the value of this variable.
This is not a graph construction method, it does not add ops to the graph.
This convenience method requires a session where the graph
containing this variable has been launched. If no session is
passed, the default session is used. See `tf.compat.v1.Session` for more
information on launching a graph and on sessions.
```python
v = tf.Variable([1, 2])
init = tf.compat.v1.global_variables_initializer()
with tf.compat.v1.Session() as sess:
sess.run(init)
# Usage passing the session explicitly.
print(v.eval(sess))
# Usage with the default session. The 'with' block
# above makes 'sess' the default session.
print(v.eval())
```
Args:
session: The session to use to evaluate this variable. If none, the
default session is used.
Returns:
A numpy `ndarray` with a copy of the value of this variable."
9538,initial_value,tensorflow/tensorflow/python/ops/variables.py,2012,method,"Returns the Tensor used as the initial value for the variable.
Note that this is different from `initialized_value()` which runs
the op that initializes the variable before returning its value.
This method returns the tensor that is used by the op that initializes
the variable.
Returns:
A `Tensor`."
9539,constraint,tensorflow/tensorflow/python/ops/variables.py,2026,method,"Returns the constraint function associated with this variable.
Returns:
The constraint function that was passed to the variable constructor.
Can be `None` if no constraint was passed."
9540,assign,tensorflow/tensorflow/python/ops/variables.py,2035,method,"Assigns a new value to the variable.
This is essentially a shortcut for `assign(self, value)`.
Args:
value: A `Tensor`. The new value for this variable.
use_locking: If `True`, use locking during the assignment.
name: The name of the operation to be created
read_value: if True, will return something which evaluates to the new
value of the variable; if False will return the assign op.
Returns:
A `Tensor` that will hold the new value of this variable after
the assignment has completed."
9541,assign_add,tensorflow/tensorflow/python/ops/variables.py,2057,method,"Adds a value to this variable.
This is essentially a shortcut for `assign_add(self, delta)`.
Args:
delta: A `Tensor`. The value to add to this variable.
use_locking: If `True`, use locking during the operation.
name: The name of the operation to be created
read_value: if True, will return something which evaluates to the new
value of the variable; if False will return the assign op.
Returns:
A `Tensor` that will hold the new value of this variable after
the addition has completed."
9542,assign_sub,tensorflow/tensorflow/python/ops/variables.py,2079,method,"Subtracts a value from this variable.
This is essentially a shortcut for `assign_sub(self, delta)`.
Args:
delta: A `Tensor`. The value to subtract from this variable.
use_locking: If `True`, use locking during the operation.
name: The name of the operation to be created
read_value: if True, will return something which evaluates to the new
value of the variable; if False will return the assign op.
Returns:
A `Tensor` that will hold the new value of this variable after
the subtraction has completed."
9543,scatter_sub,tensorflow/tensorflow/python/ops/variables.py,2101,method,"Subtracts `tf.IndexedSlices` from this variable.
Args:
sparse_delta: `tf.IndexedSlices` to be subtracted from this variable.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
A `Tensor` that will hold the new value of this variable after
the scattered subtraction has completed.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9544,scatter_add,tensorflow/tensorflow/python/ops/variables.py,2125,method,"Adds `tf.IndexedSlices` to this variable.
Args:
sparse_delta: `tf.IndexedSlices` to be added to this variable.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
A `Tensor` that will hold the new value of this variable after
the scattered addition has completed.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9545,scatter_max,tensorflow/tensorflow/python/ops/variables.py,2149,method,"Updates this variable with the max of `tf.IndexedSlices` and itself.
Args:
sparse_delta: `tf.IndexedSlices` to use as an argument of max with this
variable.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
A `Tensor` that will hold the new value of this variable after
the scattered maximization has completed.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9546,scatter_min,tensorflow/tensorflow/python/ops/variables.py,2174,method,"Updates this variable with the min of `tf.IndexedSlices` and itself.
Args:
sparse_delta: `tf.IndexedSlices` to use as an argument of min with this
variable.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
A `Tensor` that will hold the new value of this variable after
the scattered minimization has completed.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9547,scatter_mul,tensorflow/tensorflow/python/ops/variables.py,2199,method,"Multiply this variable by `tf.IndexedSlices`.
Args:
sparse_delta: `tf.IndexedSlices` to multiply this variable by.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
A `Tensor` that will hold the new value of this variable after
the scattered multiplication has completed.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9548,scatter_div,tensorflow/tensorflow/python/ops/variables.py,2223,method,"Divide this variable by `tf.IndexedSlices`.
Args:
sparse_delta: `tf.IndexedSlices` to divide this variable by.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
A `Tensor` that will hold the new value of this variable after
the scattered division has completed.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9549,scatter_update,tensorflow/tensorflow/python/ops/variables.py,2247,method,"Assigns `tf.IndexedSlices` to this variable.
Args:
sparse_delta: `tf.IndexedSlices` to be assigned to this variable.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
A `Tensor` that will hold the new value of this variable after
the scattered assignment has completed.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9550,batch_scatter_update,tensorflow/tensorflow/python/ops/variables.py,2271,method,"Assigns `tf.IndexedSlices` to this variable batch-wise.
Analogous to `batch_gather`. This assumes that this variable and the
sparse_delta IndexedSlices have a series of leading dimensions that are the
same for all of them, and the updates are performed on the last dimension of
indices. In other words, the dimensions should be the following:
`num_prefix_dims = sparse_delta.indices.ndims - 1`
`batch_dim = num_prefix_dims + 1`
`sparse_delta.updates.shape = sparse_delta.indices.shape + var.shape[
batch_dim:]`
where
`sparse_delta.updates.shape[:num_prefix_dims]`
`== sparse_delta.indices.shape[:num_prefix_dims]`
`== var.shape[:num_prefix_dims]`
And the operation performed can be expressed as:
`var[i_1, ..., i_n,
sparse_delta.indices[i_1, ..., i_n, j]] = sparse_delta.updates[
i_1, ..., i_n, j]`
When sparse_delta.indices is a 1D tensor, this operation is equivalent to
`scatter_update`.
To avoid this operation one can looping over the first `ndims` of the
variable and using `scatter_update` on the subtensors that result of slicing
the first dimension. This is a valid option for `ndims = 1`, but less
efficient than this implementation.
Args:
sparse_delta: `tf.IndexedSlices` to be assigned to this variable.
use_locking: If `True`, use locking during the operation.
name: the name of the operation.
Returns:
A `Tensor` that will hold the new value of this variable after
the scattered assignment has completed.
Raises:
TypeError: if `sparse_delta` is not an `IndexedSlices`."
9551,scatter_nd_sub,tensorflow/tensorflow/python/ops/variables.py,2323,method,"Applies sparse subtraction to individual values or slices in a Variable.
`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`.
`indices` must be integer tensor, containing indices into `ref`.
It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`.
The innermost dimension of `indices` (with length `K`) corresponds to
indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th
dimension of `ref`.
`updates` is `Tensor` of rank `Q-1+P-K` with shape:
```
[d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]].
```
For example, say we want to add 4 scattered elements to a rank-1 tensor to
8 elements. In Python, that update would look like this:
```python
ref = tf.Variable([1, 2, 3, 4, 5, 6, 7, 8])
indices = tf.constant([[4], [3], [1] ,[7]])
updates = tf.constant([9, 10, 11, 12])
op = ref.scatter_nd_sub(indices, updates)
with tf.compat.v1.Session() as sess:
print sess.run(op)
```
The resulting update to ref would look like this:
[1, -9, 3, -6, -6, 6, 7, -4]
See `tf.scatter_nd` for more details about how to make updates to
slices.
Args:
indices: The indices to be used in the operation.
updates: The values to be used in the operation.
name: the name of the operation.
Returns:
A `Tensor` that will hold the new value of this variable after
the scattered subtraction has completed."
9552,scatter_nd_add,tensorflow/tensorflow/python/ops/variables.py,2372,method,"Applies sparse addition to individual values or slices in a Variable.
`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`.
`indices` must be integer tensor, containing indices into `ref`.
It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`.
The innermost dimension of `indices` (with length `K`) corresponds to
indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th
dimension of `ref`.
`updates` is `Tensor` of rank `Q-1+P-K` with shape:
```
[d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]].
```
For example, say we want to add 4 scattered elements to a rank-1 tensor to
8 elements. In Python, that update would look like this:
```python
ref = tf.Variable([1, 2, 3, 4, 5, 6, 7, 8])
indices = tf.constant([[4], [3], [1] ,[7]])
updates = tf.constant([9, 10, 11, 12])
add = ref.scatter_nd_add(indices, updates)
with tf.compat.v1.Session() as sess:
print sess.run(add)
```
The resulting update to ref would look like this:
[1, 13, 3, 14, 14, 6, 7, 20]
See `tf.scatter_nd` for more details about how to make updates to
slices.
Args:
indices: The indices to be used in the operation.
updates: The values to be used in the operation.
name: the name of the operation.
Returns:
A `Tensor` that will hold the new value of this variable after
the scattered addition has completed."
9553,scatter_nd_update,tensorflow/tensorflow/python/ops/variables.py,2421,method,"Applies sparse assignment to individual values or slices in a Variable.
`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`.
`indices` must be integer tensor, containing indices into `ref`.
It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`.
The innermost dimension of `indices` (with length `K`) corresponds to
indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th
dimension of `ref`.
`updates` is `Tensor` of rank `Q-1+P-K` with shape:
```
[d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]].
```
For example, say we want to add 4 scattered elements to a rank-1 tensor to
8 elements. In Python, that update would look like this:
```python
ref = tf.Variable([1, 2, 3, 4, 5, 6, 7, 8])
indices = tf.constant([[4], [3], [1] ,[7]])
updates = tf.constant([9, 10, 11, 12])
op = ref.scatter_nd_update(indices, updates)
with tf.compat.v1.Session() as sess:
print sess.run(op)
```
The resulting update to ref would look like this:
[1, 11, 3, 10, 9, 6, 7, 12]
See `tf.scatter_nd` for more details about how to make updates to
slices.
Args:
indices: The indices to be used in the operation.
updates: The values to be used in the operation.
name: the name of the operation.
Returns:
A `Tensor` that will hold the new value of this variable after
the scattered assignment has completed."
9554,scatter_nd_max,tensorflow/tensorflow/python/ops/variables.py,2470,method,"Updates this variable with the max of `tf.IndexedSlices` and itself.
`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`.
`indices` must be integer tensor, containing indices into `ref`.
It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`.
The innermost dimension of `indices` (with length `K`) corresponds to
indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th
dimension of `ref`.
`updates` is `Tensor` of rank `Q-1+P-K` with shape:
```
[d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]].
```
See `tf.scatter_nd` for more details about how to make updates to
slices.
Args:
indices: The indices to be used in the operation.
updates: The values to be used in the operation.
name: the name of the operation.
Returns:
A `Tensor` that will hold the new value of this variable after
the scattered addition has completed."
9555,scatter_nd_min,tensorflow/tensorflow/python/ops/variables.py,2503,method,"Updates this variable with the min of `tf.IndexedSlices` and itself.
`ref` is a `Tensor` with rank `P` and `indices` is a `Tensor` of rank `Q`.
`indices` must be integer tensor, containing indices into `ref`.
It must be shape `[d_0, ..., d_{Q-2}, K]` where `0 < K <= P`.
The innermost dimension of `indices` (with length `K`) corresponds to
indices into elements (if `K = P`) or slices (if `K < P`) along the `K`th
dimension of `ref`.
`updates` is `Tensor` of rank `Q-1+P-K` with shape:
```
[d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]].
```
See `tf.scatter_nd` for more details about how to make updates to
slices.
Args:
indices: The indices to be used in the operation.
updates: The values to be used in the operation.
name: the name of the operation.
Returns:
A `Tensor` that will hold the new value of this variable after
the scattered addition has completed."
9556,count_up_to,tensorflow/tensorflow/python/ops/variables.py,2553,method,"Increments this variable until it reaches `limit`.
When that Op is run it tries to increment the variable by `1`. If
incrementing the variable would bring it above `limit` then the Op raises
the exception `OutOfRangeError`.
If no error is raised, the Op outputs the value of the variable before
the increment.
This is essentially a shortcut for `count_up_to(self, limit)`.
Args:
limit: value at which incrementing the variable raises an error.
Returns:
A `Tensor` that will hold the variable value before the increment. If no
other Op modifies this variable, the values produced will all be
distinct."
9557,name,tensorflow/tensorflow/python/ops/variables.py,2599,method,The name of this variable.
9558,initializer,tensorflow/tensorflow/python/ops/variables.py,2604,method,The initializer operation for this variable.
9559,device,tensorflow/tensorflow/python/ops/variables.py,2609,method,The device of this variable.
9560,dtype,tensorflow/tensorflow/python/ops/variables.py,2614,method,The `DType` of this variable.
9561,op,tensorflow/tensorflow/python/ops/variables.py,2619,method,The `Operation` of this variable.
9562,graph,tensorflow/tensorflow/python/ops/variables.py,2624,method,The `Graph` of this variable.
9563,shape,tensorflow/tensorflow/python/ops/variables.py,2634,method,"The `TensorShape` of this variable.
Returns:
A `TensorShape`."
9564,to_proto,tensorflow/tensorflow/python/ops/variables.py,2642,method,"Converts a `Variable` to a `VariableDef` protocol buffer.
Args:
export_scope: Optional `string`. Name scope to remove.
Returns:
A `VariableDef` protocol buffer, or `None` if the `Variable` is not
in the specified name scope."
9565,PartitionedVariable,tensorflow/tensorflow/python/ops/variables.py,2884,class,"A container for partitioned `Variable` objects.
@compatibility(eager) `tf.PartitionedVariable` is not compatible with
eager execution. Use `tf.Variable` instead which is compatible
with both eager execution and graph construction. See [the
TensorFlow Eager Execution
guide](https://www.tensorflow.org/guide/eager#variables_and_optimizers)
for details on how variables work in eager execution.
@end_compatibility"
9566,as_tensor,tensorflow/tensorflow/python/ops/variables.py,2994,method,"Returns the overall concatenated value as a `Tensor`.
The returned tensor will not inherit the control dependencies from the scope
where the value is used, which is similar to getting the value of
`Variable`.
Returns:
`Tensor` containing the concatenated value."
9567,name,tensorflow/tensorflow/python/ops/variables.py,3022,method,
9568,dtype,tensorflow/tensorflow/python/ops/variables.py,3026,method,
9569,shape,tensorflow/tensorflow/python/ops/variables.py,3030,method,
9570,get_shape,tensorflow/tensorflow/python/ops/variables.py,3039,method,
9571,assign,tensorflow/tensorflow/python/ops/variables.py,3074,method,
9572,assign_add,tensorflow/tensorflow/python/ops/variables.py,3082,method,
9573,assign_sub,tensorflow/tensorflow/python/ops/variables.py,3090,method,
9574,global_variables,tensorflow/tensorflow/python/ops/variables.py,3106,function,"Returns global variables.
Global variables are variables that are shared across machines in a
distributed environment. The `Variable()` constructor or `get_variable()`
automatically adds new variables to the graph collection
`GraphKeys.GLOBAL_VARIABLES`.
This convenience function returns the contents of that collection.
An alternative to global variables are local variables. See
`tf.compat.v1.local_variables`
Args:
scope: (Optional.) A string. If supplied, the resulting list is filtered to
include only items whose `name` attribute matches `scope` using
`re.match`. Items without a `name` attribute are never returned if a scope
is supplied. The choice of `re.match` means that a `scope` without special
tokens filters by prefix.
Returns:
A list of `Variable` objects."
9575,all_variables,tensorflow/tensorflow/python/ops/variables.py,3133,function,Use `tf.compat.v1.global_variables` instead.
9576,local_variables,tensorflow/tensorflow/python/ops/variables.py,3157,function,"Returns local variables.
Local variables - per process variables, usually not saved/restored to
checkpoint and used for temporary or intermediate values.
For example, they can be used as counters for metrics computation or
number of epochs this machine has read data.
The `tf.contrib.framework.local_variable()` function automatically adds the
new variable to `GraphKeys.LOCAL_VARIABLES`.
This convenience function returns the contents of that collection.
An alternative to local variables are global variables. See
`tf.compat.v1.global_variables`
Args:
scope: (Optional.) A string. If supplied, the resulting list is filtered to
include only items whose `name` attribute matches `scope` using
`re.match`. Items without a `name` attribute are never returned if a scope
is supplied. The choice of `re.match` means that a `scope` without special
tokens filters by prefix.
Returns:
A list of local `Variable` objects."
9577,model_variables,tensorflow/tensorflow/python/ops/variables.py,3185,function,"Returns all variables in the MODEL_VARIABLES collection.
Args:
scope: (Optional.) A string. If supplied, the resulting list is filtered to
include only items whose `name` attribute matches `scope` using
`re.match`. Items without a `name` attribute are never returned if a scope
is supplied. The choice of `re.match` means that a `scope` without special
tokens filters by prefix.
Returns:
A list of local Variable objects."
9578,trainable_variables,tensorflow/tensorflow/python/ops/variables.py,3202,function,"Returns all variables created with `trainable=True`.
When passed `trainable=True`, the `Variable()` constructor automatically
adds new variables to the graph collection
`GraphKeys.TRAINABLE_VARIABLES`. This convenience function returns the
contents of that collection.
Args:
scope: (Optional.) A string. If supplied, the resulting list is filtered to
include only items whose `name` attribute matches `scope` using
`re.match`. Items without a `name` attribute are never returned if a scope
is supplied. The choice of `re.match` means that a `scope` without special
tokens filters by prefix.
Returns:
A list of Variable objects."
9579,moving_average_variables,tensorflow/tensorflow/python/ops/variables.py,3224,function,"Returns all variables that maintain their moving averages.
If an `ExponentialMovingAverage` object is created and the `apply()`
method is called on a list of variables, these variables will
be added to the `GraphKeys.MOVING_AVERAGE_VARIABLES` collection.
This convenience function returns the contents of that collection.
Args:
scope: (Optional.) A string. If supplied, the resulting list is filtered to
include only items whose `name` attribute matches `scope` using
`re.match`. Items without a `name` attribute are never returned if a scope
is supplied. The choice of `re.match` means that a `scope` without special
tokens filters by prefix.
Returns:
A list of Variable objects."
9580,variables_initializer,tensorflow/tensorflow/python/ops/variables.py,3246,function,"Returns an Op that initializes a list of variables.
After you launch the graph in a session, you can run the returned Op to
initialize all the variables in `var_list`. This Op runs all the
initializers of the variables in `var_list` in parallel.
Calling `initialize_variables()` is equivalent to passing the list of
initializers to `Group()`.
If `var_list` is empty, however, the function still returns an Op that can
be run. That Op just has no effect.
Args:
var_list: List of `Variable` objects to initialize.
name: Optional name for the returned operation.
Returns:
An Op that run the initializers of all the specified variables."
9581,initialize_variables,tensorflow/tensorflow/python/ops/variables.py,3274,function,See `tf.compat.v1.variables_initializer`.
9582,global_variables_initializer,tensorflow/tensorflow/python/ops/variables.py,3280,function,"Returns an Op that initializes global variables.
This is just a shortcut for `variables_initializer(global_variables())`
Returns:
An Op that initializes global variables in the graph."
9583,initialize_all_variables,tensorflow/tensorflow/python/ops/variables.py,3296,function,See `tf.compat.v1.global_variables_initializer`.
9584,local_variables_initializer,tensorflow/tensorflow/python/ops/variables.py,3302,function,"Returns an Op that initializes all local variables.
This is just a shortcut for `variables_initializer(local_variables())`
Returns:
An Op that initializes all local variables in the graph."
9585,initialize_local_variables,tensorflow/tensorflow/python/ops/variables.py,3318,function,See `tf.compat.v1.local_variables_initializer`.
9586,is_variable_initialized,tensorflow/tensorflow/python/ops/variables.py,3325,function,"Tests if a variable has been initialized.
Args:
variable: A `Variable`.
Returns:
Returns a scalar boolean Tensor, `True` if the variable has been
initialized, `False` otherwise."
9587,assert_variables_initialized,tensorflow/tensorflow/python/ops/variables.py,3340,function,"Returns an Op to check if variables are initialized.
NOTE: This function is obsolete and will be removed in 6 months. Please
change your implementation to use `report_uninitialized_variables()`.
When run, the returned Op will raise the exception `FailedPreconditionError`
if any of the variables has not yet been initialized.
Note: This function is implemented by trying to fetch the values of the
variables. If one of the variables is not initialized a message may be
logged by the C++ runtime. This is expected.
Args:
var_list: List of `Variable` objects to check. Defaults to the value of
`global_variables().`
Returns:
An Op, or None if there are no variables."
9588,report_uninitialized_variables,tensorflow/tensorflow/python/ops/variables.py,3383,function,"Adds ops to list the names of uninitialized variables.
When run, it returns a 1-D tensor containing the names of uninitialized
variables if there are any, or an empty array if there are none.
Args:
var_list: List of `Variable` objects to check. Defaults to the value of
`global_variables() + local_variables()`
name: Optional name of the `Operation`.
Returns:
A 1-D tensor containing names of the uninitialized variables, or an empty
1-D tensor if there are no variables or no uninitialized variables."
9589,assert_broadcastable,tensorflow/tensorflow/python/ops/weights_broadcast_ops.py,63,function,"Asserts `weights` can be broadcast to `values`.
In `tf.losses` and `tf.metrics`, we support limited weight broadcasting. We
let weights be either scalar, or the same rank as the target values, with each
dimension either 1, or the same as the corresponding values dimension.
Args:
weights: `Tensor` of weights.
values: `Tensor` of values to which weights are applied.
Returns:
`Operation` raising `InvalidArgumentError` if `weights` has incorrect shape.
`no_op` if static checks determine `weights` has correct shape.
Raises:
ValueError: If static checks determine `weights` has incorrect shape."
9590,broadcast_weights,tensorflow/tensorflow/python/ops/weights_broadcast_ops.py,136,function,"Broadcast `weights` to the same shape as `values`.
This returns a version of `weights` following the same broadcast rules as
`mul(weights, values)`, but limited to the weights shapes allowed by
`assert_broadcastable`. When computing a weighted average, use this function
to broadcast `weights` before summing them; e.g.,
`reduce_sum(w * v) / reduce_sum(_broadcast_weights(w, v))`.
Args:
weights: `Tensor` whose shape is broadcastable to `values` according to the
rules of `assert_broadcastable`.
values: `Tensor` of any shape.
Returns:
`weights` broadcast to `values` shape according to the rules of
`assert_broadcastable`."
9591,while_loop,tensorflow/tensorflow/python/ops/while_v2.py,58,function,"Like tf.while_loop, except emits a single While op."
9592,rewrite_grad_indexed_slices,tensorflow/tensorflow/python/ops/while_v2_indexed_slices_rewriter.py,31,function,"Handles special case of IndexedSlices returned from while gradient.
Some gradient functions return IndexedSlices instead of a Tensor (e.g. the
gradient of Gather ops). When this happens in the gradient of a while body,
the resulting gradient body function will have mismatched inputs and outputs,
since the input is a single Tensor, but the IndexedSlices gets unnested into
three output Tensors.
This function fixes this by rewriting the gradient body to have three inputs
to match the three outputs, i.e., it effectively converts the input Tensor
into an input IndexedSlices. It also returns new `loop_vars` to reflect the
new inputs.
Args:
grads: the input gradient Tensors to the while gradient computation.
body_grad_graph: _WhileBodyGradFuncGraph.
loop_vars: list of Tensors. The inputs to body_grad_graph.
forward_inputs: list of Tensors. The (flat) inputs to the forward-pass
While op.
Returns:
The new loop_vars to pass to body_grad_graph."
9593,Bernoulli,tensorflow/tensorflow/python/ops/distributions/bernoulli.py,36,class,"Bernoulli distribution.
The Bernoulli distribution with `probs` parameter, i.e., the probability of a
`1` outcome (vs a `0` outcome)."
9594,logits,tensorflow/tensorflow/python/ops/distributions/bernoulli.py,104,method,Log-odds of a `1` outcome (vs `0`).
9595,probs,tensorflow/tensorflow/python/ops/distributions/bernoulli.py,109,method,Probability of a `1` outcome (vs `0`).
9596,Beta,tensorflow/tensorflow/python/ops/distributions/beta.py,51,class,"Beta distribution.
The Beta distribution is defined over the `(0, 1)` interval using parameters
`concentration1` (aka ""alpha"") and `concentration0` (aka ""beta"").
#### Mathematical Details
The probability density function (pdf) is,
```none
pdf(x; alpha, beta) = x**(alpha - 1) (1 - x)**(beta - 1) / Z
Z = Gamma(alpha) Gamma(beta) / Gamma(alpha + beta)
```
where:
* `concentration1 = alpha`,
* `concentration0 = beta`,
* `Z` is the normalization constant, and,
* `Gamma` is the [gamma function](
https://en.wikipedia.org/wiki/Gamma_function).
The concentration parameters represent mean total counts of a `1` or a `0`,
i.e.,
```none
concentration1 = alpha = mean * total_concentration
concentration0 = beta = (1. - mean) * total_concentration
```
where `mean` in `(0, 1)` and `total_concentration` is a positive real number
representing a mean `total_count = concentration1 + concentration0`.
Distribution parameters are automatically broadcast in all functions; see
examples for details.
Warning: The samples can be zero due to finite precision.
This happens more often when some of the concentrations are very small.
Make sure to round the samples to `np.finfo(dtype).tiny` before computing the
density.
Samples of this distribution are reparameterized (pathwise differentiable).
The derivatives are computed using the approach described in
(Figurnov et al., 2018).
#### Examples
```python
import tensorflow_probability as tfp
tfd = tfp.distributions
# Create a batch of three Beta distributions.
alpha = [1, 2, 3]
beta = [1, 2, 3]
dist = tfd.Beta(alpha, beta)
dist.sample([4, 5]) # Shape [4, 5, 3]
# `x` has three batch entries, each with two samples.
x = [[.1, .4, .5],
[.2, .3, .5]]
# Calculate the probability of each pair of samples under the corresponding
# distribution in `dist`.
dist.prob(x) # Shape [2, 3]
```
```python
# Create batch_shape=[2, 3] via parameter broadcast:
alpha = [[1.], [2]] # Shape [2, 1]
beta = [3., 4, 5] # Shape [3]
dist = tfd.Beta(alpha, beta)
# alpha broadcast as: [[1., 1, 1,],
# [2, 2, 2]]
# beta broadcast as: [[3., 4, 5],
# [3, 4, 5]]
# batch_Shape [2, 3]
dist.sample([4, 5]) # Shape [4, 5, 2, 3]
x = [.2, .3, .5]
# x will be broadcast as [[.2, .3, .5],
# [.2, .3, .5]],
# thus matching batch_shape [2, 3].
dist.prob(x) # Shape [2, 3]
```
Compute the gradients of samples w.r.t. the parameters:
```python
alpha = tf.constant(1.0)
beta = tf.constant(2.0)
dist = tfd.Beta(alpha, beta)
samples = dist.sample(5) # Shape [5]
loss = tf.reduce_mean(tf.square(samples)) # Arbitrary loss function
# Unbiased stochastic gradients of the loss function
grads = tf.gradients(loss, [alpha, beta])
```
References:
Implicit Reparameterization Gradients:
[Figurnov et al., 2018]
(http://papers.nips.cc/paper/7326-implicit-reparameterization-gradients)
([pdf]
(http://papers.nips.cc/paper/7326-implicit-reparameterization-gradients.pdf))"
9597,concentration1,tensorflow/tensorflow/python/ops/distributions/beta.py,221,method,Concentration parameter associated with a `1` outcome.
9598,concentration0,tensorflow/tensorflow/python/ops/distributions/beta.py,226,method,Concentration parameter associated with a `0` outcome.
9599,total_concentration,tensorflow/tensorflow/python/ops/distributions/beta.py,231,method,Sum of concentration parameters.
9600,BetaWithSoftplusConcentration,tensorflow/tensorflow/python/ops/distributions/beta.py,354,class,Beta with softplus transform of `concentration1` and `concentration0`.
9601,Bijector,tensorflow/tensorflow/python/ops/distributions/bijector_impl.py,136,class,"Interface for transformations of a `Distribution` sample.
Bijectors can be used to represent any differentiable and injective
(one to one) function defined on an open subset of `R^n`. Some non-injective
transformations are also supported (see ""Non Injective Transforms"" below).
#### Mathematical Details
A `Bijector` implements a [smooth covering map](
https://en.wikipedia.org/wiki/Local_diffeomorphism), i.e., a local
diffeomorphism such that every point in the target has a neighborhood evenly
covered by a map ([see also](
https://en.wikipedia.org/wiki/Covering_space#Covering_of_a_manifold)).
A `Bijector` is used by `TransformedDistribution` but can be generally used
for transforming a `Distribution` generated `Tensor`. A `Bijector` is
characterized by three operations:
1. Forward
Useful for turning one random outcome into another random outcome from a
different distribution.
2. Inverse
Useful for ""reversing"" a transformation to compute one probability in
terms of another.
3. `log_det_jacobian(x)`
""The log of the absolute value of the determinant of the matrix of all
first-order partial derivatives of the inverse function.""
Useful for inverting a transformation to compute one probability in terms
of another. Geometrically, the Jacobian determinant is the volume of the
transformation and is used to scale the probability.
We take the absolute value of the determinant before log to avoid NaN
values. Geometrically, a negative determinant corresponds to an
orientation-reversing transformation. It is ok for us to discard the sign
of the determinant because we only integrate everywhere-nonnegative
functions (probability densities) and the correct orientation is always the
one that produces a nonnegative integrand.
By convention, transformations of random variables are named in terms of the
forward transformation. The forward transformation creates samples, the
inverse is useful for computing probabilities.
#### Example Uses
- Basic properties:
```python
x = ... # A tensor.
# Evaluate forward transformation.
fwd_x = my_bijector.forward(x)
x == my_bijector.inverse(fwd_x)
x != my_bijector.forward(fwd_x) # Not equal because x != g(g(x)).
```
- Computing a log-likelihood:
```python
def transformed_log_prob(bijector, log_prob, x):
return (bijector.inverse_log_det_jacobian(x, event_ndims=0) +
log_prob(bijector.inverse(x)))
```
- Transforming a random outcome:
```python
def transformed_sample(bijector, x):
return bijector.forward(x)
```
#### Example Bijectors
- ""Exponential""
```none
Y = g(X) = exp(X)
X ~ Normal(0, 1) # Univariate.
```
Implies:
```none
g^{-1}(Y) = log(Y)
|Jacobian(g^{-1})(y)| = 1 / y
Y ~ LogNormal(0, 1), i.e.,
prob(Y=y) = |Jacobian(g^{-1})(y)| * prob(X=g^{-1}(y))
= (1 / y) Normal(log(y); 0, 1)
```
Here is an example of how one might implement the `Exp` bijector:
```python
class Exp(Bijector):
def __init__(self, validate_args=False, name=""exp""):
super(Exp, self).__init__(
validate_args=validate_args,
forward_min_event_ndims=0,
name=name)
def _forward(self, x):
return math_ops.exp(x)
def _inverse(self, y):
return math_ops.log(y)
def _inverse_log_det_jacobian(self, y):
return -self._forward_log_det_jacobian(self._inverse(y))
def _forward_log_det_jacobian(self, x):
# Notice that we needn't do any reducing, even when`event_ndims > 0`.
# The base Bijector class will handle reducing for us; it knows how
# to do so because we called `super` `__init__` with
# `forward_min_event_ndims = 0`.
return x
```
- ""Affine""
```none
Y = g(X) = sqrtSigma * X + mu
X ~ MultivariateNormal(0, I_d)
```
Implies:
```none
g^{-1}(Y) = inv(sqrtSigma) * (Y - mu)
|Jacobian(g^{-1})(y)| = det(inv(sqrtSigma))
Y ~ MultivariateNormal(mu, sqrtSigma) , i.e.,
prob(Y=y) = |Jacobian(g^{-1})(y)| * prob(X=g^{-1}(y))
= det(sqrtSigma)^(-d) *
MultivariateNormal(inv(sqrtSigma) * (y - mu); 0, I_d)
```
#### Min_event_ndims and Naming
Bijectors are named for the dimensionality of data they act on (i.e. without
broadcasting). We can think of bijectors having an intrinsic `min_event_ndims`
, which is the minimum number of dimensions for the bijector act on. For
instance, a Cholesky decomposition requires a matrix, and hence
`min_event_ndims=2`.
Some examples:
`AffineScalar: min_event_ndims=0`
`Affine: min_event_ndims=1`
`Cholesky: min_event_ndims=2`
`Exp: min_event_ndims=0`
`Sigmoid: min_event_ndims=0`
`SoftmaxCentered: min_event_ndims=1`
Note the difference between `Affine` and `AffineScalar`. `AffineScalar`
operates on scalar events, whereas `Affine` operates on vector-valued events.
More generally, there is a `forward_min_event_ndims` and an
`inverse_min_event_ndims`. In most cases, these will be the same.
However, for some shape changing bijectors, these will be different
(e.g. a bijector which pads an extra dimension at the end, might have
`forward_min_event_ndims=0` and `inverse_min_event_ndims=1`.
#### Jacobian Determinant
The Jacobian determinant is a reduction over `event_ndims - min_event_ndims`
(`forward_min_event_ndims` for `forward_log_det_jacobian` and
`inverse_min_event_ndims` for `inverse_log_det_jacobian`).
To see this, consider the `Exp` `Bijector` applied to a `Tensor` which has
sample, batch, and event (S, B, E) shape semantics. Suppose the `Tensor`'s
partitioned-shape is `(S=[4], B=[2], E=[3, 3])`. The shape of the `Tensor`
returned by `forward` and `inverse` is unchanged, i.e., `[4, 2, 3, 3]`.
However the shape returned by `inverse_log_det_jacobian` is `[4, 2]` because
the Jacobian determinant is a reduction over the event dimensions.
Another example is the `Affine` `Bijector`. Because `min_event_ndims = 1`, the
Jacobian determinant reduction is over `event_ndims - 1`.
It is sometimes useful to implement the inverse Jacobian determinant as the
negative forward Jacobian determinant. For example,
```python
def _inverse_log_det_jacobian(self, y):
return -self._forward_log_det_jac(self._inverse(y)) # Note negation.
```
The correctness of this approach can be seen from the following claim.
- Claim:
Assume `Y = g(X)` is a bijection whose derivative exists and is nonzero
for its domain, i.e., `dY/dX = d/dX g(X) != 0`. Then:
```none
(log o det o jacobian o g^{-1})(Y) = -(log o det o jacobian o g)(X)
```
- Proof:
From the bijective, nonzero differentiability of `g`, the
[inverse function theorem](
https://en.wikipedia.org/wiki/Inverse_function_theorem)
implies `g^{-1}` is differentiable in the image of `g`.
Applying the chain rule to `y = g(x) = g(g^{-1}(y))` yields
`I = g'(g^{-1}(y))*g^{-1}'(y)`.
The same theorem also implies `g^{-1}'` is non-singular therefore:
`inv[ g'(g^{-1}(y)) ] = g^{-1}'(y)`.
The claim follows from [properties of determinant](
https://en.wikipedia.org/wiki/Determinant#Multiplicativity_and_matrix_groups).
Generally its preferable to directly implement the inverse Jacobian
determinant. This should have superior numerical stability and will often
share subgraphs with the `_inverse` implementation.
#### Is_constant_jacobian
Certain bijectors will have constant jacobian matrices. For instance, the
`Affine` bijector encodes multiplication by a matrix plus a shift, with
jacobian matrix, the same aforementioned matrix.
`is_constant_jacobian` encodes the fact that the jacobian matrix is constant.
The semantics of this argument are the following:
* Repeated calls to ""log_det_jacobian"" functions with the same
`event_ndims` (but not necessarily same input), will return the first
computed jacobian (because the matrix is constant, and hence is input
independent).
* `log_det_jacobian` implementations are merely broadcastable to the true
`log_det_jacobian` (because, again, the jacobian matrix is input
independent). Specifically, `log_det_jacobian` is implemented as the
log jacobian determinant for a single input.
```python
class Identity(Bijector):
def __init__(self, validate_args=False, name=""identity""):
super(Identity, self).__init__(
is_constant_jacobian=True,
validate_args=validate_args,
forward_min_event_ndims=0,
name=name)
def _forward(self, x):
return x
def _inverse(self, y):
return y
def _inverse_log_det_jacobian(self, y):
return -self._forward_log_det_jacobian(self._inverse(y))
def _forward_log_det_jacobian(self, x):
# The full log jacobian determinant would be array_ops.zero_like(x).
# However, we circumvent materializing that, since the jacobian
# calculation is input independent, and we specify it for one input.
return constant_op.constant(0., x.dtype.base_dtype)
```
#### Subclass Requirements
- Subclasses typically implement:
- `_forward`,
- `_inverse`,
- `_inverse_log_det_jacobian`,
- `_forward_log_det_jacobian` (optional).
The `_forward_log_det_jacobian` is called when the bijector is inverted via
the `Invert` bijector. If undefined, a slightly less efficiently
calculation, `-1 * _inverse_log_det_jacobian`, is used.
If the bijector changes the shape of the input, you must also implement:
- _forward_event_shape_tensor,
- _forward_event_shape (optional),
- _inverse_event_shape_tensor,
- _inverse_event_shape (optional).
By default the event-shape is assumed unchanged from input.
- If the `Bijector`'s use is limited to `TransformedDistribution` (or friends
like `QuantizedDistribution`) then depending on your use, you may not need
to implement all of `_forward` and `_inverse` functions.
Examples:
1. Sampling (e.g., `sample`) only requires `_forward`.
2. Probability functions (e.g., `prob`, `cdf`, `survival`) only require
`_inverse` (and related).
3. Only calling probability functions on the output of `sample` means
`_inverse` can be implemented as a cache lookup.
See ""Example Uses"" [above] which shows how these functions are used to
transform a distribution. (Note: `_forward` could theoretically be
implemented as a cache lookup but this would require controlling the
underlying sample generation mechanism.)
#### Non Injective Transforms
**WARNING** Handing of non-injective transforms is subject to change.
Non injective maps `g` are supported, provided their domain `D` can be
partitioned into `k` disjoint subsets, `Union{D1, ..., Dk}`, such that,
ignoring sets of measure zero, the restriction of `g` to each subset is a
differentiable bijection onto `g(D)`. In particular, this implies that for
`y in g(D)`, the set inverse, i.e. `g^{-1}(y) = {x in D : g(x) = y}`, always
contains exactly `k` distinct points.
The property, `_is_injective` is set to `False` to indicate that the bijector
is not injective, yet satisfies the above condition.
The usual bijector API is modified in the case `_is_injective is False` (see
method docstrings for specifics). Here we show by example the `AbsoluteValue`
bijector. In this case, the domain `D = (-inf, inf)`, can be partitioned
into `D1 = (-inf, 0)`, `D2 = {0}`, and `D3 = (0, inf)`. Let `gi` be the
restriction of `g` to `Di`, then both `g1` and `g3` are bijections onto
`(0, inf)`, with `g1^{-1}(y) = -y`, and `g3^{-1}(y) = y`. We will use
`g1` and `g3` to define bijector methods over `D1` and `D3`. `D2 = {0}` is
an oddball in that `g2` is one to one, and the derivative is not well defined.
Fortunately, when considering transformations of probability densities
(e.g. in `TransformedDistribution`), sets of measure zero have no effect in
theory, and only a small effect in 32 or 64 bit precision. For that reason,
we define `inverse(0)` and `inverse_log_det_jacobian(0)` both as `[0, 0]`,
which is convenient and results in a left-semicontinuous pdf.
```python
abs = tfp.distributions.bijectors.AbsoluteValue()
abs.forward(-1.)
==> 1.
abs.forward(1.)
==> 1.
abs.inverse(1.)
==> (-1., 1.)
# The |dX/dY| is constant, == 1. So Log|dX/dY| == 0.
abs.inverse_log_det_jacobian(1., event_ndims=0)
==> (0., 0.)
# Special case handling of 0.
abs.inverse(0.)
==> (0., 0.)
abs.inverse_log_det_jacobian(0., event_ndims=0)
==> (0., 0.)
```"
9602,graph_parents,tensorflow/tensorflow/python/ops/distributions/bijector_impl.py,591,method,Returns this `Bijector`'s graph_parents as a Python list.
9603,forward_min_event_ndims,tensorflow/tensorflow/python/ops/distributions/bijector_impl.py,596,method,Returns the minimal number of dimensions bijector.forward operates on.
9604,inverse_min_event_ndims,tensorflow/tensorflow/python/ops/distributions/bijector_impl.py,601,method,Returns the minimal number of dimensions bijector.inverse operates on.
9605,is_constant_jacobian,tensorflow/tensorflow/python/ops/distributions/bijector_impl.py,606,method,"Returns true iff the Jacobian matrix is not a function of x.
Note: Jacobian matrix is either constant for both forward and inverse or
neither.
Returns:
is_constant_jacobian: Python `bool`."
9606,validate_args,tensorflow/tensorflow/python/ops/distributions/bijector_impl.py,634,method,Returns True if Tensor arguments will be validated.
9607,dtype,tensorflow/tensorflow/python/ops/distributions/bijector_impl.py,639,method,dtype of `Tensor`s transformable by this distribution.
9608,name,tensorflow/tensorflow/python/ops/distributions/bijector_impl.py,644,method,Returns the string name of this `Bijector`.
9609,forward_event_shape_tensor,tensorflow/tensorflow/python/ops/distributions/bijector_impl.py,653,method,"Shape of a single sample from a single batch as an `int32` 1D `Tensor`.
Args:
input_shape: `Tensor`, `int32` vector indicating event-portion shape
passed into `forward` function.
name: name to give to the op
Returns:
forward_event_shape_tensor: `Tensor`, `int32` vector indicating
event-portion shape after applying `forward`."
9610,forward_event_shape,tensorflow/tensorflow/python/ops/distributions/bijector_impl.py,677,method,"Shape of a single sample from a single batch as a `TensorShape`.
Same meaning as `forward_event_shape_tensor`. May be only partially defined.
Args:
input_shape: `TensorShape` indicating event-portion shape passed into
`forward` function.
Returns:
forward_event_shape_tensor: `TensorShape` indicating event-portion shape
after applying `forward`. Possibly unknown."
9611,inverse_event_shape_tensor,tensorflow/tensorflow/python/ops/distributions/bijector_impl.py,697,method,"Shape of a single sample from a single batch as an `int32` 1D `Tensor`.
Args:
output_shape: `Tensor`, `int32` vector indicating event-portion shape
passed into `inverse` function.
name: name to give to the op
Returns:
inverse_event_shape_tensor: `Tensor`, `int32` vector indicating
event-portion shape after applying `inverse`."
9612,inverse_event_shape,tensorflow/tensorflow/python/ops/distributions/bijector_impl.py,721,method,"Shape of a single sample from a single batch as a `TensorShape`.
Same meaning as `inverse_event_shape_tensor`. May be only partially defined.
Args:
output_shape: `TensorShape` indicating event-portion shape passed into
`inverse` function.
Returns:
inverse_event_shape_tensor: `TensorShape` indicating event-portion shape
after applying `inverse`. Possibly unknown."
9613,forward,tensorflow/tensorflow/python/ops/distributions/bijector_impl.py,753,method,"Returns the forward `Bijector` evaluation, i.e., X = g(Y).
Args:
x: `Tensor`. The input to the ""forward"" evaluation.
name: The name to give this op.
Returns:
`Tensor`.
Raises:
TypeError: if `self.dtype` is specified and `x.dtype` is not
`self.dtype`.
NotImplementedError: if `_forward` is not implemented."
9614,inverse,tensorflow/tensorflow/python/ops/distributions/bijector_impl.py,787,method,"Returns the inverse `Bijector` evaluation, i.e., X = g^{-1}(Y).
Args:
y: `Tensor`. The input to the ""inverse"" evaluation.
name: The name to give this op.
Returns:
`Tensor`, if this bijector is injective.
If not injective, returns the k-tuple containing the unique
`k` points `(x1, ..., xk)` such that `g(xi) = y`.
Raises:
TypeError: if `self.dtype` is specified and `y.dtype` is not
`self.dtype`.
NotImplementedError: if `_inverse` is not implemented."
9615,inverse_log_det_jacobian,tensorflow/tensorflow/python/ops/distributions/bijector_impl.py,871,method,"Returns the (log o det o Jacobian o inverse)(y).
Mathematically, returns: `log(det(dX/dY))(Y)`. (Recall that: `X=g^{-1}(Y)`.)
Note that `forward_log_det_jacobian` is the negative of this function,
evaluated at `g^{-1}(y)`.
Args:
y: `Tensor`. The input to the ""inverse"" Jacobian determinant evaluation.
event_ndims: Number of dimensions in the probabilistic events being
transformed. Must be greater than or equal to
`self.inverse_min_event_ndims`. The result is summed over the final
dimensions to produce a scalar Jacobian determinant for each event,
i.e. it has shape `y.shape.ndims - event_ndims` dimensions.
name: The name to give this op.
Returns:
`Tensor`, if this bijector is injective.
If not injective, returns the tuple of local log det
Jacobians, `log(det(Dg_i^{-1}(y)))`, where `g_i` is the restriction
of `g` to the `ith` partition `Di`.
Raises:
TypeError: if `self.dtype` is specified and `y.dtype` is not
`self.dtype`.
NotImplementedError: if `_inverse_log_det_jacobian` is not implemented."
9616,forward_log_det_jacobian,tensorflow/tensorflow/python/ops/distributions/bijector_impl.py,973,method,"Returns both the forward_log_det_jacobian.
Args:
x: `Tensor`. The input to the ""forward"" Jacobian determinant evaluation.
event_ndims: Number of dimensions in the probabilistic events being
transformed. Must be greater than or equal to
`self.forward_min_event_ndims`. The result is summed over the final
dimensions to produce a scalar Jacobian determinant for each event,
i.e. it has shape `x.shape.ndims - event_ndims` dimensions.
name: The name to give this op.
Returns:
`Tensor`, if this bijector is injective.
If not injective this is not implemented.
Raises:
TypeError: if `self.dtype` is specified and `y.dtype` is not
`self.dtype`.
NotImplementedError: if neither `_forward_log_det_jacobian`
nor {`_inverse`, `_inverse_log_det_jacobian`} are implemented, or
this is a non-injective bijector."
9617,camel_to_snake,tensorflow/tensorflow/python/ops/distributions/bijector_impl.py,581,method,
9618,assert_finite,tensorflow/tensorflow/python/ops/distributions/bijector_test_util.py,28,function,
9619,assert_strictly_increasing,tensorflow/tensorflow/python/ops/distributions/bijector_test_util.py,33,function,
9620,assert_strictly_decreasing,tensorflow/tensorflow/python/ops/distributions/bijector_test_util.py,37,function,
9621,assert_strictly_monotonic,tensorflow/tensorflow/python/ops/distributions/bijector_test_util.py,41,function,
9622,assert_scalar_congruency,tensorflow/tensorflow/python/ops/distributions/bijector_test_util.py,48,function,"Assert `bijector`'s forward/inverse/inverse_log_det_jacobian are congruent.
We draw samples `X ~ U(lower_x, upper_x)`, then feed these through the
`bijector` in order to check that:
1. the forward is strictly monotonic.
2. the forward/inverse methods are inverses of each other.
3. the jacobian is the correct change of measure.
This can only be used for a Bijector mapping open subsets of the real line
to themselves. This is due to the fact that this test compares the `prob`
before/after transformation with the Lebesgue measure on the line.
Args:
bijector: Instance of Bijector
lower_x: Python scalar.
upper_x: Python scalar. Must have `lower_x < upper_x`, and both must be in
the domain of the `bijector`. The `bijector` should probably not produce
huge variation in values in the interval `(lower_x, upper_x)`, or else
the variance based check of the Jacobian will require small `rtol` or
huge `n`.
n: Number of samples to draw for the checks.
rtol: Positive number. Used for the Jacobian check.
sess: `tf.compat.v1.Session`. Defaults to the default session.
Raises:
AssertionError: If tests fail."
9623,assert_bijective_and_finite,tensorflow/tensorflow/python/ops/distributions/bijector_test_util.py,163,function,"Assert that forward/inverse (along with jacobians) are inverses and finite.
It is recommended to use x and y values that are very very close to the edge
of the Bijector's domain.
Args:
bijector: A Bijector instance.
x: np.array of values in the domain of bijector.forward.
y: np.array of values in the domain of bijector.inverse.
event_ndims: Integer describing the number of event dimensions this bijector
operates on.
atol: Absolute tolerance.
rtol: Relative tolerance.
sess: TensorFlow session. Defaults to the default session.
Raises:
AssertionError: If tests fail."
9624,Categorical,tensorflow/tensorflow/python/ops/distributions/categorical.py,63,class,"Categorical distribution.
The Categorical distribution is parameterized by either probabilities or
log-probabilities of a set of `K` classes. It is defined over the integers
`{0, 1, ..., K}`.
The Categorical distribution is closely related to the `OneHotCategorical` and
`Multinomial` distributions. The Categorical distribution can be intuited as
generating samples according to `argmax{ OneHotCategorical(probs) }` itself
being identical to `argmax{ Multinomial(probs, total_count=1) }`.
#### Mathematical Details
The probability mass function (pmf) is,
```none
pmf(k; pi) = prod_j pi_j**[k == j]
```
#### Pitfalls
The number of classes, `K`, must not exceed:
- the largest integer representable by `self.dtype`, i.e.,
`2**(mantissa_bits+1)` (IEEE 754),
- the maximum `Tensor` index, i.e., `2**31-1`.
In other words,
```python
K <= min(2**31-1, {
tf.float16: 2**11,
tf.float32: 2**24,
tf.float64: 2**53 }[param.dtype])
```
Note: This condition is validated only when `self.validate_args = True`.
#### Examples
Creates a 3-class distribution with the 2nd class being most likely.
```python
dist = Categorical(probs=[0.1, 0.5, 0.4])
n = 1e4
empirical_prob = tf.cast(
tf.histogram_fixed_width(
dist.sample(int(n)),
[0., 2],
nbins=3),
dtype=tf.float32) / n
# ==> array([ 0.1005, 0.5037, 0.3958], dtype=float32)
```
Creates a 3-class distribution with the 2nd class being most likely.
Parameterized by [logits](https://en.wikipedia.org/wiki/Logit) rather than
probabilities.
```python
dist = Categorical(logits=np.log([0.1, 0.5, 0.4])
n = 1e4
empirical_prob = tf.cast(
tf.histogram_fixed_width(
dist.sample(int(n)),
[0., 2],
nbins=3),
dtype=tf.float32) / n
# ==> array([0.1045, 0.5047, 0.3908], dtype=float32)
```
Creates a 3-class distribution with the 3rd class being most likely.
The distribution functions can be evaluated on counts.
```python
# counts is a scalar.
p = [0.1, 0.4, 0.5]
dist = Categorical(probs=p)
dist.prob(0) # Shape []
# p will be broadcast to [[0.1, 0.4, 0.5], [0.1, 0.4, 0.5]] to match counts.
counts = [1, 0]
dist.prob(counts) # Shape [2]
# p will be broadcast to shape [3, 5, 7, 3] to match counts.
counts = [[...]] # Shape [5, 7, 3]
dist.prob(counts) # Shape [5, 7, 3]
```"
9625,event_size,tensorflow/tensorflow/python/ops/distributions/categorical.py,245,method,Scalar `int32` tensor: the number of classes.
9626,logits,tensorflow/tensorflow/python/ops/distributions/categorical.py,250,method,Vector of coordinatewise logits.
9627,probs,tensorflow/tensorflow/python/ops/distributions/categorical.py,255,method,Vector of coordinatewise probabilities.
9628,Dirichlet,tensorflow/tensorflow/python/ops/distributions/dirichlet.py,49,class,"Dirichlet distribution.
The Dirichlet distribution is defined over the
[`(k-1)`-simplex](https://en.wikipedia.org/wiki/Simplex) using a positive,
length-`k` vector `concentration` (`k > 1`). The Dirichlet is identically the
Beta distribution when `k = 2`.
#### Mathematical Details
The Dirichlet is a distribution over the open `(k-1)`-simplex, i.e.,
```none
S^{k-1} = { (x_0, ..., x_{k-1}) in R^k : sum_j x_j = 1 and all_j x_j > 0 }.
```
The probability density function (pdf) is,
```none
pdf(x; alpha) = prod_j x_j**(alpha_j - 1) / Z
Z = prod_j Gamma(alpha_j) / Gamma(sum_j alpha_j)
```
where:
* `x in S^{k-1}`, i.e., the `(k-1)`-simplex,
* `concentration = alpha = [alpha_0, ..., alpha_{k-1}]`, `alpha_j > 0`,
* `Z` is the normalization constant aka the [multivariate beta function](
https://en.wikipedia.org/wiki/Beta_function#Multivariate_beta_function),
and,
* `Gamma` is the [gamma function](
https://en.wikipedia.org/wiki/Gamma_function).
The `concentration` represents mean total counts of class occurrence, i.e.,
```none
concentration = alpha = mean * total_concentration
```
where `mean` in `S^{k-1}` and `total_concentration` is a positive real number
representing a mean total count.
Distribution parameters are automatically broadcast in all functions; see
examples for details.
Warning: Some components of the samples can be zero due to finite precision.
This happens more often when some of the concentrations are very small.
Make sure to round the samples to `np.finfo(dtype).tiny` before computing the
density.
Samples of this distribution are reparameterized (pathwise differentiable).
The derivatives are computed using the approach described in
(Figurnov et al., 2018).
#### Examples
```python
import tensorflow_probability as tfp
tfd = tfp.distributions
# Create a single trivariate Dirichlet, with the 3rd class being three times
# more frequent than the first. I.e., batch_shape=[], event_shape=[3].
alpha = [1., 2, 3]
dist = tfd.Dirichlet(alpha)
dist.sample([4, 5]) # shape: [4, 5, 3]
# x has one sample, one batch, three classes:
x = [.2, .3, .5] # shape: [3]
dist.prob(x) # shape: []
# x has two samples from one batch:
x = [[.1, .4, .5],
[.2, .3, .5]]
dist.prob(x) # shape: [2]
# alpha will be broadcast to shape [5, 7, 3] to match x.
x = [[...]] # shape: [5, 7, 3]
dist.prob(x) # shape: [5, 7]
```
```python
# Create batch_shape=[2], event_shape=[3]:
alpha = [[1., 2, 3],
[4, 5, 6]] # shape: [2, 3]
dist = tfd.Dirichlet(alpha)
dist.sample([4, 5]) # shape: [4, 5, 2, 3]
x = [.2, .3, .5]
# x will be broadcast as [[.2, .3, .5],
# [.2, .3, .5]],
# thus matching batch_shape [2, 3].
dist.prob(x) # shape: [2]
```
Compute the gradients of samples w.r.t. the parameters:
```python
alpha = tf.constant([1.0, 2.0, 3.0])
dist = tfd.Dirichlet(alpha)
samples = dist.sample(5) # Shape [5, 3]
loss = tf.reduce_mean(tf.square(samples)) # Arbitrary loss function
# Unbiased stochastic gradients of the loss function
grads = tf.gradients(loss, alpha)
```
References:
Implicit Reparameterization Gradients:
[Figurnov et al., 2018]
(http://papers.nips.cc/paper/7326-implicit-reparameterization-gradients)
([pdf]
(http://papers.nips.cc/paper/7326-implicit-reparameterization-gradients.pdf))"
9629,concentration,tensorflow/tensorflow/python/ops/distributions/dirichlet.py,213,method,Concentration parameter; expected counts for that coordinate.
9630,total_concentration,tensorflow/tensorflow/python/ops/distributions/dirichlet.py,218,method,Sum of last dim of concentration parameter.
9631,DirichletMultinomial,tensorflow/tensorflow/python/ops/distributions/dirichlet_multinomial.py,55,class,"Dirichlet-Multinomial compound distribution.
The Dirichlet-Multinomial distribution is parameterized by a (batch of)
length-`K` `concentration` vectors (`K > 1`) and a `total_count` number of
trials, i.e., the number of trials per draw from the DirichletMultinomial. It
is defined over a (batch of) length-`K` vector `counts` such that
`tf.reduce_sum(counts, -1) = total_count`. The Dirichlet-Multinomial is
identically the Beta-Binomial distribution when `K = 2`.
#### Mathematical Details
The Dirichlet-Multinomial is a distribution over `K`-class counts, i.e., a
length-`K` vector of non-negative integer `counts = n = [n_0, ..., n_{K-1}]`.
The probability mass function (pmf) is,
```none
pmf(n; alpha, N) = Beta(alpha + n) / (prod_j n_j!) / Z
Z = Beta(alpha) / N!
```
where:
* `concentration = alpha = [alpha_0, ..., alpha_{K-1}]`, `alpha_j > 0`,
* `total_count = N`, `N` a positive integer,
* `N!` is `N` factorial, and,
* `Beta(x) = prod_j Gamma(x_j) / Gamma(sum_j x_j)` is the
[multivariate beta function](
https://en.wikipedia.org/wiki/Beta_function#Multivariate_beta_function),
and,
* `Gamma` is the [gamma function](
https://en.wikipedia.org/wiki/Gamma_function).
Dirichlet-Multinomial is a [compound distribution](
https://en.wikipedia.org/wiki/Compound_probability_distribution), i.e., its
samples are generated as follows.
1. Choose class probabilities:
`probs = [p_0,...,p_{K-1}] ~ Dir(concentration)`
2. Draw integers:
`counts = [n_0,...,n_{K-1}] ~ Multinomial(total_count, probs)`
The last `concentration` dimension parametrizes a single Dirichlet-Multinomial
distribution. When calling distribution functions (e.g., `dist.prob(counts)`),
`concentration`, `total_count` and `counts` are broadcast to the same shape.
The last dimension of `counts` corresponds single Dirichlet-Multinomial
distributions.
Distribution parameters are automatically broadcast in all functions; see
examples for details.
#### Pitfalls
The number of classes, `K`, must not exceed:
- the largest integer representable by `self.dtype`, i.e.,
`2**(mantissa_bits+1)` (IEE754),
- the maximum `Tensor` index, i.e., `2**31-1`.
In other words,
```python
K <= min(2**31-1, {
tf.float16: 2**11,
tf.float32: 2**24,
tf.float64: 2**53 }[param.dtype])
```
Note: This condition is validated only when `self.validate_args = True`.
#### Examples
```python
alpha = [1., 2., 3.]
n = 2.
dist = DirichletMultinomial(n, alpha)
```
Creates a 3-class distribution, with the 3rd class is most likely to be
drawn.
The distribution functions can be evaluated on counts.
```python
# counts same shape as alpha.
counts = [0., 0., 2.]
dist.prob(counts) # Shape []
# alpha will be broadcast to [[1., 2., 3.], [1., 2., 3.]] to match counts.
counts = [[1., 1., 0.], [1., 0., 1.]]
dist.prob(counts) # Shape [2]
# alpha will be broadcast to shape [5, 7, 3] to match counts.
counts = [[...]] # Shape [5, 7, 3]
dist.prob(counts) # Shape [5, 7]
```
Creates a 2-batch of 3-class distributions.
```python
alpha = [[1., 2., 3.], [4., 5., 6.]] # Shape [2, 3]
n = [3., 3.]
dist = DirichletMultinomial(n, alpha)
# counts will be broadcast to [[2., 1., 0.], [2., 1., 0.]] to match alpha.
counts = [2., 1., 0.]
dist.prob(counts) # Shape [2]
```"
9632,total_count,tensorflow/tensorflow/python/ops/distributions/dirichlet_multinomial.py,234,method,Number of trials used to construct a sample.
9633,concentration,tensorflow/tensorflow/python/ops/distributions/dirichlet_multinomial.py,239,method,Concentration parameter; expected prior counts for that coordinate.
9634,total_concentration,tensorflow/tensorflow/python/ops/distributions/dirichlet_multinomial.py,244,method,Sum of last dim of concentration parameter.
9635,ReparameterizationType,tensorflow/tensorflow/python/ops/distributions/distribution.py,216,class,"Instances of this class represent how sampling is reparameterized.
Two static instances exist in the distributions library, signifying
one of two possible properties for samples from a distribution:
`FULLY_REPARAMETERIZED`: Samples from the distribution are fully
reparameterized, and straight-through gradients are supported.
`NOT_REPARAMETERIZED`: Samples from the distribution are not fully
reparameterized, and straight-through gradients are either partially
unsupported or are not supported at all. In this case, for purposes of
e.g. RL or variational inference, it is generally safest to wrap the
sample results in a `stop_gradients` call and use policy
gradients / surrogate loss instead."
9636,Distribution,tensorflow/tensorflow/python/ops/distributions/distribution.py,280,class,"A generic probability distribution base class.
`Distribution` is a base class for constructing and organizing properties
(e.g., mean, variance) of random variables (e.g, Bernoulli, Gaussian).
#### Subclassing
Subclasses are expected to implement a leading-underscore version of the
same-named function. The argument signature should be identical except for
the omission of `name=""...""`. For example, to enable `log_prob(value,
name=""log_prob"")` a subclass should implement `_log_prob(value)`.
Subclasses can append to public-level docstrings by providing
docstrings for their method specializations. For example:
```python
@util.AppendDocstring(""Some other details."")
def _log_prob(self, value):
...
```
would add the string ""Some other details."" to the `log_prob` function
docstring. This is implemented as a simple decorator to avoid python
linter complaining about missing Args/Returns/Raises sections in the
partial docstrings.
#### Broadcasting, batching, and shapes
All distributions support batches of independent distributions of that type.
The batch shape is determined by broadcasting together the parameters.
The shape of arguments to `__init__`, `cdf`, `log_cdf`, `prob`, and
`log_prob` reflect this broadcasting, as does the return value of `sample` and
`sample_n`.
`sample_n_shape = [n] + batch_shape + event_shape`, where `sample_n_shape` is
the shape of the `Tensor` returned from `sample_n`, `n` is the number of
samples, `batch_shape` defines how many independent distributions there are,
and `event_shape` defines the shape of samples from each of those independent
distributions. Samples are independent along the `batch_shape` dimensions, but
not necessarily so along the `event_shape` dimensions (depending on the
particulars of the underlying distribution).
Using the `Uniform` distribution as an example:
```python
minval = 3.0
maxval = [[4.0, 6.0],
[10.0, 12.0]]
# Broadcasting:
# This instance represents 4 Uniform distributions. Each has a lower bound at
# 3.0 as the `minval` parameter was broadcasted to match `maxval`'s shape.
u = Uniform(minval, maxval)
# `event_shape` is `TensorShape([])`.
event_shape = u.event_shape
# `event_shape_t` is a `Tensor` which will evaluate to [].
event_shape_t = u.event_shape_tensor()
# Sampling returns a sample per distribution. `samples` has shape
# [5, 2, 2], which is [n] + batch_shape + event_shape, where n=5,
# batch_shape=[2, 2], and event_shape=[].
samples = u.sample_n(5)
# The broadcasting holds across methods. Here we use `cdf` as an example. The
# same holds for `log_cdf` and the likelihood functions.
# `cum_prob` has shape [2, 2] as the `value` argument was broadcasted to the
# shape of the `Uniform` instance.
cum_prob_broadcast = u.cdf(4.0)
# `cum_prob`'s shape is [2, 2], one per distribution. No broadcasting
# occurred.
cum_prob_per_dist = u.cdf([[4.0, 5.0],
[6.0, 7.0]])
# INVALID as the `value` argument is not broadcastable to the distribution's
# shape.
cum_prob_invalid = u.cdf([4.0, 5.0, 6.0])
```
#### Shapes
There are three important concepts associated with TensorFlow Distributions
shapes:
- Event shape describes the shape of a single draw from the distribution;
it may be dependent across dimensions. For scalar distributions, the event
shape is `[]`. For a 5-dimensional MultivariateNormal, the event shape is
`[5]`.
- Batch shape describes independent, not identically distributed draws, aka a
""collection"" or ""bunch"" of distributions.
- Sample shape describes independent, identically distributed draws of batches
from the distribution family.
The event shape and the batch shape are properties of a Distribution object,
whereas the sample shape is associated with a specific call to `sample` or
`log_prob`.
For detailed usage examples of TensorFlow Distributions shapes, see
[this tutorial](
https://github.com/tensorflow/probability/blob/master/tensorflow_probability/examples/jupyter_notebooks/Understanding_TensorFlow_Distributions_Shapes.ipynb)
#### Parameter values leading to undefined statistics or distributions.
Some distributions do not have well-defined statistics for all initialization
parameter values. For example, the beta distribution is parameterized by
positive real numbers `concentration1` and `concentration0`, and does not have
well-defined mode if `concentration1 < 1` or `concentration0 < 1`.
The user is given the option of raising an exception or returning `NaN`.
```python
a = tf.exp(tf.matmul(logits, weights_a))
b = tf.exp(tf.matmul(logits, weights_b))
# Will raise exception if ANY batch member has a < 1 or b < 1.
dist = distributions.beta(a, b, allow_nan_stats=False)
mode = dist.mode().eval()
# Will return NaN for batch members with either a < 1 or b < 1.
dist = distributions.beta(a, b, allow_nan_stats=True) # Default behavior
mode = dist.mode().eval()
```
In all cases, an exception is raised if *invalid* parameters are passed, e.g.
```python
# Will raise an exception if any Op is run.
negative_a = -1.0 * a # beta distribution by definition has a > 0.
dist = distributions.beta(negative_a, b, allow_nan_stats=True)
dist.mean().eval()
```"
9637,param_shapes,tensorflow/tensorflow/python/ops/distributions/distribution.py,498,method,"Shapes of parameters given the desired shape of a call to `sample()`.
This is a class method that describes what key/value arguments are required
to instantiate the given `Distribution` so that a particular shape is
returned for that instance's call to `sample()`.
Subclasses should override class method `_param_shapes`.
Args:
sample_shape: `Tensor` or python list/tuple. Desired shape of a call to
`sample()`.
name: name to prepend ops with.
Returns:
`dict` of parameter name to `Tensor` shapes."
9638,param_static_shapes,tensorflow/tensorflow/python/ops/distributions/distribution.py,519,method,"param_shapes with static (i.e. `TensorShape`) shapes.
This is a class method that describes what key/value arguments are required
to instantiate the given `Distribution` so that a particular shape is
returned for that instance's call to `sample()`. Assumes that the sample's
shape is known statically.
Subclasses should override class method `_param_shapes` to return
constant-valued tensors when constant values are fed.
Args:
sample_shape: `TensorShape` or python list/tuple. Desired shape of a call
to `sample()`.
Returns:
`dict` of parameter name to `TensorShape`.
Raises:
ValueError: if `sample_shape` is a `TensorShape` and is not fully defined."
9639,name,tensorflow/tensorflow/python/ops/distributions/distribution.py,562,method,Name prepended to all ops created by this `Distribution`.
9640,dtype,tensorflow/tensorflow/python/ops/distributions/distribution.py,567,method,The `DType` of `Tensor`s handled by this `Distribution`.
9641,parameters,tensorflow/tensorflow/python/ops/distributions/distribution.py,572,method,Dictionary of parameters used to instantiate this `Distribution`.
9642,reparameterization_type,tensorflow/tensorflow/python/ops/distributions/distribution.py,581,method,"Describes how samples from the distribution are reparameterized.
Currently this is one of the static instances
`distributions.FULLY_REPARAMETERIZED`
or `distributions.NOT_REPARAMETERIZED`.
Returns:
An instance of `ReparameterizationType`."
9643,allow_nan_stats,tensorflow/tensorflow/python/ops/distributions/distribution.py,594,method,"Python `bool` describing behavior when a stat is undefined.
Stats return +/- infinity when it makes sense. E.g., the variance of a
Cauchy distribution is infinity. However, sometimes the statistic is
undefined, e.g., if a distribution's pdf does not achieve a maximum within
the support of the distribution, the mode is undefined. If the mean is
undefined, then by definition the variance is undefined. E.g. the mean for
Student's T for df = 1 is undefined (no clear way to say it is either + or -
infinity), so the variance = E[(X - mean)**2] is also undefined.
Returns:
allow_nan_stats: Python `bool`."
9644,validate_args,tensorflow/tensorflow/python/ops/distributions/distribution.py,611,method,Python `bool` indicating possibly expensive checks are enabled.
9645,copy,tensorflow/tensorflow/python/ops/distributions/distribution.py,615,method,"Creates a deep copy of the distribution.
Note: the copy distribution may continue to depend on the original
initialization arguments.
Args:
**override_parameters_kwargs: String/value dictionary of initialization
arguments to override with new values.
Returns:
distribution: A new instance of `type(self)` initialized from the union
of self.parameters and override_parameters_kwargs, i.e.,
`dict(self.parameters, **override_parameters_kwargs)`."
9646,batch_shape_tensor,tensorflow/tensorflow/python/ops/distributions/distribution.py,637,method,"Shape of a single sample from a single event index as a 1-D `Tensor`.
The batch dimensions are indexes into independent, non-identical
parameterizations of this distribution.
Args:
name: name to give to the op
Returns:
batch_shape: `Tensor`."
9647,batch_shape,tensorflow/tensorflow/python/ops/distributions/distribution.py,660,method,"Shape of a single sample from a single event index as a `TensorShape`.
May be partially defined or unknown.
The batch dimensions are indexes into independent, non-identical
parameterizations of this distribution.
Returns:
batch_shape: `TensorShape`, possibly unknown."
9648,event_shape_tensor,tensorflow/tensorflow/python/ops/distributions/distribution.py,677,method,"Shape of a single sample from a single batch as a 1-D int32 `Tensor`.
Args:
name: name to give to the op
Returns:
event_shape: `Tensor`."
9649,event_shape,tensorflow/tensorflow/python/ops/distributions/distribution.py,697,method,"Shape of a single sample from a single batch as a `TensorShape`.
May be partially defined or unknown.
Returns:
event_shape: `TensorShape`, possibly unknown."
9650,is_scalar_event,tensorflow/tensorflow/python/ops/distributions/distribution.py,707,method,"Indicates that `event_shape == []`.
Args:
name: Python `str` prepended to names of ops created by this function.
Returns:
is_scalar_event: `bool` scalar `Tensor`."
9651,is_scalar_batch,tensorflow/tensorflow/python/ops/distributions/distribution.py,721,method,"Indicates that `batch_shape == []`.
Args:
name: Python `str` prepended to names of ops created by this function.
Returns:
is_scalar_batch: `bool` scalar `Tensor`."
9652,sample,tensorflow/tensorflow/python/ops/distributions/distribution.py,752,method,"Generate samples of the specified shape.
Note that a call to `sample()` without arguments will generate a single
sample.
Args:
sample_shape: 0D or 1D `int32` `Tensor`. Shape of the generated samples.
seed: Python integer seed for RNG
name: name to give to the op.
Returns:
samples: a `Tensor` with prepended dimensions `sample_shape`."
9653,log_prob,tensorflow/tensorflow/python/ops/distributions/distribution.py,784,method,"Log probability density/mass function.
Args:
value: `float` or `double` `Tensor`.
name: Python `str` prepended to names of ops created by this function.
Returns:
log_prob: a `Tensor` of shape `sample_shape(x) + self.batch_shape` with
values of type `self.dtype`."
9654,prob,tensorflow/tensorflow/python/ops/distributions/distribution.py,813,method,"Probability density/mass function.
Args:
value: `float` or `double` `Tensor`.
name: Python `str` prepended to names of ops created by this function.
Returns:
prob: a `Tensor` of shape `sample_shape(x) + self.batch_shape` with
values of type `self.dtype`."
9655,log_cdf,tensorflow/tensorflow/python/ops/distributions/distribution.py,842,method,"Log cumulative distribution function.
Given random variable `X`, the cumulative distribution function `cdf` is:
```none
log_cdf(x) := Log[ P[X <= x] ]
```
Often, a numerical approximation can be used for `log_cdf(x)` that yields
a more accurate answer than simply taking the logarithm of the `cdf` when
`x << -1`.
Args:
value: `float` or `double` `Tensor`.
name: Python `str` prepended to names of ops created by this function.
Returns:
logcdf: a `Tensor` of shape `sample_shape(x) + self.batch_shape` with
values of type `self.dtype`."
9656,cdf,tensorflow/tensorflow/python/ops/distributions/distribution.py,881,method,"Cumulative distribution function.
Given random variable `X`, the cumulative distribution function `cdf` is:
```none
cdf(x) := P[X <= x]
```
Args:
value: `float` or `double` `Tensor`.
name: Python `str` prepended to names of ops created by this function.
Returns:
cdf: a `Tensor` of shape `sample_shape(x) + self.batch_shape` with
values of type `self.dtype`."
9657,log_survival_function,tensorflow/tensorflow/python/ops/distributions/distribution.py,917,method,"Log survival function.
Given random variable `X`, the survival function is defined:
```none
log_survival_function(x) = Log[ P[X > x] ]
= Log[ 1 - P[X <= x] ]
= Log[ 1 - cdf(x) ]
```
Typically, different numerical approximations can be used for the log
survival function, which are more accurate than `1 - cdf(x)` when `x >> 1`.
Args:
value: `float` or `double` `Tensor`.
name: Python `str` prepended to names of ops created by this function.
Returns:
`Tensor` of shape `sample_shape(x) + self.batch_shape` with values of type
`self.dtype`."
9658,survival_function,tensorflow/tensorflow/python/ops/distributions/distribution.py,957,method,"Survival function.
Given random variable `X`, the survival function is defined:
```none
survival_function(x) = P[X > x]
= 1 - P[X <= x]
= 1 - cdf(x).
```
Args:
value: `float` or `double` `Tensor`.
name: Python `str` prepended to names of ops created by this function.
Returns:
`Tensor` of shape `sample_shape(x) + self.batch_shape` with values of type
`self.dtype`."
9659,entropy,tensorflow/tensorflow/python/ops/distributions/distribution.py,982,method,Shannon entropy in nats.
9660,mean,tensorflow/tensorflow/python/ops/distributions/distribution.py,991,method,Mean.
9661,quantile,tensorflow/tensorflow/python/ops/distributions/distribution.py,1006,method,"Quantile function. Aka ""inverse cdf"" or ""percent point function"".
Given random variable `X` and `p in [0, 1]`, the `quantile` is:
```none
quantile(p) := x such that P[X <= x] == p
```
Args:
value: `float` or `double` `Tensor`.
name: Python `str` prepended to names of ops created by this function.
Returns:
quantile: a `Tensor` of shape `sample_shape(x) + self.batch_shape` with
values of type `self.dtype`."
9662,variance,tensorflow/tensorflow/python/ops/distributions/distribution.py,1029,method,"Variance.
Variance is defined as,
```none
Var = E[(X - E[X])**2]
```
where `X` is the random variable associated with this distribution, `E`
denotes expectation, and `Var.shape = batch_shape + event_shape`.
Args:
name: Python `str` prepended to names of ops created by this function.
Returns:
variance: Floating-point `Tensor` with shape identical to
`batch_shape + event_shape`, i.e., the same shape as `self.mean()`."
9663,stddev,tensorflow/tensorflow/python/ops/distributions/distribution.py,1061,method,"Standard deviation.
Standard deviation is defined as,
```none
stddev = E[(X - E[X])**2]**0.5
```
where `X` is the random variable associated with this distribution, `E`
denotes expectation, and `stddev.shape = batch_shape + event_shape`.
Args:
name: Python `str` prepended to names of ops created by this function.
Returns:
stddev: Floating-point `Tensor` with shape identical to
`batch_shape + event_shape`, i.e., the same shape as `self.mean()`."
9664,covariance,tensorflow/tensorflow/python/ops/distributions/distribution.py,1094,method,"Covariance.
Covariance is (possibly) defined only for non-scalar-event distributions.
For example, for a length-`k`, vector-valued distribution, it is calculated
as,
```none
Cov[i, j] = Covariance(X_i, X_j) = E[(X_i - E[X_i]) (X_j - E[X_j])]
```
where `Cov` is a (batch of) `k x k` matrix, `0 <= (i, j) < k`, and `E`
denotes expectation.
Alternatively, for non-vector, multivariate distributions (e.g.,
matrix-valued, Wishart), `Covariance` shall return a (batch of) matrices
under some vectorization of the events, i.e.,
```none
Cov[i, j] = Covariance(Vec(X)_i, Vec(X)_j) = [as above]
```
where `Cov` is a (batch of) `k' x k'` matrices,
`0 <= (i, j) < k' = reduce_prod(event_shape)`, and `Vec` is some function
mapping indices of this distribution's event dimensions to indices of a
length-`k'` vector.
Args:
name: Python `str` prepended to names of ops created by this function.
Returns:
covariance: Floating-point `Tensor` with shape `[B1, ..., Bn, k', k']`
where the first `n` dimensions are batch coordinates and
`k' = reduce_prod(self.event_shape)`."
9665,mode,tensorflow/tensorflow/python/ops/distributions/distribution.py,1137,method,Mode.
9666,cross_entropy,tensorflow/tensorflow/python/ops/distributions/distribution.py,1146,method,"Computes the (Shannon) cross entropy.
Denote this distribution (`self`) by `P` and the `other` distribution by
`Q`. Assuming `P, Q` are absolutely continuous with respect to
one another and permit densities `p(x) dr(x)` and `q(x) dr(x)`, (Shanon)
cross entropy is defined as:
```none
H[P, Q] = E_p[-log q(X)] = -int_F p(x) log q(x) dr(x)
```
where `F` denotes the support of the random variable `X ~ P`.
Args:
other: `tfp.distributions.Distribution` instance.
name: Python `str` prepended to names of ops created by this function.
Returns:
cross_entropy: `self.dtype` `Tensor` with shape `[B1, ..., Bn]`
representing `n` different calculations of (Shanon) cross entropy."
9667,kl_divergence,tensorflow/tensorflow/python/ops/distributions/distribution.py,1175,method,"Computes the Kullback--Leibler divergence.
Denote this distribution (`self`) by `p` and the `other` distribution by
`q`. Assuming `p, q` are absolutely continuous with respect to reference
measure `r`, the KL divergence is defined as:
```none
KL[p, q] = E_p[log(p(X)/q(X))]
= -int_F p(x) log q(x) dr(x) + int_F p(x) log p(x) dr(x)
= H[p, q] - H[p]
```
where `F` denotes the support of the random variable `X ~ p`, `H[., .]`
denotes (Shanon) cross entropy, and `H[.]` denotes (Shanon) entropy.
Args:
other: `tfp.distributions.Distribution` instance.
name: Python `str` prepended to names of ops created by this function.
Returns:
kl_divergence: `self.dtype` `Tensor` with shape `[B1, ..., Bn]`
representing `n` different calculations of the Kullback-Leibler
divergence."
9668,Exponential,tensorflow/tensorflow/python/ops/distributions/exponential.py,41,class,"Exponential distribution.
The Exponential distribution is parameterized by an event `rate` parameter.
#### Mathematical Details
The probability density function (pdf) is,
```none
pdf(x; lambda, x > 0) = exp(-lambda x) / Z
Z = 1 / lambda
```
where `rate = lambda` and `Z` is the normalizaing constant.
The Exponential distribution is a special case of the Gamma distribution,
i.e.,
```python
Exponential(rate) = Gamma(concentration=1., rate)
```
The Exponential distribution uses a `rate` parameter, or ""inverse scale"",
which can be intuited as,
```none
X ~ Exponential(rate=1)
Y = X / rate
```"
9669,rate,tensorflow/tensorflow/python/ops/distributions/exponential.py,123,method,
9670,ExponentialWithSoftplusRate,tensorflow/tensorflow/python/ops/distributions/exponential.py,147,class,Exponential with softplus transform on `rate`.
9671,Gamma,tensorflow/tensorflow/python/ops/distributions/gamma.py,47,class,"Gamma distribution.
The Gamma distribution is defined over positive real numbers using
parameters `concentration` (aka ""alpha"") and `rate` (aka ""beta"").
#### Mathematical Details
The probability density function (pdf) is,
```none
pdf(x; alpha, beta, x > 0) = x**(alpha - 1) exp(-x beta) / Z
Z = Gamma(alpha) beta**(-alpha)
```
where:
* `concentration = alpha`, `alpha > 0`,
* `rate = beta`, `beta > 0`,
* `Z` is the normalizing constant, and,
* `Gamma` is the [gamma function](
https://en.wikipedia.org/wiki/Gamma_function).
The cumulative density function (cdf) is,
```none
cdf(x; alpha, beta, x > 0) = GammaInc(alpha, beta x) / Gamma(alpha)
```
where `GammaInc` is the [lower incomplete Gamma function](
https://en.wikipedia.org/wiki/Incomplete_gamma_function).
The parameters can be intuited via their relationship to mean and stddev,
```none
concentration = alpha = (mean / stddev)**2
rate = beta = mean / stddev**2 = concentration / mean
```
Distribution parameters are automatically broadcast in all functions; see
examples for details.
Warning: The samples of this distribution are always non-negative. However,
the samples that are smaller than `np.finfo(dtype).tiny` are rounded
to this value, so it appears more often than it should.
This should only be noticeable when the `concentration` is very small, or the
`rate` is very large. See note in `tf.random.gamma` docstring.
Samples of this distribution are reparameterized (pathwise differentiable).
The derivatives are computed using the approach described in
(Figurnov et al., 2018).
#### Examples
```python
import tensorflow_probability as tfp
tfd = tfp.distributions
dist = tfd.Gamma(concentration=3.0, rate=2.0)
dist2 = tfd.Gamma(concentration=[3.0, 4.0], rate=[2.0, 3.0])
```
Compute the gradients of samples w.r.t. the parameters:
```python
concentration = tf.constant(3.0)
rate = tf.constant(2.0)
dist = tfd.Gamma(concentration, rate)
samples = dist.sample(5) # Shape [5]
loss = tf.reduce_mean(tf.square(samples)) # Arbitrary loss function
# Unbiased stochastic gradients of the loss function
grads = tf.gradients(loss, [concentration, rate])
```
References:
Implicit Reparameterization Gradients:
[Figurnov et al., 2018]
(http://papers.nips.cc/paper/7326-implicit-reparameterization-gradients)
([pdf](http://papers.nips.cc/paper/7326-implicit-reparameterization-gradients.pdf))"
9672,concentration,tensorflow/tensorflow/python/ops/distributions/gamma.py,193,method,Concentration parameter.
9673,rate,tensorflow/tensorflow/python/ops/distributions/gamma.py,198,method,Rate parameter.
9674,GammaWithSoftplusConcentrationRate,tensorflow/tensorflow/python/ops/distributions/gamma.py,291,class,`Gamma` with softplus of `concentration` and `rate`.
9675,Identity,tensorflow/tensorflow/python/ops/distributions/identity_bijector.py,31,class,"Compute Y = g(X) = X.
Example Use:
```python
# Create the Y=g(X)=X transform which is intended for Tensors with 1 batch
# ndim and 1 event ndim (i.e., vector of vectors).
identity = Identity()
x = [[1., 2],
[3, 4]]
x == identity.forward(x) == identity.inverse(x)
```"
9676,kl_divergence,tensorflow/tensorflow/python/ops/distributions/kullback_leibler.py,64,function,"Get the KL-divergence KL(distribution_a || distribution_b).
If there is no KL method registered specifically for `type(distribution_a)`
and `type(distribution_b)`, then the class hierarchies of these types are
searched.
If one KL method is registered between any pairs of classes in these two
parent hierarchies, it is used.
If more than one such registered method exists, the method whose registered
classes have the shortest sum MRO paths to the input types is used.
If more than one such shortest path exists, the first method
identified in the search is used (favoring a shorter MRO distance to
`type(distribution_a)`).
Args:
distribution_a: The first distribution.
distribution_b: The second distribution.
allow_nan_stats: Python `bool`, default `True`. When `True`,
statistics (e.g., mean, mode, variance) use the value ""`NaN`"" to
indicate the result is undefined. When `False`, an exception is raised
if one or more of the statistic's batch members are undefined.
name: Python `str` name prefixed to Ops created by this class.
Returns:
A Tensor with the batchwise KL-divergence between `distribution_a`
and `distribution_b`.
Raises:
NotImplementedError: If no KL method is defined for distribution types
of `distribution_a` and `distribution_b`."
9677,cross_entropy,tensorflow/tensorflow/python/ops/distributions/kullback_leibler.py,132,function,"Computes the (Shannon) cross entropy.
Denote two distributions by `P` (`ref`) and `Q` (`other`). Assuming `P, Q`
are absolutely continuous with respect to one another and permit densities
`p(x) dr(x)` and `q(x) dr(x)`, (Shanon) cross entropy is defined as:
```none
H[P, Q] = E_p[-log q(X)] = -int_F p(x) log q(x) dr(x)
```
where `F` denotes the support of the random variable `X ~ P`.
Args:
ref: `tfd.Distribution` instance.
other: `tfd.Distribution` instance.
allow_nan_stats: Python `bool`, default `True`. When `True`,
statistics (e.g., mean, mode, variance) use the value ""`NaN`"" to
indicate the result is undefined. When `False`, an exception is raised
if one or more of the statistic's batch members are undefined.
name: Python `str` prepended to names of ops created by this function.
Returns:
cross_entropy: `ref.dtype` `Tensor` with shape `[B1, ..., Bn]`
representing `n` different calculations of (Shanon) cross entropy."
9678,RegisterKL,tensorflow/tensorflow/python/ops/distributions/kullback_leibler.py,165,class,"Decorator to register a KL divergence implementation function.
Usage:
@distributions.RegisterKL(distributions.Normal, distributions.Normal)
def _kl_normal_mvn(norm_a, norm_b):
# Return KL(norm_a || norm_b)"
9679,Laplace,tensorflow/tensorflow/python/ops/distributions/laplace.py,47,class,"The Laplace distribution with location `loc` and `scale` parameters.
#### Mathematical details
The probability density function (pdf) of this distribution is,
```none
pdf(x; mu, sigma) = exp(-|x - mu| / sigma) / Z
Z = 2 sigma
```
where `loc = mu`, `scale = sigma`, and `Z` is the normalization constant.
Note that the Laplace distribution can be thought of two exponential
distributions spliced together ""back-to-back.""
The Lpalce distribution is a member of the [location-scale family](
https://en.wikipedia.org/wiki/Location-scale_family), i.e., it can be
constructed as,
```none
X ~ Laplace(loc=0, scale=1)
Y = loc + scale * X
```"
9680,loc,tensorflow/tensorflow/python/ops/distributions/laplace.py,135,method,Distribution parameter for the location.
9681,scale,tensorflow/tensorflow/python/ops/distributions/laplace.py,140,method,Distribution parameter for scale.
9682,LaplaceWithSoftplusScale,tensorflow/tensorflow/python/ops/distributions/laplace.py,220,class,Laplace with softplus applied to `scale`.
9683,Multinomial,tensorflow/tensorflow/python/ops/distributions/multinomial.py,56,class,"Multinomial distribution.
This Multinomial distribution is parameterized by `probs`, a (batch of)
length-`K` `prob` (probability) vectors (`K > 1`) such that
`tf.reduce_sum(probs, -1) = 1`, and a `total_count` number of trials, i.e.,
the number of trials per draw from the Multinomial. It is defined over a
(batch of) length-`K` vector `counts` such that
`tf.reduce_sum(counts, -1) = total_count`. The Multinomial is identically the
Binomial distribution when `K = 2`.
#### Mathematical Details
The Multinomial is a distribution over `K`-class counts, i.e., a length-`K`
vector of non-negative integer `counts = n = [n_0, ..., n_{K-1}]`.
The probability mass function (pmf) is,
```none
pmf(n; pi, N) = prod_j (pi_j)**n_j / Z
Z = (prod_j n_j!) / N!
```
where:
* `probs = pi = [pi_0, ..., pi_{K-1}]`, `pi_j > 0`, `sum_j pi_j = 1`,
* `total_count = N`, `N` a positive integer,
* `Z` is the normalization constant, and,
* `N!` denotes `N` factorial.
Distribution parameters are automatically broadcast in all functions; see
examples for details.
#### Pitfalls
The number of classes, `K`, must not exceed:
- the largest integer representable by `self.dtype`, i.e.,
`2**(mantissa_bits+1)` (IEE754),
- the maximum `Tensor` index, i.e., `2**31-1`.
In other words,
```python
K <= min(2**31-1, {
tf.float16: 2**11,
tf.float32: 2**24,
tf.float64: 2**53 }[param.dtype])
```
Note: This condition is validated only when `self.validate_args = True`.
#### Examples
Create a 3-class distribution, with the 3rd class is most likely to be drawn,
using logits.
```python
logits = [-50., -43, 0]
dist = Multinomial(total_count=4., logits=logits)
```
Create a 3-class distribution, with the 3rd class is most likely to be drawn.
```python
p = [.2, .3, .5]
dist = Multinomial(total_count=4., probs=p)
```
The distribution functions can be evaluated on counts.
```python
# counts same shape as p.
counts = [1., 0, 3]
dist.prob(counts) # Shape []
# p will be broadcast to [[.2, .3, .5], [.2, .3, .5]] to match counts.
counts = [[1., 2, 1], [2, 2, 0]]
dist.prob(counts) # Shape [2]
# p will be broadcast to shape [5, 7, 3] to match counts.
counts = [[...]] # Shape [5, 7, 3]
dist.prob(counts) # Shape [5, 7]
```
Create a 2-batch of 3-class distributions.
```python
p = [[.1, .2, .7], [.3, .3, .4]] # Shape [2, 3]
dist = Multinomial(total_count=[4., 5], probs=p)
counts = [[2., 1, 1], [3, 1, 1]]
dist.prob(counts) # Shape [2]
dist.sample(5) # Shape [5, 2, 3]
```"
9684,total_count,tensorflow/tensorflow/python/ops/distributions/multinomial.py,220,method,Number of trials used to construct a sample.
9685,logits,tensorflow/tensorflow/python/ops/distributions/multinomial.py,225,method,Vector of coordinatewise logits.
9686,probs,tensorflow/tensorflow/python/ops/distributions/multinomial.py,230,method,Probability of drawing a `1` in that coordinate.
9687,Normal,tensorflow/tensorflow/python/ops/distributions/normal.py,46,class,"The Normal distribution with location `loc` and `scale` parameters.
#### Mathematical details
The probability density function (pdf) is,
```none
pdf(x; mu, sigma) = exp(-0.5 (x - mu)**2 / sigma**2) / Z
Z = (2 pi sigma**2)**0.5
```
where `loc = mu` is the mean, `scale = sigma` is the std. deviation, and, `Z`
is the normalization constant.
The Normal distribution is a member of the [location-scale family](
https://en.wikipedia.org/wiki/Location-scale_family), i.e., it can be
constructed as,
```none
X ~ Normal(loc=0, scale=1)
Y = loc + scale * X
```
#### Examples
Examples of initialization of one or a batch of distributions.
```python
import tensorflow_probability as tfp
tfd = tfp.distributions
# Define a single scalar Normal distribution.
dist = tfd.Normal(loc=0., scale=3.)
# Evaluate the cdf at 1, returning a scalar.
dist.cdf(1.)
# Define a batch of two scalar valued Normals.
# The first has mean 1 and standard deviation 11, the second 2 and 22.
dist = tfd.Normal(loc=[1, 2.], scale=[11, 22.])
# Evaluate the pdf of the first distribution on 0, and the second on 1.5,
# returning a length two tensor.
dist.prob([0, 1.5])
# Get 3 samples, returning a 3 x 2 tensor.
dist.sample([3])
```
Arguments are broadcast when possible.
```python
# Define a batch of two scalar valued Normals.
# Both have mean 1, but different standard deviations.
dist = tfd.Normal(loc=1., scale=[11, 22.])
# Evaluate the pdf of both distributions on the same point, 3.0,
# returning a length 2 tensor.
dist.prob(3.0)
```"
9688,loc,tensorflow/tensorflow/python/ops/distributions/normal.py,169,method,Distribution parameter for the mean.
9689,scale,tensorflow/tensorflow/python/ops/distributions/normal.py,174,method,Distribution parameter for standard deviation.
9690,NormalWithSoftplusScale,tensorflow/tensorflow/python/ops/distributions/normal.py,249,class,Normal with softplus applied to `scale`.
9691,ndtr,tensorflow/tensorflow/python/ops/distributions/special_math.py,111,function,"Normal distribution function.
Returns the area under the Gaussian probability density function, integrated
from minus infinity to x:
```
1 / x
ndtr(x) = ---------- | exp(-0.5 t**2) dt
sqrt(2 pi) /-inf
= 0.5 (1 + erf(x / sqrt(2)))
= 0.5 erfc(x / sqrt(2))
```
Args:
x: `Tensor` of type `float32`, `float64`.
name: Python string. A name for the operation (default=""ndtr"").
Returns:
ndtr: `Tensor` with `dtype=x.dtype`.
Raises:
TypeError: if `x` is not floating-type."
9692,ndtri,tensorflow/tensorflow/python/ops/distributions/special_math.py,159,function,"The inverse of the CDF of the Normal distribution function.
Returns x such that the area under the pdf from minus infinity to x is equal
to p.
A piece-wise rational approximation is done for the function.
This is a port of the implementation in netlib.
Args:
p: `Tensor` of type `float32`, `float64`.
name: Python string. A name for the operation (default=""ndtri"").
Returns:
x: `Tensor` with `dtype=p.dtype`.
Raises:
TypeError: if `p` is not floating-type."
9693,log_ndtr,tensorflow/tensorflow/python/ops/distributions/special_math.py,282,function,"Log Normal distribution function.
For details of the Normal distribution function see `ndtr`.
This function calculates `(log o ndtr)(x)` by either calling `log(ndtr(x))` or
using an asymptotic series. Specifically:
- For `x > upper_segment`, use the approximation `-ndtr(-x)` based on
`log(1-x) ~= -x, x << 1`.
- For `lower_segment < x <= upper_segment`, use the existing `ndtr` technique
and take a log.
- For `x <= lower_segment`, we use the series approximation of erf to compute
the log CDF directly.
The `lower_segment` is set based on the precision of the input:
```
lower_segment = { -20, x.dtype=float64
{ -10, x.dtype=float32
upper_segment = { 8, x.dtype=float64
{ 5, x.dtype=float32
```
When `x < lower_segment`, the `ndtr` asymptotic series approximation is:
```
ndtr(x) = scale * (1 + sum) + R_N
scale = exp(-0.5 x**2) / (-x sqrt(2 pi))
sum = Sum{(-1)^n (2n-1)!! / (x**2)^n, n=1:N}
R_N = O(exp(-0.5 x**2) (2N+1)!! / |x|^{2N+3})
```
where `(2n-1)!! = (2n-1) (2n-3) (2n-5) ... (3) (1)` is a
[double-factorial](https://en.wikipedia.org/wiki/Double_factorial).
Args:
x: `Tensor` of type `float32`, `float64`.
series_order: Positive Python `integer`. Maximum depth to
evaluate the asymptotic expansion. This is the `N` above.
name: Python string. A name for the operation (default=""log_ndtr"").
Returns:
log_ndtr: `Tensor` with `dtype=x.dtype`.
Raises:
TypeError: if `x.dtype` is not handled.
TypeError: if `series_order` is a not Python `integer.`
ValueError: if `series_order` is not in `[0, 30]`."
9694,erfinv,tensorflow/tensorflow/python/ops/distributions/special_math.py,401,function,"The inverse function for erf, the error function.
Args:
x: `Tensor` of type `float32`, `float64`.
name: Python string. A name for the operation (default=""erfinv"").
Returns:
x: `Tensor` with `dtype=x.dtype`.
Raises:
TypeError: if `x` is not floating-type."
9695,log_cdf_laplace,tensorflow/tensorflow/python/ops/distributions/special_math.py,429,function,"Log Laplace distribution function.
This function calculates `Log[L(x)]`, where `L(x)` is the cumulative
distribution function of the Laplace distribution, i.e.
```L(x) := 0.5 * int_{-infty}^x e^{-|t|} dt```
For numerical accuracy, `L(x)` is computed in different ways depending on `x`,
```
x <= 0:
Log[L(x)] = Log[0.5] + x, which is exact
0 < x:
Log[L(x)] = Log[1 - 0.5 * e^{-x}], which is exact
```
Args:
x: `Tensor` of type `float32`, `float64`.
name: Python string. A name for the operation (default=""log_ndtr"").
Returns:
`Tensor` with `dtype=x.dtype`.
Raises:
TypeError: if `x.dtype` is not handled."
9696,StudentT,tensorflow/tensorflow/python/ops/distributions/student_t.py,47,class,"Student's t-distribution.
This distribution has parameters: degree of freedom `df`, location `loc`,
and `scale`.
#### Mathematical details
The probability density function (pdf) is,
```none
pdf(x; df, mu, sigma) = (1 + y**2 / df)**(-0.5 (df + 1)) / Z
where,
y = (x - mu) / sigma
Z = abs(sigma) sqrt(df pi) Gamma(0.5 df) / Gamma(0.5 (df + 1))
```
where:
* `loc = mu`,
* `scale = sigma`, and,
* `Z` is the normalization constant, and,
* `Gamma` is the [gamma function](
https://en.wikipedia.org/wiki/Gamma_function).
The StudentT distribution is a member of the [location-scale family](
https://en.wikipedia.org/wiki/Location-scale_family), i.e., it can be
constructed as,
```none
X ~ StudentT(df, loc=0, scale=1)
Y = loc + scale * X
```
Notice that `scale` has semantics more similar to standard deviation than
variance. However it is not actually the std. deviation; the Student's
t-distribution std. dev. is `scale sqrt(df / (df - 2))` when `df > 2`.
Samples of this distribution are reparameterized (pathwise differentiable).
The derivatives are computed using the approach described in
(Figurnov et al., 2018).
#### Examples
Examples of initialization of one or a batch of distributions.
```python
import tensorflow_probability as tfp
tfd = tfp.distributions
# Define a single scalar Student t distribution.
single_dist = tfd.StudentT(df=3)
# Evaluate the pdf at 1, returning a scalar Tensor.
single_dist.prob(1.)
# Define a batch of two scalar valued Student t's.
# The first has degrees of freedom 2, mean 1, and scale 11.
# The second 3, 2 and 22.
multi_dist = tfd.StudentT(df=[2, 3], loc=[1, 2.], scale=[11, 22.])
# Evaluate the pdf of the first distribution on 0, and the second on 1.5,
# returning a length two tensor.
multi_dist.prob([0, 1.5])
# Get 3 samples, returning a 3 x 2 tensor.
multi_dist.sample(3)
```
Arguments are broadcast when possible.
```python
# Define a batch of two Student's t distributions.
# Both have df 2 and mean 1, but different scales.
dist = tfd.StudentT(df=2, loc=1, scale=[11, 22.])
# Evaluate the pdf of both distributions on the same point, 3.0,
# returning a length 2 tensor.
dist.prob(3.0)
```
Compute the gradients of samples w.r.t. the parameters:
```python
df = tf.constant(2.0)
loc = tf.constant(2.0)
scale = tf.constant(11.0)
dist = tfd.StudentT(df=df, loc=loc, scale=scale)
samples = dist.sample(5) # Shape [5]
loss = tf.reduce_mean(tf.square(samples)) # Arbitrary loss function
# Unbiased stochastic gradients of the loss function
grads = tf.gradients(loss, [df, loc, scale])
```
References:
Implicit Reparameterization Gradients:
[Figurnov et al., 2018]
(http://papers.nips.cc/paper/7326-implicit-reparameterization-gradients)
([pdf](http://papers.nips.cc/paper/7326-implicit-reparameterization-gradients.pdf))"
9697,df,tensorflow/tensorflow/python/ops/distributions/student_t.py,217,method,Degrees of freedom in these Student's t distribution(s).
9698,loc,tensorflow/tensorflow/python/ops/distributions/student_t.py,222,method,Locations of these Student's t distribution(s).
9699,scale,tensorflow/tensorflow/python/ops/distributions/student_t.py,227,method,Scaling factors of these Student's t distribution(s).
9700,StudentTWithAbsDfSoftplusScale,tensorflow/tensorflow/python/ops/distributions/student_t.py,371,class,StudentT with `df = floor(abs(df))` and `scale = softplus(scale)`.
9701,TransformedDistribution,tensorflow/tensorflow/python/ops/distributions/transformed_distribution.py,122,class,"A Transformed Distribution.
A `TransformedDistribution` models `p(y)` given a base distribution `p(x)`,
and a deterministic, invertible, differentiable transform, `Y = g(X)`. The
transform is typically an instance of the `Bijector` class and the base
distribution is typically an instance of the `Distribution` class.
A `Bijector` is expected to implement the following functions:
- `forward`,
- `inverse`,
- `inverse_log_det_jacobian`.
The semantics of these functions are outlined in the `Bijector` documentation.
We now describe how a `TransformedDistribution` alters the input/outputs of a
`Distribution` associated with a random variable (rv) `X`.
Write `cdf(Y=y)` for an absolutely continuous cumulative distribution function
of random variable `Y`; write the probability density function `pdf(Y=y) :=
d^k / (dy_1,...,dy_k) cdf(Y=y)` for its derivative wrt to `Y` evaluated at
`y`. Assume that `Y = g(X)` where `g` is a deterministic diffeomorphism,
i.e., a non-random, continuous, differentiable, and invertible function.
Write the inverse of `g` as `X = g^{-1}(Y)` and `(J o g)(x)` for the Jacobian
of `g` evaluated at `x`.
A `TransformedDistribution` implements the following operations:
* `sample`
Mathematically: `Y = g(X)`
Programmatically: `bijector.forward(distribution.sample(...))`
* `log_prob`
Mathematically: `(log o pdf)(Y=y) = (log o pdf o g^{-1})(y)
+ (log o abs o det o J o g^{-1})(y)`
Programmatically: `(distribution.log_prob(bijector.inverse(y))
+ bijector.inverse_log_det_jacobian(y))`
* `log_cdf`
Mathematically: `(log o cdf)(Y=y) = (log o cdf o g^{-1})(y)`
Programmatically: `distribution.log_cdf(bijector.inverse(x))`
* and similarly for: `cdf`, `prob`, `log_survival_function`,
`survival_function`.
A simple example constructing a Log-Normal distribution from a Normal
distribution:
```python
ds = tfp.distributions
log_normal = ds.TransformedDistribution(
distribution=ds.Normal(loc=0., scale=1.),
bijector=ds.bijectors.Exp(),
name=""LogNormalTransformedDistribution"")
```
A `LogNormal` made from callables:
```python
ds = tfp.distributions
log_normal = ds.TransformedDistribution(
distribution=ds.Normal(loc=0., scale=1.),
bijector=ds.bijectors.Inline(
forward_fn=tf.exp,
inverse_fn=tf.math.log,
inverse_log_det_jacobian_fn=(
lambda y: -tf.reduce_sum(tf.math.log(y), axis=-1)),
name=""LogNormalTransformedDistribution"")
```
Another example constructing a Normal from a StandardNormal:
```python
ds = tfp.distributions
normal = ds.TransformedDistribution(
distribution=ds.Normal(loc=0., scale=1.),
bijector=ds.bijectors.Affine(
shift=-1.,
scale_identity_multiplier=2.)
name=""NormalTransformedDistribution"")
```
A `TransformedDistribution`'s batch- and event-shape are implied by the base
distribution unless explicitly overridden by `batch_shape` or `event_shape`
arguments. Specifying an overriding `batch_shape` (`event_shape`) is
permitted only if the base distribution has scalar batch-shape (event-shape).
The bijector is applied to the distribution as if the distribution possessed
the overridden shape(s). The following example demonstrates how to construct a
multivariate Normal as a `TransformedDistribution`.
```python
ds = tfp.distributions
# We will create two MVNs with batch_shape = event_shape = 2.
mean = [[-1., 0], # batch:0
[0., 1]] # batch:1
chol_cov = [[[1., 0],
[0, 1]], # batch:0
[[1, 0],
[2, 2]]] # batch:1
mvn1 = ds.TransformedDistribution(
distribution=ds.Normal(loc=0., scale=1.),
bijector=ds.bijectors.Affine(shift=mean, scale_tril=chol_cov),
batch_shape=[2], # Valid because base_distribution.batch_shape == [].
event_shape=[2]) # Valid because base_distribution.event_shape == [].
mvn2 = ds.MultivariateNormalTriL(loc=mean, scale_tril=chol_cov)
# mvn1.log_prob(x) == mvn2.log_prob(x)
```"
9702,distribution,tensorflow/tensorflow/python/ops/distributions/transformed_distribution.py,330,method,"Base distribution, p(x)."
9703,bijector,tensorflow/tensorflow/python/ops/distributions/transformed_distribution.py,335,method,Function transforming x => y.
9704,Uniform,tensorflow/tensorflow/python/ops/distributions/uniform.py,37,class,"Uniform distribution with `low` and `high` parameters.
#### Mathematical Details
The probability density function (pdf) is,
```none
pdf(x; a, b) = I[a <= x < b] / Z
Z = b - a
```
where
- `low = a`,
- `high = b`,
- `Z` is the normalizing constant, and
- `I[predicate]` is the [indicator function](
https://en.wikipedia.org/wiki/Indicator_function) for `predicate`.
The parameters `low` and `high` must be shaped in a way that supports
broadcasting (e.g., `high - low` is a valid operation).
#### Examples
```python
# Without broadcasting:
u1 = Uniform(low=3.0, high=4.0) # a single uniform distribution [3, 4]
u2 = Uniform(low=[1.0, 2.0],
high=[3.0, 4.0]) # 2 distributions [1, 3], [2, 4]
u3 = Uniform(low=[[1.0, 2.0],
[3.0, 4.0]],
high=[[1.5, 2.5],
[3.5, 4.5]]) # 4 distributions
```
```python
# With broadcasting:
u1 = Uniform(low=3.0, high=[5.0, 6.0, 7.0]) # 3 distributions
```"
9705,low,tensorflow/tensorflow/python/ops/distributions/uniform.py,140,method,Lower boundary of the output interval.
9706,high,tensorflow/tensorflow/python/ops/distributions/uniform.py,145,method,Upper boundary of the output interval.
9707,range,tensorflow/tensorflow/python/ops/distributions/uniform.py,149,method,`high - low`.
9708,assert_integer_form,tensorflow/tensorflow/python/ops/distributions/util.py,40,function,"Assert that x has integer components (or floats equal to integers).
Args:
x: Floating-point `Tensor`
data: The tensors to print out if the condition is `False`. Defaults to
error message and first few entries of `x` and `y`.
summarize: Print this many entries of each tensor.
message: A string to prefix to the default message.
int_dtype: A `tf.dtype` used to cast the float to. The default (`None`)
implies the smallest possible signed int will be used for casting.
name: A name for this operation (optional).
Returns:
Op raising `InvalidArgumentError` if `cast(x, int_dtype) != x`."
9709,assert_symmetric,tensorflow/tensorflow/python/ops/distributions/util.py,84,function,
9710,embed_check_nonnegative_integer_form,tensorflow/tensorflow/python/ops/distributions/util.py,90,function,"Assert x is a non-negative tensor, and optionally of integers."
9711,same_dynamic_shape,tensorflow/tensorflow/python/ops/distributions/util.py,108,function,"Returns whether a and b have the same dynamic shape.
Args:
a: `Tensor`
b: `Tensor`
Returns:
`bool` `Tensor` representing if both tensors have the same shape."
9712,maybe_get_static_value,tensorflow/tensorflow/python/ops/distributions/util.py,139,function,"Helper which tries to return a static value.
Given `x`, extract it's value statically, optionally casting to a specific
dtype. If this is not possible, None is returned.
Args:
x: `Tensor` for which to extract a value statically.
dtype: Optional dtype to cast to.
Returns:
Statically inferred value if possible, otherwise None."
9713,get_logits_and_probs,tensorflow/tensorflow/python/ops/distributions/util.py,164,function,"Converts logit to probabilities (or vice-versa), and returns both.
Args:
logits: Floating-point `Tensor` representing log-odds.
probs: Floating-point `Tensor` representing probabilities.
multidimensional: Python `bool`, default `False`. If `True`, represents
whether the last dimension of `logits` or `probs`, a `[N1, N2, ... k]`
dimensional tensor, representing the logit or probability of `shape[-1]`
classes.
validate_args: Python `bool`, default `False`. When `True`, either assert `0
<= probs <= 1` (if not `multidimensional`) or that the last dimension of
`probs` sums to one.
name: A name for this operation (optional).
dtype: `tf.DType` to prefer when converting args to `Tensor`s.
Returns:
logits, probs: Tuple of `Tensor`s. If `probs` has an entry that is `0` or
`1`, then the corresponding entry in the returned logit will be `-Inf` and
`Inf` respectively.
Raises:
ValueError: if neither `probs` nor `logits` were passed in, or both were."
9714,embed_check_categorical_event_shape,tensorflow/tensorflow/python/ops/distributions/util.py,302,function,"Embeds checks that categorical distributions don't have too many classes.
A categorical-type distribution is one which, e.g., returns the class label
rather than a one-hot encoding. E.g., `Categorical(probs)`.
Since distributions output samples in the same dtype as the parameters, we
must ensure that casting doesn't lose precision. That is, the
`parameter.dtype` implies a maximum number of classes. However, since shape is
`int32` and categorical variables are presumed to be indexes into a `Tensor`,
we must also ensure that the number of classes is no larger than the largest
possible `int32` index, i.e., `2**31-1`.
In other words the number of classes, `K`, must satisfy the following
condition:
```python
K <= min(
int(2**31 - 1), # Largest float as an index.
{
dtypes.float16: int(2**11), # Largest int as a float16.
dtypes.float32: int(2**24),
dtypes.float64: int(2**53),
}.get(categorical_param.dtype.base_dtype, 0))
```
Args:
categorical_param: Floating-point `Tensor` representing parameters of
distribution over categories. The rightmost shape is presumed to be the
number of categories.
name: A name for this operation (optional).
Returns:
categorical_param: Input `Tensor` with appropriate assertions embedded.
Raises:
TypeError: if `categorical_param` has an unknown `dtype`.
ValueError: if we can statically identify `categorical_param` as being too
large (for being closed under int32/float casting)."
9715,embed_check_integer_casting_closed,tensorflow/tensorflow/python/ops/distributions/util.py,397,function,"Ensures integers remain unaffected despite casting to/from int/float types.
Example integer-types: `uint8`, `int32`, `bool`.
Example floating-types: `float32`, `float64`.
The largest possible integer representable by an IEEE754 floating-point is
`2**(1 + mantissa_bits)` yet the largest possible integer as an int-type is
`2**(bits - 1) - 1`. This function ensures that a `Tensor` purporting to have
integer-form values can be cast to some other type without loss of precision.
The smallest representable integer is the negative of the largest
representable integer, except for types: `uint8`, `uint16`, `bool`. For these
types, the smallest representable integer is `0`.
Args:
x: `Tensor` representing integer-form values.
target_dtype: TF `dtype` under which `x` should have identical values.
assert_nonnegative: `bool` indicating `x` should contain nonnegative values.
name: A name for this operation (optional).
Returns:
x: Input `Tensor` with appropriate assertions embedded.
Raises:
TypeError: if `x` is neither integer- nor floating-type.
TypeError: if `target_dtype` is neither integer- nor floating-type.
TypeError: if neither `x` nor `target_dtype` are integer-type."
9716,log_combinations,tensorflow/tensorflow/python/ops/distributions/util.py,488,function,"Multinomial coefficient.
Given `n` and `counts`, where `counts` has last dimension `k`, we compute
the multinomial coefficient as:
```n! / sum_i n_i!```
where `i` runs over all `k` classes.
Args:
n: Floating-point `Tensor` broadcastable with `counts`. This represents `n`
outcomes.
counts: Floating-point `Tensor` broadcastable with `n`. This represents
counts in `k` classes, where `k` is the last dimension of the tensor.
name: A name for this operation (optional).
Returns:
`Tensor` representing the multinomial coefficient between `n` and `counts`."
9717,matrix_diag_transform,tensorflow/tensorflow/python/ops/distributions/util.py,522,function,"Transform diagonal of [batch-]matrix, leave rest of matrix unchanged.
Create a trainable covariance defined by a Cholesky factor:
```python
# Transform network layer into 2 x 2 array.
matrix_values = tf.contrib.layers.fully_connected(activations, 4)
matrix = tf.reshape(matrix_values, (batch_size, 2, 2))
# Make the diagonal positive. If the upper triangle was zero, this would be a
# valid Cholesky factor.
chol = matrix_diag_transform(matrix, transform=tf.nn.softplus)
# LinearOperatorLowerTriangular ignores the upper triangle.
operator = LinearOperatorLowerTriangular(chol)
```
Example of heteroskedastic 2-D linear regression.
```python
tfd = tfp.distributions
# Get a trainable Cholesky factor.
matrix_values = tf.contrib.layers.fully_connected(activations, 4)
matrix = tf.reshape(matrix_values, (batch_size, 2, 2))
chol = matrix_diag_transform(matrix, transform=tf.nn.softplus)
# Get a trainable mean.
mu = tf.contrib.layers.fully_connected(activations, 2)
# This is a fully trainable multivariate normal!
dist = tfd.MultivariateNormalTriL(mu, chol)
# Standard log loss. Minimizing this will ""train"" mu and chol, and then dist
# will be a distribution predicting labels as multivariate Gaussians.
loss = -1 * tf.reduce_mean(dist.log_prob(labels))
```
Args:
matrix: Rank `R` `Tensor`, `R >= 2`, where the last two dimensions are
equal.
transform: Element-wise function mapping `Tensors` to `Tensors`. To be
applied to the diagonal of `matrix`. If `None`, `matrix` is returned
unchanged. Defaults to `None`.
name: A name to give created ops. Defaults to ""matrix_diag_transform"".
Returns:
A `Tensor` with same shape and `dtype` as `matrix`."
9718,rotate_transpose,tensorflow/tensorflow/python/ops/distributions/util.py,584,function,"Circularly moves dims left or right.
Effectively identical to:
```python
numpy.transpose(x, numpy.roll(numpy.arange(len(x.shape)), shift))
```
When `validate_args=False` additional graph-runtime checks are
performed. These checks entail moving data from to GPU to CPU.
Example:
```python
x = tf.random.normal([1, 2, 3, 4]) # Tensor of shape [1, 2, 3, 4].
rotate_transpose(x, -1).shape == [2, 3, 4, 1]
rotate_transpose(x, -2).shape == [3, 4, 1, 2]
rotate_transpose(x, 1).shape == [4, 1, 2, 3]
rotate_transpose(x, 2).shape == [3, 4, 1, 2]
rotate_transpose(x, 7).shape == rotate_transpose(x, 3).shape # [2, 3, 4, 1]
rotate_transpose(x, -7).shape == rotate_transpose(x, -3).shape # [4, 1, 2, 3]
```
Args:
x: `Tensor`.
shift: `Tensor`. Number of dimensions to transpose left (shift<0) or
transpose right (shift>0).
name: Python `str`. The name to give this op.
Returns:
rotated_x: Input `Tensor` with dimensions circularly rotated by shift.
Raises:
TypeError: if shift is not integer type."
9719,pick_vector,tensorflow/tensorflow/python/ops/distributions/util.py,660,function,"Picks possibly different length row `Tensor`s based on condition.
Value `Tensor`s should have exactly one dimension.
If `cond` is a python Boolean or `tf.constant` then either `true_vector` or
`false_vector` is immediately returned. I.e., no graph nodes are created and
no validation happens.
Args:
cond: `Tensor`. Must have `dtype=tf.bool` and be scalar.
true_vector: `Tensor` of one dimension. Returned when cond is `True`.
false_vector: `Tensor` of one dimension. Returned when cond is `False`.
name: Python `str`. The name to give this op.
Example: ```python pick_vector(tf.less(0, 5), tf.range(10, 12), tf.range(15,
18)) # [10, 11] pick_vector(tf.less(5, 0), tf.range(10, 12), tf.range(15,
18)) # [15, 16, 17] ```
Returns:
true_or_false_vector: `Tensor`.
Raises:
TypeError: if `cond.dtype != tf.bool`
TypeError: if `cond` is not a constant and
`true_vector.dtype != false_vector.dtype`"
9720,prefer_static_broadcast_shape,tensorflow/tensorflow/python/ops/distributions/util.py,706,function,"Convenience function which statically broadcasts shape when possible.
Args:
shape1: `1-D` integer `Tensor`. Already converted to tensor!
shape2: `1-D` integer `Tensor`. Already converted to tensor!
name: A string name to prepend to created ops.
Returns:
The broadcast shape, either as `TensorShape` (if broadcast can be done
statically), or as a `Tensor`."
9721,prefer_static_rank,tensorflow/tensorflow/python/ops/distributions/util.py,751,function,"Return static rank of tensor `x` if available, else `tf.rank(x)`.
Args:
x: `Tensor` (already converted).
Returns:
Numpy array (if static rank is obtainable), else `Tensor`."
9722,prefer_static_shape,tensorflow/tensorflow/python/ops/distributions/util.py,763,function,"Return static shape of tensor `x` if available, else `tf.shape(x)`.
Args:
x: `Tensor` (already converted).
Returns:
Numpy array (if static shape is obtainable), else `Tensor`."
9723,prefer_static_value,tensorflow/tensorflow/python/ops/distributions/util.py,775,function,"Return static value of tensor `x` if available, else `x`.
Args:
x: `Tensor` (already converted).
Returns:
Numpy array (if static value is obtainable), else `Tensor`."
9724,gen_new_seed,tensorflow/tensorflow/python/ops/distributions/util.py,790,function,"Generate a new seed, from the given seed and salt."
9725,fill_triangular,tensorflow/tensorflow/python/ops/distributions/util.py,798,function,"Creates a (batch of) triangular matrix from a vector of inputs.
Created matrix can be lower- or upper-triangular. (It is more efficient to
create the matrix as upper or lower, rather than transpose.)
Triangular matrix elements are filled in a clockwise spiral. See example,
below.
If `x.get_shape()` is `[b1, b2, ..., bB, d]` then the output shape is
`[b1, b2, ..., bB, n, n]` where `n` is such that `d = n(n+1)/2`, i.e.,
`n = int(np.sqrt(0.25 + 2. * m) - 0.5)`.
Example:
```python
fill_triangular([1, 2, 3, 4, 5, 6])
# ==> [[4, 0, 0],
# [6, 5, 0],
# [3, 2, 1]]
fill_triangular([1, 2, 3, 4, 5, 6], upper=True)
# ==> [[1, 2, 3],
# [0, 5, 6],
# [0, 0, 4]]
```
For comparison, a pure numpy version of this function can be found in
`util_test.py`, function `_fill_triangular`.
Args:
x: `Tensor` representing lower (or upper) triangular elements.
upper: Python `bool` representing whether output matrix should be upper
triangular (`True`) or lower triangular (`False`, default).
name: Python `str`. The name to give this op.
Returns:
tril: `Tensor` with lower (or upper) triangular elements filled from `x`.
Raises:
ValueError: if `x` cannot be mapped to a triangular matrix."
9726,fill_triangular_inverse,tensorflow/tensorflow/python/ops/distributions/util.py,913,function,"Creates a vector from a (batch of) triangular matrix.
The vector is created from the lower-triangular or upper-triangular portion
depending on the value of the parameter `upper`.
If `x.shape` is `[b1, b2, ..., bB, n, n]` then the output shape is
`[b1, b2, ..., bB, d]` where `d = n (n + 1) / 2`.
Example:
```python
fill_triangular_inverse(
[[4, 0, 0],
[6, 5, 0],
[3, 2, 1]])
# ==> [1, 2, 3, 4, 5, 6]
fill_triangular_inverse(
[[1, 2, 3],
[0, 5, 6],
[0, 0, 4]], upper=True)
# ==> [1, 2, 3, 4, 5, 6]
```
Args:
x: `Tensor` representing lower (or upper) triangular elements.
upper: Python `bool` representing whether output matrix should be upper
triangular (`True`) or lower triangular (`False`, default).
name: Python `str`. The name to give this op.
Returns:
flat_tril: (Batch of) vector-shaped `Tensor` representing vectorized lower
(or upper) triangular elements from `x`."
9727,tridiag,tensorflow/tensorflow/python/ops/distributions/util.py,982,function,"Creates a matrix with values set above, below, and on the diagonal.
Example:
```python
tridiag(below=[1., 2., 3.],
diag=[4., 5., 6., 7.],
above=[8., 9., 10.])
# ==> array([[ 4., 8., 0., 0.],
# [ 1., 5., 9., 0.],
# [ 0., 2., 6., 10.],
# [ 0., 0., 3., 7.]], dtype=float32)
```
Warning: This Op is intended for convenience, not efficiency.
Args:
below: `Tensor` of shape `[B1, ..., Bb, d-1]` corresponding to the below
diagonal part. `None` is logically equivalent to `below = 0`.
diag: `Tensor` of shape `[B1, ..., Bb, d]` corresponding to the diagonal
part. `None` is logically equivalent to `diag = 0`.
above: `Tensor` of shape `[B1, ..., Bb, d-1]` corresponding to the above
diagonal part. `None` is logically equivalent to `above = 0`.
name: Python `str`. The name to give this op.
Returns:
tridiag: `Tensor` with values set above, below and on the diagonal.
Raises:
ValueError: if all inputs are `None`."
9728,reduce_weighted_logsumexp,tensorflow/tensorflow/python/ops/distributions/util.py,1050,function,"Computes `log(abs(sum(weight * exp(elements across tensor dimensions))))`.
If all weights `w` are known to be positive, it is more efficient to directly
use `reduce_logsumexp`, i.e., `tf.reduce_logsumexp(logx + tf.math.log(w))` is
more
efficient than `du.reduce_weighted_logsumexp(logx, w)`.
Reduces `input_tensor` along the dimensions given in `axis`.
Unless `keep_dims` is true, the rank of the tensor is reduced by 1 for each
entry in `axis`. If `keep_dims` is true, the reduced dimensions
are retained with length 1.
If `axis` has no entries, all dimensions are reduced, and a
tensor with a single element is returned.
This function is more numerically stable than log(sum(w * exp(input))). It
avoids overflows caused by taking the exp of large inputs and underflows
caused by taking the log of small inputs.
For example:
```python
x = tf.constant([[0., 0, 0],
[0, 0, 0]])
w = tf.constant([[-1., 1, 1],
[1, 1, 1]])
du.reduce_weighted_logsumexp(x, w)
# ==> log(-1*1 + 1*1 + 1*1 + 1*1 + 1*1 + 1*1) = log(4)
du.reduce_weighted_logsumexp(x, w, axis=0)
# ==> [log(-1+1), log(1+1), log(1+1)]
du.reduce_weighted_logsumexp(x, w, axis=1)
# ==> [log(-1+1+1), log(1+1+1)]
du.reduce_weighted_logsumexp(x, w, axis=1, keep_dims=True)
# ==> [[log(-1+1+1)], [log(1+1+1)]]
du.reduce_weighted_logsumexp(x, w, axis=[0, 1])
# ==> log(-1+5)
```
Args:
logx: The tensor to reduce. Should have numeric type.
w: The weight tensor. Should have numeric type identical to `logx`.
axis: The dimensions to reduce. If `None` (the default), reduces all
dimensions. Must be in the range `[-rank(input_tensor),
rank(input_tensor))`.
keep_dims: If true, retains reduced dimensions with length 1.
return_sign: If `True`, returns the sign of the result.
name: A name for the operation (optional).
Returns:
lswe: The `log(abs(sum(weight * exp(x))))` reduced tensor.
sign: (Optional) The sign of `sum(weight * exp(x))`."
9729,softplus_inverse,tensorflow/tensorflow/python/ops/distributions/util.py,1148,function,"Computes the inverse softplus, i.e., x = softplus_inverse(softplus(x)).
Mathematically this op is equivalent to:
```none
softplus_inverse = log(exp(x) - 1.)
```
Args:
x: `Tensor`. Non-negative (not enforced), floating-point.
name: A name for the operation (optional).
Returns:
`Tensor`. Has the same type/shape as input `x`."
9730,dimension_size,tensorflow/tensorflow/python/ops/distributions/util.py,1205,function,Returns the size of a specific dimension.
9731,process_quadrature_grid_and_probs,tensorflow/tensorflow/python/ops/distributions/util.py,1216,function,"Validates quadrature grid, probs or computes them as necessary.
Args:
quadrature_grid_and_probs: Python pair of `float`-like `Tensor`s
representing the sample points and the corresponding (possibly
normalized) weight. When `None`, defaults to:
`np.polynomial.hermite.hermgauss(deg=8)`.
dtype: The expected `dtype` of `grid` and `probs`.
validate_args: Python `bool`, default `False`. When `True` distribution
parameters are checked for validity despite possibly degrading runtime
performance. When `False` invalid inputs may silently render incorrect
outputs.
name: Python `str` name prefixed to Ops created by this class.
Returns:
quadrature_grid_and_probs: Python pair of `float`-like `Tensor`s
representing the sample points and the corresponding (possibly
normalized) weight.
Raises:
ValueError: if `quadrature_grid_and_probs is not None` and
`len(quadrature_grid_and_probs[0]) != len(quadrature_grid_and_probs[1])`"
9732,pad,tensorflow/tensorflow/python/ops/distributions/util.py,1283,function,"Pads `value` to the front and/or back of a `Tensor` dim, `count` times.
Args:
x: `Tensor` input.
axis: Scalar `int`-like `Tensor` representing the single dimension to pad.
(Negative indexing is supported.)
front: Python `bool`; if `True` the beginning of the `axis` dimension is
padded with `value`, `count` times. If `False` no front padding is made.
back: Python `bool`; if `True` the end of the `axis` dimension is padded
with `value`, `count` times. If `False` no end padding is made.
value: Scalar `int`-like `Tensor` representing the actual value added to the
front and/or back of the `axis` dimension of `x`.
count: Scalar `int`-like `Tensor` representing number of elements added to
the front and/or back of the `axis` dimension of `x`. E.g., if `front =
back = True` then `2 * count` elements are added.
name: Python `str` name prefixed to Ops created by this function.
Returns:
pad: The padded version of input `x`.
Raises:
ValueError: if both `front` and `back` are `False`.
TypeError: if `count` is not `int`-like."
9733,parent_frame_arguments,tensorflow/tensorflow/python/ops/distributions/util.py,1354,function,"Returns parent frame arguments.
When called inside a function, returns a dictionary with the caller's function
arguments. These are positional arguments and keyword arguments (**kwargs),
while variable arguments (*varargs) are excluded.
When called at global scope, this will return an empty dictionary, since there
are no arguments.
WARNING: If caller function argument names are overloaded before invoking
this method, then values will reflect the overloaded value. For this reason,
we recommend calling `parent_frame_arguments` at the beginning of the
function."
9734,AppendDocstring,tensorflow/tensorflow/python/ops/distributions/util.py,1391,class,"Helper class to promote private subclass docstring to public counterpart.
Example:
```python
class TransformedDistribution(Distribution):
@distribution_util.AppendDocstring(
additional_note=""A special note!"",
kwargs_dict={""foo"": ""An extra arg.""})
def _prob(self, y, foo=None):
pass
```
In this case, the `AppendDocstring` decorator appends the `additional_note` to
the docstring of `prob` (not `_prob`) and adds a new `kwargs`
section with each dictionary item as a bullet-point.
For a more detailed example, see `TransformedDistribution`."
9735,logdet,tensorflow/tensorflow/python/ops/linalg/linalg_impl.py,70,function,"Computes log of the determinant of a hermitian positive definite matrix.
```python
# Compute the determinant of a matrix while reducing the chance of over- or
underflow:
A = ... # shape 10 x 10
det = tf.exp(tf.linalg.logdet(A)) # scalar
```
Args:
matrix: A `Tensor`. Must be `float16`, `float32`, `float64`, `complex64`,
or `complex128` with shape `[..., M, M]`.
name: A name to give this `Op`. Defaults to `logdet`.
Returns:
The natural log of the determinant of `matrix`.
@compatibility(numpy)
Equivalent to numpy.linalg.slogdet, although no sign is returned since only
hermitian positive definite matrices are supported.
@end_compatibility"
9736,adjoint,tensorflow/tensorflow/python/ops/linalg/linalg_impl.py,104,function,"Transposes the last two dimensions of and conjugates tensor `matrix`.
For example:
```python
x = tf.constant([[1 + 1j, 2 + 2j, 3 + 3j],
[4 + 4j, 5 + 5j, 6 + 6j]])
tf.linalg.adjoint(x) # [[1 - 1j, 4 - 4j],
# [2 - 2j, 5 - 5j],
# [3 - 3j, 6 - 6j]]
```
Args:
matrix: A `Tensor`. Must be `float16`, `float32`, `float64`, `complex64`,
or `complex128` with shape `[..., M, M]`.
name: A name to give this `Op` (optional).
Returns:
The adjoint (a.k.a. Hermitian transpose a.k.a. conjugate transpose) of
matrix."
9737,matrix_exponential,tensorflow/tensorflow/python/ops/linalg/linalg_impl.py,234,function,"Computes the matrix exponential of one or more square matrices.
exp(A) = \sum_{n=0}^\infty A^n/n!
The exponential is computed using a combination of the scaling and squaring
method and the Pade approximation. Details can be found in:
Nicholas J. Higham, ""The scaling and squaring method for the matrix
exponential revisited,"" SIAM J. Matrix Anal. Applic., 26:1179-1193, 2005.
The input is a tensor of shape `[..., M, M]` whose inner-most 2 dimensions
form square matrices. The output is a tensor of the same shape as the input
containing the exponential for all input submatrices `[..., :, :]`.
Args:
input: A `Tensor`. Must be `float16`, `float32`, `float64`, `complex64`, or
`complex128` with shape `[..., M, M]`.
name: A name to give this `Op` (optional).
Returns:
the matrix exponential of the input.
Raises:
ValueError: An unsupported type is provided as input.
@compatibility(scipy)
Equivalent to scipy.linalg.expm
@end_compatibility"
9738,banded_triangular_solve,tensorflow/tensorflow/python/ops/linalg/linalg_impl.py,344,function,"Solve triangular systems of equations with a banded solver.
`bands` is a tensor of shape `[..., K, M]`, where `K` represents the number
of bands stored. This corresponds to a batch of `M` by `M` matrices, whose
`K` subdiagonals (when `lower` is `True`) are stored.
This operator broadcasts the batch dimensions of `bands` and the batch
dimensions of `rhs`.
Examples:
Storing 2 bands of a 3x3 matrix.
Note that first element in the second row is ignored due to
the 'LEFT_RIGHT' padding.
>>> x = [[2., 3., 4.], [1., 2., 3.]]
>>> x2 = [[2., 3., 4.], [10000., 2., 3.]]
>>> y = tf.zeros([3, 3])
>>> z = tf.linalg.set_diag(y, x, align='LEFT_RIGHT', k=(-1, 0))
>>> z
<tf.Tensor: shape=(3, 3), dtype=float32, numpy=
array([[2., 0., 0.],
[2., 3., 0.],
[0., 3., 4.]], dtype=float32)>
>>> soln = tf.linalg.banded_triangular_solve(x, tf.ones([3, 1]))
>>> soln
<tf.Tensor: shape=(3, 1), dtype=float32, numpy=
array([[0.5 ],
[0. ],
[0.25]], dtype=float32)>
>>> are_equal = soln == tf.linalg.banded_triangular_solve(x2, tf.ones([3, 1]))
>>> tf.reduce_all(are_equal).numpy()
True
>>> are_equal = soln == tf.linalg.triangular_solve(z, tf.ones([3, 1]))
>>> tf.reduce_all(are_equal).numpy()
True
Storing 2 superdiagonals of a 4x4 matrix. Because of the 'LEFT_RIGHT' padding
the last element of the first row is ignored.
>>> x = [[2., 3., 4., 5.], [-1., -2., -3., -4.]]
>>> y = tf.zeros([4, 4])
>>> z = tf.linalg.set_diag(y, x, align='LEFT_RIGHT', k=(0, 1))
>>> z
<tf.Tensor: shape=(4, 4), dtype=float32, numpy=
array([[-1., 2., 0., 0.],
[ 0., -2., 3., 0.],
[ 0., 0., -3., 4.],
[ 0., 0., -0., -4.]], dtype=float32)>
>>> soln = tf.linalg.banded_triangular_solve(x, tf.ones([4, 1]), lower=False)
>>> soln
<tf.Tensor: shape=(4, 1), dtype=float32, numpy=
array([[-4. ],
[-1.5 ],
[-0.6666667],
[-0.25 ]], dtype=float32)>
>>> are_equal = (soln == tf.linalg.triangular_solve(
... z, tf.ones([4, 1]), lower=False))
>>> tf.reduce_all(are_equal).numpy()
True
Args:
bands: A `Tensor` describing the bands of the left hand side, with shape
`[..., K, M]`. The `K` rows correspond to the diagonal to the `K - 1`-th
diagonal (the diagonal is the top row) when `lower` is `True` and
otherwise the `K - 1`-th superdiagonal to the diagonal (the diagonal is
the bottom row) when `lower` is `False`. The bands are stored with
'LEFT_RIGHT' alignment, where the superdiagonals are padded on the right
and subdiagonals are padded on the left. This is the alignment cuSPARSE
uses. See `tf.linalg.set_diag` for more details.
rhs: A `Tensor` of shape [..., M] or [..., M, N] and with the same dtype as
`diagonals`. Note that if the shape of `rhs` and/or `diags` isn't known
statically, `rhs` will be treated as a matrix rather than a vector.
lower: An optional `bool`. Defaults to `True`. Boolean indicating whether
`bands` represents a lower or upper triangular matrix.
adjoint: An optional `bool`. Defaults to `False`. Boolean indicating whether
to solve with the matrix's block-wise adjoint.
name: A name to give this `Op` (optional).
Returns:
A `Tensor` of shape [..., M] or [..., M, N] containing the solutions."
9739,tridiagonal_solve,tensorflow/tensorflow/python/ops/linalg/linalg_impl.py,441,function,"Solves tridiagonal systems of equations.
The input can be supplied in various formats: `matrix`, `sequence` and
`compact`, specified by the `diagonals_format` arg.
In `matrix` format, `diagonals` must be a tensor of shape `[..., M, M]`, with
two inner-most dimensions representing the square tridiagonal matrices.
Elements outside of the three diagonals will be ignored.
In `sequence` format, `diagonals` are supplied as a tuple or list of three
tensors of shapes `[..., N]`, `[..., M]`, `[..., N]` representing
superdiagonals, diagonals, and subdiagonals, respectively. `N` can be either
`M-1` or `M`; in the latter case, the last element of superdiagonal and the
first element of subdiagonal will be ignored.
In `compact` format the three diagonals are brought together into one tensor
of shape `[..., 3, M]`, with last two dimensions containing superdiagonals,
diagonals, and subdiagonals, in order. Similarly to `sequence` format,
elements `diagonals[..., 0, M-1]` and `diagonals[..., 2, 0]` are ignored.
The `compact` format is recommended as the one with best performance. In case
you need to cast a tensor into a compact format manually, use `tf.gather_nd`.
An example for a tensor of shape [m, m]:
```python
rhs = tf.constant([...])
matrix = tf.constant([[...]])
m = matrix.shape[0]
dummy_idx = [0, 0] # An arbitrary element to use as a dummy
indices = [[[i, i + 1] for i in range(m - 1)] + [dummy_idx], # Superdiagonal
[[i, i] for i in range(m)], # Diagonal
[dummy_idx] + [[i + 1, i] for i in range(m - 1)]] # Subdiagonal
diagonals=tf.gather_nd(matrix, indices)
x = tf.linalg.tridiagonal_solve(diagonals, rhs)
```
Regardless of the `diagonals_format`, `rhs` is a tensor of shape `[..., M]` or
`[..., M, K]`. The latter allows to simultaneously solve K systems with the
same left-hand sides and K different right-hand sides. If `transpose_rhs`
is set to `True` the expected shape is `[..., M]` or `[..., K, M]`.
The batch dimensions, denoted as `...`, must be the same in `diagonals` and
`rhs`.
The output is a tensor of the same shape as `rhs`: either `[..., M]` or
`[..., M, K]`.
The op isn't guaranteed to raise an error if the input matrix is not
invertible. `tf.debugging.check_numerics` can be applied to the output to
detect invertibility problems.
**Note**: with large batch sizes, the computation on the GPU may be slow, if
either `partial_pivoting=True` or there are multiple right-hand sides
(`K > 1`). If this issue arises, consider if it's possible to disable pivoting
and have `K = 1`, or, alternatively, consider using CPU.
On CPU, solution is computed via Gaussian elimination with or without partial
pivoting, depending on `partial_pivoting` parameter. On GPU, Nvidia's cuSPARSE
library is used: https://docs.nvidia.com/cuda/cusparse/index.html#gtsv
Args:
diagonals: A `Tensor` or tuple of `Tensor`s describing left-hand sides. The
shape depends of `diagonals_format`, see description above. Must be
`float32`, `float64`, `complex64`, or `complex128`.
rhs: A `Tensor` of shape [..., M] or [..., M, K] and with the same dtype as
`diagonals`. Note that if the shape of `rhs` and/or `diags` isn't known
statically, `rhs` will be treated as a matrix rather than a vector.
diagonals_format: one of `matrix`, `sequence`, or `compact`. Default is
`compact`.
transpose_rhs: If `True`, `rhs` is transposed before solving (has no effect
if the shape of rhs is [..., M]).
conjugate_rhs: If `True`, `rhs` is conjugated before solving.
name: A name to give this `Op` (optional).
partial_pivoting: whether to perform partial pivoting. `True` by default.
Partial pivoting makes the procedure more stable, but slower. Partial
pivoting is unnecessary in some cases, including diagonally dominant and
symmetric positive definite matrices (see e.g. theorem 9.12 in [1]).
Returns:
A `Tensor` of shape [..., M] or [..., M, K] containing the solutions.
Raises:
ValueError: An unsupported type is provided as input, or when the input
tensors have incorrect shapes.
UnimplementedError: Whenever `partial_pivoting` is true and the backend is
XLA.
[1] Nicholas J. Higham (2002). Accuracy and Stability of Numerical Algorithms:
Second Edition. SIAM. p. 175. ISBN 978-0-89871-802-7."
9740,tridiagonal_matmul,tensorflow/tensorflow/python/ops/linalg/linalg_impl.py,643,function,"Multiplies tridiagonal matrix by matrix.
`diagonals` is representation of 3-diagonal NxN matrix, which depends on
`diagonals_format`.
In `matrix` format, `diagonals` must be a tensor of shape `[..., M, M]`, with
two inner-most dimensions representing the square tridiagonal matrices.
Elements outside of the three diagonals will be ignored.
If `sequence` format, `diagonals` is list or tuple of three tensors:
`[superdiag, maindiag, subdiag]`, each having shape [..., M]. Last element
of `superdiag` first element of `subdiag` are ignored.
In `compact` format the three diagonals are brought together into one tensor
of shape `[..., 3, M]`, with last two dimensions containing superdiagonals,
diagonals, and subdiagonals, in order. Similarly to `sequence` format,
elements `diagonals[..., 0, M-1]` and `diagonals[..., 2, 0]` are ignored.
The `sequence` format is recommended as the one with the best performance.
`rhs` is matrix to the right of multiplication. It has shape `[..., M, N]`.
Example:
```python
superdiag = tf.constant([-1, -1, 0], dtype=tf.float64)
maindiag = tf.constant([2, 2, 2], dtype=tf.float64)
subdiag = tf.constant([0, -1, -1], dtype=tf.float64)
diagonals = [superdiag, maindiag, subdiag]
rhs = tf.constant([[1, 1], [1, 1], [1, 1]], dtype=tf.float64)
x = tf.linalg.tridiagonal_matmul(diagonals, rhs, diagonals_format='sequence')
```
Args:
diagonals: A `Tensor` or tuple of `Tensor`s describing left-hand sides. The
shape depends of `diagonals_format`, see description above. Must be
`float32`, `float64`, `complex64`, or `complex128`.
rhs: A `Tensor` of shape [..., M, N] and with the same dtype as `diagonals`.
diagonals_format: one of `sequence`, or `compact`. Default is `compact`.
name: A name to give this `Op` (optional).
Returns:
A `Tensor` of shape [..., M, N] containing the result of multiplication.
Raises:
ValueError: An unsupported type is provided as input, or when the input
tensors have incorrect shapes."
9741,matrix_rank,tensorflow/tensorflow/python/ops/linalg/linalg_impl.py,741,function,"Compute the matrix rank of one or more matrices.
Arguments:
a: (Batch of) `float`-like matrix-shaped `Tensor`(s) which are to be
pseudo-inverted.
tol: Threshold below which the singular value is counted as 'zero'.
Default value: `None` (i.e., `eps * max(rows, cols) * max(singular_val)`).
validate_args: When `True`, additional assertions might be embedded in the
graph.
Default value: `False` (i.e., no graph assertions are added).
name: Python `str` prefixed to ops created by this function.
Default value: 'matrix_rank'.
Returns:
matrix_rank: (Batch of) `int32` scalars representing the number of non-zero
singular values."
9742,pinv,tensorflow/tensorflow/python/ops/linalg/linalg_impl.py,780,function,"Compute the Moore-Penrose pseudo-inverse of one or more matrices.
Calculate the [generalized inverse of a matrix](
https://en.wikipedia.org/wiki/Moore%E2%80%93Penrose_inverse) using its
singular-value decomposition (SVD) and including all large singular values.
The pseudo-inverse of a matrix `A`, is defined as: 'the matrix that 'solves'
[the least-squares problem] `A @ x = b`,' i.e., if `x_hat` is a solution, then
`A_pinv` is the matrix such that `x_hat = A_pinv @ b`. It can be shown that if
`U @ Sigma @ V.T = A` is the singular value decomposition of `A`, then
`A_pinv = V @ inv(Sigma) U^T`. [(Strang, 1980)][1]
This function is analogous to [`numpy.linalg.pinv`](
https://docs.scipy.org/doc/numpy/reference/generated/numpy.linalg.pinv.html).
It differs only in default value of `rcond`. In `numpy.linalg.pinv`, the
default `rcond` is `1e-15`. Here the default is
`10. * max(num_rows, num_cols) * np.finfo(dtype).eps`.
Args:
a: (Batch of) `float`-like matrix-shaped `Tensor`(s) which are to be
pseudo-inverted.
rcond: `Tensor` of small singular value cutoffs. Singular values smaller
(in modulus) than `rcond` * largest_singular_value (again, in modulus) are
set to zero. Must broadcast against `tf.shape(a)[:-2]`.
Default value: `10. * max(num_rows, num_cols) * np.finfo(a.dtype).eps`.
validate_args: When `True`, additional assertions might be embedded in the
graph.
Default value: `False` (i.e., no graph assertions are added).
name: Python `str` prefixed to ops created by this function.
Default value: 'pinv'.
Returns:
a_pinv: (Batch of) pseudo-inverse of input `a`. Has same shape as `a` except
rightmost two dimensions are transposed.
Raises:
TypeError: if input `a` does not have `float`-like `dtype`.
ValueError: if input `a` has fewer than 2 dimensions.
#### Examples
```python
import tensorflow as tf
import tensorflow_probability as tfp
a = tf.constant([[1., 0.4, 0.5],
[0.4, 0.2, 0.25],
[0.5, 0.25, 0.35]])
tf.matmul(tf.linalg..pinv(a), a)
# ==> array([[1., 0., 0.],
[0., 1., 0.],
[0., 0., 1.]], dtype=float32)
a = tf.constant([[1., 0.4, 0.5, 1.],
[0.4, 0.2, 0.25, 2.],
[0.5, 0.25, 0.35, 3.]])
tf.matmul(tf.linalg..pinv(a), a)
# ==> array([[ 0.76, 0.37, 0.21, -0.02],
[ 0.37, 0.43, -0.33, 0.02],
[ 0.21, -0.33, 0.81, 0.01],
[-0.02, 0.02, 0.01, 1. ]], dtype=float32)
```
#### References
[1]: G. Strang. 'Linear Algebra and Its Applications, 2nd Ed.' Academic Press,
Inc., 1980, pp. 139-142."
9743,lu_solve,tensorflow/tensorflow/python/ops/linalg/linalg_impl.py,910,function,"Solves systems of linear eqns `A X = RHS`, given LU factorizations.
Note: this function does not verify the implied matrix is actually invertible
nor is this condition checked even when `validate_args=True`.
Args:
lower_upper: `lu` as returned by `tf.linalg.lu`, i.e., if `matmul(P,
matmul(L, U)) = X` then `lower_upper = L + U - eye`.
perm: `p` as returned by `tf.linag.lu`, i.e., if `matmul(P, matmul(L, U)) =
X` then `perm = argmax(P)`.
rhs: Matrix-shaped float `Tensor` representing targets for which to solve;
`A X = RHS`. To handle vector cases, use: `lu_solve(..., rhs[...,
tf.newaxis])[..., 0]`.
validate_args: Python `bool` indicating whether arguments should be checked
for correctness. Note: this function does not verify the implied matrix is
actually invertible, even when `validate_args=True`.
Default value: `False` (i.e., don't validate arguments).
name: Python `str` name given to ops managed by this object.
Default value: `None` (i.e., 'lu_solve').
Returns:
x: The `X` in `A @ X = RHS`.
#### Examples
```python
import numpy as np
import tensorflow as tf
import tensorflow_probability as tfp
x = [[[1., 2],
[3, 4]],
[[7, 8],
[3, 4]]]
inv_x = tf.linalg.lu_solve(*tf.linalg.lu(x), rhs=tf.eye(2))
tf.assert_near(tf.matrix_inverse(x), inv_x)
# ==> True
```"
9744,lu_matrix_inverse,tensorflow/tensorflow/python/ops/linalg/linalg_impl.py,1008,function,"Computes the inverse given the LU decomposition(s) of one or more matrices.
This op is conceptually identical to,
```python
inv_X = tf.lu_matrix_inverse(*tf.linalg.lu(X))
tf.assert_near(tf.matrix_inverse(X), inv_X)
# ==> True
```
Note: this function does not verify the implied matrix is actually invertible
nor is this condition checked even when `validate_args=True`.
Args:
lower_upper: `lu` as returned by `tf.linalg.lu`, i.e., if `matmul(P,
matmul(L, U)) = X` then `lower_upper = L + U - eye`.
perm: `p` as returned by `tf.linag.lu`, i.e., if `matmul(P, matmul(L, U)) =
X` then `perm = argmax(P)`.
validate_args: Python `bool` indicating whether arguments should be checked
for correctness. Note: this function does not verify the implied matrix is
actually invertible, even when `validate_args=True`.
Default value: `False` (i.e., don't validate arguments).
name: Python `str` name given to ops managed by this object.
Default value: `None` (i.e., 'lu_matrix_inverse').
Returns:
inv_x: The matrix_inv, i.e.,
`tf.matrix_inverse(tf.linalg.lu_reconstruct(lu, perm))`.
#### Examples
```python
import numpy as np
import tensorflow as tf
import tensorflow_probability as tfp
x = [[[3., 4], [1, 2]],
[[7., 8], [3, 4]]]
inv_x = tf.linalg.lu_matrix_inverse(*tf.linalg.lu(x))
tf.assert_near(tf.matrix_inverse(x), inv_x)
# ==> True
```"
9745,lu_reconstruct,tensorflow/tensorflow/python/ops/linalg/linalg_impl.py,1073,function,"The reconstruct one or more matrices from their LU decomposition(s).
Args:
lower_upper: `lu` as returned by `tf.linalg.lu`, i.e., if `matmul(P,
matmul(L, U)) = X` then `lower_upper = L + U - eye`.
perm: `p` as returned by `tf.linag.lu`, i.e., if `matmul(P, matmul(L, U)) =
X` then `perm = argmax(P)`.
validate_args: Python `bool` indicating whether arguments should be checked
for correctness.
Default value: `False` (i.e., don't validate arguments).
name: Python `str` name given to ops managed by this object.
Default value: `None` (i.e., 'lu_reconstruct').
Returns:
x: The original input to `tf.linalg.lu`, i.e., `x` as in,
`lu_reconstruct(*tf.linalg.lu(x))`.
#### Examples
```python
import numpy as np
import tensorflow as tf
import tensorflow_probability as tfp
x = [[[3., 4], [1, 2]],
[[7., 8], [3, 4]]]
x_reconstructed = tf.linalg.lu_reconstruct(*tf.linalg.lu(x))
tf.assert_near(x, x_reconstructed)
# ==> True
```"
9746,lu_reconstruct_assertions,tensorflow/tensorflow/python/ops/linalg/linalg_impl.py,1145,function,Returns list of assertions related to `lu_reconstruct` assumptions.
9747,LinearOperator,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,50,class,"Base class defining a [batch of] linear operator[s].
Subclasses of `LinearOperator` provide access to common methods on a
(batch) matrix, without the need to materialize the matrix. This allows:
* Matrix free computations
* Operators that take advantage of special structure, while providing a
consistent API to users.
#### Subclassing
To enable a public method, subclasses should implement the leading-underscore
version of the method. The argument signature should be identical except for
the omission of `name=""...""`. For example, to enable
`matmul(x, adjoint=False, name=""matmul"")` a subclass should implement
`_matmul(x, adjoint=False)`.
#### Performance contract
Subclasses should only implement the assert methods
(e.g. `assert_non_singular`) if they can be done in less than `O(N^3)`
time.
Class docstrings should contain an explanation of computational complexity.
Since this is a high-performance library, attention should be paid to detail,
and explanations can include constants as well as Big-O notation.
#### Shape compatibility
`LinearOperator` subclasses should operate on a [batch] matrix with
compatible shape. Class docstrings should define what is meant by compatible
shape. Some subclasses may not support batching.
Examples:
`x` is a batch matrix with compatible shape for `matmul` if
```
operator.shape = [B1,...,Bb] + [M, N], b >= 0,
x.shape = [B1,...,Bb] + [N, R]
```
`rhs` is a batch matrix with compatible shape for `solve` if
```
operator.shape = [B1,...,Bb] + [M, N], b >= 0,
rhs.shape = [B1,...,Bb] + [M, R]
```
#### Example docstring for subclasses.
This operator acts like a (batch) matrix `A` with shape
`[B1,...,Bb, M, N]` for some `b >= 0`. The first `b` indices index a
batch member. For every batch index `(i1,...,ib)`, `A[i1,...,ib, : :]` is
an `m x n` matrix. Again, this matrix `A` may not be materialized, but for
purposes of identifying and working with compatible arguments the shape is
relevant.
Examples:
```python
some_tensor = ... shape = ????
operator = MyLinOp(some_tensor)
operator.shape()
==> [2, 4, 4]
operator.log_abs_determinant()
==> Shape [2] Tensor
x = ... Shape [2, 4, 5] Tensor
operator.matmul(x)
==> Shape [2, 4, 5] Tensor
```
#### Shape compatibility
This operator acts on batch matrices with compatible shape.
FILL IN WHAT IS MEANT BY COMPATIBLE SHAPE
#### Performance
FILL THIS IN
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning:
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9748,dtype,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,225,method,The `DType` of `Tensor`s handled by this `LinearOperator`.
9749,name,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,230,method,Name prepended to all ops created by this `LinearOperator`.
9750,graph_parents,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,236,method,List of graph dependencies of this `LinearOperator`.
9751,is_non_singular,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,241,method,
9752,is_self_adjoint,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,245,method,
9753,is_positive_definite,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,249,method,
9754,is_square,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,253,method,Return `True/False` depending on if this operator is square.
9755,shape,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,272,method,"`TensorShape` of this `LinearOperator`.
If this operator acts like the batch matrix `A` with
`A.shape = [B1,...,Bb, M, N]`, then this returns
`TensorShape([B1,...,Bb, M, N])`, equivalent to `A.shape`.
Returns:
`TensorShape`, statically determined, may be undefined."
9756,shape_tensor,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,290,method,"Shape of this `LinearOperator`, determined at runtime.
If this operator acts like the batch matrix `A` with
`A.shape = [B1,...,Bb, M, N]`, then this returns a `Tensor` holding
`[B1,...,Bb, M, N]`, equivalent to `tf.shape(A)`.
Args:
name: A name for this `Op`.
Returns:
`int32` `Tensor`"
9757,batch_shape,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,311,method,"`TensorShape` of batch dimensions of this `LinearOperator`.
If this operator acts like the batch matrix `A` with
`A.shape = [B1,...,Bb, M, N]`, then this returns
`TensorShape([B1,...,Bb])`, equivalent to `A.shape[:-2]`
Returns:
`TensorShape`, statically determined, may be undefined."
9758,batch_shape_tensor,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,324,method,"Shape of batch dimensions of this operator, determined at runtime.
If this operator acts like the batch matrix `A` with
`A.shape = [B1,...,Bb, M, N]`, then this returns a `Tensor` holding
`[B1,...,Bb]`.
Args:
name: A name for this `Op`.
Returns:
`int32` `Tensor`"
9759,tensor_rank,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,352,method,"Rank (in the sense of tensors) of matrix corresponding to this operator.
If this operator acts like the batch matrix `A` with
`A.shape = [B1,...,Bb, M, N]`, then this returns `b + 2`.
Args:
name: A name for this `Op`.
Returns:
Python integer, or None if the tensor rank is undefined."
9760,tensor_rank_tensor,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,368,method,"Rank (in the sense of tensors) of matrix corresponding to this operator.
If this operator acts like the batch matrix `A` with
`A.shape = [B1,...,Bb, M, N]`, then this returns `b + 2`.
Args:
name: A name for this `Op`.
Returns:
`int32` `Tensor`, determined at runtime."
9761,range_dimension,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,438,method,"Dimension (in the sense of vector spaces) of the range of this operator.
If this operator acts like the batch matrix `A` with
`A.shape = [B1,...,Bb, M, N]`, then this returns `M`.
Returns:
`Dimension` object."
9762,range_dimension_tensor,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,453,method,"Dimension (in the sense of vector spaces) of the range of this operator.
Determined at runtime.
If this operator acts like the batch matrix `A` with
`A.shape = [B1,...,Bb, M, N]`, then this returns `M`.
Args:
name: A name for this `Op`.
Returns:
`int32` `Tensor`"
9763,assert_non_singular,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,510,method,"Returns an `Op` that asserts this operator is non singular.
This operator is considered non-singular if
```
ConditionNumber < max{100, range_dimension, domain_dimension} * eps,
eps := np.finfo(self.dtype.as_numpy_dtype).eps
```
Args:
name: A string name to prepend to created ops.
Returns:
An `Assert` `Op`, that, when run, will raise an `InvalidArgumentError` if
the operator is singular."
9764,assert_positive_definite,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,546,method,"Returns an `Op` that asserts this operator is positive definite.
Here, positive definite means that the quadratic form `x^H A x` has positive
real part for all nonzero `x`. Note that we do not require the operator to
be self-adjoint to be positive definite.
Args:
name: A name to give this `Op`.
Returns:
An `Assert` `Op`, that, when run, will raise an `InvalidArgumentError` if
the operator is not positive definite."
9765,assert_self_adjoint,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,573,method,"Returns an `Op` that asserts this operator is self-adjoint.
Here we check that this operator is *exactly* equal to its hermitian
transpose.
Args:
name: A string name to prepend to created ops.
Returns:
An `Assert` `Op`, that, when run, will raise an `InvalidArgumentError` if
the operator is not self-adjoint."
9766,matmul,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,600,method,"Transform [batch] matrix `x` with left multiplication: `x --> Ax`.
```python
# Make an operator acting like batch matrix A. Assume A.shape = [..., M, N]
operator = LinearOperator(...)
operator.shape = [..., M, N]
X = ... # shape [..., N, R], batch matrix, R > 0.
Y = operator.matmul(X)
Y.shape
==> [..., M, R]
Y[..., :, r] = sum_j A[..., :, j] X[j, r]
```
Args:
x: `LinearOperator` or `Tensor` with compatible shape and same `dtype` as
`self`. See class docstring for definition of compatibility.
adjoint: Python `bool`. If `True`, left multiply by the adjoint: `A^H x`.
adjoint_arg: Python `bool`. If `True`, compute `A x^H` where `x^H` is
the hermitian transpose (transposition and complex conjugation).
name: A name for this `Op`.
Returns:
A `LinearOperator` or `Tensor` with shape `[..., M, R]` and same `dtype`
as `self`."
9767,matvec,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,663,method,"Transform [batch] vector `x` with left multiplication: `x --> Ax`.
```python
# Make an operator acting like batch matrix A. Assume A.shape = [..., M, N]
operator = LinearOperator(...)
X = ... # shape [..., N], batch vector
Y = operator.matvec(X)
Y.shape
==> [..., M]
Y[..., :] = sum_j A[..., :, j] X[..., j]
```
Args:
x: `Tensor` with compatible shape and same `dtype` as `self`.
`x` is treated as a [batch] vector meaning for every set of leading
dimensions, the last dimension defines a vector.
See class docstring for definition of compatibility.
adjoint: Python `bool`. If `True`, left multiply by the adjoint: `A^H x`.
name: A name for this `Op`.
Returns:
A `Tensor` with shape `[..., M]` and same `dtype` as `self`."
9768,determinant,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,706,method,"Determinant for every batch member.
Args:
name: A name for this `Op`.
Returns:
`Tensor` with shape `self.batch_shape` and same `dtype` as `self`.
Raises:
NotImplementedError: If `self.is_square` is `False`."
9769,log_abs_determinant,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,735,method,"Log absolute value of determinant for every batch member.
Args:
name: A name for this `Op`.
Returns:
`Tensor` with shape `self.batch_shape` and same `dtype` as `self`.
Raises:
NotImplementedError: If `self.is_square` is `False`."
9770,solve,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,773,method,"Solve (exact or approx) `R` (batch) systems of equations: `A X = rhs`.
The returned `Tensor` will be close to an exact solution if `A` is well
conditioned. Otherwise closeness will vary. See class docstring for details.
Examples:
```python
# Make an operator acting like batch matrix A. Assume A.shape = [..., M, N]
operator = LinearOperator(...)
operator.shape = [..., M, N]
# Solve R > 0 linear systems for every member of the batch.
RHS = ... # shape [..., M, R]
X = operator.solve(RHS)
# X[..., :, r] is the solution to the r'th linear system
# sum_j A[..., :, j] X[..., j, r] = RHS[..., :, r]
operator.matmul(X)
==> RHS
```
Args:
rhs: `Tensor` with same `dtype` as this operator and compatible shape.
`rhs` is treated like a [batch] matrix meaning for every set of leading
dimensions, the last two dimensions defines a matrix.
See class docstring for definition of compatibility.
adjoint: Python `bool`. If `True`, solve the system involving the adjoint
of this `LinearOperator`: `A^H X = rhs`.
adjoint_arg: Python `bool`. If `True`, solve `A X = rhs^H` where `rhs^H`
is the hermitian transpose (transposition and complex conjugation).
name: A name scope to use for ops added by this method.
Returns:
`Tensor` with shape `[...,N, R]` and same `dtype` as `rhs`.
Raises:
NotImplementedError: If `self.is_non_singular` or `is_square` is False."
9771,solvevec,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,854,method,"Solve single equation with best effort: `A X = rhs`.
The returned `Tensor` will be close to an exact solution if `A` is well
conditioned. Otherwise closeness will vary. See class docstring for details.
Examples:
```python
# Make an operator acting like batch matrix A. Assume A.shape = [..., M, N]
operator = LinearOperator(...)
operator.shape = [..., M, N]
# Solve one linear system for every member of the batch.
RHS = ... # shape [..., M]
X = operator.solvevec(RHS)
# X is the solution to the linear system
# sum_j A[..., :, j] X[..., j] = RHS[..., :]
operator.matvec(X)
==> RHS
```
Args:
rhs: `Tensor` with same `dtype` as this operator.
`rhs` is treated like a [batch] vector meaning for every set of leading
dimensions, the last dimension defines a vector. See class docstring
for definition of compatibility regarding batch dimensions.
adjoint: Python `bool`. If `True`, solve the system involving the adjoint
of this `LinearOperator`: `A^H X = rhs`.
name: A name scope to use for ops added by this method.
Returns:
`Tensor` with shape `[...,N]` and same `dtype` as `rhs`.
Raises:
NotImplementedError: If `self.is_non_singular` or `is_square` is False."
9772,adjoint,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,902,method,"Returns the adjoint of the current `LinearOperator`.
Given `A` representing this `LinearOperator`, return `A*`.
Note that calling `self.adjoint()` and `self.H` are equivalent.
Args:
name: A name for this `Op`.
Returns:
`LinearOperator` which represents the adjoint of this `LinearOperator`."
9773,inverse,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,922,method,"Returns the Inverse of this `LinearOperator`.
Given `A` representing this `LinearOperator`, return a `LinearOperator`
representing `A^-1`.
Args:
name: A name scope to use for ops added by this method.
Returns:
`LinearOperator` representing inverse of this matrix.
Raises:
ValueError: When the `LinearOperator` is not hinted to be `non_singular`."
9774,cholesky,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,947,method,"Returns a Cholesky factor as a `LinearOperator`.
Given `A` representing this `LinearOperator`, if `A` is positive definite
self-adjoint, return `L`, where `A = L L^T`, i.e. the cholesky
decomposition.
Args:
name: A name for this `Op`.
Returns:
`LinearOperator` which represents the lower triangular matrix
in the Cholesky decomposition.
Raises:
ValueError: When the `LinearOperator` is not hinted to be positive
definite and self adjoint."
9775,to_dense,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,988,method,Return a dense (batch) matrix representing this operator.
9776,diag_part,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,997,method,"Efficiently get the [batch] diagonal part of this operator.
If this operator has shape `[B1,...,Bb, M, N]`, this returns a
`Tensor` `diagonal`, of shape `[B1,...,Bb, min(M, N)]`, where
`diagonal[b1,...,bb, i] = self.to_dense()[b1,...,bb, i, i]`.
```
my_operator = LinearOperatorDiag([1., 2.])
# Efficiently get the diagonal
my_operator.diag_part()
==> [1., 2.]
# Equivalent, but inefficient method
tf.linalg.diag_part(my_operator.to_dense())
==> [1., 2.]
```
Args:
name: A name for this `Op`.
Returns:
diag_part: A `Tensor` of same `dtype` as self."
9777,trace,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,1028,method,"Trace of the linear operator, equal to sum of `self.diag_part()`.
If the operator is square, this is also the sum of the eigenvalues.
Args:
name: A name for this `Op`.
Returns:
Shape `[B1,...,Bb]` `Tensor` of same `dtype` as `self`."
9778,add_to_tensor,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,1046,method,"Add matrix represented by this operator to `x`. Equivalent to `A + x`.
Args:
x: `Tensor` with same `dtype` and shape broadcastable to `self.shape`.
name: A name to give this `Op`.
Returns:
A `Tensor` with broadcast shape and same `dtype` as `self`."
9779,eigvals,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,1064,method,"Returns the eigenvalues of this linear operator.
If the operator is marked as self-adjoint (via `is_self_adjoint`)
this computation can be more efficient.
Note: This currently only supports self-adjoint operators.
Args:
name: A name for this `Op`.
Returns:
Shape `[B1,...,Bb, N]` `Tensor` of same `dtype` as `self`."
9780,cond,tensorflow/tensorflow/python/ops/linalg/linear_operator.py,1096,method,"Returns the condition number of this linear operator.
Args:
name: A name for this `Op`.
Returns:
Shape `[B1,...,Bb]` `Tensor` of same `dtype` as `self`."
9781,add_operators,tensorflow/tensorflow/python/ops/linalg/linear_operator_addition.py,38,function,"Efficiently add one or more linear operators.
Given operators `[A1, A2,...]`, this `Op` returns a possibly shorter list of
operators `[B1, B2,...]` such that
```sum_k Ak.matmul(x) = sum_k Bk.matmul(x).```
The operators `Bk` result by adding some of the `Ak`, as allowed by
`addition_tiers`.
Example of efficient adding of diagonal operators.
```python
A1 = LinearOperatorDiag(diag=[1., 1.], name=""A1"")
A2 = LinearOperatorDiag(diag=[2., 2.], name=""A2"")
# Use two tiers, the first contains an Adder that returns Diag. Since both
# A1 and A2 are Diag, they can use this Adder. The second tier will not be
# used.
addition_tiers = [
[_AddAndReturnDiag()],
[_AddAndReturnMatrix()]]
B_list = add_operators([A1, A2], addition_tiers=addition_tiers)
len(B_list)
==> 1
B_list[0].__class__.__name__
==> 'LinearOperatorDiag'
B_list[0].to_dense()
==> [[3., 0.],
[0., 3.]]
B_list[0].name
==> 'Add/A1__A2/'
```
Args:
operators: Iterable of `LinearOperator` objects with same `dtype`, domain
and range dimensions, and broadcastable batch shapes.
operator_name: String name for returned `LinearOperator`. Defaults to
concatenation of ""Add/A__B/"" that indicates the order of addition steps.
addition_tiers: List tiers, like `[tier_0, tier_1, ...]`, where `tier_i`
is a list of `Adder` objects. This function attempts to do all additions
in tier `i` before trying tier `i + 1`.
name: A name for this `Op`. Defaults to `add_operators`.
Returns:
Subclass of `LinearOperator`. Class and order of addition may change as new
(and better) addition strategies emerge.
Raises:
ValueError: If `operators` argument is empty.
ValueError: If shapes are incompatible."
9782,LinearOperatorAdjoint,tensorflow/tensorflow/python/ops/linalg/linear_operator_adjoint.py,33,class,"`LinearOperator` representing the adjoint of another operator.
This operator represents the adjoint of another operator.
```python
# Create a 2 x 2 linear operator.
operator = LinearOperatorFullMatrix([[1 - i., 3.], [0., 1. + i]])
operator_adjoint = LinearOperatorAdjoint(operator)
operator_adjoint.to_dense()
==> [[1. + i, 0.]
[3., 1 - i]]
operator_adjoint.shape
==> [2, 2]
operator_adjoint.log_abs_determinant()
==> - log(2)
x = ... Shape [2, 4] Tensor
operator_adjoint.matmul(x)
==> Shape [2, 4] Tensor, equal to operator.matmul(x, adjoint=True)
```
#### Performance
The performance of `LinearOperatorAdjoint` depends on the underlying
operators performance.
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning:
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9783,operator,tensorflow/tensorflow/python/ops/linalg/linear_operator_adjoint.py,158,method,The operator before taking the adjoint.
9784,adjoint,tensorflow/tensorflow/python/ops/linalg/linear_operator_algebra.py,76,function,"Get the adjoint associated to lin_op_a.
Args:
lin_op_a: The LinearOperator to take the adjoint of.
name: Name to use for this operation.
Returns:
A LinearOperator that represents the adjoint of `lin_op_a`.
Raises:
NotImplementedError: If no Adjoint method is defined for the LinearOperator
type of `lin_op_a`."
9785,cholesky,tensorflow/tensorflow/python/ops/linalg/linear_operator_algebra.py,99,function,"Get the Cholesky factor associated to lin_op_a.
Args:
lin_op_a: The LinearOperator to decompose.
name: Name to use for this operation.
Returns:
A LinearOperator that represents the lower Cholesky factor of `lin_op_a`.
Raises:
NotImplementedError: If no Cholesky method is defined for the LinearOperator
type of `lin_op_a`."
9786,matmul,tensorflow/tensorflow/python/ops/linalg/linear_operator_algebra.py,122,function,"Compute lin_op_a.matmul(lin_op_b).
Args:
lin_op_a: The LinearOperator on the left.
lin_op_b: The LinearOperator on the right.
name: Name to use for this operation.
Returns:
A LinearOperator that represents the matmul between `lin_op_a` and
`lin_op_b`.
Raises:
NotImplementedError: If no matmul method is defined between types of
`lin_op_a` and `lin_op_b`."
9787,solve,tensorflow/tensorflow/python/ops/linalg/linear_operator_algebra.py,147,function,"Compute lin_op_a.solve(lin_op_b).
Args:
lin_op_a: The LinearOperator on the left.
lin_op_b: The LinearOperator on the right.
name: Name to use for this operation.
Returns:
A LinearOperator that represents the solve between `lin_op_a` and
`lin_op_b`.
Raises:
NotImplementedError: If no solve method is defined between types of
`lin_op_a` and `lin_op_b`."
9788,inverse,tensorflow/tensorflow/python/ops/linalg/linear_operator_algebra.py,172,function,"Get the Inverse associated to lin_op_a.
Args:
lin_op_a: The LinearOperator to decompose.
name: Name to use for this operation.
Returns:
A LinearOperator that represents the inverse of `lin_op_a`.
Raises:
NotImplementedError: If no Inverse method is defined for the LinearOperator
type of `lin_op_a`."
9789,RegisterAdjoint,tensorflow/tensorflow/python/ops/linalg/linear_operator_algebra.py,195,class,"Decorator to register an Adjoint implementation function.
Usage:
@linear_operator_algebra.RegisterAdjoint(lin_op.LinearOperatorIdentity)
def _adjoint_identity(lin_op_a):
# Return the identity matrix."
9790,RegisterCholesky,tensorflow/tensorflow/python/ops/linalg/linear_operator_algebra.py,237,class,"Decorator to register a Cholesky implementation function.
Usage:
@linear_operator_algebra.RegisterCholesky(lin_op.LinearOperatorIdentity)
def _cholesky_identity(lin_op_a):
# Return the identity matrix."
9791,RegisterMatmul,tensorflow/tensorflow/python/ops/linalg/linear_operator_algebra.py,279,class,"Decorator to register a Matmul implementation function.
Usage:
@linear_operator_algebra.RegisterMatmul(
lin_op.LinearOperatorIdentity,
lin_op.LinearOperatorIdentity)
def _matmul_identity(a, b):
# Return the identity matrix."
9792,RegisterSolve,tensorflow/tensorflow/python/ops/linalg/linear_operator_algebra.py,325,class,"Decorator to register a Solve implementation function.
Usage:
@linear_operator_algebra.RegisterSolve(
lin_op.LinearOperatorIdentity,
lin_op.LinearOperatorIdentity)
def _solve_identity(a, b):
# Return the identity matrix."
9793,RegisterInverse,tensorflow/tensorflow/python/ops/linalg/linear_operator_algebra.py,371,class,"Decorator to register an Inverse implementation function.
Usage:
@linear_operator_algebra.RegisterInverse(lin_op.LinearOperatorIdentity)
def _inverse_identity(lin_op_a):
# Return the identity matrix."
9794,LinearOperatorBlockDiag,tensorflow/tensorflow/python/ops/linalg/linear_operator_block_diag.py,37,class,"Combines one or more `LinearOperators` in to a Block Diagonal matrix.
This operator combines one or more linear operators `[op1,...,opJ]`,
building a new `LinearOperator`, whose underlying matrix representation is
square and has each operator `opi` on the main diagonal, and zero's elsewhere.
#### Shape compatibility
If `opj` acts like a [batch] square matrix `Aj`, then `op_combined` acts like
the [batch] square matrix formed by having each matrix `Aj` on the main
diagonal.
Each `opj` is required to represent a square matrix, and hence will have
shape `batch_shape_j + [M_j, M_j]`.
If `opj` has shape `batch_shape_j + [M_j, M_j]`, then the combined operator
has shape `broadcast_batch_shape + [sum M_j, sum M_j]`, where
`broadcast_batch_shape` is the mutual broadcast of `batch_shape_j`,
`j = 1,...,J`, assuming the intermediate batch shapes broadcast.
Even if the combined shape is well defined, the combined operator's
methods may fail due to lack of broadcasting ability in the defining
operators' methods.
Arguments to `matmul`, `matvec`, `solve`, and `solvevec` may either be single
`Tensor`s or lists of `Tensor`s that are interpreted as blocks. The `j`th
element of a blockwise list of `Tensor`s must have dimensions that match
`opj` for the given method. If a list of blocks is input, then a list of
blocks is returned as well.
```python
# Create a 4 x 4 linear operator combined of two 2 x 2 operators.
operator_1 = LinearOperatorFullMatrix([[1., 2.], [3., 4.]])
operator_2 = LinearOperatorFullMatrix([[1., 0.], [0., 1.]])
operator = LinearOperatorBlockDiag([operator_1, operator_2])
operator.to_dense()
==> [[1., 2., 0., 0.],
[3., 4., 0., 0.],
[0., 0., 1., 0.],
[0., 0., 0., 1.]]
operator.shape
==> [4, 4]
operator.log_abs_determinant()
==> scalar Tensor
x1 = ... # Shape [2, 2] Tensor
x2 = ... # Shape [2, 2] Tensor
x = tf.concat([x1, x2], 0) # Shape [2, 4] Tensor
operator.matmul(x)
==> tf.concat([operator_1.matmul(x1), operator_2.matmul(x2)])
# Create a [2, 3] batch of 4 x 4 linear operators.
matrix_44 = tf.random.normal(shape=[2, 3, 4, 4])
operator_44 = LinearOperatorFullMatrix(matrix)
# Create a [1, 3] batch of 5 x 5 linear operators.
matrix_55 = tf.random.normal(shape=[1, 3, 5, 5])
operator_55 = LinearOperatorFullMatrix(matrix_55)
# Combine to create a [2, 3] batch of 9 x 9 operators.
operator_99 = LinearOperatorBlockDiag([operator_44, operator_55])
# Create a shape [2, 3, 9] vector.
x = tf.random.normal(shape=[2, 3, 9])
operator_99.matmul(x)
==> Shape [2, 3, 9] Tensor
# Create a blockwise list of vectors.
x = [tf.random.normal(shape=[2, 3, 4]), tf.random.normal(shape=[2, 3, 5])]
operator_99.matmul(x)
==> [Shape [2, 3, 4] Tensor, Shape [2, 3, 5] Tensor]
```
#### Performance
The performance of `LinearOperatorBlockDiag` on any operation is equal to
the sum of the individual operators' operations.
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning:
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9795,operators,tensorflow/tensorflow/python/ops/linalg/linear_operator_block_diag.py,233,method,
9796,matmul,tensorflow/tensorflow/python/ops/linalg/linear_operator_block_diag.py,281,method,"Transform [batch] matrix `x` with left multiplication: `x --> Ax`.
```python
# Make an operator acting like batch matrix A. Assume A.shape = [..., M, N]
operator = LinearOperator(...)
operator.shape = [..., M, N]
X = ... # shape [..., N, R], batch matrix, R > 0.
Y = operator.matmul(X)
Y.shape
==> [..., M, R]
Y[..., :, r] = sum_j A[..., :, j] X[j, r]
```
Args:
x: `LinearOperator`, `Tensor` with compatible shape and same `dtype` as
`self`, or a blockwise iterable of `LinearOperator`s or `Tensor`s. See
class docstring for definition of shape compatibility.
adjoint: Python `bool`. If `True`, left multiply by the adjoint: `A^H x`.
adjoint_arg: Python `bool`. If `True`, compute `A x^H` where `x^H` is
the hermitian transpose (transposition and complex conjugation).
name: A name for this `Op`.
Returns:
A `LinearOperator` or `Tensor` with shape `[..., M, R]` and same `dtype`
as `self`, or if `x` is blockwise, a list of `Tensor`s with shapes that
concatenate to `[..., M, R]`."
9797,matvec,tensorflow/tensorflow/python/ops/linalg/linear_operator_block_diag.py,373,method,"Transform [batch] vector `x` with left multiplication: `x --> Ax`.
```python
# Make an operator acting like batch matric A. Assume A.shape = [..., M, N]
operator = LinearOperator(...)
X = ... # shape [..., N], batch vector
Y = operator.matvec(X)
Y.shape
==> [..., M]
Y[..., :] = sum_j A[..., :, j] X[..., j]
```
Args:
x: `Tensor` with compatible shape and same `dtype` as `self`, or an
iterable of `Tensor`s (for blockwise operators). `Tensor`s are treated
a [batch] vectors, meaning for every set of leading dimensions, the last
dimension defines a vector.
See class docstring for definition of compatibility.
adjoint: Python `bool`. If `True`, left multiply by the adjoint: `A^H x`.
name: A name for this `Op`.
Returns:
A `Tensor` with shape `[..., M]` and same `dtype` as `self`."
9798,solve,tensorflow/tensorflow/python/ops/linalg/linear_operator_block_diag.py,436,method,"Solve (exact or approx) `R` (batch) systems of equations: `A X = rhs`.
The returned `Tensor` will be close to an exact solution if `A` is well
conditioned. Otherwise closeness will vary. See class docstring for details.
Examples:
```python
# Make an operator acting like batch matrix A. Assume A.shape = [..., M, N]
operator = LinearOperator(...)
operator.shape = [..., M, N]
# Solve R > 0 linear systems for every member of the batch.
RHS = ... # shape [..., M, R]
X = operator.solve(RHS)
# X[..., :, r] is the solution to the r'th linear system
# sum_j A[..., :, j] X[..., j, r] = RHS[..., :, r]
operator.matmul(X)
==> RHS
```
Args:
rhs: `Tensor` with same `dtype` as this operator and compatible shape,
or a list of `Tensor`s (for blockwise operators). `Tensor`s are treated
like a [batch] matrices meaning for every set of leading dimensions, the
last two dimensions defines a matrix.
See class docstring for definition of compatibility.
adjoint: Python `bool`. If `True`, solve the system involving the adjoint
of this `LinearOperator`: `A^H X = rhs`.
adjoint_arg: Python `bool`. If `True`, solve `A X = rhs^H` where `rhs^H`
is the hermitian transpose (transposition and complex conjugation).
name: A name scope to use for ops added by this method.
Returns:
`Tensor` with shape `[...,N, R]` and same `dtype` as `rhs`.
Raises:
NotImplementedError: If `self.is_non_singular` or `is_square` is False."
9799,solvevec,tensorflow/tensorflow/python/ops/linalg/linear_operator_block_diag.py,540,method,"Solve single equation with best effort: `A X = rhs`.
The returned `Tensor` will be close to an exact solution if `A` is well
conditioned. Otherwise closeness will vary. See class docstring for details.
Examples:
```python
# Make an operator acting like batch matrix A. Assume A.shape = [..., M, N]
operator = LinearOperator(...)
operator.shape = [..., M, N]
# Solve one linear system for every member of the batch.
RHS = ... # shape [..., M]
X = operator.solvevec(RHS)
# X is the solution to the linear system
# sum_j A[..., :, j] X[..., j] = RHS[..., :]
operator.matvec(X)
==> RHS
```
Args:
rhs: `Tensor` with same `dtype` as this operator, or list of `Tensor`s
(for blockwise operators). `Tensor`s are treated as [batch] vectors,
meaning for every set of leading dimensions, the last dimension defines
a vector. See class docstring for definition of compatibility regarding
batch dimensions.
adjoint: Python `bool`. If `True`, solve the system involving the adjoint
of this `LinearOperator`: `A^H X = rhs`.
name: A name scope to use for ops added by this method.
Returns:
`Tensor` with shape `[...,N]` and same `dtype` as `rhs`.
Raises:
NotImplementedError: If `self.is_non_singular` or `is_square` is False."
9800,LinearOperatorBlockLowerTriangular,tensorflow/tensorflow/python/ops/linalg/linear_operator_block_lower_triangular.py,39,class,"Combines `LinearOperators` into a blockwise lower-triangular matrix.
This operator is initialized with a nested list of linear operators, which
are combined into a new `LinearOperator` whose underlying matrix
representation is square and has each operator on or below the main diagonal,
and zero's elsewhere. Each element of the outer list is a list of
`LinearOperators` corresponding to a row-partition of the blockwise structure.
The number of `LinearOperator`s in row-partion `i` must be equal to `i`.
For example, a blockwise `3 x 3` `LinearOperatorBlockLowerTriangular` is
initialized with the list `[[op_00], [op_10, op_11], [op_20, op_21, op_22]]`,
where the `op_ij`, `i < 3, j <= i`, are `LinearOperator` instances. The
`LinearOperatorBlockLowerTriangular` behaves as the following blockwise
matrix, where `0` represents appropriately-sized [batch] matrices of zeros:
```none
[[op_00, 0, 0],
[op_10, op_11, 0],
[op_20, op_21, op_22]]
```
Each `op_jj` on the diagonal is required to represent a square matrix, and
hence will have shape `batch_shape_j + [M_j, M_j]`. `LinearOperator`s in row
`j` of the blockwise structure must have `range_dimension` equal to that of
`op_jj`, and `LinearOperators` in column `j` must have `domain_dimension`
equal to that of `op_jj`.
If each `op_jj` on the diagonal has shape `batch_shape_j + [M_j, M_j]`, then
the combined operator has shape `broadcast_batch_shape + [sum M_j, sum M_j]`,
where `broadcast_batch_shape` is the mutual broadcast of `batch_shape_j`,
`j = 0, 1, ..., J`, assuming the intermediate batch shapes broadcast.
Even if the combined shape is well defined, the combined operator's
methods may fail due to lack of broadcasting ability in the defining
operators' methods.
For example, to create a 4 x 4 linear operator combined of three 2 x 2
operators:
>>> operator_0 = tf.linalg.LinearOperatorFullMatrix([[1., 2.], [3., 4.]])
>>> operator_1 = tf.linalg.LinearOperatorFullMatrix([[1., 0.], [0., 1.]])
>>> operator_2 = tf.linalg.LinearOperatorLowerTriangular([[5., 6.], [7., 8]])
>>> operator = LinearOperatorBlockLowerTriangular(
... [[operator_0], [operator_1, operator_2]])
>>> operator.to_dense()
<tf.Tensor: shape=(4, 4), dtype=float32, numpy=
array([[1., 2., 0., 0.],
[3., 4., 0., 0.],
[1., 0., 5., 0.],
[0., 1., 7., 8.]], dtype=float32)>
>>> operator.shape
TensorShape([4, 4])
>>> operator.log_abs_determinant()
<tf.Tensor: shape=(), dtype=float32, numpy=4.3820267>
>>> x0 = [[1., 6.], [-3., 4.]]
>>> x1 = [[0., 2.], [4., 0.]]
>>> x = tf.concat([x0, x1], 0) # Shape [2, 4] Tensor
>>> operator.matmul(x)
<tf.Tensor: shape=(4, 2), dtype=float32, numpy=
array([[-5., 14.],
[-9., 34.],
[ 1., 16.],
[29., 18.]], dtype=float32)>
The above `matmul` is equivalent to:
>>> tf.concat([operator_0.matmul(x0),
... operator_1.matmul(x0) + operator_2.matmul(x1)], axis=0)
<tf.Tensor: shape=(4, 2), dtype=float32, numpy=
array([[-5., 14.],
[-9., 34.],
[ 1., 16.],
[29., 18.]], dtype=float32)>
#### Shape compatibility
This operator acts on [batch] matrix with compatible shape.
`x` is a batch matrix with compatible shape for `matmul` and `solve` if
```
operator.shape = [B1,...,Bb] + [M, N], with b >= 0
x.shape = [B1,...,Bb] + [N, R], with R >= 0.
```
For example:
Create a [2, 3] batch of 4 x 4 linear operators:
>>> matrix_44 = tf.random.normal(shape=[2, 3, 4, 4])
>>> operator_44 = tf.linalg.LinearOperatorFullMatrix(matrix_44)
Create a [1, 3] batch of 5 x 4 linear operators:
>>> matrix_54 = tf.random.normal(shape=[1, 3, 5, 4])
>>> operator_54 = tf.linalg.LinearOperatorFullMatrix(matrix_54)
Create a [1, 3] batch of 5 x 5 linear operators:
>>> matrix_55 = tf.random.normal(shape=[1, 3, 5, 5])
>>> operator_55 = tf.linalg.LinearOperatorFullMatrix(matrix_55)
Combine to create a [2, 3] batch of 9 x 9 operators:
>>> operator_99 = LinearOperatorBlockLowerTriangular(
... [[operator_44], [operator_54, operator_55]])
>>> operator_99.shape
TensorShape([2, 3, 9, 9])
Create a shape [2, 1, 9] batch of vectors and apply the operator to it.
>>> x = tf.random.normal(shape=[2, 1, 9])
>>> y = operator_99.matvec(x)
>>> y.shape
TensorShape([2, 3, 9])
Create a blockwise list of vectors and apply the operator to it. A blockwise
list is returned.
>>> x4 = tf.random.normal(shape=[2, 1, 4])
>>> x5 = tf.random.normal(shape=[2, 3, 5])
>>> y_blockwise = operator_99.matvec([x4, x5])
>>> y_blockwise[0].shape
TensorShape([2, 3, 4])
>>> y_blockwise[1].shape
TensorShape([2, 3, 5])
#### Performance
Suppose `operator` is a `LinearOperatorBlockLowerTriangular` consisting of `D`
row-partitions and `D` column-partitions, such that the total number of
operators is `N = D * (D + 1) // 2`.
* `operator.matmul` has complexity equal to the sum of the `matmul`
complexities of the individual operators.
* `operator.solve` has complexity equal to the sum of the `solve` complexities
of the operators on the diagonal and the `matmul` complexities of the
operators off the diagonal.
* `operator.determinant` has complexity equal to the sum of the `determinant`
complexities of the operators on the diagonal.
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning:
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9801,operators,tensorflow/tensorflow/python/ops/linalg/linear_operator_block_lower_triangular.py,335,method,
9802,matmul,tensorflow/tensorflow/python/ops/linalg/linear_operator_block_lower_triangular.py,384,method,"Transform [batch] matrix `x` with left multiplication: `x --> Ax`.
```python
# Make an operator acting like batch matrix A. Assume A.shape = [..., M, N]
operator = LinearOperator(...)
operator.shape = [..., M, N]
X = ... # shape [..., N, R], batch matrix, R > 0.
Y = operator.matmul(X)
Y.shape
==> [..., M, R]
Y[..., :, r] = sum_j A[..., :, j] X[j, r]
```
Args:
x: `LinearOperator`, `Tensor` with compatible shape and same `dtype` as
`self`, or a blockwise iterable of `LinearOperator`s or `Tensor`s. See
class docstring for definition of shape compatibility.
adjoint: Python `bool`. If `True`, left multiply by the adjoint: `A^H x`.
adjoint_arg: Python `bool`. If `True`, compute `A x^H` where `x^H` is
the hermitian transpose (transposition and complex conjugation).
name: A name for this `Op`.
Returns:
A `LinearOperator` or `Tensor` with shape `[..., M, R]` and same `dtype`
as `self`, or if `x` is blockwise, a list of `Tensor`s with shapes that
concatenate to `[..., M, R]`."
9803,matvec,tensorflow/tensorflow/python/ops/linalg/linear_operator_block_lower_triangular.py,512,method,"Transform [batch] vector `x` with left multiplication: `x --> Ax`.
```python
# Make an operator acting like batch matrix A. Assume A.shape = [..., M, N]
operator = LinearOperator(...)
X = ... # shape [..., N], batch vector
Y = operator.matvec(X)
Y.shape
==> [..., M]
Y[..., :] = sum_j A[..., :, j] X[..., j]
```
Args:
x: `Tensor` with compatible shape and same `dtype` as `self`, or an
iterable of `Tensor`s. `Tensor`s are treated a [batch] vectors, meaning
for every set of leading dimensions, the last dimension defines a
vector.
See class docstring for definition of compatibility.
adjoint: Python `bool`. If `True`, left multiply by the adjoint: `A^H x`.
name: A name for this `Op`.
Returns:
A `Tensor` with shape `[..., M]` and same `dtype` as `self`."
9804,solve,tensorflow/tensorflow/python/ops/linalg/linear_operator_block_lower_triangular.py,577,method,"Solve (exact or approx) `R` (batch) systems of equations: `A X = rhs`.
The returned `Tensor` will be close to an exact solution if `A` is well
conditioned. Otherwise closeness will vary. See class docstring for details.
Given the blockwise `n + 1`-by-`n + 1` linear operator:
op = [[A_00 0 ... 0 ... 0],
[A_10 A_11 ... 0 ... 0],
...
[A_k0 A_k1 ... A_kk ... 0],
...
[A_n0 A_n1 ... A_nk ... A_nn]]
we find `x = op.solve(y)` by observing that
`y_k = A_k0.matmul(x_0) + A_k1.matmul(x_1) + ... + A_kk.matmul(x_k)`
and therefore
`x_k = A_kk.solve(y_k -
A_k0.matmul(x_0) - ... - A_k(k-1).matmul(x_(k-1)))`
where `x_k` and `y_k` are the `k`th blocks obtained by decomposing `x`
and `y` along their appropriate axes.
We first solve `x_0 = A_00.solve(y_0)`. Proceeding inductively, we solve
for `x_k`, `k = 1..n`, given `x_0..x_(k-1)`.
The adjoint case is solved similarly, beginning with
`x_n = A_nn.solve(y_n, adjoint=True)` and proceeding backwards.
Examples:
```python
# Make an operator acting like batch matrix A. Assume A.shape = [..., M, N]
operator = LinearOperator(...)
operator.shape = [..., M, N]
# Solve R > 0 linear systems for every member of the batch.
RHS = ... # shape [..., M, R]
X = operator.solve(RHS)
# X[..., :, r] is the solution to the r'th linear system
# sum_j A[..., :, j] X[..., j, r] = RHS[..., :, r]
operator.matmul(X)
==> RHS
```
Args:
rhs: `Tensor` with same `dtype` as this operator and compatible shape,
or a list of `Tensor`s. `Tensor`s are treated like a [batch] matrices
meaning for every set of leading dimensions, the last two dimensions
defines a matrix.
See class docstring for definition of compatibility.
adjoint: Python `bool`. If `True`, solve the system involving the adjoint
of this `LinearOperator`: `A^H X = rhs`.
adjoint_arg: Python `bool`. If `True`, solve `A X = rhs^H` where `rhs^H`
is the hermitian transpose (transposition and complex conjugation).
name: A name scope to use for ops added by this method.
Returns:
`Tensor` with shape `[...,N, R]` and same `dtype` as `rhs`.
Raises:
NotImplementedError: If `self.is_non_singular` or `is_square` is False."
9805,solvevec,tensorflow/tensorflow/python/ops/linalg/linear_operator_block_lower_triangular.py,752,method,"Solve single equation with best effort: `A X = rhs`.
The returned `Tensor` will be close to an exact solution if `A` is well
conditioned. Otherwise closeness will vary. See class docstring for details.
Examples:
```python
# Make an operator acting like batch matrix A. Assume A.shape = [..., M, N]
operator = LinearOperator(...)
operator.shape = [..., M, N]
# Solve one linear system for every member of the batch.
RHS = ... # shape [..., M]
X = operator.solvevec(RHS)
# X is the solution to the linear system
# sum_j A[..., :, j] X[..., j] = RHS[..., :]
operator.matvec(X)
==> RHS
```
Args:
rhs: `Tensor` with same `dtype` as this operator, or list of `Tensor`s
(for blockwise operators). `Tensor`s are treated as [batch] vectors,
meaning for every set of leading dimensions, the last dimension defines
a vector. See class docstring for definition of compatibility regarding
batch dimensions.
adjoint: Python `bool`. If `True`, solve the system involving the adjoint
of this `LinearOperator`: `A^H X = rhs`.
name: A name scope to use for ops added by this method.
Returns:
`Tensor` with shape `[...,N]` and same `dtype` as `rhs`.
Raises:
NotImplementedError: If `self.is_non_singular` or `is_square` is False."
9806,LinearOperatorCirculant,tensorflow/tensorflow/python/ops/linalg/linear_operator_circulant.py,516,class,"`LinearOperator` acting like a circulant matrix.
This operator acts like a circulant matrix `A` with
shape `[B1,...,Bb, N, N]` for some `b >= 0`. The first `b` indices index a
batch member. For every batch index `(i1,...,ib)`, `A[i1,...,ib, : :]` is
an `N x N` matrix. This matrix `A` is not materialized, but for
purposes of broadcasting this shape will be relevant.
#### Description in terms of circulant matrices
Circulant means the entries of `A` are generated by a single vector, the
convolution kernel `h`: `A_{mn} := h_{m-n mod N}`. With `h = [w, x, y, z]`,
```
A = |w z y x|
|x w z y|
|y x w z|
|z y x w|
```
This means that the result of matrix multiplication `v = Au` has `Lth` column
given circular convolution between `h` with the `Lth` column of `u`.
#### Description in terms of the frequency spectrum
There is an equivalent description in terms of the [batch] spectrum `H` and
Fourier transforms. Here we consider `A.shape = [N, N]` and ignore batch
dimensions. Define the discrete Fourier transform (DFT) and its inverse by
```
DFT[ h[n] ] = H[k] := sum_{n = 0}^{N - 1} h_n e^{-i 2pi k n / N}
IDFT[ H[k] ] = h[n] = N^{-1} sum_{k = 0}^{N - 1} H_k e^{i 2pi k n / N}
```
From these definitions, we see that
```
H[0] = sum_{n = 0}^{N - 1} h_n
H[1] = ""the first positive frequency""
H[N - 1] = ""the first negative frequency""
```
Loosely speaking, with `*` element-wise multiplication, matrix multiplication
is equal to the action of a Fourier multiplier: `A u = IDFT[ H * DFT[u] ]`.
Precisely speaking, given `[N, R]` matrix `u`, let `DFT[u]` be the `[N, R]`
matrix with `rth` column equal to the DFT of the `rth` column of `u`.
Define the `IDFT` similarly.
Matrix multiplication may be expressed columnwise:
```(A u)_r = IDFT[ H * (DFT[u])_r ]```
#### Operator properties deduced from the spectrum.
Letting `U` be the `kth` Euclidean basis vector, and `U = IDFT[u]`.
The above formulas show that`A U = H_k * U`. We conclude that the elements
of `H` are the eigenvalues of this operator. Therefore
* This operator is positive definite if and only if `Real{H} > 0`.
A general property of Fourier transforms is the correspondence between
Hermitian functions and real valued transforms.
Suppose `H.shape = [B1,...,Bb, N]`. We say that `H` is a Hermitian spectrum
if, with `%` meaning modulus division,
```H[..., n % N] = ComplexConjugate[ H[..., (-n) % N] ]```
* This operator corresponds to a real matrix if and only if `H` is Hermitian.
* This operator is self-adjoint if and only if `H` is real.
See e.g. ""Discrete-Time Signal Processing"", Oppenheim and Schafer.
#### Example of a self-adjoint positive definite operator
```python
# spectrum is real ==> operator is self-adjoint
# spectrum is positive ==> operator is positive definite
spectrum = [6., 4, 2]
operator = LinearOperatorCirculant(spectrum)
# IFFT[spectrum]
operator.convolution_kernel()
==> [4 + 0j, 1 + 0.58j, 1 - 0.58j]
operator.to_dense()
==> [[4 + 0.0j, 1 - 0.6j, 1 + 0.6j],
[1 + 0.6j, 4 + 0.0j, 1 - 0.6j],
[1 - 0.6j, 1 + 0.6j, 4 + 0.0j]]
```
#### Example of defining in terms of a real convolution kernel
```python
# convolution_kernel is real ==> spectrum is Hermitian.
convolution_kernel = [1., 2., 1.]]
spectrum = tf.signal.fft(tf.cast(convolution_kernel, tf.complex64))
# spectrum is Hermitian ==> operator is real.
# spectrum is shape [3] ==> operator is shape [3, 3]
# We force the input/output type to be real, which allows this to operate
# like a real matrix.
operator = LinearOperatorCirculant(spectrum, input_output_dtype=tf.float32)
operator.to_dense()
==> [[ 1, 1, 2],
[ 2, 1, 1],
[ 1, 2, 1]]
```
#### Example of Hermitian spectrum
```python
# spectrum is shape [3] ==> operator is shape [3, 3]
# spectrum is Hermitian ==> operator is real.
spectrum = [1, 1j, -1j]
operator = LinearOperatorCirculant(spectrum)
operator.to_dense()
==> [[ 0.33 + 0j, 0.91 + 0j, -0.24 + 0j],
[-0.24 + 0j, 0.33 + 0j, 0.91 + 0j],
[ 0.91 + 0j, -0.24 + 0j, 0.33 + 0j]
```
#### Example of forcing real `dtype` when spectrum is Hermitian
```python
# spectrum is shape [4] ==> operator is shape [4, 4]
# spectrum is real ==> operator is self-adjoint
# spectrum is Hermitian ==> operator is real
# spectrum has positive real part ==> operator is positive-definite.
spectrum = [6., 4, 2, 4]
# Force the input dtype to be float32.
# Cast the output to float32. This is fine because the operator will be
# real due to Hermitian spectrum.
operator = LinearOperatorCirculant(spectrum, input_output_dtype=tf.float32)
operator.shape
==> [4, 4]
operator.to_dense()
==> [[4, 1, 0, 1],
[1, 4, 1, 0],
[0, 1, 4, 1],
[1, 0, 1, 4]]
# convolution_kernel = tf.signal.ifft(spectrum)
operator.convolution_kernel()
==> [4, 1, 0, 1]
```
#### Performance
Suppose `operator` is a `LinearOperatorCirculant` of shape `[N, N]`,
and `x.shape = [N, R]`. Then
* `operator.matmul(x)` is `O(R*N*Log[N])`
* `operator.solve(x)` is `O(R*N*Log[N])`
* `operator.determinant()` involves a size `N` `reduce_prod`.
If instead `operator` and `x` have shape `[B1,...,Bb, N, N]` and
`[B1,...,Bb, N, R]`, every operation increases in complexity by `B1*...*Bb`.
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning:
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way.
References:
Toeplitz and Circulant Matrices - A Review:
[Gray, 2006](https://www.nowpublishers.com/article/Details/CIT-006)
([pdf](https://ee.stanford.edu/~gray/toeplitz.pdf))"
9807,LinearOperatorCirculant2D,tensorflow/tensorflow/python/ops/linalg/linear_operator_circulant.py,762,class,"`LinearOperator` acting like a block circulant matrix.
This operator acts like a block circulant matrix `A` with
shape `[B1,...,Bb, N, N]` for some `b >= 0`. The first `b` indices index a
batch member. For every batch index `(i1,...,ib)`, `A[i1,...,ib, : :]` is
an `N x N` matrix. This matrix `A` is not materialized, but for
purposes of broadcasting this shape will be relevant.
#### Description in terms of block circulant matrices
If `A` is block circulant, with block sizes `N0, N1` (`N0 * N1 = N`):
`A` has a block circulant structure, composed of `N0 x N0` blocks, with each
block an `N1 x N1` circulant matrix.
For example, with `W`, `X`, `Y`, `Z` each circulant,
```
A = |W Z Y X|
|X W Z Y|
|Y X W Z|
|Z Y X W|
```
Note that `A` itself will not in general be circulant.
#### Description in terms of the frequency spectrum
There is an equivalent description in terms of the [batch] spectrum `H` and
Fourier transforms. Here we consider `A.shape = [N, N]` and ignore batch
dimensions.
If `H.shape = [N0, N1]`, (`N0 * N1 = N`):
Loosely speaking, matrix multiplication is equal to the action of a
Fourier multiplier: `A u = IDFT2[ H DFT2[u] ]`.
Precisely speaking, given `[N, R]` matrix `u`, let `DFT2[u]` be the
`[N0, N1, R]` `Tensor` defined by re-shaping `u` to `[N0, N1, R]` and taking
a two dimensional DFT across the first two dimensions. Let `IDFT2` be the
inverse of `DFT2`. Matrix multiplication may be expressed columnwise:
```(A u)_r = IDFT2[ H * (DFT2[u])_r ]```
#### Operator properties deduced from the spectrum.
* This operator is positive definite if and only if `Real{H} > 0`.
A general property of Fourier transforms is the correspondence between
Hermitian functions and real valued transforms.
Suppose `H.shape = [B1,...,Bb, N0, N1]`, we say that `H` is a Hermitian
spectrum if, with `%` indicating modulus division,
```
H[..., n0 % N0, n1 % N1] = ComplexConjugate[ H[..., (-n0) % N0, (-n1) % N1 ].
```
* This operator corresponds to a real matrix if and only if `H` is Hermitian.
* This operator is self-adjoint if and only if `H` is real.
See e.g. ""Discrete-Time Signal Processing"", Oppenheim and Schafer.
### Example of a self-adjoint positive definite operator
```python
# spectrum is real ==> operator is self-adjoint
# spectrum is positive ==> operator is positive definite
spectrum = [[1., 2., 3.],
[4., 5., 6.],
[7., 8., 9.]]
operator = LinearOperatorCirculant2D(spectrum)
# IFFT[spectrum]
operator.convolution_kernel()
==> [[5.0+0.0j, -0.5-.3j, -0.5+.3j],
[-1.5-.9j, 0, 0],
[-1.5+.9j, 0, 0]]
operator.to_dense()
==> Complex self adjoint 9 x 9 matrix.
```
#### Example of defining in terms of a real convolution kernel,
```python
# convolution_kernel is real ==> spectrum is Hermitian.
convolution_kernel = [[1., 2., 1.], [5., -1., 1.]]
spectrum = tf.signal.fft2d(tf.cast(convolution_kernel, tf.complex64))
# spectrum is shape [2, 3] ==> operator is shape [6, 6]
# spectrum is Hermitian ==> operator is real.
operator = LinearOperatorCirculant2D(spectrum, input_output_dtype=tf.float32)
```
#### Performance
Suppose `operator` is a `LinearOperatorCirculant` of shape `[N, N]`,
and `x.shape = [N, R]`. Then
* `operator.matmul(x)` is `O(R*N*Log[N])`
* `operator.solve(x)` is `O(R*N*Log[N])`
* `operator.determinant()` involves a size `N` `reduce_prod`.
If instead `operator` and `x` have shape `[B1,...,Bb, N, N]` and
`[B1,...,Bb, N, R]`, every operation increases in complexity by `B1*...*Bb`.
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9808,LinearOperatorCirculant3D,tensorflow/tensorflow/python/ops/linalg/linear_operator_circulant.py,939,class,"`LinearOperator` acting like a nested block circulant matrix.
This operator acts like a block circulant matrix `A` with
shape `[B1,...,Bb, N, N]` for some `b >= 0`. The first `b` indices index a
batch member. For every batch index `(i1,...,ib)`, `A[i1,...,ib, : :]` is
an `N x N` matrix. This matrix `A` is not materialized, but for
purposes of broadcasting this shape will be relevant.
#### Description in terms of block circulant matrices
If `A` is nested block circulant, with block sizes `N0, N1, N2`
(`N0 * N1 * N2 = N`):
`A` has a block structure, composed of `N0 x N0` blocks, with each
block an `N1 x N1` block circulant matrix.
For example, with `W`, `X`, `Y`, `Z` each block circulant,
```
A = |W Z Y X|
|X W Z Y|
|Y X W Z|
|Z Y X W|
```
Note that `A` itself will not in general be circulant.
#### Description in terms of the frequency spectrum
There is an equivalent description in terms of the [batch] spectrum `H` and
Fourier transforms. Here we consider `A.shape = [N, N]` and ignore batch
dimensions.
If `H.shape = [N0, N1, N2]`, (`N0 * N1 * N2 = N`):
Loosely speaking, matrix multiplication is equal to the action of a
Fourier multiplier: `A u = IDFT3[ H DFT3[u] ]`.
Precisely speaking, given `[N, R]` matrix `u`, let `DFT3[u]` be the
`[N0, N1, N2, R]` `Tensor` defined by re-shaping `u` to `[N0, N1, N2, R]` and
taking a three dimensional DFT across the first three dimensions. Let `IDFT3`
be the inverse of `DFT3`. Matrix multiplication may be expressed columnwise:
```(A u)_r = IDFT3[ H * (DFT3[u])_r ]```
#### Operator properties deduced from the spectrum.
* This operator is positive definite if and only if `Real{H} > 0`.
A general property of Fourier transforms is the correspondence between
Hermitian functions and real valued transforms.
Suppose `H.shape = [B1,...,Bb, N0, N1, N2]`, we say that `H` is a Hermitian
spectrum if, with `%` meaning modulus division,
```
H[..., n0 % N0, n1 % N1, n2 % N2]
= ComplexConjugate[ H[..., (-n0) % N0, (-n1) % N1, (-n2) % N2] ].
```
* This operator corresponds to a real matrix if and only if `H` is Hermitian.
* This operator is self-adjoint if and only if `H` is real.
See e.g. ""Discrete-Time Signal Processing"", Oppenheim and Schafer.
### Examples
See `LinearOperatorCirculant` and `LinearOperatorCirculant2D` for examples.
#### Performance
Suppose `operator` is a `LinearOperatorCirculant` of shape `[N, N]`,
and `x.shape = [N, R]`. Then
* `operator.matmul(x)` is `O(R*N*Log[N])`
* `operator.solve(x)` is `O(R*N*Log[N])`
* `operator.determinant()` involves a size `N` `reduce_prod`.
If instead `operator` and `x` have shape `[B1,...,Bb, N, N]` and
`[B1,...,Bb, N, R]`, every operation increases in complexity by `B1*...*Bb`.
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9809,LinearOperatorComposition,tensorflow/tensorflow/python/ops/linalg/linear_operator_composition.py,34,class,"Composes one or more `LinearOperators`.
This operator composes one or more linear operators `[op1,...,opJ]`,
building a new `LinearOperator` with action defined by:
```
op_composed(x) := op1(op2(...(opJ(x)...))
```
If `opj` acts like [batch] matrix `Aj`, then `op_composed` acts like the
[batch] matrix formed with the multiplication `A1 A2...AJ`.
If `opj` has shape `batch_shape_j + [M_j, N_j]`, then we must have
`N_j = M_{j+1}`, in which case the composed operator has shape equal to
`broadcast_batch_shape + [M_1, N_J]`, where `broadcast_batch_shape` is the
mutual broadcast of `batch_shape_j`, `j = 1,...,J`, assuming the intermediate
batch shapes broadcast. Even if the composed shape is well defined, the
composed operator's methods may fail due to lack of broadcasting ability in
the defining operators' methods.
```python
# Create a 2 x 2 linear operator composed of two 2 x 2 operators.
operator_1 = LinearOperatorFullMatrix([[1., 2.], [3., 4.]])
operator_2 = LinearOperatorFullMatrix([[1., 0.], [0., 1.]])
operator = LinearOperatorComposition([operator_1, operator_2])
operator.to_dense()
==> [[1., 2.]
[3., 4.]]
operator.shape
==> [2, 2]
operator.log_abs_determinant()
==> scalar Tensor
x = ... Shape [2, 4] Tensor
operator.matmul(x)
==> Shape [2, 4] Tensor
# Create a [2, 3] batch of 4 x 5 linear operators.
matrix_45 = tf.random.normal(shape=[2, 3, 4, 5])
operator_45 = LinearOperatorFullMatrix(matrix)
# Create a [2, 3] batch of 5 x 6 linear operators.
matrix_56 = tf.random.normal(shape=[2, 3, 5, 6])
operator_56 = LinearOperatorFullMatrix(matrix_56)
# Compose to create a [2, 3] batch of 4 x 6 operators.
operator_46 = LinearOperatorComposition([operator_45, operator_56])
# Create a shape [2, 3, 6, 2] vector.
x = tf.random.normal(shape=[2, 3, 6, 2])
operator.matmul(x)
==> Shape [2, 3, 4, 2] Tensor
```
#### Performance
The performance of `LinearOperatorComposition` on any operation is equal to
the sum of the individual operators' operations.
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning:
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9810,operators,tensorflow/tensorflow/python/ops/linalg/linear_operator_composition.py,190,method,
9811,LinearOperatorDiag,tensorflow/tensorflow/python/ops/linalg/linear_operator_diag.py,34,class,"`LinearOperator` acting like a [batch] square diagonal matrix.
This operator acts like a [batch] diagonal matrix `A` with shape
`[B1,...,Bb, N, N]` for some `b >= 0`. The first `b` indices index a
batch member. For every batch index `(i1,...,ib)`, `A[i1,...,ib, : :]` is
an `N x N` matrix. This matrix `A` is not materialized, but for
purposes of broadcasting this shape will be relevant.
`LinearOperatorDiag` is initialized with a (batch) vector.
```python
# Create a 2 x 2 diagonal linear operator.
diag = [1., -1.]
operator = LinearOperatorDiag(diag)
operator.to_dense()
==> [[1., 0.]
[0., -1.]]
operator.shape
==> [2, 2]
operator.log_abs_determinant()
==> scalar Tensor
x = ... Shape [2, 4] Tensor
operator.matmul(x)
==> Shape [2, 4] Tensor
# Create a [2, 3] batch of 4 x 4 linear operators.
diag = tf.random.normal(shape=[2, 3, 4])
operator = LinearOperatorDiag(diag)
# Create a shape [2, 1, 4, 2] vector. Note that this shape is compatible
# since the batch dimensions, [2, 1], are broadcast to
# operator.batch_shape = [2, 3].
y = tf.random.normal(shape=[2, 1, 4, 2])
x = operator.solve(y)
==> operator.matmul(x) = y
```
#### Shape compatibility
This operator acts on [batch] matrix with compatible shape.
`x` is a batch matrix with compatible shape for `matmul` and `solve` if
```
operator.shape = [B1,...,Bb] + [N, N], with b >= 0
x.shape = [C1,...,Cc] + [N, R],
and [C1,...,Cc] broadcasts with [B1,...,Bb] to [D1,...,Dd]
```
#### Performance
Suppose `operator` is a `LinearOperatorDiag` of shape `[N, N]`,
and `x.shape = [N, R]`. Then
* `operator.matmul(x)` involves `N * R` multiplications.
* `operator.solve(x)` involves `N` divisions and `N * R` multiplications.
* `operator.determinant()` involves a size `N` `reduce_prod`.
If instead `operator` and `x` have shape `[B1,...,Bb, N, N]` and
`[B1,...,Bb, N, R]`, every operation increases in complexity by `B1*...*Bb`.
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning:
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9812,diag,tensorflow/tensorflow/python/ops/linalg/linear_operator_diag.py,187,method,
9813,LinearOperatorFullMatrix,tensorflow/tensorflow/python/ops/linalg/linear_operator_full_matrix.py,33,class,"`LinearOperator` that wraps a [batch] matrix.
This operator wraps a [batch] matrix `A` (which is a `Tensor`) with shape
`[B1,...,Bb, M, N]` for some `b >= 0`. The first `b` indices index a
batch member. For every batch index `(i1,...,ib)`, `A[i1,...,ib, : :]` is
an `M x N` matrix.
```python
# Create a 2 x 2 linear operator.
matrix = [[1., 2.], [3., 4.]]
operator = LinearOperatorFullMatrix(matrix)
operator.to_dense()
==> [[1., 2.]
[3., 4.]]
operator.shape
==> [2, 2]
operator.log_abs_determinant()
==> scalar Tensor
x = ... Shape [2, 4] Tensor
operator.matmul(x)
==> Shape [2, 4] Tensor
# Create a [2, 3] batch of 4 x 4 linear operators.
matrix = tf.random.normal(shape=[2, 3, 4, 4])
operator = LinearOperatorFullMatrix(matrix)
```
#### Shape compatibility
This operator acts on [batch] matrix with compatible shape.
`x` is a batch matrix with compatible shape for `matmul` and `solve` if
```
operator.shape = [B1,...,Bb] + [M, N], with b >= 0
x.shape = [B1,...,Bb] + [N, R], with R >= 0.
```
#### Performance
`LinearOperatorFullMatrix` has exactly the same performance as would be
achieved by using standard `TensorFlow` matrix ops. Intelligent choices are
made based on the following initialization hints.
* If `dtype` is real, and `is_self_adjoint` and `is_positive_definite`, a
Cholesky factorization is used for the determinant and solve.
In all cases, suppose `operator` is a `LinearOperatorFullMatrix` of shape
`[M, N]`, and `x.shape = [N, R]`. Then
* `operator.matmul(x)` is `O(M * N * R)`.
* If `M=N`, `operator.solve(x)` is `O(N^3 * R)`.
* If `M=N`, `operator.determinant()` is `O(N^3)`.
If instead `operator` and `x` have shape `[B1,...,Bb, M, N]` and
`[B1,...,Bb, N, R]`, every operation increases in complexity by `B1*...*Bb`.
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning:
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9814,LinearOperatorHouseholder,tensorflow/tensorflow/python/ops/linalg/linear_operator_householder.py,35,class,"`LinearOperator` acting like a [batch] of Householder transformations.
This operator acts like a [batch] of householder reflections with shape
`[B1,...,Bb, N, N]` for some `b >= 0`. The first `b` indices index a
batch member. For every batch index `(i1,...,ib)`, `A[i1,...,ib, : :]` is
an `N x N` matrix. This matrix `A` is not materialized, but for
purposes of broadcasting this shape will be relevant.
`LinearOperatorHouseholder` is initialized with a (batch) vector.
A Householder reflection, defined via a vector `v`, which reflects points
in `R^n` about the hyperplane orthogonal to `v` and through the origin.
```python
# Create a 2 x 2 householder transform.
vec = [1 / np.sqrt(2), 1. / np.sqrt(2)]
operator = LinearOperatorHouseholder(vec)
operator.to_dense()
==> [[0., -1.]
[-1., -0.]]
operator.shape
==> [2, 2]
operator.log_abs_determinant()
==> scalar Tensor
x = ... Shape [2, 4] Tensor
operator.matmul(x)
==> Shape [2, 4] Tensor
```
#### Shape compatibility
This operator acts on [batch] matrix with compatible shape.
`x` is a batch matrix with compatible shape for `matmul` and `solve` if
```
operator.shape = [B1,...,Bb] + [N, N], with b >= 0
x.shape = [C1,...,Cc] + [N, R],
and [C1,...,Cc] broadcasts with [B1,...,Bb] to [D1,...,Dd]
```
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning:
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9815,reflection_axis,tensorflow/tensorflow/python/ops/linalg/linear_operator_householder.py,262,method,
9816,BaseLinearOperatorIdentity,tensorflow/tensorflow/python/ops/linalg/linear_operator_identity.py,42,class,Base class for Identity operators.
9817,LinearOperatorIdentity,tensorflow/tensorflow/python/ops/linalg/linear_operator_identity.py,102,class,"`LinearOperator` acting like a [batch] square identity matrix.
This operator acts like a [batch] identity matrix `A` with shape
`[B1,...,Bb, N, N]` for some `b >= 0`. The first `b` indices index a
batch member. For every batch index `(i1,...,ib)`, `A[i1,...,ib, : :]` is
an `N x N` matrix. This matrix `A` is not materialized, but for
purposes of broadcasting this shape will be relevant.
`LinearOperatorIdentity` is initialized with `num_rows`, and optionally
`batch_shape`, and `dtype` arguments. If `batch_shape` is `None`, this
operator efficiently passes through all arguments. If `batch_shape` is
provided, broadcasting may occur, which will require making copies.
```python
# Create a 2 x 2 identity matrix.
operator = LinearOperatorIdentity(num_rows=2, dtype=tf.float32)
operator.to_dense()
==> [[1., 0.]
[0., 1.]]
operator.shape
==> [2, 2]
operator.log_abs_determinant()
==> 0.
x = ... Shape [2, 4] Tensor
operator.matmul(x)
==> Shape [2, 4] Tensor, same as x.
y = tf.random.normal(shape=[3, 2, 4])
# Note that y.shape is compatible with operator.shape because operator.shape
# is broadcast to [3, 2, 2].
# This broadcast does NOT require copying data, since we can infer that y
# will be passed through without changing shape. We are always able to infer
# this if the operator has no batch_shape.
x = operator.solve(y)
==> Shape [3, 2, 4] Tensor, same as y.
# Create a 2-batch of 2x2 identity matrices
operator = LinearOperatorIdentity(num_rows=2, batch_shape=[2])
operator.to_dense()
==> [[[1., 0.]
[0., 1.]],
[[1., 0.]
[0., 1.]]]
# Here, even though the operator has a batch shape, the input is the same as
# the output, so x can be passed through without a copy. The operator is able
# to detect that no broadcast is necessary because both x and the operator
# have statically defined shape.
x = ... Shape [2, 2, 3]
operator.matmul(x)
==> Shape [2, 2, 3] Tensor, same as x
# Here the operator and x have different batch_shape, and are broadcast.
# This requires a copy, since the output is different size than the input.
x = ... Shape [1, 2, 3]
operator.matmul(x)
==> Shape [2, 2, 3] Tensor, equal to [x, x]
```
### Shape compatibility
This operator acts on [batch] matrix with compatible shape.
`x` is a batch matrix with compatible shape for `matmul` and `solve` if
```
operator.shape = [B1,...,Bb] + [N, N], with b >= 0
x.shape = [C1,...,Cc] + [N, R],
and [C1,...,Cc] broadcasts with [B1,...,Bb] to [D1,...,Dd]
```
### Performance
If `batch_shape` initialization arg is `None`:
* `operator.matmul(x)` is `O(1)`
* `operator.solve(x)` is `O(1)`
* `operator.determinant()` is `O(1)`
If `batch_shape` initialization arg is provided, and static checks cannot
rule out the need to broadcast:
* `operator.matmul(x)` is `O(D1*...*Dd*N*R)`
* `operator.solve(x)` is `O(D1*...*Dd*N*R)`
* `operator.determinant()` is `O(B1*...*Bb)`
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning:
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9818,add_to_tensor,tensorflow/tensorflow/python/ops/linalg/linear_operator_identity.py,386,method,"Add matrix represented by this operator to `mat`. Equiv to `I + mat`.
Args:
mat: `Tensor` with same `dtype` and shape broadcastable to `self`.
name: A name to give this `Op`.
Returns:
A `Tensor` with broadcast shape and same `dtype` as `self`."
9819,LinearOperatorScaledIdentity,tensorflow/tensorflow/python/ops/linalg/linear_operator_identity.py,475,class,"`LinearOperator` acting like a scaled [batch] identity matrix `A = c I`.
This operator acts like a scaled [batch] identity matrix `A` with shape
`[B1,...,Bb, N, N]` for some `b >= 0`. The first `b` indices index a
batch member. For every batch index `(i1,...,ib)`, `A[i1,...,ib, : :]` is
a scaled version of the `N x N` identity matrix.
`LinearOperatorIdentity` is initialized with `num_rows`, and a `multiplier`
(a `Tensor`) of shape `[B1,...,Bb]`. `N` is set to `num_rows`, and the
`multiplier` determines the scale for each batch member.
```python
# Create a 2 x 2 scaled identity matrix.
operator = LinearOperatorIdentity(num_rows=2, multiplier=3.)
operator.to_dense()
==> [[3., 0.]
[0., 3.]]
operator.shape
==> [2, 2]
operator.log_abs_determinant()
==> 2 * Log[3]
x = ... Shape [2, 4] Tensor
operator.matmul(x)
==> 3 * x
y = tf.random.normal(shape=[3, 2, 4])
# Note that y.shape is compatible with operator.shape because operator.shape
# is broadcast to [3, 2, 2].
x = operator.solve(y)
==> 3 * x
# Create a 2-batch of 2x2 identity matrices
operator = LinearOperatorIdentity(num_rows=2, multiplier=5.)
operator.to_dense()
==> [[[5., 0.]
[0., 5.]],
[[5., 0.]
[0., 5.]]]
x = ... Shape [2, 2, 3]
operator.matmul(x)
==> 5 * x
# Here the operator and x have different batch_shape, and are broadcast.
x = ... Shape [1, 2, 3]
operator.matmul(x)
==> 5 * x
```
### Shape compatibility
This operator acts on [batch] matrix with compatible shape.
`x` is a batch matrix with compatible shape for `matmul` and `solve` if
```
operator.shape = [B1,...,Bb] + [N, N], with b >= 0
x.shape = [C1,...,Cc] + [N, R],
and [C1,...,Cc] broadcasts with [B1,...,Bb] to [D1,...,Dd]
```
### Performance
* `operator.matmul(x)` is `O(D1*...*Dd*N*R)`
* `operator.solve(x)` is `O(D1*...*Dd*N*R)`
* `operator.determinant()` is `O(D1*...*Dd)`
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9820,add_to_tensor,tensorflow/tensorflow/python/ops/linalg/linear_operator_identity.py,708,method,"Add matrix represented by this operator to `mat`. Equiv to `I + mat`.
Args:
mat: `Tensor` with same `dtype` and shape broadcastable to `self`.
name: A name to give this `Op`.
Returns:
A `Tensor` with broadcast shape and same `dtype` as `self`."
9821,multiplier,tensorflow/tensorflow/python/ops/linalg/linear_operator_identity.py,745,method,"The [batch] scalar `Tensor`, `c` in `cI`."
9822,LinearOperatorInversion,tensorflow/tensorflow/python/ops/linalg/linear_operator_inversion.py,30,class,"`LinearOperator` representing the inverse of another operator.
This operator represents the inverse of another operator.
```python
# Create a 2 x 2 linear operator.
operator = LinearOperatorFullMatrix([[1., 0.], [0., 2.]])
operator_inv = LinearOperatorInversion(operator)
operator_inv.to_dense()
==> [[1., 0.]
[0., 0.5]]
operator_inv.shape
==> [2, 2]
operator_inv.log_abs_determinant()
==> - log(2)
x = ... Shape [2, 4] Tensor
operator_inv.matmul(x)
==> Shape [2, 4] Tensor, equal to operator.solve(x)
```
#### Performance
The performance of `LinearOperatorInversion` depends on the underlying
operators performance: `solve` and `matmul` are swapped, and determinant is
inverted.
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning:
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9823,operator,tensorflow/tensorflow/python/ops/linalg/linear_operator_inversion.py,171,method,The operator before inversion.
9824,LinearOperatorKronecker,tensorflow/tensorflow/python/ops/linalg/linear_operator_kronecker.py,66,class,"Kronecker product between two `LinearOperators`.
This operator composes one or more linear operators `[op1,...,opJ]`,
building a new `LinearOperator` representing the Kronecker product:
`op1 x op2 x .. opJ` (we omit parentheses as the Kronecker product is
associative).
If `opj` has shape `batch_shape_j + [M_j, N_j]`, then the composed operator
will have shape equal to `broadcast_batch_shape + [prod M_j, prod N_j]`,
where the product is over all operators.
```python
# Create a 4 x 4 linear operator composed of two 2 x 2 operators.
operator_1 = LinearOperatorFullMatrix([[1., 2.], [3., 4.]])
operator_2 = LinearOperatorFullMatrix([[1., 0.], [2., 1.]])
operator = LinearOperatorKronecker([operator_1, operator_2])
operator.to_dense()
==> [[1., 0., 2., 0.],
[2., 1., 4., 2.],
[3., 0., 4., 0.],
[6., 3., 8., 4.]]
operator.shape
==> [4, 4]
operator.log_abs_determinant()
==> scalar Tensor
x = ... Shape [4, 2] Tensor
operator.matmul(x)
==> Shape [4, 2] Tensor
# Create a [2, 3] batch of 4 x 5 linear operators.
matrix_45 = tf.random.normal(shape=[2, 3, 4, 5])
operator_45 = LinearOperatorFullMatrix(matrix)
# Create a [2, 3] batch of 5 x 6 linear operators.
matrix_56 = tf.random.normal(shape=[2, 3, 5, 6])
operator_56 = LinearOperatorFullMatrix(matrix_56)
# Compose to create a [2, 3] batch of 20 x 30 operators.
operator_large = LinearOperatorKronecker([operator_45, operator_56])
# Create a shape [2, 3, 20, 2] vector.
x = tf.random.normal(shape=[2, 3, 6, 2])
operator_large.matmul(x)
==> Shape [2, 3, 30, 2] Tensor
```
#### Performance
The performance of `LinearOperatorKronecker` on any operation is equal to
the sum of the individual operators' operations.
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning:
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9825,operators,tensorflow/tensorflow/python/ops/linalg/linear_operator_kronecker.py,234,method,
9826,LinearOperatorLowRankUpdate,tensorflow/tensorflow/python/ops/linalg/linear_operator_low_rank_update.py,39,class,"Perturb a `LinearOperator` with a rank `K` update.
This operator acts like a [batch] matrix `A` with shape
`[B1,...,Bb, M, N]` for some `b >= 0`. The first `b` indices index a
batch member. For every batch index `(i1,...,ib)`, `A[i1,...,ib, : :]` is
an `M x N` matrix.
`LinearOperatorLowRankUpdate` represents `A = L + U D V^H`, where
```
L, is a LinearOperator representing [batch] M x N matrices
U, is a [batch] M x K matrix. Typically K << M.
D, is a [batch] K x K matrix.
V, is a [batch] N x K matrix. Typically K << N.
V^H is the Hermitian transpose (adjoint) of V.
```
If `M = N`, determinants and solves are done using the matrix determinant
lemma and Woodbury identities, and thus require L and D to be non-singular.
Solves and determinants will be attempted unless the ""is_non_singular""
property of L and D is False.
In the event that L and D are positive-definite, and U = V, solves and
determinants can be done using a Cholesky factorization.
```python
# Create a 3 x 3 diagonal linear operator.
diag_operator = LinearOperatorDiag(
diag_update=[1., 2., 3.], is_non_singular=True, is_self_adjoint=True,
is_positive_definite=True)
# Perturb with a rank 2 perturbation
operator = LinearOperatorLowRankUpdate(
operator=diag_operator,
u=[[1., 2.], [-1., 3.], [0., 0.]],
diag_update=[11., 12.],
v=[[1., 2.], [-1., 3.], [10., 10.]])
operator.shape
==> [3, 3]
operator.log_abs_determinant()
==> scalar Tensor
x = ... Shape [3, 4] Tensor
operator.matmul(x)
==> Shape [3, 4] Tensor
```
### Shape compatibility
This operator acts on [batch] matrix with compatible shape.
`x` is a batch matrix with compatible shape for `matmul` and `solve` if
```
operator.shape = [B1,...,Bb] + [M, N], with b >= 0
x.shape = [B1,...,Bb] + [N, R], with R >= 0.
```
### Performance
Suppose `operator` is a `LinearOperatorLowRankUpdate` of shape `[M, N]`,
made from a rank `K` update of `base_operator` which performs `.matmul(x)` on
`x` having `x.shape = [N, R]` with `O(L_matmul*N*R)` complexity (and similarly
for `solve`, `determinant`. Then, if `x.shape = [N, R]`,
* `operator.matmul(x)` is `O(L_matmul*N*R + K*N*R)`
and if `M = N`,
* `operator.solve(x)` is `O(L_matmul*N*R + N*K*R + K^2*R + K^3)`
* `operator.determinant()` is `O(L_determinant + L_solve*N*K + K^2*N + K^3)`
If instead `operator` and `x` have shape `[B1,...,Bb, M, N]` and
`[B1,...,Bb, N, R]`, every operation increases in complexity by `B1*...*Bb`.
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular`, `self_adjoint`, `positive_definite`,
`diag_update_positive` and `square`. These have the following meaning:
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9827,u,tensorflow/tensorflow/python/ops/linalg/linear_operator_low_rank_update.py,298,method,"If this operator is `A = L + U D V^H`, this is the `U`."
9828,v,tensorflow/tensorflow/python/ops/linalg/linear_operator_low_rank_update.py,303,method,"If this operator is `A = L + U D V^H`, this is the `V`."
9829,is_diag_update_positive,tensorflow/tensorflow/python/ops/linalg/linear_operator_low_rank_update.py,308,method,"If this operator is `A = L + U D V^H`, this hints `D > 0` elementwise."
9830,diag_update,tensorflow/tensorflow/python/ops/linalg/linear_operator_low_rank_update.py,313,method,"If this operator is `A = L + U D V^H`, this is the diagonal of `D`."
9831,diag_operator,tensorflow/tensorflow/python/ops/linalg/linear_operator_low_rank_update.py,318,method,"If this operator is `A = L + U D V^H`, this is `D`."
9832,base_operator,tensorflow/tensorflow/python/ops/linalg/linear_operator_low_rank_update.py,323,method,"If this operator is `A = L + U D V^H`, this is the `L`."
9833,LinearOperatorLowerTriangular,tensorflow/tensorflow/python/ops/linalg/linear_operator_lower_triangular.py,35,class,"`LinearOperator` acting like a [batch] square lower triangular matrix.
This operator acts like a [batch] lower triangular matrix `A` with shape
`[B1,...,Bb, N, N]` for some `b >= 0`. The first `b` indices index a
batch member. For every batch index `(i1,...,ib)`, `A[i1,...,ib, : :]` is
an `N x N` matrix.
`LinearOperatorLowerTriangular` is initialized with a `Tensor` having
dimensions `[B1,...,Bb, N, N]`. The upper triangle of the last two
dimensions is ignored.
```python
# Create a 2 x 2 lower-triangular linear operator.
tril = [[1., 2.], [3., 4.]]
operator = LinearOperatorLowerTriangular(tril)
# The upper triangle is ignored.
operator.to_dense()
==> [[1., 0.]
[3., 4.]]
operator.shape
==> [2, 2]
operator.log_abs_determinant()
==> scalar Tensor
x = ... Shape [2, 4] Tensor
operator.matmul(x)
==> Shape [2, 4] Tensor
# Create a [2, 3] batch of 4 x 4 linear operators.
tril = tf.random.normal(shape=[2, 3, 4, 4])
operator = LinearOperatorLowerTriangular(tril)
```
#### Shape compatibility
This operator acts on [batch] matrix with compatible shape.
`x` is a batch matrix with compatible shape for `matmul` and `solve` if
```
operator.shape = [B1,...,Bb] + [N, N], with b >= 0
x.shape = [B1,...,Bb] + [N, R], with R >= 0.
```
#### Performance
Suppose `operator` is a `LinearOperatorLowerTriangular` of shape `[N, N]`,
and `x.shape = [N, R]`. Then
* `operator.matmul(x)` involves `N^2 * R` multiplications.
* `operator.solve(x)` involves `N * R` size `N` back-substitutions.
* `operator.determinant()` involves a size `N` `reduce_prod`.
If instead `operator` and `x` have shape `[B1,...,Bb, N, N]` and
`[B1,...,Bb, N, R]`, every operation increases in complexity by `B1*...*Bb`.
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning:
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9834,LinearOperatorPermutation,tensorflow/tensorflow/python/ops/linalg/linear_operator_permutation.py,39,class,"`LinearOperator` acting like a [batch] of permutation matrices.
This operator acts like a [batch] of permutations with shape
`[B1,...,Bb, N, N]` for some `b >= 0`. The first `b` indices index a
batch member. For every batch index `(i1,...,ib)`, `A[i1,...,ib, : :]` is
an `N x N` matrix. This matrix `A` is not materialized, but for
purposes of broadcasting this shape will be relevant.
`LinearOperatorPermutation` is initialized with a (batch) vector.
A permutation, is defined by an integer vector `v` whose values are unique
and are in the range `[0, ... n]`. Applying the permutation on an input
matrix has the folllowing meaning: the value of `v` at index `i`
says to move the `v[i]`-th row of the input matrix to the `i`-th row.
Because all values are unique, this will result in a permutation of the
rows the input matrix. Note, that the permutation vector `v` has the same
semantics as `tf.transpose`.
```python
# Create a 3 x 3 permutation matrix that swaps the last two columns.
vec = [0, 2, 1]
operator = LinearOperatorPermutation(vec)
operator.to_dense()
==> [[1., 0., 0.]
[0., 0., 1.]
[0., 1., 0.]]
operator.shape
==> [3, 3]
# This will be zero.
operator.log_abs_determinant()
==> scalar Tensor
x = ... Shape [3, 4] Tensor
operator.matmul(x)
==> Shape [3, 4] Tensor
```
#### Shape compatibility
This operator acts on [batch] matrix with compatible shape.
`x` is a batch matrix with compatible shape for `matmul` and `solve` if
```
operator.shape = [B1,...,Bb] + [N, N], with b >= 0
x.shape = [C1,...,Cc] + [N, R],
and [C1,...,Cc] broadcasts with [B1,...,Bb] to [D1,...,Dd]
```
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning:
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9835,perm,tensorflow/tensorflow/python/ops/linalg/linear_operator_permutation.py,251,method,
9836,OperatorShapesInfo,tensorflow/tensorflow/python/ops/linalg/linear_operator_test_util.py,45,class,"Object encoding expected shape for a test.
Encodes the expected shape of a matrix for a test. Also
allows additional metadata for the test harness."
9837,CheckTapeSafeSkipOptions,tensorflow/tensorflow/python/ops/linalg/linear_operator_test_util.py,57,class,
9838,random_positive_definite_matrix,tensorflow/tensorflow/python/ops/linalg/linear_operator_test_util.py,895,function,"[batch] positive definite Wisart matrix.
A Wishart(N, S) matrix is the S sample covariance matrix of an N-variate
(standard) Normal random variable.
Args:
shape: `TensorShape` or Python list. Shape of the returned matrix.
dtype: `TensorFlow` `dtype` or Python dtype.
oversampling_ratio: S / N in the above. If S < N, the matrix will be
singular (unless `force_well_conditioned is True`).
force_well_conditioned: Python bool. If `True`, add `1` to the diagonal
of the Wishart matrix, then divide by 2, ensuring most eigenvalues are
close to 1.
Returns:
`Tensor` with desired shape and dtype."
9839,random_tril_matrix,tensorflow/tensorflow/python/ops/linalg/linear_operator_test_util.py,938,function,"[batch] lower triangular matrix.
Args:
shape: `TensorShape` or Python `list`. Shape of the returned matrix.
dtype: `TensorFlow` `dtype` or Python dtype
force_well_conditioned: Python `bool`. If `True`, returned matrix will have
eigenvalues with modulus in `(1, 2)`. Otherwise, eigenvalues are unit
normal random variables.
remove_upper: Python `bool`.
If `True`, zero out the strictly upper triangle.
If `False`, the lower triangle of returned matrix will have desired
properties, but will not have the strictly upper triangle zero'd out.
Returns:
`Tensor` with desired shape and dtype."
9840,random_normal,tensorflow/tensorflow/python/ops/linalg/linear_operator_test_util.py,974,function,"Tensor with (possibly complex) Gaussian entries.
Samples are distributed like
```
N(mean, stddev^2), if dtype is real,
X + iY, where X, Y ~ N(mean, stddev^2) if dtype is complex.
```
Args:
shape: `TensorShape` or Python list. Shape of the returned tensor.
mean: `Tensor` giving mean of normal to sample from.
stddev: `Tensor` giving stdev of normal to sample from.
dtype: `TensorFlow` `dtype` or numpy dtype
seed: Python integer seed for the RNG.
Returns:
`Tensor` with desired shape and dtype."
9841,random_uniform,tensorflow/tensorflow/python/ops/linalg/linear_operator_test_util.py,1008,function,"Tensor with (possibly complex) Uniform entries.
Samples are distributed like
```
Uniform[minval, maxval], if dtype is real,
X + iY, where X, Y ~ Uniform[minval, maxval], if dtype is complex.
```
Args:
shape: `TensorShape` or Python list. Shape of the returned tensor.
minval: `0-D` `Tensor` giving the minimum values.
maxval: `0-D` `Tensor` giving the maximum values.
dtype: `TensorFlow` `dtype` or Python dtype
seed: Python integer seed for the RNG.
Returns:
`Tensor` with desired shape and dtype."
9842,random_sign_uniform,tensorflow/tensorflow/python/ops/linalg/linear_operator_test_util.py,1050,function,"Tensor with (possibly complex) random entries from a ""sign Uniform"".
Letting `Z` be a random variable equal to `-1` and `1` with equal probability,
Samples from this `Op` are distributed like
```
Z * X, where X ~ Uniform[minval, maxval], if dtype is real,
Z * (X + iY), where X, Y ~ Uniform[minval, maxval], if dtype is complex.
```
Args:
shape: `TensorShape` or Python list. Shape of the returned tensor.
minval: `0-D` `Tensor` giving the minimum values.
maxval: `0-D` `Tensor` giving the maximum values.
dtype: `TensorFlow` `dtype` or Python dtype
seed: Python integer seed for the RNG.
Returns:
`Tensor` with desired shape and dtype."
9843,random_normal_correlated_columns,tensorflow/tensorflow/python/ops/linalg/linear_operator_test_util.py,1087,function,"Batch matrix with (possibly complex) Gaussian entries and correlated cols.
Returns random batch matrix `A` with specified element-wise `mean`, `stddev`,
living close to an embedded hyperplane.
Suppose `shape[-2:] = (M, N)`.
If `M < N`, `A` is a random `M x N` [batch] matrix with iid Gaussian entries.
If `M >= N`, then the columns of `A` will be made almost dependent as follows:
```
L = random normal N x N-1 matrix, mean = 0, stddev = 1 / sqrt(N - 1)
B = random normal M x N-1 matrix, mean = 0, stddev = stddev.
G = (L B^H)^H, a random normal M x N matrix, living on N-1 dim hyperplane
E = a random normal M x N matrix, mean = 0, stddev = eps
mu = a constant M x N matrix, equal to the argument ""mean""
A = G + E + mu
```
Args:
shape: Python list of integers.
Shape of the returned tensor. Must be at least length two.
mean: `Tensor` giving mean of normal to sample from.
stddev: `Tensor` giving stdev of normal to sample from.
dtype: `TensorFlow` `dtype` or numpy dtype
eps: Distance each column is perturbed from the low-dimensional subspace.
seed: Python integer seed for the RNG.
Returns:
`Tensor` with desired shape and dtype.
Raises:
ValueError: If `shape` is not at least length 2."
9844,LinearOperatorToeplitz,tensorflow/tensorflow/python/ops/linalg/linear_operator_toeplitz.py,37,class,"`LinearOperator` acting like a [batch] of toeplitz matrices.
This operator acts like a [batch] Toeplitz matrix `A` with shape
`[B1,...,Bb, N, N]` for some `b >= 0`. The first `b` indices index a
batch member. For every batch index `(i1,...,ib)`, `A[i1,...,ib, : :]` is
an `N x N` matrix. This matrix `A` is not materialized, but for
purposes of broadcasting this shape will be relevant.
#### Description in terms of toeplitz matrices
Toeplitz means that `A` has constant diagonals. Hence, `A` can be generated
with two vectors. One represents the first column of the matrix, and the
other represents the first row.
Below is a 4 x 4 example:
```
A = |a b c d|
|e a b c|
|f e a b|
|g f e a|
```
#### Example of a Toeplitz operator.
```python
# Create a 3 x 3 Toeplitz operator.
col = [1., 2., 3.]
row = [1., 4., -9.]
operator = LinearOperatorToeplitz(col, row)
operator.to_dense()
==> [[1., 4., -9.],
[2., 1., 4.],
[3., 2., 1.]]
operator.shape
==> [3, 3]
operator.log_abs_determinant()
==> scalar Tensor
x = ... Shape [3, 4] Tensor
operator.matmul(x)
==> Shape [3, 4] Tensor
```
#### Shape compatibility
This operator acts on [batch] matrix with compatible shape.
`x` is a batch matrix with compatible shape for `matmul` and `solve` if
```
operator.shape = [B1,...,Bb] + [N, N], with b >= 0
x.shape = [C1,...,Cc] + [N, R],
and [C1,...,Cc] broadcasts with [B1,...,Bb] to [D1,...,Dd]
```
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning:
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9845,col,tensorflow/tensorflow/python/ops/linalg/linear_operator_toeplitz.py,266,method,
9846,row,tensorflow/tensorflow/python/ops/linalg/linear_operator_toeplitz.py,270,method,
9847,LinearOperatorTridiag,tensorflow/tensorflow/python/ops/linalg/linear_operator_tridiag.py,42,class,"`LinearOperator` acting like a [batch] square tridiagonal matrix.
This operator acts like a [batch] square tridiagonal matrix `A` with shape
`[B1,...,Bb, N, N]` for some `b >= 0`. The first `b` indices index a
batch member. For every batch index `(i1,...,ib)`, `A[i1,...,ib, : :]` is
an `N x M` matrix. This matrix `A` is not materialized, but for
purposes of broadcasting this shape will be relevant.
Example usage:
Create a 3 x 3 tridiagonal linear operator.
>>> superdiag = [3., 4., 5.]
>>> diag = [1., -1., 2.]
>>> subdiag = [6., 7., 8]
>>> operator = tf.linalg.LinearOperatorTridiag(
... [superdiag, diag, subdiag],
... diagonals_format='sequence')
>>> operator.to_dense()
<tf.Tensor: shape=(3, 3), dtype=float32, numpy=
array([[ 1., 3., 0.],
[ 7., -1., 4.],
[ 0., 8., 2.]], dtype=float32)>
>>> operator.shape
TensorShape([3, 3])
Scalar Tensor output.
>>> operator.log_abs_determinant()
<tf.Tensor: shape=(), dtype=float32, numpy=4.3307333>
Create a [2, 3] batch of 4 x 4 linear operators.
>>> diagonals = tf.random.normal(shape=[2, 3, 3, 4])
>>> operator = tf.linalg.LinearOperatorTridiag(
... diagonals,
... diagonals_format='compact')
Create a shape [2, 1, 4, 2] vector. Note that this shape is compatible
since the batch dimensions, [2, 1], are broadcast to
operator.batch_shape = [2, 3].
>>> y = tf.random.normal(shape=[2, 1, 4, 2])
>>> x = operator.solve(y)
>>> x
<tf.Tensor: shape=(2, 3, 4, 2), dtype=float32, numpy=...,
dtype=float32)>
#### Shape compatibility
This operator acts on [batch] matrix with compatible shape.
`x` is a batch matrix with compatible shape for `matmul` and `solve` if
```
operator.shape = [B1,...,Bb] + [N, N], with b >= 0
x.shape = [C1,...,Cc] + [N, R],
and [C1,...,Cc] broadcasts with [B1,...,Bb].
```
#### Performance
Suppose `operator` is a `LinearOperatorTridiag` of shape `[N, N]`,
and `x.shape = [N, R]`. Then
* `operator.matmul(x)` will take O(N * R) time.
* `operator.solve(x)` will take O(N * R) time.
If instead `operator` and `x` have shape `[B1,...,Bb, N, N]` and
`[B1,...,Bb, N, R]`, every operation increases in complexity by `B1*...*Bb`.
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning:
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9848,diagonals,tensorflow/tensorflow/python/ops/linalg/linear_operator_tridiag.py,368,method,
9849,diagonals_format,tensorflow/tensorflow/python/ops/linalg/linear_operator_tridiag.py,372,method,
9850,convert_nonref_to_tensor,tensorflow/tensorflow/python/ops/linalg/linear_operator_util.py,40,function,"Converts the given `value` to a `Tensor` if input is nonreference type.
This function converts Python objects of various types to `Tensor` objects
except if the input has nonreference semantics. Reference semantics are
characterized by `is_ref` and is any object which is a
`tf.Variable` or instance of `tf.Module`. This function accepts any input
which `tf.convert_to_tensor` would also.
Note: This function diverges from default Numpy behavior for `float` and
`string` types when `None` is present in a Python list or scalar. Rather
than silently converting `None` values, an error will be thrown.
Args:
value: An object whose type has a registered `Tensor` conversion function.
dtype: Optional element type for the returned tensor. If missing, the
type is inferred from the type of `value`.
dtype_hint: Optional element type for the returned tensor,
used when dtype is None. In some cases, a caller may not have a
dtype in mind when converting to a tensor, so dtype_hint
can be used as a soft preference. If the conversion to
`dtype_hint` is not possible, this argument has no effect.
name: Optional name to use if a new `Tensor` is created.
Returns:
tensor: A `Tensor` based on `value`.
Raises:
TypeError: If no conversion function is registered for `value` to `dtype`.
RuntimeError: If a registered conversion function returns an invalid value.
ValueError: If the `value` is a tensor not of given `dtype` in graph mode.
#### Examples:
```python
x = tf.Variable(0.)
y = convert_nonref_to_tensor(x)
x is y
# ==> True
x = tf.constant(0.)
y = convert_nonref_to_tensor(x)
x is y
# ==> True
x = np.array(0.)
y = convert_nonref_to_tensor(x)
x is y
# ==> False
tf.is_tensor(y)
# ==> True
x = tfp.util.DeferredTensor(13.37, lambda x: x)
y = convert_nonref_to_tensor(x)
x is y
# ==> True
tf.is_tensor(y)
# ==> False
tf.equal(y, 13.37)
# ==> True
```"
9851,base_dtype,tensorflow/tensorflow/python/ops/linalg/linear_operator_util.py,121,function,Returns a non-reference `dtype` based on this `dtype`.
9852,dtype_name,tensorflow/tensorflow/python/ops/linalg/linear_operator_util.py,129,function,Returns the string name for this `dtype`.
9853,check_dtype,tensorflow/tensorflow/python/ops/linalg/linear_operator_util.py,139,function,Check that arg.dtype == self.dtype.
9854,is_ref,tensorflow/tensorflow/python/ops/linalg/linear_operator_util.py,147,function,"Evaluates if the object has reference semantics.
An object is deemed ""reference"" if it is a `tf.Variable` instance or is
derived from a `tf.Module` with `dtype` and `shape` properties.
Args:
x: Any object.
Returns:
is_ref: Python `bool` indicating input is has nonreference semantics, i.e.,
is a `tf.Variable` or a `tf.Module` with `dtype` and `shape` properties."
9855,assert_not_ref_type,tensorflow/tensorflow/python/ops/linalg/linear_operator_util.py,168,function,
9856,assert_no_entries_with_modulus_zero,tensorflow/tensorflow/python/ops/linalg/linear_operator_util.py,179,function,"Returns `Op` that asserts Tensor `x` has no entries with modulus zero.
Args:
x: Numeric `Tensor`, real, integer, or complex.
message: A string message to prepend to failure message.
name: A name to give this `Op`.
Returns:
An `Op` that asserts `x` has no entries with modulus zero."
9857,assert_zero_imag_part,tensorflow/tensorflow/python/ops/linalg/linear_operator_util.py,199,function,"Returns `Op` that asserts Tensor `x` has no non-zero imaginary parts.
Args:
x: Numeric `Tensor`, real, integer, or complex.
message: A string message to prepend to failure message.
name: A name to give this `Op`.
Returns:
An `Op` that asserts `x` has no entries with modulus zero."
9858,assert_compatible_matrix_dimensions,tensorflow/tensorflow/python/ops/linalg/linear_operator_util.py,221,function,"Assert that an argument to solve/matmul has proper domain dimension.
If `operator.shape[-2:] = [M, N]`, and `x.shape[-2:] = [Q, R]`, then
`operator.matmul(x)` is defined only if `N = Q`. This `Op` returns an
`Assert` that ""fires"" if this is not the case. Static checks are already
done by the base class `LinearOperator`.
Args:
operator: `LinearOperator`.
x: `Tensor`.
Returns:
`Assert` `Op`."
9859,assert_is_batch_matrix,tensorflow/tensorflow/python/ops/linalg/linear_operator_util.py,248,function,Static assert that `tensor` has rank `2` or higher.
9860,shape_tensor,tensorflow/tensorflow/python/ops/linalg/linear_operator_util.py,257,function,"Convert Tensor using default type, unless empty list or tuple."
9861,broadcast_matrix_batch_dims,tensorflow/tensorflow/python/ops/linalg/linear_operator_util.py,273,function,"Broadcast leading dimensions of zero or more [batch] matrices.
Example broadcasting one batch dim of two simple matrices.
```python
x = [[1, 2],
[3, 4]] # Shape [2, 2], no batch dims
y = [[[1]]] # Shape [1, 1, 1], 1 batch dim of shape [1]
x_bc, y_bc = broadcast_matrix_batch_dims([x, y])
x_bc
==> [[[1, 2],
[3, 4]]] # Shape [1, 2, 2], 1 batch dim of shape [1].
y_bc
==> same as y
```
Example broadcasting many batch dims
```python
x = tf.random.normal(shape=(2, 3, 1, 4, 4))
y = tf.random.normal(shape=(1, 3, 2, 5, 5))
x_bc, y_bc = broadcast_matrix_batch_dims([x, y])
x_bc.shape
==> (2, 3, 2, 4, 4)
y_bc.shape
==> (2, 3, 2, 5, 5)
```
Args:
batch_matrices: Iterable of `Tensor`s, each having two or more dimensions.
name: A string name to prepend to created ops.
Returns:
bcast_matrices: List of `Tensor`s, with `bcast_matrices[i]` containing
the values from `batch_matrices[i]`, with possibly broadcast batch dims.
Raises:
ValueError: If any input `Tensor` is statically determined to have less
than two dimensions."
9862,matrix_solve_with_broadcast,tensorflow/tensorflow/python/ops/linalg/linear_operator_util.py,366,function,Solve systems of linear equations.
9863,use_operator_or_provided_hint_unless_contradicting,tensorflow/tensorflow/python/ops/linalg/linear_operator_util.py,484,function,"Get combined hint in the case where operator.hint should equal hint.
Args:
operator: LinearOperator that a meta-operator was initialized with.
hint_attr_name: String name for the attribute.
provided_hint_value: Bool or None. Value passed by user in initialization.
message: Error message to print if hints contradict.
Returns:
True, False, or None.
Raises:
ValueError: If hints contradict."
9864,arg_is_blockwise,tensorflow/tensorflow/python/ops/linalg/linear_operator_util.py,519,function,Detect if input should be interpreted as a list of blocks.
9865,split_arg_into_blocks,tensorflow/tensorflow/python/ops/linalg/linear_operator_util.py,566,function,"Split `x` into blocks matching `operators`'s `domain_dimension`.
Specifically, if we have a blockwise lower-triangular matrix, with block
sizes along the diagonal `[M_j, M_j] j = 0,1,2..J`, this method splits `arg`
on `axis` into `J` tensors, whose shape at `axis` is `M_j`.
Args:
block_dims: Iterable of `TensorShapes`.
block_dims_fn: Callable returning an iterable of `Tensor`s.
arg: `Tensor`. `arg` is split into `J` tensors.
axis: Python `Integer` representing the axis to split `arg` on.
Returns:
A list of `Tensor`s."
9866,LinearOperatorZeros,tensorflow/tensorflow/python/ops/linalg/linear_operator_zeros.py,43,class,"`LinearOperator` acting like a [batch] zero matrix.
This operator acts like a [batch] zero matrix `A` with shape
`[B1,...,Bb, N, M]` for some `b >= 0`. The first `b` indices index a
batch member. For every batch index `(i1,...,ib)`, `A[i1,...,ib, : :]` is
an `N x M` matrix. This matrix `A` is not materialized, but for
purposes of broadcasting this shape will be relevant.
`LinearOperatorZeros` is initialized with `num_rows`, and optionally
`num_columns, `batch_shape`, and `dtype` arguments. If `num_columns` is
`None`, then this operator will be initialized as a square matrix. If
`batch_shape` is `None`, this operator efficiently passes through all
arguments. If `batch_shape` is provided, broadcasting may occur, which will
require making copies.
```python
# Create a 2 x 2 zero matrix.
operator = LinearOperatorZero(num_rows=2, dtype=tf.float32)
operator.to_dense()
==> [[0., 0.]
[0., 0.]]
operator.shape
==> [2, 2]
operator.determinant()
==> 0.
x = ... Shape [2, 4] Tensor
operator.matmul(x)
==> Shape [2, 4] Tensor, same as x.
# Create a 2-batch of 2x2 zero matrices
operator = LinearOperatorZeros(num_rows=2, batch_shape=[2])
operator.to_dense()
==> [[[0., 0.]
[0., 0.]],
[[0., 0.]
[0., 0.]]]
# Here, even though the operator has a batch shape, the input is the same as
# the output, so x can be passed through without a copy. The operator is able
# to detect that no broadcast is necessary because both x and the operator
# have statically defined shape.
x = ... Shape [2, 2, 3]
operator.matmul(x)
==> Shape [2, 2, 3] Tensor, same as tf.zeros_like(x)
# Here the operator and x have different batch_shape, and are broadcast.
# This requires a copy, since the output is different size than the input.
x = ... Shape [1, 2, 3]
operator.matmul(x)
==> Shape [2, 2, 3] Tensor, equal to tf.zeros_like([x, x])
```
### Shape compatibility
This operator acts on [batch] matrix with compatible shape.
`x` is a batch matrix with compatible shape for `matmul` and `solve` if
```
operator.shape = [B1,...,Bb] + [N, M], with b >= 0
x.shape = [C1,...,Cc] + [M, R],
and [C1,...,Cc] broadcasts with [B1,...,Bb] to [D1,...,Dd]
```
#### Matrix property hints
This `LinearOperator` is initialized with boolean flags of the form `is_X`,
for `X = non_singular, self_adjoint, positive_definite, square`.
These have the following meaning:
* If `is_X == True`, callers should expect the operator to have the
property `X`. This is a promise that should be fulfilled, but is *not* a
runtime assert. For example, finite floating point precision may result
in these promises being violated.
* If `is_X == False`, callers should expect the operator to not have `X`.
* If `is_X == None` (the default), callers should have no expectation either
way."
9867,add_to_tensor,tensorflow/tensorflow/python/ops/linalg/linear_operator_zeros.py,338,method,"Add matrix represented by this operator to `mat`. Equiv to `I + mat`.
Args:
mat: `Tensor` with same `dtype` and shape broadcastable to `self`.
name: A name to give this `Op`.
Returns:
A `Tensor` with broadcast shape and same `dtype` as `self`."
9868,combined_commuting_self_adjoint_hint,tensorflow/tensorflow/python/ops/linalg/registrations_util.py,23,function,Get combined hint for self-adjoint-ness.
9869,is_square,tensorflow/tensorflow/python/ops/linalg/registrations_util.py,47,function,Return a hint to whether the composition is square.
9870,combined_commuting_positive_definite_hint,tensorflow/tensorflow/python/ops/linalg/registrations_util.py,68,function,Get combined PD hint for compositions.
9871,combined_non_singular_hint,tensorflow/tensorflow/python/ops/linalg/registrations_util.py,81,function,Get combined hint for when .
9872,conjugate_gradient,tensorflow/tensorflow/python/ops/linalg/sparse/conjugate_gradient.py,36,function,"Conjugate gradient solver.
Solves a linear system of equations `A*x = rhs` for self-adjoint, positive
definite matrix `A` and right-hand side vector `rhs`, using an iterative,
matrix-free algorithm where the action of the matrix A is represented by
`operator`. The iteration terminates when either the number of iterations
exceeds `max_iter` or when the residual norm has been reduced to `tol`
times its initial value, i.e. \\(||rhs - A x_k|| <= tol ||rhs||\\).
Args:
operator: A `LinearOperator` that is self-adjoint and positive definite.
rhs: A possibly batched vector of shape `[..., N]` containing the right-hand
size vector.
preconditioner: A `LinearOperator` that approximates the inverse of `A`.
An efficient preconditioner could dramatically improve the rate of
convergence. If `preconditioner` represents matrix `M`(`M` approximates
`A^{-1}`), the algorithm uses `preconditioner.apply(x)` to estimate
`A^{-1}x`. For this to be useful, the cost of applying `M` should be
much lower than computing `A^{-1}` directly.
x: A possibly batched vector of shape `[..., N]` containing the initial
guess for the solution.
tol: A float scalar convergence tolerance.
max_iter: An integer giving the maximum number of iterations.
name: A name scope for the operation.
Returns:
output: A namedtuple representing the final state with fields:
- i: A scalar `int32` `Tensor`. Number of iterations executed.
- x: A rank-1 `Tensor` of shape `[..., N]` containing the computed
solution.
- r: A rank-1 `Tensor` of shape `[.., M]` containing the residual vector.
- p: A rank-1 `Tensor` of shape `[..., N]`. `A`-conjugate basis vector.
- gamma: \\(r \dot M \dot r\\), equivalent to \\(||r||_2^2\\) when
`preconditioner=None`."
9873,DenseShapeAndType,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,50,class,
9874,get_shape_and_type,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,78,function,Return matrix's shape and type if available.
9875,dense_shape_and_type,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,90,function,"Get dense shape and dtype of the tf.Tensor containing the matrix.
Args:
matrix: A `tf.Tensor` of type `tf.variant` storing a sparse matrix.
Returns:
An instance of `ShapeAndType` with properties `shape` (a `tf.TensorShape`)
and `dtype` (a `tf.DType`).
Raises:
TypeError: if `matrix` is not a tensor or its dtype is not variant.
ValueError: if `matrix` lacks static handle data containing the dense
shape and dtype."
9876,matmul_shape_inference,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,121,function,Helper function for matmul to set the result matrix's handle data.
9877,matmul,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,147,function,"Perform a sparse matrix matmul between `a` and `b`.
Performs a contraction between `a` and `b` along the two innermost dimensions.
If both `a` and `b` are instances of `SparseMatrix`, returns a new instance
of `SparseMatrix` (same type as `a`). If one is not an instance of
`SparseMatrix`, returns a dense `Tensor`:
```
c = opA(a) . opB(b)
```
where `opA` (resp. `opB`) is the transpose or hermitian transpose depending
on the values of `transpose_a` (resp. `transpose_b`) and `adjoint_a`
(resp. `adjoint_b`).
Args:
a: `Tensor` or `SparseMatrix`, having rank `2` or `3`.
b: `Tensor` or `SparseMatrix`, having rank `2` or `3`.
transpose_a: Python `bool`.
transpose_b: Python `bool`.
adjoint_a: Python `bool`.
adjoint_b: Python `bool`.
name: Optional name to use when creating ops.
Returns:
A `SparseMatrix` if both `a` and `b` are instances of `SparseMatrix`,
otherwise a dense `Tensor`."
9878,SparseMatrix,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,248,class,Abstract class for sparse matrix types.
9879,to_dense,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,264,method,
9880,to_sparse_tensor,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,268,method,
9881,graph,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,272,method,
9882,shape,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,276,method,
9883,dtype,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,280,method,
9884,eager_handle_data,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,284,method,Return the matrix's handle data iff in eager mode.
9885,conj,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,288,method,
9886,hermitian_transpose,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,292,method,Return the hermitian transpose of the matrix.
9887,nnz,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,299,method,"Number of stored values, including explicit zeros."
9888,sorted_indices,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,305,method,
9889,transpose,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,309,method,
9890,CSRSparseMatrix,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,315,class,(Optionally batched) CSR Sparse Matrix.
9891,to_dense,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,372,method,
9892,to_sparse_tensor,tensorflow/tensorflow/python/ops/linalg/sparse/sparse_csr_matrix_ops.py,375,method,
9893,ReductionV2,tensorflow/tensorflow/python/ops/losses/loss_reduction.py,21,class,"Types of loss reduction.
Contains the following values:
* `AUTO`: Indicates that the reduction option will be determined by the usage
context. For almost all cases this defaults to `SUM_OVER_BATCH_SIZE`. When
used with `tf.distribute.Strategy`, outside of built-in training loops such
as `tf.keras` `compile` and `fit`, we expect reduction value to be
`SUM` or `NONE`. Using `AUTO` in that case will raise an error.
* `NONE`: Weighted losses with one dimension reduced (axis=-1, or axis
specified by loss function). When this reduction type used with built-in
Keras training loops like `fit`/`evaluate`, the unreduced vector loss is
passed to the optimizer but the reported loss will be a scalar value.
* `SUM`: Scalar sum of weighted losses.
* `SUM_OVER_BATCH_SIZE`: Scalar `SUM` divided by number of elements in losses.
This reduction type is not supported when used with
`tf.distribute.Strategy` outside of built-in training loops like `tf.keras`
`compile`/`fit`.
You can implement 'SUM_OVER_BATCH_SIZE' using global batch size like:
```
with strategy.scope():
loss_obj = tf.keras.losses.CategoricalCrossentropy(
reduction=tf.keras.losses.Reduction.NONE)
....
loss = tf.reduce_sum(loss_object(labels, predictions)) *
(1. / global_batch_size)
```
Please see the
[custom training guide](https://www.tensorflow.org/tutorials/distribute/custom_training) # pylint: disable=line-too-long
for more details on this."
9894,all,tensorflow/tensorflow/python/ops/losses/loss_reduction.py,62,method,
9895,validate,tensorflow/tensorflow/python/ops/losses/loss_reduction.py,66,method,
9896,Reduction,tensorflow/tensorflow/python/ops/losses/losses_impl.py,39,class,"Types of loss reduction.
Contains the following values:
* `NONE`: Un-reduced weighted losses with the same shape as input.
* `SUM`: Scalar sum of weighted losses.
* `MEAN`: Scalar `SUM` divided by sum of weights. DEPRECATED.
* `SUM_OVER_BATCH_SIZE`: Scalar `SUM` divided by number of elements in losses.
* `SUM_OVER_NONZERO_WEIGHTS`: Scalar `SUM` divided by number of non-zero
weights. DEPRECATED.
* `SUM_BY_NONZERO_WEIGHTS`: Same as `SUM_OVER_NONZERO_WEIGHTS`. DEPRECATED."
9897,all,tensorflow/tensorflow/python/ops/losses/losses_impl.py,61,method,
9898,validate,tensorflow/tensorflow/python/ops/losses/losses_impl.py,71,method,
9899,compute_weighted_loss,tensorflow/tensorflow/python/ops/losses/losses_impl.py,141,function,"Computes the weighted loss.
Args:
losses: `Tensor` of shape `[batch_size, d1, ... dN]`.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`losses`, and must be broadcastable to `losses` (i.e., all dimensions must
be either `1`, or the same as the corresponding `losses` dimension).
scope: the scope for the operations performed in computing the loss.
loss_collection: the loss will be added to these collections.
reduction: Type of reduction to apply to loss.
Returns:
Weighted loss `Tensor` of the same type as `losses`. If `reduction` is
`NONE`, this has the same shape as `losses`; otherwise, it is scalar.
Raises:
ValueError: If `weights` is `None` or the shape is not compatible with
`losses`, or if the number of dimensions (rank) of either `losses` or
`weights` is missing.
Note:
When calculating the gradient of a weighted loss contributions from
both `losses` and `weights` are considered. If your `weights` depend
on some model parameters but you do not want this to affect the loss
gradient, you need to apply `tf.stop_gradient` to `weights` before
passing them to `compute_weighted_loss`.
@compatibility(eager)
The `loss_collection` argument is ignored when executing eagerly. Consider
holding on to the return value or collecting losses via a `tf.keras.Model`.
@end_compatibility"
9900,absolute_difference,tensorflow/tensorflow/python/ops/losses/losses_impl.py,210,function,"Adds an Absolute Difference loss to the training procedure.
`weights` acts as a coefficient for the loss. If a scalar is provided, then
the loss is simply scaled by the given value. If `weights` is a `Tensor` of
shape `[batch_size]`, then the total loss for each sample of the batch is
rescaled by the corresponding element in the `weights` vector. If the shape of
`weights` matches the shape of `predictions`, then the loss of each
measurable element of `predictions` is scaled by the corresponding value of
`weights`.
Args:
labels: The ground truth output tensor, same dimensions as 'predictions'.
predictions: The predicted outputs.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `losses` dimension).
scope: The scope for the operations performed in computing the loss.
loss_collection: collection to which this loss will be added.
reduction: Type of reduction to apply to loss.
Returns:
Weighted loss float `Tensor`. If `reduction` is `NONE`, this has the same
shape as `labels`; otherwise, it is scalar.
Raises:
ValueError: If the shape of `predictions` doesn't match that of
`labels` or if the shape of `weights` is invalid or if `labels`
or `predictions` is None.
@compatibility(eager)
The `loss_collection` argument is ignored when executing eagerly. Consider
holding on to the return value or collecting losses via a `tf.keras.Model`.
@end_compatibility"
9901,cosine_distance,tensorflow/tensorflow/python/ops/losses/losses_impl.py,265,function,"Adds a cosine-distance loss to the training procedure.
Note that the function assumes that `predictions` and `labels` are already
unit-normalized.
Args:
labels: `Tensor` whose shape matches 'predictions'
predictions: An arbitrary matrix.
axis: The dimension along which the cosine distance is computed.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `losses` dimension).
scope: The scope for the operations performed in computing the loss.
loss_collection: collection to which this loss will be added.
reduction: Type of reduction to apply to loss.
dim: The old (deprecated) name for `axis`.
Returns:
Weighted loss float `Tensor`. If `reduction` is `NONE`, this has the same
shape as `labels`; otherwise, it is scalar.
Raises:
ValueError: If `predictions` shape doesn't match `labels` shape, or
`axis`, `labels`, `predictions` or `weights` is `None`.
@compatibility(eager)
The `loss_collection` argument is ignored when executing eagerly. Consider
holding on to the return value or collecting losses via a `tf.keras.Model`.
@end_compatibility"
9902,hinge_loss,tensorflow/tensorflow/python/ops/losses/losses_impl.py,321,function,"Adds a hinge loss to the training procedure.
Args:
labels: The ground truth output tensor. Its shape should match the shape of
logits. The values of the tensor are expected to be 0.0 or 1.0. Internally
the {0,1} labels are converted to {-1,1} when calculating the hinge loss.
logits: The logits, a float tensor. Note that logits are assumed to be
unbounded and 0-centered. A value > 0 (resp. < 0) is considered a positive
(resp. negative) binary prediction.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `losses` dimension).
scope: The scope for the operations performed in computing the loss.
loss_collection: collection to which the loss will be added.
reduction: Type of reduction to apply to loss.
Returns:
Weighted loss float `Tensor`. If `reduction` is `NONE`, this has the same
shape as `labels`; otherwise, it is scalar.
Raises:
ValueError: If the shapes of `logits` and `labels` don't match or
if `labels` or `logits` is None.
@compatibility(eager)
The `loss_collection` argument is ignored when executing eagerly. Consider
holding on to the return value or collecting losses via a `tf.keras.Model`.
@end_compatibility"
9903,huber_loss,tensorflow/tensorflow/python/ops/losses/losses_impl.py,372,function,"Adds a [Huber Loss](https://en.wikipedia.org/wiki/Huber_loss) term to the training procedure.
For each value x in `error=labels-predictions`, the following is calculated:
```
0.5 * x^2 if |x| <= d
0.5 * d^2 + d * (|x| - d) if |x| > d
```
where d is `delta`.
`weights` acts as a coefficient for the loss. If a scalar is provided, then
the loss is simply scaled by the given value. If `weights` is a tensor of size
`[batch_size]`, then the total loss for each sample of the batch is rescaled
by the corresponding element in the `weights` vector. If the shape of
`weights` matches the shape of `predictions`, then the loss of each
measurable element of `predictions` is scaled by the corresponding value of
`weights`.
Args:
labels: The ground truth output tensor, same dimensions as 'predictions'.
predictions: The predicted outputs.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `losses` dimension).
delta: `float`, the point where the huber loss function changes from a
quadratic to linear.
scope: The scope for the operations performed in computing the loss.
loss_collection: collection to which the loss will be added.
reduction: Type of reduction to apply to loss.
Returns:
Weighted loss float `Tensor`. If `reduction` is `NONE`, this has the same
shape as `labels`; otherwise, it is scalar.
Raises:
ValueError: If the shape of `predictions` doesn't match that of `labels` or
if the shape of `weights` is invalid. Also if `labels` or
`predictions` is None.
@compatibility(eager)
The `loss_collection` argument is ignored when executing eagerly. Consider
holding on to the return value or collecting losses via a `tf.keras.Model`.
@end_compatibility"
9904,log_loss,tensorflow/tensorflow/python/ops/losses/losses_impl.py,449,function,"Adds a Log Loss term to the training procedure.
`weights` acts as a coefficient for the loss. If a scalar is provided, then
the loss is simply scaled by the given value. If `weights` is a tensor of size
`[batch_size]`, then the total loss for each sample of the batch is rescaled
by the corresponding element in the `weights` vector. If the shape of
`weights` matches the shape of `predictions`, then the loss of each
measurable element of `predictions` is scaled by the corresponding value of
`weights`.
Args:
labels: The ground truth output tensor, same dimensions as 'predictions'.
predictions: The predicted outputs.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `losses` dimension).
epsilon: A small increment to add to avoid taking a log of zero.
scope: The scope for the operations performed in computing the loss.
loss_collection: collection to which the loss will be added.
reduction: Type of reduction to apply to loss.
Returns:
Weighted loss float `Tensor`. If `reduction` is `NONE`, this has the same
shape as `labels`; otherwise, it is scalar.
Raises:
ValueError: If the shape of `predictions` doesn't match that of `labels` or
if the shape of `weights` is invalid. Also if `labels` or `predictions`
is None.
@compatibility(eager)
The `loss_collection` argument is ignored when executing eagerly. Consider
holding on to the return value or collecting losses via a `tf.keras.Model`.
@end_compatibility"
9905,mean_pairwise_squared_error,tensorflow/tensorflow/python/ops/losses/losses_impl.py,507,function,"Adds a pairwise-errors-squared loss to the training procedure.
Unlike `mean_squared_error`, which is a measure of the differences between
corresponding elements of `predictions` and `labels`,
`mean_pairwise_squared_error` is a measure of the differences between pairs of
corresponding elements of `predictions` and `labels`.
For example, if `labels`=[a, b, c] and `predictions`=[x, y, z], there are
three pairs of differences are summed to compute the loss:
loss = [ ((a-b) - (x-y)).^2 + ((a-c) - (x-z)).^2 + ((b-c) - (y-z)).^2 ] / 3
Note that since the inputs are of shape `[batch_size, d0, ... dN]`, the
corresponding pairs are computed within each batch sample but not across
samples within a batch. For example, if `predictions` represents a batch of
16 grayscale images of dimension [batch_size, 100, 200], then the set of pairs
is drawn from each image, but not across images.
`weights` acts as a coefficient for the loss. If a scalar is provided, then
the loss is simply scaled by the given value. If `weights` is a tensor of size
`[batch_size]`, then the total loss for each sample of the batch is rescaled
by the corresponding element in the `weights` vector.
Args:
labels: The ground truth output tensor, whose shape must match the shape of
`predictions`.
predictions: The predicted outputs, a tensor of size
`[batch_size, d0, .. dN]` where N+1 is the total number of dimensions in
`predictions`.
weights: Coefficients for the loss a scalar, a tensor of shape
`[batch_size]` or a tensor whose shape matches `predictions`.
scope: The scope for the operations performed in computing the loss.
loss_collection: collection to which the loss will be added.
Returns:
A scalar `Tensor` that returns the weighted loss.
Raises:
ValueError: If the shape of `predictions` doesn't match that of `labels` or
if the shape of `weights` is invalid. Also if `labels` or `predictions`
is None.
@compatibility(eager)
The `loss_collection` argument is ignored when executing eagerly. Consider
holding on to the return value or collecting losses via a `tf.keras.Model`.
@end_compatibility"
9906,mean_squared_error,tensorflow/tensorflow/python/ops/losses/losses_impl.py,604,function,"Adds a Sum-of-Squares loss to the training procedure.
`weights` acts as a coefficient for the loss. If a scalar is provided, then
the loss is simply scaled by the given value. If `weights` is a tensor of size
`[batch_size]`, then the total loss for each sample of the batch is rescaled
by the corresponding element in the `weights` vector. If the shape of
`weights` matches the shape of `predictions`, then the loss of each
measurable element of `predictions` is scaled by the corresponding value of
`weights`.
Args:
labels: The ground truth output tensor, same dimensions as 'predictions'.
predictions: The predicted outputs.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`labels`, and must be broadcastable to `labels` (i.e., all dimensions must
be either `1`, or the same as the corresponding `losses` dimension).
scope: The scope for the operations performed in computing the loss.
loss_collection: collection to which the loss will be added.
reduction: Type of reduction to apply to loss.
Returns:
Weighted loss float `Tensor`. If `reduction` is `NONE`, this has the same
shape as `labels`; otherwise, it is scalar.
Raises:
ValueError: If the shape of `predictions` doesn't match that of `labels` or
if the shape of `weights` is invalid. Also if `labels` or `predictions`
is None.
@compatibility(eager)
The `loss_collection` argument is ignored when executing eagerly. Consider
holding on to the return value or collecting losses via a `tf.keras.Model`.
@end_compatibility"
9907,sigmoid_cross_entropy,tensorflow/tensorflow/python/ops/losses/losses_impl.py,658,function,"Creates a cross-entropy loss using tf.nn.sigmoid_cross_entropy_with_logits.
`weights` acts as a coefficient for the loss. If a scalar is provided,
then the loss is simply scaled by the given value. If `weights` is a
tensor of shape `[batch_size]`, then the loss weights apply to each
corresponding sample.
If `label_smoothing` is nonzero, smooth the labels towards 1/2:
new_multiclass_labels = multiclass_labels * (1 - label_smoothing)
+ 0.5 * label_smoothing
Args:
multi_class_labels: `[batch_size, num_classes]` target integer labels in
`{0, 1}`.
logits: Float `[batch_size, num_classes]` logits outputs of the network.
weights: Optional `Tensor` whose rank is either 0, or the same rank as
`multi_class_labels`, and must be broadcastable to `multi_class_labels`
(i.e., all dimensions must be either `1`, or the same as the
corresponding `losses` dimension).
label_smoothing: If greater than `0` then smooth the labels.
scope: The scope for the operations performed in computing the loss.
loss_collection: collection to which the loss will be added.
reduction: Type of reduction to apply to loss.
Returns:
Weighted loss `Tensor` of the same type as `logits`. If `reduction` is
`NONE`, this has the same shape as `logits`; otherwise, it is scalar.
Raises:
ValueError: If the shape of `logits` doesn't match that of
`multi_class_labels` or if the shape of `weights` is invalid, or if
`weights` is None. Also if `multi_class_labels` or `logits` is None.
@compatibility(eager)
The `loss_collection` argument is ignored when executing eagerly. Consider
holding on to the return value or collecting losses via a `tf.keras.Model`.
@end_compatibility"
9908,softmax_cross_entropy,tensorflow/tensorflow/python/ops/losses/losses_impl.py,724,function,"Creates a cross-entropy loss using tf.nn.softmax_cross_entropy_with_logits_v2.
`weights` acts as a coefficient for the loss. If a scalar is provided,
then the loss is simply scaled by the given value. If `weights` is a
tensor of shape `[batch_size]`, then the loss weights apply to each
corresponding sample.
If `label_smoothing` is nonzero, smooth the labels towards 1/num_classes:
new_onehot_labels = onehot_labels * (1 - label_smoothing)
+ label_smoothing / num_classes
Note that `onehot_labels` and `logits` must have the same shape,
e.g. `[batch_size, num_classes]`. The shape of `weights` must be
broadcastable to loss, whose shape is decided by the shape of `logits`.
In case the shape of `logits` is `[batch_size, num_classes]`, loss is
a `Tensor` of shape `[batch_size]`.
Args:
onehot_labels: One-hot-encoded labels.
logits: Logits outputs of the network.
weights: Optional `Tensor` that is broadcastable to loss.
label_smoothing: If greater than 0 then smooth the labels.
scope: the scope for the operations performed in computing the loss.
loss_collection: collection to which the loss will be added.
reduction: Type of reduction to apply to loss.
Returns:
Weighted loss `Tensor` of the same type as `logits`. If `reduction` is
`NONE`, this has shape `[batch_size]`; otherwise, it is scalar.
Raises:
ValueError: If the shape of `logits` doesn't match that of `onehot_labels`
or if the shape of `weights` is invalid or if `weights` is None. Also if
`onehot_labels` or `logits` is None.
@compatibility(eager)
The `loss_collection` argument is ignored when executing eagerly. Consider
holding on to the return value or collecting losses via a `tf.keras.Model`.
@end_compatibility"
9909,sparse_softmax_cross_entropy,tensorflow/tensorflow/python/ops/losses/losses_impl.py,847,function,"Cross-entropy loss using `tf.nn.sparse_softmax_cross_entropy_with_logits`.
`weights` acts as a coefficient for the loss. If a scalar is provided,
then the loss is simply scaled by the given value. If `weights` is a
tensor of shape `[batch_size]`, then the loss weights apply to each
corresponding sample.
Args:
labels: `Tensor` of shape `[d_0, d_1, ..., d_{r-1}]` (where `r` is rank of
`labels` and result) and dtype `int32` or `int64`. Each entry in `labels`
must be an index in `[0, num_classes)`. Other values will raise an
exception when this op is run on CPU, and return `NaN` for corresponding
loss and gradient rows on GPU.
logits: Unscaled log probabilities of shape
`[d_0, d_1, ..., d_{r-1}, num_classes]` and dtype `float16`, `float32` or
`float64`.
weights: Coefficients for the loss. This must be scalar or broadcastable to
`labels` (i.e. same rank and each dimension is either 1 or the same).
scope: the scope for the operations performed in computing the loss.
loss_collection: collection to which the loss will be added.
reduction: Type of reduction to apply to loss.
Returns:
Weighted loss `Tensor` of the same type as `logits`. If `reduction` is
`NONE`, this has the same shape as `labels`; otherwise, it is scalar.
Raises:
ValueError: If the shapes of `logits`, `labels`, and `weights` are
incompatible, or if any of them are None.
@compatibility(eager)
The `loss_collection` argument is ignored when executing eagerly. Consider
holding on to the return value or collecting losses via a `tf.keras.Model`.
@end_compatibility"
9910,squeeze_or_expand_dimensions,tensorflow/tensorflow/python/ops/losses/util.py,34,function,"Squeeze or expand last dimension if needed.
1. Squeezes last dim of `y_pred` or `y_true` if their rank differs by 1
(using `confusion_matrix.remove_squeezable_dimensions`).
2. Squeezes or expands last dim of `sample_weight` if its rank differs by 1
from the new rank of `y_pred`.
If `sample_weight` is scalar, it is kept scalar.
This will use static shape if available. Otherwise, it will add graph
operations, which could result in a performance hit.
Args:
y_pred: Predicted values, a `Tensor` of arbitrary dimensions.
y_true: Optional label `Tensor` whose dimensions match `y_pred`.
sample_weight: Optional weight scalar or `Tensor` whose dimensions match
`y_pred`.
Returns:
Tuple of `y_pred`, `y_true` and `sample_weight`. Each of them possibly has
the last dimension squeezed,
`sample_weight` could be extended by one dimension.
If `sample_weight` is None, (y_pred, y_true) is returned."
9911,scale_losses_by_sample_weight,tensorflow/tensorflow/python/ops/losses/util.py,123,function,"Scales loss values by the given sample weights.
`sample_weight` dimensions are updated to match with the dimension of `losses`
if possible by using squeeze/expand/broadcast.
Args:
losses: Loss tensor.
sample_weight: Sample weights tensor.
Returns:
`losses` scaled by `sample_weight` with dtype float32."
9912,check_per_example_loss_rank,tensorflow/tensorflow/python/ops/losses/util.py,148,function,"Context manager that checks that the rank of per_example_loss is atleast 1.
Args:
per_example_loss: Per example loss tensor.
Yields:
A context manager."
9913,add_loss,tensorflow/tensorflow/python/ops/losses/util.py,178,function,"Adds a externally defined loss to the collection of losses.
Args:
loss: A loss `Tensor`.
loss_collection: Optional collection to add the loss to."
9914,get_losses,tensorflow/tensorflow/python/ops/losses/util.py,193,function,"Gets the list of losses from the loss_collection.
Args:
scope: An optional scope name for filtering the losses to return.
loss_collection: Optional losses collection.
Returns:
a list of loss tensors."
9915,get_regularization_losses,tensorflow/tensorflow/python/ops/losses/util.py,207,function,"Gets the list of regularization losses.
Args:
scope: An optional scope name for filtering the losses to return.
Returns:
A list of regularization losses as Tensors."
9916,get_regularization_loss,tensorflow/tensorflow/python/ops/losses/util.py,220,function,"Gets the total regularization loss.
Args:
scope: An optional scope name for filtering the losses to return.
name: The name of the returned tensor.
Returns:
A scalar regularization loss."
9917,get_total_loss,tensorflow/tensorflow/python/ops/losses/util.py,238,function,"Returns a tensor whose value represents the total loss.
In particular, this adds any losses you have added with `tf.add_loss()` to
any regularization losses that have been added by regularization parameters
on layers constructors e.g. `tf.layers`. Be very sure to use this if you
are constructing a loss_op manually. Otherwise regularization arguments
on `tf.layers` methods will not function.
Args:
add_regularization_losses: A boolean indicating whether or not to use the
regularization losses in the sum.
name: The name of the returned tensor.
scope: An optional scope name for filtering the losses to return. Note that
this filters the losses added with `tf.add_loss()` as well as the
regularization losses to that scope.
Returns:
A `Tensor` whose value represents the total loss.
Raises:
ValueError: if `losses` is not iterable."
9918,max,tensorflow/tensorflow/python/ops/numpy_ops/__init__.py,185,function,
9919,min,tensorflow/tensorflow/python/ops/numpy_ops/__init__.py,190,function,
9920,round,tensorflow/tensorflow/python/ops/numpy_ops/__init__.py,195,function,
9921,empty,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,49,function,
9922,empty_like,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,54,function,
9923,zeros,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,59,function,
9924,zeros_like,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,68,function,
9925,ones,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,83,function,
9926,ones_like,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,92,function,
9927,eye,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,103,function,
9928,identity,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,135,function,
9929,full,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,140,function,
9930,full_like,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,152,function,"order, subok and shape arguments mustn't be changed."
9931,array,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,225,function,"Since Tensors are immutable, a copy is made only if val is placed on a
different device than the current one. Even if `copy` is False, a new Tensor
may need to be built to satisfy `dtype` and `ndim`. This is used only if `val`
is an ndarray or a Tensor."
9932,asarray,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,241,function,
9933,asanyarray,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,250,function,
9934,ascontiguousarray,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,255,function,
9935,arange,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,261,function,"Returns `step`-separated values in the range [start, stop).
Args:
start: Start of the interval. Included in the range.
stop: End of the interval. If not specified, `start` is treated as 0 and
`start` value is used as `stop`. If specified, it is not included in the
range if `step` is integer. When `step` is floating point, it may or may
not be included.
step: The difference between 2 consecutive values in the output range. It is
recommended to use `linspace` instead of using non-integer values for
`step`.
dtype: Optional. Type of the resulting ndarray. Could be a python type, a
NumPy type or a TensorFlow `DType`. If not provided, the largest type of
`start`, `stop`, `step` is used.
Raises:
ValueError: If step is zero."
9936,diag,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,303,function,Raises an error if input is not 1- or 2-d.
9937,diagonal,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,338,function,
9938,diagflat,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,369,function,
9939,all,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,393,function,
9940,any,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,400,function,
9941,compress,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,407,function,
9942,copy,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,438,function,
9943,cumprod,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,454,function,
9944,cumsum,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,470,function,
9945,imag,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,486,function,
9946,sum,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,566,function,
9947,prod,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,577,function,
9948,mean,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,588,function,
9949,amax,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,599,function,
9950,amin,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,612,function,
9951,var,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,625,function,
9952,std,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,672,function,
9953,ravel,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,683,function,
9954,real,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,695,function,
9955,repeat,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,703,function,
9956,around,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,734,function,
9957,reshape,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,756,function,order argument can only b 'C' or 'F'.
9958,expand_dims,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,788,function,
9959,squeeze,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,794,function,
9960,transpose,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,800,function,
9961,swapaxes,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,808,function,
9962,moveaxis,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,824,function,"Raises ValueError if source, destination not in (-ndim(a), ndim(a))."
9963,pad,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,888,function,"Only supports modes 'constant', 'reflect' and 'symmetric' currently."
9964,take,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,905,function,"out argument is not supported, and default mode is clip."
9965,where,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,932,function,Raises ValueError if exactly one of x or y is not None.
9966,select,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,945,function,
9967,shape,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,962,function,
9968,ndim,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,968,function,
9969,isscalar,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,974,function,
9970,split,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1009,function,
9971,broadcast_to,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1032,function,
9972,stack,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1037,function,
9973,hstack,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1052,function,
9974,vstack,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1066,function,
9975,dstack,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1076,function,
9976,atleast_1d,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1123,function,
9977,atleast_2d,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1128,function,
9978,atleast_3d,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1133,function,
9979,nonzero,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1150,function,
9980,diag_indices,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1164,function,
9981,tri,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1177,function,
9982,tril,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1204,function,
9983,triu,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1226,function,
9984,flip,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1248,function,
9985,flipud,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1261,function,
9986,fliplr,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1266,function,
9987,roll,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1271,function,
9988,rot90,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1284,function,
9989,vander,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1304,function,
9990,ix_,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1343,function,
9991,broadcast_arrays,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1375,function,
9992,sign,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1388,function,
9993,take_along_axis,tensorflow/tensorflow/python/ops/numpy_ops/np_array_ops.py,1409,function,
9994,convert_to_tensor,tensorflow/tensorflow/python/ops/numpy_ops/np_arrays.py,37,function,"Wrapper over `tf.convert_to_tensor`.
Args:
value: value to convert
dtype: (optional) the type we would like it to be converted to.
dtype_hint: (optional) soft preference for the type we would like it to be
converted to. `tf.convert_to_tensor` will attempt to convert value to this
type first, but will not fail if conversion is not possible falling back
to inferring the type instead.
Returns:
Value converted to tf.Tensor."
9995,NdarraySpec,tensorflow/tensorflow/python/ops/numpy_ops/np_arrays.py,61,class,Type specification for a `tf.experiemntal.numpy.ndarray`.
9996,ndarray,tensorflow/tensorflow/python/ops/numpy_ops/np_arrays.py,99,class,"Equivalent of numpy.ndarray backed by TensorFlow tensors.
This does not support all features of NumPy ndarrays e.g. strides and
memory order since, unlike NumPy, the backing storage is not a raw memory
buffer.
TODO(srbs): Clearly specify which attributes and methods are not supported
or if there are any differences in behavior."
9997,from_tensor,tensorflow/tensorflow/python/ops/numpy_ops/np_arrays.py,165,method,
9998,data,tensorflow/tensorflow/python/ops/numpy_ops/np_arrays.py,182,method,"Tensor object containing the array data.
This has a few key differences from the Python buffer object used in
NumPy arrays.
1. Tensors are immutable. So operations requiring in-place edit, e.g.
__setitem__, are performed by replacing the underlying buffer with a new
one.
2. Tensors do not provide access to their raw buffer.
Returns:
A Tensor."
9999,shape,tensorflow/tensorflow/python/ops/numpy_ops/np_arrays.py,198,method,Returns a tuple or tf.Tensor of array dimensions.
10000,dtype,tensorflow/tensorflow/python/ops/numpy_ops/np_arrays.py,207,method,
10001,ndim,tensorflow/tensorflow/python/ops/numpy_ops/np_arrays.py,216,method,
10002,size,tensorflow/tensorflow/python/ops/numpy_ops/np_arrays.py,224,method,Returns the number of elements in the array.
10003,T,tensorflow/tensorflow/python/ops/numpy_ops/np_arrays.py,233,method,
10004,astype,tensorflow/tensorflow/python/ops/numpy_ops/np_arrays.py,245,method,
10005,tolist,tensorflow/tensorflow/python/ops/numpy_ops/np_arrays.py,316,method,
10006,tensor_to_ndarray,tensorflow/tensorflow/python/ops/numpy_ops/np_arrays.py,326,function,
10007,ndarray_to_tensor,tensorflow/tensorflow/python/ops/numpy_ops/np_arrays.py,330,function,
10008,is_allow_float64,tensorflow/tensorflow/python/ops/numpy_ops/np_dtypes.py,69,function,
10009,set_allow_float64,tensorflow/tensorflow/python/ops/numpy_ops/np_dtypes.py,73,function,
10010,canonicalize_dtype,tensorflow/tensorflow/python/ops/numpy_ops/np_dtypes.py,78,function,
10011,default_float_type,tensorflow/tensorflow/python/ops/numpy_ops/np_dtypes.py,104,function,"Gets the default float type.
Returns:
If `is_allow_float64()` is true, returns float64; otherwise returns float32."
10012,public_name,tensorflow/tensorflow/python/ops/numpy_ops/np_export.py,24,function,
10013,np_export,tensorflow/tensorflow/python/ops/numpy_ops/np_export.py,28,function,
10014,np_export_constant,tensorflow/tensorflow/python/ops/numpy_ops/np_export.py,32,function,
10015,make_numpy_compatible,tensorflow/tensorflow/python/ops/numpy_ops/np_logic_test.py,104,function,
10016,dot,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,54,function,
10017,add,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,81,function,
10018,subtract,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,93,function,
10019,multiply,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,98,function,
10020,true_divide,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,110,function,
10021,divide,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,134,function,
10022,floor_divide,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,139,function,
10023,mod,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,152,function,
10024,divmod,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,170,function,
10025,maximum,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,175,function,
10026,minimum,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,194,function,
10027,clip,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,206,function,
10028,matmul,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,221,function,
10029,tensordot,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,242,function,
10030,inner,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,247,function,
10031,cross,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,260,function,
10032,vdot,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,336,function,
10033,power,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,346,function,
10034,float_power,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,351,function,
10035,arctan2,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,356,function,
10036,nextafter,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,361,function,
10037,heaviside,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,366,function,
10038,hypot,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,380,function,
10039,kron,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,385,function,
10040,outer,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,416,function,
10041,logaddexp,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,426,function,
10042,logaddexp2,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,436,function,
10043,polyval,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,446,function,
10044,isclose,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,464,function,
10045,allclose,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,482,function,
10046,gcd,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,517,function,
10047,lcm,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,523,function,
10048,bitwise_and,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,554,function,
10049,bitwise_or,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,559,function,
10050,bitwise_xor,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,564,function,
10051,bitwise_not,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,569,function,
10052,log,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,601,function,
10053,exp,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,606,function,
10054,sqrt,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,611,function,
10055,abs,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,616,function,
10056,absolute,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,621,function,
10057,fabs,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,626,function,
10058,ceil,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,631,function,
10059,floor,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,636,function,
10060,conj,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,641,function,
10061,negative,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,646,function,
10062,reciprocal,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,651,function,
10063,signbit,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,656,function,
10064,sin,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,667,function,
10065,cos,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,672,function,
10066,tan,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,677,function,
10067,sinh,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,682,function,
10068,cosh,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,687,function,
10069,tanh,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,692,function,
10070,arcsin,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,697,function,
10071,arccos,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,702,function,
10072,arctan,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,707,function,
10073,arcsinh,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,712,function,
10074,arccosh,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,717,function,
10075,arctanh,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,722,function,
10076,deg2rad,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,727,function,
10077,rad2deg,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,736,function,
10078,angle,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,746,function,
10079,cbrt,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,762,function,
10080,conjugate,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,773,function,
10081,exp2,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,778,function,
10082,expm1,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,787,function,
10083,fix,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,792,function,
10084,iscomplex,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,801,function,
10085,isreal,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,806,function,
10086,iscomplexobj,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,811,function,
10087,isrealobj,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,817,function,
10088,isnan,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,822,function,
10089,nanmean,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,847,function,
10090,isfinite,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,860,function,
10091,isinf,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,865,function,
10092,isneginf,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,870,function,
10093,isposinf,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,875,function,
10094,log2,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,880,function,
10095,log10,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,885,function,
10096,log1p,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,890,function,
10097,positive,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,895,function,
10098,sinc,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,900,function,
10099,square,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,911,function,
10100,diff,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,916,function,
10101,equal,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,993,function,
10102,not_equal,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,998,function,
10103,greater,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1003,function,
10104,greater_equal,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1008,function,
10105,less,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1013,function,
10106,less_equal,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1018,function,
10107,array_equal,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1023,function,
10108,logical_and,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1046,function,
10109,logical_or,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1051,function,
10110,logical_xor,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1056,function,
10111,logical_not,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1061,function,
10112,linspace,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1076,function,
10113,logspace,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1115,function,
10114,geomspace,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1126,function,
10115,ptp,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1150,function,
10116,concatenate,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1156,function,
10117,tile,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1167,function,
10118,count_nonzero,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1184,function,
10119,argsort,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1190,function,
10120,sort,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1215,function,
10121,argmax,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1241,function,
10122,argmin,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1246,function,
10123,append,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1251,function,
10124,average,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1259,function,
10125,trace,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1322,function,
10126,meshgrid,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1340,function,This currently requires copy=True and sparse=False.
10127,einsum,tensorflow/tensorflow/python/ops/numpy_ops/np_math_ops.py,1364,function,
10128,seed,tensorflow/tensorflow/python/ops/numpy_ops/np_random.py,37,function,"Sets the seed for the random number generator.
Uses `tf.set_random_seed`.
Args:
s: an integer."
10129,randn,tensorflow/tensorflow/python/ops/numpy_ops/np_random.py,54,function,"Returns samples from a normal distribution.
Uses `tf.random_normal`.
Args:
*args: The shape of the output array.
Returns:
An ndarray with shape `args` and dtype `float64`."
10130,uniform,tensorflow/tensorflow/python/ops/numpy_ops/np_random.py,73,function,
10131,random,tensorflow/tensorflow/python/ops/numpy_ops/np_random.py,85,function,
10132,rand,tensorflow/tensorflow/python/ops/numpy_ops/np_random.py,90,function,
10133,randint,tensorflow/tensorflow/python/ops/numpy_ops/np_random.py,95,function,
10134,isscalar,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,90,function,Returns whether `val` is a scalar value or scalar Tensor.
10135,get_np_doc_form,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,231,function,"Gets the form of the original numpy docstrings.
Returns:
See `set_np_doc_form` for the list of valid values."
10136,set_np_doc_form,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,240,function,"Selects the form of the original numpy docstrings.
This function sets a global variable that controls how a tf-numpy symbol's
docstring should refer to the original numpy docstring. If `value` is
`'inlined'`, the numpy docstring will be verbatim copied into the tf-numpy
docstring. Otherwise, a link to the original numpy docstring will be
added. Which numpy version the link points to depends on `value`:
* `'stable'`: the current stable version;
* `'dev'`: the current development version;
* pattern `\d+(\.\d+(\.\d+)?)?`: `value` will be treated as a version number,
e.g. '1.16'.
Args:
value: the value to set the global variable to."
10137,is_sig_mismatch_an_error,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,305,function,
10138,set_is_sig_mismatch_an_error,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,309,function,
10139,np_doc,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,314,function,"Attachs numpy docstring to a function.
Args:
np_fun_name: name for the np_fun symbol. At least one of np_fun or
np_fun_name shoud be set.
np_fun: (optional) the numpy function whose docstring will be used.
export: whether to export this symbol under module
`tf.experimental.numpy`. Note that if `export` is `True`, `np_fun` must be
a function directly under the `numpy` module, not under any submodule of
`numpy` (e.g. `numpy.random`).
Returns:
A function decorator that attaches the docstring from `np_fun` to the
decorated function."
10140,np_doc_only,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,378,function,"Attachs numpy docstring to a function.
This differs from np_doc in that it doesn't check for a match in signature.
Args:
np_fun_name: name for the np_fun symbol. At least one of np_fun or
np_fun_name shoud be set.
np_fun: (optional) the numpy function whose docstring will be used.
export: whether to export this symbol under module
`tf.experimental.numpy`. Note that if `export` is `True`, `np_f` must be a
function directly under the `numpy` module, not under any submodule of
`numpy` (e.g. `numpy.random`).
Returns:
A function decorator that attaches the docstring from `np_fun` to the
decorated function."
10141,finfo,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,410,function,"Note that currently it just forwards to the numpy namesake, while
tensorflow and numpy dtypes may have different properties."
10142,result_type,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,436,function,
10143,promote_types,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,456,function,
10144,tf_broadcast,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,462,function,"Broadcast tensors.
Args:
*args: a list of tensors whose shapes are broadcastable against each other.
Returns:
Tensors broadcasted to the common shape."
10145,get_static_value,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,483,function,"A version of tf.get_static_value that returns None on float dtypes.
It returns None on float dtypes in order to avoid breaking gradients.
Args:
x: a tensor.
Returns:
Same as `tf.get_static_value`, except that it returns None when `x` has a
float dtype."
10146,cond,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,512,function,A version of tf.cond that tries to evaluate the condition.
10147,add,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,523,function,A version of tf.add that eagerly evaluates if possible.
10148,subtract,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,528,function,A version of tf.subtract that eagerly evaluates if possible.
10149,greater,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,533,function,A version of tf.greater that eagerly evaluates if possible.
10150,greater_equal,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,538,function,A version of tf.greater_equal that eagerly evaluates if possible.
10151,less_equal,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,543,function,A version of tf.less_equal that eagerly evaluates if possible.
10152,logical_and,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,548,function,A version of tf.logical_and that eagerly evaluates if possible.
10153,logical_or,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,563,function,A version of tf.logical_or that eagerly evaluates if possible.
10154,getitem,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,578,function,A version of __getitem__ that eagerly evaluates if possible.
10155,reduce_all,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,583,function,A version of tf.reduce_all that eagerly evaluates if possible.
10156,reduce_any,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,592,function,A version of tf.reduce_any that eagerly evaluates if possible.
10157,tf_rank,tensorflow/tensorflow/python/ops/numpy_ops/np_utils.py,601,function,
10158,MicroBenchmarks,tensorflow/tensorflow/python/ops/numpy_ops/integration_test/benchmarks/micro_benchmarks.py,46,class,Main micro benchmark class.
10159,benchmark_tf_np_mlp_inference_batch_1_cpu,tensorflow/tensorflow/python/ops/numpy_ops/integration_test/benchmarks/micro_benchmarks.py,83,method,
10160,benchmark_tf_np_tf_function_mlp_inference_batch_1_cpu,tensorflow/tensorflow/python/ops/numpy_ops/integration_test/benchmarks/micro_benchmarks.py,89,method,
10161,benchmark_numpy_mlp_inference_batch_1_cpu,tensorflow/tensorflow/python/ops/numpy_ops/integration_test/benchmarks/micro_benchmarks.py,96,method,
10162,benchmark_count_nonzero,tensorflow/tensorflow/python/ops/numpy_ops/integration_test/benchmarks/micro_benchmarks.py,136,method,
10163,benchmark_log,tensorflow/tensorflow/python/ops/numpy_ops/integration_test/benchmarks/micro_benchmarks.py,139,method,
10164,benchmark_exp,tensorflow/tensorflow/python/ops/numpy_ops/integration_test/benchmarks/micro_benchmarks.py,142,method,
10165,benchmark_tanh,tensorflow/tensorflow/python/ops/numpy_ops/integration_test/benchmarks/micro_benchmarks.py,145,method,
10166,benchmark_matmul,tensorflow/tensorflow/python/ops/numpy_ops/integration_test/benchmarks/micro_benchmarks.py,148,method,
10167,MLP,tensorflow/tensorflow/python/ops/numpy_ops/integration_test/benchmarks/numpy_mlp.py,27,class,"MLP model.
T = Relu(Add(MatMul(A, B), C))
R = Relu(Add(MatMul(T, D), E))"
10168,inference,tensorflow/tensorflow/python/ops/numpy_ops/integration_test/benchmarks/numpy_mlp.py,45,method,
10169,MLP,tensorflow/tensorflow/python/ops/numpy_ops/integration_test/benchmarks/tf_numpy_mlp.py,29,class,"MLP model.
T = Relu(Add(MatMul(A, B), C))
R = Relu(Add(MatMul(T, D), E))"
10170,inference,tensorflow/tensorflow/python/ops/numpy_ops/integration_test/benchmarks/tf_numpy_mlp.py,47,method,
10171,for_loop,tensorflow/tensorflow/python/ops/parallel_for/control_flow_ops.py,44,function,"Runs `loop_fn` `iters` times and stacks the outputs.
Runs `loop_fn` `iters` times, with input values from 0 to `iters - 1`, and
stacks corresponding outputs of the different runs.
Args:
loop_fn: A function that takes an int32 scalar tf.Tensor object representing
the iteration number, and returns a possibly nested structure of tensor
objects. The shape of these outputs should not depend on the input.
loop_fn_dtypes: dtypes for the outputs of `loop_fn`.
iters: Number of iterations for which to run `loop_fn`.
parallel_iterations: The number of iterations that can be dispatched in
parallel. This knob can be used to control the total memory usage.
Returns:
Returns a nested structure of stacked output tensor objects with the same
nested structure as the output of `loop_fn`."
10172,pfor,tensorflow/tensorflow/python/ops/parallel_for/control_flow_ops.py,136,function,"Equivalent to running `loop_fn` `iters` times and stacking the outputs.
`pfor` has functionality similar to `for_loop`, i.e. running `loop_fn` `iters`
times, with input from 0 to `iters - 1`, and stacking corresponding output of
each iteration. However the implementation does not use a `tf.while_loop`.
Instead it adds new operations to the graph that collectively compute the same
value as what running `loop_fn` in a loop would compute.
This is an experimental feature and currently has a lot of limitations:
- There should be no data dependency between the different iterations. For
example, a future iteration should not depend on a value or side-effect of
a previous iteration.
- Stateful kernels may mostly not be supported since these often imply a
data dependency or ordering of the iterations. We do support a limited set
of such stateful kernels though (like RandomFoo, Variable operations like
reads, etc).
- Conversion works only on a limited set of kernels for which a converter
has been registered.
- `loop_fn` has limited support for control flow operations. `tf.cond` in
particular is not supported.
- `loop_fn` should return nested structure of Tensors or Operations. However
if an Operation is returned, it should have zero outputs.
- The shape and dtype of `loop_fn` outputs should not depend on the input
to loop_fn.
Args:
loop_fn: A function that takes an int32 scalar tf.Tensor object representing
the iteration number, and optionally a keyword argument `pfor_config` set
to a PForConfig object. It returns a possibly nested structure of Tensor
or Operation objects. Note that if setting `parallel_iterations` argument
to something other than None, `loop_fn` may be called more than once
during graph construction. So it may need to avoid mutating global state.
iters: Number of iterations for which to run `loop_fn`.
fallback_to_while_loop: If true, on failing to vectorize an operation, pfor
fallbacks to using a `tf.while_loop` to dispatch the iterations.
parallel_iterations: A knob to control how many iterations are vectorized
and dispatched in parallel. The default value of None corresponds to
vectorizing all the iterations. If `parallel_iterations` is smaller than
`iters`, then chunks of at most that many iterations are dispatched in
sequence. This knob can be used to control the total memory usage.
Returns:
Returns a nested structure of stacked tensor objects with the same nested
structure as the output of `loop_fn`.
Raises:
ValueError: If parallel_iterations is not None and not an integer > 1."
10173,vectorized_map,tensorflow/tensorflow/python/ops/parallel_for/control_flow_ops.py,353,function,"Parallel map on the list of tensors unpacked from `elems` on dimension 0.
This method works similar to `tf.map_fn` but is optimized to run much faster,
possibly with a much larger memory footprint. The speedups are obtained by
vectorization (see [Auto-Vectorizing TensorFlow Graphs: Jacobians,
Auto-Batching and Beyond](https://arxiv.org/pdf/1903.04243.pdf)). The idea
behind vectorization is to semantically launch all the invocations of `fn` in
parallel and fuse corresponding operations across all these invocations. This
fusion is done statically at graph generation time and the generated code is
often similar in performance to a manually fused version.
Because `tf.vectorized_map` fully parallelizes the batch, this method will
generally be significantly faster than using `tf.map_fn`, especially in eager
mode. However this is an experimental feature and currently has a lot of
limitations:
- There should be no data dependency between the different semantic
invocations of `fn`, i.e. it should be safe to map the elements of the
inputs in any order.
- Stateful kernels may mostly not be supported since these often imply a
data dependency. We do support a limited set of such stateful kernels
though (like RandomFoo, Variable operations like reads, etc).
- `fn` has limited support for control flow operations.
- `fn` should return nested structure of Tensors or Operations. However
if an Operation is returned, it should have zero outputs.
- The shape and dtype of any intermediate or output tensors in the
computation of `fn` should not depend on the input to `fn`.
Examples:
```python
def outer_product(a):
return tf.tensordot(a, a, 0)
batch_size = 100
a = tf.ones((batch_size, 32, 32))
c = tf.vectorized_map(outer_product, a)
assert c.shape == (batch_size, 32, 32, 32, 32)
```
```python
# Computing per-example gradients
batch_size = 10
num_features = 32
layer = tf.keras.layers.Dense(1)
def model_fn(arg):
with tf.GradientTape() as g:
inp, label = arg
inp = tf.expand_dims(inp, 0)
label = tf.expand_dims(label, 0)
prediction = layer(inp)
loss = tf.nn.l2_loss(label - prediction)
return g.gradient(loss, (layer.kernel, layer.bias))
inputs = tf.random.uniform([batch_size, num_features])
labels = tf.random.uniform([batch_size, 1])
per_example_gradients = tf.vectorized_map(model_fn, (inputs, labels))
assert per_example_gradients[0].shape == (batch_size, num_features, 1)
assert per_example_gradients[1].shape == (batch_size, 1)
```
Args:
fn: The callable to be performed. It accepts one argument, which will have
the same (possibly nested) structure as `elems`, and returns a possibly
nested structure of Tensors and Operations, which may be different than
the structure of `elems`.
elems: A tensor or (possibly nested) sequence of tensors, each of which will
be unpacked along their first dimension. The nested sequence of the
resulting slices will be mapped over by `fn`.
fallback_to_while_loop: If true, on failing to vectorize an operation,
the unsupported op is wrapped in a tf.while_loop to execute the map
iterations. Note that this fallback only happens for unsupported ops and
other parts of `fn` are still vectorized. If false, on encountering an
unsupported op, a ValueError is thrown. Note that the fallbacks can result
in slowdowns since vectorization often yields speedup of one to two orders
of magnitude.
Returns:
A tensor or (possibly nested) sequence of tensors. Each tensor packs the
results of applying fn to tensors unpacked from elems along the first
dimension, from first to last.
Raises:
ValueError: If vectorization fails and fallback_to_while_loop is False."
10174,dynamic_lstm_input_fn,tensorflow/tensorflow/python/ops/parallel_for/control_flow_ops_test.py,1288,function,
10175,create_dynamic_lstm,tensorflow/tensorflow/python/ops/parallel_for/control_flow_ops_test.py,1298,function,
10176,Benchmarks,tensorflow/tensorflow/python/ops/parallel_for/control_flow_ops_test.py,1634,class,
10177,benchmark_sess_run_overhead,tensorflow/tensorflow/python/ops/parallel_for/control_flow_ops_test.py,1660,method,
10178,benchmark_add,tensorflow/tensorflow/python/ops/parallel_for/control_flow_ops_test.py,1665,method,
10179,benchmark_matmul,tensorflow/tensorflow/python/ops/parallel_for/control_flow_ops_test.py,1685,method,
10180,benchmark_map_fn,tensorflow/tensorflow/python/ops/parallel_for/control_flow_ops_test.py,1704,method,
10181,benchmark_basic_while,tensorflow/tensorflow/python/ops/parallel_for/control_flow_ops_test.py,1721,method,
10182,benchmark_dynamic_rnn,tensorflow/tensorflow/python/ops/parallel_for/control_flow_ops_test.py,1736,method,
10183,benchmark_reduction,tensorflow/tensorflow/python/ops/parallel_for/control_flow_ops_test.py,1743,method,
10184,loop_fn,tensorflow/tensorflow/python/ops/parallel_for/control_flow_ops_test.py,1672,method,
10185,loop_fn,tensorflow/tensorflow/python/ops/parallel_for/control_flow_ops_test.py,1692,method,
10186,pfor_map_fn,tensorflow/tensorflow/python/ops/parallel_for/control_flow_ops_test.py,1711,method,
10187,loop_fn,tensorflow/tensorflow/python/ops/parallel_for/control_flow_ops_test.py,1724,method,
10188,loop_fn,tensorflow/tensorflow/python/ops/parallel_for/control_flow_ops_test.py,1749,method,
10189,jacobian,tensorflow/tensorflow/python/ops/parallel_for/gradients.py,28,function,"Computes jacobian of `output` w.r.t. `inputs`.
Args:
output: A tensor.
inputs: A tensor or a nested structure of tensor objects.
use_pfor: If true, uses pfor for computing the jacobian. Else uses
tf.while_loop.
parallel_iterations: A knob to control how many iterations and dispatched in
parallel. This knob can be used to control the total memory usage.
Returns:
A tensor or a nested structure of tensors with the same structure as
`inputs`. Each entry is the jacobian of `output` w.r.t. to the corresponding
value in `inputs`. If output has shape [y_1, ..., y_n] and inputs_i has
shape [x_1, ..., x_m], the corresponding jacobian has shape
[y_1, ..., y_n, x_1, ..., x_m]. Note that in cases where the gradient is
sparse (IndexedSlices), jacobian function currently makes it dense and
returns a Tensor instead. This may change in the future."
10190,batch_jacobian,tensorflow/tensorflow/python/ops/parallel_for/gradients.py,83,function,"Computes and stacks jacobians of `output[i,...]` w.r.t. `input[i,...]`.
e.g.
x = tf.constant([[1, 2], [3, 4]], dtype=tf.float32)
y = x * x
jacobian = batch_jacobian(y, x)
# => [[[2, 0], [0, 4]], [[6, 0], [0, 8]]]
Args:
output: A tensor with shape [b, y1, ..., y_n]. `output[i,...]` should
only depend on `inp[i,...]`.
inp: A tensor with shape [b, x1, ..., x_m]
use_pfor: If true, uses pfor for computing the Jacobian. Else uses a
tf.while_loop.
parallel_iterations: A knob to control how many iterations are vectorized
and dispatched in parallel. The default value of None, when use_pfor is
true, corresponds to vectorizing all the iterations. When use_pfor is
false, the default value of None corresponds to parallel_iterations=10.
This knob can be used to control the total memory usage.
Returns:
A tensor `t` with shape [b, y_1, ..., y_n, x1, ..., x_m] where `t[i, ...]`
is the jacobian of `output[i, ...]` w.r.t. `inp[i, ...]`, i.e. stacked
per-example jacobians.
Raises:
ValueError: if first dimension of `output` and `inp` do not match."
10191,FullyConnectedModel,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,51,class,
10192,fully_connected_model_fn,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,66,function,
10193,lstm_model_fn,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,72,function,
10194,dynamic_lstm_model_fn,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,85,function,
10195,create_fc_batch_jacobian,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,103,function,
10196,create_lstm_batch_jacobian,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,111,function,
10197,create_dynamic_lstm_batch_jacobian,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,119,function,
10198,create_lstm_batch_hessian,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,133,function,
10199,create_lstm_hessian,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,145,function,
10200,create_fc_per_eg_grad,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,161,function,
10201,create_lstm_per_eg_grad,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,185,function,
10202,Mnist,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,217,class,
10203,create_mnist_autobatch,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,264,function,
10204,create_mnist_per_eg_grad,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,280,function,
10205,create_mnist_batch_jacobian,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,303,function,
10206,create_mnist_per_eg_jacobian,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,313,function,
10207,create_fc_per_eg_jacobians,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,332,function,
10208,GradientsBenchmarks,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,571,class,
10209,benchmark_fc_batch_jacobian,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,596,method,
10210,benchmark_lstm_batch_jacobian,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,602,method,
10211,benchmark_lstm_hessian,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,609,method,
10212,benchmark_lstm_batch_hessian,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,615,method,
10213,benchmark_fc_per_eg_grad,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,621,method,
10214,benchmark_lstm_per_eg_grad,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,627,method,
10215,benchmark_mnist_autobatch,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,633,method,
10216,benchmark_mnist_per_eg_grad,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,643,method,
10217,benchmark_mnist_per_eg_jacobian,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,652,method,
10218,benchmark_mnist_batch_jacobian,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,663,method,
10219,benchmark_fc_per_eg_jacobian,tensorflow/tensorflow/python/ops/parallel_for/gradients_test.py,674,method,
10220,WhileOp,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,144,class,Object for storing state for converting the outputs of a while_loop.
10221,inputs,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,264,method,Input to all the Enter nodes.
10222,control_inputs,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,269,method,Control input to all the Enter nodes.
10223,outputs,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,277,method,Outputs of all the Exit nodes.
10224,name,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,282,method,Context name for the while loop.
10225,is_inside_loop,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,287,method,Returns true if the while_loop was created inside the pfor.
10226,op_is_inside_loop,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,291,method,True if op was created inside the pfor loop body.
10227,is_stateful,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,300,method,
10228,pfor_converter,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,304,method,Return a converter for the while loop.
10229,true_fn,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,553,method,"Converts the body function for all but last iteration.
This essentially converts body_output. Additionally, it needs to handle
any control dependencies on the NextIteration node. So it creates another
Identity node with the converted dependencies."
10230,cond,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,644,method,
10231,body,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,647,method,
10232,ConversionNotImplementedError,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,715,class,
10233,RegisterPFor,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,846,class,"Utility to register converters for pfor.
Usage:
@RegisterPFor(foo_op_type)
def _foo_converter(pfor_input):
...
The above will register conversion function `_foo_converter` for handling
conversion of `foo_op_type`. These converters are called during vectorization
of a `pfor` loop body. For each operation node in this loop body,
the vectorization process will call the converter corresponding to the
operation type of the node.
During conversion, the registered function will be called with a single
argument `pfor_input`, of type `PForInput`, which will contain state needed
for the conversion. When the converter is called for a node, all its inputs
should already have been converted and these converted values are stored in
`pfor_input.inputs`. This registered function should output a list of
WrappedTensor objects with the same length as the number of outputs of the
node being converted. If the node had zero outputs, then it should return an
ops.Operation object. These new sets of nodes should implement the
functionality of running that operation for the number of iterations specified
by `pfor_input.pfor.loop_len_vector[0]` where the inputs of the node for each
iteration are picked from `pfor_inputs.inputs()`.
One tricky aspect of the conversion process is keeping track of, and
leveraging loop invariance of computation. Each converted input is a
WrappedTensor which indicates whether the input was loop invariant or not. If
the converted value is loop invariant, its rank should match the rank of the
corresponding tensor in the loop body, else its rank is larger by 1. The
converter should look at the loop invariance of the inputs and generate new
nodes based on that. Note that the converter will not be called if all inputs
are loop invariant and the operation is not stateful. The converter should
determine if its own output is loop invariant and `wrap` its output
accordingly.
Example:
Here, the converter is trying to convert a Reshape node in the loop body. This
node will have two inputs: the tensor to reshape, and the new shape. The
example here only handles the case where the shape is loop invariant.
@RegisterPFor(""Reshape"")
def _convert_reshape(pfor_input):
# We assume that input is not loop invariant. Call to `stacked_input`
# asserts that and returns the converted value. This value will have a rank
# larger by 1 compared to the rank of the input in the loop body.
t = pfor_input.stacked_input(0)
# We assume that shape input is loop invariant. Call to `unstacked_input`
# asserts that and returns the converted value.
shape = pfor_input.unstacked_input(1)
# We compute `new_shape` by prepending the number of iterations to the
# original shape.
new_shape = array_ops.concat([pfor_input.pfor.loop_len_vector, shape],
axis=0)
# The vectorized output involves reshaping the converted input `t` using
# `new_shape`.
new_output = array_ops.reshape(t, new_shape)
# The converted output is marked as not loop invariant using the call to
# wrap.
return wrap(new_output, True)"
10234,RegisterPForWithArgs,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,925,class,"Utility to register converters for pfor.
Usage:
@RegisteRPFor(foo_op_type, foo=value, ....)
def _foo_converter(pfor_input, foo=None, ....):
...
See RegisterPFor for details on the conversion function.
`RegisterPForWithArgs` allows binding extra arguments to the
conversion function at registration time."
10235,wrap,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,984,function,Helper to create a WrappedTensor object.
10236,PForConfig,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,1028,class,A configuration object used to communicate with loop body function.
10237,reduce,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,1046,method,"Performs reduction `fn` on `args` vectorized across pfor iterations.
Note that `fn` is traced once inside the loop function context. Hence any
captures or side-effects will happen in that context. Call to the traced
version of `fn` happens during the construction of the vectorized code.
Note that this currently may not work inside a control flow construct.
Args:
fn: a reduction function. It will be called with arguments that have the
same structure as *args but with individual values whose rank may be
higher by 1 since they represent loop invariant vectorized versions of
the corresponding Tensors in *args.
*args: unvectorized Tensors.
Returns:
The result of running `fn` on the vectorized versions of `*args`. These
outputs will be available as loop invariant values to all the iterations."
10238,reduce_concat,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,1106,method,"Performs a concat reduction on `x` across pfor iterations.
Note that this currently may not work inside a control flow construct.
Args:
x: an unvectorized Tensor.
Returns:
A Tensor that has rank one higher than `x`. The value is the vectorized
version of `x`, i.e. stacking the value of `x` across different pfor
iterations."
10239,reduce_mean,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,1120,method,"Performs a mean reduction on `x` across pfor iterations.
Note that this currently may not work inside a control flow construct.
Args:
x: an unvectorized Tensor.
Returns:
A Tensor that has same rank as `x`. The value is the mean of the values
of `x` across the pfor iterations."
10240,reduce_sum,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,1133,method,"Performs a sum reduction on `x` across pfor iterations.
Note that this currently may not work inside a control flow construct.
Args:
x: an unvectorized Tensor.
Returns:
A Tensor that has same rank as `x`. The value is the sum of the values
of `x` across the pfor iterations."
10241,PFor,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,1152,class,"Implementation of rewrite of parallel-for loops.
This class takes a DAG or a set of DAGs representing the body of a
parallel-for loop, and adds new operations to the graph that implements
functionality equivalent to running that loop body for a specified number of
iterations. This new set of nodes may or may not use a tensorflow loop
construct.
The process of conversion does not delete or change any existing operations.
It only adds operations that efficiently implement the equivalent
functionality. We refer to the added ops as ""converted ops"".
The conversion process uses a simple greedy heuristic. It walks the loop body
and tries to express the functionality of running each node in a loop with a
new set of nodes. When converting an op several cases are possible:
- The op is not inside the loop body. Hence it can be used as is.
- The op does not depend on the iteration number and is stateless. In this
case, it can be used as is.
- The op is not stateful, and depends on iteration number only through control
dependencies. In this case, we can create a single op with same inputs and
attributes, but with ""converted"" control dependencies.
- The op is not stateful, and all its inputs are loop invariant. In this
case, similar to above, we can create a single op with same inputs and
attributes, but with ""converted"" control dependencies.
- The op is stateful or at least one of the inputs is not loop invariant. In
this case, we run the registered converter for that op to create a set of
converted ops. All nodes in the set will have converted control dependencies
corresponding to control dependencies of the original op. If the op returned
multiple outputs, ""converted outputs"" could be produced by different ops in
this set."
10242,op_is_inside_loop,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,1233,method,True if op was created inside the pfor loop body.
10243,convert,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,1315,method,"Returns the converted value corresponding to y.
Args:
y: A ops.Tensor or a ops.Operation object. If latter, y should not have
any outputs.
Returns:
If y does not need to be converted, it returns y as is. Else it returns
the ""converted value"" corresponding to y."
10244,loop_len_vector,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,1579,method,Returns a single element vector whose value is number of iterations.
10245,loop_var,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,1584,method,Returns placeholder loop variable.
10246,pfor_ops,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,1589,method,
10247,pfor_config,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,1593,method,
10248,all_indices_partitioned,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,1597,method,"all_indices_partitioned property.
Returns:
True if we are inside a control flow construct and not all pfor iterations
may be active."
10249,fallback_to_while_loop,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,1607,method,
10250,fn,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,1294,method,
10251,WhileV2,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,4067,class,Object for vectorizing V2 while_loop op.
10252,true_fn,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,4180,method,Converts the body function for all but last iteration.
10253,cond,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,4270,method,
10254,body,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,4273,method,
10255,while_fn,tensorflow/tensorflow/python/ops/parallel_for/pfor.py,4333,method,
10256,boolean_mask,tensorflow/tensorflow/python/ops/ragged/ragged_array_ops.py,45,function,"Applies a boolean mask to `data` without flattening the mask dimensions.
Returns a potentially ragged tensor that is formed by retaining the elements
in `data` where the corresponding value in `mask` is `True`.
* `output[a1...aA, i, b1...bB] = data[a1...aA, j, b1...bB]`
Where `j` is the `i`th `True` entry of `mask[a1...aA]`.
Note that `output` preserves the mask dimensions `a1...aA`; this differs
from `tf.boolean_mask`, which flattens those dimensions.
Args:
data: A potentially ragged tensor.
mask: A potentially ragged boolean tensor. `mask`'s shape must be a prefix
of `data`'s shape. `rank(mask)` must be known statically.
name: A name prefix for the returned tensor (optional).
Returns:
A potentially ragged tensor that is formed by retaining the elements in
`data` where the corresponding value in `mask` is `True`.
* `rank(output) = rank(data)`.
* `output.ragged_rank = max(data.ragged_rank, rank(mask) - 1)`.
Raises:
ValueError: if `rank(mask)` is not known statically; or if `mask.shape` is
not a prefix of `data.shape`.
#### Examples:
>>> # Aliases for True & False so data and mask line up.
>>> T, F = (True, False)
>>> tf.ragged.boolean_mask( # Mask a 2D Tensor.
... data=[[1, 2, 3], [4, 5, 6], [7, 8, 9]],
... mask=[[T, F, T], [F, F, F], [T, F, F]]).to_list()
[[1, 3], [], [7]]
>>> tf.ragged.boolean_mask( # Mask a 2D RaggedTensor.
... tf.ragged.constant([[1, 2, 3], [4], [5, 6]]),
... tf.ragged.constant([[F, F, T], [F], [T, T]])).to_list()
[[3], [], [5, 6]]
>>> tf.ragged.boolean_mask( # Mask rows of a 2D RaggedTensor.
... tf.ragged.constant([[1, 2, 3], [4], [5, 6]]),
... tf.ragged.constant([True, False, True])).to_list()
[[1, 2, 3], [5, 6]]"
10257,tile,tensorflow/tensorflow/python/ops/ragged/ragged_array_ops.py,211,function,"Constructs a `RaggedTensor` by tiling a given `RaggedTensor`.
The values of `input` are replicated `multiples[i]` times along the
`i`th dimension (for each dimension `i`). For every dimension `axis` in
`input`, the length of each output element in that dimension is the
length of corresponding input element multiplied by `multiples[axis]`.
Args:
input: A `RaggedTensor`.
multiples: A 1-D integer `Tensor`. Length must be the same as the number of
dimensions in `input`.
name: A name for the operation (optional).
Returns:
A `RaggedTensor` with the same type, rank, and ragged_rank as `input`.
#### Example:
>>> rt = tf.ragged.constant([[1, 2], [3]])
>>> tf.tile(rt, [3, 2]).to_list()
[[1, 2, 1, 2], [3, 3], [1, 2, 1, 2], [3, 3], [1, 2, 1, 2], [3, 3]]"
10258,expand_dims,tensorflow/tensorflow/python/ops/ragged/ragged_array_ops.py,384,function,"Inserts a dimension with shape 1 into a potentially ragged tensor's shape.
Given a potentially ragged tenor `input`, this operation inserts a
dimension with size 1 at the dimension `axis` of `input`'s shape.
The following table gives some examples showing how `ragged.expand_dims`
impacts the shapes of different input tensors. Ragged dimensions are
indicated by enclosing them in parentheses.
input.shape | axis | result.shape
----------------------- | ---- | -----------------------------
`[D1, D2]` | `0` | `[1, D1, D2]`
`[D1, D2]` | `1` | `[D1, 1, D2]`
`[D1, D2]` | `2` | `[D1, D2, 1]`
`[D1, (D2), (D3), D4]` | `0` | `[1, D1, (D2), (D3), D4]`
`[D1, (D2), (D3), D4]` | `1` | `[D1, 1, (D2), (D3), D4]`
`[D1, (D2), (D3), D4]` | `2` | `[D1, (D2), 1, (D3), D4]`
`[D1, (D2), (D3), D4]` | `3` | `[D1, (D2), (D3), 1, D4]`
`[D1, (D2), (D3), D4]` | `4` | `[D1, (D2), (D3), D4, 1]`
Args:
input: The potentially tensor that should be expanded with a new dimension.
axis: An integer constant indicating where the new dimension should be
inserted.
name: A name for the operation (optional).
Returns:
A tensor with the same values as `input`, with an added dimension of
size 1 at `axis`.
#### Examples:
>>> rt = tf.ragged.constant([[1, 2], [3]])
>>> print(rt.shape)
(2, None)
>>> expanded = tf.expand_dims(rt, axis=0)
>>> print(expanded.shape, expanded)
(1, 2, None) <tf.RaggedTensor [[[1, 2], [3]]]>
>>> expanded = tf.expand_dims(rt, axis=1)
>>> print(expanded.shape, expanded)
(2, 1, None) <tf.RaggedTensor [[[1, 2]], [[3]]]>
>>> expanded = tf.expand_dims(rt, axis=2)
>>> print(expanded.shape, expanded)
(2, None, 1) <tf.RaggedTensor [[[1], [2]], [[3]]]>"
10259,size,tensorflow/tensorflow/python/ops/ragged/ragged_array_ops.py,458,function,"Returns the size of a potentially ragged tensor.
The size of a ragged tensor is the size of its inner values.
#### Example:
>>> tf.size(tf.ragged.constant([[1, 2], [3]])).numpy()
3
Args:
input: A potentially ragged `Tensor`.
out_type: The numeric output type for the operation.
name: A name for the operation (optional).
Returns:
A Tensor of type `out_type`."
10260,rank,tensorflow/tensorflow/python/ops/ragged/ragged_array_ops.py,485,function,"Returns the rank of a RaggedTensor.
Returns a 0-D `int32` `Tensor` representing the rank of `input`.
#### Example:
>>> # shape of tensor 't' is [2, None, None]
>>> t = tf.ragged.constant([[[1], [2, 2]], [[3, 3, 3], [4, 4, 4, 4]]])
>>> tf.rank(t).numpy()
3
Args:
input: A `RaggedTensor`
name: A name for the operation (optional).
Returns:
A `Tensor` of type `int32`."
10261,ragged_one_hot,tensorflow/tensorflow/python/ops/ragged/ragged_array_ops.py,514,function,Applies tf.one_hot along the values of a RaggedTensor.
10262,stack_dynamic_partitions,tensorflow/tensorflow/python/ops/ragged/ragged_array_ops.py,544,function,"Stacks dynamic partitions of a Tensor or RaggedTensor.
Returns a RaggedTensor `output` with `num_partitions` rows, where the row
`output[i]` is formed by stacking all slices `data[j1...jN]` such that
`partitions[j1...jN] = i`. Slices of `data` are stacked in row-major
order.
If `num_partitions` is an `int` (not a `Tensor`), then this is equivalent to
`tf.ragged.stack(tf.dynamic_partition(data, partitions, num_partitions))`.
#### Example:
>>> data = ['a', 'b', 'c', 'd', 'e']
>>> partitions = [ 3, 0, 2, 2, 3]
>>> num_partitions = 5
>>> tf.ragged.stack_dynamic_partitions(data, partitions, num_partitions)
<tf.RaggedTensor [[b'b'], [], [b'c', b'd'], [b'a', b'e'], []]>
Args:
data: A `Tensor` or `RaggedTensor` containing the values to stack.
partitions: An `int32` or `int64` `Tensor` or `RaggedTensor` specifying the
partition that each slice of `data` should be added to. `partitions.shape`
must be a prefix of `data.shape`. Values must be greater than or equal to
zero, and less than `num_partitions`. `partitions` is not required to be
sorted.
num_partitions: An `int32` or `int64` scalar specifying the number of
partitions to output. This determines the number of rows in `output`.
name: A name prefix for the returned tensor (optional).
Returns:
A `RaggedTensor` containing the stacked partitions. The returned tensor
has the same dtype as `data`, and its shape is
`[num_partitions, (D)] + data.shape[partitions.rank:]`, where `(D)` is a
ragged dimension whose length is the number of data slices stacked for
each `partition`."
10263,reverse,tensorflow/tensorflow/python/ops/ragged/ragged_array_ops.py,648,function,"Reverses a RaggedTensor along the specified axes.
#### Example:
>>> data = tf.ragged.constant([
... [[1, 2], [3, 4]], [[5, 6]], [[7, 8], [9, 10], [11, 12]]])
>>> tf.reverse(data, axis=[0, 2])
<tf.RaggedTensor [[[8, 7], [10, 9], [12, 11]], [[6, 5]], [[2, 1], [4, 3]]]>
Args:
tensor: A 'RaggedTensor' to reverse.
axis: A list or tuple of 'int' or a constant 1D 'tf.Tensor'. The indices of
the axes to reverse.
name: A name prefix for the returned tensor (optional).
Returns:
A 'RaggedTensor'."
10264,cross,tensorflow/tensorflow/python/ops/ragged/ragged_array_ops.py,706,function,"Generates feature cross from a list of tensors.
The input tensors must have `rank=2`, and must all have the same number of
rows. The result is a `RaggedTensor` with the same number of rows as the
inputs, where `result[row]` contains a list of all combinations of values
formed by taking a single value from each input's corresponding row
(`inputs[i][row]`). Values are combined by joining their strings with '_X_'.
E.g.:
>>> tf.ragged.cross([tf.ragged.constant([['a'], ['b', 'c']]),
... tf.ragged.constant([['d'], ['e']]),
... tf.ragged.constant([['f'], ['g']])])
<tf.RaggedTensor [[b'a_X_d_X_f'], [b'b_X_e_X_g', b'c_X_e_X_g']]>
Args:
inputs: A list of `RaggedTensor` or `Tensor` or `SparseTensor`.
name: Optional name for the op.
Returns:
A 2D `RaggedTensor` of type `string`."
10265,cross_hashed,tensorflow/tensorflow/python/ops/ragged/ragged_array_ops.py,733,function,"Generates hashed feature cross from a list of tensors.
The input tensors must have `rank=2`, and must all have the same number of
rows. The result is a `RaggedTensor` with the same number of rows as the
inputs, where `result[row]` contains a list of all combinations of values
formed by taking a single value from each input's corresponding row
(`inputs[i][row]`). Values are combined by hashing together their
fingerprints. E.g.:
>>> tf.ragged.cross_hashed([tf.ragged.constant([['a'], ['b', 'c']]),
... tf.ragged.constant([['d'], ['e']]),
... tf.ragged.constant([['f'], ['g']])],
... num_buckets=100)
<tf.RaggedTensor [[78], [66, 74]]>
Args:
inputs: A list of `RaggedTensor` or `Tensor` or `SparseTensor`.
num_buckets: A non-negative `int` that used to bucket the hashed values. If
`num_buckets != 0`, then `output = hashed_value % num_buckets`.
hash_key: Integer hash_key that will be used by the `FingerprintCat64`
function. If not given, a default key is used.
name: Optional name for the op.
Returns:
A 2D `RaggedTensor` of type `int64`."
10266,batch_gather,tensorflow/tensorflow/python/ops/ragged/ragged_batch_gather_ops.py,27,function,"Gathers slices from `params` according to `indices` with batch dims.
This operation is similar to `gather`, but it assumes that the leading `N`
dimensions of `indices` and `params` are batch dimensions, and performs a
gather within each batch. In particular, when using this operation with `N`
batch dimensions `B1...BN`:
* `indices` has shape `[B1...BN, I]`
* `params` has shape `[B1...BN, P1...PM]`.
* `result` has shape `[B1...BN, I, P2...PM]`.
* `result[b1...bN, i, p2...pM] =
params[b1...bN, indices[b1...bN, i], p2...pM]`
Args:
params: A potentially ragged tensor with shape `[B1...BN, P1...PM]` (`N>=0`,
`M>0`).
indices: A potentially ragged tensor with shape `[B1...BN, I]` (`N>=0`).
name: A name for the operation (optional).
Returns:
A potentially ragged tensor with shape `[B1...BN, I, P2...PM]`.
`result.ragged_rank = max(indices.ragged_rank, params.ragged_rank)`.
#### Example:
>>> params = tf.ragged.constant([['a', 'b', 'c'], ['d'], [], ['e']])
>>> indices = tf.ragged.constant([[1, 2, 0], [], [], [0, 0]])
>>> tf.compat.v1.batch_gather(params, indices)
<tf.RaggedTensor [[b'b', b'c', b'a'], [], [], [b'e', b'e']]>"
10267,batch_gather_with_default,tensorflow/tensorflow/python/ops/ragged/ragged_batch_gather_with_default_op.py,38,function,"Same as `batch_gather` but inserts `default_value` for invalid indices.
This operation is similar to `batch_gather` except that it will substitute
the value for invalid indices with `default_value` as the contents.
See `batch_gather` for more details.
Args:
params: A potentially ragged tensor with shape `[B1...BN, P1...PM]` (`N>=0`,
`M>0`).
indices: A potentially ragged tensor with shape `[B1...BN, I]` (`N>=0`).
default_value: A value to be inserted in places where `indices` are out of
bounds. Must be the same dtype as params and either a scalar or rank 1.
name: A name for the operation (optional).
Returns:
A potentially ragged tensor with shape `[B1...BN, I, P2...PM]`.
`result.ragged_rank = max(indices.ragged_rank, params.ragged_rank)`.
#### Example:
>>> params = tf.ragged.constant([['a', 'b', 'c'], ['d'], [], ['e']])
>>> indices = tf.ragged.constant([[1, 2, -1], [], [], [0, 10]])
>>> batch_gather_with_default(params, indices, 'FOO')
<tf.RaggedTensor [[b'b', b'c', b'FOO'], [], [], [b'e', b'FOO']]>"
10268,concat,tensorflow/tensorflow/python/ops/ragged/ragged_concat_ops.py,34,function,"Concatenates potentially ragged tensors along one dimension.
Given a list of tensors with the same rank `K` (`K >= axis`), returns a
rank-`K` `RaggedTensor` `result` such that `result[i0...iaxis]` is the
concatenation of `[rt[i0...iaxis] for rt in values]`.
Args:
values: A list of potentially ragged tensors. May not be empty. All
`values` must have the same rank and the same dtype; but unlike
`tf.concat`, they can have arbitrary shapes.
axis: A python integer, indicating the dimension along which to concatenate.
(Note: Unlike `tf.concat`, the `axis` parameter must be statically known.)
Negative values are supported only if the rank of at least one
`values` value is statically known.
name: A name prefix for the returned tensor (optional).
Returns:
A `RaggedTensor` with rank `K`.
`result.ragged_rank=max(axis, max(rt.ragged_rank for rt in values]))`.
Raises:
ValueError: If `values` is empty, if `axis` is out of bounds or if
the input tensors have different ranks.
#### Example:
>>> t1 = tf.ragged.constant([[1, 2], [3, 4, 5]])
>>> t2 = tf.ragged.constant([[6], [7, 8, 9]])
>>> tf.concat([t1, t2], axis=0)
<tf.RaggedTensor [[1, 2], [3, 4, 5], [6], [7, 8, 9]]>
>>> tf.concat([t1, t2], axis=1)
<tf.RaggedTensor [[1, 2, 6], [3, 4, 5, 7, 8, 9]]>"
10269,stack,tensorflow/tensorflow/python/ops/ragged/ragged_concat_ops.py,76,function,"Stacks a list of rank-`R` tensors into one rank-`(R+1)` `RaggedTensor`.
Given a list of tensors or ragged tensors with the same rank `R`
(`R >= axis`), returns a rank-`R+1` `RaggedTensor` `result` such that
`result[i0...iaxis]` is `[value[i0...iaxis] for value in values]`.
#### Examples:
>>> # Stacking two ragged tensors.
>>> t1 = tf.ragged.constant([[1, 2], [3, 4, 5]])
>>> t2 = tf.ragged.constant([[6], [7, 8, 9]])
>>> tf.ragged.stack([t1, t2], axis=0)
<tf.RaggedTensor [[[1, 2], [3, 4, 5]], [[6], [7, 8, 9]]]>
>>> tf.ragged.stack([t1, t2], axis=1)
<tf.RaggedTensor [[[1, 2], [6]], [[3, 4, 5], [7, 8, 9]]]>
>>> # Stacking two dense tensors with different sizes.
>>> t3 = tf.constant([[1, 2, 3], [4, 5, 6]])
>>> t4 = tf.constant([[5], [6], [7]])
>>> tf.ragged.stack([t3, t4], axis=0)
<tf.RaggedTensor [[[1, 2, 3], [4, 5, 6]], [[5], [6], [7]]]>
Args:
values: A list of `tf.Tensor` or `tf.RaggedTensor`. May not be empty. All
`values` must have the same rank and the same dtype; but unlike
`tf.stack`, they can have arbitrary dimension sizes.
axis: A python integer, indicating the dimension along which to stack.
(Note: Unlike `tf.stack`, the `axis` parameter must be statically known.)
Negative values are supported only if the rank of at least one
`values` value is statically known.
name: A name prefix for the returned tensor (optional).
Returns:
A `RaggedTensor` with rank `R+1`.
`result.ragged_rank=1+max(axis, max(rt.ragged_rank for rt in values]))`.
Raises:
ValueError: If `values` is empty, if `axis` is out of bounds or if
the input tensors have different ranks."
10270,auto_cast_partition_dtype,tensorflow/tensorflow/python/ops/ragged/ragged_config.py,22,function,"Whether incompatible row-partitioning dtypes should be auto-converted.
If true, then operations that combine RaggedTensors but have different
row-partitioning tensor dtypes will be automatically cast to a
compatible dtype (`tf.int64`). If false, then such operations will result
in an error.
Returns:
`bool`"
10271,from_tensor,tensorflow/tensorflow/python/ops/ragged/ragged_conversion_ops.py,30,function,
10272,to_tensor,tensorflow/tensorflow/python/ops/ragged/ragged_conversion_ops.py,48,function,
10273,ragged_to_dense,tensorflow/tensorflow/python/ops/ragged/ragged_conversion_ops.py,55,function,Create a dense tensor from a ragged tensor.
10274,to_sparse,tensorflow/tensorflow/python/ops/ragged/ragged_conversion_ops.py,140,function,
10275,from_sparse,tensorflow/tensorflow/python/ops/ragged/ragged_conversion_ops.py,144,function,
10276,sparse_const,tensorflow/tensorflow/python/ops/ragged/ragged_cross_op_test.py,40,function,
10277,UnaryRaggedElementwiseDispatcher,tensorflow/tensorflow/python/ops/ragged/ragged_dispatch.py,107,class,OpDispatcher for unary ops that map a base op across ragged values.
10278,handle,tensorflow/tensorflow/python/ops/ragged/ragged_dispatch.py,120,method,
10279,BinaryRaggedElementwiseDispatcher,tensorflow/tensorflow/python/ops/ragged/ragged_dispatch.py,163,class,"OpDispatcher for binary ops that map a base op across ragged values.
Supports broadcasting."
10280,handle,tensorflow/tensorflow/python/ops/ragged/ragged_dispatch.py,180,method,
10281,RaggedDispatcher,tensorflow/tensorflow/python/ops/ragged/ragged_dispatch.py,234,class,"OpDispatcher for ragged ops.
Dispatches to a wrapped op-handler if at least one of the `tensor_args`
arguments is a RaggedTensor or a RaggedTensorValue; and all of the
`tensor_args` arguments are convertible to Tensor or RaggedTensor."
10282,handle,tensorflow/tensorflow/python/ops/ragged/ragged_dispatch.py,257,method,
10283,is_supported,tensorflow/tensorflow/python/ops/ragged/ragged_dispatch.py,263,method,
10284,register_dispatchers,tensorflow/tensorflow/python/ops/ragged/ragged_dispatch.py,528,function,Constructs & registers OpDispatchers for ragged ops.
10285,ragged_op_list,tensorflow/tensorflow/python/ops/ragged/ragged_dispatch.py,591,function,Returns a string listing operators that have dispathers registered.
10286,constant,tensorflow/tensorflow/python/ops/ragged/ragged_factory_ops.py,39,function,"Constructs a constant RaggedTensor from a nested Python list.
Example:
>>> tf.ragged.constant([[1, 2], [3], [4, 5, 6]])
<tf.RaggedTensor [[1, 2], [3], [4, 5, 6]]>
All scalar values in `pylist` must have the same nesting depth `K`, and the
returned `RaggedTensor` will have rank `K`. If `pylist` contains no scalar
values, then `K` is one greater than the maximum depth of empty lists in
`pylist`. All scalar values in `pylist` must be compatible with `dtype`.
Args:
pylist: A nested `list`, `tuple` or `np.ndarray`. Any nested element that
is not a `list`, `tuple` or `np.ndarray` must be a scalar value
compatible with `dtype`.
dtype: The type of elements for the returned `RaggedTensor`. If not
specified, then a default is chosen based on the scalar values in
`pylist`.
ragged_rank: An integer specifying the ragged rank of the returned
`RaggedTensor`. Must be nonnegative and less than `K`. Defaults to
`max(0, K - 1)` if `inner_shape` is not specified. Defaults to
`max(0, K - 1 - len(inner_shape))` if `inner_shape` is specified.
inner_shape: A tuple of integers specifying the shape for individual inner
values in the returned `RaggedTensor`. Defaults to `()` if `ragged_rank`
is not specified. If `ragged_rank` is specified, then a default is chosen
based on the contents of `pylist`.
name: A name prefix for the returned tensor (optional).
row_splits_dtype: data type for the constructed `RaggedTensor`'s row_splits.
One of `tf.int32` or `tf.int64`.
Returns:
A potentially ragged tensor with rank `K` and the specified `ragged_rank`,
containing the values from `pylist`.
Raises:
ValueError: If the scalar values in `pylist` have inconsistent nesting
depth; or if ragged_rank or inner_shape are incompatible with `pylist`."
10287,constant_value,tensorflow/tensorflow/python/ops/ragged/ragged_factory_ops.py,92,function,"Constructs a RaggedTensorValue from a nested Python list.
Warning: This function returns a `RaggedTensorValue`, not a `RaggedTensor`.
If you wish to construct a constant `RaggedTensor`, use
[`ragged.constant(...)`](constant.md) instead.
Example:
>>> tf.compat.v1.ragged.constant_value([[1, 2], [3], [4, 5, 6]])
tf.RaggedTensorValue(values=array([1, 2, 3, 4, 5, 6]),
row_splits=array([0, 2, 3, 6]))
All scalar values in `pylist` must have the same nesting depth `K`, and the
returned `RaggedTensorValue` will have rank `K`. If `pylist` contains no
scalar values, then `K` is one greater than the maximum depth of empty lists
in `pylist`. All scalar values in `pylist` must be compatible with `dtype`.
Args:
pylist: A nested `list`, `tuple` or `np.ndarray`. Any nested element that
is not a `list` or `tuple` must be a scalar value compatible with `dtype`.
dtype: `numpy.dtype`. The type of elements for the returned `RaggedTensor`.
If not specified, then a default is chosen based on the scalar values in
`pylist`.
ragged_rank: An integer specifying the ragged rank of the returned
`RaggedTensorValue`. Must be nonnegative and less than `K`. Defaults to
`max(0, K - 1)` if `inner_shape` is not specified. Defaults to `max(0, K
- 1 - len(inner_shape))` if `inner_shape` is specified.
inner_shape: A tuple of integers specifying the shape for individual inner
values in the returned `RaggedTensorValue`. Defaults to `()` if
`ragged_rank` is not specified. If `ragged_rank` is specified, then a
default is chosen based on the contents of `pylist`.
row_splits_dtype: data type for the constructed `RaggedTensorValue`'s
row_splits. One of `numpy.int32` or `numpy.int64`.
Returns:
A `tf.RaggedTensorValue` or `numpy.array` with rank `K` and the specified
`ragged_rank`, containing the values from `pylist`.
Raises:
ValueError: If the scalar values in `pylist` have inconsistent nesting
depth; or if ragged_rank or inner_shape are incompatible with `pylist`."
10288,placeholder,tensorflow/tensorflow/python/ops/ragged/ragged_factory_ops.py,318,function,"Creates a placeholder for a `tf.RaggedTensor` that will always be fed.
**Important**: This ragged tensor will produce an error if evaluated.
Its value must be fed using the `feed_dict` optional argument to
`Session.run()`, `Tensor.eval()`, or `Operation.run()`.
@compatibility{eager} Placeholders are not compatible with eager execution.
Args:
dtype: The data type for the `RaggedTensor`.
ragged_rank: The ragged rank for the `RaggedTensor`
value_shape: The shape for individual flat values in the `RaggedTensor`.
name: A name for the operation (optional).
Returns:
A `RaggedTensor` that may be used as a handle for feeding a value, but
not evaluated directly.
Raises:
RuntimeError: if eager execution is enabled"
10289,map_flat_values,tensorflow/tensorflow/python/ops/ragged/ragged_functional_ops.py,33,function,"Applies `op` to the values of one or more RaggedTensors.
Replaces any `RaggedTensor` in `args` or `kwargs` with its `flat_values`
tensor, and then calls `op`. Returns a `RaggedTensor` that is constructed
from the input `RaggedTensor`s' `nested_row_splits` and the value returned by
the `op`.
If the input arguments contain multiple `RaggedTensor`s, then they must have
identical `nested_row_splits`.
Examples:
>>> rt = tf.ragged.constant([[1, 2, 3], [], [4, 5], [6]])
>>> map_flat_values(tf.ones_like, rt).to_list()
[[1, 1, 1], [], [1, 1], [1]]
>>> map_flat_values(tf.multiply, rt, rt).to_list()
[[1, 4, 9], [], [16, 25], [36]]
>>> map_flat_values(tf.add, rt, 5).to_list()
[[6, 7, 8], [], [9, 10], [11]]
Args:
op: The operation that should be applied to the RaggedTensor `flat_values`.
`op` is typically an element-wise operation (such as math_ops.add), but
any operation that preserves the size of the outermost dimension can be
used. I.e., `shape[0]` of the value returned by `op` must match
`shape[0]` of the `RaggedTensor`s' `flat_values` tensors.
*args: Arguments for `op`.
**kwargs: Keyword arguments for `op`.
Returns:
A `RaggedTensor` whose `ragged_rank` matches the `ragged_rank` of all
input `RaggedTensor`s.
Raises:
ValueError: If args contains no `RaggedTensors`, or if the `nested_splits`
of the input `RaggedTensor`s are not identical."
10290,gather,tensorflow/tensorflow/python/ops/ragged/ragged_gather_ops.py,36,function,"Gathers ragged slices from `params` axis `0` according to `indices`.
See `tf.gather` for full documentation. (This version has the same API
as `tf.gather`, but supports ragged `params` and `indices`.)
Examples:
>>> params = tf.constant(['a', 'b', 'c', 'd', 'e'])
>>> indices = tf.constant([3, 1, 2, 1, 0])
>>> ragged_params = tf.ragged.constant([['a', 'b', 'c'], ['d'], [], ['e']])
>>> ragged_indices = tf.ragged.constant([[3, 1, 2], [1], [], [0]])
>>> tf.gather(params, ragged_indices)
<tf.RaggedTensor [[b'd', b'b', b'c'], [b'b'], [], [b'a']]>
>>> tf.gather(ragged_params, indices)
<tf.RaggedTensor [[b'e'], [b'd'], [], [b'd'], [b'a', b'b', b'c']]>
>>> tf.gather(ragged_params, ragged_indices)
<tf.RaggedTensor [[[b'e'], [b'd'], []], [[b'd']], [], [[b'a', b'b', b'c']]]>
Args:
params: The potentially ragged tensor from which to gather values. Must be
at least rank 1.
indices: The potentially ragged tensor indicating which values to gather.
Must have dtype `int32` or `int64`. Values must be in the range `[0,
params.shape[0]]`.
validate_indices: Ignored.
axis: The axis in `params` to gather `indices` from.
batch_dims: The number of batch dimensions.
name: A name for the operation (optional).
Returns:
A `RaggedTensor`, where `output.dtype=params.dtype` and
`output.shape=indices.shape + params.shape[1:]` and
`output.ragged_rank=indices.shape.ndims + params.ragged_rank`.
Raises:
ValueError: If indices.shape.ndims is not known statically."
10291,gather_nd,tensorflow/tensorflow/python/ops/ragged/ragged_gather_ops.py,330,function,"Gather slices from `params` using `n`-dimensional indices.
This operation is similar to `gather`, but it uses the innermost dimension
of `indices` to define a slice into `params`. In particular, if:
* `indices` has shape `[A1...AN, I]`
* `params` has shape `[B1...BM]`
Then:
* `result` has shape `[A1...AN, B_{I+1}...BM]`.
* `result[a1...aN] = params[indices[a1...aN, :]]`
Args:
params: A potentially ragged tensor with shape `[A1...AN, I]`.
indices: A potentially ragged tensor with shape `[B1...BM]`.
batch_dims: Must be zero.
name: A name for the operation (optional).
Returns:
A potentially ragged tensor with shape `[A1...AN, B_{I+1}...BM]`.
#### Examples:
>>> params = tf.ragged.constant(
... [ [ ['000', '001'], ['010' ] ],
... [ ['100' ], ['110', '111', '112'], ['120'] ],
... [ [ ], ['210' ] ] ])
>>> # Gather 2D slices from a 3D tensor
>>> tf.gather_nd(params, [[2], [0]])
<tf.RaggedTensor [[[], [b'210']], [[b'000', b'001'], [b'010']]]>
>>> # Gather 1D slices from a 3D tensor
>>> tf.gather_nd(params, [[2, 1], [0, 0]])
<tf.RaggedTensor [[b'210'], [b'000', b'001']]>
>>> # Gather scalars from a 3D tensor
>>> tf.gather_nd(params, [[0, 0, 1], [1, 1, 2]]).numpy()
array([b'001', b'112'], dtype=object)"
10292,ragged_tensor_getitem,tensorflow/tensorflow/python/ops/ragged/ragged_getitem.py,35,function,"Returns the specified piece of this RaggedTensor.
Supports multidimensional indexing and slicing, with one restriction:
indexing into a ragged inner dimension is not allowed. This case is
problematic because the indicated value may exist in some rows but not
others. In such cases, it's not obvious whether we should (1) report an
IndexError; (2) use a default value; or (3) skip that value and return a
tensor with fewer rows than we started with. Following the guiding
principles of Python (""In the face of ambiguity, refuse the temptation to
guess""), we simply disallow this operation.
Args:
self: The RaggedTensor to slice.
key: Indicates which piece of the RaggedTensor to return, using standard
Python semantics (e.g., negative values index from the end). `key`
may have any of the following types:
* `int` constant
* Scalar integer `Tensor`
* `slice` containing integer constants and/or scalar integer
`Tensor`s
* `Ellipsis`
* `tf.newaxis`
* `tuple` containing any of the above (for multidimensional indexing)
Returns:
A `Tensor` or `RaggedTensor` object. Values that include at least one
ragged dimension are returned as `RaggedTensor`. Values that include no
ragged dimensions are returned as `Tensor`. See above for examples of
expressions that return `Tensor`s vs `RaggedTensor`s.
Raises:
ValueError: If `key` is out of bounds.
ValueError: If `key` is not supported.
TypeError: If the indices in `key` have an unsupported type.
Examples:
>>> # A 2-D ragged tensor with 1 ragged dimension.
>>> rt = tf.ragged.constant([['a', 'b', 'c'], ['d', 'e'], ['f'], ['g']])
>>> rt[0].numpy() # First row (1-D `Tensor`)
array([b'a', b'b', b'c'], dtype=object)
>>> rt[:3].to_list() # First three rows (2-D RaggedTensor)
[[b'a', b'b', b'c'], [b'd', b'e'], [b'f']]
>>> rt[3, 0].numpy() # 1st element of 4th row (scalar)
b'g'
>>> # A 3-D ragged tensor with 2 ragged dimensions.
>>> rt = tf.ragged.constant([[[1, 2, 3], [4]],
... [[5], [], [6]],
... [[7]],
... [[8, 9], [10]]])
>>> rt[1].to_list() # Second row (2-D RaggedTensor)
[[5], [], [6]]
>>> rt[3, 0].numpy() # First element of fourth row (1-D Tensor)
array([8, 9], dtype=int32)
>>> rt[:, 1:3].to_list() # Items 1-3 of each row (3-D RaggedTensor)
[[[4]], [[], [6]], [], [[10]]]
>>> rt[:, -1:].to_list() # Last item of each row (3-D RaggedTensor)
[[[4]], [[6]], [[7]], [[10]]]"
10293,map_fn,tensorflow/tensorflow/python/ops/ragged/ragged_map_ops.py,30,function,"map on the list of tensors unpacked from `elems` on dimension 0.
The simplest version of `map_fn` repeatedly applies the callable `fn` to a
sequence of elements from first to last. The elements are made of the
tensors unpacked from `elems`. `dtype` is the data type of the return
value of `fn`. Users must provide `dtype` if it is different from
the data type of `elems`.
Suppose that `elems` is unpacked into `values`, a list of tensors. The shape
of the result tensor is `[values.shape[0]] + fn(values[0]).shape`.
This method also allows multi-arity `elems` and output of `fn`. If `elems`
is a (possibly nested) list or tuple of tensors, then each of these tensors
must have a matching first (unpack) dimension. The signature of `fn` may
match the structure of `elems`. That is, if `elems` is
`(t1, [t2, t3, [t4, t5]])`, then an appropriate signature for `fn` is:
`fn = lambda (t1, [t2, t3, [t4, t5]]):`.
Furthermore, `fn` may emit a different structure than its input. For example,
`fn` may look like: `fn = lambda t1: return (t1 + 1, t1 - 1)`. In this case,
the `dtype` parameter is not optional: `dtype` must be a type or (possibly
nested) tuple of types matching the output of `fn`.
To apply a functional operation to the nonzero elements of a SparseTensor
one of the following methods is recommended. First, if the function is
expressible as TensorFlow ops, use
```python
result = SparseTensor(input.indices, fn(input.values), input.dense_shape)
```
If, however, the function is not expressible as a TensorFlow op, then use
```python
result = SparseTensor(
input.indices, map_fn(fn, input.values), input.dense_shape)
```
instead.
When executing eagerly, map_fn does not execute in parallel even if
`parallel_iterations` is set to a value > 1. You can still get the
performance benefits of running a function in parallel by using the
`tf.contrib.eager.defun` decorator,
```python
# Assume the function being used in map_fn is fn.
# To ensure map_fn calls fn in parallel, use the defun decorator.
@tf.contrib.eager.defun
def func(tensor):
return tf.map_fn(fn, tensor)
```
Note that if you use the defun decorator, any non-TensorFlow Python code
that you may have written in your function won't get executed. See
`tf.contrib.eager.defun` for more details. The recommendation would be to
debug without defun but switch to defun to get performance benefits of
running map_fn in parallel.
Args:
fn: The callable to be performed. It accepts one argument, which will have
the same (possibly nested) structure as `elems`. Its output must have the
same structure as `dtype` if one is provided, otherwise it must have the
same structure as `elems`.
elems: A tensor or (possibly nested) sequence of tensors, each of which will
be unpacked along their first dimension. The nested sequence of the
resulting slices will be applied to `fn`.
dtype: (optional) The output type(s) of `fn`. If `fn` returns a structure
of Tensors differing from the structure of `elems`, then `dtype` is not
optional and must have the same structure as the output of `fn`. Use
`RaggedTensorType` to declare an output of type `RaggedTensor`.
parallel_iterations: (optional) The number of iterations allowed to run in
parallel. When graph building, the default value is 10. While executing
eagerly, the default value is set to 1.
back_prop: (optional) True enables support for back propagation.
swap_memory: (optional) True enables GPU-CPU memory swapping.
infer_shape: (optional) False disables tests for consistent output shapes.
name: (optional) Name prefix for the returned tensors.
Returns:
A possibly nested sequence of potentially ragged tensors. Each
tensor packs the results of applying `fn` to tensors unpacked from `elems`
along the first dimension, from first to last.
Raises:
TypeError: if `fn` is not callable or the structure of the output of
`fn` and `dtype` do not match, or if elems is a SparseTensor.
ValueError: if the lengths of the output of `fn` and `dtype` do not match.
#### Examples:
```python
elems = np.array([1, 2, 3, 4, 5, 6])
squares = map_fn(lambda x: x * x, elems)
# squares == [1, 4, 9, 16, 25, 36]
```
```python
elems = (np.array([1, 2, 3]), np.array([-1, 1, -1]))
alternate = map_fn(lambda x: x[0] * x[1], elems, dtype=tf.int64)
# alternate == [-1, 2, -3]
```
```python
elems = np.array([1, 2, 3])
alternates = map_fn(lambda x: (x, -x), elems, dtype=(tf.int64, tf.int64))
# alternates[0] == [1, 2, 3]
# alternates[1] == [-1, -2, -3]
```
```python
elems=ragged.constant([[1, 2, 3], [4, 5], [6, 7]])
mean = map_fn(tf.reduce_mean, elems)
# mean == [2, 4, 6]
```
```python
elems=ragged.constant([[1, 2, 3], [4, 5], [6, 7]], dtype=tf.int64)
out = map_fn(fn=lambda x: x+1, elems,
dtype=ragged.RaggedTensorType(type=tf.int64, ragged_rank=0))
# out = tf.ragged.constant([[2, 3, 4], [5, 6], [7, 8]])
```"
10294,range,tensorflow/tensorflow/python/ops/ragged/ragged_math_ops.py,43,function,"Returns a `RaggedTensor` containing the specified sequences of numbers.
Each row of the returned `RaggedTensor` contains a single sequence:
```python
ragged.range(starts, limits, deltas)[i] ==
tf.range(starts[i], limits[i], deltas[i])
```
If `start[i] < limits[i] and deltas[i] > 0`, then `output[i]` will be an
empty list. Similarly, if `start[i] > limits[i] and deltas[i] < 0`, then
`output[i]` will be an empty list. This behavior is consistent with the
Python `range` function, but differs from the `tf.range` op, which returns
an error for these cases.
Examples:
>>> tf.ragged.range([3, 5, 2]).to_list()
[[0, 1, 2], [0, 1, 2, 3, 4], [0, 1]]
>>> tf.ragged.range([0, 5, 8], [3, 3, 12]).to_list()
[[0, 1, 2], [], [8, 9, 10, 11]]
>>> tf.ragged.range([0, 5, 8], [3, 3, 12], 2).to_list()
[[0, 2], [], [8, 10]]
The input tensors `starts`, `limits`, and `deltas` may be scalars or vectors.
The vector inputs must all have the same size. Scalar inputs are broadcast
to match the size of the vector inputs.
Args:
starts: Vector or scalar `Tensor`. Specifies the first entry for each range
if `limits` is not `None`; otherwise, specifies the range limits, and the
first entries default to `0`.
limits: Vector or scalar `Tensor`. Specifies the exclusive upper limits for
each range.
deltas: Vector or scalar `Tensor`. Specifies the increment for each range.
Defaults to `1`.
dtype: The type of the elements of the resulting tensor. If not specified,
then a value is chosen based on the other args.
name: A name for the operation.
row_splits_dtype: `dtype` for the returned `RaggedTensor`'s `row_splits`
tensor. One of `tf.int32` or `tf.int64`.
Returns:
A `RaggedTensor` of type `dtype` with `ragged_rank=1`."
10295,segment_sum,tensorflow/tensorflow/python/ops/ragged/ragged_math_ops.py,262,function,
10296,segment_prod,tensorflow/tensorflow/python/ops/ragged/ragged_math_ops.py,271,function,
10297,segment_min,tensorflow/tensorflow/python/ops/ragged/ragged_math_ops.py,280,function,
10298,segment_max,tensorflow/tensorflow/python/ops/ragged/ragged_math_ops.py,289,function,
10299,segment_mean,tensorflow/tensorflow/python/ops/ragged/ragged_math_ops.py,298,function,"For docs, see: _RAGGED_SEGMENT_DOCSTRING."
10300,segment_sqrt_n,tensorflow/tensorflow/python/ops/ragged/ragged_math_ops.py,313,function,"For docs, see: _RAGGED_SEGMENT_DOCSTRING."
10301,ragged_reduce_aggregate,tensorflow/tensorflow/python/ops/ragged/ragged_math_ops.py,426,function,"Aggregates across axes of a RaggedTensor using the given `Tensor` ops.
Reduces `rt_input` along the dimensions given in `axis`. The rank of the
tensor is reduced by 1 for each entry in `axis`. If `axis` is not specified,
then all dimensions are reduced, and a scalar value is returned.
This op assumes that `reduce_op` and `unsorted_segment_op` are associative;
if not, then reducing multiple axes will return incorrect results. (In
particular, reducing multiple axes is currently implemented by reducing the
axes one at a time.)
Args:
reduce_op: The tensorflow `op` that should be used to reduce values in
uniform dimensions. Must have the same signature and basic behavior as
`reduce_sum`, `reduce_max`, etc.
unsorted_segment_op: The tensorflow `op` that should be used to combine
values in ragged dimensions. Must have the same signature and basic
behavior as `unsorted_segment_sum`, `unsorted_segment_max`, etc.
rt_input: A `Tensor` or `RaggedTensor` containing the values to be reduced.
axis: The axis or axes to reduce. May be `None` (to reduce all axes), an
`int` (to reduce a single axis), a `list` or `tuple` of `int` (to reduce a
given set of axes), or a `Tensor` with a constant value. Must be in the
range `[0, rt_input.rank)`.
keepdims: If true, retains reduced dimensions with length 1.
separator: An optional string. Defaults to None. The separator to use when
joining. The separator must not be set for non-string data types. (i.e.
if separator is not None then it uses string ops)
name: A name prefix for the returned tensor (optional).
Returns:
A `RaggedTensor` containing the reduced values. The returned tensor
has the same dtype as `data`, and its shape is given by removing the
dimensions specified in `axis` from `rt_input.shape`. The `ragged_rank`
of the returned tensor is given by substracting any ragged dimensions
specified in `axis` from `rt_input.ragged_rank`.
Raises:
ValueError: If `axis` contains a `Tensor` whose value is not constant."
10302,reduce_sum,tensorflow/tensorflow/python/ops/ragged/ragged_math_ops.py,550,function,"For docs, see: _RAGGED_REDUCE_DOCSTRING."
10303,reduce_prod,tensorflow/tensorflow/python/ops/ragged/ragged_math_ops.py,561,function,"For docs, see: _RAGGED_REDUCE_DOCSTRING."
10304,reduce_min,tensorflow/tensorflow/python/ops/ragged/ragged_math_ops.py,572,function,"For docs, see: _RAGGED_REDUCE_DOCSTRING."
10305,reduce_max,tensorflow/tensorflow/python/ops/ragged/ragged_math_ops.py,583,function,"For docs, see: _RAGGED_REDUCE_DOCSTRING."
10306,reduce_mean,tensorflow/tensorflow/python/ops/ragged/ragged_math_ops.py,594,function,"For docs, see: _RAGGED_REDUCE_DOCSTRING."
10307,reduce_all,tensorflow/tensorflow/python/ops/ragged/ragged_math_ops.py,618,function,"For docs, see: _RAGGED_REDUCE_DOCSTRING."
10308,reduce_any,tensorflow/tensorflow/python/ops/ragged/ragged_math_ops.py,626,function,"For docs, see: _RAGGED_REDUCE_DOCSTRING."
10309,mean,tensorflow/tensorflow/python/ops/ragged/ragged_reduce_op_test.py,38,function,
10310,RaggedRowLengthsOp,tensorflow/tensorflow/python/ops/ragged/ragged_row_lengths_op_test.py,31,class,
10311,prod,tensorflow/tensorflow/python/ops/ragged/ragged_segment_op_test.py,35,function,
10312,mean,tensorflow/tensorflow/python/ops/ragged/ragged_segment_op_test.py,43,function,
10313,sqrt_n,tensorflow/tensorflow/python/ops/ragged/ragged_segment_op_test.py,47,function,
10314,squeeze,tensorflow/tensorflow/python/ops/ragged/ragged_squeeze_op.py,31,function,"Ragged compatible squeeze.
If `input` is a `tf.Tensor`, then this calls `tf.squeeze`.
If `input` is a `tf.RaggedTensor`, then this operation takes `O(N)` time,
where `N` is the number of elements in the squeezed dimensions.
Args:
input: A potentially ragged tensor. The input to squeeze.
axis: An optional list of ints. Defaults to `None`. If the `input` is
ragged, it only squeezes the dimensions listed. It fails if `input` is
ragged and axis is []. If `input` is not ragged it calls tf.squeeze. Note
that it is an error to squeeze a dimension that is not 1. It must be in
the range of [-rank(input), rank(input)).
name: A name for the operation (optional).
Returns:
A potentially ragged tensor. Contains the same data as input,
but has one or more dimensions of size 1 removed."
10315,string_bytes_split,tensorflow/tensorflow/python/ops/ragged/ragged_string_ops.py,46,function,"Split string elements of `input` into bytes.
Examples:
>>> tf.strings.bytes_split('hello').numpy()
array([b'h', b'e', b'l', b'l', b'o'], dtype=object)
>>> tf.strings.bytes_split(['hello', '123'])
<tf.RaggedTensor [[b'h', b'e', b'l', b'l', b'o'], [b'1', b'2', b'3']]>
Note that this op splits strings into bytes, not unicode characters. To
split strings into unicode characters, use `tf.strings.unicode_split`.
See also: `tf.io.decode_raw`, `tf.strings.split`, `tf.strings.unicode_split`.
Args:
input: A string `Tensor` or `RaggedTensor`: the strings to split. Must
have a statically known rank (`N`).
name: A name for the operation (optional).
Returns:
A `RaggedTensor` of rank `N+1`: the bytes that make up the source strings."
10316,unicode_encode,tensorflow/tensorflow/python/ops/ragged/ragged_string_ops.py,94,function,"Encodes each sequence of Unicode code points in `input` into a string.
`result[i1...iN]` is the string formed by concatenating the Unicode
codepoints `input[1...iN, :]`, encoded using `output_encoding`.
Args:
input: An `N+1` dimensional potentially ragged integer tensor with shape
`[D1...DN, num_chars]`.
output_encoding: Unicode encoding that should be used to encode each
codepoint sequence. Can be `""UTF-8""`, `""UTF-16-BE""`, or `""UTF-32-BE""`.
errors: Specifies the response when an invalid codepoint is encountered
(optional). One of:
* `'replace'`: Replace invalid codepoint with the
`replacement_char`. (default)
* `'ignore'`: Skip invalid codepoints.
* `'strict'`: Raise an exception for any invalid codepoint.
replacement_char: The replacement character codepoint to be used in place of
any invalid input when `errors='replace'`. Any valid unicode codepoint may
be used. The default value is the default unicode replacement character
which is 0xFFFD (U+65533).
name: A name for the operation (optional).
Returns:
A `N` dimensional `string` tensor with shape `[D1...DN]`.
#### Example:
>>> input = tf.ragged.constant(
... [[71, 246, 246, 100, 110, 105, 103, 104, 116], [128522]])
>>> print(unicode_encode(input, 'UTF-8'))
tf.Tensor([b'G\xc3\xb6\xc3\xb6dnight' b'\xf0\x9f\x98\x8a'],
shape=(2,), dtype=string)"
10317,unicode_decode,tensorflow/tensorflow/python/ops/ragged/ragged_string_ops.py,192,function,"Decodes each string in `input` into a sequence of Unicode code points.
`result[i1...iN, j]` is the Unicode codepoint for the `j`th character in
`input[i1...iN]`, when decoded using `input_encoding`.
Args:
input: An `N` dimensional potentially ragged `string` tensor with shape
`[D1...DN]`. `N` must be statically known.
input_encoding: String name for the unicode encoding that should be used to
decode each string.
errors: Specifies the response when an input string can't be converted
using the indicated encoding. One of:
* `'strict'`: Raise an exception for any illegal substrings.
* `'replace'`: Replace illegal substrings with `replacement_char`.
* `'ignore'`: Skip illegal substrings.
replacement_char: The replacement codepoint to be used in place of invalid
substrings in `input` when `errors='replace'`; and in place of C0 control
characters in `input` when `replace_control_characters=True`.
replace_control_characters: Whether to replace the C0 control characters
`(U+0000 - U+001F)` with the `replacement_char`.
name: A name for the operation (optional).
Returns:
A `N+1` dimensional `int32` tensor with shape `[D1...DN, (num_chars)]`.
The returned tensor is a `tf.Tensor` if `input` is a scalar, or a
`tf.RaggedTensor` otherwise.
#### Example:
>>> input = [s.encode('utf8') for s in (u'G\xf6\xf6dnight', u'\U0001f60a')]
>>> tf.strings.unicode_decode(input, 'UTF-8').to_list()
[[71, 246, 246, 100, 110, 105, 103, 104, 116], [128522]]"
10318,unicode_decode_with_offsets,tensorflow/tensorflow/python/ops/ragged/ragged_string_ops.py,238,function,"Decodes each string into a sequence of code points with start offsets.
This op is similar to `tf.strings.decode(...)`, but it also returns the
start offset for each character in its respective string. This information
can be used to align the characters with the original byte sequence.
Returns a tuple `(codepoints, start_offsets)` where:
* `codepoints[i1...iN, j]` is the Unicode codepoint for the `j`th character
in `input[i1...iN]`, when decoded using `input_encoding`.
* `start_offsets[i1...iN, j]` is the start byte offset for the `j`th
character in `input[i1...iN]`, when decoded using `input_encoding`.
Args:
input: An `N` dimensional potentially ragged `string` tensor with shape
`[D1...DN]`. `N` must be statically known.
input_encoding: String name for the unicode encoding that should be used to
decode each string.
errors: Specifies the response when an input string can't be converted
using the indicated encoding. One of:
* `'strict'`: Raise an exception for any illegal substrings.
* `'replace'`: Replace illegal substrings with `replacement_char`.
* `'ignore'`: Skip illegal substrings.
replacement_char: The replacement codepoint to be used in place of invalid
substrings in `input` when `errors='replace'`; and in place of C0 control
characters in `input` when `replace_control_characters=True`.
replace_control_characters: Whether to replace the C0 control characters
`(U+0000 - U+001F)` with the `replacement_char`.
name: A name for the operation (optional).
Returns:
A tuple of `N+1` dimensional tensors `(codepoints, start_offsets)`.
* `codepoints` is an `int32` tensor with shape `[D1...DN, (num_chars)]`.
* `offsets` is an `int64` tensor with shape `[D1...DN, (num_chars)]`.
The returned tensors are `tf.Tensor`s if `input` is a scalar, or
`tf.RaggedTensor`s otherwise.
#### Example:
>>> input = [s.encode('utf8') for s in (u'G\xf6\xf6dnight', u'\U0001f60a')]
>>> result = tf.strings.unicode_decode_with_offsets(input, 'UTF-8')
>>> result[0].to_list() # codepoints
[[71, 246, 246, 100, 110, 105, 103, 104, 116], [128522]]
>>> result[1].to_list() # offsets
[[0, 1, 3, 5, 6, 7, 8, 9, 10], [0]]"
10319,unicode_split,tensorflow/tensorflow/python/ops/ragged/ragged_string_ops.py,300,function,"Splits each string in `input` into a sequence of Unicode code points.
`result[i1...iN, j]` is the substring of `input[i1...iN]` that encodes its
`j`th character, when decoded using `input_encoding`.
Args:
input: An `N` dimensional potentially ragged `string` tensor with shape
`[D1...DN]`. `N` must be statically known.
input_encoding: String name for the unicode encoding that should be used to
decode each string.
errors: Specifies the response when an input string can't be converted
using the indicated encoding. One of:
* `'strict'`: Raise an exception for any illegal substrings.
* `'replace'`: Replace illegal substrings with `replacement_char`.
* `'ignore'`: Skip illegal substrings.
replacement_char: The replacement codepoint to be used in place of invalid
substrings in `input` when `errors='replace'`.
name: A name for the operation (optional).
Returns:
A `N+1` dimensional `int32` tensor with shape `[D1...DN, (num_chars)]`.
The returned tensor is a `tf.Tensor` if `input` is a scalar, or a
`tf.RaggedTensor` otherwise.
#### Example:
>>> input = [s.encode('utf8') for s in (u'G\xf6\xf6dnight', u'\U0001f60a')]
>>> tf.strings.unicode_split(input, 'UTF-8').to_list()
[[b'G', b'\xc3\xb6', b'\xc3\xb6', b'd', b'n', b'i', b'g', b'h', b't'],
[b'\xf0\x9f\x98\x8a']]"
10320,unicode_split_with_offsets,tensorflow/tensorflow/python/ops/ragged/ragged_string_ops.py,348,function,"Splits each string into a sequence of code points with start offsets.
This op is similar to `tf.strings.decode(...)`, but it also returns the
start offset for each character in its respective string. This information
can be used to align the characters with the original byte sequence.
Returns a tuple `(chars, start_offsets)` where:
* `chars[i1...iN, j]` is the substring of `input[i1...iN]` that encodes its
`j`th character, when decoded using `input_encoding`.
* `start_offsets[i1...iN, j]` is the start byte offset for the `j`th
character in `input[i1...iN]`, when decoded using `input_encoding`.
Args:
input: An `N` dimensional potentially ragged `string` tensor with shape
`[D1...DN]`. `N` must be statically known.
input_encoding: String name for the unicode encoding that should be used to
decode each string.
errors: Specifies the response when an input string can't be converted
using the indicated encoding. One of:
* `'strict'`: Raise an exception for any illegal substrings.
* `'replace'`: Replace illegal substrings with `replacement_char`.
* `'ignore'`: Skip illegal substrings.
replacement_char: The replacement codepoint to be used in place of invalid
substrings in `input` when `errors='replace'`.
name: A name for the operation (optional).
Returns:
A tuple of `N+1` dimensional tensors `(codepoints, start_offsets)`.
* `codepoints` is an `int32` tensor with shape `[D1...DN, (num_chars)]`.
* `offsets` is an `int64` tensor with shape `[D1...DN, (num_chars)]`.
The returned tensors are `tf.Tensor`s if `input` is a scalar, or
`tf.RaggedTensor`s otherwise.
#### Example:
>>> input = [s.encode('utf8') for s in (u'G\xf6\xf6dnight', u'\U0001f60a')]
>>> result = tf.strings.unicode_split_with_offsets(input, 'UTF-8')
>>> result[0].to_list() # character substrings
[[b'G', b'\xc3\xb6', b'\xc3\xb6', b'd', b'n', b'i', b'g', b'h', b't'],
[b'\xf0\x9f\x98\x8a']]
>>> result[1].to_list() # offsets
[[0, 1, 3, 5, 6, 7, 8, 9, 10], [0]]"
10321,string_split_v2,tensorflow/tensorflow/python/ops/ragged/ragged_string_ops.py,472,function,"Split elements of `input` based on `sep` into a `RaggedTensor`.
Let N be the size of `input` (typically N will be the batch size). Split each
element of `input` based on `sep` and return a `RaggedTensor` containing the
split tokens. Empty tokens are ignored.
Example:
>>> tf.strings.split('hello world').numpy()
array([b'hello', b'world'], dtype=object)
>>> tf.strings.split(['hello world', 'a b c'])
<tf.RaggedTensor [[b'hello', b'world'], [b'a', b'b', b'c']]>
If `sep` is given, consecutive delimiters are not grouped together and are
deemed to delimit empty strings. For example, `input` of `""1<>2<><>3""` and
`sep` of `""<>""` returns `[""1"", ""2"", """", ""3""]`. If `sep` is None or an empty
string, consecutive whitespace are regarded as a single separator, and the
result will contain no empty strings at the start or end if the string has
leading or trailing whitespace.
Note that the above mentioned behavior matches python's str.split.
Args:
input: A string `Tensor` of rank `N`, the strings to split. If
`rank(input)` is not known statically, then it is assumed to be `1`.
sep: `0-D` string `Tensor`, the delimiter string.
maxsplit: An `int`. If `maxsplit > 0`, limit of the split of the result.
name: A name for the operation (optional).
Raises:
ValueError: If sep is not a string.
Returns:
A `RaggedTensor` of rank `N+1`, the strings split according to the
delimiter."
10322,string_split,tensorflow/tensorflow/python/ops/ragged/ragged_string_ops.py,537,function,"Split elements of `source` based on `delimiter`.
Let N be the size of `source` (typically N will be the batch size). Split each
element of `source` based on `delimiter` and return a `SparseTensor`
or `RaggedTensor` containing the split tokens. Empty tokens are ignored.
If `sep` is an empty string, each element of the `source` is split
into individual strings, each containing one byte. (This includes splitting
multibyte sequences of UTF-8.) If delimiter contains multiple bytes, it is
treated as a set of delimiters with each considered a potential split point.
Examples:
>>> print(tf.compat.v1.string_split(['hello world', 'a b c']))
SparseTensor(indices=tf.Tensor( [[0 0] [0 1] [1 0] [1 1] [1 2]], ...),
values=tf.Tensor([b'hello' b'world' b'a' b'b' b'c'], ...),
dense_shape=tf.Tensor([2 3], shape=(2,), dtype=int64))
>>> print(tf.compat.v1.string_split(['hello world', 'a b c'],
... result_type=""RaggedTensor""))
<tf.RaggedTensor [[b'hello', b'world'], [b'a', b'b', b'c']]>
Args:
source: `1-D` string `Tensor`, the strings to split.
sep: `0-D` string `Tensor`, the delimiter character, the string should
be length 0 or 1. Default is ' '.
skip_empty: A `bool`. If `True`, skip the empty strings from the result.
delimiter: deprecated alias for `sep`.
result_type: The tensor type for the result: one of `""RaggedTensor""` or
`""SparseTensor""`.
name: A name for the operation (optional).
Raises:
ValueError: If delimiter is not a string.
Returns:
A `SparseTensor` or `RaggedTensor` of rank `2`, the strings split according
to the delimiter. The first column of the indices corresponds to the row
in `source` and the second column corresponds to the index of the split
component in this row."
10323,strings_split_v1,tensorflow/tensorflow/python/ops/ragged/ragged_string_ops.py,599,function,"Split elements of `input` based on `sep`.
Let N be the size of `input` (typically N will be the batch size). Split each
element of `input` based on `sep` and return a `SparseTensor` or
`RaggedTensor` containing the split tokens. Empty tokens are ignored.
Examples:
>>> print(tf.compat.v1.strings.split(['hello world', 'a b c']))
SparseTensor(indices=tf.Tensor( [[0 0] [0 1] [1 0] [1 1] [1 2]], ...),
values=tf.Tensor([b'hello' b'world' b'a' b'b' b'c'], ...),
dense_shape=tf.Tensor([2 3], shape=(2,), dtype=int64))
>>> print(tf.compat.v1.strings.split(['hello world', 'a b c'],
... result_type=""RaggedTensor""))
<tf.RaggedTensor [[b'hello', b'world'], [b'a', b'b', b'c']]>
If `sep` is given, consecutive delimiters are not grouped together and are
deemed to delimit empty strings. For example, `input` of `""1<>2<><>3""` and
`sep` of `""<>""` returns `[""1"", ""2"", """", ""3""]`. If `sep` is None or an empty
string, consecutive whitespace are regarded as a single separator, and the
result will contain no empty strings at the start or end if the string has
leading or trailing whitespace.
Note that the above mentioned behavior matches python's str.split.
Args:
input: A string `Tensor` of rank `N`, the strings to split. If
`rank(input)` is not known statically, then it is assumed to be `1`.
sep: `0-D` string `Tensor`, the delimiter character.
maxsplit: An `int`. If `maxsplit > 0`, limit of the split of the result.
result_type: The tensor type for the result: one of `""RaggedTensor""` or
`""SparseTensor""`.
source: alias for ""input"" argument.
name: A name for the operation (optional).
Raises:
ValueError: If sep is not a string.
Returns:
A `SparseTensor` or `RaggedTensor` of rank `N+1`, the strings split
according to the delimiter."
10324,reduce_join,tensorflow/tensorflow/python/ops/ragged/ragged_string_ops.py,664,function,"For docs, see: _RAGGED_REDUCE_DOCSTRING."
10325,ngrams,tensorflow/tensorflow/python/ops/ragged/ragged_string_ops.py,673,function,"Create a tensor of n-grams based on `data`.
Creates a tensor of n-grams based on `data`. The n-grams are created by
joining windows of `width` adjacent strings from the inner axis of `data`
using `separator`.
The input data can be padded on both the start and end of the sequence, if
desired, using the `pad_values` argument. If set, `pad_values` should contain
either a tuple of strings or a single string; the 0th element of the tuple
will be used to pad the left side of the sequence and the 1st element of the
tuple will be used to pad the right side of the sequence. The `padding_width`
arg controls how many padding values are added to each side; it defaults to
`ngram_width-1`.
If this op is configured to not have padding, or if it is configured to add
padding with `padding_width` set to less than ngram_width-1, it is possible
that a sequence, or a sequence plus padding, is smaller than the ngram
width. In that case, no ngrams will be generated for that sequence. This can
be prevented by setting `preserve_short_sequences`, which will cause the op
to always generate at least one ngram per non-empty sequence.
Examples:
>>> tf.strings.ngrams([""A"", ""B"", ""C"", ""D""], 2).numpy()
array([b'A B', b'B C', b'C D'], dtype=object)
>>> tf.strings.ngrams([""TF"", ""and"", ""keras""], 1).numpy()
array([b'TF', b'and', b'keras'], dtype=object)
Args:
data: A Tensor or RaggedTensor containing the source data for the ngrams.
ngram_width: The width(s) of the ngrams to create. If this is a list or
tuple, the op will return ngrams of all specified arities in list order.
Values must be non-Tensor integers greater than 0.
separator: The separator string used between ngram elements. Must be a
string constant, not a Tensor.
pad_values: A tuple of (left_pad_value, right_pad_value), a single string,
or None. If None, no padding will be added; if a single string, then that
string will be used for both left and right padding. Values must be Python
strings.
padding_width: If set, `padding_width` pad values will be added to both
sides of each sequence. Defaults to `ngram_width`-1. Must be greater than
0. (Note that 1-grams are never padded, regardless of this value.)
preserve_short_sequences: If true, then ensure that at least one ngram is
generated for each input sequence. In particular, if an input sequence is
shorter than `min(ngram_width) + 2*pad_width`, then generate a single
ngram containing the entire sequence. If false, then no ngrams are
generated for these short input sequences.
name: The op name.
Returns:
A RaggedTensor of ngrams. If `data.shape=[D1...DN, S]`, then
`output.shape=[D1...DN, NUM_NGRAMS]`, where
`NUM_NGRAMS=S-ngram_width+1+2*padding_width`.
Raises:
TypeError: if `pad_values` is set to an invalid type.
ValueError: if `pad_values`, `padding_width`, or `ngram_width` is set to an
invalid value."
10326,string_format,tensorflow/tensorflow/python/ops/ragged/ragged_string_ops.py,826,function,Version of tf.strings.format that handles RaggedTensors.
10327,ragged_tensor_to_string,tensorflow/tensorflow/python/ops/ragged/ragged_string_ops.py,851,function,"Returns a scalar string tensor with the contents of a RaggedTensor.
Requires that `rt.shape.rank` is not `None`.
Note: this converts the entire `RaggedTensor` into a single string scalar.
If you want to convert individual elements, use `tf.strings.as_string(rt)`.
>>> rt1 = tf.ragged.constant([[1, 2, 3], [4, 5]])
>>> ragged_tensor_to_string(rt1).numpy()
b'[[1, 2, 3], [4, 5]]'
>>> rt2 = tf.ragged.constant([[['a'], ['b', 'c']], [['d', 'e', 'f'], []]])
>>> ragged_tensor_to_string(rt2).numpy()
b""[[['a'], ['b', 'c']], [['d', 'e', 'f'], []]]""
>>> rt3 = tf.ragged.constant([[1], [2, 3, 4, 5, 6], [], [], [7], [8, 9]])
>>> ragged_tensor_to_string(rt3, summarize=2).numpy()
b'[[1], [2, 3, ..., 5, 6], ..., [7], [8, 9]]'
Args:
rt: The RaggedTensor that should be converted to a string.
summarize: If specified, then only the first and last `summarize` elements
within each dimension are included in the string. If `-1` or `None`, then
all elements are included."
10328,RaggedTensor,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,59,class,"Represents a ragged tensor.
A `RaggedTensor` is a tensor with one or more *ragged dimensions*, which are
dimensions whose slices may have different lengths. For example, the inner
(column) dimension of `rt=[[3, 1, 4, 1], [], [5, 9, 2], [6], []]` is ragged,
since the column slices (`rt[0, :]`, ..., `rt[4, :]`) have different lengths.
Dimensions whose slices all have the same length are called *uniform
dimensions*. The outermost dimension of a `RaggedTensor` is always uniform,
since it consists of a single slice (and so there is no possibility for
differing slice lengths).
The total number of dimensions in a `RaggedTensor` is called its *rank*,
and the number of ragged dimensions in a `RaggedTensor` is called its
*ragged-rank*. A `RaggedTensor`'s ragged-rank is fixed at graph creation
time: it can't depend on the runtime values of `Tensor`s, and can't vary
dynamically for different session runs.
### Potentially Ragged Tensors
Many ops support both `Tensor`s and `RaggedTensor`s. The term ""potentially
ragged tensor"" may be used to refer to a tensor that might be either a
`Tensor` or a `RaggedTensor`. The ragged-rank of a `Tensor` is zero.
### Documenting RaggedTensor Shapes
When documenting the shape of a RaggedTensor, ragged dimensions can be
indicated by enclosing them in parentheses. For example, the shape of
a 3-D `RaggedTensor` that stores the fixed-size word embedding for each
word in a sentence, for each sentence in a batch, could be written as
`[num_sentences, (num_words), embedding_size]`. The parentheses around
`(num_words)` indicate that dimension is ragged, and that the length
of each element list in that dimension may vary for each item.
### Component Tensors
Internally, a `RaggedTensor` consists of a concatenated list of values that
are partitioned into variable-length rows. In particular, each `RaggedTensor`
consists of:
* A `values` tensor, which concatenates the variable-length rows into a
flattened list. For example, the `values` tensor for
`[[3, 1, 4, 1], [], [5, 9, 2], [6], []]` is `[3, 1, 4, 1, 5, 9, 2, 6]`.
* A `row_splits` vector, which indicates how those flattened values are
divided into rows. In particular, the values for row `rt[i]` are stored
in the slice `rt.values[rt.row_splits[i]:rt.row_splits[i+1]]`.
Example:
>>> print(tf.RaggedTensor.from_row_splits(
... values=[3, 1, 4, 1, 5, 9, 2, 6],
... row_splits=[0, 4, 4, 7, 8, 8]))
<tf.RaggedTensor [[3, 1, 4, 1], [], [5, 9, 2], [6], []]>
### Alternative Row-Partitioning Schemes
In addition to `row_splits`, ragged tensors provide support for five other
row-partitioning schemes:
* `row_lengths`: a vector with shape `[nrows]`, which specifies the length
of each row.
* `value_rowids` and `nrows`: `value_rowids` is a vector with shape
`[nvals]`, corresponding one-to-one with `values`, which specifies
each value's row index. In particular, the row `rt[row]` consists of the
values `rt.values[j]` where `value_rowids[j]==row`. `nrows` is an
integer scalar that specifies the number of rows in the
`RaggedTensor`. (`nrows` is used to indicate trailing empty rows.)
* `row_starts`: a vector with shape `[nrows]`, which specifies the start
offset of each row. Equivalent to `row_splits[:-1]`.
* `row_limits`: a vector with shape `[nrows]`, which specifies the stop
offset of each row. Equivalent to `row_splits[1:]`.
* `uniform_row_length`: A scalar tensor, specifying the length of every
row. This row-partitioning scheme may only be used if all rows have
the same length.
Example: The following ragged tensors are equivalent, and all represent the
nested list `[[3, 1, 4, 1], [], [5, 9, 2], [6], []]`.
>>> values = [3, 1, 4, 1, 5, 9, 2, 6]
>>> rt1 = RaggedTensor.from_row_splits(values, row_splits=[0, 4, 4, 7, 8, 8])
>>> rt2 = RaggedTensor.from_row_lengths(values, row_lengths=[4, 0, 3, 1, 0])
>>> rt3 = RaggedTensor.from_value_rowids(
... values, value_rowids=[0, 0, 0, 0, 2, 2, 2, 3], nrows=5)
>>> rt4 = RaggedTensor.from_row_starts(values, row_starts=[0, 4, 4, 7, 8])
>>> rt5 = RaggedTensor.from_row_limits(values, row_limits=[4, 4, 7, 8, 8])
### Multiple Ragged Dimensions
`RaggedTensor`s with multiple ragged dimensions can be defined by using
a nested `RaggedTensor` for the `values` tensor. Each nested `RaggedTensor`
adds a single ragged dimension.
>>> inner_rt = RaggedTensor.from_row_splits( # =rt1 from above
... values=[3, 1, 4, 1, 5, 9, 2, 6], row_splits=[0, 4, 4, 7, 8, 8])
>>> outer_rt = RaggedTensor.from_row_splits(
... values=inner_rt, row_splits=[0, 3, 3, 5])
>>> print(outer_rt.to_list())
[[[3, 1, 4, 1], [], [5, 9, 2]], [], [[6], []]]
>>> print(outer_rt.ragged_rank)
2
The factory function `RaggedTensor.from_nested_row_splits` may be used to
construct a `RaggedTensor` with multiple ragged dimensions directly, by
providing a list of `row_splits` tensors:
>>> RaggedTensor.from_nested_row_splits(
... flat_values=[3, 1, 4, 1, 5, 9, 2, 6],
... nested_row_splits=([0, 3, 3, 5], [0, 4, 4, 7, 8, 8])).to_list()
[[[3, 1, 4, 1], [], [5, 9, 2]], [], [[6], []]]
### Uniform Inner Dimensions
`RaggedTensor`s with uniform inner dimensions can be defined
by using a multidimensional `Tensor` for `values`.
>>> rt = RaggedTensor.from_row_splits(values=tf.ones([5, 3], tf.int32),
... row_splits=[0, 2, 5])
>>> print(rt.to_list())
[[[1, 1, 1], [1, 1, 1]],
[[1, 1, 1], [1, 1, 1], [1, 1, 1]]]
>>> print(rt.shape)
(2, None, 3)
### Uniform Outer Dimensions
`RaggedTensor`s with uniform outer dimensions can be defined by using
one or more `RaggedTensor` with a `uniform_row_length` row-partitioning
tensor. For example, a `RaggedTensor` with shape `[2, 2, None]` can be
constructed with this method from a `RaggedTensor` values with shape
`[4, None]`:
>>> values = tf.ragged.constant([[1, 2, 3], [4], [5, 6], [7, 8, 9, 10]])
>>> print(values.shape)
(4, None)
>>> rt6 = tf.RaggedTensor.from_uniform_row_length(values, 2)
>>> print(rt6)
<tf.RaggedTensor [[[1, 2, 3], [4]], [[5, 6], [7, 8, 9, 10]]]>
>>> print(rt6.shape)
(2, 2, None)
Note that `rt6` only contains one ragged dimension (the innermost
dimension). In contrast, if `from_row_splits` is used to construct a similar
`RaggedTensor`, then that `RaggedTensor` will have two ragged dimensions:
>>> rt7 = tf.RaggedTensor.from_row_splits(values, [0, 2, 4])
>>> print(rt7.shape)
(2, None, None)
Uniform and ragged outer dimensions may be interleaved, meaning that a
tensor with any combination of ragged and uniform dimensions may be created.
For example, a RaggedTensor `t4` with shape `[3, None, 4, 8, None, 2]` could
be constructed as follows:
```python
t0 = tf.zeros([1000, 2]) # Shape: [1000, 2]
t1 = RaggedTensor.from_row_lengths(t0, [...]) # [160, None, 2]
t2 = RaggedTensor.from_uniform_row_length(t1, 8) # [20, 8, None, 2]
t3 = RaggedTensor.from_uniform_row_length(t2, 4) # [5, 4, 8, None, 2]
t4 = RaggedTensor.from_row_lengths(t3, [...]) # [3, None, 4, 8, None, 2]
```"
10329,from_value_rowids,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,333,method,"Creates a `RaggedTensor` with rows partitioned by `value_rowids`.
The returned `RaggedTensor` corresponds with the python list defined by:
```python
result = [[values[i] for i in range(len(values)) if value_rowids[i] == row]
for row in range(nrows)]
```
Args:
values: A potentially ragged tensor with shape `[nvals, ...]`.
value_rowids: A 1-D integer tensor with shape `[nvals]`, which corresponds
one-to-one with `values`, and specifies each value's row index. Must be
nonnegative, and must be sorted in ascending order.
nrows: An integer scalar specifying the number of rows. This should be
specified if the `RaggedTensor` may containing empty training rows. Must
be greater than `value_rowids[-1]` (or zero if `value_rowids` is empty).
Defaults to `value_rowids[-1]` (or zero if `value_rowids` is empty).
name: A name prefix for the RaggedTensor (optional).
validate: If true, then use assertions to check that the arguments form
a valid `RaggedTensor`. Note: these assertions incur a runtime cost,
since they must be checked for each tensor value.
Returns:
A `RaggedTensor`. `result.rank = values.rank + 1`.
`result.ragged_rank = values.ragged_rank + 1`.
Raises:
ValueError: If `nrows` is incompatible with `value_rowids`.
#### Example:
>>> print(tf.RaggedTensor.from_value_rowids(
... values=[3, 1, 4, 1, 5, 9, 2, 6],
... value_rowids=[0, 0, 0, 0, 2, 2, 2, 3],
... nrows=5))
<tf.RaggedTensor [[3, 1, 4, 1], [], [5, 9, 2], [6], []]>"
10330,from_row_splits,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,391,method,"Creates a `RaggedTensor` with rows partitioned by `row_splits`.
The returned `RaggedTensor` corresponds with the python list defined by:
```python
result = [values[row_splits[i]:row_splits[i + 1]]
for i in range(len(row_splits) - 1)]
```
Args:
values: A potentially ragged tensor with shape `[nvals, ...]`.
row_splits: A 1-D integer tensor with shape `[nrows+1]`. Must not be
empty, and must be sorted in ascending order. `row_splits[0]` must be
zero and `row_splits[-1]` must be `nvals`.
name: A name prefix for the RaggedTensor (optional).
validate: If true, then use assertions to check that the arguments form
a valid `RaggedTensor`. Note: these assertions incur a runtime cost,
since they must be checked for each tensor value.
Returns:
A `RaggedTensor`. `result.rank = values.rank + 1`.
`result.ragged_rank = values.ragged_rank + 1`.
Raises:
ValueError: If `row_splits` is an empty list.
#### Example:
>>> print(tf.RaggedTensor.from_row_splits(
... values=[3, 1, 4, 1, 5, 9, 2, 6],
... row_splits=[0, 4, 4, 7, 8, 8]))
<tf.RaggedTensor [[3, 1, 4, 1], [], [5, 9, 2], [6], []]>"
10331,from_row_lengths,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,437,method,"Creates a `RaggedTensor` with rows partitioned by `row_lengths`.
The returned `RaggedTensor` corresponds with the python list defined by:
```python
result = [[values.pop(0) for i in range(length)]
for length in row_lengths]
```
Args:
values: A potentially ragged tensor with shape `[nvals, ...]`.
row_lengths: A 1-D integer tensor with shape `[nrows]`. Must be
nonnegative. `sum(row_lengths)` must be `nvals`.
name: A name prefix for the RaggedTensor (optional).
validate: If true, then use assertions to check that the arguments form
a valid `RaggedTensor`. Note: these assertions incur a runtime cost,
since they must be checked for each tensor value.
Returns:
A `RaggedTensor`. `result.rank = values.rank + 1`.
`result.ragged_rank = values.ragged_rank + 1`.
#### Example:
>>> print(tf.RaggedTensor.from_row_lengths(
... values=[3, 1, 4, 1, 5, 9, 2, 6],
... row_lengths=[4, 0, 3, 1, 0]))
<tf.RaggedTensor [[3, 1, 4, 1], [], [5, 9, 2], [6], []]>"
10332,from_row_starts,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,479,method,"Creates a `RaggedTensor` with rows partitioned by `row_starts`.
Equivalent to: `from_row_splits(values, concat([row_starts, nvals]))`.
Args:
values: A potentially ragged tensor with shape `[nvals, ...]`.
row_starts: A 1-D integer tensor with shape `[nrows]`. Must be
nonnegative and sorted in ascending order. If `nrows>0`, then
`row_starts[0]` must be zero.
name: A name prefix for the RaggedTensor (optional).
validate: If true, then use assertions to check that the arguments form
a valid `RaggedTensor`. Note: these assertions incur a runtime cost,
since they must be checked for each tensor value.
Returns:
A `RaggedTensor`. `result.rank = values.rank + 1`.
`result.ragged_rank = values.ragged_rank + 1`.
#### Example:
>>> print(tf.RaggedTensor.from_row_starts(
... values=[3, 1, 4, 1, 5, 9, 2, 6],
... row_starts=[0, 4, 4, 7, 8]))
<tf.RaggedTensor [[3, 1, 4, 1], [], [5, 9, 2], [6], []]>"
10333,from_row_limits,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,518,method,"Creates a `RaggedTensor` with rows partitioned by `row_limits`.
Equivalent to: `from_row_splits(values, concat([0, row_limits]))`.
Args:
values: A potentially ragged tensor with shape `[nvals, ...]`.
row_limits: A 1-D integer tensor with shape `[nrows]`. Must be sorted in
ascending order. If `nrows>0`, then `row_limits[-1]` must be `nvals`.
name: A name prefix for the RaggedTensor (optional).
validate: If true, then use assertions to check that the arguments form
a valid `RaggedTensor`. Note: these assertions incur a runtime cost,
since they must be checked for each tensor value.
Returns:
A `RaggedTensor`. `result.rank = values.rank + 1`.
`result.ragged_rank = values.ragged_rank + 1`.
#### Example:
>>> print(tf.RaggedTensor.from_row_limits(
... values=[3, 1, 4, 1, 5, 9, 2, 6],
... row_limits=[4, 4, 7, 8, 8]))
<tf.RaggedTensor [[3, 1, 4, 1], [], [5, 9, 2], [6], []]>"
10334,from_uniform_row_length,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,554,method,"Creates a `RaggedTensor` with rows partitioned by `uniform_row_length`.
This method can be used to create `RaggedTensor`s with multiple uniform
outer dimensions. For example, a `RaggedTensor` with shape `[2, 2, None]`
can be constructed with this method from a `RaggedTensor` values with shape
`[4, None]`:
>>> values = tf.ragged.constant([[1, 2, 3], [4], [5, 6], [7, 8, 9, 10]])
>>> print(values.shape)
(4, None)
>>> rt1 = tf.RaggedTensor.from_uniform_row_length(values, 2)
>>> print(rt1)
<tf.RaggedTensor [[[1, 2, 3], [4]], [[5, 6], [7, 8, 9, 10]]]>
>>> print(rt1.shape)
(2, 2, None)
Note that `rt1` only contains one ragged dimension (the innermost
dimension). In contrast, if `from_row_splits` is used to construct a similar
`RaggedTensor`, then that `RaggedTensor` will have two ragged dimensions:
>>> rt2 = tf.RaggedTensor.from_row_splits(values, [0, 2, 4])
>>> print(rt2.shape)
(2, None, None)
Args:
values: A potentially ragged tensor with shape `[nvals, ...]`.
uniform_row_length: A scalar integer tensor. Must be nonnegative. The
size of the outer axis of `values` must be evenly divisible by
`uniform_row_length`.
nrows: The number of rows in the constructed RaggedTensor. If not
specified, then it defaults to `nvals/uniform_row_length` (or `0` if
`uniform_row_length==0`). `nrows` only needs to be specified if
`uniform_row_length` might be zero. `uniform_row_length*nrows` must
be `nvals`.
validate: If true, then use assertions to check that the arguments form
a valid `RaggedTensor`. Note: these assertions incur a runtime cost,
since they must be checked for each tensor value.
name: A name prefix for the RaggedTensor (optional).
Returns:
A `RaggedTensor` that corresponds with the python list defined by:
```python
result = [[values.pop(0) for i in range(uniform_row_length)]
for _ in range(nrows)]
```
`result.rank = values.rank + 1`.
`result.ragged_rank = values.ragged_rank + 1`."
10335,from_nested_value_rowids,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,628,method,"Creates a `RaggedTensor` from a nested list of `value_rowids` tensors.
Equivalent to:
```python
result = flat_values
for (rowids, nrows) in reversed(zip(nested_value_rowids, nested_nrows)):
result = from_value_rowids(result, rowids, nrows)
```
Args:
flat_values: A potentially ragged tensor.
nested_value_rowids: A list of 1-D integer tensors. The `i`th tensor is
used as the `value_rowids` for the `i`th ragged dimension.
nested_nrows: A list of integer scalars. The `i`th scalar is used as the
`nrows` for the `i`th ragged dimension.
name: A name prefix for the RaggedTensor (optional).
validate: If true, then use assertions to check that the arguments form
a valid `RaggedTensor`. Note: these assertions incur a runtime cost,
since they must be checked for each tensor value.
Returns:
A `RaggedTensor` (or `flat_values` if `nested_value_rowids` is empty).
Raises:
ValueError: If `len(nested_values_rowids) != len(nested_nrows)`."
10336,from_nested_row_splits,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,684,method,"Creates a `RaggedTensor` from a nested list of `row_splits` tensors.
Equivalent to:
```python
result = flat_values
for row_splits in reversed(nested_row_splits):
result = from_row_splits(result, row_splits)
```
Args:
flat_values: A potentially ragged tensor.
nested_row_splits: A list of 1-D integer tensors. The `i`th tensor is
used as the `row_splits` for the `i`th ragged dimension.
name: A name prefix for the RaggedTensor (optional).
validate: If true, then use assertions to check that the arguments form
a valid `RaggedTensor`. Note: these assertions incur a runtime cost,
since they must be checked for each tensor value.
Returns:
A `RaggedTensor` (or `flat_values` if `nested_row_splits` is empty)."
10337,from_nested_row_lengths,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,723,method,"Creates a `RaggedTensor` from a nested list of `row_lengths` tensors.
Equivalent to:
```python
result = flat_values
for row_lengths in reversed(nested_row_lengths):
result = from_row_lengths(result, row_lengths)
```
Args:
flat_values: A potentially ragged tensor.
nested_row_lengths: A list of 1-D integer tensors. The `i`th tensor is
used as the `row_lengths` for the `i`th ragged dimension.
name: A name prefix for the RaggedTensor (optional).
validate: If true, then use assertions to check that the arguments form
a valid `RaggedTensor`. Note: these assertions incur a runtime cost,
since they must be checked for each tensor value.
Returns:
A `RaggedTensor` (or `flat_values` if `nested_row_lengths` is empty)."
10338,dtype,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,802,method,The `DType` of values in this tensor.
10339,shape,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,807,method,"The statically known shape of this ragged tensor.
Returns:
A `TensorShape` containing the statically known shape of this ragged
tensor. Ragged dimensions have a size of `None`.
Examples:
>>> tf.ragged.constant([[0], [1, 2]]).shape
TensorShape([2, None])
>>> tf.ragged.constant([[[0, 1]], [[1, 2], [3, 4]]], ragged_rank=1).shape
TensorShape([2, None, 2])"
10340,get_shape,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,828,method,"The statically known shape of this ragged tensor.
Returns:
A `TensorShape` containing the statically known shape of this ragged
tensor. Ragged dimensions have a size of `None`.
Alias for `shape` property.
Examples:
>>> tf.ragged.constant([[0], [1, 2]]).get_shape()
TensorShape([2, None])
>>> tf.ragged.constant(
... [[[0, 1]], [[1, 2], [3, 4]]], ragged_rank=1).get_shape()
TensorShape([2, None, 2])"
10341,ragged_rank,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,850,method,"The number of times the RaggedTensor's flat_values is partitioned.
Examples:
>>> values = tf.ragged.constant([[1, 2, 3], [4], [5, 6], [7, 8, 9, 10]])
>>> values.ragged_rank
1
>>> rt = tf.RaggedTensor.from_uniform_row_length(values, 2)
>>> rt.ragged_rank
2
Returns:
A Python `int` indicating the number of times the underlying `flat_values`
Tensor has been partitioned to add a new dimension.
I.e., `tf.rank(rt) = tf.rank(rt.flat_values) + rt.ragged_rank`."
10342,values,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,872,method,"The concatenated rows for this ragged tensor.
`rt.values` is a potentially ragged tensor formed by flattening the two
outermost dimensions of `rt` into a single dimension.
`rt.values.shape = [nvals] + rt.shape[2:]` (where `nvals` is the
number of items in the outer two dimensions of `rt`).
`rt.ragged_rank = self.ragged_rank - 1`
Returns:
A potentially ragged tensor.
#### Example:
>>> rt = tf.ragged.constant([[3, 1, 4, 1], [], [5, 9, 2], [6], []])
>>> print(rt.values)
tf.Tensor([3 1 4 1 5 9 2 6], shape=(8,), dtype=int32)"
10343,row_splits,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,907,method,"The row-split indices for this ragged tensor's `values`.
`rt.row_splits` specifies where the values for each row begin and end in
`rt.values`. In particular, the values for row `rt[i]` are stored in
the slice `rt.values[rt.row_splits[i]:rt.row_splits[i+1]]`.
Returns:
A 1-D integer `Tensor` with shape `[self.nrows+1]`.
The returned tensor is non-empty, and is sorted in ascending order.
`self.row_splits[0]` is zero, and `self.row_splits[-1]` is equal to
`self.values.shape[0]`.
#### Example:
>>> rt = tf.ragged.constant([[3, 1, 4, 1], [], [5, 9, 2], [6], []])
>>> print(rt.row_splits) # indices of row splits in rt.values
tf.Tensor([0 4 4 7 8 8], shape=(6,), dtype=int64)"
10344,uniform_row_length,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,930,method,"The length of each row in this ragged tensor, or None if rows are ragged.
>>> rt1 = tf.ragged.constant([[1, 2, 3], [4], [5, 6], [7, 8, 9, 10]])
>>> print(rt1.uniform_row_length) # rows are ragged.
None
>>> rt2 = tf.RaggedTensor.from_uniform_row_length(
... values=rt1, uniform_row_length=2)
>>> print(rt2)
<tf.RaggedTensor [[[1, 2, 3], [4]], [[5, 6], [7, 8, 9, 10]]]>
>>> print(rt2.uniform_row_length) # rows are not ragged (all have size 2).
tf.Tensor(2, shape=(), dtype=int64)
A RaggedTensor's rows are only considered to be uniform (i.e. non-ragged)
if it can be determined statically (at graph construction time) that the
rows all have the same length.
Returns:
A scalar integer `Tensor`, specifying the length of every row in this
ragged tensor (for ragged tensors whose rows are uniform); or `None`
(for ragged tensors whose rows are ragged)."
10345,flat_values,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,956,method,"The innermost `values` tensor for this ragged tensor.
Concretely, if `rt.values` is a `Tensor`, then `rt.flat_values` is
`rt.values`; otherwise, `rt.flat_values` is `rt.values.flat_values`.
Conceptually, `flat_values` is the tensor formed by flattening the
outermost dimension and all of the ragged dimensions into a single
dimension.
`rt.flat_values.shape = [nvals] + rt.shape[rt.ragged_rank + 1:]`
(where `nvals` is the number of items in the flattened dimensions).
Returns:
A `Tensor`.
#### Example:
>>> rt = tf.ragged.constant([[[3, 1, 4, 1], [], [5, 9, 2]], [], [[6], []]])
>>> print(rt.flat_values)
tf.Tensor([3 1 4 1 5 9 2 6], shape=(8,), dtype=int32)"
10346,nested_row_splits,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,985,method,"A tuple containing the row_splits for all ragged dimensions.
`rt.nested_row_splits` is a tuple containing the `row_splits` tensors for
all ragged dimensions in `rt`, ordered from outermost to innermost. In
particular, `rt.nested_row_splits = (rt.row_splits,) + value_splits` where:
* `value_splits = ()` if `rt.values` is a `Tensor`.
* `value_splits = rt.values.nested_row_splits` otherwise.
Returns:
A `tuple` of 1-D integer `Tensor`s.
#### Example:
>>> rt = tf.ragged.constant(
... [[[[3, 1, 4, 1], [], [5, 9, 2]], [], [[6], []]]])
>>> for i, splits in enumerate(rt.nested_row_splits):
... print('Splits for dimension %d: %s' % (i+1, splits.numpy()))
Splits for dimension 1: [0 3]
Splits for dimension 2: [0 3 3 5]
Splits for dimension 3: [0 4 4 7 8 8]"
10347,value_rowids,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,1016,method,"Returns the row indices for the `values` in this ragged tensor.
`rt.value_rowids()` corresponds one-to-one with the outermost dimension of
`rt.values`, and specifies the row containing each value. In particular,
the row `rt[row]` consists of the values `rt.values[j]` where
`rt.value_rowids()[j] == row`.
Args:
name: A name prefix for the returned tensor (optional).
Returns:
A 1-D integer `Tensor` with shape `self.values.shape[:1]`.
The returned tensor is nonnegative, and is sorted in ascending order.
#### Example:
>>> rt = tf.ragged.constant([[3, 1, 4, 1], [], [5, 9, 2], [6], []])
>>> print(rt.values)
tf.Tensor([3 1 4 1 5 9 2 6], shape=(8,), dtype=int32)
>>> print(rt.value_rowids()) # corresponds 1:1 with rt.values
tf.Tensor([0 0 0 0 2 2 2 3], shape=(8,), dtype=int64)"
10348,nested_value_rowids,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,1043,method,"Returns a tuple containing the value_rowids for all ragged dimensions.
`rt.nested_value_rowids` is a tuple containing the `value_rowids` tensors
for
all ragged dimensions in `rt`, ordered from outermost to innermost. In
particular, `rt.nested_value_rowids = (rt.value_rowids(),) + value_ids`
where:
* `value_ids = ()` if `rt.values` is a `Tensor`.
* `value_ids = rt.values.nested_value_rowids` otherwise.
Args:
name: A name prefix for the returned tensors (optional).
Returns:
A `tuple` of 1-D integer `Tensor`s.
#### Example:
>>> rt = tf.ragged.constant(
... [[[[3, 1, 4, 1], [], [5, 9, 2]], [], [[6], []]]])
>>> for i, ids in enumerate(rt.nested_value_rowids()):
... print('row ids for dimension %d: %s' % (i+1, ids.numpy()))
row ids for dimension 1: [0 0 0]
row ids for dimension 2: [0 0 0 2 2]
row ids for dimension 3: [0 0 0 0 2 2 2 3]"
10349,nrows,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,1080,method,"Returns the number of rows in this ragged tensor.
I.e., the size of the outermost dimension of the tensor.
Args:
out_type: `dtype` for the returned tensor. Defaults to
`self.row_splits.dtype`.
name: A name prefix for the returned tensor (optional).
Returns:
A scalar `Tensor` with dtype `out_type`.
#### Example:
>>> rt = tf.ragged.constant([[3, 1, 4, 1], [], [5, 9, 2], [6], []])
>>> print(rt.nrows()) # rt has 5 rows.
tf.Tensor(5, shape=(), dtype=int64)"
10350,row_starts,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,1103,method,"Returns the start indices for rows in this ragged tensor.
These indices specify where the values for each row begin in
`self.values`. `rt.row_starts()` is equal to `rt.row_splits[:-1]`.
Args:
name: A name prefix for the returned tensor (optional).
Returns:
A 1-D integer Tensor with shape `[nrows]`.
The returned tensor is nonnegative, and is sorted in ascending order.
#### Example:
>>> rt = tf.ragged.constant([[3, 1, 4, 1], [], [5, 9, 2], [6], []])
>>> print(rt.values)
tf.Tensor([3 1 4 1 5 9 2 6], shape=(8,), dtype=int32)
>>> print(rt.row_starts()) # indices of row starts in rt.values
tf.Tensor([0 4 4 7 8], shape=(5,), dtype=int64)"
10351,row_limits,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,1128,method,"Returns the limit indices for rows in this ragged tensor.
These indices specify where the values for each row end in
`self.values`. `rt.row_limits(self)` is equal to `rt.row_splits[:-1]`.
Args:
name: A name prefix for the returned tensor (optional).
Returns:
A 1-D integer Tensor with shape `[nrows]`.
The returned tensor is nonnegative, and is sorted in ascending order.
#### Example:
>>> rt = tf.ragged.constant([[3, 1, 4, 1], [], [5, 9, 2], [6], []])
>>> print(rt.values)
tf.Tensor([3 1 4 1 5 9 2 6], shape=(8,), dtype=int32)
>>> print(rt.row_limits()) # indices of row limits in rt.values
tf.Tensor([4 4 7 8 8], shape=(5,), dtype=int64)"
10352,row_lengths,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,1153,method,"Returns the lengths of the rows in this ragged tensor.
`rt.row_lengths()[i]` indicates the number of values in the
`i`th row of `rt`.
Args:
axis: An integer constant indicating the axis whose row lengths should be
returned.
name: A name prefix for the returned tensor (optional).
Returns:
A potentially ragged integer Tensor with shape `self.shape[:axis]`.
Raises:
ValueError: If `axis` is out of bounds.
#### Example:
>>> rt = tf.ragged.constant(
... [[[3, 1, 4], [1]], [], [[5, 9], [2]], [[6]], []])
>>> print(rt.row_lengths()) # lengths of rows in rt
tf.Tensor([2 0 2 1 0], shape=(5,), dtype=int64)
>>> print(rt.row_lengths(axis=2)) # lengths of axis=2 rows.
<tf.RaggedTensor [[3, 1], [], [2, 1], [1], []]>"
10353,nested_row_lengths,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,1202,method,"Returns a tuple containing the row_lengths for all ragged dimensions.
`rt.nested_row_lengths()` is a tuple containing the `row_lengths` tensors
for all ragged dimensions in `rt`, ordered from outermost to innermost.
Args:
name: A name prefix for the returned tensors (optional).
Returns:
A `tuple` of 1-D integer `Tensors`. The length of the tuple is equal to
`self.ragged_rank`."
10354,bounding_shape,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,1223,method,"Returns the tight bounding box shape for this `RaggedTensor`.
Args:
axis: An integer scalar or vector indicating which axes to return the
bounding box for. If not specified, then the full bounding box is
returned.
name: A name prefix for the returned tensor (optional).
out_type: `dtype` for the returned tensor. Defaults to
`self.row_splits.dtype`.
Returns:
An integer `Tensor` (`dtype=self.row_splits.dtype`). If `axis` is not
specified, then `output` is a vector with
`output.shape=[self.shape.ndims]`. If `axis` is a scalar, then the
`output` is a scalar. If `axis` is a vector, then `output` is a vector,
where `output[i]` is the bounding size for dimension `axis[i]`.
#### Example:
>>> rt = tf.ragged.constant([[1, 2, 3, 4], [5], [], [6, 7, 8, 9], [10]])
>>> rt.bounding_shape().numpy()
array([5, 4])"
10355,with_values,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,1284,method,"Returns a copy of `self` with `values` replaced by `new_value`.
Preserves cached row-partitioning tensors such as `self.cached_nrows` and
`self.cached_value_rowids` if they have values.
Args:
new_values: Potentially ragged tensor to use as the `values` for the
returned `RaggedTensor`. Must have `rank > 0`, and must have the same
number of rows as `self.values`.
Returns:
A `RaggedTensor`. `result.rank = 1 + new_values.rank`.
`result.ragged_rank = 1 + new_values.ragged_rank`"
10356,with_flat_values,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,1312,method,"Returns a copy of `self` with `flat_values` replaced by `new_value`.
Preserves cached row-partitioning tensors such as `self.cached_nrows` and
`self.cached_value_rowids` if they have values.
Args:
new_values: Potentially ragged tensor that should replace
`self.flat_values`. Must have `rank > 0`, and must have the same number
of rows as `self.flat_values`.
Returns:
A `RaggedTensor`.
`result.rank = self.ragged_rank + new_values.rank`.
`result.ragged_rank = self.ragged_rank + new_values.ragged_rank`."
10357,with_row_splits_dtype,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,1333,method,"Returns a copy of this RaggedTensor with the given `row_splits` dtype.
For RaggedTensors with multiple ragged dimensions, the `row_splits` for all
nested `RaggedTensor` objects are cast to the given dtype.
Args:
dtype: The dtype for `row_splits`. One of `tf.int32` or `tf.int64`.
Returns:
A copy of this RaggedTensor, with the `row_splits` cast to the given
type."
10358,merge_dims,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,1363,method,"Merges outer_axis...inner_axis into a single dimension.
Returns a copy of this RaggedTensor with the specified range of dimensions
flattened into a single dimension, with elements in row-major order.
#### Examples:
>>> rt = tf.ragged.constant([[[1, 2], [3]], [[4, 5, 6]]])
>>> print(rt.merge_dims(0, 1))
<tf.RaggedTensor [[1, 2], [3], [4, 5, 6]]>
>>> print(rt.merge_dims(1, 2))
<tf.RaggedTensor [[1, 2, 3], [4, 5, 6]]>
>>> print(rt.merge_dims(0, 2))
tf.Tensor([1 2 3 4 5 6], shape=(6,), dtype=int32)
To mimic the behavior of `np.flatten` (which flattens all dimensions), use
`rt.merge_dims(0, -1). To mimic the behavior of `tf.layers.Flatten` (which
flattens all dimensions except the outermost batch dimension), use
`rt.merge_dims(1, -1)`.
Args:
outer_axis: `int`: The first dimension in the range of dimensions to
merge. May be negative if `self.shape.rank` is statically known.
inner_axis: `int`: The last dimension in the range of dimensions to merge.
May be negative if `self.shape.rank` is statically known.
Returns:
A copy of this tensor, with the specified dimensions merged into a
single dimension. The shape of the returned tensor will be
`self.shape[:outer_axis] + [N] + self.shape[inner_axis + 1:]`, where `N`
is the total number of slices in the merged dimensions."
10359,from_tensor,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,1470,method,"Converts a `tf.Tensor` into a `RaggedTensor`.
The set of absent/default values may be specified using a vector of lengths
or a padding value (but not both). If `lengths` is specified, then the
output tensor will satisfy `output[row] = tensor[row][:lengths[row]]`. If
'lengths' is a list of lists or tuple of lists, those lists will be used
as nested row lengths. If `padding` is specified, then any row *suffix*
consisting entirely of `padding` will be excluded from the returned
`RaggedTensor`. If neither `lengths` nor `padding` is specified, then the
returned `RaggedTensor` will have no absent/default values.
Examples:
>>> dt = tf.constant([[5, 7, 0], [0, 3, 0], [6, 0, 0]])
>>> tf.RaggedTensor.from_tensor(dt)
<tf.RaggedTensor [[5, 7, 0], [0, 3, 0], [6, 0, 0]]>
>>> tf.RaggedTensor.from_tensor(dt, lengths=[1, 0, 3])
<tf.RaggedTensor [[5], [], [6, 0, 0]]>
>>> tf.RaggedTensor.from_tensor(dt, padding=0)
<tf.RaggedTensor [[5, 7], [0, 3], [6]]>
>>> dt = tf.constant([[[5, 0], [7, 0], [0, 0]],
... [[0, 0], [3, 0], [0, 0]],
... [[6, 0], [0, 0], [0, 0]]])
>>> tf.RaggedTensor.from_tensor(dt, lengths=([2, 0, 3], [1, 1, 2, 0, 1]))
<tf.RaggedTensor [[[5], [7]], [], [[6, 0], [], [0]]]>
Args:
tensor: The `Tensor` to convert. Must have rank `ragged_rank + 1` or
higher.
lengths: An optional set of row lengths, specified using a 1-D integer
`Tensor` whose length is equal to `tensor.shape[0]` (the number of rows
in `tensor`). If specified, then `output[row]` will contain
`tensor[row][:lengths[row]]`. Negative lengths are treated as zero. You
may optionally pass a list or tuple of lengths to this argument, which
will be used as nested row lengths to construct a ragged tensor with
multiple ragged dimensions.
padding: An optional padding value. If specified, then any row suffix
consisting entirely of `padding` will be excluded from the returned
RaggedTensor. `padding` is a `Tensor` with the same dtype as `tensor`
and with `shape=tensor.shape[ragged_rank + 1:]`.
ragged_rank: Integer specifying the ragged rank for the returned
`RaggedTensor`. Must be greater than zero.
name: A name prefix for the returned tensors (optional).
row_splits_dtype: `dtype` for the returned `RaggedTensor`'s `row_splits`
tensor. One of `tf.int32` or `tf.int64`.
Returns:
A `RaggedTensor` with the specified `ragged_rank`. The shape of the
returned ragged tensor is compatible with the shape of `tensor`.
Raises:
ValueError: If both `lengths` and `padding` are specified."
10360,to_tensor,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,1675,method,"Converts this `RaggedTensor` into a `tf.Tensor`.
If `shape` is specified, then the result is padded and/or truncated to
the specified shape.
Examples:
>>> rt = tf.ragged.constant([[9, 8, 7], [], [6, 5], [4]])
>>> print(rt.to_tensor())
tf.Tensor(
[[9 8 7] [0 0 0] [6 5 0] [4 0 0]], shape=(4, 3), dtype=int32)
>>> print(rt.to_tensor(shape=[5, 2]))
tf.Tensor(
[[9 8] [0 0] [6 5] [4 0] [0 0]], shape=(5, 2), dtype=int32)
Args:
default_value: Value to set for indices not specified in `self`. Defaults
to zero. `default_value` must be broadcastable to
`self.shape[self.ragged_rank + 1:]`.
name: A name prefix for the returned tensors (optional).
shape: The shape of the resulting dense tensor. In particular,
`result.shape[i]` is `shape[i]` (if `shape[i]` is not None), or
`self.bounding_shape(i)` (otherwise).`shape.rank` must be `None` or
equal to `self.rank`.
Returns:
A `Tensor` with shape `ragged.bounding_shape(self)` and the
values specified by the non-empty values in `self`. Empty values are
assigned `default_value`."
10361,from_sparse,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,1742,method,"Converts a 2D `tf.sparse.SparseTensor` to a `RaggedTensor`.
Each row of the `output` `RaggedTensor` will contain the explicit values
from the same row in `st_input`. `st_input` must be ragged-right. If not
it is not ragged-right, then an error will be generated.
Example:
>>> indices = [[0, 0], [0, 1], [0, 2], [1, 0], [3, 0]]
>>> st = tf.sparse.SparseTensor(indices=indices,
... values=[1, 2, 3, 4, 5],
... dense_shape=[4, 3])
>>> tf.RaggedTensor.from_sparse(st).to_list()
[[1, 2, 3], [4], [], [5]]
Currently, only two-dimensional `SparseTensors` are supported.
Args:
st_input: The sparse tensor to convert. Must have rank 2.
name: A name prefix for the returned tensors (optional).
row_splits_dtype: `dtype` for the returned `RaggedTensor`'s `row_splits`
tensor. One of `tf.int32` or `tf.int64`.
Returns:
A `RaggedTensor` with the same values as `st_input`.
`output.ragged_rank = rank(st_input) - 1`.
`output.shape = [st_input.dense_shape[0], None]`.
Raises:
ValueError: If the number of dimensions in `st_input` is not known
statically, or is not two."
10362,to_sparse,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,1804,method,"Converts this `RaggedTensor` into a `tf.sparse.SparseTensor`.
Example:
>>> rt = tf.ragged.constant([[1, 2, 3], [4], [], [5, 6]])
>>> print(rt.to_sparse())
SparseTensor(indices=tf.Tensor(
[[0 0] [0 1] [0 2] [1 0] [3 0] [3 1]],
shape=(6, 2), dtype=int64),
values=tf.Tensor([1 2 3 4 5 6], shape=(6,), dtype=int32),
dense_shape=tf.Tensor([4 3], shape=(2,), dtype=int64))
Args:
name: A name prefix for the returned tensors (optional).
Returns:
A SparseTensor with the same values as `self`."
10363,numpy,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,1944,method,"Returns a numpy `array` with the values for this `RaggedTensor`.
Requires that this `RaggedTensor` was constructed in eager execution mode.
Ragged dimensions are encoded using numpy `arrays` with `dtype=object` and
`rank=1`, where each element is a single row.
#### Examples
In the following example, the value returned by `RaggedTensor.numpy()`
contains three numpy `array` objects: one for each row (with `rank=1` and
`dtype=int64`), and one to combine them (with `rank=1` and `dtype=object`):
>>> tf.ragged.constant([[1, 2, 3], [4, 5]], dtype=tf.int64).numpy()
array([array([1, 2, 3]), array([4, 5])], dtype=object)
Uniform dimensions are encoded using multidimensional numpy `array`s. In
the following example, the value returned by `RaggedTensor.numpy()` contains
a single numpy `array` object, with `rank=2` and `dtype=int64`:
>>> tf.ragged.constant([[1, 2, 3], [4, 5, 6]], dtype=tf.int64).numpy()
array([[1, 2, 3], [4, 5, 6]])
Returns:
A numpy `array`."
10364,to_list,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,1984,method,"Returns a nested Python `list` with the values for this `RaggedTensor`.
Requires that `rt` was constructed in eager execution mode.
Returns:
A nested Python `list`."
10365,consumers,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,2088,method,
10366,stub,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,2022,method,
10367,is_ragged,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,2092,function,Returns true if `value` is a ragged tensor or ragged tensor value.
10368,match_row_splits_dtypes,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,2098,function,"Return a copy of `tensors` with row_splits all having the same dtype.
Args:
*tensors: A list of Tensors or RaggedTensors.
**kwargs: If 'return_dtype=True', then return a tuple (dtype, tensors),
where `dtype` is the data type used by row-splits, and `tensors` is the
converted list of `Tensors` and `RaggedTensors`.
Returns:
The converted list of `Tensors` and `RaggedTensors`."
10369,RaggedTensorSpec,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,2149,class,Type specification for a `tf.RaggedTensor`.
10370,dtype,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,2155,method,"The `tf.dtypes.DType` specified by this type for the RaggedTensor.
Examples:
>>> rt = tf.ragged.constant([[""a""], [""b"", ""c""]], dtype=tf.string)
>>> tf.type_spec_from_value(rt).dtype
tf.string
Returns:
A `tf.dtypes.DType` of the values in the RaggedTensor."
10371,shape,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,2170,method,"The statically known shape of the RaggedTensor.
Examples:
>>> rt = tf.ragged.constant([[0], [1, 2]])
>>> tf.type_spec_from_value(rt).shape
TensorShape([2, None])
>>> rt = tf.ragged.constant([[[0, 1]], [[1, 2], [3, 4]]], ragged_rank=1)
>>> tf.type_spec_from_value(rt).shape
TensorShape([2, None, 2])
Returns:
A `tf.TensorShape` containing the statically known shape of the
RaggedTensor. Ragged dimensions have a size of `None`."
10372,ragged_rank,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,2190,method,"The number of times the RaggedTensor's flat_values is partitioned.
Defaults to `shape.ndims - 1`.
Examples:
>>> values = tf.ragged.constant([[1, 2, 3], [4], [5, 6], [7, 8, 9, 10]])
>>> tf.type_spec_from_value(values).ragged_rank
1
>>> rt1 = tf.RaggedTensor.from_uniform_row_length(values, 2)
>>> tf.type_spec_from_value(rt1).ragged_rank
2
Returns:
A Python `int` indicating the number of times the underlying `flat_values`
Tensor has been partitioned to add a new dimension.
I.e., `tf.rank(rt) = tf.rank(rt.flat_values) + rt.ragged_rank`."
10373,row_splits_dtype,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,2213,method,"The `tf.dtypes.DType` of the the RaggedTensor's `row_splits`.
Examples:
>>> rt = tf.ragged.constant([[1, 2, 3], [4]], row_splits_dtype=tf.int64)
>>> tf.type_spec_from_value(rt).row_splits_dtype
tf.int64
Returns:
A `tf.dtypes.DType` for the RaggedTensor's `row_splits` tensor. One
of `tf.int32` or `tf.int64`."
10374,value_type,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,2229,method,
10375,is_compatible_with,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,2266,method,
10376,from_value,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,2397,method,
10377,convert_to_tensor_or_ragged_tensor,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,2412,function,"Converts value to a `RaggedTensor` or `Tensor`.
* If `value` is a `RaggedTensor`, then return it as-is.
* If `value` is a `RaggedTensorValue`, return a corresponding constant
`RaggedTensor`.
* Otherwise, use `convert_to_tensor` to convert `value` to a `Tensor`.
Args:
value: A `RaggedTensor`, a `RaggedTensorValue`, or an object whose type has
a registered `Tensor` conversion function.
dtype: Optional element type for the returned tensor. If missing the type
is inferred from the type of `value`.
preferred_dtype: Optional element type for the returned tensor, used when
dtype is None. This argument has no effect if `value` is already a
tensor, or when conversion is not possible.
name: Optional name to use if a new `Tensor` is created.
Returns:
A `Tensor` or `RaggedTensor`."
10378,RaggedTensorType,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,2490,class,"Encoding of a static type for a `RaggedTensor`.
Use this type to express/declare that an output must have the type of
`RaggedTensor`."
10379,merge_dims,tensorflow/tensorflow/python/ops/ragged/ragged_tensor.py,2601,function,"Merges value[outer_axis...inner_axis] into a single dimension.
See `RaggedTensor.merge_dims()` for more details. This helper differs from
`RaggedTensor.merge_dims()` in that `value` may be a dense or ragged tensor.
Args:
value: A `RaggedTensor` or `Tensor`
outer_axis: `int`
inner_axis: `int`
Returns:
A flattened `RaggedTensor` or `Tensor`."
10380,RaggedTensorBoundingShapeOp,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_bounding_shape_op_test.py,31,class,
10381,RaggedTensorDynamicShape,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_shape.py,35,class,"A collection of tensors encoding the shape of a potentially ragged tensor.
Each `RaggedTensorDynamicShape` consists of an ordered list of dimension
sizes. There are two dimension types:
* ""Uniform dimensions"" are dimensions where all slices have the same
length. `RaggedTensorDynamicShape` records the size of each uniform
dimension using a single scalar integer.
* ""Ragged dimensions"" are dimensions whose slices may have different
lengths. `RaggedTensorDynamicShape` records the size of each ragged
dimension using an integer vector containing the slice lengths for all
the slices across that dimension.
Furthermore, there are two ways a dimension might be encoded:
* ""Partitioned dimensions"" are dimensions that are encoded using a
`RaggedTensor`'s `nested_row_splits`. The outermostmost partitioned
dimension must be uniform, and the innermost partitioned dimension must
be ragged.
* ""Inner dimensions"" are dimensions that are encoded using a
`RaggedTensor`'s `flat_values`. Inner dimensions are always uniform.
The sizes of partitioned dimensions are recorded using `partitioned_dim_sizes`
and `inner_dim_sizes`:
* `partitioned_dim_sizes` is a list of tensors (one for each partitioned
dimension).
* For uniform dimensions, the tensor is an integer scalar specifying the
size of all slices across that dimension.
* For ragged dimensions, the tensor is an integer vector specifying the
size of each slice across that dimension.
* `inner_dim_sizes` is a single integer vector, where each element
specifies the size of a single inner dimension.
Examples:
Tensor | Ragged | Partitioned Dim Sizes | Inner Dim
: Rank : : Sizes
------------------------------ | ------ | ---------------------- | ----------
`[[1, 2, 3], [4, 5, 6]]` | 0 | | `2, 3`
`[[1, 2], [], [3, 4, 5]]` | 1 | `3, (2, 0, 3)` |
`[[[1, 2], [3, 4]], [[5, 6]]]` | 1 | `2, (2, 1)` | 2
`[[[1, 2], [3]], [[4, 5]]]` | 2 | `2, (2, 1), (2, 1, 2)` |"
10382,from_dim_sizes,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_shape.py,152,method,"Constructs a ragged shape from a list of dimension sizes.
This list contains a single tensor for each dimension, where the tensor
is a scalar if the dimension is uniform, or a vector if the dimension is
ragged.
Args:
dim_sizes: List of int32 or int64 scalars or vectors.
Returns:
A RaggedTensorDynamicShape."
10383,from_tensor,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_shape.py,181,method,Constructs a ragged shape for a potentially ragged tensor.
10384,dimension_size,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_shape.py,195,method,Returns the size of slices across the specified dimension.
10385,is_ragged,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_shape.py,205,method,Returns true if the indicated dimension is ragged.
10386,rank,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_shape.py,219,method,"The number of dimensions in this shape, or None if unknown."
10387,partitioned_dim_sizes,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_shape.py,228,method,"The partitioned dimension sizes for this shape.
Returns:
A `list` of 0-D or 1-D integer `Tensor`."
10388,inner_dim_sizes,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_shape.py,237,method,"The inner dimension sizes for this shape.
Returns:
A 1-D integer `Tensor`."
10389,num_partitioned_dimensions,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_shape.py,246,method,The number of partitioned dimensions in this shape.
10390,num_inner_dimensions,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_shape.py,251,method,"The number of inner dimensions, or `None` if not statically known."
10391,dim_size_dtype,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_shape.py,256,method,DType used by this shape for dimension sizes.
10392,broadcast_to_rank,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_shape.py,260,method,"Adds leading size-1 dimensions to broadcast `self` to the given rank.
E.g., if `shape1` is `[3, (D2), 4]`, then `shape1.broadcast_to_rank(5)`
is `[1, 1, 3, (D2), 4]`.
Args:
rank: The rank for the returned shape.
Returns:
A RaggedTensorDynamicShape with `rank` dimensions, whose inner dimensions
have the same size as `self` and whose outer dimensions have size `1`.
Raises:
ValueError: If `self.rank` is unknown or greater than `rank`."
10393,broadcast_dimension,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_shape.py,294,method,"Returns a shape that is broadcast-compatible with self & lengths.
* If dimension[axis] is uniform and lengths is a scalar, the check
that either lengths==1 or axis==1 or lengths==axis, and tile
dimension[axis] with tf.where(lengths==axis, 1, axis) repeats.
* If dimension[axis] is uniform and lengths is a vector, then check
that dimension[axis]==1, and raggedly tile dimension[axis] with
lengths repeats. (we can skip tiling if we statically know that
slice_lengths == 1??)
* If dimension[axis] is ragged and lengths is a scalar, then check
that lengths==1.
* If dimension[axis] is ragged and lengths is a vector, then check
that self.dimension_size(axis) == lengths.
Args:
axis: `int`. The dimension to broadcast.
lengths: 0-D or 1-D integer `Tensor`.
Returns:
A `RaggedTensorDynamicShape`."
10394,num_slices_in_dimension,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_shape.py,374,method,Returns the total number of slices across the indicated dimension.
10395,with_dim_size_dtype,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_shape.py,434,method,
10396,broadcast_dynamic_shape,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_shape.py,444,function,"Returns the shape formed by broadcasting two shapes to be compatible.
Args:
shape_x: A `RaggedTensorDynamicShape`
shape_y: A `RaggedTensorDynamicShape`
Returns:
A `RaggedTensorDynamicShape`.
Raises:
ValueError: If `shape_x` and `shape_y` are not broadcast-compatible."
10397,broadcast_to,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_shape.py,476,function,"Broadcasts a potentially ragged tensor to a ragged shape.
Tiles `rt_input` as necessary to match the given shape.
Behavior is undefined if `rt_input` is not broadcast-compatible with `shape`.
Args:
rt_input: The potentially ragged tensor to broadcast.
shape: A `RaggedTensorDynamicShape`
broadcast_inner_dimensions: If false, then inner dimensions will not be
tiled.
Returns:
A potentially ragged tensor whose values are taken from
`rt_input`, and whose shape matches `shape`."
10398,int32array,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_test.py,47,function,
10399,RaggedTensorValue,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_value.py,27,class,"Represents the value of a `RaggedTensor`.
Warning: `RaggedTensorValue` should only be used in graph mode; in
eager mode, the `tf.RaggedTensor` class contains its value directly.
See `tf.RaggedTensor` for a description of ragged tensors."
10400,flat_values,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_value.py,66,method,The innermost `values` array for this ragged tensor value.
10401,nested_row_splits,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_value.py,74,method,The row_splits for all ragged dimensions in this ragged tensor value.
10402,ragged_rank,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_value.py,84,method,The number of ragged dimensions in this ragged tensor value.
10403,shape,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_value.py,90,method,A tuple indicating the shape of this RaggedTensorValue.
10404,to_list,tensorflow/tensorflow/python/ops/ragged/ragged_tensor_value.py,101,method,Returns this ragged tensor value as a nested Python list.
10405,make_placeholder,tensorflow/tensorflow/python/ops/ragged/ragged_to_tensor_op_test.py,46,function,
10406,rebuild_ragged_tensor_with_value_rowids,tensorflow/tensorflow/python/ops/ragged/ragged_to_tensor_op_test.py,50,function,"Returns a copy of `rt`, built using `from_value_rowids`.
This ensures that RaggedTensor._cached_value_rowids is populated, which
triggers a different code-path for converting ragged tensors to tensors.
If `feed_dict` and `sess` are specified, then build the new `RaggedTensor`
using placeholder tensors, and populate a feed dictionary that can be used
to feed the placeholders.
Args:
rt: The RaggedTensor to copy.
feed_dict: If specified, then build the new `RaggedTensor` using
placeholders, and populate this dict with entries to feed those
placeholders.
sess: A session used to evaluate tensors; required if feed_dict is
specified.
Returns:
A copy of `rt`, built using `from_value_rowids`."
10407,RaggedToDenseBenchmark,tensorflow/tensorflow/python/ops/ragged/ragged_to_tensor_op_test.py,725,class,
10408,run_benchmark,tensorflow/tensorflow/python/ops/ragged/ragged_to_tensor_op_test.py,747,method,"Run a benchmark with the specified configuration parameters.
Args:
shape: Bounding box for the input ragged tensor.
ragged_rank: Ragged rank for the input ragged tensor. Defaults to
`len(shape)-1`.
dtype: Data type for the input ragged tensor.
fill: How full each dimension should be (0-1). Corresponds 1:1 with
`shape`. Defaults to 0.8 for each dimension.
default_shape: Shape for the default (padding) value.
output_shape: Output shape -- ragged tensor will be padded or cropped to
this shape.
min_iters: Minimum iterations for benchmark."
10409,benchmark_ragged_to_dense,tensorflow/tensorflow/python/ops/ragged/ragged_to_tensor_op_test.py,842,method,
10410,assert_splits_match,tensorflow/tensorflow/python/ops/ragged/ragged_util.py,31,function,"Checks that the given splits lists are identical.
Performs static tests to ensure that the given splits lists are identical,
and returns a list of control dependency op tensors that check that they are
fully identical.
Args:
nested_splits_lists: A list of nested_splits_lists, where each split_list is
a list of `splits` tensors from a `RaggedTensor`, ordered from outermost
ragged dimension to innermost ragged dimension.
Returns:
A list of control dependency op tensors.
Raises:
ValueError: If the splits are not identical."
10411,lengths_to_splits,tensorflow/tensorflow/python/ops/ragged/ragged_util.py,65,function,Returns splits corresponding to the given lengths.
10412,repeat_ranges,tensorflow/tensorflow/python/ops/ragged/ragged_util.py,70,function,"Repeats each range of `params` (as specified by `splits`) `repeats` times.
Let the `i`th range of `params` be defined as
`params[splits[i]:splits[i + 1]]`. Then this function returns a tensor
containing range 0 repeated `repeats[0]` times, followed by range 1 repeated
`repeats[1]`, ..., followed by the last range repeated `repeats[-1]` times.
Args:
params: The `Tensor` whose values should be repeated.
splits: A splits tensor indicating the ranges of `params` that should be
repeated.
repeats: The number of times each range should be repeated. Supports
broadcasting from a scalar value.
Returns:
A `Tensor` with the same rank and type as `params`.
#### Example:
>>> print(repeat_ranges(
... params=tf.constant(['a', 'b', 'c']),
... splits=tf.constant([0, 2, 3]),
... repeats=tf.constant(3)))
tf.Tensor([b'a' b'b' b'a' b'b' b'a' b'b' b'c' b'c' b'c'],
shape=(9,), dtype=string)"
10413,where,tensorflow/tensorflow/python/ops/ragged/ragged_where_op.py,30,function,"Return the elements, either from `x` or `y`, depending on the `condition`.
: If both `x` and `y` are `None`:
Returns the coordinates of true elements of `condition`. The coordinates
are returned in a 2-D tensor with shape
`[num_true_values, dim_size(condition)]`, where `result[i]` is the
coordinates of the `i`th true value (in row-major order).
: If both `x` and `y` are non-`None`:
Returns a tensor formed by selecting values from `x` where condition is
true, and from `y` when condition is false. In particular:
: If `condition`, `x`, and `y` all have the same shape:
* `result[i1...iN] = x[i1...iN]` if `condition[i1...iN]` is true.
* `result[i1...iN] = y[i1...iN]` if `condition[i1...iN]` is false.
: Otherwise:
* `condition` must be a vector.
* `x` and `y` must have the same number of dimensions.
* The outermost dimensions of `condition`, `x`, and `y` must all have the
same size.
* `result[i] = x[i]` if `condition[i]` is true.
* `result[i] = y[i]` if `condition[i]` is false.
Args:
condition: A potentially ragged tensor of type `bool`
x: A potentially ragged tensor (optional).
y: A potentially ragged tensor (optional). Must be specified if `x` is
specified. Must have the same rank and type as `x`.
name: A name of the operation (optional)
Returns:
: If both `x` and `y` are `None`:
A `Tensor` with shape `(num_true, dim_size(condition))`.
: Otherwise:
A potentially ragged tensor with the same type, rank, and outermost
dimension size as `x` and `y`.
`result.ragged_rank = max(x.ragged_rank, y.ragged_rank)`.
Raises:
ValueError: When exactly one of `x` or `y` is non-`None`; or when
`condition`, `x`, and `y` have incompatible shapes.
#### Examples:
>>> # Coordinates where condition is true.
>>> condition = tf.ragged.constant([[True, False, True], [False, True]])
>>> print(where(condition))
tf.Tensor( [[0 0] [0 2] [1 1]], shape=(3, 2), dtype=int64)
>>> # Elementwise selection between x and y, based on condition.
>>> condition = tf.ragged.constant([[True, False, True], [False, True]])
>>> x = tf.ragged.constant([['A', 'B', 'C'], ['D', 'E']])
>>> y = tf.ragged.constant([['a', 'b', 'c'], ['d', 'e']])
>>> print(where(condition, x, y))
<tf.RaggedTensor [[b'A', b'b', b'C'], [b'd', b'E']]>
>>> # Row selection between x and y, based on condition.
>>> condition = [True, False]
>>> x = tf.ragged.constant([['A', 'B', 'C'], ['D', 'E']])
>>> y = tf.ragged.constant([['a', 'b', 'c'], ['d', 'e']])
>>> print(where(condition, x, y))
<tf.RaggedTensor [[b'A', b'B', b'C'], [b'd', b'e']]>"
10414,RowPartition,tensorflow/tensorflow/python/ops/ragged/row_partition.py,51,class,"Partitioning of a sequence of values into contiguous subsequences (""rows"").
A `RowPartition` describes how a sequence with `nvals` items should be
divided into `nrows` contiguous subsequences (""rows""). For example, a
`RowPartition` could be used to partition the vector `[1, 2, 3, 4, 5]` into
subsequences `[[1, 2], [3], [], [4, 5]]`. Note that `RowPartition` stores
information about how values are partitioned, but does not include the
partitioned values themselves. `tf.RaggedTensor` is used to pair a `values`
tensor with one or more `RowPartition`s, providing a complete encoding for a
ragged tensor (i.e. a tensor with variable-length dimensions).
`RowPartition`s may be defined using several different schemes:
* `row_lengths`: an integer vector with shape `[nrows]`, which specifies
the length of each row.
* `row_splits`: an integer vector with shape `[nrows+1]`, specifying the
""split points"" between each row.
* `row_starts`: an integer vector with shape `[nrows]`, which specifies
the start offset for each row. Equivalent to `row_splits[:-1]`.
* `row_limits`: an integer vector with shape `[nrows]`, which specifies
the stop offset for each row. Equivalent to `row_splits[1:]`.
* `value_rowids` is an integer vector with shape `[nvals]`, corresponding
one-to-one with sequence values, which specifies the row that each value
belongs to. If the partition has empty trailing rows, then `nrows`
must also be specified.
* `uniform_row_length` is an integer scalar, specifying the length of every
row. This scheme may only be used if all rows have the same length.
For example, the following `RowPartition`s all represent the partitioning of
8 values into 5 sublists as follows: `[[*, *, *, *], [], [*, *, *], [*], []]`.
>>> p1 = RowPartition.from_row_lengths([4, 0, 3, 1, 0])
>>> p2 = RowPartition.from_row_splits([0, 4, 4, 7, 8, 8])
>>> p3 = RowPartition.from_row_starts([0, 4, 4, 7, 8], nvals=8)
>>> p4 = RowPartition.from_row_limits([4, 4, 7, 8, 8])
>>> p5 = RowPartition.from_value_rowids([0, 0, 0, 0, 2, 2, 2, 3], nrows=5)
For more information about each scheme, see the documentation for the
its factory method. For additional examples, see the documentation on
`tf.RaggedTensor`.
### Precomputed Encodings
`RowPartition` always stores at least one encoding of the partitioning, but
it can be configured to cache additional encodings as well. This can
avoid unnecessary recomputation in eager mode. (In graph mode, optimizations
such as common subexpression elimination will typically prevent these
unnecessary recomputations.) To check which encodings are precomputed, use
`RowPartition.has_precomputed_<encoding>`. To cache an additional
encoding, use `RowPartition.with_precomputed_<encoding>`."
10415,from_value_rowids,tensorflow/tensorflow/python/ops/ragged/row_partition.py,188,method,"Creates a `RowPartition` with rows partitioned by `value_rowids`.
This `RowPartition` divides a sequence `values` into rows by specifying
which row each value should be added to:
```python
partitioned_rows = [[] for _ in nrows]
for (value, rowid) in zip(values, value_rowids):
partitioned_rows[rowid].append(value)
``
Args:
value_rowids: A 1-D integer tensor with shape `[nvals]`, which corresponds
one-to-one with `values`, and specifies each value's row index. Must be
nonnegative, and must be sorted in ascending order.
nrows: An integer scalar specifying the number of rows. This should be
specified if the `RowPartition` may containing empty training rows. Must
be greater than `value_rowids[-1]` (or greater than or equal to zero if
`value_rowids` is empty). Defaults to `value_rowids[-1]` (or zero if
`value_rowids` is empty).
validate: If true, then use assertions to check that the arguments form a
valid `RowPartition`.
preferred_dtype: The dtype to encode value_rowids if it doesn't already
have one. The default is tf.int64.
Returns:
A `RowPartition`.
Raises:
ValueError: If `nrows` is incompatible with `value_rowids`.
#### Example:
>>> print(RowPartition.from_value_rowids(
... value_rowids=[0, 0, 0, 0, 2, 2, 2, 3],
... nrows=4))
tf.RowPartition(row_splits=tf.Tensor([0 4 4 7 8], shape=(5,), dtype=int64))"
10416,from_row_splits,tensorflow/tensorflow/python/ops/ragged/row_partition.py,302,method,"Creates a `RowPartition` with rows partitioned by `row_splits`.
This `RowPartition` divides a sequence `values` into rows by indicating
where each row begins and ends:
```python
partitioned_rows = []
for i in range(len(row_splits) - 1):
row_start = row_splits[i]
row_end = row_splits[i + 1]
partitioned_rows.append(values[row_start:row_end])
```
Args:
row_splits: A 1-D integer tensor with shape `[nrows+1]`. Must not be
empty, and must be sorted in ascending order. `row_splits[0]` must be
zero.
validate: If true, then use assertions to check that the arguments form a
valid `RowPartition`.
preferred_dtype: If row_splits has an unspecified type, use this one. If
preferred_dtype is None, defaults to dtypes.int64.
Returns:
A `RowPartition`.
Raises:
ValueError: If `row_splits` is an empty list."
10417,from_row_lengths,tensorflow/tensorflow/python/ops/ragged/row_partition.py,356,method,"Creates a `RowPartition` with rows partitioned by `row_lengths`.
This `RowPartition` divides a sequence `values` into rows by indicating
the length of each row:
```python
partitioned_rows = [[values.pop(0) for _ in range(length)]
for length in row_lengths]
```
Args:
row_lengths: A 1-D integer tensor with shape `[nrows]`. Must be
nonnegative.
validate: If true, then use assertions to check that the arguments form a
valid `RowPartition`.
preferred_dtype: If row_lengths has an unspecified type, use this one. If
preferred_dtype is None, defaults to dtypes.int64.
Returns:
A `RowPartition`."
10418,from_row_starts,tensorflow/tensorflow/python/ops/ragged/row_partition.py,401,method,"Creates a `RowPartition` with rows partitioned by `row_starts`.
Equivalent to: `from_row_splits(concat([row_starts, nvals], axis=0))`.
Args:
row_starts: A 1-D integer tensor with shape `[nrows]`. Must be
nonnegative and sorted in ascending order. If `nrows>0`, then
`row_starts[0]` must be zero.
nvals: A scalar tensor indicating the number of values.
validate: If true, then use assertions to check that the arguments form a
valid `RowPartition`.
preferred_dtype: If row_limits has an unspecified type, use this one. If
preferred_dtype is None, defaults to dtypes.int64.
Returns:
A `RowPartition`."
10419,from_row_limits,tensorflow/tensorflow/python/ops/ragged/row_partition.py,444,method,"Creates a `RowPartition` with rows partitioned by `row_limits`.
Equivalent to: `from_row_splits(values, concat([0, row_limits], axis=0))`.
Args:
row_limits: A 1-D integer tensor with shape `[nrows]`. Must be sorted in
ascending order.
validate: If true, then use assertions to check that the arguments form a
valid `RowPartition`.
preferred_dtype: If row_limits has an unspecified type, use this one. If
preferred_dtype is None, defaults to dtypes.int64.
Returns:
A `RowPartition`."
10420,from_uniform_row_length,tensorflow/tensorflow/python/ops/ragged/row_partition.py,483,method,"Creates a `RowPartition` with rows partitioned by `uniform_row_length`.
This `RowPartition` divides a sequence `values` into rows that all have
the same length:
```python
partitioned_rows = [[values.pop(0) for _ in range(uniform_row_length)]
for _ in range(nrows)]
```
Args:
uniform_row_length: A scalar integer tensor. Must be nonnegative. The
size of the outer axis of `values` must be evenly divisible by
`uniform_row_length`.
nvals: a non-negative scalar integer tensor for the number of values.
nrows: The number of rows in the constructed RowPartition. If not
specified, then it defaults to `nvals/uniform_row_length` (or `0` if
`uniform_row_length==0`). `nrows` only needs to be specified if
`uniform_row_length` might be zero. `uniform_row_length*nrows` must be
`nvals`.
validate: If true, then use assertions to check that the arguments form a
valid `RowPartition`.
preferred_dtype: if uniform_row_length has no dtype, use this one.
Returns:
A `RowPartition`."
10421,with_dependencies,tensorflow/tensorflow/python/ops/ragged/row_partition.py,622,method,"Returns a new RowPartition equal to self with control dependencies.
Specifically, self._row_splits is gated by the given control dependencies.
Used to add sanity checks to the constructors.
Args:
dependencies: a list of tensors to use as dependencies.
Returns:
A new RowPartition object."
10422,dtype,tensorflow/tensorflow/python/ops/ragged/row_partition.py,649,method,The `DType` used to encode the row partition (either int32 or int64).
10423,row_splits,tensorflow/tensorflow/python/ops/ragged/row_partition.py,653,method,"Returns the row-split indices for this row partition.
`row_splits` specifies where the values for each row begin and end.
In particular, the values for row `i` are stored in the slice
`values[row_splits[i]:row_splits[i+1]]`.
Returns:
A 1-D integer `Tensor` with shape `[self.nrows+1]`.
The returned tensor is non-empty, and is sorted in ascending order.
`self.row_splits()[0] == 0`.
`self.row_splits()[-1] == self.nvals()`."
10424,value_rowids,tensorflow/tensorflow/python/ops/ragged/row_partition.py,668,method,"Returns the row indices for this row partition.
`value_rowids` specifies the row index fo reach value. In particular,
`value_rowids[i]` is the row index for `values[i]`.
Returns:
A 1-D integer `Tensor` with shape `[self.nvals()]`.
The returned tensor is nonnegative, and is sorted in ascending order."
10425,nvals,tensorflow/tensorflow/python/ops/ragged/row_partition.py,682,method,"Returns the number of values partitioned by this `RowPartition`.
If the sequence partitioned by this `RowPartition` is a tensor, then
`nvals` is the size of that tensor's outermost dimension -- i.e.,
`nvals == values.shape[0]`.
Args:
out_type: `dtype` for the returned tensor. Defaults to `self.dtype`.
Returns:
scalar integer Tensor"
10426,nrows,tensorflow/tensorflow/python/ops/ragged/row_partition.py,701,method,"Returns the number of rows created by this `RowPartition`.
Args:
out_type: `dtype` for the returned tensor. Defaults to `self.dtype`.
Returns:
scalar integer Tensor"
10427,uniform_row_length,tensorflow/tensorflow/python/ops/ragged/row_partition.py,722,method,"Returns the length of each row in this partition, if rows are uniform.
If all rows in this `RowPartition` have the same length, then this returns
that length as a scalar integer `Tensor`. Otherwise, it returns `None`.
Returns:
scalar Tensor with `type=self.dtype`, or `None`."
10428,row_starts,tensorflow/tensorflow/python/ops/ragged/row_partition.py,733,method,"Returns the start indices for rows in this row partition.
These indices specify where the values for each row begin.
`partition.row_starts()` is equal to `partition.row_splits()[:-1]`.
Returns:
A 1-D integer Tensor with shape `[self.nrows()]`.
The returned tensor is nonnegative, and is sorted in ascending order.
`self.row_starts()[0] == 0`.
`self.row_starts()[-1] <= self.nvals()`."
10429,row_limits,tensorflow/tensorflow/python/ops/ragged/row_partition.py,747,method,"Returns the limit indices for rows in this row partition.
These indices specify where the values for each row end.
`partition.row_limits()` is equal to `partition.row_splits()[:-1]`.
Returns:
A 1-D integer Tensor with shape `[self.nrows]`.
The returned tensor is nonnegative, and is sorted in ascending order.
`self.row_limits()[-1] == self.nvals()`."
10430,row_lengths,tensorflow/tensorflow/python/ops/ragged/row_partition.py,760,method,"Returns the lengths of rows in this `RowPartition`.
Returns:
A 1-D integer Tensor with shape `[self.nrows]`.
The returned tensor is nonnegative.
`tf.reduce_sum(self.row_lengths) == self.nvals()`."
10431,static_nrows,tensorflow/tensorflow/python/ops/ragged/row_partition.py,774,method,"The number of rows in this partition, if statically known.
```python
self.row_lengths().shape == [self.static_nrows]
self.row_starts().shape == [self.static_nrows]
self.row_limits().shape == [self.static_nrows]
self.row_splits().shape == [self.static_nrows + 1]
```
Returns:
The number of rows in this partition as an `int` (if statically known);
or `None` (otherwise)."
10432,static_nvals,tensorflow/tensorflow/python/ops/ragged/row_partition.py,801,method,"The number of values in this partition, if statically known.
```python
self.value_rowids().shape == [self.static_vals]
```
Returns:
The number of values in this partition as an `int` (if statically known);
or `None` (otherwise)."
10433,static_uniform_row_length,tensorflow/tensorflow/python/ops/ragged/row_partition.py,819,method,"The number of values in each row of this partition, if statically known.
Returns:
The number of values in each row of this partition as an `int` (if
statically known); or `None` (otherwise)."
10434,with_row_splits_dtype,tensorflow/tensorflow/python/ops/ragged/row_partition.py,834,method,"Returns a copy of this RowPartition with the given `row_splits` dtype.
For RaggedTensors with multiple ragged dimensions, the `row_splits` for all
nested `RaggedTensor` objects are cast to the given dtype.
Args:
dtype: The dtype for `row_splits`. One of `tf.int32` or `tf.int64`.
Returns:
A copy of this RaggedTensor, with the `row_splits` cast to the given
type."
10435,has_precomputed_row_splits,tensorflow/tensorflow/python/ops/ragged/row_partition.py,872,method,"Returns true if `row_splits` has already been computed.
If true, then `self.row_splits()` will return its value without calling
any TensorFlow ops."
10436,has_precomputed_row_lengths,tensorflow/tensorflow/python/ops/ragged/row_partition.py,880,method,"Returns true if `row_lengths` has already been computed.
If true, then `self.row_lengths()` will return its value without calling
any TensorFlow ops."
10437,has_precomputed_value_rowids,tensorflow/tensorflow/python/ops/ragged/row_partition.py,888,method,"Returns true if `value_rowids` has already been computed.
If true, then `self.value_rowids()` will return its value without calling
any TensorFlow ops."
10438,has_precomputed_nrows,tensorflow/tensorflow/python/ops/ragged/row_partition.py,896,method,"Returns true if `nrows` has already been computed.
If true, then `self.nrows()` will return its value without calling
any TensorFlow ops."
10439,with_precomputed_row_splits,tensorflow/tensorflow/python/ops/ragged/row_partition.py,904,method,Returns a copy of `self` with `row_splits` precomputed.
10440,with_precomputed_row_lengths,tensorflow/tensorflow/python/ops/ragged/row_partition.py,914,method,Returns a copy of `self` with `row_lengths` precomputed.
10441,with_precomputed_value_rowids,tensorflow/tensorflow/python/ops/ragged/row_partition.py,924,method,Returns a copy of `self` with `value_rowids` precomputed.
10442,with_precomputed_nrows,tensorflow/tensorflow/python/ops/ragged/row_partition.py,934,method,Returns a copy of `self` with `nrows` precomputed.
10443,merge_precomputed_encodings,tensorflow/tensorflow/python/ops/ragged/row_partition.py,944,method,"Returns a RowPartition that merges encodings from `self` and `other`.
Requires that `self` and `other` describe the same partition.
Args:
other: A `RowPartition` that encodes the same partition as `self`.
validate: If true, then add runtime checks to verify that `self` and
`other` encode the same row partition.
Returns:
A `RowPartition`."
10444,RowPartitionSpec,tensorflow/tensorflow/python/ops/ragged/row_partition.py,1029,class,Type specification for a `tf.RowPartition`.
10445,is_compatible_with,tensorflow/tensorflow/python/ops/ragged/row_partition.py,1096,method,
10446,nrows,tensorflow/tensorflow/python/ops/ragged/row_partition.py,1116,method,
10447,nvals,tensorflow/tensorflow/python/ops/ragged/row_partition.py,1120,method,
10448,uniform_row_length,tensorflow/tensorflow/python/ops/ragged/row_partition.py,1124,method,
10449,dtype,tensorflow/tensorflow/python/ops/ragged/row_partition.py,1128,method,
10450,from_value,tensorflow/tensorflow/python/ops/ragged/row_partition.py,1144,method,
10451,row_splits_to_segment_ids,tensorflow/tensorflow/python/ops/ragged/segment_id_ops.py,36,function,"Generates the segmentation corresponding to a RaggedTensor `row_splits`.
Returns an integer vector `segment_ids`, where `segment_ids[i] == j` if
`splits[j] <= i < splits[j+1]`. Example:
>>> print(tf.ragged.row_splits_to_segment_ids([0, 3, 3, 5, 6, 9]))
tf.Tensor([0 0 0 2 2 3 4 4 4], shape=(9,), dtype=int64)
Args:
splits: A sorted 1-D integer Tensor. `splits[0]` must be zero.
name: A name prefix for the returned tensor (optional).
out_type: The dtype for the return value. Defaults to `splits.dtype`,
or `tf.int64` if `splits` does not have a dtype.
Returns:
A sorted 1-D integer Tensor, with `shape=[splits[-1]]`
Raises:
ValueError: If `splits` is invalid."
10452,segment_ids_to_row_splits,tensorflow/tensorflow/python/ops/ragged/segment_id_ops.py,80,function,"Generates the RaggedTensor `row_splits` corresponding to a segmentation.
Returns an integer vector `splits`, where `splits[0] = 0` and
`splits[i] = splits[i-1] + count(segment_ids==i)`. Example:
>>> print(tf.ragged.segment_ids_to_row_splits([0, 0, 0, 2, 2, 3, 4, 4, 4]))
tf.Tensor([0 3 3 5 6 9], shape=(6,), dtype=int64)
Args:
segment_ids: A 1-D integer Tensor.
num_segments: A scalar integer indicating the number of segments. Defaults
to `max(segment_ids) + 1` (or zero if `segment_ids` is empty).
out_type: The dtype for the return value. Defaults to `segment_ids.dtype`,
or `tf.int64` if `segment_ids` does not have a dtype.
name: A name prefix for the returned tensor (optional).
Returns:
A sorted 1-D integer Tensor, with `shape=[num_segments + 1]`."
10453,dct,tensorflow/tensorflow/python/ops/signal/dct_ops.py,55,function,"Computes the 1D [Discrete Cosine Transform (DCT)][dct] of `input`.
Types I, II, III and IV are supported.
Type I is implemented using a length `2N` padded `tf.signal.rfft`.
Type II is implemented using a length `2N` padded `tf.signal.rfft`, as
described here: [Type 2 DCT using 2N FFT padded (Makhoul)]
(https://dsp.stackexchange.com/a/10606).
Type III is a fairly straightforward inverse of Type II
(i.e. using a length `2N` padded `tf.signal.irfft`).
Type IV is calculated through 2N length DCT2 of padded signal and
picking the odd indices.
@compatibility(scipy)
Equivalent to [scipy.fftpack.dct]
(https://docs.scipy.org/doc/scipy-1.4.0/reference/generated/scipy.fftpack.dct.html)
for Type-I, Type-II, Type-III and Type-IV DCT.
@end_compatibility
Args:
input: A `[..., samples]` `float32`/`float64` `Tensor` containing the
signals to take the DCT of.
type: The DCT type to perform. Must be 1, 2, 3 or 4.
n: The length of the transform. If length is less than sequence length,
only the first n elements of the sequence are considered for the DCT.
If n is greater than the sequence length, zeros are padded and then
the DCT is computed as usual.
axis: For future expansion. The axis to compute the DCT along. Must be `-1`.
norm: The normalization to apply. `None` for no normalization or `'ortho'`
for orthonormal normalization.
name: An optional name for the operation.
Returns:
A `[..., samples]` `float32`/`float64` `Tensor` containing the DCT of
`input`.
Raises:
ValueError: If `type` is not `1`, `2`, `3` or `4`, `axis` is
not `-1`, `n` is not `None` or greater than 0,
or `norm` is not `None` or `'ortho'`.
ValueError: If `type` is `1` and `norm` is `ortho`.
[dct]: https://en.wikipedia.org/wiki/Discrete_cosine_transform"
10454,idct,tensorflow/tensorflow/python/ops/signal/dct_ops.py,187,function,"Computes the 1D [Inverse Discrete Cosine Transform (DCT)][idct] of `input`.
Currently Types I, II, III, IV are supported. Type III is the inverse of
Type II, and vice versa.
Note that you must re-normalize by 1/(2n) to obtain an inverse if `norm` is
not `'ortho'`. That is:
`signal == idct(dct(signal)) * 0.5 / signal.shape[-1]`.
When `norm='ortho'`, we have:
`signal == idct(dct(signal, norm='ortho'), norm='ortho')`.
@compatibility(scipy)
Equivalent to [scipy.fftpack.idct]
(https://docs.scipy.org/doc/scipy-1.4.0/reference/generated/scipy.fftpack.idct.html)
for Type-I, Type-II, Type-III and Type-IV DCT.
@end_compatibility
Args:
input: A `[..., samples]` `float32`/`float64` `Tensor` containing the
signals to take the DCT of.
type: The IDCT type to perform. Must be 1, 2, 3 or 4.
n: For future expansion. The length of the transform. Must be `None`.
axis: For future expansion. The axis to compute the DCT along. Must be `-1`.
norm: The normalization to apply. `None` for no normalization or `'ortho'`
for orthonormal normalization.
name: An optional name for the operation.
Returns:
A `[..., samples]` `float32`/`float64` `Tensor` containing the IDCT of
`input`.
Raises:
ValueError: If `type` is not `1`, `2` or `3`, `n` is not `None, `axis` is
not `-1`, or `norm` is not `None` or `'ortho'`.
[idct]:
https://en.wikipedia.org/wiki/Discrete_cosine_transform#Inverse_transforms"
10455,fftshift,tensorflow/tensorflow/python/ops/signal/fft_ops.py,374,function,"Shift the zero-frequency component to the center of the spectrum.
This function swaps half-spaces for all axes listed (defaults to all).
Note that ``y[0]`` is the Nyquist component only if ``len(x)`` is even.
@compatibility(numpy)
Equivalent to numpy.fft.fftshift.
https://docs.scipy.org/doc/numpy/reference/generated/numpy.fft.fftshift.html
@end_compatibility
For example:
```python
x = tf.signal.fftshift([ 0., 1., 2., 3., 4., -5., -4., -3., -2., -1.])
x.numpy() # array([-5., -4., -3., -2., -1., 0., 1., 2., 3., 4.])
```
Args:
x: `Tensor`, input tensor.
axes: `int` or shape `tuple`, optional Axes over which to shift. Default is
None, which shifts all axes.
name: An optional name for the operation.
Returns:
A `Tensor`, The shifted tensor."
10456,ifftshift,tensorflow/tensorflow/python/ops/signal/fft_ops.py,419,function,"The inverse of fftshift.
Although identical for even-length x,
the functions differ by one sample for odd-length x.
@compatibility(numpy)
Equivalent to numpy.fft.ifftshift.
https://docs.scipy.org/doc/numpy/reference/generated/numpy.fft.ifftshift.html
@end_compatibility
For example:
```python
x = tf.signal.ifftshift([[ 0., 1., 2.],[ 3., 4., -4.],[-3., -2., -1.]])
x.numpy() # array([[ 4., -4., 3.],[-2., -1., -3.],[ 1., 2., 0.]])
```
Args:
x: `Tensor`, input tensor.
axes: `int` or shape `tuple` Axes over which to calculate. Defaults to None,
which shifts all axes.
name: An optional name for the operation.
Returns:
A `Tensor`, The shifted tensor."
10457,linear_to_mel_weight_matrix,tensorflow/tensorflow/python/ops/signal/mel_ops.py,95,function,"Returns a matrix to warp linear scale spectrograms to the [mel scale][mel].
Returns a weight matrix that can be used to re-weight a `Tensor` containing
`num_spectrogram_bins` linearly sampled frequency information from
`[0, sample_rate / 2]` into `num_mel_bins` frequency information from
`[lower_edge_hertz, upper_edge_hertz]` on the [mel scale][mel].
This function follows the [Hidden Markov Model Toolkit
(HTK)](http://htk.eng.cam.ac.uk/) convention, defining the mel scale in
terms of a frequency in hertz according to the following formula:
$$ extrm{mel}(f) = 2595 * extrm{log}_{10}(1 + rac{f}{700})$$
In the returned matrix, all the triangles (filterbanks) have a peak value
of 1.0.
For example, the returned matrix `A` can be used to right-multiply a
spectrogram `S` of shape `[frames, num_spectrogram_bins]` of linear
scale spectrum values (e.g. STFT magnitudes) to generate a ""mel spectrogram""
`M` of shape `[frames, num_mel_bins]`.
# `S` has shape [frames, num_spectrogram_bins]
# `M` has shape [frames, num_mel_bins]
M = tf.matmul(S, A)
The matrix can be used with `tf.tensordot` to convert an arbitrary rank
`Tensor` of linear-scale spectral bins into the mel scale.
# S has shape [..., num_spectrogram_bins].
# M has shape [..., num_mel_bins].
M = tf.tensordot(S, A, 1)
Args:
num_mel_bins: Python int. How many bands in the resulting mel spectrum.
num_spectrogram_bins: An integer `Tensor`. How many bins there are in the
source spectrogram data, which is understood to be `fft_size // 2 + 1`,
i.e. the spectrogram only contains the nonredundant FFT bins.
sample_rate: An integer or float `Tensor`. Samples per second of the input
signal used to create the spectrogram. Used to figure out the frequencies
corresponding to each spectrogram bin, which dictates how they are mapped
into the mel scale.
lower_edge_hertz: Python float. Lower bound on the frequencies to be
included in the mel spectrum. This corresponds to the lower edge of the
lowest triangular band.
upper_edge_hertz: Python float. The desired top edge of the highest
frequency band.
dtype: The `DType` of the result matrix. Must be a floating point type.
name: An optional name for the operation.
Returns:
A `Tensor` of shape `[num_spectrogram_bins, num_mel_bins]`.
Raises:
ValueError: If `num_mel_bins`/`num_spectrogram_bins`/`sample_rate` are not
positive, `lower_edge_hertz` is negative, frequency edges are incorrectly
ordered, `upper_edge_hertz` is larger than the Nyquist frequency.
[mel]: https://en.wikipedia.org/wiki/Mel_scale"
10458,mfccs_from_log_mel_spectrograms,tensorflow/tensorflow/python/ops/signal/mfcc_ops.py,31,function,"Computes [MFCCs][mfcc] of `log_mel_spectrograms`.
Implemented with GPU-compatible ops and supports gradients.
[Mel-Frequency Cepstral Coefficient (MFCC)][mfcc] calculation consists of
taking the DCT-II of a log-magnitude mel-scale spectrogram. [HTK][htk]'s MFCCs
use a particular scaling of the DCT-II which is almost orthogonal
normalization. We follow this convention.
All `num_mel_bins` MFCCs are returned and it is up to the caller to select
a subset of the MFCCs based on their application. For example, it is typical
to only use the first few for speech recognition, as this results in
an approximately pitch-invariant representation of the signal.
For example:
```python
batch_size, num_samples, sample_rate = 32, 32000, 16000.0
# A Tensor of [batch_size, num_samples] mono PCM samples in the range [-1, 1].
pcm = tf.random.normal([batch_size, num_samples], dtype=tf.float32)
# A 1024-point STFT with frames of 64 ms and 75% overlap.
stfts = tf.signal.stft(pcm, frame_length=1024, frame_step=256,
fft_length=1024)
spectrograms = tf.abs(stfts)
# Warp the linear scale spectrograms into the mel-scale.
num_spectrogram_bins = stfts.shape[-1].value
lower_edge_hertz, upper_edge_hertz, num_mel_bins = 80.0, 7600.0, 80
linear_to_mel_weight_matrix = tf.signal.linear_to_mel_weight_matrix(
num_mel_bins, num_spectrogram_bins, sample_rate, lower_edge_hertz,
upper_edge_hertz)
mel_spectrograms = tf.tensordot(
spectrograms, linear_to_mel_weight_matrix, 1)
mel_spectrograms.set_shape(spectrograms.shape[:-1].concatenate(
linear_to_mel_weight_matrix.shape[-1:]))
# Compute a stabilized log to get log-magnitude mel-scale spectrograms.
log_mel_spectrograms = tf.math.log(mel_spectrograms + 1e-6)
# Compute MFCCs from log_mel_spectrograms and take the first 13.
mfccs = tf.signal.mfccs_from_log_mel_spectrograms(
log_mel_spectrograms)[..., :13]
```
Args:
log_mel_spectrograms: A `[..., num_mel_bins]` `float32`/`float64` `Tensor`
of log-magnitude mel-scale spectrograms.
name: An optional name for the operation.
Returns:
A `[..., num_mel_bins]` `float32`/`float64` `Tensor` of the MFCCs of
`log_mel_spectrograms`.
Raises:
ValueError: If `num_mel_bins` is not positive.
[mfcc]: https://en.wikipedia.org/wiki/Mel-frequency_cepstrum
[htk]: https://en.wikipedia.org/wiki/HTK_(software)"
10459,overlap_and_add,tensorflow/tensorflow/python/ops/signal/reconstruction_ops.py,32,function,"Reconstructs a signal from a framed representation.
Adds potentially overlapping frames of a signal with shape
`[..., frames, frame_length]`, offsetting subsequent frames by `frame_step`.
The resulting tensor has shape `[..., output_size]` where
output_size = (frames - 1) * frame_step + frame_length
Args:
signal: A [..., frames, frame_length] `Tensor`. All dimensions may be
unknown, and rank must be at least 2.
frame_step: An integer or scalar `Tensor` denoting overlap offsets. Must be
less than or equal to `frame_length`.
name: An optional name for the operation.
Returns:
A `Tensor` with shape `[..., output_size]` containing the overlap-added
frames of `signal`'s inner-most two dimensions.
Raises:
ValueError: If `signal`'s rank is less than 2, or `frame_step` is not a
scalar integer."
10460,frame,tensorflow/tensorflow/python/ops/signal/shape_ops.py,60,function,"Expands `signal`'s `axis` dimension into frames of `frame_length`.
Slides a window of size `frame_length` over `signal`'s `axis` dimension
with a stride of `frame_step`, replacing the `axis` dimension with
`[frames, frame_length]` frames.
If `pad_end` is True, window positions that are past the end of the `axis`
dimension are padded with `pad_value` until the window moves fully past the
end of the dimension. Otherwise, only window positions that fully overlap the
`axis` dimension are produced.
For example:
>>> # A batch size 3 tensor of 9152 audio samples.
>>> audio = tf.random.normal([3, 9152])
>>>
>>> # Compute overlapping frames of length 512 with a step of 180 (frames overlap
>>> # by 332 samples). By default, only 49 frames are generated since a frame
>>> # with start position j*180 for j > 48 would overhang the end.
>>> frames = tf.signal.frame(audio, 512, 180)
>>> frames.shape.assert_is_compatible_with([3, 49, 512])
>>>
>>> # When pad_end is enabled, the final two frames are kept (padded with zeros).
>>> frames = tf.signal.frame(audio, 512, 180, pad_end=True)
>>> frames.shape.assert_is_compatible_with([3, 51, 512])
If the dimension along `axis` is N, and `pad_end=False`, the number of frames
can be computed by:
```python
num_frames = 1 + (N - frame_size) // frame_step
```
If `pad_end=True`, the number of frames can be computed by:
```python
num_frames = -(-N // frame_step) # ceiling division
```
Args:
signal: A `[..., samples, ...]` `Tensor`. The rank and dimensions
may be unknown. Rank must be at least 1.
frame_length: The frame length in samples. An integer or scalar `Tensor`.
frame_step: The frame hop size in samples. An integer or scalar `Tensor`.
pad_end: Whether to pad the end of `signal` with `pad_value`.
pad_value: An optional scalar `Tensor` to use where the input signal
does not exist when `pad_end` is True.
axis: A scalar integer `Tensor` indicating the axis to frame. Defaults to
the last axis. Supports negative values for indexing from the end.
name: An optional name for the operation.
Returns:
A `Tensor` of frames with shape `[..., num_frames, frame_length, ...]`.
Raises:
ValueError: If `frame_length`, `frame_step`, `pad_value`, or `axis` are not
scalar."
10461,stft,tensorflow/tensorflow/python/ops/signal/spectral_ops.py,40,function,"Computes the [Short-time Fourier Transform][stft] of `signals`.
Implemented with TPU/GPU-compatible ops and supports gradients.
Args:
signals: A `[..., samples]` `float32`/`float64` `Tensor` of real-valued
signals.
frame_length: An integer scalar `Tensor`. The window length in samples.
frame_step: An integer scalar `Tensor`. The number of samples to step.
fft_length: An integer scalar `Tensor`. The size of the FFT to apply.
If not provided, uses the smallest power of 2 enclosing `frame_length`.
window_fn: A callable that takes a window length and a `dtype` keyword
argument and returns a `[window_length]` `Tensor` of samples in the
provided datatype. If set to `None`, no windowing is used.
pad_end: Whether to pad the end of `signals` with zeros when the provided
frame length and step produces a frame that lies partially past its end.
name: An optional name for the operation.
Returns:
A `[..., frames, fft_unique_bins]` `Tensor` of `complex64`/`complex128`
STFT values where `fft_unique_bins` is `fft_length // 2 + 1` (the unique
components of the FFT).
Raises:
ValueError: If `signals` is not at least rank 1, `frame_length` is
not scalar, or `frame_step` is not scalar.
[stft]: https://en.wikipedia.org/wiki/Short-time_Fourier_transform"
10462,inverse_stft_window_fn,tensorflow/tensorflow/python/ops/signal/spectral_ops.py,101,function,"Generates a window function that can be used in `inverse_stft`.
Constructs a window that is equal to the forward window with a further
pointwise amplitude correction. `inverse_stft_window_fn` is equivalent to
`forward_window_fn` in the case where it would produce an exact inverse.
See examples in `inverse_stft` documentation for usage.
Args:
frame_step: An integer scalar `Tensor`. The number of samples to step.
forward_window_fn: window_fn used in the forward transform, `stft`.
name: An optional name for the operation.
Returns:
A callable that takes a window length and a `dtype` keyword argument and
returns a `[window_length]` `Tensor` of samples in the provided datatype.
The returned window is suitable for reconstructing original waveform in
inverse_stft."
10463,inverse_stft,tensorflow/tensorflow/python/ops/signal/spectral_ops.py,163,function,"Computes the inverse [Short-time Fourier Transform][stft] of `stfts`.
To reconstruct an original waveform, a complementary window function should
be used with `inverse_stft`. Such a window function can be constructed with
`tf.signal.inverse_stft_window_fn`.
Example:
```python
frame_length = 400
frame_step = 160
waveform = tf.random.normal(dtype=tf.float32, shape=[1000])
stft = tf.signal.stft(waveform, frame_length, frame_step)
inverse_stft = tf.signal.inverse_stft(
stft, frame_length, frame_step,
window_fn=tf.signal.inverse_stft_window_fn(frame_step))
```
If a custom `window_fn` is used with `tf.signal.stft`, it must be passed to
`tf.signal.inverse_stft_window_fn`:
```python
frame_length = 400
frame_step = 160
window_fn = tf.signal.hamming_window
waveform = tf.random.normal(dtype=tf.float32, shape=[1000])
stft = tf.signal.stft(
waveform, frame_length, frame_step, window_fn=window_fn)
inverse_stft = tf.signal.inverse_stft(
stft, frame_length, frame_step,
window_fn=tf.signal.inverse_stft_window_fn(
frame_step, forward_window_fn=window_fn))
```
Implemented with TPU/GPU-compatible ops and supports gradients.
Args:
stfts: A `complex64`/`complex128` `[..., frames, fft_unique_bins]`
`Tensor` of STFT bins representing a batch of `fft_length`-point STFTs
where `fft_unique_bins` is `fft_length // 2 + 1`
frame_length: An integer scalar `Tensor`. The window length in samples.
frame_step: An integer scalar `Tensor`. The number of samples to step.
fft_length: An integer scalar `Tensor`. The size of the FFT that produced
`stfts`. If not provided, uses the smallest power of 2 enclosing
`frame_length`.
window_fn: A callable that takes a window length and a `dtype` keyword
argument and returns a `[window_length]` `Tensor` of samples in the
provided datatype. If set to `None`, no windowing is used.
name: An optional name for the operation.
Returns:
A `[..., samples]` `Tensor` of `float32`/`float64` signals representing
the inverse STFT for each input STFT in `stfts`.
Raises:
ValueError: If `stfts` is not at least rank 2, `frame_length` is not scalar,
`frame_step` is not scalar, or `fft_length` is not scalar.
[stft]: https://en.wikipedia.org/wiki/Short-time_Fourier_transform"
10464,mdct,tensorflow/tensorflow/python/ops/signal/spectral_ops.py,299,function,"Computes the [Modified Discrete Cosine Transform][mdct] of `signals`.
Implemented with TPU/GPU-compatible ops and supports gradients.
Args:
signals: A `[..., samples]` `float32`/`float64` `Tensor` of real-valued
signals.
frame_length: An integer scalar `Tensor`. The window length in samples
which must be divisible by 4.
window_fn: A callable that takes a frame_length and a `dtype` keyword
argument and returns a `[frame_length]` `Tensor` of samples in the
provided datatype. If set to `None`, a rectangular window with a scale of
1/sqrt(2) is used. For perfect reconstruction of a signal from `mdct`
followed by `inverse_mdct`, please use `tf.signal.vorbis_window`,
`tf.signal.kaiser_bessel_derived_window` or `None`. If using another
window function, make sure that w[n]^2 + w[n + frame_length // 2]^2 = 1
and w[n] = w[frame_length - n - 1] for n = 0,...,frame_length // 2 - 1 to
achieve perfect reconstruction.
pad_end: Whether to pad the end of `signals` with zeros when the provided
frame length and step produces a frame that lies partially past its end.
norm: If it is None, unnormalized dct4 is used, if it is ""ortho""
orthonormal dct4 is used.
name: An optional name for the operation.
Returns:
A `[..., frames, frame_length // 2]` `Tensor` of `float32`/`float64`
MDCT values where `frames` is roughly `samples // (frame_length // 2)`
when `pad_end=False`.
Raises:
ValueError: If `signals` is not at least rank 1, `frame_length` is
not scalar, or `frame_length` is not a multiple of `4`.
[mdct]: https://en.wikipedia.org/wiki/Modified_discrete_cosine_transform"
10465,inverse_mdct,tensorflow/tensorflow/python/ops/signal/spectral_ops.py,375,function,"Computes the inverse modified DCT of `mdcts`.
To reconstruct an original waveform, the same window function should
be used with `mdct` and `inverse_mdct`.
Example usage:
>>> @tf.function
... def compare_round_trip():
... samples = 1000
... frame_length = 400
... halflen = frame_length // 2
... waveform = tf.random.normal(dtype=tf.float32, shape=[samples])
... waveform_pad = tf.pad(waveform, [[halflen, 0],])
... mdct = tf.signal.mdct(waveform_pad, frame_length, pad_end=True,
... window_fn=tf.signal.vorbis_window)
... inverse_mdct = tf.signal.inverse_mdct(mdct,
... window_fn=tf.signal.vorbis_window)
... inverse_mdct = inverse_mdct[halflen: halflen + samples]
... return waveform, inverse_mdct
>>> waveform, inverse_mdct = compare_round_trip()
>>> np.allclose(waveform.numpy(), inverse_mdct.numpy(), rtol=1e-3, atol=1e-4)
True
Implemented with TPU/GPU-compatible ops and supports gradients.
Args:
mdcts: A `float32`/`float64` `[..., frames, frame_length // 2]`
`Tensor` of MDCT bins representing a batch of `frame_length // 2`-point
MDCTs.
window_fn: A callable that takes a frame_length and a `dtype` keyword
argument and returns a `[frame_length]` `Tensor` of samples in the
provided datatype. If set to `None`, a rectangular window with a scale of
1/sqrt(2) is used. For perfect reconstruction of a signal from `mdct`
followed by `inverse_mdct`, please use `tf.signal.vorbis_window`,
`tf.signal.kaiser_bessel_derived_window` or `None`. If using another
window function, make sure that w[n]^2 + w[n + frame_length // 2]^2 = 1
and w[n] = w[frame_length - n - 1] for n = 0,...,frame_length // 2 - 1 to
achieve perfect reconstruction.
norm: If ""ortho"", orthonormal inverse DCT4 is performed, if it is None,
a regular dct4 followed by scaling of `1/frame_length` is performed.
name: An optional name for the operation.
Returns:
A `[..., samples]` `Tensor` of `float32`/`float64` signals representing
the inverse MDCT for each input MDCT in `mdcts` where `samples` is
`(frames - 1) * (frame_length // 2) + frame_length`.
Raises:
ValueError: If `mdcts` is not at least rank 2.
[mdct]: https://en.wikipedia.org/wiki/Modified_discrete_cosine_transform"
10466,gcd,tensorflow/tensorflow/python/ops/signal/util_ops.py,32,function,"Returns the greatest common divisor via Euclid's algorithm.
Args:
a: The dividend. A scalar integer `Tensor`.
b: The divisor. A scalar integer `Tensor`.
name: An optional name for the operation.
Returns:
A scalar `Tensor` representing the greatest common divisor between `a` and
`b`.
Raises:
ValueError: If `a` or `b` are not scalar integers."
10467,kaiser_window,tensorflow/tensorflow/python/ops/signal/window_ops.py,58,function,"Generate a [Kaiser window][kaiser].
Args:
window_length: A scalar `Tensor` indicating the window length to generate.
beta: Beta parameter for Kaiser window, see reference below.
dtype: The data type to produce. Must be a floating point type.
name: An optional name for the operation.
Returns:
A `Tensor` of shape `[window_length]` of type `dtype`.
[kaiser]:
https://docs.scipy.org/doc/numpy/reference/generated/numpy.kaiser.html"
10468,kaiser_bessel_derived_window,tensorflow/tensorflow/python/ops/signal/window_ops.py,98,function,"Generate a [Kaiser Bessel derived window][kbd].
Args:
window_length: A scalar `Tensor` indicating the window length to generate.
beta: Beta parameter for Kaiser window.
dtype: The data type to produce. Must be a floating point type.
name: An optional name for the operation.
Returns:
A `Tensor` of shape `[window_length]` of type `dtype`.
[kbd]:
https://en.wikipedia.org/wiki/Kaiser_window#Kaiser%E2%80%93Bessel-derived_(KBD)_window"
10469,vorbis_window,tensorflow/tensorflow/python/ops/signal/window_ops.py,126,function,"Generate a [Vorbis power complementary window][vorbis].
Args:
window_length: A scalar `Tensor` indicating the window length to generate.
dtype: The data type to produce. Must be a floating point type.
name: An optional name for the operation.
Returns:
A `Tensor` of shape `[window_length]` of type `dtype`.
[vorbis]:
https://en.wikipedia.org/wiki/Modified_discrete_cosine_transform#Window_functions"
10470,hann_window,tensorflow/tensorflow/python/ops/signal/window_ops.py,151,function,"Generate a [Hann window][hann].
Args:
window_length: A scalar `Tensor` indicating the window length to generate.
periodic: A bool `Tensor` indicating whether to generate a periodic or
symmetric window. Periodic windows are typically used for spectral
analysis while symmetric windows are typically used for digital
filter design.
dtype: The data type to produce. Must be a floating point type.
name: An optional name for the operation.
Returns:
A `Tensor` of shape `[window_length]` of type `dtype`.
Raises:
ValueError: If `dtype` is not a floating point type.
[hann]: https://en.wikipedia.org/wiki/Window_function#Hann_and_Hamming_windows"
10471,hamming_window,tensorflow/tensorflow/python/ops/signal/window_ops.py,177,function,"Generate a [Hamming][hamming] window.
Args:
window_length: A scalar `Tensor` indicating the window length to generate.
periodic: A bool `Tensor` indicating whether to generate a periodic or
symmetric window. Periodic windows are typically used for spectral
analysis while symmetric windows are typically used for digital
filter design.
dtype: The data type to produce. Must be a floating point type.
name: An optional name for the operation.
Returns:
A `Tensor` of shape `[window_length]` of type `dtype`.
Raises:
ValueError: If `dtype` is not a floating point type.
[hamming]:
https://en.wikipedia.org/wiki/Window_function#Hann_and_Hamming_windows"
10472,StructuredTensor,tensorflow/tensorflow/python/ops/structured/structured_tensor.py,43,class,"A multidimensional collection of structures with the same schema.
A **`StructuredTensor`** is a multi-dimensional collection of ***structures***
with the same ***schema***, where:
* A ***schema*** is a collection of fields, each of which has a name and type.
* A ***structure*** maps each field in the schema to a tensor value (which
could be a nested StructuredTensor).
As an important special case, a 1D `StructuredTensor` encodes a 2D table,
where columns are heterogeneous `Tensor`s, and rows are the aligned elements
in each of those `Tensor`s.
Internally, StructuredTensors use a ""field-major"" encoding: for each leaf
field, there is a single tensor that stores the value of that field for all
structures in the `StructuredTensor`.
### Examples
>>> # A scalar StructuredTensor describing a single person.
>>> s1 = StructuredTensor.from_pyval(
... {""age"": 82, ""nicknames"": [""Bob"", ""Bobby""]})
>>> s1.shape
TensorShape([])
>>> s1[""age""]
<tf.Tensor: shape=(), dtype=int32, numpy=82>
>>> # A vector StructuredTensor describing three people.
>>> s2 = StructuredTensor.from_pyval([
... {""age"": 12, ""nicknames"": [""Josaphine""]},
... {""age"": 82, ""nicknames"": [""Bob"", ""Bobby""]},
... {""age"": 42, ""nicknames"": [""Elmo""]}])
>>> s2.shape
TensorShape([3])
>>> s2[0][""age""]
<tf.Tensor: shape=(), dtype=int32, numpy=12>
### Field Paths
A *field path* is a tuple of field names, specifying the path to a nested
field."
10473,from_fields,tensorflow/tensorflow/python/ops/structured/structured_tensor.py,122,method,"Creates a `StructuredTensor` from a dictionary of fields.
Args:
fields: A dictionary mapping from string to `Tensor`, `RaggedTensor`, or
`StructuredTensor`, providing the values for individual fields in each
structure. If `shape.rank > 0`, then every tensor in `fields` must have
the same shape in the first `shape.rank` dimensions; and that shape must
be compatible with `shape`; and
`result[i1...iN][key] = fields[key][i1...iN]` (where `N==shape.rank`).
shape: A `TensorShape`: static information about the shape of the
`StructuredTensor`. Must have a known `rank`. Defaults to scalar
shape (i.e. `rank=0`).
nrows: scalar integer tensor containing the number of rows in this
`StructuredTensor`. Should only be specified if `shape.rank > 0`.
Default value is inferred from the `fields` values. If `fields` is
empty, then this must be specified.
row_partitions: A list of `RowPartition`s describing the (possibly ragged)
shape of this `StructuredTensor`. Should only be specified if
`shape.rank > 1`. Default value is inferred from the `fields` values.
If `fields` is empty, then this must be specified.
validate: If true, then add runtime validation ops that check that the
field values all have compatible shapes in the outer `shape.rank`
dimensions.
Returns:
A `StructuredTensor`.
Examples:
>>> StructuredTensor.from_fields({'x': 1, 'y': [1, 2, 3]})
<StructuredTensor(
fields={
""x"": tf.Tensor(1, shape=(), dtype=int32),
""y"": tf.Tensor([1 2 3], shape=(3,), dtype=int32)},
shape=())>
>>> StructuredTensor.from_fields({'foo': [1, 2], 'bar': [3, 4]},
... shape=[2])
<StructuredTensor(
fields={
""bar"": tf.Tensor([3 4], shape=(2,), dtype=int32),
""foo"": tf.Tensor([1 2], shape=(2,), dtype=int32)},
shape=(2,))>"
10474,rank,tensorflow/tensorflow/python/ops/structured/structured_tensor.py,260,method,The rank of this StructuredTensor. Guaranteed not to be `None`.
10475,shape,tensorflow/tensorflow/python/ops/structured/structured_tensor.py,265,method,"The static shape of this StructuredTensor.
The returned `TensorShape` is guaranteed to have a known rank, but the
individual dimension sizes may be unknown.
Returns:
`tf.TensorShape`"
10476,row_partitions,tensorflow/tensorflow/python/ops/structured/structured_tensor.py,279,method,"A tuple of `RowPartition`s defining the shape of this `StructuredTensor`.
If this `StructuredTensor` has a ragged shape, then all fields will be
encoded as either `RaggedTensor`s or `StructuredTensor`s with these
`RowPartition`s used to define their outermost `self.rank` dimensions.
If this `StructuredTensor` has a uniform (non-ragged) shape, then these
row partitions will all be defined using `uniform_row_length`.
Returns:
A `tuple` of `RowPartition` objects with length `self.rank - 1`
(or `0` if `self.rank < 2`)."
10477,nrows,tensorflow/tensorflow/python/ops/structured/structured_tensor.py,295,method,"The number of rows in this StructuredTensor (if rank>0).
Returns:
A scalar integer `Tensor` (or `None` if `self.rank == 0`)."
10478,field_names,tensorflow/tensorflow/python/ops/structured/structured_tensor.py,312,method,Returns the string field names for this `StructuredTensor`.
10479,field_value,tensorflow/tensorflow/python/ops/structured/structured_tensor.py,316,method,"Returns the tensor value for the specified field or path.
If `field_name` is a `string`, then it names a field directly owned by this
`StructuredTensor`. If this `StructuredTensor` has shape `[D1...DN]`, then
the returned tensor will have shape `[D1...DN, V1...VM]`, where the slice
`result[d1...dN]` contains the field value for the structure at
`self[d1...dN]`.
If `field_name` is a `tuple` of `string`, then it specifies a path to a
field owned by nested `StructuredTensor`. In particular,
`struct.field_value((f1, f2, ..., fN))` is equivalent to
`struct.field_value(f1).field_value(f2)....field_value(fN)`
Args:
field_name: `string` or `tuple` of `string`: The field whose values should
be returned.
Returns:
`Tensor`, `StructuredTensor`, or `RaggedTensor`.
Raises:
KeyError: If the given field_name is not found."
10480,to_pyval,tensorflow/tensorflow/python/ops/structured/structured_tensor.py,454,method,"Returns this StructuredTensor as a nested Python dict or list of dicts.
Converts this `StructuredTensor` to a nested python value:
* `StructTensors` with `rank=0` are converted into a dictionary, with an
entry for each field. Field names are used as keys and field values are
converted to python values. In particular:
* Scalar Tensor fields are converted to simple values (such as
`int` or `float` or `string`)
* Non-scalar Tensor fields and RaggedTensor fields are converted to
nested lists of simple values.
* StructuredTensor fields are converted recursively using `to_pyval`.
* `StructTensors` with `rank>0` are converted to nested python `list`s,
containing one dictionary for each structure (where each structure's
dictionary is defined as described above).
Requires that all fields are Eager tensors.
>>> StructuredTensor.from_fields(
... {'a': [1, 2, 3]}, [3]).to_pyval()
[{'a': 1}, {'a': 2}, {'a': 3}]
Note that `StructuredTensor.from_pyval(pyval).to_pyval() == pyval`.
Returns:
A nested Python dict or list of dicts."
10481,from_pyval,tensorflow/tensorflow/python/ops/structured/structured_tensor.py,510,method,"Constructs a StructuredTensor from a nested Python structure.
>>> StructuredTensor.from_pyval(
... {'a': [1, 2, 3], 'b': [[4, 5], [6, 7]]})
<StructuredTensor(
fields={
""a"": tf.Tensor([1 2 3], shape=(3,), dtype=int32),
""b"": <tf.RaggedTensor [[4, 5], [6, 7]]>},
shape=())>
Note that `StructuredTensor.from_pyval(pyval).to_pyval() == pyval`.
Args:
pyval: The nested Python structure that should be used to create the new
`StructuredTensor`.
typespec: A `StructuredTensorSpec` specifying the expected type for each
field. If not specified, then all nested dictionaries are turned into
StructuredTensors, and all nested lists are turned into Tensors (if
rank<2) or RaggedTensors (if rank>=2).
Returns:
A `StructuredTensor`."
10482,partition_outer_dimension,tensorflow/tensorflow/python/ops/structured/structured_tensor.py,637,method,"Partitions the outer dimension of this StructuredTensor.
Returns a new `StructuredTensor` with the same values as `self`, where
the outer dimension is partitioned into two (possibly ragged) dimensions.
Requires that this StructuredTensor have an outer dimension (i.e.,
`self.shape.rank > 0`).
>>> st = StructuredTensor.from_pyval(
... [{'foo': 12}, {'foo': 33}, {'foo': 99}])
>>> partition = RowPartition.from_row_lengths([2, 0, 1])
>>> st.partition_outer_dimension(partition)
<StructuredTensor(
fields={
""foo"": <tf.RaggedTensor [[12, 33], [], [99]]>},
shape=(3, None))>
Args:
row_partition: A `RowPartition`.
Returns:
A `StructuredTensor` with rank `values.rank + 1`."
10483,merge_dims,tensorflow/tensorflow/python/ops/structured/structured_tensor.py,666,method,"Merges outer_axis...inner_axis into a single dimension.
Returns a copy of this RaggedTensor with the specified range of dimensions
flattened into a single dimension, with elements in row-major order.
>>> st = StructuredTensor.from_pyval(
... [[{'foo': 12}, {'foo': 33}], [], [{'foo': 99}]])
>>> st.merge_dims(0, 1)
<StructuredTensor(
fields={
""foo"": tf.Tensor([12 33 99], shape=(3,), dtype=int32)},
shape=(3,))>
Args:
outer_axis: `int`: The first dimension in the range of dimensions to
merge. May be negative (to index from the last dimension).
inner_axis: `int`: The last dimension in the range of dimensions to merge.
May be negative (to index from the last dimension).
Returns:
A copy of this tensor, with the specified dimensions merged into a
single dimension. The shape of the returned tensor will be
`self.shape[:outer_axis] + [N] + self.shape[inner_axis + 1:]`, where `N`
is the total number of slices in the merged dimensions."
10484,StructuredTensorSpec,tensorflow/tensorflow/python/ops/structured/structured_tensor.py,716,class,Type specification for `StructuredTensor`s.
10485,value_type,tensorflow/tensorflow/python/ops/structured/structured_tensor.py,752,method,
10486,from_value,tensorflow/tensorflow/python/ops/structured/structured_tensor.py,766,method,
10487,track_usage,tensorflow/tensorflow/python/platform/analytics.py,21,function,"No usage tracking for external library.
Args:
tool_id: A string identifier for tool to be tracked.
tags: list of string tags that will be added to the tracking."
10488,run,tensorflow/tensorflow/python/platform/app.py,35,function,Runs the program with an optional 'main' function and 'argv' list.
10489,ParameterizedBenchmark,tensorflow/tensorflow/python/platform/benchmark.py,171,class,Metaclass to generate parameterized benchmarks.
10490,create_benchmark_function,tensorflow/tensorflow/python/platform/benchmark.py,192,method,
10491,Benchmark,tensorflow/tensorflow/python/platform/benchmark.py,203,class,"Abstract class that provides helper functions for running benchmarks.
Any class subclassing this one is immediately registered in the global
benchmark registry.
Only methods whose names start with the word ""benchmark"" will be run during
benchmarking."
10492,is_abstract,tensorflow/tensorflow/python/platform/benchmark.py,214,method,
10493,report_benchmark,tensorflow/tensorflow/python/platform/benchmark.py,242,method,"Report a benchmark.
Args:
iters: (optional) How many iterations were run
cpu_time: (optional) Median or mean cpu time in seconds.
wall_time: (optional) Median or mean wall time in seconds.
throughput: (optional) Throughput (in MB/s)
extras: (optional) Dict mapping string keys to additional benchmark info.
Values may be either floats or values that are convertible to strings.
name: (optional) Override the BenchmarkEntry name with `name`.
Otherwise it is inferred from the top-level method name.
metrics: (optional) A list of dict, where each dict has the keys below
name (required), string, metric name
value (required), double, metric value
min_value (optional), double, minimum acceptable metric value
max_value (optional), double, maximum acceptable metric value"
10494,benchmark_config,tensorflow/tensorflow/python/platform/benchmark.py,275,function,"Returns a tf.compat.v1.ConfigProto for disabling the dependency optimizer.
Returns:
A TensorFlow ConfigProto object."
10495,TensorFlowBenchmark,tensorflow/tensorflow/python/platform/benchmark.py,288,class,Abstract class that provides helpers for TensorFlow benchmarks.
10496,is_abstract,tensorflow/tensorflow/python/platform/benchmark.py,298,method,
10497,run_op_benchmark,tensorflow/tensorflow/python/platform/benchmark.py,303,method,"Run an op or tensor in the given session. Report the results.
Args:
sess: `Session` object to use for timing.
op_or_tensor: `Operation` or `Tensor` to benchmark.
feed_dict: A `dict` of values to feed for each op iteration (see the
`feed_dict` parameter of `Session.run`).
burn_iters: Number of burn-in iterations to run.
min_iters: Minimum number of iterations to use for timing.
store_trace: Boolean, whether to run an extra untimed iteration and
store the trace of iteration in returned extras.
The trace will be stored as a string in Google Chrome trace format
in the extras field ""full_trace_chrome_format"". Note that trace
will not be stored in test_log_pb2.TestResults proto.
store_memory_usage: Boolean, whether to run an extra untimed iteration,
calculate memory usage, and store that in extras fields.
name: (optional) Override the BenchmarkEntry name with `name`.
Otherwise it is inferred from the top-level method name.
extras: (optional) Dict mapping string keys to additional benchmark info.
Values may be either floats or values that are convertible to strings.
mbs: (optional) The number of megabytes moved by this op, used to
calculate the ops throughput.
Returns:
A `dict` containing the key-value pairs that were passed to
`report_benchmark`. If `store_trace` option is used, then
`full_chrome_trace_format` will be included in return dictionary even
though it is not passed to `report_benchmark` with `extras`."
10498,evaluate,tensorflow/tensorflow/python/platform/benchmark.py,410,method,"Evaluates tensors and returns numpy values.
Args:
tensors: A Tensor or a nested list/tuple of Tensors.
Returns:
tensors numpy values."
10499,enclosing_tpu_context,tensorflow/tensorflow/python/platform/device_context.py,21,function,
10500,GFile,tensorflow/tensorflow/python/platform/gfile.py,41,class,"File I/O wrappers without thread locking.
The main roles of the `tf.io.gfile` module are:
1. To provide an API that is close to Python's file I/O objects, and
2. To provide an implementation based on TensorFlow's C++ FileSystem API.
The C++ FileSystem API supports multiple file system implementations,
including local files, Google Cloud Storage (using a `gs://` prefix, and
HDFS (using an `hdfs://` prefix). TensorFlow exports these as `tf.io.gfile`,
so that you can use these implementations for saving and loading checkpoints,
writing to TensorBoard logs, and accessing training data (among other uses).
However, if all your files are local, you can use the regular Python file
API without any problem.
*Note*: though similar to Python's I/O implementation, there are semantic
differences to make `tf.io.gfile` more efficient for backing filesystems. For
example, a write mode file will not be opened until the first write call, to
minimize RPC invocations in network filesystems."
10501,FastGFile,tensorflow/tensorflow/python/platform/gfile.py,68,class,"File I/O wrappers without thread locking.
Note, that this is somewhat like builtin Python file I/O, but
there are semantic differences to make it more efficient for
some backing filesystems. For example, a write mode file will
not be opened until the first write call (to minimize RPC
invocations in network filesystems)."
10502,GetTempDir,tensorflow/tensorflow/python/platform/googletest.py,69,function,Return a temporary directory for tests to use.
10503,StatefulSessionAvailable,tensorflow/tensorflow/python/platform/googletest.py,111,function,
10504,get_default_communication_protocol,tensorflow/tensorflow/python/platform/remote_utils.py,21,function,
10505,load_resource,tensorflow/tensorflow/python/platform/resource_loader.py,35,function,"Load the resource at given path, where path is relative to tensorflow/.
Args:
path: a string resource path relative to tensorflow/.
Returns:
The contents of that resource.
Raises:
IOError: If the path is not found, or the resource can't be opened."
10506,get_data_files_path,tensorflow/tensorflow/python/platform/resource_loader.py,53,function,"Get a direct path to the data files colocated with the script.
Returns:
The directory where files specified in data attribute of py_test
and py_binary are stored."
10507,get_root_dir_with_all_resources,tensorflow/tensorflow/python/platform/resource_loader.py,64,function,"Get a root directory containing all the data attributes in the build rule.
Returns:
The path to the specified file present in the data attribute of py_test
or py_binary. Falls back to returning the same as get_data_files_path if it
fails to detect a bazel runfiles directory."
10508,get_path_to_datafile,tensorflow/tensorflow/python/platform/resource_loader.py,104,function,"Get the path to the specified file in the data dependencies.
The path is relative to tensorflow/
Args:
path: a string resource path relative to tensorflow/
Returns:
The path to the specified file present in the data attribute of py_test
or py_binary.
Raises:
IOError: If the path is not found, or the resource can't be opened."
10509,readahead_file_path,tensorflow/tensorflow/python/platform/resource_loader.py,134,function,Readahead files not implemented; simply returns given path.
10510,preload_check,tensorflow/tensorflow/python/platform/self_check.py,34,function,"Raises an exception if the environment is not correctly configured.
Raises:
ImportError: If the check detects that the environment is not correctly
configured, and attempting to load the TensorFlow runtime will fail."
10511,SetupStatusBarInsideGoogle,tensorflow/tensorflow/python/platform/status_bar.py,23,function,
10512,get_include,tensorflow/tensorflow/python/platform/sysconfig.py,33,function,"Get the directory containing the TensorFlow C++ header files.
Returns:
The directory as string."
10513,get_lib,tensorflow/tensorflow/python/platform/sysconfig.py,48,function,"Get the directory containing the TensorFlow framework library.
Returns:
The directory as string."
10514,get_compile_flags,tensorflow/tensorflow/python/platform/sysconfig.py,59,function,"Get the compilation flags for custom operators.
Returns:
The compilation flags."
10515,get_link_flags,tensorflow/tensorflow/python/platform/sysconfig.py,72,function,"Get the link flags for custom operators.
Returns:
The link flags."
10516,get_build_info,tensorflow/tensorflow/python/platform/sysconfig.py,91,function,"Get a dictionary describing TensorFlow's build environment.
Values are generated when TensorFlow is compiled, and are static for each
TensorFlow package. The return value is a dictionary with string keys such as:
- cuda_version
- cudnn_version
- is_cuda_build
- is_rocm_build
- msvcp_dll_names
- nvcuda_dll_name
- cudart_dll_name
- cudnn_dll_name
Note that the actual keys and values returned by this function is subject to
change across different versions of TensorFlow or across platforms.
Returns:
A Dictionary describing TensorFlow's build environment."
10517,get_temp_dir,tensorflow/tensorflow/python/platform/test.py,62,function,"Returns a temporary directory for use during tests.
There is no need to delete the directory after the test.
Returns:
The temporary directory."
10518,is_built_with_cuda,tensorflow/tensorflow/python/platform/test.py,88,function,Returns whether TensorFlow was built with CUDA (GPU) support.
10519,is_built_with_rocm,tensorflow/tensorflow/python/platform/test.py,94,function,Returns whether TensorFlow was built with ROCm (GPU) support.
10520,is_built_with_gpu_support,tensorflow/tensorflow/python/platform/test.py,100,function,Returns whether TensorFlow was built with GPU (i.e. CUDA or ROCm) support.
10521,is_built_with_xla,tensorflow/tensorflow/python/platform/test.py,106,function,Returns whether TensorFlow was built with XLA support.
10522,get_logger,tensorflow/tensorflow/python/platform/tf_logging.py,94,function,Return TF logger instance.
10523,log,tensorflow/tensorflow/python/platform/tf_logging.py,147,function,
10524,debug,tensorflow/tensorflow/python/platform/tf_logging.py,152,function,
10525,error,tensorflow/tensorflow/python/platform/tf_logging.py,157,function,
10526,fatal,tensorflow/tensorflow/python/platform/tf_logging.py,162,function,
10527,info,tensorflow/tensorflow/python/platform/tf_logging.py,167,function,
10528,warn,tensorflow/tensorflow/python/platform/tf_logging.py,172,function,
10529,warning,tensorflow/tensorflow/python/platform/tf_logging.py,177,function,
10530,TaskLevelStatusMessage,tensorflow/tensorflow/python/platform/tf_logging.py,200,function,
10531,flush,tensorflow/tensorflow/python/platform/tf_logging.py,205,function,
10532,vlog,tensorflow/tensorflow/python/platform/tf_logging.py,211,function,
10533,log_every_n,tensorflow/tensorflow/python/platform/tf_logging.py,231,function,"Log 'msg % args' at level 'level' once per 'n' times.
Logs the 1st call, (N+1)st call, (2N+1)st call, etc.
Not threadsafe.
Args:
level: The level at which to log.
msg: The message to be logged.
n: The number of times this should be called before it is logged.
*args: The args to be substituted into the msg."
10534,log_first_n,tensorflow/tensorflow/python/platform/tf_logging.py,248,function,"Log 'msg % args' at level 'level' only first 'n' times.
Not threadsafe.
Args:
level: The level at which to log.
msg: The message to be logged.
n: The number of times this should be called before it is logged.
*args: The args to be substituted into the msg."
10535,log_if,tensorflow/tensorflow/python/platform/tf_logging.py,264,function,Log 'msg % args' at level 'level' only if condition is fulfilled.
10536,google2_log_prefix,tensorflow/tensorflow/python/platform/tf_logging.py,278,function,Assemble a logline prefix using the google2 format.
10537,get_verbosity,tensorflow/tensorflow/python/platform/tf_logging.py,313,function,Return how much logging output will be produced.
10538,set_verbosity,tensorflow/tensorflow/python/platform/tf_logging.py,319,function,Sets the threshold for what messages will be logged.
10539,Profiler,tensorflow/tensorflow/python/profiler/model_analyzer.py,126,class,"TensorFlow multi-step profiler.
https://github.com/tensorflow/tensorflow/tree/master/tensorflow/core/profiler/README.md
```python
Typical use case:
# Currently we are only allowed to create 1 profiler per process.
profiler = Profiler(sess.graph)
for i in xrange(total_steps):
if i % 10000 == 0:
run_meta = tf.compat.v1.RunMetadata()
_ = sess.run(...,
options=tf.compat.v1.RunOptions(
trace_level=tf.RunOptions.FULL_TRACE),
run_metadata=run_meta)
profiler.add_step(i, run_meta)
# Profile the parameters of your model.
profiler.profile_name_scope(options=(option_builder.ProfileOptionBuilder
.trainable_variables_parameter()))
# Or profile the timing of your model operations.
opts = option_builder.ProfileOptionBuilder.time_and_memory()
profiler.profile_operations(options=opts)
# Or you can generate a timeline:
opts = (option_builder.ProfileOptionBuilder(
option_builder.ProfileOptionBuilder.time_and_memory())
.with_step(i)
.with_timeline_output(filename).build())
profiler.profile_graph(options=opts)
else:
_ = sess.run(...)
# Auto detect problems and generate advice.
profiler.advise()
```"
10540,add_step,tensorflow/tensorflow/python/profiler/model_analyzer.py,189,method,"Add statistics of a step.
Args:
step: int, An id used to group one or more different `run_meta` together.
When profiling with the profile_xxx APIs, user can use the `step`
id in the `options` to profile these `run_meta` together.
run_meta: RunMetadata proto that contains statistics of a session run."
10541,profile_python,tensorflow/tensorflow/python/profiler/model_analyzer.py,207,method,"Profile the statistics of the Python codes.
By default, it shows the call stack from root. To avoid
redundant output, you may use options to filter as below
options['show_name_regexes'] = ['.*my_code.py.*']
Args:
options: A dict of options. See core/profiler/g3doc/options.md.
Returns:
a MultiGraphNodeProto that records the results."
10542,profile_operations,tensorflow/tensorflow/python/profiler/model_analyzer.py,228,method,"Profile the statistics of the Operation types (e.g. MatMul, Conv2D).
Args:
options: A dict of options. See core/profiler/g3doc/options.md.
Returns:
a MultiGraphNodeProto that records the results."
10543,profile_name_scope,tensorflow/tensorflow/python/profiler/model_analyzer.py,245,method,"Profile the statistics of graph nodes, organized by name scope.
Args:
options: A dict of options. See core/profiler/g3doc/options.md.
Returns:
a GraphNodeProto that records the results."
10544,profile_graph,tensorflow/tensorflow/python/profiler/model_analyzer.py,262,method,"Profile the statistics of graph nodes, organized by dataflow graph.
Args:
options: A dict of options. See core/profiler/g3doc/options.md.
Returns:
a GraphNodeProto that records the results."
10545,advise,tensorflow/tensorflow/python/profiler/model_analyzer.py,279,method,"Automatically detect problems and generate reports.
Args:
options: A dict of options. See ALL_ADVICE example above.
Returns:
An Advise proto that contains the reports from all checkers."
10546,serialize_to_string,tensorflow/tensorflow/python/profiler/model_analyzer.py,293,method,"Serialize the ProfileProto to a binary string.
Users can write it to file for offline analysis by tfprof commandline
or graphical interface.
Returns:
ProfileProto binary string."
10547,profile,tensorflow/tensorflow/python/profiler/model_analyzer.py,310,function,"Profile model.
Tutorials and examples can be found in:
https://github.com/tensorflow/tensorflow/blob/master/tensorflow/core/profiler/g3doc/python_api.md
Args:
graph: tf.Graph. If None and eager execution is not enabled, use
default graph.
run_meta: optional tensorflow.RunMetadata proto. It is necessary to
to support run time information profiling, such as time and memory.
op_log: tensorflow.tfprof.OpLogProto proto. User can assign ""types"" to
graph nodes with op_log. ""types"" allow user to flexibly group and
account profiles using options['accounted_type_regexes'].
cmd: string. Either 'op', 'scope', 'graph' or 'code'.
'op' view organizes profile using operation type. (e.g. MatMul)
'scope' view organizes profile using graph node name scope.
'graph' view organizes profile using graph node inputs/outputs.
'code' view organizes profile using Python call stack.
options: A dict of options. See core/profiler/g3doc/options.md.
Returns:
If cmd is 'scope' or 'graph', returns GraphNodeProto proto.
If cmd is 'op' or 'code', returns MultiGraphNodeProto proto.
Side effect: stdout/file/timeline.json depending on options['output']"
10548,advise,tensorflow/tensorflow/python/profiler/model_analyzer.py,385,function,"Auto profile and advise.
Builds profiles and automatically check anomalies of various
aspects. For more details:
https://github.com/tensorflow/tensorflow/tree/master/tensorflow/core/profiler/README.md
Args:
graph: tf.Graph. If None and eager execution is not enabled, use
default graph.
run_meta: optional tensorflow.RunMetadata proto. It is necessary to
to support run time information profiling, such as time and memory.
options: see ALL_ADVICE example above. Default checks everything.
Returns:
Returns AdviceProto proto"
10549,ProfileOptionBuilder,tensorflow/tensorflow/python/profiler/option_builder.py,27,class,"Option Builder for Profiling API.
For tutorial on the options, see
https://github.com/tensorflow/tensorflow/tree/master/tensorflow/core/profiler/g3doc/options.md
```python
# Users can use pre-built options:
opts = (
tf.profiler.ProfileOptionBuilder.trainable_variables_parameter())
# Or, build your own options:
opts = (tf.compat.v1.profiler.ProfileOptionBuilder()
.with_max_depth(10)
.with_min_micros(1000)
.select(['accelerator_micros'])
.with_stdout_output()
.build()
# Or customize the pre-built options:
opts = (tf.compat.v1.profiler.ProfileOptionBuilder(
tf.profiler.ProfileOptionBuilder.time_and_memory())
.with_displaying_options(show_name_regexes=['.*rnn.*'])
.build())
# Finally, profiling with the options:
_ = tf.compat.v1.profiler.profile(tf.compat.v1.get_default_graph(),
run_meta=run_meta,
cmd='scope',
options=opts)
```"
10550,trainable_variables_parameter,tensorflow/tensorflow/python/profiler/option_builder.py,89,method,"Options used to profile trainable variable parameters.
Normally used together with 'scope' view.
Returns:
A dict of profiling options."
10551,float_operation,tensorflow/tensorflow/python/profiler/option_builder.py,115,method,"Options used to profile float operations.
Please see https://github.com/tensorflow/tensorflow/tree/master/tensorflow/core/profiler/g3doc/profile_model_architecture.md
on the caveats of calculating float operations.
Returns:
A dict of profiling options."
10552,time_and_memory,tensorflow/tensorflow/python/profiler/option_builder.py,144,method,"Show operation time and memory consumptions.
Args:
min_micros: Only show profiler nodes with execution time
no less than this. It sums accelerator and cpu times.
min_bytes: Only show profiler nodes requested to allocate no less bytes
than this.
min_accelerator_micros: Only show profiler nodes spend no less than
this time on accelerator (e.g. GPU).
min_cpu_micros: Only show profiler nodes spend no less than
this time on cpu.
min_peak_bytes: Only show profiler nodes using no less than this bytes
at peak (high watermark). For profiler nodes consist of multiple
graph nodes, it sums the graph nodes' peak_bytes.
min_residual_bytes: Only show profiler nodes have no less than
this bytes not being de-allocated after Compute() ends. For
profiler nodes consist of multiple graph nodes, it sums the
graph nodes' residual_bytes.
min_output_bytes: Only show profiler nodes have no less than this bytes
output. The output are not necessarily allocated by this profiler
nodes.
Returns:
A dict of profiling options."
10553,build,tensorflow/tensorflow/python/profiler/option_builder.py,193,method,"Build a profiling option.
Returns:
A dict of profiling options."
10554,with_max_depth,tensorflow/tensorflow/python/profiler/option_builder.py,201,method,"Set the maximum depth of display.
The depth depends on profiling view. For 'scope' view, it's the
depth of name scope hierarchy (tree), for 'op' view, it's the number
of operation types (list), etc.
Args:
max_depth: Maximum depth of the data structure to display.
Returns:
self"
10555,with_min_memory,tensorflow/tensorflow/python/profiler/option_builder.py,216,method,"Only show profiler nodes consuming no less than 'min_bytes'.
Args:
min_bytes: Only show profiler nodes requested to allocate no less bytes
than this.
min_peak_bytes: Only show profiler nodes using no less than this bytes
at peak (high watermark). For profiler nodes consist of multiple
graph nodes, it sums the graph nodes' peak_bytes.
min_residual_bytes: Only show profiler nodes have no less than
this bytes not being de-allocated after Compute() ends. For
profiler nodes consist of multiple graph nodes, it sums the
graph nodes' residual_bytes.
min_output_bytes: Only show profiler nodes have no less than this bytes
output. The output are not necessarily allocated by this profiler
nodes.
Returns:
self"
10556,with_min_execution_time,tensorflow/tensorflow/python/profiler/option_builder.py,245,method,"Only show profiler nodes consuming no less than 'min_micros'.
Args:
min_micros: Only show profiler nodes with execution time
no less than this. It sums accelerator and cpu times.
min_accelerator_micros: Only show profiler nodes spend no less than
this time on accelerator (e.g. GPU).
min_cpu_micros: Only show profiler nodes spend no less than
this time on cpu.
Returns:
self"
10557,with_min_parameters,tensorflow/tensorflow/python/profiler/option_builder.py,266,method,"Only show profiler nodes holding no less than 'min_params' parameters.
'Parameters' normally refers the weights of in TensorFlow variables.
It reflects the 'capacity' of models.
Args:
min_params: Only show profiler nodes holding number parameters
no less than this.
Returns:
self"
10558,with_min_occurrence,tensorflow/tensorflow/python/profiler/option_builder.py,281,method,"Only show profiler nodes including no less than 'min_occurrence' graph nodes.
A ""node"" means a profiler output node, which can be a python line
(code view), an operation type (op view), or a graph node
(graph/scope view). A python line includes all graph nodes created by that
line, while an operation type includes all graph nodes of that type.
Args:
min_occurrence: Only show nodes including no less than this.
Returns:
self"
10559,with_min_float_operations,tensorflow/tensorflow/python/profiler/option_builder.py,299,method,"Only show profiler nodes consuming no less than 'min_float_ops'.
Please see https://github.com/tensorflow/tensorflow/tree/master/tensorflow/core/profiler/g3doc/profile_model_architecture.md
on the caveats of calculating float operations.
Args:
min_float_ops: Only show profiler nodes with float operations
no less than this.
Returns:
self"
10560,with_accounted_types,tensorflow/tensorflow/python/profiler/option_builder.py,316,method,"Selectively counting statistics based on node types.
Here, 'types' means the profiler nodes' properties. Profiler by default
consider device name (e.g. /job:xx/.../device:GPU:0) and operation type
(e.g. MatMul) as profiler nodes' properties. User can also associate
customized 'types' to profiler nodes through OpLogProto proto.
For example, user can select profiler nodes placed on gpu:0 with:
`account_type_regexes=['.*gpu:0.*']`
If none of a node's properties match the specified regexes, the node is
not displayed nor accounted.
Args:
account_type_regexes: A list of regexes specifying the types.
Returns:
self."
10561,with_node_names,tensorflow/tensorflow/python/profiler/option_builder.py,338,method,"Regular expressions used to select profiler nodes to display.
After 'with_accounted_types' is evaluated, 'with_node_names' are
evaluated as follows:
For a profile data structure, profiler first finds the profiler
nodes matching 'start_name_regexes', and starts displaying profiler
nodes from there. Then, if a node matches 'show_name_regexes' and
doesn't match 'hide_name_regexes', it's displayed. If a node matches
'trim_name_regexes', profiler stops further searching that branch.
Args:
start_name_regexes: list of node name regexes to start displaying.
show_name_regexes: list of node names regexes to display.
hide_name_regexes: list of node_names regexes that should be hidden.
trim_name_regexes: list of node name regexes from where to stop.
Returns:
self"
10562,account_displayed_op_only,tensorflow/tensorflow/python/profiler/option_builder.py,372,method,"Whether only account the statistics of displayed profiler nodes.
Args:
is_true: If true, only account statistics of nodes eventually
displayed by the outputs.
Otherwise, a node's statistics are accounted by its parents
as long as it's types match 'account_type_regexes', even if
it is hidden from the output, say, by hide_name_regexes.
Returns:
self"
10563,with_empty_output,tensorflow/tensorflow/python/profiler/option_builder.py,387,method,Do not generate side-effect outputs.
10564,with_stdout_output,tensorflow/tensorflow/python/profiler/option_builder.py,392,method,Print the result to stdout.
10565,with_file_output,tensorflow/tensorflow/python/profiler/option_builder.py,397,method,Print the result to a file.
10566,with_timeline_output,tensorflow/tensorflow/python/profiler/option_builder.py,402,method,Generate a timeline json file.
10567,with_pprof_output,tensorflow/tensorflow/python/profiler/option_builder.py,407,method,"Generate a pprof profile gzip file.
To use the pprof file:
pprof -png --nodecount=100 --sample_index=1 <pprof_file>
Args:
pprof_file: filename for output, usually suffixed with .pb.gz.
Returns:
self."
10568,order_by,tensorflow/tensorflow/python/profiler/option_builder.py,421,method,"Order the displayed profiler nodes based on a attribute.
Supported attribute includes micros, bytes, occurrence, params, etc.
https://github.com/tensorflow/tensorflow/tree/master/tensorflow/core/profiler/g3doc/options.md
Args:
attribute: An attribute the profiler node has.
Returns:
self"
10569,select,tensorflow/tensorflow/python/profiler/option_builder.py,437,method,"Select the attributes to display.
See https://github.com/tensorflow/tensorflow/tree/master/tensorflow/core/profiler/g3doc/options.md
for supported attributes.
Args:
attributes: A list of attribute the profiler node has.
Returns:
self"
10570,with_step,tensorflow/tensorflow/python/profiler/option_builder.py,453,method,"Which profile step to use for profiling.
The 'step' here refers to the step defined by `Profiler.add_step()` API.
Args:
step: When multiple steps of profiles are available, select which step's
profile to use. If -1, use average of all available steps.
Returns:
self"
10571,StringTable,tensorflow/tensorflow/python/profiler/pprof_profiler.py,63,class,Keeps track of strings to add to string_table in pprof proto.
10572,index_of,tensorflow/tensorflow/python/profiler/pprof_profiler.py,71,method,"Get index of value_str in the string table.
If value_str is not in the string table, we will add it at the end
and then return the new index.
Args:
value_str: (string) Value to lookup/add in/to the string table.
Returns:
Index of value_str in the string table."
10573,next_index,tensorflow/tensorflow/python/profiler/pprof_profiler.py,91,method,"Gets index that would be assigned to the next added string.
Returns:
Index of the next string if it was added."
10574,string_table,tensorflow/tensorflow/python/profiler/pprof_profiler.py,99,method,Returns a list of strings to store in pprof's string_table.
10575,Functions,tensorflow/tensorflow/python/profiler/pprof_profiler.py,104,class,Keeps track of `Function` protos for pprof profile.
10576,index_of,tensorflow/tensorflow/python/profiler/pprof_profiler.py,118,method,"Returns index of the function, adding the function if needed.
Args:
file_path: (string) Path to file where the function is defined.
function_name: (string) Function name.
function_start_line: (integer) Start line number of function definition.
Returns:
Function index."
10577,function_protos,tensorflow/tensorflow/python/profiler/pprof_profiler.py,143,method,Returns list of `profile_pb2.Function` protos.
10578,Locations,tensorflow/tensorflow/python/profiler/pprof_profiler.py,148,class,"Keeps track of `Location` protos for pprof profile.
`Locations` store information about function call locations."
10579,index_of,tensorflow/tensorflow/python/profiler/pprof_profiler.py,165,method,"Returns index of the location, adding the location if needed.
Args:
file_path: (string) Path to file that makes the call.
line_number: (integer) Call line number.
called_function_name: (string) Function name of the function called at
`file_path` and `line_number`.
called_file_path: (string) Path to file where the called function is
defined.
called_function_start_line: (integer) Start line number of called
function definition in `called_file_path` file.
Returns:
Index of location."
10580,location_protos,tensorflow/tensorflow/python/profiler/pprof_profiler.py,200,method,Returns list of `profile_pb2.Location` protos.
10581,Samples,tensorflow/tensorflow/python/profiler/pprof_profiler.py,205,class,"Keeps track of `Sample` protos for pprof profile.
Samples store the following statistics in order:
count, all_time, op_time"
10582,add,tensorflow/tensorflow/python/profiler/pprof_profiler.py,223,method,"Adds a sample data point.
Args:
datum: `ProfileDatum` to add a sample for.
location_ids: List of numberic location ids for this
sample."
10583,get_sample_protos,tensorflow/tensorflow/python/profiler/pprof_profiler.py,253,method,Returns list of `Sample` protos for pprof profile.
10584,PprofProfiler,tensorflow/tensorflow/python/profiler/pprof_profiler.py,258,class,Creates profiles in pprof format.
10585,profile,tensorflow/tensorflow/python/profiler/pprof_profiler.py,274,method,"Generates pprof profiles.
Returns:
Dictionary mapping from device name to proto in `profile_pb2.Profile`
format."
10586,profile_data_generator,tensorflow/tensorflow/python/profiler/pprof_profiler.py,377,method,
10587,get_profiles,tensorflow/tensorflow/python/profiler/pprof_profiler.py,389,function,"Generate profiles in pprof format.
See https://github.com/google/pprof/blob/master/proto/profile.proto
for pprof proto format.
Args:
graph: A `Graph` object.
run_metadata: A `RunMetadata` proto.
Returns:
A dictionary mapping from device name to pprof proto for that device."
10588,profile,tensorflow/tensorflow/python/profiler/pprof_profiler.py,405,function,"Generate profiles in pprof format.
See https://github.com/google/pprof/blob/master/proto/profile.proto
for pprof proto format.
Args:
graph: A `Graph` object.
run_metadata: A `RunMetadata` proto.
output_dir: (string) Directory to output pprof profile to.
Profile files for each device will be stored in compressed
serialized proto format. If output_dir is None, profile protos
will be printed to stdout instead.
Returns:
List of output files created by this profile call.
(Note: this list will be empty if output_dir is None)"
10589,ProfileContext,tensorflow/tensorflow/python/profiler/profile_context.py,113,class,"A Context that captures RunMetadata and performs profiling.
```python
# Trace steps 100~200, profile at [150, 200] and dump profile at 200.
with profile_context.ProfileContext('/tmp/train_dir',
trace_steps=range(100, 200, 3),
dump_steps=[200]) as pctx:
opts = tf.profiler.ProfileOptionBuilder.time_and_memory()
pctx.add_auto_profiling('op', opts, [150, 200])
train_loop().
# Tracing only.
with profile_context.tfprof.ProfileContext('/tmp/train_dir') as pctx:
# Run train/eval loop for at least few hundred steps. Profiles will be
# dumped to train_dir. Use web UI or command line to do profiling.
train_loop().
# When session object is available, do explicit trace, profile and dump.
with profile_context.ProfileContext('/tmp/train_dir',
trace_steps=[],
dump_steps=[]) as pctx:
opts = tf.profiler.ProfileOptionBuilder.time_and_memory()
pctx.trace_next_step()
_ = session.run(train_op)
pctx.profiler.profile_operations(options=opts)
```
Args:
profile_dir: Directory to store profiles.
trace_steps: A list of session run steps to trace. If None, use
pre-defined steps.
dump_steps: A list of steps to dump the profile to `profile_dir`. If None,
use pre-defined steps.
enabled: If false, everything is disabled with minimal overhead. It allows
user to only enable profiling when needed.
debug: If true, also dumps the raw trace RunMetadata text file to
profile_dir. And print debugging message. Useful for bug report."
10590,get_profiles,tensorflow/tensorflow/python/profiler/profile_context.py,194,method,"Returns profiling results for each step at which `cmd` was run.
Args:
cmd: string, profiling command used in an `add_auto_profiling` call.
Returns:
dict[int: (MultiGraphNodeProto | GraphNodeProto)]. Keys are steps at which
the profiling command was run. Values are the outputs of profiling.
For ""code"" and ""op"" commands this will be a `MultiGraphNodeProto`, for
""scope"" and ""graph"" commands this will be a `GraphNodeProto.
Raises:
ValueError: if `cmd` was never run (either because no session.run call was
made or because there was no `add_auto_profiling` call with the specified
`cmd`."
10591,add_auto_profiling,tensorflow/tensorflow/python/profiler/profile_context.py,215,method,"Traces and profiles at some session run steps.
Args:
cmd: The profiling commands. (i.e. scope, op, python, graph)
options: The profiling options.
profile_steps: A list/set of integers. The profiling command and options
will be run automatically at these integer steps. Each step is
a session.run."
10592,profiler,tensorflow/tensorflow/python/profiler/profile_context.py,232,method,Returns the current profiler object.
10593,trace_next_step,tensorflow/tensorflow/python/profiler/profile_context.py,240,method,Enables tracing and adds traces to profiler at next step.
10594,dump_next_step,tensorflow/tensorflow/python/profiler/profile_context.py,247,method,Enable tracing and dump profiles at next step.
10595,trace,tensorflow/tensorflow/python/profiler/profiler_client.py,29,function,"Sends grpc requests to profiler server to perform on-demand profiling.
This method will block caller thread until it receives tracing result. This
method supports CPU, GPU, and Cloud TPU. This method supports profiling a
single host for CPU, GPU, TPU, as well as multiple TPU workers.
The profiled results will be saved to your specified TensorBoard log
directory (e.g. the directory you save your model checkpoints). Use the
TensorBoard profile plugin to view the visualization and analysis results.
Args:
service_addr: gRPC address of profiler service e.g. grpc://localhost:6009.
logdir: Path of TensorBoard log directory e.g. /tmp/tb_log.
duration_ms: Duration of tracing or monitoring in ms.
worker_list: Optional. The list of workers that we are about to profile in
the current session (TPU only).
num_tracing_attempts: Optional. Automatically retry N times when no trace
event is collected (default 3).
options: profiler.experimental.ProfilerOptions namedtuple for miscellaneous
profiler options.
Raises:
UnavailableError: If no trace event is collected.
Example usage (CPU/GPU):
# Start a profiler server before your model runs.
```python
tf.profiler.experimental.server.start(6009)
# your model code.
# Send gRPC request to the profiler server to collect a trace of your model.
```python
tf.profiler.experimental.client.trace('grpc://localhost:6009',
'/tmp/tb_log', 2000)
Example usage (TPU):
# Send gRPC request to a TPU worker to collect a trace of your model. A
# profiler service has been started in the TPU worker at port 8466.
```python
# E.g. your TPU IP address is 10.0.0.2 and you want to profile for 2 seconds.
tf.profiler.experimental.client.trace('grpc://10.0.0.2:8466',
'gs://your_tb_dir', 2000)
Example usage (Multiple TPUs):
# Send gRPC request to a TPU pod to collect a trace of your model on multiple
# TPUs. A profiler service has been started in all the TPU workers at the
# port 8466.
```python
# E.g. your TPU IP addresses are 10.0.0.2, 10.0.0.3, 10.0.0.4, and you want to
# profile for 2 seconds.
tf.profiler.experimental.client.trace('grpc://10.0.0.2:8466',
'gs://your_tb_dir',
2000, '10.0.0.3,10.0.0.4')
Launch TensorBoard and point it to the same logdir you provided to this API.
$ tensorboard --logdir=/tmp/tb_log (or gs://your_tb_dir in the above examples)
Open your browser and go to localhost:6006/#profile to view profiling results."
10596,monitor,tensorflow/tensorflow/python/profiler/profiler_client.py,99,function,"Sends grpc requests to profiler server to perform on-demand monitoring.
The monitoring result is a light weight performance summary of your model
execution. This method will block the caller thread until it receives the
monitoring result. This method currently supports Cloud TPU only.
Args:
service_addr: gRPC address of profiler service e.g. grpc://10.0.0.2:8466.
duration_ms: Duration of monitoring in ms.
level: Choose a monitoring level between 1 and 2 to monitor your job. Level
2 is more verbose than level 1 and shows more metrics.
Returns:
A string of monitoring output.
Example usage:
# Continuously send gRPC requests to the Cloud TPU to monitor the model
# execution.
```python
for query in range(0, 100):
print(tf.profiler.experimental.client.monitor('grpc://10.0.0.2:8466', 1000))"
10597,ProfilerOptions,tensorflow/tensorflow/python/profiler/profiler_v2.py,50,class,"Options for finer control over the profiler.
Use `tf.profiler.ProfilerOptions` to control `tf.profiler`
behavior.
Fields:
host_tracer_level: Adjust CPU tracing level. Values are: 1 - critical info
only, 2 - info, 3 - verbose. [default value is 2]
python_tracer_level: Toggle tracing of Python function calls. Values are: 1
- enabled, 0 - disabled [default value is 0]
device_tracer_level: Adjust device (TPU/GPU) tracing level. Values are: 1 -
enabled, 0 - disabled [default value is 1]"
10598,start,tensorflow/tensorflow/python/profiler/profiler_v2.py,78,function,"Start profiling TensorFlow performance.
Args:
logdir: Profiling results log directory.
options: `ProfilerOptions` namedtuple to specify miscellaneous profiler
options. See example usage below.
Raises:
AlreadyExistsError: If a profiling session is already running.
Example usage:
```python
options = tf.profiler.experimental.ProfilerOptions(host_tracer_level = 3,
python_tracer_level = 1,
device_tracer_level = 1)
tf.profiler.experimental.start('logdir_path', options = options)
# Training code here
tf.profiler.experimental.stop()
```
To view the profiling results, launch TensorBoard and point it to `logdir`.
Open your browser and go to `localhost:6006/#profile` to view profiling
results."
10599,stop,tensorflow/tensorflow/python/profiler/profiler_v2.py,127,function,"Stops the current profiling session.
The profiler session will be stopped and profile results can be saved.
Args:
save: An optional variable to save the results to TensorBoard. Default True.
Raises:
UnavailableError: If there is no active profiling session."
10600,warmup,tensorflow/tensorflow/python/profiler/profiler_v2.py,153,function,"Warm-up the profiler session.
The profiler session will set up profiling context, including loading CUPTI
library for GPU profiling. This is used for improving the accuracy of
the profiling results."
10601,start_server,tensorflow/tensorflow/python/profiler/profiler_v2.py,166,function,"Start a profiler grpc server that listens to given port.
The profiler server will exit when the process finishes. The service is
defined in tensorflow/core/profiler/profiler_service.proto.
Args:
port: port profiler server listens to.
Example usage: ```python tf.profiler.experimental.server.start('6009') # do
your training here."
10602,Profile,tensorflow/tensorflow/python/profiler/profiler_v2.py,181,class,"Context-manager profile API.
Profiling will start when entering the scope, and stop and save the results to
the logdir when exits the scope. Open TensorBoard profile tab to view results.
Example usage:
```python
with tf.profiler.experimental.Profile(""/path/to/logdir""):
# do some work
```"
10603,merge_default_with_oplog,tensorflow/tensorflow/python/profiler/tfprof_logger.py,145,function,"Merge the tfprof default extra info with caller's op_log.
Args:
graph: tf.Graph. If None and eager execution is not enabled, use
default graph.
op_log: OpLogProto proto.
run_meta: RunMetadata proto used to complete shape information.
add_trace: Whether to add op trace information.
add_trainable_var: Whether to assign tf.compat.v1.trainable_variables() op
type '_trainable_variables'.
Returns:
tmp_op_log: Merged OpLogProto proto."
10604,write_op_log,tensorflow/tensorflow/python/profiler/tfprof_logger.py,193,function,"Log provided 'op_log', and add additional model information below.
The API also assigns ops in tf.compat.v1.trainable_variables() an op type
called '_trainable_variables'.
The API also logs 'flops' statistics for ops with op.RegisterStatistics()
defined. flops calculation depends on Tensor shapes defined in 'graph',
which might not be complete. 'run_meta', if provided, completes the shape
information with best effort.
Args:
graph: tf.Graph. If None and eager execution is not enabled, use
default graph.
log_dir: directory to write the log file.
op_log: (Optional) OpLogProto proto to be written. If not provided, an new
one is created.
run_meta: (Optional) RunMetadata proto that helps flops computation using
run time shape information.
add_trace: Whether to add python code trace information.
Used to support ""code"" view."
10605,Trace,tensorflow/tensorflow/python/profiler/trace.py,30,class,"Context manager that generates a trace event in the profiler.
A trace event will start when entering the context, and stop and save the
result to the profiler when exiting the context. Open TensorBoard Profile tab
and choose trace viewer to view the trace event in the timeline.
Trace events are created only when the profiler is enabled. More information
on how to use the profiler can be found at
https://tensorflow.org/guide/profiler
Example usage:
```python
tf.profiler.experimental.start('logdir')
for step in range(num_steps):
# Creates a trace event for each training step with the step number.
with tf.profiler.experimental.Trace(""Train"", step_num=step):
train_fn()
tf.profiler.experimental.stop()
```"
10606,set_metadata,tensorflow/tensorflow/python/profiler/trace.py,89,method,"Sets metadata in this trace event.
Args:
**kwargs: metadata in key-value pairs.
This method enables setting metadata in a trace event after it is
created.
Example usage:
```python
def call(function):
with tf.profiler.experimental.Trace(""call"",
function_name=function.name) as tm:
binary, in_cache = jit_compile(function)
tm.set_metadata(in_cache=in_cache)
execute(binary)
```
In this example, we want to trace how much time spent on
calling a function, which includes compilation and execution.
The compilation can be either getting a cached copy of the
binary or actually generating the binary, which is indicated
by the boolean ""in_cache"" returned by jit_compile(). We need
to use set_metadata() to pass in_cache because we did not know
the in_cache value when the trace was created (and we cannot
create the trace after jit_compile(), because we want
to measure the entire duration of call())."
10607,traceme_wrapper,tensorflow/tensorflow/python/profiler/traceme.py,24,function,
10608,BuildSmallModel,tensorflow/tensorflow/python/profiler/internal/model_analyzer_testlib.py,38,function,Build a small forward conv model.
10609,BuildFullModel,tensorflow/tensorflow/python/profiler/internal/model_analyzer_testlib.py,58,function,"Build the full model with conv,rnn,opt."
10610,BuildSplittableModel,tensorflow/tensorflow/python/profiler/internal/model_analyzer_testlib.py,75,function,Build a small model that can be run partially in each step.
10611,SearchTFProfNode,tensorflow/tensorflow/python/profiler/internal/model_analyzer_testlib.py,95,function,Search a node in the tree.
10612,ProfilerFromFile,tensorflow/tensorflow/python/profiler/internal/model_analyzer_testlib.py,106,function,Initialize a profiler from profile file.
10613,CheckAndRemoveDoc,tensorflow/tensorflow/python/profiler/internal/model_analyzer_testlib.py,114,function,
10614,SavedModelBuilder,tensorflow/tensorflow/python/saved_model/builder_impl.py,432,class,
10615,add_meta_graph,tensorflow/tensorflow/python/saved_model/builder_impl.py,516,method,
10616,add_meta_graph_and_variables,tensorflow/tensorflow/python/saved_model/builder_impl.py,560,method,
10617,get_asset_filename_to_add,tensorflow/tensorflow/python/saved_model/builder_impl.py,670,function,"Get a unique basename to add to the SavedModel if this file is unseen.
Assets come from users as full paths, and we save them out to the
SavedModel as basenames. In some cases, the basenames collide. Here,
we dedupe asset basenames by first checking if the file is the same,
and, if different, generate and return an index-suffixed basename
that can be used to add the asset to the SavedModel.
Args:
asset_filepath: the full path to the asset that is being saved
asset_filename_map: a dict of filenames used for saving the asset in
the SavedModel to full paths from which the filenames were derived.
Returns:
Uniquified filename string if the file is not a duplicate, or the original
filename if the file has already been seen and saved."
10618,copy_assets_to_destination_dir,tensorflow/tensorflow/python/saved_model/builder_impl.py,757,function,Copy all assets from source path to destination path.
10619,setup_bare_concrete_function,tensorflow/tensorflow/python/saved_model/function_deserialization.py,153,function,Makes a restored bare concrete function callable.
10620,RestoredFunction,tensorflow/tensorflow/python/saved_model/function_deserialization.py,170,class,"Wrapper class for a function that has been restored from saved state.
See `def_function.Function`."
10621,recreate_function,tensorflow/tensorflow/python/saved_model/function_deserialization.py,192,function,"Creates a `Function` from a `SavedFunction`.
Args:
saved_function: `SavedFunction` proto.
concrete_functions: map from function name to `ConcreteFunction`.
As a side effect of this function, the `FunctionSpec` from
`saved_function` is added to each `ConcreteFunction` in this map.
Returns:
A `Function`."
10622,load_function_def_library,tensorflow/tensorflow/python/saved_model/function_deserialization.py,278,function,"Load a set of functions as concrete functions without captured inputs.
Functions names are manipulated during load such that they do not overlap
with previously created ones.
Args:
library: FunctionDefLibrary proto message.
load_shared_name_suffix: If specified, used to uniquify shared
names. Otherwise, a unique name is generated.
Returns:
Map of original function names in the library to instances of
`ConcreteFunction` without captured inputs.
Raises:
ValueError: if functions dependencies have a cycle."
10623,fix_node_def,tensorflow/tensorflow/python/saved_model/function_deserialization.py,393,function,Replace functions calls and shared names in `node_def`.
10624,serialize_concrete_function,tensorflow/tensorflow/python/saved_model/function_serialization.py,52,function,Build a SavedConcreteFunction.
10625,serialize_bare_concrete_function,tensorflow/tensorflow/python/saved_model/function_serialization.py,78,function,Build a SavedBareConcreteFunction.
10626,serialize_function,tensorflow/tensorflow/python/saved_model/function_serialization.py,94,function,Build a SavedFunction proto.
10627,wrap_cached_variables,tensorflow/tensorflow/python/saved_model/function_serialization.py,110,function,"Wraps the concrete function if it uses cached read tensors.
This function creates a new concrete function that captures variables
instead of the cached read tensors.
Args:
concrete_function: A Concrete function that maybe captures cached read
tensors.
Returns:
A concrete function that wraps the original concrete function, which
captures variables instead. If the original function did not capture any
cached values, then the function is not wrapped and the original object is
returned."
10628,Loader,tensorflow/tensorflow/python/saved_model/load.py,109,class,Helper class to load an object-based SavedModel.
10629,adjust_debug_info_func_names,tensorflow/tensorflow/python/saved_model/load.py,360,method,Rewrite func names in the debug info by using the concrete func names.
10630,get,tensorflow/tensorflow/python/saved_model/load.py,373,method,
10631,uninitialized_variable_creator,tensorflow/tensorflow/python/saved_model/load.py,438,method,A variable creator that creates uninitialized variables.
10632,load,tensorflow/tensorflow/python/saved_model/load.py,513,function,"Load a SavedModel from `export_dir`.
Signatures associated with the SavedModel are available as functions:
```python
imported = tf.saved_model.load(path)
f = imported.signatures[""serving_default""]
print(f(x=tf.constant([[1.]])))
```
Objects exported with `tf.saved_model.save` additionally have trackable
objects and functions assigned to attributes:
```python
exported = tf.train.Checkpoint(v=tf.Variable(3.))
exported.f = tf.function(
lambda x: exported.v * x,
input_signature=[tf.TensorSpec(shape=None, dtype=tf.float32)])
tf.saved_model.save(exported, path)
imported = tf.saved_model.load(path)
assert 3. == imported.v.numpy()
assert 6. == imported.f(x=tf.constant(2.)).numpy()
```
_Loading Keras models_
Keras models are trackable, so they can be saved to SavedModel. The object
returned by `tf.saved_model.load` is not a Keras object (i.e. doesn't have
`.fit`, `.predict`, etc. methods). A few attributes and functions are still
available: `.variables`, `.trainable_variables` and `.__call__`.
```python
model = tf.keras.Model(...)
tf.saved_model.save(model, path)
imported = tf.saved_model.load(path)
outputs = imported(inputs)
```
Use `tf.keras.models.load_model` to restore the Keras model.
_Importing SavedModels from TensorFlow 1.x_
SavedModels from `tf.estimator.Estimator` or 1.x SavedModel APIs have a flat
graph instead of `tf.function` objects. These SavedModels will be loaded with
the following attributes:
* `.signatures`: A dictionary mapping signature names to functions.
* `.prune(feeds, fetches) `: A method which allows you to extract
functions for new subgraphs. This is equivalent to importing the SavedModel
and naming feeds and fetches in a Session from TensorFlow 1.x.
```python
imported = tf.saved_model.load(path_to_v1_saved_model)
pruned = imported.prune(""x:0"", ""out:0"")
pruned(tf.ones([]))
```
See `tf.compat.v1.wrap_function` for details.
* `.variables`: A list of imported variables.
* `.graph`: The whole imported graph.
* `.restore(save_path)`: A function that restores variables from a checkpoint
saved from `tf.compat.v1.Saver`.
_Consuming SavedModels asynchronously_
When consuming SavedModels asynchronously (the producer is a separate
process), the SavedModel directory will appear before all files have been
written, and `tf.saved_model.load` will fail if pointed at an incomplete
SavedModel. Rather than checking for the directory, check for
""saved_model_dir/saved_model.pb"". This file is written atomically as the last
`tf.saved_model.save` file operation.
Args:
export_dir: The SavedModel directory to load from.
tags: A tag or sequence of tags identifying the MetaGraph to load. Optional
if the SavedModel contains a single MetaGraph, as for those exported from
`tf.saved_model.save`.
options: Optional, `tf.saved_model.LoadOptions` object that specifies
options for loading.
Returns:
A trackable object with a `signatures` attribute mapping from signature
keys to functions. If the SavedModel was exported by `tf.saved_model.load`,
it also points to trackable objects, functions, debug info which it has been
saved.
Raises:
ValueError: If `tags` don't match a MetaGraph in the SavedModel."
10633,load_internal,tensorflow/tensorflow/python/saved_model/load.py,606,function,Loader implementation.
10634,LoadContext,tensorflow/tensorflow/python/saved_model/load_context.py,25,class,A context for loading a model.
10635,set_load_options,tensorflow/tensorflow/python/saved_model/load_context.py,32,method,
10636,clear_load_options,tensorflow/tensorflow/python/saved_model/load_context.py,35,method,
10637,load_options,tensorflow/tensorflow/python/saved_model/load_context.py,38,method,
10638,load_context,tensorflow/tensorflow/python/saved_model/load_context.py,46,function,
10639,get_load_options,tensorflow/tensorflow/python/saved_model/load_context.py,54,function,Returns whether under a load context.
10640,LoadOptions,tensorflow/tensorflow/python/saved_model/load_options.py,25,class,"Options for loading a SavedModel.
This function may be used in the `options` argument in functions that
load a SavedModel (`tf.saved_model.load`, `tf.keras.models.load_model`)."
10641,cycle,tensorflow/tensorflow/python/saved_model/load_test.py,68,function,
10642,load,tensorflow/tensorflow/python/saved_model/load_v1_in_v2.py,260,function,Load a v1-style SavedModel as an object.
10643,parse_saved_model_with_debug_info,tensorflow/tensorflow/python/saved_model/loader_impl.py,43,function,"Reads the savedmodel as well as the graph debug info.
Args:
export_dir: Directory containing the SavedModel and GraphDebugInfo files.
Returns:
`SavedModel` and `GraphDebugInfo` protocol buffers.
Raises:
IOError: If the saved model file does not exist, or cannot be successfully
parsed. Missing graph debug info file is fine."
10644,parse_saved_model,tensorflow/tensorflow/python/saved_model/loader_impl.py,72,function,"Reads the savedmodel.pb or savedmodel.pbtxt file containing `SavedModel`.
Args:
export_dir: String or Pathlike, path to the directory containing the
SavedModel file.
Returns:
A `SavedModel` protocol buffer.
Raises:
IOError: If the file does not exist, or cannot be successfully parsed."
10645,get_asset_tensors,tensorflow/tensorflow/python/saved_model/loader_impl.py,122,function,"Gets the asset tensors, if defined in the meta graph def to load.
Args:
export_dir: Directory where the SavedModel is located.
meta_graph_def_to_load: The meta graph def from the SavedModel to be loaded.
import_scope: Optional `string` -- if specified, prepend this followed by
'/' to all returned asset tensor names.
Returns:
A dictionary of asset tensors, keyed by the name of the asset tensor. The
value in the map corresponds to the absolute path of the asset file."
10646,get_init_op,tensorflow/tensorflow/python/saved_model/loader_impl.py,208,function,
10647,get_train_op,tensorflow/tensorflow/python/saved_model/loader_impl.py,215,function,
10648,maybe_saved_model_directory,tensorflow/tensorflow/python/saved_model/loader_impl.py,230,function,"Checks whether the provided export directory could contain a SavedModel.
Note that the method does not load any data by itself. If the method returns
`false`, the export directory definitely does not contain a SavedModel. If the
method returns `true`, the export directory may contain a SavedModel but
provides no guarantee that it can be loaded.
Args:
export_dir: Absolute string path to possible export location. For example,
'/my/foo/model'.
Returns:
True if the export directory contains SavedModel files, False otherwise."
10649,contains_saved_model,tensorflow/tensorflow/python/saved_model/loader_impl.py,251,function,"Checks whether the provided export directory could contain a SavedModel.
Note that the method does not load any data by itself. If the method returns
`false`, the export directory definitely does not contain a SavedModel. If the
method returns `true`, the export directory may contain a SavedModel but
provides no guarantee that it can be loaded.
Args:
export_dir: Absolute string path to possible export location. For example,
'/my/foo/model'.
Returns:
True if the export directory contains SavedModel files, False otherwise."
10650,load,tensorflow/tensorflow/python/saved_model/loader_impl.py,276,function,"Loads the model from a SavedModel as specified by tags.
Args:
sess: The TensorFlow session to restore the variables.
tags: Set of string tags to identify the required MetaGraphDef. These should
correspond to the tags used when saving the variables using the
SavedModel `save()` API.
export_dir: Directory in which the SavedModel protocol buffer and variables
to be loaded are located.
import_scope: Optional `string` -- if specified, prepend this string
followed by '/' to all loaded tensor names. This scope is applied to
tensor instances loaded into the passed session, but it is *not* written
through to the static `MetaGraphDef` protocol buffer that is returned.
**saver_kwargs: Optional keyword arguments passed through to Saver.
Returns:
The `MetaGraphDef` protocol buffer loaded in the provided session. This
can be used to further extract signature-defs, collection-defs, etc.
Raises:
RuntimeError: MetaGraphDef associated with the tags cannot be found."
10651,SavedModelLoader,tensorflow/tensorflow/python/saved_model/loader_impl.py,303,class,Load graphs and restore variable values from a `SavedModel`.
10652,export_dir,tensorflow/tensorflow/python/saved_model/loader_impl.py,318,method,Directory containing the SavedModel.
10653,variables_path,tensorflow/tensorflow/python/saved_model/loader_impl.py,323,method,Path to variable checkpoint files.
10654,saved_model,tensorflow/tensorflow/python/saved_model/loader_impl.py,328,method,SavedModel object parsed from the export directory.
10655,get_meta_graph_def_from_tags,tensorflow/tensorflow/python/saved_model/loader_impl.py,332,method,"Return MetaGraphDef with the exact specified tags.
Args:
tags: A list or set of string tags that identify the MetaGraphDef.
Returns:
MetaGraphDef with the same tags.
Raises:
RuntimeError: if no metagraphs were found with the associated tags."
10656,load_graph,tensorflow/tensorflow/python/saved_model/loader_impl.py,361,method,"Load ops and nodes from SavedModel MetaGraph into graph.
Args:
graph: tf.Graph object.
tags: a set of string tags identifying a MetaGraphDef.
import_scope: Optional `string` -- if specified, prepend this string
followed by '/' to all loaded tensor names. This scope is applied to
tensor instances loaded into the passed session, but it is *not* written
through to the static `MetaGraphDef` protocol buffer that is returned.
**saver_kwargs: keyword arguments to pass to tf.train.import_meta_graph.
Returns:
A tuple of
* Saver defined by the MetaGraph, which can be used to restore the
variable values.
* List of `Operation`/`Tensor` objects returned from
`tf.import_graph_def` (may be `None`)."
10657,restore_variables,tensorflow/tensorflow/python/saved_model/loader_impl.py,385,method,"Restore SavedModel variable values into the session.
Args:
sess: tf.compat.v1.Session to restore variable values.
saver: a tf.compat.v1.train.Saver object. Can be None if there are no
variables in graph. This may be the saver returned by the load_graph()
function, or a default `tf.compat.v1.train.Saver()`.
import_scope: Optional `string` -- if specified, prepend this string
followed by '/' to all loaded tensor names. This scope is applied to
tensor instances loaded into the passed session, but it is *not* written
through to the static `MetaGraphDef` protocol buffer that is returned.
Raises:
ValueError: if no saver was passed to the saver argument, and there are
variables in the graph."
10658,run_init_ops,tensorflow/tensorflow/python/saved_model/loader_impl.py,415,method,"Run initialization ops defined in the `MetaGraphDef`.
Args:
sess: tf.compat.v1.Session to restore variable values.
tags: a set of string tags identifying a MetaGraphDef.
import_scope: Optional `string` -- if specified, prepend this string
followed by '/' to all loaded tensor names. This scope is applied to
tensor instances loaded into the passed session, but it is *not* written
through to the static `MetaGraphDef` protocol buffer that is returned."
10659,load,tensorflow/tensorflow/python/saved_model/loader_impl.py,436,method,"Load the MetaGraphDef graph and restore variable values into the session.
Args:
sess: tf.compat.v1.Session to restore variable values.
tags: a set of string tags identifying a MetaGraphDef.
import_scope: Optional `string` -- if specified, prepend this string
followed by '/' to all loaded tensor names. This scope is applied to
tensor instances loaded into the passed session, but it is *not* written
through to the static `MetaGraphDef` protocol buffer that is returned.
**saver_kwargs: keyword arguments to pass to tf.train.import_meta_graph.
Returns:
`MetagraphDef` proto of the graph that was loaded."
10660,build_graph_helper,tensorflow/tensorflow/python/saved_model/loader_test.py,48,function,
10661,MethodNameUpdater,tensorflow/tensorflow/python/saved_model/method_name_updater.py,37,class,"Updates the method name(s) of the SavedModel stored in the given path.
The `MethodNameUpdater` class provides the functionality to update the method
name field in the signature_defs of the given SavedModel. For example, it
can be used to replace the `predict` `method_name` to `regress`.
Typical usages of the `MethodNameUpdater`
```python
...
updater = tf.compat.v1.saved_model.MethodNameUpdater(export_dir)
# Update all signature_defs with key ""foo"" in all meta graph defs.
updater.replace_method_name(signature_key=""foo"", method_name=""regress"")
# Update a single signature_def with key ""bar"" in the meta graph def with
# tags [""serve""]
updater.replace_method_name(signature_key=""bar"", method_name=""classify"",
tags=""serve"")
updater.save(new_export_dir)
```
Note: This function will only be available through the v1 compatibility
library as tf.compat.v1.saved_model.builder.MethodNameUpdater."
10662,replace_method_name,tensorflow/tensorflow/python/saved_model/method_name_updater.py,74,method,"Replaces the method_name in the specified signature_def.
This will match and replace multiple sig defs iff tags is None (i.e when
multiple `MetaGraph`s have a signature_def with the same key).
If tags is not None, this will only replace a single signature_def in the
`MetaGraph` with matching tags.
Args:
signature_key: Key of the signature_def to be updated.
method_name: new method_name to replace the existing one.
tags: A tag or sequence of tags identifying the `MetaGraph` to update. If
None, all meta graphs will be updated.
Raises:
ValueError: if signature_key or method_name are not defined or
if no metagraphs were found with the associated tags or
if no meta graph has a signature_def that matches signature_key."
10663,save,tensorflow/tensorflow/python/saved_model/method_name_updater.py,119,method,"Saves the updated `SavedModel`.
Args:
new_export_dir: Path where the updated `SavedModel` will be saved. If
None, the input `SavedModel` will be overriden with the updates.
Raises:
errors.OpError: If there are errors during the file save operation."
10664,NotEncodableError,tensorflow/tensorflow/python/saved_model/nested_structure_coder.py,57,class,Error raised when a coder cannot encode an object.
10665,StructureCoder,tensorflow/tensorflow/python/saved_model/nested_structure_coder.py,61,class,Encoder and decoder for nested structures into protos.
10666,register_codec,tensorflow/tensorflow/python/saved_model/nested_structure_coder.py,67,method,
10667,encode_structure,tensorflow/tensorflow/python/saved_model/nested_structure_coder.py,86,method,"Encodes nested structures composed of encodable types into a proto.
Args:
nested_structure: Structure to encode.
Returns:
Encoded proto.
Raises:
NotEncodableError: For values for which there are no encoders."
10668,can_encode,tensorflow/tensorflow/python/saved_model/nested_structure_coder.py,100,method,"Determines whether a nested structure can be encoded into a proto.
Args:
nested_structure: Structure to encode.
Returns:
True if the nested structured can be encoded."
10669,decode_proto,tensorflow/tensorflow/python/saved_model/nested_structure_coder.py,115,method,"Decodes proto representing a nested structure.
Args:
proto: Proto to decode.
Returns:
Decoded structure.
Raises:
NotEncodableError: For values for which there are no encoders."
10670,VersionedTypeRegistration,tensorflow/tensorflow/python/saved_model/revived_types.py,25,class,Holds information about one version of a revived type.
10671,to_proto,tensorflow/tensorflow/python/saved_model/revived_types.py,73,method,Create a SavedUserObject proto.
10672,from_proto,tensorflow/tensorflow/python/saved_model/revived_types.py,85,method,Recreate a trackable object from a SavedUserObject proto.
10673,should_load,tensorflow/tensorflow/python/saved_model/revived_types.py,89,method,Checks if this object should load the SavedUserObject `proto`.
10674,register_revived_type,tensorflow/tensorflow/python/saved_model/revived_types.py,108,function,"Register a type for revived objects.
Args:
identifier: A unique string identifying this class of objects.
predicate: A Boolean predicate for this registration. Takes a
trackable object as an argument. If True, `type_registration` may be
used to save and restore the object.
versions: A list of `VersionedTypeRegistration` objects."
10675,serialize,tensorflow/tensorflow/python/saved_model/revived_types.py,140,function,Create a SavedUserObject from a trackable object.
10676,deserialize,tensorflow/tensorflow/python/saved_model/revived_types.py,150,function,"Create a trackable object from a SavedUserObject proto.
Args:
proto: A SavedUserObject to deserialize.
Returns:
A tuple of (trackable, assignment_fn) where assignment_fn has the same
signature as setattr and should be used to add dependencies to
`trackable` when they are available."
10677,registered_identifiers,tensorflow/tensorflow/python/saved_model/revived_types.py,170,function,
10678,get_setter,tensorflow/tensorflow/python/saved_model/revived_types.py,174,function,
10679,save,tensorflow/tensorflow/python/saved_model/save.py,810,function,"Exports the Trackable object `obj` to [SavedModel format](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/saved_model/README.md).
Example usage:
```python
class Adder(tf.Module):
@tf.function(input_signature=[tf.TensorSpec(shape=None, dtype=tf.float32)])
def add(self, x):
return x + x + 1.
to_export = Adder()
tf.saved_model.save(to_export, '/tmp/adder')
```
The resulting SavedModel is then servable with an input named ""x"", its value
having any shape and dtype float32.
The optional `signatures` argument controls which methods in `obj` will be
available to programs which consume `SavedModel`s, for example, serving
APIs. Python functions may be decorated with
`@tf.function(input_signature=...)` and passed as signatures directly, or
lazily with a call to `get_concrete_function` on the method decorated with
`@tf.function`.
If the `signatures` argument is omitted, `obj` will be searched for
`@tf.function`-decorated methods. If exactly one `@tf.function` is found, that
method will be used as the default signature for the SavedModel. This behavior
is expected to change in the future, when a corresponding
`tf.saved_model.load` symbol is added. At that point signatures will be
completely optional, and any `@tf.function` attached to `obj` or its
dependencies will be exported for use with `load`.
When invoking a signature in an exported SavedModel, `Tensor` arguments are
identified by name. These names will come from the Python function's argument
names by default. They may be overridden by specifying a `name=...` argument
in the corresponding `tf.TensorSpec` object. Explicit naming is required if
multiple `Tensor`s are passed through a single argument to the Python
function.
The outputs of functions used as `signatures` must either be flat lists, in
which case outputs will be numbered, or a dictionary mapping string keys to
`Tensor`, in which case the keys will be used to name outputs.
Signatures are available in objects returned by `tf.saved_model.load` as a
`.signatures` attribute. This is a reserved attribute: `tf.saved_model.save`
on an object with a custom `.signatures` attribute will raise an exception.
Since `tf.keras.Model` objects are also Trackable, this function can be
used to export Keras models. For example, exporting with a signature
specified:
```python
class Model(tf.keras.Model):
@tf.function(input_signature=[tf.TensorSpec(shape=[None], dtype=tf.string)])
def serve(self, serialized):
...
m = Model()
tf.saved_model.save(m, '/tmp/saved_model/')
```
Exporting from a function without a fixed signature:
```python
class Model(tf.keras.Model):
@tf.function
def call(self, x):
...
m = Model()
tf.saved_model.save(
m, '/tmp/saved_model/',
signatures=m.call.get_concrete_function(
tf.TensorSpec(shape=[None, 3], dtype=tf.float32, name=""inp"")))
```
`tf.keras.Model` instances constructed from inputs and outputs already have a
signature and so do not require a `@tf.function` decorator or a `signatures`
argument. If neither are specified, the model's forward pass is exported.
```python
x = input_layer.Input((4,), name=""x"")
y = core.Dense(5, name=""out"")(x)
model = training.Model(x, y)
tf.saved_model.save(model, '/tmp/saved_model/')
# The exported SavedModel takes ""x"" with shape [None, 4] and returns ""out""
# with shape [None, 5]
```
Variables must be tracked by assigning them to an attribute of a tracked
object or to an attribute of `obj` directly. TensorFlow objects (e.g. layers
from `tf.keras.layers`, optimizers from `tf.train`) track their variables
automatically. This is the same tracking scheme that `tf.train.Checkpoint`
uses, and an exported `Checkpoint` object may be restored as a training
checkpoint by pointing `tf.train.Checkpoint.restore` to the SavedModel's
""variables/"" subdirectory. Currently, variables are the only stateful objects
supported by `tf.saved_model.save`, but others (e.g. tables) will be supported
in the future.
`tf.function` does not hard-code device annotations from outside the function
body, instead of using the calling context's device. This means for example
that exporting a model that runs on a GPU and serving it on a CPU will
generally work, with some exceptions. `tf.device` annotations inside the body
of the function will be hard-coded in the exported model; this type of
annotation is discouraged. Device-specific operations, e.g. with ""cuDNN"" in
the name or with device-specific layouts, may cause issues. Currently a
`DistributionStrategy` is another exception: active distribution strategies
will cause device placements to be hard-coded in a function. Exporting a
single-device computation and importing under a `DistributionStrategy` is
not currently supported, but may be in the future.
SavedModels exported with `tf.saved_model.save` [strip default-valued
attributes](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/saved_model/README.md#stripping-default-valued-attributes)
automatically, which removes one source of incompatibilities when the consumer
of a SavedModel is running an older TensorFlow version than the
producer. There are however other sources of incompatibilities which are not
handled automatically, such as when the exported model contains operations
which the consumer does not have definitions for.
A single tf.function can generate many ConcreteFunctions. If a downstream tool
wants to refer to all concrete functions generated by a single tf.function you
can use the `function_aliases` argument to store a map from the alias name to
all concrete function names.
E.g.
```python
class MyModel:
@tf.function
def func():
...
@tf.function
def serve():
...
func()
model = MyModel()
signatures = {
'serving_default': model.serve.get_concrete_function(),
}
options = tf.saved_model.SaveOptions(function_aliases={
'my_func': func,
})
tf.saved_model.save(model, export_dir, signatures, options)
```
Args:
obj: A trackable object to export.
export_dir: A directory in which to write the SavedModel.
signatures: Optional, one of three types:
* a `tf.function` with an input signature specified, which will use the
default serving signature key,
* the result of `f.get_concrete_function` on a `@tf.function`-decorated
function `f`, in which case `f` will be used to generate a signature for
the SavedModel under the default serving signature key,
* a dictionary, which maps signature keys to either `tf.function`
instances with input signatures or concrete functions. Keys of such a
dictionary may be arbitrary strings, but will typically be from the
`tf.saved_model.signature_constants` module.
options: Optional, `tf.saved_model.SaveOptions` object that specifies
options for saving.
Raises:
ValueError: If `obj` is not trackable.
@compatibility(eager)
Not well supported when graph building. From TensorFlow 1.x,
`tf.compat.v1.enable_eager_execution()` should run first. Calling
tf.saved_model.save in a loop when graph building from TensorFlow 1.x will
add new save operations to the default graph each iteration.
May not be called from within a function body.
@end_compatibility"
10680,export_meta_graph,tensorflow/tensorflow/python/saved_model/save.py,1038,function,"Exports the MetaGraph proto to a file.
This function goes through the same procedures saved_model.save goes to
produce the given object's MetaGraph, then saves it to the given file. It
skips saving checkpoint information, and is useful when all one wants is the
graph defining the model.
Args:
obj: A trackable object to build the MetaGraph from.
filename: The file into which to write the MetaGraph.
signatures: Optional, either a `tf.function` with an input signature
specified or the result of `f.get_concrete_function` on a
`@tf.function`-decorated function `f`, in which case `f` will be used to
generate a signature for the SavedModel under the default serving
signature key. `signatures` may also be a dictionary, in which case it
maps from signature keys to either `tf.function` instances with input
signatures or concrete functions. The keys of such a dictionary may be
arbitrary strings, but will typically be from the
`tf.saved_model.signature_constants` module.
options: Optional, `tf.saved_model.SaveOptions` object that specifies
options for saving."
10681,SaveContext,tensorflow/tensorflow/python/saved_model/save_context.py,25,class,A context for building a graph of SavedModel.
10682,options,tensorflow/tensorflow/python/saved_model/save_context.py,33,method,
10683,enter_save_context,tensorflow/tensorflow/python/saved_model/save_context.py,38,method,
10684,exit_save_context,tensorflow/tensorflow/python/saved_model/save_context.py,42,method,
10685,in_save_context,tensorflow/tensorflow/python/saved_model/save_context.py,46,method,
10686,save_context,tensorflow/tensorflow/python/saved_model/save_context.py,53,function,
10687,in_save_context,tensorflow/tensorflow/python/saved_model/save_context.py,63,function,Returns whether under a save context.
10688,get_save_options,tensorflow/tensorflow/python/saved_model/save_context.py,68,function,Returns the save options if under a save context.
10689,VariablePolicy,tensorflow/tensorflow/python/saved_model/save_options.py,29,class,"Enum defining options for variable handling when saving.
NONE
No policy applied: Distributed variables are saved as one variable, with no
device attached.
SAVE_VARIABLE_DEVICES
When saving variables, also save their device assignment.
This is useful if one wants to hardcode devices in saved models, but it also
makes them non-portable if soft device placement is disabled (more details
in `tf.config.set_soft_device_placement`). This is currently not
fully supported by `saved_model.load`, and is mainly intended to be used
when one will be reading the saved model at a lower API level. In the
example below, the graph saved by the call to `saved_model.save` will have
the variable devices correctly specified:
```python
exported = tf.train.Checkpoint()
with tf.device('/GPU:0'):
exported.x_gpu = tf.Variable(1.0)
with tf.device('/CPU:0'):
exported.x_cpu = tf.Variable(1.0)
tf.saved_model.save(exported, export_dir,
options = tf.saved_model.SaveOptions(
experimental_variable_policy=
tf.saved_model.experimental.VariablePolicy.SAVE_VARIABLE_DEVICES))
```
Distributed variables are still saved as one variable under this policy.
EXPAND_DISTRIBUTED_VARIABLES
Distributed variables will be explicitly expanded into their respective
distributed replicas, and their assigned devices will be saved. This is
useful when one wants to use the model for training in environments where
the original distribution strategy is not available. Checkpoints are
currently incompatible with this option, so it is not implemented in
`saved_model.save` (only the internal `saved_model.export_meta_graph` API
supports it for now)."
10690,from_obj,tensorflow/tensorflow/python/saved_model/save_options.py,83,method,Tries to convert `obj` to a VariablePolicy instance.
10691,SaveOptions,tensorflow/tensorflow/python/saved_model/save_options.py,97,class,"Options for saving to SavedModel.
This function may be used in the `options` argument in functions that
save a SavedModel (`tf.saved_model.save`, `tf.keras.models.save_model`)."
10692,tearDownModule,tensorflow/tensorflow/python/saved_model/saved_model_test.py,55,function,
10693,build_signature_def,tensorflow/tensorflow/python/saved_model/signature_def_utils_impl.py,39,function,"Utility function to build a SignatureDef protocol buffer.
Args:
inputs: Inputs of the SignatureDef defined as a proto map of string to
tensor info.
outputs: Outputs of the SignatureDef defined as a proto map of string to
tensor info.
method_name: Method name of the SignatureDef as a string.
Returns:
A SignatureDef protocol buffer constructed based on the supplied arguments."
10694,regression_signature_def,tensorflow/tensorflow/python/saved_model/signature_def_utils_impl.py,71,function,"Creates regression signature from given examples and predictions.
This function produces signatures intended for use with the TensorFlow Serving
Regress API (tensorflow_serving/apis/prediction_service.proto), and so
constrains the input and output types to those allowed by TensorFlow Serving.
Args:
examples: A string `Tensor`, expected to accept serialized tf.Examples.
predictions: A float `Tensor`.
Returns:
A regression-flavored signature_def.
Raises:
ValueError: If examples is `None`."
10695,classification_signature_def,tensorflow/tensorflow/python/saved_model/signature_def_utils_impl.py,119,function,"Creates classification signature from given examples and predictions.
This function produces signatures intended for use with the TensorFlow Serving
Classify API (tensorflow_serving/apis/prediction_service.proto), and so
constrains the input and output types to those allowed by TensorFlow Serving.
Args:
examples: A string `Tensor`, expected to accept serialized tf.Examples.
classes: A string `Tensor`. Note that the ClassificationResponse message
requires that class labels are strings, not integers or anything else.
scores: a float `Tensor`.
Returns:
A classification-flavored signature_def.
Raises:
ValueError: If examples is `None`."
10696,predict_signature_def,tensorflow/tensorflow/python/saved_model/signature_def_utils_impl.py,178,function,"Creates prediction signature from given inputs and outputs.
This function produces signatures intended for use with the TensorFlow Serving
Predict API (tensorflow_serving/apis/prediction_service.proto). This API
imposes no constraints on the input and output types.
Args:
inputs: dict of string to `Tensor`.
outputs: dict of string to `Tensor`.
Returns:
A prediction-flavored signature_def.
Raises:
ValueError: If inputs or outputs is `None`."
10697,supervised_train_signature_def,tensorflow/tensorflow/python/saved_model/signature_def_utils_impl.py,212,function,
10698,supervised_eval_signature_def,tensorflow/tensorflow/python/saved_model/signature_def_utils_impl.py,219,function,
10699,is_valid_signature,tensorflow/tensorflow/python/saved_model/signature_def_utils_impl.py,275,function,Determine whether a SignatureDef can be served by TensorFlow Serving.
10700,op_signature_def,tensorflow/tensorflow/python/saved_model/signature_def_utils_impl.py,350,function,"Creates a signature def with the output pointing to an op.
Note that op isn't strictly enforced to be an Op object, and may be a Tensor.
It is recommended to use the build_signature_def() function for Tensors.
Args:
op: An Op (or possibly Tensor).
key: Key to graph element in the SignatureDef outputs.
Returns:
A SignatureDef with a single output pointing to the op."
10701,load_op_from_signature_def,tensorflow/tensorflow/python/saved_model/signature_def_utils_impl.py,368,function,"Load an Op from a SignatureDef created by op_signature_def().
Args:
signature_def: a SignatureDef proto
key: string key to op in the SignatureDef outputs.
import_scope: Scope used to import the op
Returns:
Op (or possibly Tensor) in the graph with the same name as saved in the
SignatureDef.
Raises:
NotFoundError: If the op could not be found in the graph."
10702,find_function_to_export,tensorflow/tensorflow/python/saved_model/signature_serialization.py,71,function,"Function to export, None if no suitable function was found."
10703,canonicalize_signatures,tensorflow/tensorflow/python/saved_model/signature_serialization.py,96,function,Converts `signatures` into a dictionary of concrete functions.
10704,create_signature_map,tensorflow/tensorflow/python/saved_model/signature_serialization.py,245,function,Creates an object containing `signatures`.
10705,validate_saveable_view,tensorflow/tensorflow/python/saved_model/signature_serialization.py,265,function,Performs signature-related sanity checks on `saveable_view`.
10706,simple_save,tensorflow/tensorflow/python/saved_model/simple_save.py,35,function,"Convenience function to build a SavedModel suitable for serving.
In many common cases, saving models for serving will be as simple as:
simple_save(session,
export_dir,
inputs={""x"": x, ""y"": y},
outputs={""z"": z})
Although in many cases it's not necessary to understand all of the many ways
to configure a SavedModel, this method has a few practical implications:
- It will be treated as a graph for inference / serving (i.e. uses the tag
`saved_model.SERVING`)
- The SavedModel will load in TensorFlow Serving and supports the
[Predict
API](https://github.com/tensorflow/serving/blob/master/tensorflow_serving/apis/predict.proto).
To use the Classify, Regress, or MultiInference APIs, please
use either
[tf.Estimator](https://www.tensorflow.org/api_docs/python/tf/estimator/Estimator)
or the lower level
[SavedModel
APIs](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/saved_model/README.md).
- Some TensorFlow ops depend on information on disk or other information
called ""assets"". These are generally handled automatically by adding the
assets to the `GraphKeys.ASSET_FILEPATHS` collection. Only assets in that
collection are exported; if you need more custom behavior, you'll need to
use the
[SavedModelBuilder](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/saved_model/builder.py).
More information about SavedModel and signatures can be found here:
https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/saved_model/README.md.
Args:
session: The TensorFlow session from which to save the meta graph and
variables.
export_dir: The path to which the SavedModel will be stored.
inputs: dict mapping string input names to tensors. These are added
to the SignatureDef as the inputs.
outputs: dict mapping string output names to tensors. These are added
to the SignatureDef as the outputs.
legacy_init_op: Legacy support for op or group of ops to execute after the
restore op upon a load."
10707,build_tensor_info,tensorflow/tensorflow/python/saved_model/utils_impl.py,51,function,"Utility function to build TensorInfo proto from a Tensor.
Args:
tensor: Tensor or SparseTensor whose name, dtype and shape are used to
build the TensorInfo. For SparseTensors, the names of the three
constituent Tensors are used.
Returns:
A TensorInfo protocol buffer constructed based on the supplied argument.
Raises:
RuntimeError: If eager execution is enabled."
10708,build_tensor_info_internal,tensorflow/tensorflow/python/saved_model/utils_impl.py,70,function,Utility function to build TensorInfo proto from a Tensor.
10709,build_tensor_info_from_op,tensorflow/tensorflow/python/saved_model/utils_impl.py,101,function,"Utility function to build TensorInfo proto from an Op.
Note that this function should be used with caution. It is strictly restricted
to TensorFlow internal use-cases only. Please make sure you do need it before
using it.
This utility function overloads the TensorInfo proto by setting the name to
the Op's name, dtype to DT_INVALID and tensor_shape as None. One typical usage
is for the Op of the call site for the defunned function:
```python
@function.defun
def some_variable_initialization_fn(value_a, value_b):
a = value_a
b = value_b
value_a = constant_op.constant(1, name=""a"")
value_b = constant_op.constant(2, name=""b"")
op_info = utils.build_op_info(
some_variable_initialization_fn(value_a, value_b))
```
Args:
op: An Op whose name is used to build the TensorInfo. The name that points
to the Op could be fetched at run time in the Loader session.
Returns:
A TensorInfo protocol buffer constructed based on the supplied argument."
10710,get_tensor_from_tensor_info,tensorflow/tensorflow/python/saved_model/utils_impl.py,143,function,"Returns the Tensor or CompositeTensor described by a TensorInfo proto.
Args:
tensor_info: A TensorInfo proto describing a Tensor or SparseTensor or
CompositeTensor.
graph: The tf.Graph in which tensors are looked up. If None, the
current default graph is used.
import_scope: If not None, names in `tensor_info` are prefixed with this
string before lookup.
Returns:
The Tensor or SparseTensor or CompositeTensor in `graph` described by
`tensor_info`.
Raises:
KeyError: If `tensor_info` does not correspond to a tensor in `graph`.
ValueError: If `tensor_info` is malformed."
10711,get_element_from_tensor_info,tensorflow/tensorflow/python/saved_model/utils_impl.py,186,function,"Returns the element in the graph described by a TensorInfo proto.
Args:
tensor_info: A TensorInfo proto describing an Op or Tensor by name.
graph: The tf.Graph in which tensors are looked up. If None, the current
default graph is used.
import_scope: If not None, names in `tensor_info` are prefixed with this
string before lookup.
Returns:
Op or tensor in `graph` described by `tensor_info`.
Raises:
KeyError: If `tensor_info` does not correspond to an op or tensor in `graph`"
10712,get_or_create_variables_dir,tensorflow/tensorflow/python/saved_model/utils_impl.py,210,function,"Return variables sub-directory, or create one if it doesn't exist."
10713,get_variables_dir,tensorflow/tensorflow/python/saved_model/utils_impl.py,218,function,Return variables sub-directory in the SavedModel.
10714,get_variables_path,tensorflow/tensorflow/python/saved_model/utils_impl.py,225,function,"Return the variables path, used as the prefix for checkpoint files."
10715,get_or_create_assets_dir,tensorflow/tensorflow/python/saved_model/utils_impl.py,232,function,"Return assets sub-directory, or create one if it doesn't exist."
10716,get_assets_dir,tensorflow/tensorflow/python/saved_model/utils_impl.py,242,function,Return path to asset directory in the SavedModel.
10717,get_or_create_debug_dir,tensorflow/tensorflow/python/saved_model/utils_impl.py,249,function,"Returns path to the debug sub-directory, creating if it does not exist."
10718,get_debug_dir,tensorflow/tensorflow/python/saved_model/utils_impl.py,259,function,Returns path to the debug sub-directory in the SavedModel.
10719,ExportOutput,tensorflow/tensorflow/python/saved_model/model_utils/export_output.py,32,class,"Represents an output of a model that can be served.
These typically correspond to model heads."
10720,as_signature_def,tensorflow/tensorflow/python/saved_model/model_utils/export_output.py,43,method,"Generate a SignatureDef proto for inclusion in a MetaGraphDef.
The SignatureDef will specify outputs as described in this ExportOutput,
and will use the provided receiver_tensors as inputs.
Args:
receiver_tensors: a `Tensor`, or a dict of string to `Tensor`, specifying
input nodes that will be fed."
10721,ClassificationOutput,tensorflow/tensorflow/python/saved_model/model_utils/export_output.py,102,class,"Represents the output of a classification head.
Either classes or scores or both must be set.
The classes `Tensor` must provide string labels, not integer class IDs.
If only classes is set, it is interpreted as providing top-k results in
descending order.
If only scores is set, it is interpreted as providing a score for every class
in order of class ID.
If both classes and scores are set, they are interpreted as zipped, so each
score corresponds to the class at the same index. Clients should not depend
on the order of the entries."
10722,scores,tensorflow/tensorflow/python/saved_model/model_utils/export_output.py,151,method,
10723,classes,tensorflow/tensorflow/python/saved_model/model_utils/export_output.py,155,method,
10724,as_signature_def,tensorflow/tensorflow/python/saved_model/model_utils/export_output.py,158,method,
10725,RegressionOutput,tensorflow/tensorflow/python/saved_model/model_utils/export_output.py,170,class,Represents the output of a regression head.
10726,value,tensorflow/tensorflow/python/saved_model/model_utils/export_output.py,188,method,
10727,as_signature_def,tensorflow/tensorflow/python/saved_model/model_utils/export_output.py,191,method,
10728,PredictOutput,tensorflow/tensorflow/python/saved_model/model_utils/export_output.py,202,class,"Represents the output of a generic prediction head.
A generic prediction need not be either a classification or a regression.
Named outputs must be provided as a dict from string to `Tensor`,"
10729,outputs,tensorflow/tensorflow/python/saved_model/model_utils/export_output.py,227,method,
10730,as_signature_def,tensorflow/tensorflow/python/saved_model/model_utils/export_output.py,230,method,
10731,TrainOutput,tensorflow/tensorflow/python/saved_model/model_utils/export_output.py,386,class,"Represents the output of a supervised training process.
This class generates the appropriate signature def for exporting
training output by type-checking and wrapping loss, predictions, and metrics
values."
10732,EvalOutput,tensorflow/tensorflow/python/saved_model/model_utils/export_output.py,398,class,"Represents the output of a supervised eval process.
This class generates the appropriate signature def for exporting
eval output by type-checking and wrapping loss, predictions, and metrics
values."
10733,MockSupervisedOutput,tensorflow/tensorflow/python/saved_model/model_utils/export_output_test.py,231,class,So that we can test the abstract class methods directly.
10734,build_all_signature_defs,tensorflow/tensorflow/python/saved_model/model_utils/export_utils.py,63,function,"Build `SignatureDef`s for all export outputs.
Args:
receiver_tensors: a `Tensor`, or a dict of string to `Tensor`, specifying
input nodes where this receiver expects to be fed by default. Typically,
this is a single placeholder expecting serialized `tf.Example` protos.
export_outputs: a dict of ExportOutput instances, each of which has
an as_signature_def instance method that will be called to retrieve
the signature_def for all export output tensors.
receiver_tensors_alternatives: a dict of string to additional
groups of receiver tensors, each of which may be a `Tensor` or a dict of
string to `Tensor`. These named receiver tensor alternatives generate
additional serving signatures, which may be used to feed inputs at
different points within the input receiver subgraph. A typical usage is
to allow feeding raw feature `Tensor`s *downstream* of the
tf.io.parse_example() op. Defaults to None.
serving_only: boolean; if true, resulting signature defs will only include
valid serving signatures. If false, all requested signatures will be
returned.
Returns:
signature_def representing all passed args.
Raises:
ValueError: if export_outputs is not a dict"
10735,get_timestamped_export_dir,tensorflow/tensorflow/python/saved_model/model_utils/export_utils.py,193,function,"Builds a path to a new subdirectory within the base directory.
Each export is written into a new subdirectory named using the
current time. This guarantees monotonically increasing version
numbers even across multiple runs of the pipeline.
The timestamp used is the number of seconds since epoch UTC.
Args:
export_dir_base: A string containing a directory to write the exported
graph and checkpoints.
Returns:
The full path of the new subdirectory (which is not actually created yet).
Raises:
RuntimeError: if repeated attempts fail to obtain a unique timestamped
directory name."
10736,get_temp_export_dir,tensorflow/tensorflow/python/saved_model/model_utils/export_utils.py,230,function,"Builds a directory name based on the argument but starting with 'temp-'.
This relies on the fact that TensorFlow Serving ignores subdirectories of
the base directory that can't be parsed as integers.
Args:
timestamped_export_dir: the name of the eventual export directory, e.g.
/foo/bar/<timestamp>
Returns:
A sister directory prefixed with 'temp-', e.g. /foo/bar/temp-<timestamp>."
10737,export_outputs_for_mode,tensorflow/tensorflow/python/saved_model/model_utils/export_utils.py,250,function,"Util function for constructing a `ExportOutput` dict given a mode.
The returned dict can be directly passed to `build_all_signature_defs` helper
function as the `export_outputs` argument, used for generating a SignatureDef
map.
Args:
mode: A `ModeKeys` specifying the mode.
serving_export_outputs: Describes the output signatures to be exported to
`SavedModel` and used during serving. Should be a dict or None.
predictions: A dict of Tensors or single Tensor representing model
predictions. This argument is only used if serving_export_outputs is not
set.
loss: A dict of Tensors or single Tensor representing calculated loss.
metrics: A dict of (metric_value, update_op) tuples, or a single tuple.
metric_value must be a Tensor, and update_op must be a Tensor or Op
Returns:
Dictionary mapping the a key to an `tf.estimator.export.ExportOutput` object
The key is the expected SignatureDef key for the mode.
Raises:
ValueError: if an appropriate ExportOutput cannot be found for the mode."
10738,get_export_outputs,tensorflow/tensorflow/python/saved_model/model_utils/export_utils.py,294,function,"Validate export_outputs or create default export_outputs.
Args:
export_outputs: Describes the output signatures to be exported to
`SavedModel` and used during serving. Should be a dict or None.
predictions: Predictions `Tensor` or dict of `Tensor`.
Returns:
Valid export_outputs dict
Raises:
TypeError: if export_outputs is not a dict or its values are not
ExportOutput instances."
10739,KerasModeKeys,tensorflow/tensorflow/python/saved_model/model_utils/mode_keys.py,25,class,"Standard names for model modes.
The following standard keys are defined:
* `TRAIN`: training/fitting mode.
* `TEST`: testing/evaluation mode.
* `PREDICT`: prediction/inference mode."
10740,EstimatorModeKeys,tensorflow/tensorflow/python/saved_model/model_utils/mode_keys.py,41,class,"Standard names for Estimator model modes.
The following standard keys are defined:
* `TRAIN`: training/fitting mode.
* `EVAL`: testing/evaluation mode.
* `PREDICT`: predication/inference mode."
10741,is_predict,tensorflow/tensorflow/python/saved_model/model_utils/mode_keys.py,56,function,
10742,is_eval,tensorflow/tensorflow/python/saved_model/model_utils/mode_keys.py,60,function,
10743,is_train,tensorflow/tensorflow/python/saved_model/model_utils/mode_keys.py,64,function,
10744,ModeKeyMap,tensorflow/tensorflow/python/saved_model/model_utils/mode_keys.py,68,class,"Map using ModeKeys as keys.
This class creates an immutable mapping from modes to values. For example,
SavedModel export of Keras and Estimator models use this to map modes to their
corresponding MetaGraph tags/SignatureDef keys.
Since this class uses modes, rather than strings, as keys, both ""predict""
(Keras's PREDICT ModeKey) and ""infer"" (Estimator's PREDICT ModeKey) map to the
same value."
10745,get_plugin_asset,tensorflow/tensorflow/python/summary/plugin_asset.py,42,function,"Acquire singleton PluginAsset instance from a graph.
PluginAssets are always singletons, and are stored in tf Graph collections.
This way, they can be defined anywhere the graph is being constructed, and
if the same plugin is configured at many different points, the user can always
modify the same instance.
Args:
plugin_asset_cls: The PluginAsset class
graph: (optional) The graph to retrieve the instance from. If not specified,
the default graph is used.
Returns:
An instance of the plugin_asset_class
Raises:
ValueError: If we have a plugin name collision, or if we unexpectedly find
the wrong number of items in a collection."
10746,get_all_plugin_assets,tensorflow/tensorflow/python/summary/plugin_asset.py,84,function,"Retrieve all PluginAssets stored in the graph collection.
Args:
graph: Optionally, the graph to get assets from. If unspecified, the default
graph is used.
Returns:
A list with all PluginAsset instances in the graph.
Raises:
ValueError: if we unexpectedly find a collection with the wrong number of
PluginAssets."
10747,PluginAsset,tensorflow/tensorflow/python/summary/plugin_asset.py,113,class,"This abstract base class allows TensorBoard to serialize assets to disk.
Plugin authors are expected to extend the PluginAsset class, so that it:
- has a unique plugin_name
- provides an assets method that returns an {asset_name: asset_contents}
dictionary. For now, asset_contents are strings, although we may add
StringIO support later.
LifeCycle of a PluginAsset instance:
- It is constructed when get_plugin_asset is called on the class for
the first time.
- It is configured by code that follows the calls to get_plugin_asset
- When the containing graph is serialized by the
tf.compat.v1.summary.FileWriter, the writer calls assets and the
PluginAsset instance provides its contents to be written to disk."
10748,assets,tensorflow/tensorflow/python/summary/plugin_asset.py,134,method,"Provide all of the assets contained by the PluginAsset instance.
The assets method should return a dictionary structured as
{asset_name: asset_contents}. asset_contents is a string.
This method will be called by the tf.compat.v1.summary.FileWriter when it
is time to write the assets out to disk."
10749,scalar,tensorflow/tensorflow/python/summary/summary.py,58,function,"Outputs a `Summary` protocol buffer containing a single scalar value.
The generated Summary has a Tensor.proto containing the input Tensor.
Args:
name: A name for the generated node. Will also serve as the series name in
TensorBoard.
tensor: A real numeric Tensor containing a single value.
collections: Optional list of graph collections keys. The new summary op is
added to these collections. Defaults to `[GraphKeys.SUMMARIES]`.
family: Optional; if provided, used as the prefix of the summary tag name,
which controls the tab name used for display on Tensorboard.
Returns:
A scalar `Tensor` of type `string`. Which contains a `Summary` protobuf.
Raises:
ValueError: If tensor has the wrong shape or type."
10750,image,tensorflow/tensorflow/python/summary/summary.py,88,function,"Outputs a `Summary` protocol buffer with images.
The summary has up to `max_outputs` summary values containing images. The
images are built from `tensor` which must be 4-D with shape `[batch_size,
height, width, channels]` and where `channels` can be:
* 1: `tensor` is interpreted as Grayscale.
* 3: `tensor` is interpreted as RGB.
* 4: `tensor` is interpreted as RGBA.
The images have the same number of channels as the input tensor. For float
input, the values are normalized one image at a time to fit in the range
`[0, 255]`. `uint8` values are unchanged. The op uses two different
normalization algorithms:
* If the input values are all positive, they are rescaled so the largest one
is 255.
* If any input value is negative, the values are shifted so input value 0.0
is at 127. They are then rescaled so that either the smallest value is 0,
or the largest one is 255.
The `tag` in the outputted Summary.Value protobufs is generated based on the
name, with a suffix depending on the max_outputs setting:
* If `max_outputs` is 1, the summary value tag is '*name*/image'.
* If `max_outputs` is greater than 1, the summary value tags are
generated sequentially as '*name*/image/0', '*name*/image/1', etc.
Args:
name: A name for the generated node. Will also serve as a series name in
TensorBoard.
tensor: A 4-D `uint8` or `float32` `Tensor` of shape `[batch_size, height,
width, channels]` where `channels` is 1, 3, or 4.
max_outputs: Max number of batch elements to generate images for.
collections: Optional list of ops.GraphKeys. The collections to add the
summary to. Defaults to [_ops.GraphKeys.SUMMARIES]
family: Optional; if provided, used as the prefix of the summary tag name,
which controls the tab name used for display on Tensorboard.
Returns:
A scalar `Tensor` of type `string`. The serialized `Summary` protocol
buffer."
10751,histogram,tensorflow/tensorflow/python/summary/summary.py,144,function,"Outputs a `Summary` protocol buffer with a histogram.
Adding a histogram summary makes it possible to visualize your data's
distribution in TensorBoard. You can see a detailed explanation of the
TensorBoard histogram dashboard
[here](https://www.tensorflow.org/get_started/tensorboard_histograms).
The generated
[`Summary`](https://www.tensorflow.org/code/tensorflow/core/framework/summary.proto)
has one summary value containing a histogram for `values`.
This op reports an `InvalidArgument` error if any value is not finite.
Args:
name: A name for the generated node. Will also serve as a series name in
TensorBoard.
values: A real numeric `Tensor`. Any shape. Values to use to
build the histogram.
collections: Optional list of graph collections keys. The new summary op is
added to these collections. Defaults to `[GraphKeys.SUMMARIES]`.
family: Optional; if provided, used as the prefix of the summary tag name,
which controls the tab name used for display on Tensorboard.
Returns:
A scalar `Tensor` of type `string`. The serialized `Summary` protocol
buffer."
10752,audio,tensorflow/tensorflow/python/summary/summary.py,185,function,"Outputs a `Summary` protocol buffer with audio.
The summary has up to `max_outputs` summary values containing audio. The
audio is built from `tensor` which must be 3-D with shape `[batch_size,
frames, channels]` or 2-D with shape `[batch_size, frames]`. The values are
assumed to be in the range of `[-1.0, 1.0]` with a sample rate of
`sample_rate`.
The `tag` in the outputted Summary.Value protobufs is generated based on the
name, with a suffix depending on the max_outputs setting:
* If `max_outputs` is 1, the summary value tag is '*name*/audio'.
* If `max_outputs` is greater than 1, the summary value tags are
generated sequentially as '*name*/audio/0', '*name*/audio/1', etc
Args:
name: A name for the generated node. Will also serve as a series name in
TensorBoard.
tensor: A 3-D `float32` `Tensor` of shape `[batch_size, frames, channels]`
or a 2-D `float32` `Tensor` of shape `[batch_size, frames]`.
sample_rate: A Scalar `float32` `Tensor` indicating the sample rate of the
signal in hertz.
max_outputs: Max number of batch elements to generate audio for.
collections: Optional list of ops.GraphKeys. The collections to add the
summary to. Defaults to [_ops.GraphKeys.SUMMARIES]
family: Optional; if provided, used as the prefix of the summary tag name,
which controls the tab name used for display on Tensorboard.
Returns:
A scalar `Tensor` of type `string`. The serialized `Summary` protocol
buffer."
10753,text,tensorflow/tensorflow/python/summary/summary.py,234,function,"Summarizes textual data.
Text data summarized via this plugin will be visible in the Text Dashboard
in TensorBoard. The standard TensorBoard Text Dashboard will render markdown
in the strings, and will automatically organize 1d and 2d tensors into tables.
If a tensor with more than 2 dimensions is provided, a 2d subarray will be
displayed along with a warning message. (Note that this behavior is not
intrinsic to the text summary api, but rather to the default TensorBoard text
plugin.)
Args:
name: A name for the generated node. Will also serve as a series name in
TensorBoard.
tensor: a string-type Tensor to summarize.
collections: Optional list of ops.GraphKeys. The collections to add the
summary to. Defaults to [_ops.GraphKeys.SUMMARIES]
Returns:
A TensorSummary op that is configured so that TensorBoard will recognize
that it contains textual data. The TensorSummary is a scalar `Tensor` of
type `string` which contains `Summary` protobufs.
Raises:
ValueError: If tensor has the wrong type."
10754,tensor_summary,tensorflow/tensorflow/python/summary/summary.py,275,function,"Outputs a `Summary` protocol buffer with a serialized tensor.proto.
Args:
name: A name for the generated node. If display_name is not set, it will
also serve as the tag name in TensorBoard. (In that case, the tag
name will inherit tf name scopes.)
tensor: A tensor of any type and shape to serialize.
summary_description: A long description of the summary sequence. Markdown
is supported.
collections: Optional list of graph collections keys. The new summary op is
added to these collections. Defaults to `[GraphKeys.SUMMARIES]`.
summary_metadata: Optional SummaryMetadata proto (which describes which
plugins may use the summary value).
family: Optional; if provided, used as the prefix of the summary tag,
which controls the name used for display on TensorBoard when
display_name is not set.
display_name: A string used to name this data in TensorBoard. If this is
not set, then the node name will be used instead.
Returns:
A scalar `Tensor` of type `string`. The serialized `Summary` protocol
buffer."
10755,merge,tensorflow/tensorflow/python/summary/summary.py,331,function,"Merges summaries.
This op creates a
[`Summary`](https://www.tensorflow.org/code/tensorflow/core/framework/summary.proto)
protocol buffer that contains the union of all the values in the input
summaries.
When the Op is run, it reports an `InvalidArgument` error if multiple values
in the summaries to merge use the same tag.
Args:
inputs: A list of `string` `Tensor` objects containing serialized `Summary`
protocol buffers.
collections: Optional list of graph collections keys. The new summary op is
added to these collections. Defaults to `[]`.
name: A name for the operation (optional).
Returns:
A scalar `Tensor` of type `string`. The serialized `Summary` protocol
buffer resulting from the merging.
Raises:
RuntimeError: If called with eager mode enabled.
@compatibility(eager)
Not compatible with eager execution. To write TensorBoard
summaries under eager execution, use `tf.contrib.summary` instead.
@end_compatibility"
10756,merge_all,tensorflow/tensorflow/python/summary/summary.py,377,function,"Merges all summaries collected in the default graph.
Args:
key: `GraphKey` used to collect the summaries. Defaults to
`GraphKeys.SUMMARIES`.
scope: Optional scope used to filter the summary ops, using `re.match`
Returns:
If no summaries were collected, returns None. Otherwise returns a scalar
`Tensor` of type `string` containing the serialized `Summary` protocol
buffer resulting from the merging.
Raises:
RuntimeError: If called with eager execution enabled.
@compatibility(eager)
Not compatible with eager execution. To write TensorBoard
summaries under eager execution, use `tf.contrib.summary` instead.
@end_compatibility"
10757,get_summary_description,tensorflow/tensorflow/python/summary/summary.py,410,function,"Given a TensorSummary node_def, retrieve its SummaryDescription.
When a Summary op is instantiated, a SummaryDescription of associated
metadata is stored in its NodeDef. This method retrieves the description.
Args:
node_def: the node_def_pb2.NodeDef of a TensorSummary op
Returns:
a summary_pb2.SummaryDescription
Raises:
ValueError: if the node is not a summary op.
@compatibility(eager)
Not compatible with eager execution. To write TensorBoard
summaries under eager execution, use `tf.contrib.summary` instead.
@end_compatibility"
10758,summary_iterator,tensorflow/tensorflow/python/summary/summary_iterator.py,44,function,"Returns a iterator for reading `Event` protocol buffers from an event file.
You can use this function to read events written to an event file. It returns
a Python iterator that yields `Event` protocol buffers.
Example: Print the contents of an events file.
```python
for e in tf.compat.v1.train.summary_iterator(path to events file):
print(e)
```
Example: Print selected summary values.
```python
# This example supposes that the events file contains summaries with a
# summary value tag 'loss'. These could have been added by calling
# `add_summary()`, passing the output of a scalar summary op created with
# with: `tf.compat.v1.summary.scalar('loss', loss_tensor)`.
for e in tf.compat.v1.train.summary_iterator(path to events file):
for v in e.summary.value:
if v.tag == 'loss':
print(v.simple_value)
```
Example: Continuously check for new summary values.
```python
summaries = tf.compat.v1.train.summary_iterator(path to events file)
while True:
for e in summaries:
for v in e.summary.value:
if v.tag == 'loss':
print(v.simple_value)
# Wait for a bit before checking the file for any new events
time.sleep(wait time)
```
See the protocol buffer definitions of
[Event](https://www.tensorflow.org/code/tensorflow/core/util/event.proto)
and
[Summary](https://www.tensorflow.org/code/tensorflow/core/framework/summary.proto)
for more information about their attributes.
Args:
path: The path to an event file created by a `SummaryWriter`.
Returns:
A iterator that yields `Event` protocol buffers"
10759,EventFileWriter,tensorflow/tensorflow/python/summary/writer/event_file_writer.py,35,class,"Writes `Event` protocol buffers to an event file.
The `EventFileWriter` class creates an event file in the specified directory,
and asynchronously writes Event protocol buffers to the file. The Event file
is encoded using the tfrecord format, which is similar to RecordIO.
This class is not thread-safe."
10760,get_logdir,tensorflow/tensorflow/python/summary/writer/event_file_writer.py,98,method,Returns the directory where event file will be written.
10761,reopen,tensorflow/tensorflow/python/summary/writer/event_file_writer.py,102,method,"Reopens the EventFileWriter.
Can be called after `close()` to add more events in the same directory.
The events will go into a new events file.
Does nothing if the EventFileWriter was not closed."
10762,add_event,tensorflow/tensorflow/python/summary/writer/event_file_writer.py,114,method,"Adds an event to the event file.
Args:
event: An `Event` protocol buffer."
10763,flush,tensorflow/tensorflow/python/summary/writer/event_file_writer.py,139,method,"Flushes the event file to disk.
Call this method to make sure that all pending events have been written to
disk."
10764,close,tensorflow/tensorflow/python/summary/writer/event_file_writer.py,155,method,"Flushes the event file to disk and close the file.
Call this method when you do not need the summary writer anymore."
10765,CloseableQueue,tensorflow/tensorflow/python/summary/writer/event_file_writer.py,235,class,Stripped-down fork of the standard library Queue that is closeable.
10766,get,tensorflow/tensorflow/python/summary/writer/event_file_writer.py,256,method,"Remove and return an item from the queue.
If the queue is empty, blocks until an item is available.
Returns:
an item from the queue"
10767,put,tensorflow/tensorflow/python/summary/writer/event_file_writer.py,271,method,"Put an item into the queue.
If the queue is closed, fails immediately.
If the queue is full, blocks until space is available or until the queue
is closed by a call to close(), at which point this call fails.
Args:
item: an item to add to the queue
Raises:
QueueClosedError: if insertion failed because the queue is closed"
10768,close,tensorflow/tensorflow/python/summary/writer/event_file_writer.py,296,method,"Closes the queue, causing any pending or future `put()` calls to fail."
10769,QueueClosedError,tensorflow/tensorflow/python/summary/writer/event_file_writer.py,303,class,Raised when CloseableQueue.put() fails because the queue is closed.
10770,EventFileWriterV2,tensorflow/tensorflow/python/summary/writer/event_file_writer_v2.py,29,class,"Writes `Event` protocol buffers to an event file via the graph.
The `EventFileWriterV2` class is backed by the summary file writer in the v2
summary API (currently in tf.contrib.summary), so it uses a shared summary
writer resource and graph ops to write events.
As with the original EventFileWriter, this class will asynchronously write
Event protocol buffers to the backing file. The Event file is encoded using
the tfrecord format, which is similar to RecordIO."
10771,get_logdir,tensorflow/tensorflow/python/summary/writer/event_file_writer_v2.py,98,method,Returns the directory where event file will be written.
10772,reopen,tensorflow/tensorflow/python/summary/writer/event_file_writer_v2.py,102,method,"Reopens the EventFileWriter.
Can be called after `close()` to add more events in the same directory.
The events will go into a new events file.
Does nothing if the EventFileWriter was not closed."
10773,add_event,tensorflow/tensorflow/python/summary/writer/event_file_writer_v2.py,114,method,"Adds an event to the event file.
Args:
event: An `Event` protocol buffer."
10774,flush,tensorflow/tensorflow/python/summary/writer/event_file_writer_v2.py,125,method,"Flushes the event file to disk.
Call this method to make sure that all pending events have been written to
disk."
10775,close,tensorflow/tensorflow/python/summary/writer/event_file_writer_v2.py,133,method,"Flushes the event file to disk and close the file.
Call this method when you do not need the summary writer anymore."
10776,FakeSummaryWriter,tensorflow/tensorflow/python/summary/writer/fake_summary_writer.py,27,class,Fake summary writer.
10777,install,tensorflow/tensorflow/python/summary/writer/fake_summary_writer.py,33,method,
10778,uninstall,tensorflow/tensorflow/python/summary/writer/fake_summary_writer.py,41,method,
10779,summaries,tensorflow/tensorflow/python/summary/writer/fake_summary_writer.py,58,method,
10780,assert_summaries,tensorflow/tensorflow/python/summary/writer/fake_summary_writer.py,61,method,Assert expected items have been added to summary writer.
10781,add_summary,tensorflow/tensorflow/python/summary/writer/fake_summary_writer.py,98,method,Add summary.
10782,add_graph,tensorflow/tensorflow/python/summary/writer/fake_summary_writer.py,112,method,Add graph.
10783,add_meta_graph,tensorflow/tensorflow/python/summary/writer/fake_summary_writer.py,120,method,Add metagraph.
10784,add_session_log,tensorflow/tensorflow/python/summary/writer/fake_summary_writer.py,127,method,
10785,add_run_metadata,tensorflow/tensorflow/python/summary/writer/fake_summary_writer.py,131,method,
10786,flush,tensorflow/tensorflow/python/summary/writer/fake_summary_writer.py,136,method,
10787,reopen,tensorflow/tensorflow/python/summary/writer/fake_summary_writer.py,139,method,
10788,close,tensorflow/tensorflow/python/summary/writer/fake_summary_writer.py,142,method,
10789,SummaryToEventTransformer,tensorflow/tensorflow/python/summary/writer/writer.py,42,class,"Abstractly implements the SummaryWriter API.
This API basically implements a number of endpoints (add_summary,
add_session_log, etc). The endpoints all generate an event protobuf, which is
passed to the contained event_writer."
10790,add_summary,tensorflow/tensorflow/python/summary/writer/writer.py,101,method,"Adds a `Summary` protocol buffer to the event file.
This method wraps the provided summary in an `Event` protocol buffer
and adds it to the event file.
You can pass the result of evaluating any summary op, using
`tf.Session.run` or
`tf.Tensor.eval`, to this
function. Alternatively, you can pass a `tf.compat.v1.Summary` protocol
buffer that you populate with your own data. The latter is
commonly done to report evaluation results in event files.
Args:
summary: A `Summary` protocol buffer, optionally serialized as a string.
global_step: Number. Optional global step value to record with the
summary."
10791,add_session_log,tensorflow/tensorflow/python/summary/writer/writer.py,144,method,"Adds a `SessionLog` protocol buffer to the event file.
This method wraps the provided session in an `Event` protocol buffer
and adds it to the event file.
Args:
session_log: A `SessionLog` protocol buffer.
global_step: Number. Optional global step value to record with the
summary."
10792,add_graph,tensorflow/tensorflow/python/summary/writer/writer.py,163,method,"Adds a `Graph` to the event file.
The graph described by the protocol buffer will be displayed by
TensorBoard. Most users pass a graph in the constructor instead.
Args:
graph: A `Graph` object, such as `sess.graph`.
global_step: Number. Optional global step counter to record with the
graph.
graph_def: DEPRECATED. Use the `graph` parameter instead.
Raises:
ValueError: If both graph and graph_def are passed to the method."
10793,add_meta_graph,tensorflow/tensorflow/python/summary/writer/writer.py,229,method,"Adds a `MetaGraphDef` to the event file.
The `MetaGraphDef` allows running the given graph via
`saver.import_meta_graph()`.
Args:
meta_graph_def: A `MetaGraphDef` object, often as returned by
`saver.export_meta_graph()`.
global_step: Number. Optional global step counter to record with the
graph.
Raises:
TypeError: If both `meta_graph_def` is not an instance of `MetaGraphDef`."
10794,add_run_metadata,tensorflow/tensorflow/python/summary/writer/writer.py,251,method,"Adds a metadata information for a single session.run() call.
Args:
run_metadata: A `RunMetadata` protobuf object.
tag: The tag name for this metadata.
global_step: Number. Optional global step counter to record with the
StepStats.
Raises:
ValueError: If the provided tag was already used for this type of event."
10795,FileWriter,tensorflow/tensorflow/python/summary/writer/writer.py,283,class,"Writes `Summary` protocol buffers to event files.
The `FileWriter` class provides a mechanism to create an event file in a
given directory and add summaries and events to it. The class updates the
file contents asynchronously. This allows a training program to call methods
to add data to the file directly from the training loop, without slowing down
training.
When constructed with a `tf.compat.v1.Session` parameter, a `FileWriter`
instead forms a compatibility layer over new graph-based summaries
to facilitate the use of new summary writing with
pre-existing code that expects a `FileWriter` instance.
This class is not thread-safe."
10796,get_logdir,tensorflow/tensorflow/python/summary/writer/writer.py,381,method,Returns the directory where event file will be written.
10797,add_event,tensorflow/tensorflow/python/summary/writer/writer.py,395,method,"Adds an event to the event file.
Args:
event: An `Event` protocol buffer."
10798,flush,tensorflow/tensorflow/python/summary/writer/writer.py,404,method,"Flushes the event file to disk.
Call this method to make sure that all pending events have been written to
disk."
10799,close,tensorflow/tensorflow/python/summary/writer/writer.py,415,method,"Flushes the event file to disk and close the file.
Call this method when you do not need the summary writer anymore."
10800,reopen,tensorflow/tensorflow/python/summary/writer/writer.py,423,method,"Reopens the EventFileWriter.
Can be called after `close()` to add more events in the same directory.
The events will go into a new events file.
Does nothing if the EventFileWriter was not closed."
10801,FileWriterCache,tensorflow/tensorflow/python/summary/writer/writer_cache.py,29,class,"Cache for file writers.
This class caches file writers, one per directory."
10802,clear,tensorflow/tensorflow/python/summary/writer/writer_cache.py,41,method,Clear cached summary writers. Currently only used for unit tests.
10803,get,tensorflow/tensorflow/python/summary/writer/writer_cache.py,51,method,"Returns the FileWriter for the specified directory.
Args:
logdir: str, name of the directory.
Returns:
A `FileWriter`."
10804,FakeWriteError,tensorflow/tensorflow/python/summary/writer/writer_test.py,463,class,
10805,SymbolTable,tensorflow/tensorflow/python/tf_program/mlir_gen.py,41,class,Symbol Table for python code.
10806,enter_scope,tensorflow/tensorflow/python/tf_program/mlir_gen.py,48,method,Enter a new scope - at function level.
10807,insert_symbol,tensorflow/tensorflow/python/tf_program/mlir_gen.py,53,method,
10808,insert_type,tensorflow/tensorflow/python/tf_program/mlir_gen.py,58,method,
10809,exit_scope,tensorflow/tensorflow/python/tf_program/mlir_gen.py,61,method,
10810,lookup,tensorflow/tensorflow/python/tf_program/mlir_gen.py,65,method,
10811,lookup_type,tensorflow/tensorflow/python/tf_program/mlir_gen.py,73,method,
10812,ProcessType,tensorflow/tensorflow/python/tf_program/mlir_gen.py,87,class,"Visit a node and return processed type Currently only visits annotations and gives their type.
"
10813,visit_Attribute,tensorflow/tensorflow/python/tf_program/mlir_gen.py,95,method,
10814,visit_Name,tensorflow/tensorflow/python/tf_program/mlir_gen.py,106,method,
10815,MLIRGen,tensorflow/tensorflow/python/tf_program/mlir_gen.py,115,class,"Visit the AST and generate MLIR code Requires liveness, reading_definitions.
"
10816,visit_block,tensorflow/tensorflow/python/tf_program/mlir_gen.py,125,method,
10817,process_type,tensorflow/tensorflow/python/tf_program/mlir_gen.py,128,method,
10818,visit_Assign,tensorflow/tensorflow/python/tf_program/mlir_gen.py,131,method,
10819,visit_BinOp,tensorflow/tensorflow/python/tf_program/mlir_gen.py,142,method,
10820,visit_BoolOp,tensorflow/tensorflow/python/tf_program/mlir_gen.py,153,method,
10821,visit_Call,tensorflow/tensorflow/python/tf_program/mlir_gen.py,162,method,
10822,visit_Compare,tensorflow/tensorflow/python/tf_program/mlir_gen.py,173,method,
10823,visit_Constant,tensorflow/tensorflow/python/tf_program/mlir_gen.py,199,method,
10824,visit_FunctionDef,tensorflow/tensorflow/python/tf_program/mlir_gen.py,209,method,
10825,visit_If,tensorflow/tensorflow/python/tf_program/mlir_gen.py,237,method,
10826,visit_Name,tensorflow/tensorflow/python/tf_program/mlir_gen.py,287,method,
10827,visit_Return,tensorflow/tensorflow/python/tf_program/mlir_gen.py,292,method,
10828,visit_Tuple,tensorflow/tensorflow/python/tf_program/mlir_gen.py,300,method,
10829,visit_UnaryOp,tensorflow/tensorflow/python/tf_program/mlir_gen.py,303,method,
10830,visit_While,tensorflow/tensorflow/python/tf_program/mlir_gen.py,371,method,
10831,mlir_gen_internal,tensorflow/tensorflow/python/tf_program/mlir_gen.py,417,function,Returns mlir module for unprocessed node `node`.
10832,mlir_gen,tensorflow/tensorflow/python/tf_program/mlir_gen.py,432,function,Parse a function and return TFProgram.
10833,mlir_gen_from_source,tensorflow/tensorflow/python/tf_program/mlir_gen.py,444,function,"Parse a function as either a string or from a supplied file path and return a TFProgram.
"
10834,IfOp,tensorflow/tensorflow/python/tf_program/pywrap_tfd.py,59,class,"tfp.if(cond) ({body}, {orelse}) : type If `cond` is true, `body` is
executed, otherwise `orelse` is executed."
10835,create,tensorflow/tensorflow/python/tf_program/pywrap_tfd.py,66,method,
10836,OrOp,tensorflow/tensorflow/python/tf_program/pywrap_tfd.py,75,class,"tfp.Or(ops...) This is like tf.Any, except that the first dimension is opened
into `ops`.
Returns a tensor of 1-bit integers which is ""Logical OR"" of the
coressponding elements in ops..."
10837,create,tensorflow/tensorflow/python/tf_program/pywrap_tfd.py,85,method,
10838,AndOp,tensorflow/tensorflow/python/tf_program/pywrap_tfd.py,93,class,"tfp.And(ops...) This is like tf.All, except that the first dimension is opened
to `ops`.
Returns a tensor of 1-bit integers which is ""Logical AND"" of the
coressponding elements in ops..."
10839,create,tensorflow/tensorflow/python/tf_program/pywrap_tfd.py,103,method,
10840,WhileOp,tensorflow/tensorflow/python/tf_program/pywrap_tfd.py,111,class,"tfp.While(init-vals, {
^bb1(cond-args):
cond-region
return cond
}, {
^bb1(body-args):
body-region
})
As long as `cond-region` returns a ""true""-like value, the body-region
is executed and the arguments are replaced by its return values for the next
iteration."
10841,create,tensorflow/tensorflow/python/tf_program/pywrap_tfd.py,127,method,
10842,TFProgram,tensorflow/tensorflow/python/tf_program/pywrap_tfd.py,136,class,Python wrap for a Tensorflow Program (essentially an mlir Module).
10843,add_function,tensorflow/tensorflow/python/tf_program/pywrap_tfd.py,146,method,
10844,get_function_type,tensorflow/tensorflow/python/tf_program/pywrap_tfd.py,152,method,
10845,dump,tensorflow/tensorflow/python/tf_program/pywrap_tfd.py,155,method,
10846,freeze_graph_with_def_protos,tensorflow/tensorflow/python/tools/freeze_graph.py,77,function,"Converts all variables in a graph and checkpoint into constants.
Args:
input_graph_def: A `GraphDef`.
input_saver_def: A `SaverDef` (optional).
input_checkpoint: The prefix of a V1 or V2 checkpoint, with V2 taking
priority. Typically the result of `Saver.save()` or that of
`tf.train.latest_checkpoint()`, regardless of sharded/non-sharded or
V1/V2.
output_node_names: The name(s) of the output nodes, comma separated.
restore_op_name: Unused.
filename_tensor_name: Unused.
output_graph: String where to write the frozen `GraphDef`.
clear_devices: A Bool whether to remove device specifications.
initializer_nodes: Comma separated string of initializer nodes to run before
freezing.
variable_names_whitelist: The set of variable names to convert (optional, by
default, all variables are converted).
variable_names_denylist: The set of variable names to omit converting
to constants (optional).
input_meta_graph_def: A `MetaGraphDef` (optional),
input_saved_model_dir: Path to the dir with TensorFlow 'SavedModel' file
and variables (optional).
saved_model_tags: Group of comma separated tag(s) of the MetaGraphDef to
load, in string format (optional).
checkpoint_version: Tensorflow variable file format (saver_pb2.SaverDef.V1
or saver_pb2.SaverDef.V2)
Returns:
Location of the output_graph_def."
10847,freeze_graph,tensorflow/tensorflow/python/tools/freeze_graph.py,286,function,"Converts all variables in a graph and checkpoint into constants.
Args:
input_graph: A `GraphDef` file to load.
input_saver: A TensorFlow Saver file.
input_binary: A Bool. True means input_graph is .pb, False indicates .pbtxt.
input_checkpoint: The prefix of a V1 or V2 checkpoint, with V2 taking
priority. Typically the result of `Saver.save()` or that of
`tf.train.latest_checkpoint()`, regardless of sharded/non-sharded or
V1/V2.
output_node_names: The name(s) of the output nodes, comma separated.
restore_op_name: Unused.
filename_tensor_name: Unused.
output_graph: String where to write the frozen `GraphDef`.
clear_devices: A Bool whether to remove device specifications.
initializer_nodes: Comma separated list of initializer nodes to run before
freezing.
variable_names_whitelist: The set of variable names to convert (optional, by
default, all variables are converted),
variable_names_denylist: The set of variable names to omit converting
to constants (optional).
input_meta_graph: A `MetaGraphDef` file to load (optional).
input_saved_model_dir: Path to the dir with TensorFlow 'SavedModel' file and
variables (optional).
saved_model_tags: Group of comma separated tag(s) of the MetaGraphDef to
load, in string format.
checkpoint_version: Tensorflow variable file format (saver_pb2.SaverDef.V1
or saver_pb2.SaverDef.V2).
Returns:
String that is the location of frozen GraphDef."
10848,import_to_tensorboard,tensorflow/tensorflow/python/tools/import_pb_to_tensorboard.py,43,function,"View an SavedModel as a graph in Tensorboard.
Args:
model_dir: The directory containing the SavedModel to import.
log_dir: The location for the Tensorboard log to begin visualization from.
tag_set: Group of tag(s) of the MetaGraphDef to load, in string format,
separated by ','. For tag-set contains multiple tags, all tags must be
passed in.
Usage: Call this function with your SavedModel location and desired log
directory. Launch Tensorboard by pointing it to the log directory. View your
imported SavedModel as a graph."
10849,print_tensors_in_checkpoint_file,tensorflow/tensorflow/python/tools/inspect_checkpoint.py,57,function,"Prints tensors in a checkpoint file.
If no `tensor_name` is provided, prints the tensor names and shapes
in the checkpoint file.
If `tensor_name` is provided, prints the content of the tensor.
Args:
file_name: Name of the checkpoint file.
tensor_name: Name of the tensor in the checkpoint file to print.
all_tensors: Boolean indicating whether to print all tensors.
all_tensor_names: Boolean indicating whether to print all tensor names.
count_exclude_pattern: Regex string, pattern to exclude tensors when count."
10850,parse_numpy_printoption,tensorflow/tensorflow/python/tools/inspect_checkpoint.py,115,function,"Sets a single numpy printoption from a string of the form 'x=y'.
See documentation on numpy.set_printoptions() for details about what values
x and y can take. x can be any option listed there other than 'formatter'.
Args:
kv_str: A string of the form 'x=y', such as 'threshold=100000'
Raises:
argparse.ArgumentTypeError: If the string couldn't be used to set any
nump printoption."
10851,get_parent_dir,tensorflow/tensorflow/python/tools/module_util.py,29,function,
10852,get_parent_dir_for_name,tensorflow/tensorflow/python/tools/module_util.py,33,function,"Get parent directory for module with the given name.
Args:
module_name: Module name for e.g.
tensorflow_estimator.python.estimator.api._v1.estimator.
Returns:
Path to the parent directory if module is found and None otherwise.
Given example above, it should return:
/pathtoestimator/tensorflow_estimator/python/estimator/api/_v1."
10853,parse_args,tensorflow/tensorflow/python/tools/optimize_for_inference.py,110,function,Parses command line arguments.
10854,optimize_for_inference,tensorflow/tensorflow/python/tools/optimize_for_inference_lib.py,93,function,"Applies a series of inference optimizations on the input graph.
Args:
input_graph_def: A GraphDef containing a training model.
input_node_names: A list of names of the nodes that are fed inputs during
inference.
output_node_names: A list of names of the nodes that produce the final
results.
placeholder_type_enum: The AttrValue enum for the placeholder data type, or
a list that specifies one value per input node name.
toco_compatible: Boolean, if True, only runs optimizations that result in
TOCO compatible graph operations (default=False).
Returns:
An optimized version of the input graph."
10855,ensure_graph_is_valid,tensorflow/tensorflow/python/tools/optimize_for_inference_lib.py,126,function,"Makes sure that the graph is internally consistent.
Checks basic properties of the graph def and raises an exception if there are
input references to missing nodes, duplicated names, or other logic errors.
Args:
graph_def: Definition of a graph to be checked.
Raises:
ValueError: If the graph is incorrectly constructed."
10856,node_name_from_input,tensorflow/tensorflow/python/tools/optimize_for_inference_lib.py,151,function,Strips off ports and other decorations to get the underlying node name.
10857,node_from_map,tensorflow/tensorflow/python/tools/optimize_for_inference_lib.py,161,function,"Pulls a node def from a dictionary for a given name.
Args:
node_map: Dictionary containing an entry indexed by name for every node.
name: Identifies the node we want to find.
Returns:
NodeDef of the node with the given name.
Raises:
ValueError: If the node isn't present in the dictionary."
10858,values_from_const,tensorflow/tensorflow/python/tools/optimize_for_inference_lib.py,180,function,"Extracts the values from a const NodeDef as a numpy ndarray.
Args:
node_def: Const NodeDef that has the values we want to access.
Returns:
Numpy ndarray containing the values.
Raises:
ValueError: If the node isn't a Const."
10859,scale_after_normalization,tensorflow/tensorflow/python/tools/optimize_for_inference_lib.py,202,function,
10860,fold_batch_norms,tensorflow/tensorflow/python/tools/optimize_for_inference_lib.py,208,function,"Removes batch normalization ops by folding them into convolutions.
Batch normalization during training has multiple dynamic parameters that are
updated, but once the graph is finalized these become constants. That means
there's an opportunity to reduce the computations down to a scale and
addition, rather than the more expensive multiple ops, and even bake the
scaling into the convolution weights. This function identifies the typical
pattern of batch normalization subgraphs, and performs the transformation to
fold the computations down into a simpler form. It currently only supports
batch normalization that's performed by the BatchNormWithGlobalNormalization
FusedBatchNorm and FusedBatchNormV3 ops, and will need to be extended in the
future to handle the newer style.
Args:
input_graph_def: A GraphDef containing a model.
Returns:
Modified graph with BN ops removed, and modified weights.
Raises:
ValueError: If the graph is badly formed with duplicate node names."
10861,fuse_resize_and_conv,tensorflow/tensorflow/python/tools/optimize_for_inference_lib.py,419,function,"Merges preceding resize and mirror pad ops into a specialized convolution.
There's a common pattern of enlarging the input to a convolution using a
resize operation, and also using MirrorPad to extend the boundaries to that
zero edge pixels don't bleed inwards when convolving. This routine looks for
that pattern of operations, and fuses them together into a Conv2DWithResizeOp.
Args:
input_graph_def: A GraphDef containing a model.
output_node_names: A list of names of the nodes that produce the final
results.
Returns:
Modified graph with resize and pad ops merged.
Raises:
ValueError: If the graph is badly formed with duplicate node names."
10862,aot_compile_cpu_meta_graph_def,tensorflow/tensorflow/python/tools/saved_model_aot_compile.py,212,function,"Compile a `MetaGraphDef` to header+object files in `output_prefix`.
Use XLA AOT (`tfcompile`) to convert the given meta graph and
signature into a header + object files. Also create an include makefile
that helps identify the appropriate necessary include and library paths
to incorporate these files into your C++ program.
The graph is always optimized with grappler, and optionally (by default)
variables are frozen as constants, before compilation happens.
If the `freeze_graph` is `True`, all variables are embedded as constants
into the graph and binary objects. If it is `False`, then the variable
values become inputs and outputs of the compiled class and the C++
caller must set these values manually.
Args:
checkpoint_path: Python string. Path to checkpoints/variables.
meta_graph_def: Instance of `MetaGraphDef`.
output_prefix: Python string. Path prefix for outputs.
signature_def_key: String, the signature_def to use in the SavedModel.
cpp_class: String, Name of output C++ class.
target_triple: String, LLVM target triple.
target_cpu: String, LLVM target cpu name.
variables_to_feed: A list of strings, the variables that will be fed by the
user; these won't be frozen. If `None`, then we will extract all the
variables in the graph and mark them as to-feed. The default behavior is
an empty tuple: all variables must be frozen.
enable_multithreading: Not implemented. Enable multithreading in the
compiled computation.
Raises:
RuntimeError: If tensorflow was not built with XLA.
ImportError: If tensorflow was built with XLA but there was another
issue importing the tfcompile python wrapper.
ValueError: If `meta_graph_def.signature_def[signature_def_key]` is
missing or has empty outputs.
NotImplementedError: If `enable_multithreading is True`."
10863,get_meta_graph_def,tensorflow/tensorflow/python/tools/saved_model_cli.py,310,function,"DEPRECATED: Use saved_model_utils.get_meta_graph_def instead.
Gets MetaGraphDef from SavedModel. Returns the MetaGraphDef for the given
tag-set and SavedModel directory.
Args:
saved_model_dir: Directory containing the SavedModel to inspect or execute.
tag_set: Group of tag(s) of the MetaGraphDef to load, in string format,
separated by ','. For tag-set contains multiple tags, all tags must be
passed in.
Raises:
RuntimeError: An error when the given tag-set does not exist in the
SavedModel.
Returns:
A MetaGraphDef corresponding to the tag-set."
10864,get_signature_def_map,tensorflow/tensorflow/python/tools/saved_model_cli.py,332,function,"Gets SignatureDef map from a MetaGraphDef in a SavedModel.
Returns the SignatureDef map for the given tag-set in the SavedModel
directory.
Args:
saved_model_dir: Directory containing the SavedModel to inspect or execute.
tag_set: Group of tag(s) of the MetaGraphDef with the SignatureDef map, in
string format, separated by ','. For tag-set contains multiple tags, all
tags must be passed in.
Returns:
A SignatureDef map that maps from string keys to SignatureDefs."
10865,scan_meta_graph_def,tensorflow/tensorflow/python/tools/saved_model_cli.py,351,function,"Scans meta_graph_def and reports if there are ops on denylist.
Print ops if they are on black list, or print success if no denylisted ops
found.
Args:
meta_graph_def: MetaGraphDef protocol buffer."
10866,run_saved_model_with_feed_dict,tensorflow/tensorflow/python/tools/saved_model_cli.py,373,function,"Runs SavedModel and fetch all outputs.
Runs the input dictionary through the MetaGraphDef within a SavedModel
specified by the given tag_set and SignatureDef. Also save the outputs to file
if outdir is not None.
Args:
saved_model_dir: Directory containing the SavedModel to execute.
tag_set: Group of tag(s) of the MetaGraphDef with the SignatureDef map, in
string format, separated by ','. For tag-set contains multiple tags, all
tags must be passed in.
signature_def_key: A SignatureDef key string.
input_tensor_key_feed_dict: A dictionary maps input keys to numpy ndarrays.
outdir: A directory to save the outputs to. If the directory doesn't exist,
it will be created.
overwrite_flag: A boolean flag to allow overwrite output file if file with
the same name exists.
worker: If provided, the session will be run on the worker. Valid worker
specification is a bns or gRPC path.
init_tpu: If true, the TPU system will be initialized after the session
is created.
tf_debug: A boolean flag to use TensorFlow Debugger (TFDBG) to observe the
intermediate Tensor values and runtime GraphDefs while running the
SavedModel.
Raises:
ValueError: When any of the input tensor keys is not valid.
RuntimeError: An error when output file already exists and overwrite is not
enabled."
10867,preprocess_inputs_arg_string,tensorflow/tensorflow/python/tools/saved_model_cli.py,474,function,"Parses input arg into dictionary that maps input to file/variable tuple.
Parses input string in the format of, for example,
""input1=filename1[variable_name1],input2=filename2"" into a
dictionary looks like
{'input_key1': (filename1, variable_name1),
'input_key2': (file2, None)}
, which maps input keys to a tuple of file name and variable name(None if
empty).
Args:
inputs_str: A string that specified where to load inputs. Inputs are
separated by semicolons.
* For each input key:
'<input_key>=<filename>' or
'<input_key>=<filename>[<variable_name>]'
* The optional 'variable_name' key will be set to None if not specified.
Returns:
A dictionary that maps input keys to a tuple of file name and variable name.
Raises:
RuntimeError: An error when the given input string is in a bad format."
10868,preprocess_input_exprs_arg_string,tensorflow/tensorflow/python/tools/saved_model_cli.py,521,function,"Parses input arg into dictionary that maps input key to python expression.
Parses input string in the format of 'input_key=<python expression>' into a
dictionary that maps each input_key to its python expression.
Args:
input_exprs_str: A string that specifies python expression for input keys.
Each input is separated by semicolon. For each input key:
'input_key=<python expression>'
Returns:
A dictionary that maps input keys to their values.
Raises:
RuntimeError: An error when the given input string is in a bad format."
10869,preprocess_input_examples_arg_string,tensorflow/tensorflow/python/tools/saved_model_cli.py,550,function,"Parses input into dict that maps input keys to lists of tf.Example.
Parses input string in the format of 'input_key1=[{feature_name:
feature_list}];input_key2=[{feature_name:feature_list}];' into a dictionary
that maps each input_key to its list of serialized tf.Example.
Args:
input_examples_str: A string that specifies a list of dictionaries of
feature_names and their feature_lists for each input.
Each input is separated by semicolon. For each input key:
'input=[{feature_name1: feature_list1, feature_name2:feature_list2}]'
items in feature_list can be the type of float, int, long or str.
Returns:
A dictionary that maps input keys to lists of serialized tf.Example.
Raises:
ValueError: An error when the given tf.Example is not a list."
10870,load_inputs_from_input_arg_string,tensorflow/tensorflow/python/tools/saved_model_cli.py,605,function,"Parses input arg strings and create inputs feed_dict.
Parses '--inputs' string for inputs to be loaded from file, and parses
'--input_exprs' string for inputs to be evaluated from python expression.
'--input_examples' string for inputs to be created from tf.example feature
dictionary list.
Args:
inputs_str: A string that specified where to load inputs. Each input is
separated by semicolon.
* For each input key:
'<input_key>=<filename>' or
'<input_key>=<filename>[<variable_name>]'
* The optional 'variable_name' key will be set to None if not specified.
* File specified by 'filename' will be loaded using numpy.load. Inputs
can be loaded from only .npy, .npz or pickle files.
* The ""[variable_name]"" key is optional depending on the input file type
as descripted in more details below.
When loading from a npy file, which always contains a numpy ndarray, the
content will be directly assigned to the specified input tensor. If a
variable_name is specified, it will be ignored and a warning will be
issued.
When loading from a npz zip file, user can specify which variable within
the zip file to load for the input tensor inside the square brackets. If
nothing is specified, this function will check that only one file is
included in the zip and load it for the specified input tensor.
When loading from a pickle file, if no variable_name is specified in the
square brackets, whatever that is inside the pickle file will be passed
to the specified input tensor, else SavedModel CLI will assume a
dictionary is stored in the pickle file and the value corresponding to
the variable_name will be used.
input_exprs_str: A string that specifies python expressions for inputs.
* In the format of: '<input_key>=<python expression>'.
* numpy module is available as np.
input_examples_str: A string that specifies tf.Example with dictionary.
* In the format of: '<input_key>=<[{feature:value list}]>'
Returns:
A dictionary that maps input tensor keys to numpy ndarrays.
Raises:
RuntimeError: An error when a key is specified, but the input file contains
multiple numpy ndarrays, none of which matches the given key.
RuntimeError: An error when no key is specified, but the input file contains
more than one numpy ndarrays."
10871,show,tensorflow/tensorflow/python/tools/saved_model_cli.py,708,function,"Function triggered by show command.
Args:
args: A namespace parsed from command line."
10872,run,tensorflow/tensorflow/python/tools/saved_model_cli.py,730,function,"Function triggered by run command.
Args:
args: A namespace parsed from command line.
Raises:
AttributeError: An error when neither --inputs nor --input_exprs is passed
to run command."
10873,scan,tensorflow/tensorflow/python/tools/saved_model_cli.py,752,function,"Function triggered by scan command.
Args:
args: A namespace parsed from command line."
10874,convert_with_tensorrt,tensorflow/tensorflow/python/tools/saved_model_cli.py,767,function,"Function triggered by 'convert tensorrt' command.
Args:
args: A namespace parsed from command line."
10875,aot_compile_cpu,tensorflow/tensorflow/python/tools/saved_model_cli.py,806,function,"Function triggered by aot_compile_cpu command.
Args:
args: A namespace parsed from command line."
10876,add_show_subparser,tensorflow/tensorflow/python/tools/saved_model_cli.py,834,function,Add parser for `show`.
10877,add_run_subparser,tensorflow/tensorflow/python/tools/saved_model_cli.py,880,function,Add parser for `run`.
10878,add_scan_subparser,tensorflow/tensorflow/python/tools/saved_model_cli.py,958,function,Add parser for `scan`.
10879,add_convert_subparser,tensorflow/tensorflow/python/tools/saved_model_cli.py,980,function,Add parser for `convert`.
10880,add_aot_compile_cpu_subparser,tensorflow/tensorflow/python/tools/saved_model_cli.py,1041,function,Add parser for `aot_compile_cpu`.
10881,create_parser,tensorflow/tensorflow/python/tools/saved_model_cli.py,1149,function,"Creates a parser that parse the command line arguments.
Returns:
A namespace parsed from command line arguments."
10882,captured_output,tensorflow/tensorflow/python/tools/saved_model_cli_test.py,51,function,
10883,read_saved_model,tensorflow/tensorflow/python/tools/saved_model_utils.py,31,function,"Reads the saved_model.pb or saved_model.pbtxt file containing `SavedModel`.
Args:
saved_model_dir: Directory containing the SavedModel file.
Returns:
A `SavedModel` protocol buffer.
Raises:
IOError: If the file does not exist, or cannot be successfully parsed."
10884,get_saved_model_tag_sets,tensorflow/tensorflow/python/tools/saved_model_utils.py,79,function,"Retrieves all the tag-sets available in the SavedModel.
Args:
saved_model_dir: Directory containing the SavedModel.
Returns:
List of all tag-sets in the SavedModel, where a tag-set is represented as a
list of strings."
10885,get_meta_graph_def,tensorflow/tensorflow/python/tools/saved_model_utils.py,96,function,"Gets MetaGraphDef from SavedModel.
Returns the MetaGraphDef for the given tag-set and SavedModel directory.
Args:
saved_model_dir: Directory containing the SavedModel to inspect.
tag_set: Group of tag(s) of the MetaGraphDef to load, in string format,
separated by ','. The empty string tag is ignored so that passing ''
means the empty tag set. For tag-set contains multiple tags, all tags
must be passed in.
Raises:
RuntimeError: An error when the given tag-set does not exist in the
SavedModel.
Returns:
A MetaGraphDef corresponding to the tag-set."
10886,tearDownModule,tensorflow/tensorflow/python/tools/saved_model_utils_test.py,33,function,
10887,get_ops_and_kernels,tensorflow/tensorflow/python/tools/selective_registration_header_lib.py,79,function,Gets the ops and kernels needed from the model files.
10888,get_header_from_ops_and_kernels,tensorflow/tensorflow/python/tools/selective_registration_header_lib.py,110,function,"Returns a header for use with tensorflow SELECTIVE_REGISTRATION.
Args:
ops_and_kernels: a set of (op_name, kernel_class_name) pairs to include.
include_all_ops_and_kernels: if True, ops_and_kernels is ignored and all op
kernels are included.
Returns:
the string of the header that should be written as ops_to_register.h."
10889,get_header,tensorflow/tensorflow/python/tools/selective_registration_header_lib.py,194,function,"Computes a header for use with tensorflow SELECTIVE_REGISTRATION.
Args:
graphs: a list of paths to GraphDef files to include.
proto_fileformat: optional format of proto file, either 'textproto',
'rawproto' (default) or ops_list. The ops_list is the file contain the
list of ops in JSON format, Ex: ""[[""Transpose"", ""TransposeCpuOp""]]"".
default_ops: optional comma-separated string of operator:kernel pairs to
always include implementation for. Pass 'all' to have all operators and
kernels included. Default: 'NoOp:NoOp,_Recv:RecvOp,_Send:SendOp'.
Returns:
the string of the header that should be written as ops_to_register.h."
10890,strip_unused,tensorflow/tensorflow/python/tools/strip_unused_lib.py,32,function,"Removes unused nodes from a GraphDef.
Args:
input_graph_def: A graph with nodes we want to prune.
input_node_names: A list of the nodes we use as inputs.
output_node_names: A list of the output nodes.
placeholder_type_enum: The AttrValue enum for the placeholder data type, or
a list that specifies one value per input node name.
Returns:
A `GraphDef` with all unnecessary ops removed.
Raises:
ValueError: If any element in `input_node_names` refers to a tensor instead
of an operation.
KeyError: If any element in `input_node_names` is not found in the graph."
10891,strip_unused_from_files,tensorflow/tensorflow/python/tools/strip_unused_lib.py,92,function,Removes unused nodes from a graph file.
10892,SymbolExposedTwiceError,tensorflow/tensorflow/python/tools/api/generator/create_python_api.py,73,class,Raised when different symbols are exported with the same name.
10893,get_canonical_import,tensorflow/tensorflow/python/tools/api/generator/create_python_api.py,78,function,"Obtain one single import from a set of possible sources of a symbol.
One symbol might come from multiple places as it is being imported and
reexported. To simplify API changes, we always use the same import for the
same module, and give preference based on higher priority and alphabetical
ordering.
Args:
import_set: (set) Imports providing the same symbol. This is a set of tuples
in the form (import, priority). We want to pick an import with highest
priority.
Returns:
A module name to import"
10894,add_nested_compat_imports,tensorflow/tensorflow/python/tools/api/generator/create_python_api.py,316,function,"Adds compat.vN.compat.vK modules to module builder.
To avoid circular imports, we want to add __init__.py files under
compat.vN.compat.vK and under compat.vN.compat.vK.compat. For all other
imports, we point to corresponding modules under compat.vK.
Args:
module_builder: `_ModuleInitCodeBuilder` instance.
compat_api_versions: Supported compatibility versions.
output_package: Base output python package where generated API will be
added."
10895,add_imports_for_symbol,tensorflow/tensorflow/python/tools/api/generator/create_python_api.py,405,function,"Add imports for the given symbol to `module_code_builder`.
Args:
module_code_builder: `_ModuleInitCodeBuilder` instance.
symbol: A symbol.
source_module_name: Module that we can import the symbol from.
source_name: Name we can import the symbol with.
api_name: API name. Currently, must be either `tensorflow` or `estimator`.
api_version: API version.
output_module_prefix: Prefix to prepend to destination module."
10896,get_api_init_text,tensorflow/tensorflow/python/tools/api/generator/create_python_api.py,450,function,"Get a map from destination module to __init__.py code for that module.
Args:
packages: Base python packages containing python with target tf_export
decorators.
output_package: Base output python package where generated API will be
added.
api_name: API you want to generate (e.g. `tensorflow` or `estimator`).
api_version: API version you want to generate (1 or 2).
compat_api_versions: Additional API versions to generate under compat/
directory.
lazy_loading: Boolean flag. If True, a lazy loading `__init__.py` file is
produced and if `False`, static imports are used.
use_relative_imports: True if we should use relative imports when importing
submodules.
Returns:
A dictionary where
key: (string) destination module (for e.g. tf or tf.consts).
value: (string) text that should be in __init__.py files for
corresponding modules."
10897,get_module,tensorflow/tensorflow/python/tools/api/generator/create_python_api.py,522,function,"Get module that corresponds to path relative to relative_to_dir.
Args:
dir_path: Path to directory.
relative_to_dir: Get module relative to this directory.
Returns:
Name of module that corresponds to the given directory."
10898,get_module_docstring,tensorflow/tensorflow/python/tools/api/generator/create_python_api.py,538,function,"Get docstring for the given module.
This method looks for docstring in the following order:
1. Checks if module has a docstring specified in doc_srcs.
2. Checks if module has a docstring source module specified
in doc_srcs. If it does, gets docstring from that module.
3. Checks if module with module_name exists under base package.
If it does, gets docstring from that module.
4. Returns a default docstring.
Args:
module_name: module name relative to tensorflow (excluding 'tensorflow.'
prefix) to get a docstring for.
package: Base python package containing python with target tf_export
decorators.
api_name: API you want to generate (e.g. `tensorflow` or `estimator`).
Returns:
One-line docstring to describe the module."
10899,create_api_files,tensorflow/tensorflow/python/tools/api/generator/create_python_api.py,586,function,"Creates __init__.py files for the Python API.
Args:
output_files: List of __init__.py file paths to create.
packages: Base python packages containing python with target tf_export
decorators.
root_init_template: Template for top-level __init__.py file. ""# API IMPORTS
PLACEHOLDER"" comment in the template file will be replaced with imports.
output_dir: output API root directory.
output_package: Base output package where generated API will be added.
api_name: API you want to generate (e.g. `tensorflow` or `estimator`).
api_version: API version to generate (`v1` or `v2`).
compat_api_versions: Additional API versions to generate in compat/
subdirectory.
compat_init_templates: List of templates for top level compat init files in
the same order as compat_api_versions.
lazy_loading: Boolean flag. If True, a lazy loading `__init__.py` file is
produced and if `False`, static imports are used.
use_relative_imports: True if we should use relative imports when import
submodules.
Raises:
ValueError: if output_files list is missing a required file."
10900,DocSource,tensorflow/tensorflow/python/tools/api/generator/doc_srcs.py,23,class,"Specifies docstring source for a module.
Only one of docstring or docstring_module_name should be set.
* If docstring is set, then we will use this docstring when
for the module.
* If docstring_module_name is set, then we will copy the docstring
from docstring source module."
10901,get_doc_sources,tensorflow/tensorflow/python/tools/api/generator/doc_srcs.py,86,function,"Get a map from module to a DocSource object.
Args:
api_name: API you want to generate (e.g. `tensorflow` or `estimator`).
Returns:
Map from module name to DocSource object."
10902,AsyncCheckpointSaverHook,tensorflow/tensorflow/python/tpu/async_checkpoint.py,39,class,Saves checkpoints every N steps or seconds.
10903,begin,tensorflow/tensorflow/python/tpu/async_checkpoint.py,92,method,
10904,after_create_session,tensorflow/tensorflow/python/tpu/async_checkpoint.py,101,method,
10905,before_run,tensorflow/tensorflow/python/tpu/async_checkpoint.py,125,method,
10906,after_run,tensorflow/tensorflow/python/tpu/async_checkpoint.py,128,method,
10907,end,tensorflow/tensorflow/python/tpu/async_checkpoint.py,136,method,
10908,input_fn,tensorflow/tensorflow/python/tpu/async_checkpoint_test.py,56,function,Return a dataset of source and target sequences for training.
10909,model_fn,tensorflow/tensorflow/python/tpu/async_checkpoint_test.py,65,function,
10910,bfloat16_scope,tensorflow/tensorflow/python/tpu/bfloat16.py,73,function,"Scope class for bfloat16 variables so that the model uses custom getter.
This enables variables to be read as bfloat16 type when using get_variable."
10911,StreamingFilesDataset,tensorflow/tensorflow/python/tpu/datasets.py,50,function,"StreamingFilesDataset constructs a dataset to stream from workers (GCE VM).
Because Cloud TPUs are allocated over the network, a Cloud TPU cannot read
files local to your GCE VM. In order to train using files stored on your local
VM (e.g. on local SSD for extreme performance), use the StreamingFilesDataset
helper to generate a dataset to feed your Cloud TPU with files from your GCE
VM.
The resulting dataset may return an OutOfRangeError if there are no files
found as a result of the fileglob expansion.
Note: StreamingFilesDataset assumes that the session is using a
TPUClusterResolver and has therefore a worker and a coordinator job. File
loading will be done on the coordinator job.
Args:
files: A string glob to match files, or a `tf.data.Dataset` generating file
names.
filetype: A string (one of 'tfrecord', or 'textline') or a single-argument
TensorFlow function that when given a filename returns a dataset.
file_reader_job: An optional string that corresponds to the job that should
perform the file reads.
worker_job: An optional string that corresponds to the job that should
process the tensors (i.e. your GPU or TPU worker).
num_epochs: The number of epochs through the training set that should be
generated. By default, it will repeat infinitely.
filename_shuffle_buffer_size: An optional integer whose value controls the
shuffling of the file names. If you would like to read from the files in
the same order, set to 0 or False.
num_parallel_reads: An optional integer controlling the number of files to
read from concurrently. (Set to 1 for no parallelism.)
batch_transfer_size: An optional integer controlling the batching used to
amortize the remote function invocation overhead. Set to a very large
number to increase throughput. Set to a very small number to reduce memory
consumption. Set to False to skip batching.
sloppy: (Optional.) If `False`, read input data while maintaining a
deterministic order. (This may have significant performance impacts.)
sloppy defaults to: True.
Returns:
A `tf.data.Dataset` with an infinite stream of elements generated by a
parallel interleaving of the set of files matched (or generated) by `files`
with a type is the output of the dataset specified by `filetype`.
Raises:
ValueError: if any argument is not of the expected type."
10912,DeviceAssignment,tensorflow/tensorflow/python/tpu/device_assignment.py,60,class,"Mapping from logical cores in a computation to the physical TPU topology.
Prefer to use the `DeviceAssignment.build()` helper to construct a
`DeviceAssignment`; it is easier if less flexible than constructing a
`DeviceAssignment` directly."
10913,topology,tensorflow/tensorflow/python/tpu/device_assignment.py,106,method,A `Topology` that describes the TPU topology.
10914,num_cores_per_replica,tensorflow/tensorflow/python/tpu/device_assignment.py,111,method,The number of cores per replica.
10915,num_replicas,tensorflow/tensorflow/python/tpu/device_assignment.py,116,method,The number of replicas of the computation.
10916,core_assignment,tensorflow/tensorflow/python/tpu/device_assignment.py,121,method,"The logical to physical core mapping.
Returns:
An integer numpy array of rank 3, with shape
`[num_replicas, num_cores_per_replica, topology_rank]`. Maps
(replica, logical core) pairs to physical topology coordinates."
10917,coordinates,tensorflow/tensorflow/python/tpu/device_assignment.py,131,method,Returns the physical topology coordinates of a logical core.
10918,lookup_replicas,tensorflow/tensorflow/python/tpu/device_assignment.py,135,method,"Lookup replica ids by task number and logical core.
Args:
task_id: TensorFlow task number.
logical_core: An integer, identifying a logical core.
Returns:
A sorted list of the replicas that are attached to that task and
logical_core.
Raises:
ValueError: If no replica exists in the task which contains the logical
core."
10919,tpu_ordinal,tensorflow/tensorflow/python/tpu/device_assignment.py,155,method,Returns the ordinal of the TPU device assigned to a logical core.
10920,host_device,tensorflow/tensorflow/python/tpu/device_assignment.py,160,method,Returns the CPU device attached to a logical core.
10921,tpu_device,tensorflow/tensorflow/python/tpu/device_assignment.py,165,method,Returns the name of the TPU device assigned to a logical core.
10922,build,tensorflow/tensorflow/python/tpu/device_assignment.py,171,method,
10923,device_assignment,tensorflow/tensorflow/python/tpu/device_assignment.py,316,function,"Computes a device_assignment of a computation across a TPU topology.
Attempts to choose a compact grid of cores for locality.
Returns a `DeviceAssignment` that describes the cores in the topology assigned
to each core of each replica.
`computation_shape` and `computation_stride` values should be powers of 2 for
optimal packing.
Args:
topology: A `Topology` object that describes the TPU cluster topology.
To obtain a TPU topology, evaluate the `Tensor` returned by
`initialize_system` using `Session.run`. Either a serialized
`TopologyProto` or a `Topology` object may be passed. Note: you must
evaluate the `Tensor` first; you cannot pass an unevaluated `Tensor` here.
computation_shape: A rank 1 int32 numpy array with size equal to the
topology rank, describing the shape of the computation's block of cores.
If None, the `computation_shape` is `[1] * topology_rank`.
computation_stride: A rank 1 int32 numpy array of size `topology_rank`,
describing the inter-core spacing of the `computation_shape` cores in the
TPU topology. If None, the `computation_stride` is `[1] * topology_rank`.
num_replicas: The number of computation replicas to run. The replicas will
be packed into the free spaces of the topology.
Returns:
A DeviceAssignment object, which describes the mapping between the logical
cores in each computation replica and the physical cores in the TPU
topology.
Raises:
ValueError: If `topology` is not a valid `Topology` object.
ValueError: If `computation_shape` or `computation_stride` are not 1D int32
numpy arrays with shape [3] where all values are positive.
ValueError: If computation's replicas cannot fit into the TPU topology."
10924,embedding_column,tensorflow/tensorflow/python/tpu/feature_column.py,55,function,"TPU embedding_column for `tf.feature_column.embedding_column`.
Note that the interface for TPU embedding_column is different from the non-TPU
version. The following args available for the non-TPU version are NOT
supported: ckpt_to_load_from, tensor_name_in_ckp, max_norm and trainable.
Args:
categorical_column: A categorical_column returned from
categorical_column_with_identity, weighted_categorical_column,
categorical_column_with_vocabulary_file,
categorical_column_with_vocabulary_list,
sequence_categorical_column_with_identity,
sequence_categorical_column_with_vocabulary_file,
sequence_categorical_column_with_vocabulary_list
dimension: An integer specifying dimension of the embedding, must be > 0.
combiner: A string specifying how to reduce if there are multiple entries
in a single row for a non-sequence column. For more information, see
`tf.feature_column.embedding_column`.
initializer: A variable initializer function to be used in embedding
variable initialization. If not specified, defaults to
`tf.compat.v1.truncated_normal_initializer` with mean `0.0` and
standard deviation `1/sqrt(dimension)`.
max_sequence_length: An non-negative integer specifying the max sequence
length. Any sequence shorter then this will be padded with 0 embeddings
and any sequence longer will be truncated. This must be positive for
sequence features and 0 for non-sequence features.
learning_rate_fn: A function that takes global step and returns learning
rate for the embedding table. If you intend to use the same learning rate
for multiple embedding tables, please ensure that you pass the exact same
python function to all calls of embedding_column, otherwise performence
may suffer.
use_safe_embedding_lookup: If true, uses safe_embedding_lookup_sparse
instead of embedding_lookup_sparse. safe_embedding_lookup_sparse ensures
there are no empty rows and all weights and ids are positive at the
expense of extra compute cost. This only applies to rank 2 (NxM) shaped
input tensors. Defaults to true, consider turning off if the above checks
are not needed. Note that having empty rows will not trigger any error
though the output result might be 0 or omitted.
Returns:
A _TPUEmbeddingColumn.
Raises:
ValueError: if `dimension` not > 0.
ValueError: if `initializer` is specified but not callable.
TypeError: if categorical_column is not a supported type."
10925,shared_embedding_columns,tensorflow/tensorflow/python/tpu/feature_column.py,161,function,"List of dense columns that convert from sparse, categorical input.
Note that the interface for TPU embedding_column is different from the non-TPU
version. The following args available for the non-TPU version are NOT
supported: ckpt_to_load_from, tensor_name_in_ckp, max_norm and trainable.
Args:
categorical_columns: A list of categorical_columns returned from
categorical_column_with_identity, weighted_categorical_column,
categorical_column_with_vocabulary_file,
categorical_column_with_vocabulary_list,
sequence_categorical_column_with_identity,
sequence_categorical_column_with_vocabulary_file,
sequence_categorical_column_with_vocabulary_list
dimension: An integer specifying dimension of the embedding, must be > 0.
combiner: A string specifying how to reduce if there are multiple entries
in a single row for a non-sequence column. For more information, see
`tf.feature_column.embedding_column`.
initializer: A variable initializer function to be used in embedding
variable initialization. If not specified, defaults to
`tf.truncated_normal_initializer` with mean `0.0` and standard deviation
`1/sqrt(dimension)`.
shared_embedding_collection_name: Optional name of the collection where
shared embedding weights are added. If not given, a reasonable name will
be chosen based on the names of `categorical_columns`. This is also used
in `variable_scope` when creating shared embedding weights.
max_sequence_lengths: An list of non-negative integers, either None or
empty or the same length as the argument categorical_columns. Entries
corresponding to non-sequence columns must be 0 and entries corresponding
to sequence columns specify the max sequence length for the column. Any
sequence shorter then this will be padded with 0 embeddings and any
sequence longer will be truncated.
learning_rate_fn: A function that takes global step and returns learning
rate for the embedding table. If you intend to use the same learning rate
for multiple embedding tables, please ensure that you pass the exact same
python function to all calls of shared_embedding_columns, otherwise
performence may suffer.
use_safe_embedding_lookup: If true, uses safe_embedding_lookup_sparse
instead of embedding_lookup_sparse. safe_embedding_lookup_sparse ensures
there are no empty rows and all weights and ids are positive at the
expense of extra compute cost. This only applies to rank 2 (NxM) shaped
input tensors. Defaults to true, consider turning off if the above checks
are not needed. Note that having empty rows will not trigger any error
though the output result might be 0 or omitted.
Returns:
A _TPUEmbeddingColumn.
Raises:
ValueError: if `dimension` not > 0.
ValueError: if `initializer` is specified but not callable.
ValueError: if `max_sequence_lengths` is specified and not the same length
as `categorical_columns`.
ValueError: if `max_sequence_lengths` is positive for a non sequence column
or 0 for a sequence column."
10926,get_sequence_length_feature_key_name_from_feature_key_name,tensorflow/tensorflow/python/tpu/feature_column.py,666,function,"Gets the name of the sequence length feature from that of the base feature.
Args:
feature_name: The feature key of a sequence column.
Returns:
A string which is the feature key for the associated feature length column."
10927,split_sequence_columns,tensorflow/tensorflow/python/tpu/feature_column.py,678,function,"Split a list of _TPUEmbeddingColumn into sequence and non-sequence columns.
For use in a TPUEstimator model_fn function. E.g.
def model_fn(features):
sequence_columns, feature_columns = (
tf.tpu.feature_column.split_sequence_columns(feature_columns))
input = tf.feature_column.input_layer(
features=features, feature_columns=feature_columns)
sequence_features, sequence_lengths = (
tf.contrib.feature_column.sequence_input_layer(
features=features, feature_columns=sequence_columns))
Args:
feature_columns: A list of _TPUEmbeddingColumns to split.
Returns:
Two lists of _TPUEmbeddingColumns, the first is the sequence columns and the
second is the non-sequence columns."
10928,EmbeddingDevice,tensorflow/tensorflow/python/tpu/feature_column_v2.py,48,class,
10929,embedding_column_v2,tensorflow/tensorflow/python/tpu/feature_column_v2.py,55,function,"TPU version of `tf.compat.v1.feature_column.embedding_column`.
Note that the interface for `tf.tpu.experimental.embedding_column` is
different from that of `tf.compat.v1.feature_column.embedding_column`: The
following arguments are NOT supported: `ckpt_to_load_from`,
`tensor_name_in_ckpt`, `max_norm` and `trainable`.
Use this function in place of `tf.compat.v1.feature_column.embedding_column`
when you want to use the TPU to accelerate your embedding lookups via TPU
embeddings.
```
column = tf.feature_column.categorical_column_with_identity(...)
tpu_column = tf.tpu.experimental.embedding_column(column, 10)
...
def model_fn(features):
dense_feature = tf.keras.layers.DenseFeature(tpu_column)
embedded_feature = dense_feature(features)
...
estimator = tf.estimator.tpu.TPUEstimator(
model_fn=model_fn,
...
embedding_config_spec=tf.estimator.tpu.experimental.EmbeddingConfigSpec(
column=[tpu_column],
...))
```
Args:
categorical_column: A categorical column returned from
`categorical_column_with_identity`, `weighted_categorical_column`,
`categorical_column_with_vocabulary_file`,
`categorical_column_with_vocabulary_list`,
`sequence_categorical_column_with_identity`,
`sequence_categorical_column_with_vocabulary_file`,
`sequence_categorical_column_with_vocabulary_list`
dimension: An integer specifying dimension of the embedding, must be > 0.
combiner: A string specifying how to reduce if there are multiple entries
in a single row for a non-sequence column. For more information, see
`tf.feature_column.embedding_column`.
initializer: A variable initializer function to be used in embedding
variable initialization. If not specified, defaults to
`tf.compat.v1.truncated_normal_initializer` with mean `0.0` and
standard deviation `1/sqrt(dimension)`.
max_sequence_length: An non-negative integer specifying the max sequence
length. Any sequence shorter then this will be padded with 0 embeddings
and any sequence longer will be truncated. This must be positive for
sequence features and 0 for non-sequence features.
learning_rate_fn: A function that takes global step and returns learning
rate for the embedding table. If you intend to use the same learning rate
for multiple embedding tables, please ensure that you pass the exact same
python function to all calls of embedding_column, otherwise performence
may suffer.
embedding_lookup_device: The device on which to run the embedding lookup.
Valid options are ""cpu"", ""tpu_tensor_core"", and ""tpu_embedding_core"".
If specifying ""tpu_tensor_core"", a tensor_core_shape must be supplied.
If not specified, the default behavior is embedding lookup on
""tpu_embedding_core"" for training and ""cpu"" for inference.
Valid options for training : [""tpu_embedding_core"", ""tpu_tensor_core""]
Valid options for serving : [""cpu"", ""tpu_tensor_core""]
For training, tpu_embedding_core is good for large embedding vocab (>1M),
otherwise, tpu_tensor_core is often sufficient.
For serving, doing embedding lookup on tpu_tensor_core during serving is
a way to reduce host cpu usage in cases where that is a bottleneck.
tensor_core_shape: If supplied, a list of integers which specifies
the intended dense shape to run embedding lookup for this feature on
TensorCore. The batch dimension can be left None or -1 to indicate
a dynamic shape. Only rank 2 shapes currently supported.
use_safe_embedding_lookup: If true, uses safe_embedding_lookup_sparse
instead of embedding_lookup_sparse. safe_embedding_lookup_sparse ensures
there are no empty rows and all weights and ids are positive at the
expense of extra compute cost. This only applies to rank 2 (NxM) shaped
input tensors. Defaults to true, consider turning off if the above checks
are not needed. Note that having empty rows will not trigger any error
though the output result might be 0 or omitted.
Returns:
A `_TPUEmbeddingColumnV2`.
Raises:
ValueError: if `dimension` not > 0.
ValueError: if `initializer` is specified but not callable."
10930,shared_embedding_columns_v2,tensorflow/tensorflow/python/tpu/feature_column_v2.py,211,function,"TPU version of `tf.compat.v1.feature_column.shared_embedding_columns`.
Note that the interface for `tf.tpu.experimental.shared_embedding_columns` is
different from that of `tf.compat.v1.feature_column.shared_embedding_columns`:
The following arguments are NOT supported: `ckpt_to_load_from`,
`tensor_name_in_ckpt`, `max_norm` and `trainable`.
Use this function in place of
tf.compat.v1.feature_column.shared_embedding_columns` when you want to use the
TPU to accelerate your embedding lookups via TPU embeddings.
```
column_a = tf.feature_column.categorical_column_with_identity(...)
column_b = tf.feature_column.categorical_column_with_identity(...)
tpu_columns = tf.tpu.experimental.shared_embedding_columns(
[column_a, column_b], 10)
...
def model_fn(features):
dense_feature = tf.keras.layers.DenseFeature(tpu_columns)
embedded_feature = dense_feature(features)
...
estimator = tf.estimator.tpu.TPUEstimator(
model_fn=model_fn,
...
embedding_config_spec=tf.estimator.tpu.experimental.EmbeddingConfigSpec(
column=tpu_columns,
...))
```
Args:
categorical_columns: A list of categorical columns returned from
`categorical_column_with_identity`, `weighted_categorical_column`,
`categorical_column_with_vocabulary_file`,
`categorical_column_with_vocabulary_list`,
`sequence_categorical_column_with_identity`,
`sequence_categorical_column_with_vocabulary_file`,
`sequence_categorical_column_with_vocabulary_list`
dimension: An integer specifying dimension of the embedding, must be > 0.
combiner: A string specifying how to reduce if there are multiple entries in
a single row for a non-sequence column. For more information, see
`tf.feature_column.embedding_column`.
initializer: A variable initializer function to be used in embedding
variable initialization. If not specified, defaults to
`tf.truncated_normal_initializer` with mean `0.0` and standard deviation
`1/sqrt(dimension)`.
shared_embedding_collection_name: Optional name of the collection where
shared embedding weights are added. If not given, a reasonable name will
be chosen based on the names of `categorical_columns`. This is also used
in `variable_scope` when creating shared embedding weights.
max_sequence_lengths: An list of non-negative integers, either None or empty
or the same length as the argument categorical_columns. Entries
corresponding to non-sequence columns must be 0 and entries corresponding
to sequence columns specify the max sequence length for the column. Any
sequence shorter then this will be padded with 0 embeddings and any
sequence longer will be truncated.
learning_rate_fn: A function that takes global step and returns learning
rate for the embedding table. If you intend to use the same learning rate
for multiple embedding tables, please ensure that you pass the exact same
python function to all calls of shared_embedding_columns, otherwise
performence may suffer.
embedding_lookup_device: The device on which to run the embedding lookup.
Valid options are ""cpu"", ""tpu_tensor_core"", and ""tpu_embedding_core"". If
specifying ""tpu_tensor_core"", a tensor_core_shape must be supplied.
Defaults to ""cpu"". If not specified, the default behavior is embedding
lookup on ""tpu_embedding_core"" for training and ""cpu"" for inference.
Valid options for training : [""tpu_embedding_core"", ""tpu_tensor_core""]
Valid options for serving : [""cpu"", ""tpu_tensor_core""]
For training, tpu_embedding_core is good for large embedding vocab (>1M),
otherwise, tpu_tensor_core is often sufficient.
For serving, doing embedding lookup on tpu_tensor_core during serving is
a way to reduce host cpu usage in cases where that is a bottleneck.
tensor_core_shape: If supplied, a list of integers which specifies the
intended dense shape to run embedding lookup for this feature on
TensorCore. The batch dimension can be left None or -1 to indicate a
dynamic shape. Only rank 2 shapes currently supported.
use_safe_embedding_lookup: If true, uses safe_embedding_lookup_sparse
instead of embedding_lookup_sparse. safe_embedding_lookup_sparse ensures
there are no empty rows and all weights and ids are positive at the
expense of extra compute cost. This only applies to rank 2 (NxM) shaped
input tensors. Defaults to true, consider turning off if the above checks
are not needed. Note that having empty rows will not trigger any error
though the output result might be 0 or omitted.
Returns:
A list of `_TPUSharedEmbeddingColumnV2`.
Raises:
ValueError: if `dimension` not > 0.
ValueError: if `initializer` is specified but not callable.
ValueError: if `max_sequence_lengths` is specified and not the same length
as `categorical_columns`.
ValueError: if `max_sequence_lengths` is positive for a non sequence column
or 0 for a sequence column."
10931,split_sequence_columns_v2,tensorflow/tensorflow/python/tpu/feature_column_v2.py,742,function,"Split a list of _TPUEmbeddingColumn into sequence and non-sequence columns.
For use in a TPUEstimator model_fn function. E.g.
def model_fn(features):
sequence_columns, feature_columns = (
tf.tpu.feature_column.split_sequence_columns(feature_columns))
input = tf.feature_column.input_layer(
features=features, feature_columns=feature_columns)
sequence_features, sequence_lengths = (
tf.contrib.feature_column.sequence_input_layer(
features=features, feature_columns=sequence_columns))
Args:
feature_columns: A list of _TPUEmbeddingColumns to split.
Returns:
Two lists of _TPUEmbeddingColumns, the first is the sequence columns and the
second is the non-sequence columns."
10932,sparse_embedding_aggregate_slice,tensorflow/tensorflow/python/tpu/feature_column_v2.py,778,function,"Uses XLA's dynamic slice operations to perform embedding lookups.
From third_party/cloud_tpu/models/movielens/tpu_embedding.py
Args:
params: Tensor of embedding table. Rank 2 (table_size x embedding dim)
values_and_values_mask: is a two-tuple that contains: values - Tensor of
embedding indices. Rank 2 (batch x n_indices) values_mask - Tensor of mask
/ weights. Rank 2 (batch x n_indices)
combiner: The combiner to use for the embedding lookup. Currently supports
'sum' and 'mean'.
name: Optional name scope for created ops
Returns:
Rank 2 tensor of aggregated (per batch element) embedding vectors.
Raises:
ValueError: Combiner is not supported."
10933,pad_sparse_embedding_lookup_indices,tensorflow/tensorflow/python/tpu/feature_column_v2.py,832,function,"Creates statically-sized Tensors containing indices and weights.
From third_party/cloud_tpu/models/movielens/tpu_embedding.py
Also computes sparse_indices.values % embedding_table_size, for equivalent
functionality to sparse_column_with_integerized_feature. The returned
padded weight Tensor also doubles as a mask indicating which values in
the returned padded indices Tensor are indices versus padded zeros.
Args:
sparse_indices: SparseTensor of embedding lookup indices.
padded_size: Number of columns of the returned Tensors. Indices which fall
out of bounds will be truncated to the padded size.
Returns:
(sparse_indices.values padded to the specified size,
a mask the same size as the returned padded values in which 0s
indicate padded locations and 1s (or values from sparse_weights)
indicate actual values)"
10934,CloudTPUPreemptedHook,tensorflow/tensorflow/python/tpu/preempted_hook.py,31,class,"The SessionRunHook for preemptible Cloud TPUs.
This is an implementation of SessionRunHook for the pre-emptible Google Cloud
TPU service. It attempts to close the session if the TPU is preempted, and
exits the coordinator process if the session cannot be closed."
10935,after_create_session,tensorflow/tensorflow/python/tpu/preempted_hook.py,42,method,
10936,end,tensorflow/tensorflow/python/tpu/preempted_hook.py,47,method,
10937,CoordinatorResetError,tensorflow/tensorflow/python/tpu/session_support.py,41,class,Raised when the monitored session should reset.
10938,WorkerHeartbeatManager,tensorflow/tensorflow/python/tpu/session_support.py,56,class,Manages the status/heartbeat monitor for a set of workers.
10939,from_devices,tensorflow/tensorflow/python/tpu/session_support.py,77,method,Construct a heartbeat manager for the given devices.
10940,num_workers,tensorflow/tensorflow/python/tpu/session_support.py,94,method,
10941,configure,tensorflow/tensorflow/python/tpu/session_support.py,97,method,"Configure heartbeat manager for all devices.
Args:
message: `event_pb2.WorkerHeartbeatRequest`
Returns: `None`"
10942,ping,tensorflow/tensorflow/python/tpu/session_support.py,109,method,"Ping all workers, returning the parsed status results."
10943,lame_workers,tensorflow/tensorflow/python/tpu/session_support.py,126,method,"Ping all workers, returning manager containing lame workers (or None)."
10944,shutdown,tensorflow/tensorflow/python/tpu/session_support.py,148,method,Shutdown all workers after `shutdown_timeout_secs`.
10945,all_worker_devices,tensorflow/tensorflow/python/tpu/session_support.py,163,function,Return a list of devices for each worker in the system.
10946,WatchdogManager,tensorflow/tensorflow/python/tpu/session_support.py,178,class,"Configures worker watchdog timer and handles periodic pings.
Usage:
# Ping workers every minute, shutting down workers if they haven't received
# a ping after 1 hour.
watchdog_manager = WatchdogManager(
ping_interval=60, shutdown_timeout=3600
)
# Use as a context manager, resetting watchdog on context exit:
with watchdog_manager:
session.run(...)
# Or setup globally; watchdog will remain active until program exit.
watchdog_manager.configure_and_run()"
10947,configure_and_run,tensorflow/tensorflow/python/tpu/session_support.py,252,method,
10948,stop,tensorflow/tensorflow/python/tpu/session_support.py,261,method,
10949,run,tensorflow/tensorflow/python/tpu/session_support.py,273,method,
10950,start_worker_watchdog,tensorflow/tensorflow/python/tpu/session_support.py,287,function,Start global worker watchdog to shutdown workers on coordinator exit.
10951,stop_worker_watchdog,tensorflow/tensorflow/python/tpu/session_support.py,301,function,Stop global worker watchdog.
10952,GracefulShutdownHook,tensorflow/tensorflow/python/tpu/session_support.py,309,class,"Session hook that watches for shutdown events.
If a shutdown is indicated, `saver.save(checkpoint_prefix)` is executed, and a
SystemShutdown exception is raised to terminate the main session. If `saver`
is None the `SAVERS` collection will be read to find a saver.
`on_shutdown_hooks` is an optional list of functions that should be called
after checkpointing. The function is called with (`run_context`,
`all_workers`, `lame_workers`).
If `heartbeat_group` is not specified, it will default to all CPU workers
in the system."
10953,after_create_session,tensorflow/tensorflow/python/tpu/session_support.py,335,method,
10954,saver,tensorflow/tensorflow/python/tpu/session_support.py,363,method,
10955,after_run,tensorflow/tensorflow/python/tpu/session_support.py,383,method,
10956,ResetComputation,tensorflow/tensorflow/python/tpu/session_support.py,409,class,"Hook to reset a TPUEstimator computation loop.
This hook shuts down all workers and resets the monitored session loop by
throwing a CoordinatorResetError."
10957,ShutdownLameWorkers,tensorflow/tensorflow/python/tpu/session_support.py,427,class,"Shutdown lamed workers.
Processing will continue normally (typically by waiting for the down
workers to be restarted)."
10958,ShutdownAllWorkers,tensorflow/tensorflow/python/tpu/session_support.py,441,class,"Shutdown all workers.
Processing will continue normally (typically by waiting for the down
workers to be restarted)."
10959,set_parameters,tensorflow/tensorflow/python/tpu/tensor_tracer.py,106,function,"Enables tensor tracer and sets its parameters.
Example usage:
tensor_tracer_parameters = {'trace_dir': '/usr/tmp/trace_dir',
'trace_mode': 'norm',
'report_file': '/usr/tmp/trace_dir/report.all'}
tensor_tracer.set_parameters(tensor_tracer_parameters)
This sets up the parameters for tensor tracer. A call to tensor tracer as
below is necessary to enable debugging on CPUs and GPUs. On TPUs below can be
skipped as this call is hooked into tpu.rewrite.
tt = tensor_tracer.TensorTracer()
loss = tt.trace_cpu(tf.get_default_graph(), tensor_fetches=loss)
Args:
tensor_tracer_params: Tensor tracer parameter dictionary. Below gives
examples of these parameters: See tensor_tracer_report.py for all
parameters.
- enable: If set, tensor tracer will be enabled. Calling
enable_tensor_tracer automatically adds this parameters.
- trace_mode: The trace_mode to be used by tensor tracer. These include:
- summary: Collects multiple statistics for traced tensors, and writes
them a summary file that can be visualized using tensorboard. This
mode currently only works for TPUEstimator. It can be also be used
for other models, but outfeed must be handled by the user.
- norm: Collects norm of each traced tensor and writes them into a
text file pointed by 'trace_dir' flag. (Default mode).
- nan-inf: Checks the existince of NaNs and Infs in the tensor, and
writes a boolean value to a text file pointed by 'trace_dir' flag.
Note that 'norm' mode can also capture this information with more
numerical info.
- max-abs: Collects the absolute max for each traced tensors and
writes it into a text file pointed by 'trace_dir' flag.
- full-tensor: Writes the full tensor content of the traced tensors
into a text file pointed by 'trace_dir' flag.
- part-tensor: Writes a part of the tensor content of the traced
tensors into a text file pointed by 'trace_dir' flag.
- full_tensor_summary: Writes the full tensors as binary event files.
The outputs can be read using: trace =
tensor_tracer.read_tensor_tracer_event_file(event_file_path)
- trace-back-if-nan: This mode will write the full tensor content only
when the tensor has a NaN or Inf in it. It is possible to also print
the inputs coming to this op using 'trace_stack_size' parameter.
E.g., if trace_stack_size=2, then the tensor with NaN/Inf, its
inputs, and its inputs' inputs will also be printed.
- report_file: Path to the metadata file that is written during graph
construction. If not set, metadata will be printed to stdout during
graph construction.
- trace_dir: Path where the execution traces will be written during the
graph execution. If not set, trace will be printed to stderr.
- trace_level: Tensor tracer aims to trace everything it can. This
introduces some overhead on graph execution and graph compilation
times. Using trace_level parameter, it is possible to trace operation
based on their priorities. For example, - trace_level=7 is the highest
trace_level, in which every op is traced. - trace_level=6 will skip
constant operations such as tf.constant. - trace_level=5 will skip
less important ops such as tf.identities. - The default trace_level=3,
that will skip concat ops, or random number generators. - To reduce
the graph compile time overhead, trace_level can be set to 0, that
will skip additions, and substractions, and multiplications as well.
- excluded_opnames: If set, any matching op name will not be traced.
excluded_opnames can be set as a regular expression. E.g,
excluded_opnames=.* will exclude everything.
- excluded_optypes: If set, any matching op type will not be traced.
excluded_optypes can be set as a regular expression. E.g,
excluded_optypes=.* will exclude everything. excluded_optypes=MatMul
will exclude all MatMul ops from tracing.
- included_opnames: If set, any matching op name will be forced to be
traced. included_opnames can be set as a regular expression. E.g,
'--included_opnames=some_op --excluded_opname=*.' will only trace
some_op.
- included_optypes: If set, any matching op type will be forced to be
traced. included_optypes can be set as a regular expression. E.g,
'--included_optypes=some_op_type --excluded_optypes=*.' will trace
only the ops with type 'some_op_type'
Advanced Flags:
- compact_trace: If not set, statistics per tensor is written as soon as
they are executed. If set, then statistics for all traced tensors will
be stored in a cache and will be written only once per step. This flag
is ignored for full-tensor and part-tensor trace modes. If the
trace_dir is a remote directory, compact_trace will be forced.
- trace_scalar: Scalar values are not traced by default. If this flag is
set, scalar values will also be traced.
- included_cores: Accepts a list string. Tracing will only be dumped for
these cores. E.g, setting it to '[0,2,4,6]' will result in a trace
only for those cores.
- op_range: In the form of '%d:%d' that limits the tracing to the ops
within this limit. --op_range='5:10' will trace only the ops that have
topological order between 5-10.
- trace_before_included_ops: If set to a number-k, it will also trace
distance-k inputs of each traced tensor. E.g., k=1, then in addition
to each traced_tensor, their input tensors will also be traced.
- trace_after_included_ops: Same as trace_before_included_ops, where it
will also trace distance-k outputs of each traced tensor.
- submode: 'brief' or 'detailed'. If the trace mode is not compact,
brief mode will print only the id of each traced tensor to save some
space. 'detailed' mode prints the full tensor name.
- trace_stack_size: Used only for trace_mode=trace-back-if-nan mode. It
determines how many ops to print back from a nan op. E.g, op4 -> op3
-> op2 -> op1 -> op0, if op0 has a NaN and trace_stack_size is 1, the
result of op1 will also be printed. trace_stack_size is 2, the result
of op1 and op2 will be printed.
- use_fingerprint_subdirectory: The trace directory will be chosen as
using the fingerprint of the trace metadata under the provided
trace_dir."
10960,op_priority,tensorflow/tensorflow/python/tpu/tensor_tracer.py,220,function,"Returns the priority of the op.
If the priority of the op is k, it will be traced if trace_level>=k.
Args:
op_type: String name of the operation type.
Returns:
Integer value corresponding the priority of the op."
10961,read_tensor_tracer_event_file,tensorflow/tensorflow/python/tpu/tensor_tracer.py,263,function,"Reads the event file written by tensor tracer.
This can be used to read the full tensors written into binary event files by
by TensorTracer with trace_mode=full_tensor_summary.
Example usage:
result_dict = tensor_tracer.read_tensor_tracer_event_file(event_file_path)
for step, tensor_dict in result_dict.items():
for tensor_name, full_tensor_content in tensor_dict.items():
logging.info(tensor_name, full_tensor_content)
Args:
event_file: Path to the event file that contains only tensor tracer events.
Returns:
An event dictionary in the form of
{step_number: {tensor_name: tensor_content}}
Raises:
ValueError: If an unexpected trace is found."
10962,trace_tensor,tensorflow/tensorflow/python/tpu/tensor_tracer.py,307,function,"Programmatic interface to trace a tensor with Tensor Tracer.
Tensor Tracer, by default, traces all tensors in the execution. This function
can be used to limit traced tensors. If this function is called for a subset
of the tensors, only those will be traced.
For example, Tensor Traacer will only trace c below.
c = tf.MatMul(a, b)
tensor_tracer.trace_tensor(c)
d = tf.add(c, 1)
Args:
tensor: the tensor object for which the tracing is requested.
tracepoint_name: an optional tensor tracepoint name string. A tracepoint
name is an Tensor Tracer internal name for the tensor. It is useful when
comparing equivalent traces from different models that have different
tensor namings. Equivalent tensors (with different names) can be mapped
to each other by assigning a common tracepoint_name.
Returns:
The provided tensor."
10963,keras_layer_tracepoint,tensorflow/tensorflow/python/tpu/tensor_tracer.py,337,function,"An interface for adding the tensor outputs of a keras layer.
Encapsulates trace_tensor.
Args:
layer: A keras layer.
checkpoint_name: a string name for the checkpoint. This name has to be a
unique name if used within model comparison. The tensors that have the same
checkpoint identifier is compared in model comparison.
Returns:
The provided layer."
10964,TensorTracer,tensorflow/tensorflow/python/tpu/tensor_tracer.py,386,class,"A software construct for tracing tensor values in a TF graph.
This utility is disabled by default. It is hooked into tpu.rewrite, so it can
easily be enabled on TPUs by setting the TENSOR_TRACER_FLAGS env variable as
below without a code change.
export TENSOR_TRACER_FLAGS=""--enable=1""
Below is the use example to enable it on CPUs or GPUs, or for more advance use
cases on TPUs.
a = x + 1
b = a * 2
rs = tf.reduce_sum(b)
tensor_tracer.set_parameters({'trace_dir': 'path/to/trace_dir',
'report_file: 'path/to/report/file'})
tt = tensor_tracer.TensorTracer()
if on_tpu:
rs = tt.trace_tpu(tf.get_default_graph(),
tensor_fetches=rs)
else:
rs = tt.trace_cpu(tf.get_default_graph(),
tensor_fetches=rs)
session.run(rs)
If it is enabled, it will trace the output tensor values of
selected Ops in the graph. It has two outputs: (1) the traces and (2)
a report. The traces are dumped to a specified directory during the graph
execution, while the report is dumped during the graph construction.
By passing options via the env variable, users can change:
(1) the trace mode (e.g., detecting NaN/Inf, printing partial or
full tensor values)
(2) which Ops to be traced (via op.name or op.type)
(3) output trace file path."
10965,is_enabled,tensorflow/tensorflow/python/tpu/tensor_tracer.py,426,method,Returns True if TensorTracer is enabled.
10966,check_device_type,tensorflow/tensorflow/python/tpu/tensor_tracer.py,431,method,Checks if the given device type is valid.
10967,check_trace_mode,tensorflow/tensorflow/python/tpu/tensor_tracer.py,438,method,"Checks if the given trace mode work on the given device type.
Args:
device_type: Device type, TPU, GPU, CPU.
trace_mode: Tensor tracer trace mode.
Raises:
ValueError: If the given trace mode is not supported for the device."
10968,loop_cond_op,tensorflow/tensorflow/python/tpu/tensor_tracer.py,453,method,
10969,while_loop_op,tensorflow/tensorflow/python/tpu/tensor_tracer.py,457,method,"Returns true if op is one of the special ops of in a while loop.
Args:
op: A tf.Operation.
Returns:
True if the given op is one of [Switch, Merge, Enter, Exit,
NextIteration, LoopCond], which are all building blocks for TF while
loops."
10970,control_flow_op,tensorflow/tensorflow/python/tpu/tensor_tracer.py,476,method,"Returns true if op is one of the special ops of in a while loop.
Args:
op: A tf.Operation.
Returns:
True if the given op is one of [Switch, Merge, Enter, Exit,
NextIteration, LoopCond], which are all building blocks for TF while
loops."
10971,unsafe_op,tensorflow/tensorflow/python/tpu/tensor_tracer.py,491,method,Returns True if this op is not safe to be traced.
10972,device_mismatch,tensorflow/tensorflow/python/tpu/tensor_tracer.py,501,method,
10973,unsafe_scalar_trace,tensorflow/tensorflow/python/tpu/tensor_tracer.py,509,method,Return true if scalar output tensor from Op is not safe to be traced.
10974,reason,tensorflow/tensorflow/python/tpu/tensor_tracer.py,535,method,Returns reason why the Op at op_idx is traced or not.
10975,report_proto,tensorflow/tensorflow/python/tpu/tensor_tracer.py,555,method,"Getter for tensor_tracer.proto object for summary and full_tensor_summary modes.
Returns:
A tensor_tracer.proto object.
Raises:
ValueError if called before tracing happens, or when trace mode is not
summary or full_tensor_summary."
10976,report_proto_path,tensorflow/tensorflow/python/tpu/tensor_tracer.py,571,method,"Getter for path where tensor_tracer.proto object should be written.
Returns:
A string path."
10977,merge_caches_on_tpu,tensorflow/tensorflow/python/tpu/tensor_tracer.py,1564,method,"Merges the given caches on tpu.
Args:
local_tpu_cache_tensor: A local tensor that needs to be merged
by concanting data from other tpu cores.
Returns:
A merged tf.Tensor.
Raises:
RuntimeError: if there is no aggregate function defined for a signature."
10978,aggregate_global_cache,tensorflow/tensorflow/python/tpu/tensor_tracer.py,1588,method,"Merges the given caches on tpu.
Args:
global_tt_summary_cache: The global tensor tracer summary cache tensor
with shape (num_cores, num_traced_tensors, num_traced_signatures). First
dimension corresponds to core_id, where global_tpu_cache_tensor[i]
correspond to the local cache from core-i.
Returns:
An aggregated tf.Tensor.
Raises:
RuntimeError: if there is no aggregate function defined for a signature."
10979,host_call_deps_and_fn,tensorflow/tensorflow/python/tpu/tensor_tracer.py,1723,method,
10980,get_traced_op_names,tensorflow/tensorflow/python/tpu/tensor_tracer.py,1726,method,Returns the set of traced op names.
10981,trace_tpu,tensorflow/tensorflow/python/tpu/tensor_tracer.py,1936,method,"Traces the tensors generated by TPU Ops in a TF graph.
Args:
graph: the graph of Ops executed on the TPU.
tensor_fetches: a (list,tuple,or a single object) of tensor fetches
returned by model_fn given to session.run. Function must be provided
with as least one tensor to fetch.
op_fetches: A list of op fetches returned by model_fn given to
session.run. op_fetches and tensor_fetches are used to determine the
nodes that will be executed. Can be None.
num_replicas: number of replicas used on the TPU.
num_replicas_per_host: number of replicas per TPU host.
num_hosts: total number of TPU hosts.
Returns:
tensor_fetches: an exact copy of tensor_fetches that has additional
dependencies.
Raises:
RuntimeError: If num_replicas_per_host > 8.
RuntimeError: If tensor_fetches is None or empty."
10982,trace_cpu,tensorflow/tensorflow/python/tpu/tensor_tracer.py,1995,method,"Traces the tensors generated by CPU Ops in a TF graph.
Args:
graph: the graph of Ops executed on the CPU.
tensor_fetches: a (list,tuple,or a single object) of tensor fetches
returned by model_fn given to session.run. Function must be provided
with as least one tensor to fetch.
op_fetches: A list of op fetches returned by model_fn given to
session.run. op_fetches and tensor_fetches are used to determine the
nodes that will be executed. Can be None.
Returns:
tensor_fetches: an exact copy of tensor_fetches that has additional
dependencies.
Raises:
RuntimeError: If tensor_fetches is None or empty."
10983,tpu_wrap_trace_fn,tensorflow/tensorflow/python/tpu/tensor_tracer.py,1852,method,Wraps the trace_fn with outside compilation if on TPUs.
10984,conditional_trace_fn,tensorflow/tensorflow/python/tpu/tensor_tracer.py,1861,method,Creates a cond op that traces the out_tensor if predicate is satisfied.
10985,TTParameters,tensorflow/tensorflow/python/tpu/tensor_tracer_flags.py,103,class,A class that handles the parameters of Tensor Tracer.
10986,is_brief_mode,tensorflow/tensorflow/python/tpu/tensor_tracer_flags.py,217,method,
10987,match_next_flag,tensorflow/tensorflow/python/tpu/tensor_tracer_flags.py,236,method,"Returns the match for the next TensorTracer flag.
Args:
flags: a string that contains the flags.
pos: where in flags to start the search.
Returns:
A pair where the first element is the regular-expression
match found and the second element indicates if the match
has a value."
10988,get_signature_to_agg_fn_map,tensorflow/tensorflow/python/tpu/tensor_tracer_flags.py,322,method,Returns a map that contains the aggregate function for each signature.
10989,get_flag_value,tensorflow/tensorflow/python/tpu/tensor_tracer_flags.py,392,method,"Returns the value of a TensorTracer flags.
Args:
wanted_flag_name: the name of the flag we are looking for.
Returns:
A pair where the first element indicates if the flag is
found and the second element is the value of the flag.
Raises:
RuntimeError: If supposedly deadcode is reached."
10990,is_flag_on,tensorflow/tensorflow/python/tpu/tensor_tracer_flags.py,438,method,Returns True if the given flag is on.
10991,is_enabled,tensorflow/tensorflow/python/tpu/tensor_tracer_flags.py,451,method,Returns True if TensorTracer is enabled.
10992,report_proto_path,tensorflow/tensorflow/python/tpu/tensor_tracer_report.py,58,function,"Returns the path where report proto should be written.
Args:
trace_dir: String denoting the trace directory.
Returns:
A string denoting the path to the report proto."
10993,topological_sort,tensorflow/tensorflow/python/tpu/tensor_tracer_report.py,70,function,"Performs topological sort on the given graph.
Args:
g: the graph.
Returns:
A pair where the first element indicates if the topological
sort succeeded (True if there is no cycle found; False if a
cycle is found) and the second element is either the sorted
list of nodes or the cycle of nodes found."
10994,TensorTracerConfig,tensorflow/tensorflow/python/tpu/tensor_tracer_report.py,136,class,Tensor Tracer config object.
10995,TensorTraceOrder,tensorflow/tensorflow/python/tpu/tensor_tracer_report.py,147,class,Class that is responsible from storing the trace-id of the tensors.
10996,sort_tensors_and_ops,tensorflow/tensorflow/python/tpu/tensor_tracer_report.py,178,function,Returns a wrapper that has consistent tensor and op orders.
10997,OpenReportFile,tensorflow/tensorflow/python/tpu/tensor_tracer_report.py,203,class,Context manager for writing report file.
10998,proto_fingerprint,tensorflow/tensorflow/python/tpu/tensor_tracer_report.py,223,function,
10999,TTReportHandle,tensorflow/tensorflow/python/tpu/tensor_tracer_report.py,229,class,Utility class responsible from creating a tensor tracer report.
11000,instrument,tensorflow/tensorflow/python/tpu/tensor_tracer_report.py,236,method,
11001,instrument_op,tensorflow/tensorflow/python/tpu/tensor_tracer_report.py,239,method,
11002,instrument_tensor,tensorflow/tensorflow/python/tpu/tensor_tracer_report.py,242,method,
11003,create_report_proto,tensorflow/tensorflow/python/tpu/tensor_tracer_report.py,245,method,"Creates and returns a proto that stores tensor tracer configuration.
Args:
tt_config: TensorTracerConfig object holding information about the run
environment (device, # cores, # hosts), and tensor tracer version
information.
tt_parameters: TTParameters objects storing the user provided parameters
for tensor tracer.
tensor_trace_order: TensorTraceOrder object storing a topological order of
the graph.
tensor_trace_points: Progromatically added trace_points/checkpoints.
collected_signature_types: The signature types collected, e,g, norm,
max, min, mean...
Returns:
TensorTracerReport proto."
11004,write_report_proto,tensorflow/tensorflow/python/tpu/tensor_tracer_report.py,307,method,Writes the given report proto under trace_dir.
11005,create_report,tensorflow/tensorflow/python/tpu/tensor_tracer_report.py,314,method,Creates a report file and writes the trace information.
11006,Topology,tensorflow/tensorflow/python/tpu/topology.py,45,class,"Describes a set of TPU devices.
Represents both the shape of the physical mesh, and the mapping between
TensorFlow TPU devices to physical mesh coordinates."
11007,mesh_shape,tensorflow/tensorflow/python/tpu/topology.py,150,method,A rank 1 int32 array describing the shape of the TPU topology.
11008,mesh_rank,tensorflow/tensorflow/python/tpu/topology.py,155,method,Returns the number of dimensions in the mesh.
11009,device_coordinates,tensorflow/tensorflow/python/tpu/topology.py,160,method,"Describes the mapping from TPU devices to topology coordinates.
Returns:
A rank 3 int32 array with shape `[tasks, devices, axis]`.
`tasks` is the number of tasks in the TPU cluster, `devices` is the number
of TPU devices per task, and `axis` is the number of axes in the TPU
cluster topology. Each entry gives the `axis`-th coordinate in the
topology of a task/device pair. TPU topologies are 4-dimensional, with
dimensions `(x, y, z, core number)`."
11010,missing_devices,tensorflow/tensorflow/python/tpu/topology.py,174,method,Array of indices of missing devices.
11011,task_ordinal_at_coordinates,tensorflow/tensorflow/python/tpu/topology.py,178,method,"Returns the TensorFlow task number attached to `device_coordinates`.
Args:
device_coordinates: An integer sequence describing a device's physical
coordinates in the TPU fabric.
Returns:
Returns the TensorFlow task number that contains the TPU device with those
physical coordinates."
11012,tpu_device_ordinal_at_coordinates,tensorflow/tensorflow/python/tpu/topology.py,191,method,"Returns the TensorFlow device number at `device_coordinates`.
Args:
device_coordinates: An integer sequence describing a device's physical
coordinates in the TPU fabric.
Returns:
Returns the TensorFlow device number within the task corresponding to
attached to the device with those physical coordinates."
11013,cpu_device_name_at_coordinates,tensorflow/tensorflow/python/tpu/topology.py,204,method,Returns the CPU device attached to a logical core.
11014,tpu_device_name_at_coordinates,tensorflow/tensorflow/python/tpu/topology.py,209,method,Returns the name of the TPU device assigned to a logical core.
11015,num_tasks,tensorflow/tensorflow/python/tpu/topology.py,216,method,Returns the number of TensorFlow tasks in the TPU slice.
11016,num_tpus_per_task,tensorflow/tensorflow/python/tpu/topology.py,221,method,Returns the number of TPU devices per task in the TPU slice.
11017,serialized,tensorflow/tensorflow/python/tpu/topology.py,225,method,Returns the serialized form of the topology.
11018,initialize_system,tensorflow/tensorflow/python/tpu/tpu.py,99,function,"Initializes a distributed TPU system for use with TensorFlow.
Args:
embedding_config: If not None, a `TPUEmbeddingConfiguration` proto
describing the desired configuration of the hardware embedding lookup
tables. If embedding_config is None, no hardware embeddings can be used.
job: The job (the XXX in TensorFlow device specification /job:XXX) that
contains the TPU devices that will be initialized. If job=None it is
assumed there is only one job in the TensorFlow flock, and an error will
be returned if this assumption does not hold.
compilation_failure_closes_chips: Set the configuration whether
we want to close TPU chips when there is a compilation failure.
Returns:
A serialized `TopologyProto` that describes the TPU system. Note:
the topology must be evaluated using `Session.run` before it can be used."
11019,initialize_system_for_tpu_embedding,tensorflow/tensorflow/python/tpu/tpu.py,137,function,"Initializes a distributed TPU Embedding system for use with TensorFlow.
The following two are equivalent:
1. initialize_system() with embedding_config.
2. initialize_system() without embedding_config, then
initialize_system_for_tpu_embedding().
initialize_system() should not be called with embedding_config if
initialize_system_for_tpu_embedding() is meant to be called later.
Args:
embedding_config: a `TPUEmbeddingConfiguration` proto describing the desired
configuration of the hardware embedding lookup tables.
job: The job (the XXX in TensorFlow device specification /job:XXX) that
contains the TPU devices that will be initialized. If job=None it is
assumed there is only one job in the TensorFlow flock, and an error will
be returned if this assumption does not hold.
Returns:
A no-op."
11020,shutdown_system,tensorflow/tensorflow/python/tpu/tpu.py,164,function,"Shuts down a running a distributed TPU system.
Args:
job: The job (the XXX in TensorFlow device specification /job:XXX) that
contains the TPU devices that will be shutdown. If job=None it is
assumed there is only one job in the TensorFlow flock, and an error will
be returned if this assumption does not hold."
11021,core,tensorflow/tensorflow/python/tpu/tpu.py,179,function,"Returns the device name for a core in a replicated TPU computation.
Args:
num: the virtual core number within each replica to which operators should
be assigned.
Returns:
A device name, suitable for passing to `tf.device()`."
11022,is_tpu_strategy,tensorflow/tensorflow/python/tpu/tpu.py,208,function,
11023,tpu_replicated_input_resolver,tensorflow/tensorflow/python/tpu/tpu.py,224,function,Replaces TPUReplicatedInput outputs with its inputs in resource_inputs.
11024,TPUReplicateContext,tensorflow/tensorflow/python/tpu/tpu.py,256,class,"A `ControlFlowContext` for nodes inside a TPU computation.
The primary role of `TPUReplicateContext` is to mark operators inside a
tpu.replicate() computation with the attribute ""_tpu_replicate=XYZ"", where XYZ
is a unique name.
We use a `ControlFlowContext` to perform the annotation since it integrates
with Tensorflow constructs like ResourceVariables. For example, if a
`ResourceVariable` is constructed inside a tpu.replicate() block, the
`ResourceVariable` implementation can use
`with ops.control_dependencies(None)` to build the variable's definition
outside the replicated computation."
11025,get_replicated_var_handle,tensorflow/tensorflow/python/tpu/tpu.py,301,method,"Returns a variable handle for replicated TPU variable 'var'.
This is a method used by an experimental replicated variable implementation
and is not intended as a public API.
Args:
name: The common name of the variable.
vars_: The replicated TPU variables.
is_mirrored: Whether the variables are mirrored, which guarantees the
values in each replica are always the same.
is_packed: Whether the replicated variables are packed into one variable.
Returns:
The handle of the TPU replicated input node."
11026,report_unsupported_operations,tensorflow/tensorflow/python/tpu/tpu.py,368,method,
11027,EnterGradientColocation,tensorflow/tensorflow/python/tpu/tpu.py,378,method,
11028,ExitGradientColocation,tensorflow/tensorflow/python/tpu/tpu.py,417,method,
11029,Enter,tensorflow/tensorflow/python/tpu/tpu.py,489,method,
11030,HostComputeCore,tensorflow/tensorflow/python/tpu/tpu.py,499,method,
11031,AddOp,tensorflow/tensorflow/python/tpu/tpu.py,526,method,
11032,AddValue,tensorflow/tensorflow/python/tpu/tpu.py,605,method,Add `val` to the current context and its outer context recursively.
11033,AddInnerOp,tensorflow/tensorflow/python/tpu/tpu.py,622,method,
11034,grad_state,tensorflow/tensorflow/python/tpu/tpu.py,628,method,
11035,back_prop,tensorflow/tensorflow/python/tpu/tpu.py,636,method,"Forwards to the enclosing while context, if any."
11036,GetControlPivot,tensorflow/tensorflow/python/tpu/tpu.py,642,method,
11037,RequiresUniqueFunctionRetracing,tensorflow/tensorflow/python/tpu/tpu.py,645,method,
11038,type,tensorflow/tensorflow/python/tpu/tpu.py,447,method,
11039,device,tensorflow/tensorflow/python/tpu/tpu.py,451,method,
11040,OutsideCompilationV2Context,tensorflow/tensorflow/python/tpu/tpu.py,652,class,"The context for outside compilation in Tensorflow 2.0.
Every op added in this context will be assigned an _xla_outside_compilation
attribute."
11041,AddOp,tensorflow/tensorflow/python/tpu/tpu.py,663,method,
11042,AddInnerOp,tensorflow/tensorflow/python/tpu/tpu.py,671,method,
11043,to_control_flow_context_def,tensorflow/tensorflow/python/tpu/tpu.py,679,method,
11044,outside_compilation,tensorflow/tensorflow/python/tpu/tpu.py,684,function,"Builds part of a computation outside any current TPU replicate scope.
`tf.tpu.outside_compilation()` is used to run ops in `computation` on CPU
instead of running on TPU. For example, users can run ops that are not
supported on TPU's (e.g. tf.summary.write()) by explicitly placing those
ops on CPU's. Below usage of outside compilation will place ops in
`computation_with_string_ops` on CPU.
Example usage:
```python
def computation_with_string_ops(x):
# strings types are not supported on TPU's and below ops must
# run on CPU instead.
output = tf.strings.format('1{}', x)
return tf.strings.to_number(output)
def tpu_computation():
# Expected output is 11.
output = tf.tpu.outside_compilation(computation_with_string_ops, 1)
```
Outside compilation should be called inside TPUReplicateContext. That is,
`tf.tpu.outside_compilation()` should be called inside a function that is
passed to `tpu.split_compile_and_replicate()` -- this is implied when
outside compilation is invoked inside a function passed to TPUStrategy
`run()`. If invoked outside of TPUReplicateContext,
then this simply returns the result of `computation`, and therefore,
would be a no-op. Note that outside compilation is different from
`tf.distribute.experimental.TPUStrategy.merge_call()` as logic in
outside compilation is replicated and executed separately for each
replica. On the other hand, `merge_call()` requires a `merge_fn`
to aggregate the inputs from different replicas and is executed only
once.
For variables placed in TPU device, which includes variables created inside
TPUStrategy scope, outside compilation logic must not include variable
read/write. For variables placed on host, which is the case when variables
created via TPUEstimator, variable read/write is only allowed if the variable
is not accessed by any other ops in the TPU computation. Variable read/write
from outside compilation cluster is not visible from TPU computation and
vice versa. Therefore, if outside compilation logic contains such host
variables read/write ops and if the variables are accessed by TPU
computation as well, then this may lead to deadlock.
Internally, `tf.tpu.outside_compilation()` adds outside compilation
attributes to all ops in `computation`. During later graph pass, these
ops with outside compilation attribute is extracted out and replicated
into a host-side graph. Inputs to this extract host-side graph is sent
from TPU computation graph to host graph via a pair of XlaSendToHost and
XlaRecvFromHost ops. Note that using `tf.tpu.outside_compilation()`
may result in tensor transfer between TPU and CPU, leading to non-trivial
performance impact.
Args:
computation: A Python function that builds the computation to
place on the host.
*args: the positional arguments for the computation.
**kwargs: the keyword arguments for the computation.
Returns:
The Tensors returned by computation."
11045,PaddingSpec,tensorflow/tensorflow/python/tpu/tpu.py,806,class,Represents the type of padding policies for tpu.replicate.
11046,XLAOptions,tensorflow/tensorflow/python/tpu/tpu.py,816,class,"XLA compilation options.
Attributes:
use_spmd_for_xla_partitioning: Boolean. Whether to use XLA's SPMD
partitioner instead of MPMD partitioner when compiler partitioning is
requested."
11047,replicate,tensorflow/tensorflow/python/tpu/tpu.py,833,function,"Builds a graph operator that runs a replicated TPU computation.
Example for the basic usage that `inputs` has static shape:
```python
def computation(x):
x = x + 1
return tf.math.reduce_mean(x)
x = tf.convert_to_tensor([1., 2., 3.])
y = tf.convert_to_tensor([4., 5., 6.])
tf.compat.v1.tpu.replicate(computation, inputs=[[x], [y]])
```
If the `inputs` has dynamic shapes and you would like to automatically
bucketize the inputs to avoid XLA recompilation. See the advanced example
below:
```python
def computation(x):
x = x + 1
return tf.math.reduce_mean(x)
# Assume input tensors in two replicas `x` and `y` both have dynamic shape
# ([None, 2]).
tf.compat.v1.tpu.replicate(
computation,
inputs=[x, y],
maximum_shapes=[tf.TensorShape([None, None])],
padding_spec=tf.compat.v1.tpu.PaddingSpec.POWER_OF_TWO)
```
Args:
computation: A Python function that builds the computation to replicate.
inputs: A list of lists of input tensors or `None` (equivalent to
`[[]]`), indexed by `[replica_num][input_num]`. All replicas must
have the same number of inputs. Each input can be a nested structure
containing values that are convertible to tensors. Note that passing an
N-dimension list of compatible values will result in a N-dimension list of
scalar tensors rather than a single Rank-N tensors. If you need different
behavior, convert part of inputs to tensors with `tf.convert_to_tensor`.
infeed_queue: If not `None`, the `InfeedQueue` from which to append a tuple
of arguments as inputs to computation.
device_assignment: If not `None`, a `DeviceAssignment` describing the
mapping between logical cores in the computation with physical cores in
the TPU topology. Uses a default device assignment if `None`. The
`DeviceAssignment` may be omitted if each replica of the computation uses
only one core, and there is either only one replica, or the number of
replicas is equal to the number of cores in the TPU system.
name: (Deprecated) Does nothing.
maximum_shapes: A nested structure of tf.TensorShape representing the shape
to which the respective component of each input element in each replica
should be padded. Any unknown dimensions (e.g.
tf.compat.v1.Dimension(None) in a tf.TensorShape or -1 in a tensor-like
object) will be padded to the maximum size of that dimension over all
replicas. The structure of `maximum_shapes` needs to be the same as
`inputs[0]`.
padding_spec: An enum specified by `tpu.PaddingSpec`. This describes the
padding policy when the `inputs` to `tpu.replicate` is dynamic.
One usage is to enable automatic bucketizing on the inputs by setting the
value to `tpu.PaddingSpec.POWER_OF_TWO`, which can help to reduce the
recompilation in the XLA side.
xla_options: An instance of `tpu.XLAOptions` which indicates the options
passed to XLA compiler. Use `None` for default options.
Returns:
A list of outputs, indexed by `[replica_num]` each output can be a nested
structure same as what computation() returns with a few exceptions.
Exceptions include:
1) None output: a NoOp would be returned which control-depends on
computation.
2) Single value output: A tuple containing the value would be returned.
3) Operation-only outputs: a NoOp would be returned which
control-depends on computation.
TODO(b/121383831): Investigate into removing these special cases.
Raises:
ValueError: If all replicas do not have equal numbers of input tensors.
ValueError: If the number of inputs per replica does not match
the number of formal parameters to `computation`.
ValueError: If the static `inputs` dimensions don't match with the values
given in `maximum_shapes`.
ValueError: If the structure of inputs per replica does not match
the structure of `maximum_shapes`."
11048,split_compile_and_replicate,tensorflow/tensorflow/python/tpu/tpu.py,1121,function,"Builds graph operators that runs compilation and replicated computation.
This is a lower level interface than replicate that returns a separate compile
and execute output tensor. In the generated graph the compile op feeds into
the execute op and no additional compilation is incurred when running the
compile op before the execute op. The compile op returns additional
information about the compilation but does not return the compiled program.
Args:
computation: A Python function that builds the computation to replicate.
inputs: A list of lists of input tensors or `None` (equivalent to
`[[]]`), indexed by `[replica_num][input_num]`. All replicas must
have the same number of inputs. Each input can be a nested structure
containing values that are convertible to tensors. Note that passing an
N-dimension list of compatible values will result in a N-dimension list of
scalar tensors rather than a single Rank-N tensors. If you need different
behavior, convert part of inputs to tensors with `tf.convert_to_tensor`.
infeed_queue: If not `None`, the `InfeedQueue` from which to append a tuple
of arguments as inputs to computation.
device_assignment: If not `None`, a `DeviceAssignment` describing the
mapping between logical cores in the computation with physical cores in
the TPU topology. Uses a default device assignment if `None`. The
`DeviceAssignment` may be omitted if each replica of the computation uses
only one core, and there is either only one replica, or the number of
replicas is equal to the number of cores in the TPU system.
name: (Deprecated) Does nothing.
use_tpu: When false, the input `computation` is executed on the XLA CPU/GPU
backends. Currently, only supports a default placement (computation is
placed on GPU if one is available, and on CPU if not).
maximum_shapes: A nested structure of tf.TensorShape representing the shape
to which the respective component of each input element in each replica
should be padded. Any unknown dimensions (e.g.
tf.compat.v1.Dimension(None) in a tf.TensorShape or -1 in a tensor-like
object) will be padded to the maximum size of that dimension over all
replicas. The structure of `maximum_shapes` needs to be the same as
`inputs[0]`.
padding_spec: An enum specified by `tf.tpu.PaddingSpec`. This describes the
padding policy when the `inputs` to `tf.tpu.replicate` is dynamic.
One usage is to enable automatic bucketizing on the inputs by setting the
value to `tpu.PaddingSpec.POWER_OF_TWO`, which can help to reduce the
recompilation in the XLA side.
xla_options: An instance of `tpu.XLAOptions` which indicates the options
passed to XLA compiler. Use `None` for default options.
Returns:
A list of lists with the first list corresponding to the compile op and the
second a list of output tensors, indexed by `[replica_num][output_num]`.
Raises:
ValueError: If all replicas do not have equal numbers of input tensors.
ValueError: If the number of inputs per replica does not match
the number of formal parameters to `computation`.
ValueError: If the static `inputs` dimensions don't match with the values
given in `maximum_shapes`.
ValueError: If the structure of inputs per replica does not match
the structure of `maximum_shapes`."
11049,split_compile_and_shard,tensorflow/tensorflow/python/tpu/tpu.py,1609,function,"Shards `computation` for parallel execution.
`inputs` must be a list of Tensors or None (equivalent to an empty list), each
of which has a corresponding split axis (from `input_shard_axes`). Each input
is split into `num_shards` pieces along the corresponding axis, and
computation is applied to each shard in parallel.
Tensors are broadcast to all shards if they are lexically captured by
`computation`. e.g.,
x = tf.constant(7)
def computation():
return x + 3
... = shard(computation, ...)
If `outputs_from_all_shards` is true, the outputs from all shards of
`computation` are concatenated back together along their `output_shard_axes`.
Otherwise, each output is taken from an arbitrary shard.
Inputs and outputs of the computation must be at least rank-1 Tensors.
Args:
computation: A Python function that builds a computation to apply to each
shard of the input.
inputs: A list of input tensors or None (equivalent to an empty list). Each
input tensor has a corresponding shard axes, given by `input_shard_axes`,
which must have size divisible by `num_shards`.
num_shards: The number of shards.
input_shard_axes: A list of dimensions along which to shard `inputs`, or
`None`. `None` means ""shard all inputs along dimension 0"". If not `None`,
there must be one dimension per input.
outputs_from_all_shards: Boolean or list of boolean. For each output, if
`True`, outputs from all shards are concatenated along the corresponding
`output_shard_axes` entry. Otherwise, each output is taken
from an arbitrary shard. If the argument is a boolean, the argument's
value is used for each output.
output_shard_axes: A list of dimensions along which to concatenate the
outputs of `computation`, or `None`. `None` means ""concatenate all outputs
along dimension 0"". If not `None`, there must be one dimension per output.
Ignored if `outputs_from_all_shards` is False.
infeed_queue: If not `None`, the `InfeedQueue` to use to augment the inputs
of `computation`.
device_assignment: If not `None`, a `DeviceAssignment` describing the
mapping between logical cores in the computation with physical cores in
the TPU topology. Uses a default device assignment if `None`. The
`DeviceAssignment` may be omitted if each shard of the computation uses
only one core, and there is either only one shard, or the number of shards
is equal to the number of cores in the TPU system.
name: (Deprecated) Does nothing.
xla_options: An instance of `tpu.XLAOptions` which indicates the options
passed to XLA compiler. Use `None` for default options.
Returns:
A tuple of (compile op, [output tensors]).
Raises:
ValueError: If num_shards <= 0
ValueError: If len(input_shard_axes) != len(inputs)
ValueError: If len(output_shard_axes) != len(outputs from `computation`)"
11050,shard,tensorflow/tensorflow/python/tpu/tpu.py,1764,function,"Shards `computation` for parallel execution.
`inputs` must be a list of Tensors or None (equivalent to an empty list), each
of which has a corresponding split axis (from `input_shard_axes`). Each input
is split into `num_shards` pieces along the corresponding axis, and
computation is applied to each shard in parallel.
Tensors are broadcast to all shards if they are lexically captured by
`computation`. e.g.,
x = tf.constant(7)
def computation():
return x + 3
... = shard(computation, ...)
TODO(phawkins): consider adding support for broadcasting Tensors passed
as inputs.
If `outputs_from_all_shards` is true, the outputs from all shards of
`computation` are concatenated back together along their `output_shard_axes`.
Otherwise, each output is taken from an arbitrary shard.
Inputs and outputs of the computation must be at least rank-1 Tensors.
Args:
computation: A Python function that builds a computation to apply to each
shard of the input.
inputs: A list of input tensors or None (equivalent to an empty list). Each
input tensor has a corresponding shard axes, given by `input_shard_axes`,
which must have size divisible by `num_shards`.
num_shards: The number of shards.
input_shard_axes: A list of dimensions along which to shard `inputs`, or
`None`. `None` means ""shard all inputs along dimension 0"". If not `None`,
there must be one dimension per input.
outputs_from_all_shards: Boolean or list of boolean. For each output, if
`True`, outputs from all shards are concatenated along the corresponding
`output_shard_axes` entry. Otherwise, each output is taken
from an arbitrary shard. If the argument is a boolean, the argument's
value is used for each output.
output_shard_axes: A list of dimensions along which to concatenate the
outputs of `computation`, or `None`. `None` means ""concatenate all outputs
along dimension 0"". If not `None`, there must be one dimension per output.
Ignored if `outputs_from_all_shards` is False.
infeed_queue: If not `None`, the `InfeedQueue` to use to augment the inputs
of `computation`.
device_assignment: If not `None`, a `DeviceAssignment` describing the
mapping between logical cores in the computation with physical cores in
the TPU topology. Uses a default device assignment if `None`. The
`DeviceAssignment` may be omitted if each shard of the computation uses
only one core, and there is either only one shard, or the number of shards
is equal to the number of cores in the TPU system.
name: (Deprecated) Does nothing.
xla_options: An instance of `tpu.XLAOptions` which indicates the options
passed to XLA compiler. Use `None` for default options.
Returns:
A list of output tensors.
Raises:
ValueError: If num_shards <= 0
ValueError: If len(input_shard_axes) != len(inputs)
ValueError: If len(output_shard_axes) != len(outputs from `computation`)"
11051,batch_parallel,tensorflow/tensorflow/python/tpu/tpu.py,1849,function,"Shards `computation` along the batch dimension for parallel execution.
Convenience wrapper around shard().
`inputs` must be a list of Tensors or None (equivalent to an empty list).
Each input is split into `num_shards` pieces along the 0-th dimension, and
computation is applied to each shard in parallel.
Tensors are broadcast to all shards if they are lexically captured by
`computation`. e.g.,
x = tf.constant(7)
def computation():
return x + 3
... = shard(computation, ...)
The outputs from all shards are concatenated back together along their 0-th
dimension.
Inputs and outputs of the computation must be at least rank-1 Tensors.
Args:
computation: A Python function that builds a computation to apply to each
shard of the input.
inputs: A list of input tensors or None (equivalent to an empty list). The
0-th dimension of each Tensor must have size divisible by `num_shards`.
num_shards: The number of shards.
infeed_queue: If not `None`, the `InfeedQueue` from which to append a tuple
of arguments as inputs to `computation`.
device_assignment: If not `None`, a `DeviceAssignment` describing the
mapping between logical cores in the computation with physical cores in
the TPU topology. Uses a default device assignment if `None`. The
`DeviceAssignment` may be omitted if each shard of the computation uses
only one core, and there is either only one shard, or the number of shards
is equal to the number of cores in the TPU system.
name: (Deprecated) Does nothing.
xla_options: An instance of `tpu.XLAOptions` which indicates the options
passed to XLA compiler. Use `None` for default options.
Returns:
A list of output tensors.
Raises:
ValueError: If `num_shards <= 0`"
11052,rewrite,tensorflow/tensorflow/python/tpu/tpu.py,1910,function,"Rewrites `computation` for execution on a TPU system.
Args:
computation: A Python function that builds a computation to apply to the
input. If the function takes n inputs, 'inputs' should be a list of n
tensors.
`computation` may return a list of operations and tensors. Tensors must
come before operations in the returned list. The return value of
`rewrite` is a list of tensors corresponding to the tensors from the
output of `computation`.
All `Operation`s constructed during `computation` will be executed when
evaluating any of the returned output tensors, not just the ones returned.
inputs: A list of input tensors or `None` (equivalent to an empty list).
Each input can be a nested structure containing values that are
convertible to tensors. Note that passing an N-dimension list of
compatible values will result in a N-dimension list of scalar tensors
rather than a single Rank-N tensors. If you need different behavior,
convert part of inputs to tensors with `tf.convert_to_tensor`.
infeed_queue: If not `None`, the `InfeedQueue` from which to append a tuple
of arguments as inputs to `computation`.
device_assignment: if not `None`, a `DeviceAssignment` describing the
mapping between logical cores in the computation with physical cores in
the TPU topology. May be omitted for a single-core computation, in which
case the core attached to task 0, TPU device 0 is used.
name: (Deprecated) Does nothing.
xla_options: An instance of `tpu.XLAOptions` which indicates the options
passed to XLA compiler. Use `None` for default options.
Returns:
Same data structure as if computation(*inputs) is called directly with some
exceptions for correctness. Exceptions include:
1) None output: a NoOp would be returned which control-depends on
computation.
2) Single value output: A tuple containing the value would be returned.
3) Operation-only outputs: a NoOp would be returned which
control-depends on computation.
TODO(b/121383831): Investigate into removing these special cases."
11053,under_tpu_inference_context,tensorflow/tensorflow/python/tpu/tpu.py,1980,function,Check if it is currently under `_TPUInferenceContext`.
11054,validate_inference_rewrite_for_variables,tensorflow/tensorflow/python/tpu/tpu.py,2037,function,"Validates whether rewrite_for_inference() 'worked' for variables.
The rewrite_for_inference() method is supposed to append GuaranteeConstOps
after ReadVariableOps, but this mechanism works only if you are using
tf.compat.v1.get_variable() to create and access variables in your tpu
computation. This validation method can be called immediately after calling
tpu.rewrite_for_inference() to check whether GuaranteeConstOps where added
to the graph.
Typical usages:
tpu.validate_inference_rewrite_for_variables(
tf.compat.v1.get_default_graph())
tpu.validate_inference_rewrite_for_variables(sess.graph)
Args:
graph: The graph which needs to be validated.
Raises:
RuntimeError: if validation failed."
11055,rewrite_for_inference,tensorflow/tensorflow/python/tpu/tpu.py,2066,function,"Rewrites `computation` for inference on a TPU system.
Other than 'rewriting' the computation to run on a TPU, if using variables
in your computation, it moves the ReadVariableOps outside the TPU
computation, and adds GuaranteeConst ops just after the ReadVariableOps.
This mechanism works only if you are using tf.compat.v1.get_variable() to
create and access variables in your tpu computation. You can validate
whether this worked, by calling validate_inference_rewrite_for_variables()
method immediately after this method to check whether GuaranteeConstOps
where added to the graph.
Args:
computation: A Python function that builds a computation to apply to the
input. If the function takes n inputs, 'inputs' should be a list of n
tensors. If the function returns m outputs, rewrite will return a list of
m tensors.
inputs: A list of input tensors or `None` (equivalent to an empty list).
infeed_queue: If not `None`, the `InfeedQueue` from which to append a tuple
of arguments as inputs to `computation`.
device_assignment: if not `None`, a `DeviceAssignment` describing the
mapping between logical cores in the computation with physical cores in
the TPU topology. May be omitted for a single-core computation, in which
case the core attached to task 0, TPU device 0 is used.
name: The name of the operator.
Returns:
A list of output tensors."
11056,prune_unconnected_ops_from_xla,tensorflow/tensorflow/python/tpu/tpu.py,2135,function,"Prunes unconnected ops as listed in _UNCONNECTED_OPS_TO_PRUNE.
Args:
prune_graph: A tensorflow graph from which we wish to prune unconnected ops
as listed in _UNCONNECTED_OPS_TO_PRUNE. In general, these ops should have
no inputs and no consumers. These can often be left behind due to graph
construction rewiring (for instance TF-Hub). While they never execute,
they will cause XLA compile to fail so we strip them from XLA compile by
removing the tpu_replicate attribute."
11057,TableConfig,tensorflow/tensorflow/python/tpu/tpu_embedding.py,51,class,Embedding table configuration.
11058,FeatureConfig,tensorflow/tensorflow/python/tpu/tpu_embedding.py,144,class,Feature configuration.
11059,EnqueueData,tensorflow/tensorflow/python/tpu/tpu_embedding.py,179,class,Data to be enqueued through generate_enqueue_ops().
11060,from_sparse_tensor,tensorflow/tensorflow/python/tpu/tpu_embedding.py,214,method,
11061,RaggedEnqueueData,tensorflow/tensorflow/python/tpu/tpu_embedding.py,221,class,RaggedTensor Data to be enqueued through generate_enqueue_ops().
11062,from_ragged_tensor,tensorflow/tensorflow/python/tpu/tpu_embedding.py,256,method,
11063,get_enqueue_datas_list_from_sparse_tensors_list,tensorflow/tensorflow/python/tpu/tpu_embedding.py,263,function,"Convenient function for generate_enqueue_ops().
Args:
sp_tensors_list: a list of dictionary mapping from string of feature names
to SparseTensor. Each dictionary is for one TPU core. Dictionaries for the
same host should be contiguous on the list.
Returns:
enqueue_datas_list: a list of dictionary mapping from string
of feature names to EnqueueData. Each dictionary is for one
TPU core. Dictionaries for the same host should be contiguous
on the list."
11064,get_enqueue_datas_list_from_ragged_tensors_list,tensorflow/tensorflow/python/tpu/tpu_embedding.py,287,function,"Convenient function for generate_enqueue_ops().
Args:
rg_tensors_list: a list of dictionary mapping from string of feature names
to RaggedTensor. Each dictionary is for one TPU core. Dictionaries for the
same host should be contiguous on the list.
Returns:
enqueue_datas_list: a list of dictionary mapping from string
of feature names to RaggedEnqueueData. Each dictionary is for one
TPU core. Dictionaries for the same host should be contiguous
on the list."
11065,AdagradParameters,tensorflow/tensorflow/python/tpu/tpu_embedding.py,364,class,"Optimization parameters for Adagrad with TPU embeddings.
Pass this to `tf.estimator.tpu.experimental.EmbeddingConfigSpec` via the
`optimization_parameters` argument to set the optimizer and its parameters.
See the documentation for `tf.estimator.tpu.experimental.EmbeddingConfigSpec`
for more details.
```
estimator = tf.estimator.tpu.TPUEstimator(
...
embedding_spec=tf.estimator.tpu.experimental.EmbeddingConfigSpec(
...
optimization_parameters=tf.tpu.experimental.AdagradParameters(0.1),
...))
```"
11066,ProximalAdagradParameters,tensorflow/tensorflow/python/tpu/tpu_embedding.py,416,class,"Optimization parameters for ProximalAdagrad with TPU embeddings.
Pass this to `tf.estimator.tpu.experimental.EmbeddingConfigSpec` via the
`optimization_parameters` argument to set the optimizer and its parameters.
See the documentation for `tf.estimator.tpu.experimental.EmbeddingConfigSpec`
for more details."
11067,AdamParameters,tensorflow/tensorflow/python/tpu/tpu_embedding.py,474,class,"Optimization parameters for Adam with TPU embeddings.
Pass this to `tf.estimator.tpu.experimental.EmbeddingConfigSpec` via the
`optimization_parameters` argument to set the optimizer and its parameters.
See the documentation for `tf.estimator.tpu.experimental.EmbeddingConfigSpec`
for more details.
```
estimator = tf.estimator.tpu.TPUEstimator(
...
embedding_config_spec=tf.estimator.tpu.experimental.EmbeddingConfigSpec(
...
optimization_parameters=tf.tpu.experimental.AdamParameters(0.1),
...))
```"
11068,FtrlParameters,tensorflow/tensorflow/python/tpu/tpu_embedding.py,551,class,"Optimization parameters for Ftrl with TPU embeddings.
Pass this to `tf.estimator.tpu.experimental.EmbeddingConfigSpec` via the
`optimization_parameters` argument to set the optimizer and its parameters.
See the documentation for `tf.estimator.tpu.experimental.EmbeddingConfigSpec`
for more details.
```
estimator = tf.estimator.tpu.TPUEstimator(
...
embedding_config_spec=tf.estimator.tpu.experimental.EmbeddingConfigSpec(
...
optimization_parameters=tf.tpu.experimental.FtrlParameters(0.1),
...))
```"
11069,ProximalYogiParameters,tensorflow/tensorflow/python/tpu/tpu_embedding.py,633,class,"Optimization parameters for Proximal Yogi with TPU embeddings.
Implements the Yogi optimizer as described in
[Adaptive Methods for Nonconvex Optimization](https://papers.nips.cc/paper/8186-adaptive-methods-for-nonconvex-optimization).
Pass this to `tf.estimator.tpu.experimental.EmbeddingConfigSpec` via the
`optimization_parameters` argument to set the optimizer and its parameters.
See the documentation for `tf.estimator.tpu.experimental.EmbeddingConfigSpec`
for more details."
11070,StochasticGradientDescentParameters,tensorflow/tensorflow/python/tpu/tpu_embedding.py,711,class,"Optimization parameters for stochastic gradient descent for TPU embeddings.
Pass this to `tf.estimator.tpu.experimental.EmbeddingConfigSpec` via the
`optimization_parameters` argument to set the optimizer and its parameters.
See the documentation for `tf.estimator.tpu.experimental.EmbeddingConfigSpec`
for more details.
```
estimator = tf.estimator.tpu.TPUEstimator(
...
embedding_config_spec=tf.estimator.tpu.experimental.EmbeddingConfigSpec(
...
optimization_parameters=(
tf.tpu.experimental.StochasticGradientDescentParameters(0.1))))
```"
11071,TPUEmbedding,tensorflow/tensorflow/python/tpu/tpu_embedding.py,757,class,"API for using TPU for embedding.
Example:
```
table_config_user = tpu_embedding.TableConfig(
vocabulary_size=4, dimension=2,
initializer=initializer, combiner='mean')
table_to_config_dict = {'video': table_config_video,
'user': table_config_user}
feature_to_config_dict = {'watched': tpu_embedding.FeatureConfig('video'),
'favorited': tpu_embedding.FeatureConfig('video'),
'friends': tpu_embedding.FeatureConfig('user')}
batch_size = 4
num_hosts = 1
optimization_parameters = tpu_embedding.AdagradParameters(1., 1.)
mode = tpu_embedding.TRAINING
embedding = tpu_embedding.TPUEmbedding(
table_to_config_dict, feature_to_config_dict,
batch_size, num_hosts, mode, optimization_parameters)
batch_size_per_core = embedding.batch_size_per_core
sparse_features_list = []
for host in hosts:
with ops.device(host):
for _ in range(embedding.num_cores_per_host):
sparse_features = {}
sparse_features['watched'] = sparse_tensor.SparseTensor(...)
sparse_features['favorited'] = sparse_tensor.SparseTensor(...)
sparse_features['friends'] = sparse_tensor.SparseTensor(...)
sparse_features_list.append(sparse_features)
enqueue_ops = embedding.generate_enqueue_ops(sparse_features_list)
embedding_variables_and_ops = embedding.create_variables_and_ops()
def computation():
activations = embedding.get_activations()
loss = compute_loss(activations)
base_optimizer = gradient_descent.GradientDescentOptimizer(
learning_rate=1)
cross_shard_optimizer = tpu_optimizer.CrossShardOptimizer(
base_optimizer)
train_op = cross_shard_optimizer.minimize(loss)
gradients = (
tpu_embedding_gradient.get_gradients_through_compute_gradients(
cross_shard_optimizer, loss, activations)
send_gradients_op = embedding.generate_send_gradients_op(gradients)
with ops.control_dependencies([train_op, send_gradients_op]):
loss = array_ops.identity(loss)
loss = tpu.shard(computation,
num_shards=embedding.num_cores)
with self.test_session() as sess:
sess.run(tpu.initialize_system(embedding_config=
embedding.config_proto))
sess.run(variables.global_variables_initializer())
sess.run(embedding_variables_and_ops.load_ops())
sess.run(enqueue_ops)
loss_val = sess.run(loss)
```
Example with weight decay:
>>> def learning_rate_fn(global_step):
... return tf.compat.v1.train.polynomial_decay(
... learning_rate=5e-5,
... global_step=global_step,
... decay_steps=100000,
... end_learning_rate=0.0)
>>> wordpiece_table_config = TableConfig(
... vocabulary_size=119547,
... dimension=256,
... learning_rate_fn=learning_rate_fn)
>>> wordpiece_feature_config = FeatureConfig(
... table_id='bert/embeddings/word_embeddings',
... max_sequence_length=512)
>>> optimization_parameters = AdamParameters(
... learning_rate=5e-5,
... epsilon=1e-6,
... weight_decay_factor=0.01,
... multiply_weight_decay_factor_by_learning_rate=True)
>>> tpu_embedding = TPUEmbedding(
... table_to_config_dict={
... 'bert/embeddings/word_embeddings': wordpiece_table_config,
... },
... feature_to_config_dict={'input_ids': wordpiece_feature_config},
... batch_size=128,
... mode=TRAINING,
... optimization_parameters=optimization_parameters,
... master='')
>>> with tf.Graph().as_default():
... init_tpu_op = tf.compat.v1.tpu.initialize_system(
... embedding_config=tpu_embedding.config_proto)
... tf.compat.v1.Session().run(init_tpu_op)"
11072,hosts,tensorflow/tensorflow/python/tpu/tpu_embedding.py,1011,method,"A list of device names for CPU hosts.
Returns:
A list of device names for CPU hosts."
11073,num_cores_per_host,tensorflow/tensorflow/python/tpu/tpu_embedding.py,1022,method,"Number of TPU cores on a CPU host.
Returns:
Number of TPU cores on a CPU host."
11074,num_cores,tensorflow/tensorflow/python/tpu/tpu_embedding.py,1031,method,"Total number of TPU cores on all hosts.
Returns:
Total number of TPU cores on all hosts."
11075,batch_size_per_core,tensorflow/tensorflow/python/tpu/tpu_embedding.py,1040,method,"Batch size for each TPU core.
The sparse tensors in `sparse_features_list` to `generate_enqueue_ops`
must have batch dimension equal to this.
Returns:
Batch size for each TPU core."
11076,config_proto,tensorflow/tensorflow/python/tpu/tpu_embedding.py,1052,method,"Create embedding config proto for `tpu.initialize_system()`.
Returns:
an `TPUEmbeddingConfiguration` proto describing the desired
configuration of the hardware embedding lookup tables, which
is passed to `tpu.initialize_system()`."
11077,table_to_config_dict,tensorflow/tensorflow/python/tpu/tpu_embedding.py,1063,method,
11078,feature_to_config_dict,tensorflow/tensorflow/python/tpu/tpu_embedding.py,1067,method,
11079,table_to_features_dict,tensorflow/tensorflow/python/tpu/tpu_embedding.py,1071,method,
11080,optimization_parameters,tensorflow/tensorflow/python/tpu/tpu_embedding.py,1075,method,
11081,create_variables_and_ops,tensorflow/tensorflow/python/tpu/tpu_embedding.py,1141,method,"Create embedding and slot variables, with ops to load and retrieve them.
N.B.: the retrieve embedding variables (including slot variables) ops are
returned as lambda fn, as the call side might want to impose control
dependencies between the TPU computation and retrieving actions. For
example, the following code snippet ensures the TPU computation finishes
first, and then we pull the variables back from TPU to CPU.
```
updates_ops = []
with ops.control_dependencies([loss]):
for op_fn in retrieve_parameters_op_fns:
update_ops.append(op_fn())
```
Args:
embedding_variable_name_by_table: A dictionary mapping from string of
table name to string of embedding variable name. If `None`,
defaults from `get_default_slot_variable_names()` will be used.
slot_variable_names_by_table: A dictionary mapping from string of table
name to `AdamSlotVariableNames`, `AdagradSlotVariableNames` etc. If
`None`, defaults from `get_default_slot_variable_names()` will be used.
Returns:
`tpu_embedding.VariablesAndOps` with:
A dictionary mapping from string of table name to embedding variables,
A dictionary mapping from string of table name to AdagradSlotVariable,
AdamSlotVariables etc with slot variables,
A function which returns a list of ops to load embedding and slot
variables from CPU to TPU.
A function which returns a list of ops to retrieve embedding and slot
variables from TPU to CPU."
11082,generate_enqueue_ops,tensorflow/tensorflow/python/tpu/tpu_embedding.py,1248,method,"Generate enqueue ops.
Args:
enqueue_datas_list: a list of dictionary mapping from string
of feature names to EnqueueData. Each dictionary is for one
TPU core. Dictionaries for the same host should be contiguous
on the list.
mode_override: A string input that overrides the mode specified in the
TPUEmbeddingConfiguration. Supported values are {'unspecified',
'inference', 'training', 'backward_pass_only'}. When set to
'unspecified', the mode set in TPUEmbeddingConfiguration is used,
otherwise mode_override is used (optional).
ragged: If True, creates RaggedTensor enqueue ops rather than
SparseTensor.
Returns:
Ops to enqueue to TPU for embedding."
11083,get_activations,tensorflow/tensorflow/python/tpu/tpu_embedding.py,1469,method,"Get activations for features.
This should be called within `computation` that is passed to
`tpu.replicate` and friends.
Returns:
A dictionary mapping from `String` of feature name to `Tensor`
of activation."
11084,generate_send_gradients_op,tensorflow/tensorflow/python/tpu/tpu_embedding.py,1503,method,"Send gradient to TPU embedding.
Args:
feature_to_gradient_dict: dict mapping feature names to gradient wrt
activations.
step: the current global step, used for dynamic learning rate.
Returns:
SendTPUEmbeddingGradients Op.
Raises:
RuntimeError: If `mode` is not `TRAINING`."
11085,load_ops,tensorflow/tensorflow/python/tpu/tpu_embedding.py,1222,method,"Calls and returns the load ops for each embedding table.
Returns:
A list of ops to load embedding and slot variables from CPU to TPU."
11086,retrieve_ops,tensorflow/tensorflow/python/tpu/tpu_embedding.py,1233,method,"Calls and returns the retrieve ops for each embedding table.
Returns:
A list of ops to retrieve embedding and slot variables from TPU to CPU."
11087,get_gradients_through_compute_gradients,tensorflow/tensorflow/python/tpu/tpu_embedding_gradient.py,32,function,"Compute gradients to send to TPU embedding.
Args:
optimizer: a subclass of optimizer.Optimizer, usually CrossShardOptimizer.
Used to call compute_gradients().
loss: a Tensor to call optimizer.compute_gradients() on.
activations: an OrderedDict mapping feature_name to Tensors of activations.
Returns:
An OrderedDict mapping from feature name Strings to Tensors of gradients of
the loss wrt the activations of the features."
11088,create_dummy_table_variables,tensorflow/tensorflow/python/tpu/tpu_embedding_gradient.py,53,function,"Create dummy embedding table variables.
The sole purpose of these dummy variables are to trigger gradient
calculation wrt them so that the gradients wrt activation can be captured
and later sent to TPU embedding.
Args:
tpu_embedding: TPUEmbedding, dummy table variables will be created for use
with tpu_embedding.
Returns:
A tuple of dummy variables and their initializer.
Raises:
RuntimeError: if collection to store gradients already exists and is not
empty."
11089,hook_dummy_table_variables_to_activations,tensorflow/tensorflow/python/tpu/tpu_embedding_gradient.py,103,function,"Have activations depend on dummy table variables for gradient intercept.
Args:
tpu_embedding: TPUEmbedding, activations and dummy_table_variables are from
tpu_embedding.
activations: An OrderedDict of feature name String to activation tensors.
dummy_table_variables: An OrderedDict of table name String to dummy table
variables.
Returns:
An OrderedDict of feature name String to activation tensors, which can be
used just as the activations input."
11090,get_gradients_through_dummy_table_variables,tensorflow/tensorflow/python/tpu/tpu_embedding_gradient.py,129,function,"Get gradients wrt the activations of each feature.
Args:
tpu_embedding: TPUEmbedding, create dummy table variable to be used with
tpu_embedding.
Returns:
An OrderedDict mapping feature name to gradient.
Raises:
ValueError: if some gradients are not defined."
11091,TPUShardedVariable,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,62,class,A ShardedVariable class for TPU.
11092,TPUEmbedding,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,87,class,"The TPUEmbedding mid level API.
NOTE: When instantiated under a TPUStrategy, this class can only be created
once per call to `tf.tpu.experimental.initialize_tpu_system`. If you wish to
re-initialize the embedding engine you must re-initialize the tpu as well.
Doing this will clear any variables from TPU, so ensure you have checkpointed
before you do this. If a further instances of the class are needed,
set the `initialize_tpu_embedding` argument to `False`.
This class can be used to support training large embeddings on TPU. When
creating an instance of this class, you must specify the complete set of
tables and features you expect to lookup in those tables. See the
documentation of `tf.tpu.experimental.embedding.TableConfig` and
`tf.tpu.experimental.embedding.FeatureConfig` for more details on the complete
set of options. We will cover the basic usage here.
NOTE: multiple `FeatureConfig` objects can use the same `TableConfig` object,
allowing different features to share the same table:
```python
table_config_one = tf.tpu.experimental.embedding.TableConfig(
vocabulary_size=...,
dim=...)
table_config_two = tf.tpu.experimental.embedding.TableConfig(
vocabulary_size=...,
dim=...)
feature_config = {
'feature_one': tf.tpu.experimental.embedding.FeatureConfig(
table=table_config_one),
'feature_two': tf.tpu.experimental.embedding.FeatureConfig(
table=table_config_one),
'feature_three': tf.tpu.experimental.embedding.FeatureConfig(
table=table_config_two)}
```
There are two modes under which the `TPUEmbedding` class can used. This
depends on if the class was created under a `TPUStrategy` scope or not.
Under `TPUStrategy`, we allow access to the method `enqueue`, `dequeue` and
`apply_gradients`. We will show examples below of how to use these to train
and evaluate your model. Under CPU, we only access to the `embedding_tables`
property which allow access to the embedding tables so that you can use them
to run model evaluation/prediction on CPU.
First lets look at the `TPUStrategy` mode. Initial setup looks like:
```python
strategy = tf.distribute.TPUStrategy(...)
with strategy.scope():
embedding = tf.tpu.experimental.embedding.TPUEmbedding(
feature_config=feature_config,
batch_size=1024,
optimizer=tf.tpu.experimental.embedding.SGD(0.1))
```
When creating a distributed dataset that is to be passed to the enqueue
operation a special input option must be specified:
```python
distributed_dataset = (
strategy.experimental_distribute_datasets_from_function(
dataset_fn=...,
options=tf.distribute.InputOptions(
experimental_prefetch_to_device=False))
dataset_iterator = iter(distributed_dataset)
```
To use this API on TPU you should use a custom training loop. Below is an
example of a training and evaluation step:
```python
@tf.function
def training_step(dataset_iterator, num_steps):
def tpu_step(tpu_features):
with tf.GradientTape() as tape:
activations = embedding.dequeue()
tape.watch(activations)
model_output = model(activations)
loss = ... # some function of labels and model_output
embedding_gradients = tape.gradient(loss, activations)
embedding.apply_gradients(embedding_gradients)
# Insert your model gradient and optimizer application here
for _ in tf.range(num_steps):
embedding_features, tpu_features = next(dataset_iterator)
embedding.enqueue(embedding_features, training=True)
strategy.run(tpu_step, args=(embedding_features, ))
@tf.function
def evalution_step(dataset_iterator, num_steps):
def tpu_step(tpu_features):
activations = embedding.dequeue()
model_output = model(activations)
# Insert your evaluation code here.
for _ in tf.range(num_steps):
embedding_features, tpu_features = next(dataset_iterator)
embedding.enqueue(embedding_features, training=False)
strategy.run(tpu_step, args=(embedding_features, ))
```
NOTE: The calls to `enqueue` have `training` set to `True` when
`embedding.apply_gradients` is used and set to `False` when
`embedding.apply_gradients` is not present in the function. If you don't
follow this pattern you may cause an error to be raised or the tpu may
deadlock.
In the above examples, we assume that the user has a dataset which returns
a tuple where the first element of the tuple matches the structure of what
was passed as the `feature_config` argument to the object initializer. Also we
utilize `tf.range` to get a `tf.while_loop` in order to increase performance.
When checkpointing your model, you should include your
`tf.tpu.experimental.embedding.TPUEmbedding` object in the checkpoint. It is a
trackable object and saving it will save the embedding tables and their
optimizer slot variables:
```python
checkpoint = tf.train.Checkpoint(model=model, embedding=embedding)
checkpoint.save(...)
```
On CPU, only the `embedding_table` property is usable. This will allow you to
restore a checkpoint to the object and have access to the table variables:
```python
model = model_fn(...)
embedding = tf.tpu.experimental.embedding.TPUEmbedding(
feature_config=feature_config,
batch_size=1024,
optimizer=tf.tpu.experimental.embedding.SGD(0.1))
checkpoint = tf.train.Checkpoint(model=model, embedding=embedding)
checkpoint.restore(...)
tables = embedding.embedding_tables
```
You can now use table in functions like `tf.nn.embedding_lookup` to perform
your embedding lookup and pass to your model."
11093,embedding_tables,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,353,method,"Returns a dict of embedding tables, keyed by `TableConfig`.
This property only works when the `TPUEmbedding` object is created under a
non-TPU strategy. This is intended to be used to for CPU based lookup when
creating a serving checkpoint.
Returns:
A dict of embedding tables, keyed by `TableConfig`.
Raises:
RuntimeError: If object was created under a `TPUStrategy`."
11094,apply_gradients,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,506,method,"Applies the gradient update to the embedding tables.
If a gradient of `None` is passed in any position of the nested structure,
then an gradient update with a zero gradient is applied for that feature.
For optimizers like SGD or Adagrad, this is the same as applying no update
at all. For lazy Adam and other sparsely applied optimizers with decay,
ensure you understand the effect of applying a zero gradient.
```python
strategy = tf.distribute.TPUStrategy(...)
with strategy.scope():
embedding = tf.tpu.experimental.embedding.TPUEmbedding(...)
distributed_dataset = (
strategy.experimental_distribute_datasets_from_function(
dataset_fn=...,
options=tf.distribute.InputOptions(
experimental_prefetch_to_device=False))
dataset_iterator = iter(distributed_dataset)
@tf.function
def training_step():
def tpu_step(tpu_features):
with tf.GradientTape() as tape:
activations = embedding.dequeue()
tape.watch(activations)
loss = ... # some computation involving activations
embedding_gradients = tape.gradient(loss, activations)
embedding.apply_gradients(embedding_gradients)
embedding_features, tpu_features = next(dataset_iterator)
embedding.enqueue(embedding_features, training=True)
strategy.run(tpu_step, args=(embedding_features, ))
training_step()
```
Args:
gradients: A nested structure of gradients, with structure matching the
`feature_config` passed to this object.
name: A name for the underlying op.
Raises:
RuntimeError: If called when object wasn't created under a `TPUStrategy`.
ValueError: If a non-`tf.Tensor` non-`None` gradient is passed in, or a
`tf.Tensor` of the incorrect shape is passed in. Also if
the size of any sequence in `gradients` does not match corresponding
sequence in `feature_config`.
TypeError: If the type of any sequence in `gradients` does not match
corresponding sequence in `feature_config`."
11095,dequeue,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,595,method,"Get the embedding results.
Returns a nested structure of `tf.Tensor` objects, matching the structure of
the `feature_config` argument to the `TPUEmbedding` class. The output shape
of the tensors is `(batch_size, dim)`, where `batch_size` is the per core
batch size, `dim` is the dimension of the corresponding `TableConfig`. If
the feature's corresponding `FeatureConfig` has `max_sequence_length`
greater than 0, the output will be a sequence of shape
`(batch_size, max_sequence_length, dim)` instead.
```python
strategy = tf.distribute.TPUStrategy(...)
with strategy.scope():
embedding = tf.tpu.experimental.embedding.TPUEmbedding(...)
distributed_dataset = (
strategy.experimental_distribute_datasets_from_function(
dataset_fn=...,
options=tf.distribute.InputOptions(
experimental_prefetch_to_device=False))
dataset_iterator = iter(distributed_dataset)
@tf.function
def training_step():
def tpu_step(tpu_features):
with tf.GradientTape() as tape:
activations = embedding.dequeue()
tape.watch(activations)
loss = ... # some computation involving activations
embedding_gradients = tape.gradient(loss, activations)
embedding.apply_gradients(embedding_gradients)
embedding_features, tpu_features = next(dataset_iterator)
embedding.enqueue(embedding_features, training=True)
strategy.run(tpu_step, args=(embedding_features, ))
training_step()
```
Args:
name: A name for the underlying op.
Returns:
A nested structure of tensors, with the same structure as `feature_config`
passed to this instance of the `TPUEmbedding` object.
Raises:
RuntimeError: If called when object wasn't created under a `TPUStrategy`."
11096,enqueue,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,1051,method,"Enqueues id tensors for embedding lookup.
This function enqueues a structure of features to be looked up in the
embedding tables. We expect that the batch size of each of the tensors in
features matches the per core batch size. This will automatically happen if
your input dataset is batched to the global batch size and you use
`tf.distribute.TPUStrategy`'s `experimental_distribute_dataset`
or if you use `experimental_distribute_datasets_from_function` and batch
to the per core batch size computed by the context passed to your input
function.
```python
strategy = tf.distribute.TPUStrategy(...)
with strategy.scope():
embedding = tf.tpu.experimental.embedding.TPUEmbedding(...)
distributed_dataset = (
strategy.experimental_distribute_datasets_from_function(
dataset_fn=...,
options=tf.distribute.InputOptions(
experimental_prefetch_to_device=False))
dataset_iterator = iter(distributed_dataset)
@tf.function
def training_step():
def tpu_step(tpu_features):
with tf.GradientTape() as tape:
activations = embedding.dequeue()
tape.watch(activations)
loss = ... # some computation involving activations
embedding_gradients = tape.gradient(loss, activations)
embedding.apply_gradients(embedding_gradients)
embedding_features, tpu_features = next(dataset_iterator)
embedding.enqueue(embedding_features, training=True)
strategy.run(tpu_step, args=(embedding_features,))
training_step()
```
NOTE: You should specify `training=True` when using
`embedding.apply_gradients` as above and `training=False` when not using
`embedding.apply_gradients` (e.g. for frozen embeddings or when doing
evaluation).
Args:
features: A nested structure of `tf.Tensor`s, `tf.SparseTensor`s or
`tf.RaggedTensor`s, with the same structure as `feature_config`. Inputs
will be downcast to `tf.int32`. Only one type out of `tf.SparseTensor`
or `tf.RaggedTensor` is supported per call.
weights: If not `None`, a nested structure of `tf.Tensor`s,
`tf.SparseTensor`s or `tf.RaggedTensor`s, matching the above, except
that the tensors should be of float type (and they will be downcast to
`tf.float32`). For `tf.SparseTensor`s we assume the `indices` are the
same for the parallel entries from `features` and similarly for
`tf.RaggedTensor`s we assume the row_splits are the same.
training: Defaults to `True`. If `False`, enqueue the batch as inference
batch (forward pass only). Do not call `apply_gradients` when this is
`False` as this may lead to a deadlock.
name: A name for the underlying op.
Raises:
ValueError: When called inside a strategy.run call and input is not
directly taken from the args of the `strategy.run` call. Also if
the size of any sequence in `features` does not match corresponding
sequence in `feature_config`. Similarly for `weights`, if not `None`.
RuntimeError: When called inside a strategy.run call and inside XLA
control flow.
TypeError: If the type of any sequence in `features` does not match
corresponding sequence in `feature_config`. Similarly for `weights`, if
not `None`."
11097,create_variables,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,713,method,Create all variables.
11098,select_fn,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,764,method,
11099,factory,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,823,method,
11100,check_device,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,1028,method,
11101,getter,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,717,method,
11102,variable_creator,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,723,method,
11103,slot_creator,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,738,method,
11104,generate_enqueue_ops,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,1148,method,Generate enqueue ops for outside compilation.
11105,load_config,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,338,method,
11106,TPUEmbeddingSaveable,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,1203,class,Save/Restore hook to Retrieve/Load TPUEmbedding variables.
11107,before_save,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,1211,method,
11108,after_restore,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,1215,method,
11109,cpu_embedding_lookup,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,1245,function,"Uses CPU embedding lookup for embedding ids in features.
Args:
inputs: a nested structure of Tensors, SparseTensors or RaggedTensors.
weights: a nested structure of Tensors, SparseTensors or RaggedTensors or
None for no weights.
tables: a dict of mapping TableConfig objects to Variables.
feature_config: a nested structure of FeatureConfig objects with the same
structure as inputs.
Returns:
A nested structure of Tensors with the same structure as inputs."
11110,get_list_of_hosts,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,1302,function,"Returns a sorted list of CPU devices for the remote jobs.
Args:
strategy: A TPUStrategy object.
Returns:
A sort list of device strings."
11111,extract_variable_info,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,1321,function,"Extracts the variable creation attributes from the kwargs.
Args:
kwargs: a dict of keyword arguments that were passed to a variable creator
scope.
Returns:
A tuple of variable name, initialization function, shape, and dtype."
11112,make_sharded_variable_creator,tensorflow/tensorflow/python/tpu/tpu_embedding_v2.py,1355,function,"Makes a sharded variable creator given a list of hosts.
Args:
hosts: a list of tensorflow devices on which to shard the tensors.
Returns:
A variable creator function."
11113,TPUEmbeddingCorrectness,tensorflow/tensorflow/python/tpu/tpu_embedding_v2_correctness_test.py,58,class,
11114,setUp,tensorflow/tensorflow/python/tpu/tpu_embedding_v2_correctness_test.py,60,method,
11115,tearDown,tensorflow/tensorflow/python/tpu/tpu_embedding_v2_correctness_test.py,127,method,
11116,input_fn,tensorflow/tensorflow/python/tpu/tpu_embedding_v2_correctness_test.py,255,method,
11117,select_replica,tensorflow/tensorflow/python/tpu/tpu_embedding_v2_correctness_test.py,423,method,
11118,embedding_and_set_gradients,tensorflow/tensorflow/python/tpu/tpu_embedding_v2_correctness_test.py,487,method,
11119,embedding_only,tensorflow/tensorflow/python/tpu/tpu_embedding_v2_correctness_test.py,497,method,
11120,step,tensorflow/tensorflow/python/tpu/tpu_embedding_v2_correctness_test.py,173,method,Create and run computation that returns the embedding activations.
11121,step,tensorflow/tensorflow/python/tpu/tpu_embedding_v2_correctness_test.py,442,method,
11122,tpu_fn,tensorflow/tensorflow/python/tpu/tpu_embedding_v2_correctness_test.py,488,method,
11123,tpu_fn,tensorflow/tensorflow/python/tpu/tpu_embedding_v2_correctness_test.py,498,method,
11124,SGD,tensorflow/tensorflow/python/tpu/tpu_embedding_v2_utils.py,125,class,"Optimization parameters for stochastic gradient descent for TPU embeddings.
Pass this to `tf.tpu.experimental.embedding.TPUEmbedding` via the `optimizer`
argument to set the global optimizer and its parameters:
```
embedding = tf.tpu.experimental.embedding.TPUEmbedding(
...
optimizer=tf.tpu.experimental.embedding.SGD(0.1))
```
This can also be used in a `tf.tpu.experimental.embedding.TableConfig` as the
optimizer parameter to set a table specific optimizer. This will override the
optimizer and parameters for global embedding optimizer defined above:
```
table_one = tf.tpu.experimental.embedding.TableConfig(
vocabulary_size=...,
dim=...,
optimizer=tf.tpu.experimental.embedding.SGD(0.2))
table_two = tf.tpu.experimental.embedding.TableConfig(
vocabulary_size=...,
dim=...)
feature_config = (
tf.tpu.experimental.embedding.FeatureConfig(
table=table_one),
tf.tpu.experimental.embedding.FeatureConfig(
table=table_two))
embedding = tf.tpu.experimental.embedding.TPUEmbedding(
feature_config=feature_config,
batch_size=...
optimizer=tf.tpu.experimental.embedding.SGD(0.1))
```
In the above example, the first feature will be looked up in a table that has
a learning rate of 0.2 while the second feature will be looked up in a table
that has a learning rate of 0.1.
See 'tensorflow/core/protobuf/tpu/optimization_parameters.proto' for a
complete description of these parameters and their impacts on the optimizer
algorithm."
11125,Adagrad,tensorflow/tensorflow/python/tpu/tpu_embedding_v2_utils.py,212,class,"Optimization parameters for Adagrad with TPU embeddings.
Pass this to `tf.tpu.experimental.embedding.TPUEmbedding` via the `optimizer`
argument to set the global optimizer and its parameters:
```python
embedding = tf.tpu.experimental.embedding.TPUEmbedding(
...
optimizer=tf.tpu.experimental.embedding.Adagrad(0.1))
```
This can also be used in a `tf.tpu.experimental.embedding.TableConfig` as the
optimizer parameter to set a table specific optimizer. This will override the
optimizer and parameters for global embedding optimizer defined above:
```python
table_one = tf.tpu.experimental.embedding.TableConfig(
vocabulary_size=...,
dim=...,
optimizer=tf.tpu.experimental.embedding.Adagrad(0.2))
table_two = tf.tpu.experimental.embedding.TableConfig(
vocabulary_size=...,
dim=...)
feature_config = (
tf.tpu.experimental.embedding.FeatureConfig(
table=table_one),
tf.tpu.experimental.embedding.FeatureConfig(
table=table_two))
embedding = tf.tpu.experimental.embedding.TPUEmbedding(
feature_config=feature_config,
batch_size=...
optimizer=tf.tpu.experimental.embedding.Adagrad(0.1))
```
In the above example, the first feature will be looked up in a table that has
a learning rate of 0.2 while the second feature will be looked up in a table
that has a learning rate of 0.1.
See 'tensorflow/core/protobuf/tpu/optimization_parameters.proto' for a
complete description of these parameters and their impacts on the optimizer
algorithm."
11126,Adam,tensorflow/tensorflow/python/tpu/tpu_embedding_v2_utils.py,315,class,"Optimization parameters for Adam with TPU embeddings.
Pass this to `tf.tpu.experimental.embedding.TPUEmbedding` via the `optimizer`
argument to set the global optimizer and its parameters:
NOTE: By default this optimizer is lazy, i.e. it will not apply the gradient
update of zero to rows that were not looked up. You can change this behavior
by setting `lazy_adam` to `False`.
```python
embedding = tf.tpu.experimental.embedding.TPUEmbedding(
...
optimizer=tf.tpu.experimental.embedding.Adam(0.1))
```
This can also be used in a `tf.tpu.experimental.embedding.TableConfig` as the
optimizer parameter to set a table specific optimizer. This will override the
optimizer and parameters for global embedding optimizer defined above:
```python
table_one = tf.tpu.experimental.embedding.TableConfig(
vocabulary_size=...,
dim=...,
optimizer=tf.tpu.experimental.embedding.Adam(0.2))
table_two = tf.tpu.experimental.embedding.TableConfig(
vocabulary_size=...,
dim=...)
feature_config = (
tf.tpu.experimental.embedding.FeatureConfig(
table=table_one),
tf.tpu.experimental.embedding.FeatureConfig(
table=table_two))
embedding = tf.tpu.experimental.embedding.TPUEmbedding(
feature_config=feature_config,
batch_size=...
optimizer=tf.tpu.experimental.embedding.Adam(0.1))
```
In the above example, the first feature will be looked up in a table that has
a learning rate of 0.2 while the second feature will be looked up in a table
that has a learning rate of 0.1.
See 'tensorflow/core/protobuf/tpu/optimization_parameters.proto' for a
complete description of these parameters and their impacts on the optimizer
algorithm."
11127,TableConfig,tensorflow/tensorflow/python/tpu/tpu_embedding_v2_utils.py,455,class,"Configuration data for one embedding table.
This class holds the configuration data for a single embedding table. It is
used as the `table` parameter of a
`tf.tpu.experimental.embedding.FeatureConfig`. Multiple
`tf.tpu.experimental.embedding.FeatureConfig` objects can use the same
`tf.tpu.experimental.embedding.TableConfig` object. In this case a shared
table will be created for those feature lookups.
```python
table_config_one = tf.tpu.experimental.embedding.TableConfig(
vocabulary_size=...,
dim=...)
table_config_two = tf.tpu.experimental.embedding.TableConfig(
vocabulary_size=...,
dim=...)
feature_config = {
'feature_one': tf.tpu.experimental.embedding.FeatureConfig(
table=table_config_one),
'feature_two': tf.tpu.experimental.embedding.FeatureConfig(
table=table_config_one),
'feature_three': tf.tpu.experimental.embedding.FeatureConfig(
table=table_config_two)}
embedding = tf.tpu.experimental.embedding.TPUEmbedding(
feature_config=feature_config,
batch_size=...
optimizer=tf.tpu.experimental.embedding.Adam(0.1))
```
The above configuration has 2 tables, and three features. The first two
features will be looked up in the first table and the third feature will be
looked up in the second table."
11128,FeatureConfig,tensorflow/tensorflow/python/tpu/tpu_embedding_v2_utils.py,548,class,"Configuration data for one embedding feature.
This class holds the configuration data for a single embedding feature. The
main use is to assign features to `tf.tpu.experimental.embedding.TableConfig`s
via the table parameter:
```python
table_config_one = tf.tpu.experimental.embedding.TableConfig(
vocabulary_size=...,
dim=...)
table_config_two = tf.tpu.experimental.embedding.TableConfig(
vocabulary_size=...,
dim=...)
feature_config = {
'feature_one': tf.tpu.experimental.embedding.FeatureConfig(
table=table_config_one),
'feature_two': tf.tpu.experimental.embedding.FeatureConfig(
table=table_config_one),
'feature_three': tf.tpu.experimental.embedding.FeatureConfig(
table=table_config_two)}
embedding = tf.tpu.experimental.embedding.TPUEmbedding(
feature_config=feature_config,
batch_size=...
optimizer=tf.tpu.experimental.embedding.Adam(0.1))
```
The above configuration has 2 tables, and three features. The first two
features will be looked up in the first table and the third feature will be
looked up in the second table.
When feeding features into `embedding.enqueue` they can be `tf.Tensor`s,
`tf.SparseTensor`s or `tf.RaggedTensor`s. When the argument
`max_sequence_length` is 0, the default, you should expect a output of
`embedding.dequeue` for this feature of shape `(batch_size, dim)`. If
`max_sequence_length` is greater than 0, the feature is embedded as a sequence
and padded up to the given length. The shape of the output for this feature
will be `(batch_size, max_sequence_length, dim)`."
11129,partition_or_replicate_on_host,tensorflow/tensorflow/python/tpu/tpu_feed.py,40,function,"Partitions or replicates the input tensor.
The ops inside this function are placed on the host side.
Args:
tensor: The input tensor which will be partitioned or replicated.
dims: A list of integer describes how to partition the input tensor.
Returns:
An iterator of `Tensor`s or a list of partitioned tensors."
11130,tag_sharding_attribute_for_dequeued_tensors,tensorflow/tensorflow/python/tpu/tpu_feed.py,110,function,"Tags appropriate XLA sharding attribute to the dequeued tensors.
Args:
dequeues: A list of dequeued tensors on TPU.
dims: A list of integer describes how the tensor is partitioned.
Returns:
The same dequeues with appropriate xla_sharding attribute."
11131,InfeedQueue,tensorflow/tensorflow/python/tpu/tpu_feed.py,125,class,"A helper object to build a device infeed queue.
The InfeedQueue builds the host-side and device-side Ops to enqueue and
dequeue elements, respectively, and ensures that their types and
shapes match."
11132,number_of_tuple_elements,tensorflow/tensorflow/python/tpu/tpu_feed.py,212,method,Returns the number of InfeedQueue tuple elements.
11133,tuple_types,tensorflow/tensorflow/python/tpu/tpu_feed.py,217,method,Returns the types of the InfeedQueue tuple elements.
11134,set_tuple_types,tensorflow/tensorflow/python/tpu/tpu_feed.py,221,method,"Sets the type of each element of the queue.
tuple_types must be a list of length
self.number_of_tuple_elements, and each element must be
convertible to a dtype.
Args:
tuple_types: the types of each queue element.
Raises:
ValueError: if tuple_types is not of length
self.number_of_tuple_elements.
TypeError: if an element of tuple_types cannot be converted to a
dtype."
11135,tuple_shapes,tensorflow/tensorflow/python/tpu/tpu_feed.py,256,method,Returns the shapes of the InfeedQueue tuple elements.
11136,set_tuple_shapes,tensorflow/tensorflow/python/tpu/tpu_feed.py,260,method,"Sets the shape of each element of the queue.
tuple_shapes must be a list of length
self.number_of_tuple_elements, and each element must be
convertible to a TensorShape.
Args:
tuple_shapes: the shapes of each queue element.
Raises:
ValueError: if tuple_shapes is not of length
self.number_of_tuple_elements.
TypeError: if an element of tuple_shapes cannot be converted to
a TensorShape."
11137,sharding_policies,tensorflow/tensorflow/python/tpu/tpu_feed.py,298,method,Returns the sharding policies of the InfeedQueue tuple elements.
11138,shard_dimensions,tensorflow/tensorflow/python/tpu/tpu_feed.py,303,method,"Gets the shard dimension of each tuple element.
Returns:
A list of length number_of_tuple_elements, where each list entry
is the shard dimension of that tuple element or None if the
shard dimension has not been set."
11139,set_shard_dimensions,tensorflow/tensorflow/python/tpu/tpu_feed.py,314,method,"Sets the shard_dimension of each element of the queue.
shard_dimensions must be a list of length
self.number_of_tuple_elements, and each element must be
convertible to a Dimension compatible with self.tuple_shapes.
Args:
shard_dimensions: the dimensions of each queue element.
Raises:
ValueError: if shard_dimensions is not of length
self.number_of_tuple_elements; or an element of
shard_dimensions cannot be converted to a Dimension; or an
element of shard_dimensions is a Dimension that is out of
range for the corresponding tuple element shape."
11140,number_of_shards,tensorflow/tensorflow/python/tpu/tpu_feed.py,340,method,"Gets the number of shards to use for the InfeedQueue.
Returns:
Number of shards or None if the number of shards has not been set."
11141,set_number_of_shards,tensorflow/tensorflow/python/tpu/tpu_feed.py,349,method,"Sets the number of shards to use for the InfeedQueue.
Args:
number_of_shards: number of ways to shard the InfeedQueue.
Raises:
ValueError: if number_of_shards is not > 0; or the policies have
been frozen and number_of_shards was already set to something
else."
11142,set_configuration_from_input_tensors,tensorflow/tensorflow/python/tpu/tpu_feed.py,364,method,"Sets the shapes and types of the queue tuple elements.
input_tensors is a list of Tensors whose types and shapes are used
to set the queue configuration.
Args:
input_tensors: list of Tensors of the same types and shapes as
the desired queue Tuple.
Raises:
ValueError: if input_tensors is not a list of length
self.number_of_tuple_elements"
11143,set_configuration_from_sharded_input_tensors,tensorflow/tensorflow/python/tpu/tpu_feed.py,384,method,"Sets the shapes and types of the queue tuple elements.
input_tensors is a list of lists of Tensors whose types and shapes are used
to set the queue configuration. The length of the outer list is the number
of shards required, and each inner list is the tuple of Tensors to use to
determine the types and shapes of the corresponding shard. This method
depends on the shard dimension, and calling it freezes the shard policy.
Args:
input_tensors: list of lists of Tensors. The outer list length corresponds
to the desired number of shards, and each inner list is the size
and shape of the desired configuration of the corresponding shard.
Raises:
ValueError: if any inner list is not a list of length
self.number_of_tuple_elements; or the inner lists do not combine to
form a consistent unsharded shape.
TypeError: if the types of the Tensors in the inner lists do not match."
11144,freeze,tensorflow/tensorflow/python/tpu/tpu_feed.py,434,method,"Freezes the InfeedQueue so it can no longer be modified.
The configuration is implicitly frozen before any host-side or
device-side Ops are generated. The configuration cannot be frozen
until the types and shapes of the tuple elements have been set.
Raises:
ValueError: if the types or shapes of the tuple elements have not been
set."
11145,generate_dequeue_op,tensorflow/tensorflow/python/tpu/tpu_feed.py,460,method,"Generates the device-side Op to dequeue a tuple from the queue.
Implicitly freezes the queue configuration if it is not already
frozen, which will raise errors if the shapes and types have not
been fully specified.
Args:
tpu_device: The TPU device ordinal where the infeed instruction should be
placed. If None, no explicit placement will be performed, and it is up
to the user to call this API from within a proper TPU device scope.
The XLA code will fail if the TPU dequeue instruction is not bound to
any device.
Returns:
A list of Outputs corresponding to a shard of infeed dequeued
into XLA, suitable for use within a replicated block.
Raises:
ValueError: if the types or shapes of the tuple elements have not been
set; or if a dequeue op has already been generated."
11146,generate_enqueue_ops,tensorflow/tensorflow/python/tpu/tpu_feed.py,552,method,"Generates the host-side Ops to enqueue the shards of a tuple.
sharded_inputs is a list, one for each shard, of lists of
Tensors. sharded_inputs[i] is the tuple of Tensors to use to feed
shard i of the queue. Returns the host-side Ops that must be run to
enqueue the sharded tuple. The Op for shard i is colocated with the inputs
for shard i.
Implicitly freezes the queue configuration if it is not already
frozen. If the configuration has already been frozen, and is not
compatible with the types and shapes of sharded_inputs, an error
will be raised.
Args:
sharded_inputs: a list of lists of Tensors. The length of the outer list
determines the number of shards. Each inner list indicates the types
and shapes of the tuples in the corresponding shard.
tpu_ordinal_function: if not None, a function that takes the
shard index as input and returns the ordinal of the TPU device
the shard's infeed should be placed on. tpu_ordinal_function must be
set if the inputs are placed on CPU devices.
placement_function: if not None, a function that takes the shard index as
input and returns the host device where the enqueue op should be placed
on.
Returns:
A list of host-side Ops, one for each shard, that when executed together
will enqueue a full-size element of infeed.
Raises:
ValueError: if the queue configuration has previously been frozen and the
shapes of the elements of sharded_inputs are not compatible with the
frozen configuration; or if the shapes of the elements of sharded_inputs
don't form a consistent unsharded tuple; or if the elements of a tuple
have different device constraints.
TypeError: if the queue configuration has previously been frozen and the
types of the elements of sharded_inputs are not compatible with the
frozen configuration; or if the types of the elements of sharded_inputs
don't form a consistent unsharded tuple."
11147,split_inputs_and_generate_enqueue_ops,tensorflow/tensorflow/python/tpu/tpu_feed.py,625,method,"POORLY-PERFORMING ON MULTI-HOST SYSTEMS.
Generates the host-side Ops to enqueue a tuple.
This method performs poorly because it takes an entire input on a single
host, splits it, and distributes it to all of the cores. It is present only
to simplify tutorial examples.
inputs is a list of Tensors to use to feed the queue. Each input is split
into self.number_of_shards shards. Returns an Op for each shard to enqueue
the shard. The Op for shard i is placed on device placement_function(i).
Implicitly freezes the queue configuration if it is not already
frozen. If the configuration has already been frozen, and is not
compatible with the types and shapes of inputs, an error
will be raised.
Args:
inputs: a list of Tensors which indicates the types and shapes of the
queue tuple.
device_assignment: if not `None`, a TPU `DeviceAssignment`. If
device_assignment is not `None`, but `placement_function` and
`ordinal_function` are None, then `device_assignment` will be used to
place infeeds on the first k TPU shards, where k is the number of shards
in the queue. If all three are `None`, then default placement and
ordinal functions are used.
placement_function: if not None, a function that takes the shard
index as input and returns a device string indicating which
device the shard's infeed should be placed on. If placement_function
and tpu_ordinal_function are None, inputs are sharded round-robin
across the devices in the system.
tpu_ordinal_function: if not None, a function that takes the
shard index as input and returns the ordinal of the TPU device
the shard's infeed should be placed on. If placement_function
and tpu_ordinal_function are None, inputs are sharded round-robin
across the devices in the system.
Returns:
A list of host-side Ops, one for each shard, that when executed together
will enqueue a full-size element of infeed.
Raises:
ValueError: if the queue configuration has previously been frozen and the
shapes of the elements of inputs are not compatible with the frozen
configuration.
TypeError: if the queue configuration has previously been frozen and the
types of the elements of inputs are not compatible with the frozen
configuration."
11148,split_fn,tensorflow/tensorflow/python/tpu/tpu_feed.py,706,method,
11149,TpuContext,tensorflow/tensorflow/python/tpu/tpu_function.py,26,class,A context object holding state about the TPU computation being built.
11150,number_of_shards,tensorflow/tensorflow/python/tpu/tpu_function.py,34,method,
11151,set_number_of_shards,tensorflow/tensorflow/python/tpu/tpu_function.py,37,method,
11152,tpu_shard_context,tensorflow/tensorflow/python/tpu/tpu_function.py,47,function,
11153,get_tpu_context,tensorflow/tensorflow/python/tpu/tpu_function.py,57,function,
11154,on_device_training_loop,tensorflow/tensorflow/python/tpu/tpu_function.py,64,function,
11155,CrossShardOptimizer,tensorflow/tensorflow/python/tpu/tpu_optimizer.py,33,class,An optimizer that averages gradients across TPU shards.
11156,compute_gradients,tensorflow/tensorflow/python/tpu/tpu_optimizer.py,111,method,"Compute gradients of ""loss"" for the variables in ""var_list"".
This simply wraps `compute_gradients()` from the real optimizer. The
gradients will be aggregated in `apply_gradients()` so that user can
modify the gradients like clipping with per replica global norm if needed.
The global norm with aggregated gradients can be bad as one replica's huge
gradients can hurt the gradients from other replicas.
When the CrossShardOptimizer is constructed with
`reduction == losses.Reduction.MEAN` (default), this function scales the
loss by `1.0 / num_shards` before computing the gradients. Assuming the
optimizer uses the default implementation of `compute_gradients()`, the
gradients of the scaled loss are scaled by `1.0 / num_shards` compared to
the gradients of the original loss. This scaling factor is important because
`apply_gradients()` sums gradients across shards, rather than averaging
them. However, the scaling factor must be taken into account when clipping
the norm of the gradients or performing other postprocessing.
Args:
loss: A Tensor containing the value to minimize.
var_list: Optional list or tuple of `tf.Variable` to update to minimize
`loss`. Defaults to the list of variables collected in the graph
under the key `GraphKey.TRAINABLE_VARIABLES`.
**kwargs: Keyword arguments for compute_gradients().
Returns:
A list of (gradient, variable) pairs.
Raises:
ValueError: If not within a tpu_shard_context or group_assignment is
invalid."
11157,apply_gradients,tensorflow/tensorflow/python/tpu/tpu_optimizer.py,163,method,"Apply gradients to variables.
Calls tpu_ops.cross_replica_sum() to sum gradient contributions across
replicas, and then applies the real optimizer.
Args:
grads_and_vars: List of (gradient, variable) pairs as returned by
compute_gradients().
global_step: Optional Variable to increment by one after the
variables have been updated.
name: Optional name for the returned operation. Default to the
name passed to the Optimizer constructor.
Returns:
An `Operation` that applies the gradients. If `global_step` was not None,
that operation also increments `global_step`.
Raises:
ValueError: If the grads_and_vars is malformed."
11158,get_slot,tensorflow/tensorflow/python/tpu/tpu_optimizer.py,194,method,"Return a slot named ""name"" created for ""var"" by the Optimizer.
This simply wraps the get_slot() from the actual optimizer.
Args:
*args: Arguments for get_slot().
**kwargs: Keyword arguments for get_slot().
Returns:
The `Variable` for the slot if it was created, `None` otherwise."
11159,get_slot_names,tensorflow/tensorflow/python/tpu/tpu_optimizer.py,208,method,"Return a list of the names of slots created by the `Optimizer`.
This simply wraps the get_slot_names() from the actual optimizer.
Args:
*args: Arguments for get_slot().
**kwargs: Keyword arguments for get_slot().
Returns:
A list of strings."
11160,variables,tensorflow/tensorflow/python/tpu/tpu_optimizer.py,222,method,Forwarding the variables from the underlying optimizer.
11161,get_tpu_cluster_resolver,tensorflow/tensorflow/python/tpu/tpu_outside_compilation_test.py,48,function,
11162,get_tpu_strategy,tensorflow/tensorflow/python/tpu/tpu_outside_compilation_test.py,57,function,
11163,ShardingPolicy,tensorflow/tensorflow/python/tpu/tpu_sharding.py,31,class,"An object use to hold the sharding policy for a Tensor.
"
11164,freeze,tensorflow/tensorflow/python/tpu/tpu_sharding.py,54,method,"Prevents further modification to the sharding policy.
Any values that have not been set when freeze is called are set to
defaults. If the ShardingPolicy is already frozen, this is a NoOp."
11165,number_of_shards,tensorflow/tensorflow/python/tpu/tpu_sharding.py,65,method,Returns the number of shards in the policy or None if unspecified.
11166,set_number_of_shards,tensorflow/tensorflow/python/tpu/tpu_sharding.py,69,method,"Sets the number of shards for the current policy.
If the policy has been frozen then number_of_shards must match the
existing setting.
Args:
number_of_shards: The number of shards to use in the policy.
Raises:
ValueError: If the policy has been frozen and number_of_shards
differs from the frozen value; or number_of_shards <= 0."
11167,shard_dimension,tensorflow/tensorflow/python/tpu/tpu_sharding.py,96,method,Returns the shard dimension of the policy or None if unspecified.
11168,set_shard_dimension,tensorflow/tensorflow/python/tpu/tpu_sharding.py,100,method,"Sets the shard dimension for the current policy.
If the policy has been frozen then shard_dimension must match the
existing setting.
Args:
shard_dimension: The shard dimension to use in the policy.
Raises:
ValueError: If the policy has been frozen and shard_dimension
differs from the frozen value, or shard_dimension can't be
interpreted as a Dimension."
11169,merge,tensorflow/tensorflow/python/tpu/tpu_sharding.py,122,method,"Merges the policy of another policy into the current policy.
Args:
other: The policy to merge into this one.
Raises:
ValueError: If this policy has been frozen and the merge conflicts with
the frozen policy."
11170,get_sharded_shape,tensorflow/tensorflow/python/tpu/tpu_sharding.py,137,method,"Returns the shape of a shard of a full Tensor.
When given the shape of a 'full-size' Tensor, returns the shape of
the sub-Tensor after it has been sharded. Freezes the policy if it
has not yet been frozen.
Args:
shape: The shape of the full-size Tensor to be sharded.
shard_index: The index of the shard whose shape should be returned.
shard_index can be None for sharding policies that use the same
shape for every shard.
Returns:
The shape of the sharded version of the Tensor.
Raises:
ValueError: If shard_index is None when shards are of different
shapes; or shard_index is not None and
!(0<=shard_index<number_of_shards); or shape does not have at
least self.shard_dimension+1 dimensions; or the value of
shape's shard dimension is not a multiple of
self.number_of_shards"
11171,get_unsharded_shape,tensorflow/tensorflow/python/tpu/tpu_sharding.py,218,method,"Returns the shape of an unsharded Tensor given a list of shards.
When given a list of shapes of shards, returns the shape of the
unsharded Tensor that would generate the shards. Sets defaults for the
policy if number_of_shards or shard_dimension is None.
Args:
shapes: The shapes of the Tensor shards to be combined.
Returns:
The shape of the unsharded version of the Tensor.
Raises:
ValueError: if shapes is not a list of length
self.number_of_shards; or any element of shapes is not a valid
shape consistent with the sharding policy; or the list of
shapes is not a valid sharding of a full shape.
TypeError: if an element of shapes is not convertible to a
TensorShape"
11172,initialize_tpu_system,tensorflow/tensorflow/python/tpu/tpu_strategy_util.py,41,function,"Initialize the TPU devices.
Args:
cluster_resolver: A tf.distribute.cluster_resolver.TPUClusterResolver,
which provides information about the TPU cluster.
Returns:
The tf.tpu.Topology object for the topology of the TPU cluster.
Raises:
RuntimeError: If running inside a tf.function.
NotFoundError: If no TPU devices found in eager mode."
11173,shutdown_tpu_system,tensorflow/tensorflow/python/tpu/tpu_strategy_util.py,139,function,"Shuts down the TPU devices.
This will clear all caches, even those that are maintained through sequential
calls to tf.tpu.experimental.initialize_tpu_system, such as the compilation
cache.
Args:
cluster_resolver: A tf.distribute.cluster_resolver.TPUClusterResolver,
which provides information about the TPU cluster.
Raises:
RuntimeError: If no TPU devices found for eager execution or if run in a
tf.function."
11174,TPUSystemMetadata,tensorflow/tensorflow/python/tpu/tpu_system_metadata.py,45,class,"Describes some metadata about the TPU system.
Attributes:
num_cores: interger. Total number of TPU cores in the TPU system.
num_hosts: interger. Total number of hosts (TPU workers) in the TPU system.
num_of_cores_per_host: interger. Number of TPU cores per host (TPU worker).
topology: an instance of `tf.tpu.experimental.Topology`, which describes the
physical topology of TPU system.
devices: a tuple of strings, which describes all the TPU devices in the
system."
11175,get_session_config_with_timeout,tensorflow/tensorflow/python/tpu/tpu_system_metadata.py,191,function,Returns a session given a timeout and a cluster configuration.
11176,master_job,tensorflow/tensorflow/python/tpu/tpu_system_metadata.py,198,function,"Returns the canonical job name to use to place TPU computations on.
Args:
master: A `string` representing the TensorFlow master to use.
cluster_def: A ClusterDef object describing the TPU cluster.
Returns:
A string containing the job name, or None if no job should be specified.
Raises:
ValueError: If the user needs to specify a tpu_job_name, because we are
unable to infer the job name automatically, or if the user-specified job
names are inappropriate."
11177,do_einsum,tensorflow/tensorflow/python/tpu/tpu_test.py,144,function,
11178,find_einsum,tensorflow/tensorflow/python/tpu/tpu_test.py,150,function,
11179,find_xla_einsum,tensorflow/tensorflow/python/tpu/tpu_test.py,158,function,
11180,maybe_define_flags,tensorflow/tensorflow/python/tpu/tpu_test_wrapper.py,58,function,Defines any required flags that are missing.
11181,calculate_parent_python_path,tensorflow/tensorflow/python/tpu/tpu_test_wrapper.py,73,function,"Returns the absolute import path for the containing directory.
Args:
test_filepath: The filepath which Bazel invoked
(ex: /filesystem/path/tensorflow/tensorflow/python/tpu/tpu_test)
Returns:
Absolute import path of parent (ex: tensorflow.python.tpu).
Raises:
ValueError: if bazel_repo_root does not appear within test_filepath."
11182,import_user_module,tensorflow/tensorflow/python/tpu/tpu_test_wrapper.py,100,function,"Imports the flag-specified user test code.
This runs all top-level statements in the user module, specifically flag
definitions.
Returns:
The user test module."
11183,while_loop,tensorflow/tensorflow/python/tpu/training_loop.py,30,function,"Builds a training loop for TPUs.
The set of loop-carried tensors corresponds to `inputs`. Both
`condition` and `body` take the current value of the loop-carried
tensors. 'body' additionally takes a tuple of infeed from
infeed_queue if infeed_queue is not None. `condition` must return a
single boolean value that determines whether iteration
continues. `body` must return an updated list of values for the
loop-carried tensors.
Args:
condition: a Python function that builds the loop condition.
body: a Python function that builds the loop body.
inputs: a list of initial values passed into the training loop, or
None (equivalent to an empty list).
infeed_queue: if not None, the infeed queue from which to append a tuple
of arguments as inputs to condition.
name: (Deprecated) Does nothing.
Returns:
The final values of the loop-carried tensors.
Raises:
TypeError: if body or condition has the wrong signature."
11184,repeat,tensorflow/tensorflow/python/tpu/training_loop.py,181,function,"Builds a training loop that executes a fixed number of iterations.
The set of loop-carried tensors correspond to `inputs`.
`body` must be a function that takes and returns the values of the
loop-carried tensors.
Args:
n: the number of loop iterations
body: a Python function that builds the loop body.
inputs: a list of initial values passed into the training loop or
None (equivalent to an empty list).
infeed_queue: if not None, the infeed queue from which to append a tuple
of arguments as inputs to condition.
name: (Deprecated) Does nothing.
Returns:
The final values of the loop-carried tensors.
Raises:
ValueError: if there is a type error."
11185,Client,tensorflow/tensorflow/python/tpu/client/client.py,112,class,"Client for working with the Cloud TPU API.
This client is intended to be used for resolving tpu name to ip addresses.
It's recommended to use this library as a contextlib to utilize all
functionality."
11186,recoverable,tensorflow/tensorflow/python/tpu/client/client.py,256,method,"Returns true if the TPU is in a state where training should eventually resume.
If false the TPU is in a unrecoverable state and should be recreated."
11187,symptoms,tensorflow/tensorflow/python/tpu/client/client.py,268,method,Return Cloud TPU Symptoms of the TPU.
11188,state,tensorflow/tensorflow/python/tpu/client/client.py,272,method,Return state of the TPU.
11189,health,tensorflow/tensorflow/python/tpu/client/client.py,276,method,Return health of the TPU.
11190,runtime_version,tensorflow/tensorflow/python/tpu/client/client.py,280,method,Return runtime version of the TPU.
11191,accelerator_type,tensorflow/tensorflow/python/tpu/client/client.py,300,method,Return accelerator type of the TPU.
11192,api_available,tensorflow/tensorflow/python/tpu/client/client.py,304,method,"Return if the Cloud TPU API is available, if not certain features will not work."
11193,name,tensorflow/tensorflow/python/tpu/client/client.py,308,method,"Return the name of the tpu, or the ip address if name is not provided."
11194,get_local_ip,tensorflow/tensorflow/python/tpu/client/client.py,312,method,Return the local ip address of the Google Cloud VM the workload is running on.
11195,network_endpoints,tensorflow/tensorflow/python/tpu/client/client.py,316,method,Return a list of tpu endpoints.
11196,wait_for_healthy,tensorflow/tensorflow/python/tpu/client/client.py,330,method,"Wait for TPU to become healthy or raise error if timeout reached.
Args:
timeout_s (int): The timeout in seconds for waiting TPU to become healthy.
interval (int): The interval in seconds to poll the TPU for health.
Raises:
RuntimeError: If the TPU doesn't become healthy by the timeout."
11197,configure_tpu_version,tensorflow/tensorflow/python/tpu/client/client.py,353,method,"Configure TPU software version.
Args:
version (string): Version of software to configure the TPU with.
restart_type (string): Restart behaviour when switching versions,
defaults to always restart. Options are 'always', 'ifNeeded'."
11198,configure_worker,tensorflow/tensorflow/python/tpu/client/client.py,363,method,"Configure individual TPU worker.
Args:
worker: A dict with the field ipAddress where the configure request will
be sent."
11199,mock_utcnow,tensorflow/tensorflow/python/tpu/client/client_test.py,39,function,
11200,mock_request_compute_metadata,tensorflow/tensorflow/python/tpu/client/client_test.py,43,function,
11201,MockRequestClass,tensorflow/tensorflow/python/tpu/client/client_test.py,53,class,
11202,execute,tensorflow/tensorflow/python/tpu/client/client_test.py,59,method,
11203,MockNodeClass,tensorflow/tensorflow/python/tpu/client/client_test.py,72,class,
11204,get,tensorflow/tensorflow/python/tpu/client/client_test.py,77,method,
11205,all_to_all,tensorflow/tensorflow/python/tpu/ops/tpu_ops.py,45,function,"Exchange data across TPU replicas.
Args:
x: The local tensor.
concat_dimension: The dimension number to concatenate.
split_dimension: The dimension number to split.
split_count: The number of splits, this number must equal to the sub-group
size(group_assignment.get_shape()[1])
group_assignment: Optional 2d int32 lists with shape [num_groups,
num_replicas_per_group]. `group_assignment[i]` represents the replica
ids in the ith subgroup.
name: Optional op name.
Returns:
A `Tensor` which is concatenated by data from different replicas."
11206,cross_replica_sum,tensorflow/tensorflow/python/tpu/ops/tpu_ops.py,94,function,"Sum the input tensor across replicas according to group_assignment.
Args:
x: The local tensor to the sum.
group_assignment: Optional 2d int32 lists with shape [num_groups,
num_replicas_per_group]. `group_assignment[i]` represents the replica
ids in the ith subgroup.
name: Optional op name.
Returns:
A `Tensor` which is summed across replicas."
11207,collective_permute,tensorflow/tensorflow/python/tpu/ops/tpu_ops.py,113,function,"Permute the input tensor across replicas given source_target_pairs.
For each source_target_pair <a, b>, we send replica a's input to replica b.
Each replica id must only appear once in the source column. Also it must
only appear once in the target column.
For the replica id not in the target column, this op returns a zero tensor
with the same shape and dtype of the input x.
For example, suppose there are 4 TPU instances: `[A, B, C, D]`. Passing
source_target_pairs=`[[0,1],[1,2],[2,3]]` gets the outputs:
`[0, A, B, C]`.
Args:
x: The local tensor to be permuted.
source_target_pairs: 2d int lists with shape [num_pairs, 2].
source_target_pairs[i][0] represents the source replica id and
source_target_pairs[i][1] represents the target replica id.
name: Optional op name.
Returns:
A `Tensor` which is permuted."
11208,infeed_dequeue,tensorflow/tensorflow/python/tpu/ops/tpu_ops.py,196,function,"A placeholder op for a value that will be fed into the computation.
Args:
dtype: A `tf.DType`. The type of elements in the tensor.
shape: A `tf.TensorShape` or list of `ints`. The shape of the tensor.
name: A name for the operation (optional).
Returns:
A `Tensor` of type `dtype`.
A tensor that will be provided using the infeed mechanism.
Raises:
TypeError: If 'dtype` is not a supported infeed type."
11209,infeed_dequeue_tuple,tensorflow/tensorflow/python/tpu/ops/tpu_ops.py,221,function,"A placeholder op for values fed into the TPU simultaneously as a tuple.
Args:
dtypes: A list of `tf.DType`s that has length `>= 1`.
The element types of each element in `outputs`.
shapes: A list of shapes (each a `tf.TensorShape` or list of `ints`).
The shapes of each tensor in `outputs`.
name: A name for the operation (optional).
Returns:
A list of `Tensor` objects of type `dtypes`.
A list of tensors that will be provided using the infeed mechanism.
Raises:
TypeError: If a type in 'dtypes` is not a supported infeed type."
11210,send_tpu_embedding_gradients,tensorflow/tensorflow/python/tpu/ops/tpu_ops.py,248,function,"A placeholder op for feeding per-sample gradients to the embedding layer.
Args:
inputs: A TensorList of gradients with which to update embedding tables.
This argument has the same length and shapes as the return value of
RecvTPUEmbeddingActivations, but contains gradients of the model's
loss with respect to the embedding activations. The embedding tables
are updated from these gradients via the optimizers specified in the
TPU embedding configuration given to tpu.initialize_system.
config: Serialized TPUEmbeddingConfiguration proto.
learning_rates: A TensorList of float32 scalars, one for each dynamic
learning rate tag: see the comments in
//third_party/tensorflow/core/protobuf/tpu/
optimization_parameters.proto.
Multiple tables can share the same dynamic learning rate tag as
specified in the configuration. If the learning rates for all tables
are constant, this list should be empty.
name: A name for the operation (optional).
Returns:
A SendTPUEmbeddingGradients operation."
11211,enqueue_tpu_embedding_integer_batch,tensorflow/tensorflow/python/tpu/ops/tpu_ops.py,285,function,"A placeholder op for enqueueing embedding IDs to the TPU.
Args:
batch: A list of 1D tensors, one for each embedding table, containing the
indices into the tables.
device_ordinal: The TPU device to use. Should be >= 0 and less than the
number of TPU cores in the task on which the node is placed.
mode_override: A string input that overrides the mode specified in the
TPUEmbeddingConfiguration. Supported values are {'unspecified',
'inference', 'train', 'backward_pass_only'}. When set to 'unspecified',
the mode set in TPUEmbeddingConfiguration is used, otherwise mode_override
is used (optional).
name: A name for the operation (optional).
Returns:
An EnqueueTPUEmbeddingIntegerBatch operation."
11212,enqueue_tpu_embedding_sparse_batch,tensorflow/tensorflow/python/tpu/ops/tpu_ops.py,320,function,"A placeholder op for enqueueing embedding IDs to the TPU.
Args:
sample_indices: A list of rank 1 Tensors specifying the training example
and feature to which the corresponding embedding_indices and
aggregation_weights values belong. sample_indices[i] must equal b * nf +
f, where nf is the number of features from the corresponding table, f is
in [0, nf), and b is in [0, batch size). Both int32 and int64 are allowed,
and will be converted to int32 internally.
embedding_indices: A list of rank 1 Tensors, indices into the embedding
tables. Both int32 and int64 are allowed and will be converted to int32
internally.
aggregation_weights: A list of rank 1 Tensors containing per sample --
i.e. per (training example, feature) -- aggregation weights. Both float32
and float64 are allowed and will be converted to float32 internally.
device_ordinal: The TPU device to use. Should be >= 0 and less than the
number of TPU cores in the task on which the node is placed.
combiners: A list of string scalars, one for each embedding table that
specify how to normalize the embedding activations after weighted
summation. Supported combiners are 'mean', 'sum', or 'sqrtn'. It is
invalid to have the sum of the weights be 0 for 'mean' or the sum of the
squared weights be 0 for 'sqrtn'. If combiners isn't passed, the default
is to use 'sum' for all tables (optional).
mode_override: A string input that overrides the mode specified in the
TPUEmbeddingConfiguration. Supported values are {'unspecified',
'inference', 'train', 'backward_pass_only'}. When set to 'unspecified',
the mode set in TPUEmbeddingConfiguration is used, otherwise mode_override
is used (optional).
name: A name for the operation (optional).
Returns:
An EnqueueTPUEmbeddingSparseBatch operation."
11213,enqueue_tpu_embedding_sparse_tensor_batch,tensorflow/tensorflow/python/tpu/ops/tpu_ops.py,377,function,"A placeholder op for enqueueing embedding IDs to the TPU.
Args:
sample_indices: A list of rank 2 Tensors specifying the training example
to which the corresponding embedding_indices and aggregation_weights
values belong. It corresponds to sp_ids.indices in
embedding_lookup_sparse(). If the size of its first dimension is 0, we
assume each embedding_indices belongs to a different sample. Both int32
and int64 are allowed and will be converted to int32 internally.
embedding_indices: A list of rank 1 Tensors, indices into the embedding
tables. It corresponds to sp_ids.values in embedding_lookup_sparse(). Both
int32 and int64 are allowed and will be converted to int32 internally.
aggregation_weights: A list of rank 1 Tensors containing per training
example aggregation weights. It corresponds to sp_weights.values in
embedding_lookup_sparse(). If the size of its first dimension is 0, we
assume all weights are 1. Both float32 and float64 are allowed and will
be converted to float32 internally.
table_ids: A list of integers specifying the identifier of the embedding
table (offset of TableDescriptor in the TPUEmbeddingConfiguration) to
lookup the corresponding input. The ith input is looked up using
table_ids[i]. The size of the table_ids list must be equal to that of
sample_indices, embedding_indices and aggregation_weights.
device_ordinal: The TPU device to use. Should be >= 0 and less than the
number of TPU cores in the task on which the node is placed.
max_sequence_lengths: A list of integers, the size of which is equal to
sample_indices. If equal to 0, the corresponding feature is considered to
be a non-sequence feature, If greater than 0, the corresponding feature is
a sequence feature with the given maximal length. If None, then we assume
a list of all zeroes.
combiners: A list of string scalars, one for each embedding table that
specify how to normalize the embedding activations after weighted
summation. Supported combiners are 'mean', 'sum', or 'sqrtn'. It is
invalid to have the sum of the weights be 0 for 'mean' or the sum of the
squared weights be 0 for 'sqrtn'. If combiners isn't passed, the default
is to use 'sum' for all tables (optional).
mode_override: A string input that overrides the mode specified in the
TPUEmbeddingConfiguration. Supported values are {'unspecified',
'inference', 'train', 'backward_pass_only'}. When set to 'unspecified',
the mode set in TPUEmbeddingConfiguration is used, otherwise mode_override
is used (optional).
name: A name for the operation (optional).
Returns:
An EnqueueTPUEmbeddingSparseTensorBatch operation."
11214,enqueue_tpu_embedding_ragged_tensor_batch,tensorflow/tensorflow/python/tpu/ops/tpu_ops.py,450,function,"A placeholder op for enqueueing embedding IDs to the TPU.
Args:
sample_splits: A list of rank 1 Tensors specifying the break points for
splitting embedding_indices and aggregation_weights into rows. It
corresponds to ids.row_splits in embedding_lookup(), when ids is a
RaggedTensor. Both int32 and int64 are allowed and will be converted to
int32 internally.
embedding_indices: A list of rank 1 Tensors, indices into the embedding
tables. It corresponds to ids.values in embedding_lookup(), when ids is a
RaggedTensor. Both int32 and int64 are allowed and will be converted to
int32 internally.
aggregation_weights: A list of rank 1 Tensors containing per training
example aggregation weights. It corresponds to the values field of a
RaggedTensor with the same row_splits as ids in embedding_lookup(), when
ids is a RaggedTensor. Both float32 and float64 are allowed and will be
converted to float32 internally.
table_ids: A list of integers specifying the identifier of the embedding
table (offset of TableDescriptor in the TPUEmbeddingConfiguration) to
lookup the corresponding input. The ith input is looked up using
table_ids[i]. The size of the table_ids list must be equal to that of
sample_indices, embedding_indices and aggregation_weights.
device_ordinal: The TPU device to use. Should be >= 0 and less than the
number of TPU cores in the task on which the node is placed.
max_sequence_lengths: A list of integers, the size of which is equal to
sample_indices. If equal to 0, the corresponding feature is considered to
be a non-sequence feature, If greater than 0, the corresponding feature is
a sequence feature with the given maximal length. If None, then we assume
a list of all zeroes.
combiners: A list of string scalars, one for each embedding table that
specify how to normalize the embedding activations after weighted
summation. Supported combiners are 'mean', 'sum', or 'sqrtn'. It is
invalid to have the sum of the weights be 0 for 'mean' or the sum of the
squared weights be 0 for 'sqrtn'. If combiners isn't passed, the default
is to use 'sum' for all tables (optional).
mode_override: A string input that overrides the mode specified in the
TPUEmbeddingConfiguration. Supported values are {'unspecified',
'inference', 'training', 'backward_pass_only'}. When set to 'unspecified',
the mode set in TPUEmbeddingConfiguration is used, otherwise mode_override
is used (optional).
name: A name for the operation (optional).
Returns:
An EnqueueTPUEmbeddingRaggedTensorBatch operation."
11215,get_workers_list,tensorflow/tensorflow/python/tpu/profiler/capture_tpu_profile.py,85,function,"Returns a comma separated list of TPU worker IP addresses.
Gets cluster_spec from cluster_resolver. Use the worker's task indices to
obtain and return a list of ip addresses.
Args:
cluster_resolver: TensorFlow TPUClusterResolver instance.
Returns:
A string of comma separated list of IP addresses. For example:
'10.2.0.1,10.2.0.2,10.2.0.3,10.2.0.4'
Raises:
UnavailableError: cluster_resolver doesn't contain a valid cluster_spec."
11216,monitoring_helper,tensorflow/tensorflow/python/tpu/profiler/capture_tpu_profile.py,115,function,"Helper function to print monitoring results.
Helper function to print monitoring results for num_queries times.
Args:
service_addr: Address of the TPU profiler service.
duration_ms: Duration of one monitoring sample in milliseconds.
monitoring_level: An integer between 1 and 2. Level 2 is more verbose than
level 1 and shows more metrics.
num_queries: Number of monitoring samples to collect."
11217,ProfileAnalysisStub,tensorflow/tensorflow/python/tpu/profiler/profiler_analysis_pb2_grpc.py,28,class,"//////////////////////////////////////////////////////////////////////////////
ProfileAnalysis service provide entry point for profiling TPU and for
serving profiled data to Tensorboard through GRPC
//////////////////////////////////////////////////////////////////////////////"
11218,ProfileAnalysisServicer,tensorflow/tensorflow/python/tpu/profiler/profiler_analysis_pb2_grpc.py,65,class,"//////////////////////////////////////////////////////////////////////////////
ProfileAnalysis service provide entry point for profiling TPU and for
serving profiled data to Tensorboard through GRPC
//////////////////////////////////////////////////////////////////////////////"
11219,NewSession,tensorflow/tensorflow/python/tpu/profiler/profiler_analysis_pb2_grpc.py,73,method,"Starts a profiling session, blocks until it completes.
TPUProfileAnalysis service delegate this to TPUProfiler service.
Populate the profiled data in repository, then return status to caller."
11220,EnumSessions,tensorflow/tensorflow/python/tpu/profiler/profiler_analysis_pb2_grpc.py,83,method,Enumerate existing sessions and return available profile tools.
11221,GetSessionToolData,tensorflow/tensorflow/python/tpu/profiler/profiler_analysis_pb2_grpc.py,89,method,Retrieve specific tool's data for specific session.
11222,add_ProfileAnalysisServicer_to_server,tensorflow/tensorflow/python/tpu/profiler/profiler_analysis_pb2_grpc.py,96,function,
11223,AdadeltaOptimizer,tensorflow/tensorflow/python/training/adadelta.py,29,class,"Optimizer that implements the Adadelta algorithm.
References:
ADADELTA - An Adaptive Learning Rate Method:
[Zeiler, 2012](http://arxiv.org/abs/1212.5701)
([pdf](http://arxiv.org/pdf/1212.5701v1.pdf))"
11224,AdagradOptimizer,tensorflow/tensorflow/python/training/adagrad.py,32,class,"Optimizer that implements the Adagrad algorithm.
References:
Adaptive Subgradient Methods for Online Learning and Stochastic Optimization
:[Duchi et al., 2011](http://jmlr.org/papers/v12/duchi11a.html)
([pdf](http://www.jmlr.org/papers/volume12/duchi11a/duchi11a.pdf))"
11225,init,tensorflow/tensorflow/python/training/adagrad.py,84,method,
11226,AdagradDAOptimizer,tensorflow/tensorflow/python/training/adagrad_da.py,30,class,"Adagrad Dual Averaging algorithm for sparse linear models.
This optimizer takes care of regularization of unseen features in a mini batch
by updating them when they are seen with a closed form update rule that is
equivalent to having updated them on every mini-batch.
AdagradDA is typically used when there is a need for large sparsity in the
trained model. This optimizer only guarantees sparsity for linear models. Be
careful when using AdagradDA for deep networks as it will require careful
initialization of the gradient accumulators for it to train.
References:
Adaptive Subgradient Methods for Online Learning and Stochastic Optimization
:[Duchi et al., 2011](http://jmlr.org/papers/v12/duchi11a.html)
([pdf](http://www.jmlr.org/papers/volume12/duchi11a/duchi11a.pdf))"
11227,AdamOptimizer,tensorflow/tensorflow/python/training/adam.py,32,class,"Optimizer that implements the Adam algorithm.
References:
Adam - A Method for Stochastic Optimization:
[Kingma et al., 2015](https://arxiv.org/abs/1412.6980)
([pdf](https://arxiv.org/pdf/1412.6980.pdf))"
11228,adam_update_numpy,tensorflow/tensorflow/python/training/adam_test.py,37,function,
11229,basic_train_loop,tensorflow/tensorflow/python/training/basic_loops.py,25,function,"Basic loop to train a model.
Calls `train_step_fn` in a loop to train a model. The function is called as:
```python
train_step_fn(session, *args, **kwargs)
```
It is passed a `tf.compat.v1.Session` in addition to `args` and `kwargs`. The
function
typically runs one training step in the session.
Args:
supervisor: `tf.compat.v1.train.Supervisor` to run the training services.
train_step_fn: Callable to execute one training step. Called repeatedly as
`train_step_fn(session, *args **kwargs)`.
args: Optional positional arguments passed to `train_step_fn`.
kwargs: Optional keyword arguments passed to `train_step_fn`.
master: Master to use to create the training session. Defaults to `""""`
which causes the session to be created in the local process."
11230,SecondOrStepTimer,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,91,class,"Timer that triggers at most once every N seconds or once every N steps.
This symbol is also exported to v2 in tf.estimator namespace. See
https://github.com/tensorflow/estimator/blob/master/tensorflow_estimator/python/estimator/hooks/basic_session_run_hooks.py"
11231,reset,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,110,method,
11232,should_trigger_for_step,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,114,method,"Return true if the timer should trigger for the specified step.
Args:
step: Training step to trigger on.
Returns:
True if the difference between the current time and the time of the last
trigger exceeds `every_secs`, or if the difference between the current
step and the last triggered step exceeds `every_steps`. False otherwise."
11233,update_last_triggered_step,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,141,method,
11234,last_triggered_step,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,154,method,
11235,NeverTriggerTimer,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,158,class,Timer that never triggers.
11236,should_trigger_for_step,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,161,method,
11237,update_last_triggered_step,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,165,method,
11238,last_triggered_step,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,169,method,
11239,LoggingTensorHook,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,174,class,"Prints the given tensors every N local steps, every N seconds, or at end.
The tensors will be printed to the log, with `INFO` severity. If you are not
seeing the logs, you might want to add the following line after your imports:
```python
tf.compat.v1.logging.set_verbosity(tf.compat.v1.logging.INFO)
```
Note that if `at_end` is True, `tensors` should not include any tensor
whose evaluation produces a side effect such as consuming additional inputs."
11240,begin,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,233,method,
11241,before_run,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,242,method,
11242,after_run,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,265,method,
11243,end,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,272,method,
11244,get_or_create_steps_per_run_variable,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,278,function,"Gets or creates the steps_per_run variable.
In Estimator, the user provided computation, the model_fn, is wrapped
inside a tf.while_loop for peak performance. The iterations of the loop are
specified by this variable, which adjusts its value on the CPU after each
device program execution and before the next execution.
The purpose of using a variable, rather than a constant, is to allow
Estimator adapt the device training iterations according to the final steps
specified by users. For example, if the user sets the steps_per_run as
4 and steps as 10 in Estimator.train(), the steps_per_run
variable will have the following value before each training run.
- 1-st execution: steps_per_run = 4
- 2-nd execution: steps_per_run = 4
- 3-rd execution: steps_per_run = 2
As model_fn increases the global step once per train_op invocation, the global
step is 10 after all executions, matching the steps=10 inputs passed in by
users.
Returns:
A TF non-trainable resource variable.
Raises:
RuntimeError: If multi steps_per_run variables were found."
11245,StopAtStepHook,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,390,class,Hook that requests stop at a specified step.
11246,begin,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,419,method,
11247,after_create_session,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,424,method,
11248,before_run,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,429,method,
11249,after_run,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,432,method,
11250,CheckpointSaverListener,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,446,class,"Interface for listeners that take action before or after checkpoint save.
`CheckpointSaverListener` triggers only in steps when `CheckpointSaverHook` is
triggered, and provides callbacks at the following points:
- before using the session
- before each call to `Saver.save()`
- after each call to `Saver.save()`
- at the end of session
To use a listener, implement a class and pass the listener to a
`CheckpointSaverHook`, as in this example:
```python
class ExampleCheckpointSaverListener(CheckpointSaverListener):
def begin(self):
# You can add ops to the graph here.
print('Starting the session.')
self.your_tensor = ...
def before_save(self, session, global_step_value):
print('About to write a checkpoint')
def after_save(self, session, global_step_value):
print('Done writing checkpoint.')
if decided_to_stop_training():
return True
def end(self, session, global_step_value):
print('Done with the session.')
...
listener = ExampleCheckpointSaverListener()
saver_hook = tf.estimator.CheckpointSaverHook(
checkpoint_dir, listeners=[listener])
with
tf.compat.v1.train.MonitoredTrainingSession(chief_only_hooks=[saver_hook]):
...
```
A `CheckpointSaverListener` may simply take some action after every
checkpoint save. It is also possible for the listener to use its own schedule
to act less frequently, e.g. based on global_step_value. In this case,
implementors should implement the `end()` method to handle actions related to
the last checkpoint save. But the listener should not act twice if
`after_save()` already handled this last checkpoint save.
A `CheckpointSaverListener` can request training to be stopped, by returning
True in `after_save`. Please note that, in replicated distributed training
setting, only `chief` should use this behavior. Otherwise each worker will do
their own evaluation, which may be wasteful of resources."
11251,begin,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,499,method,
11252,before_save,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,502,method,
11253,after_save,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,505,method,
11254,end,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,508,method,
11255,CheckpointSaverHook,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,513,class,Saves checkpoints every N steps or seconds.
11256,begin,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,562,method,
11257,after_create_session,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,571,method,
11258,before_run,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,590,method,
11259,after_run,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,593,method,
11260,end,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,604,method,
11261,StepCounterHook,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,660,class,Hook that counts steps per second.
11262,begin,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,683,method,
11263,before_run,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,692,method,
11264,after_run,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,704,method,
11265,NanLossDuringTrainingError,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,740,class,
11266,NanTensorHook,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,747,class,"Monitors the loss tensor and stops training if loss is NaN.
Can either fail with exception or just stop training."
11267,before_run,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,763,method,
11268,after_run,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,766,method,
11269,SummarySaverHook,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,779,class,Saves summaries every N steps.
11270,begin,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,821,method,
11271,before_run,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,830,method,
11272,after_run,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,841,method,
11273,end,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,863,method,
11274,GlobalStepWaiterHook,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,888,class,"Delays execution until global step reaches `wait_until_step`.
This hook delays execution until global step reaches to `wait_until_step`. It
is used to gradually start workers in distributed settings. One example usage
would be setting `wait_until_step=int(K*log(task_id+1))` assuming that
task_id=0 is the chief."
11275,begin,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,905,method,
11276,before_run,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,912,method,
11277,FinalOpsHook,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,937,class,A hook which evaluates `Tensors` at the end of a session.
11278,final_ops_values,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,954,method,
11279,end,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,957,method,
11280,FeedFnHook,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,981,class,Runs `feed_fn` and sets the `feed_dict` accordingly.
11281,before_run,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,993,method,
11282,ProfilerHook,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,999,class,"Captures CPU/GPU profiling information every N steps or seconds.
This produces files called ""timeline-<step>.json"", which are in Chrome
Trace format.
For more information see:
https://github.com/catapult-project/catapult/blob/master/tracing/README.md"
11283,begin,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,1040,method,
11284,before_run,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,1046,method,
11285,after_run,tensorflow/tensorflow/python/training/basic_session_run_hooks.py,1057,method,
11286,MockCheckpointSaverListener,tensorflow/tensorflow/python/training/basic_session_run_hooks_test.py,59,class,
11287,begin,tensorflow/tensorflow/python/training/basic_session_run_hooks_test.py,69,method,
11288,before_save,tensorflow/tensorflow/python/training/basic_session_run_hooks_test.py,72,method,
11289,after_save,tensorflow/tensorflow/python/training/basic_session_run_hooks_test.py,75,method,
11290,end,tensorflow/tensorflow/python/training/basic_session_run_hooks_test.py,80,method,
11291,get_counts,tensorflow/tensorflow/python/training/basic_session_run_hooks_test.py,83,method,
11292,generate_checkpoint_state_proto,tensorflow/tensorflow/python/training/checkpoint_management.py,67,function,"Generates a checkpoint state proto.
Args:
save_dir: Directory where the model was saved.
model_checkpoint_path: The checkpoint file.
all_model_checkpoint_paths: List of strings. Paths to all not-yet-deleted
checkpoints, sorted from oldest to newest. If this is a non-empty list,
the last element must be equal to model_checkpoint_path. These paths
are also saved in the CheckpointState proto.
all_model_checkpoint_timestamps: A list of floats, indicating the number of
seconds since the Epoch when each checkpoint was generated.
last_preserved_timestamp: A float, indicating the number of seconds since
the Epoch when the last preserved checkpoint was written, e.g. due to a
`keep_checkpoint_every_n_hours` parameter (see
`tf.train.CheckpointManager` for an implementation).
Returns:
CheckpointState proto with model_checkpoint_path and
all_model_checkpoint_paths updated to either absolute paths or
relative paths to the current save_dir.
Raises:
ValueError: If `all_model_checkpoint_timestamps` was provided but its length
does not match `all_model_checkpoint_paths`."
11293,update_checkpoint_state,tensorflow/tensorflow/python/training/checkpoint_management.py,136,function,"Updates the content of the 'checkpoint' file.
This updates the checkpoint file containing a CheckpointState
proto.
Args:
save_dir: Directory where the model was saved.
model_checkpoint_path: The checkpoint file.
all_model_checkpoint_paths: List of strings. Paths to all not-yet-deleted
checkpoints, sorted from oldest to newest. If this is a non-empty list,
the last element must be equal to model_checkpoint_path. These paths
are also saved in the CheckpointState proto.
latest_filename: Optional name of the checkpoint file. Default to
'checkpoint'.
all_model_checkpoint_timestamps: Optional list of timestamps (floats,
seconds since the Epoch) indicating when the checkpoints in
`all_model_checkpoint_paths` were created.
last_preserved_timestamp: A float, indicating the number of seconds since
the Epoch when the last preserved checkpoint was written, e.g. due to a
`keep_checkpoint_every_n_hours` parameter (see
`tf.train.CheckpointManager` for an implementation).
Raises:
RuntimeError: If any of the model checkpoint paths conflict with the file
containing CheckpointSate."
11294,update_checkpoint_state_internal,tensorflow/tensorflow/python/training/checkpoint_management.py,177,function,"Updates the content of the 'checkpoint' file.
This updates the checkpoint file containing a CheckpointState
proto.
Args:
save_dir: Directory where the model was saved.
model_checkpoint_path: The checkpoint file.
all_model_checkpoint_paths: List of strings. Paths to all not-yet-deleted
checkpoints, sorted from oldest to newest. If this is a non-empty list,
the last element must be equal to model_checkpoint_path. These paths
are also saved in the CheckpointState proto.
latest_filename: Optional name of the checkpoint file. Default to
'checkpoint'.
save_relative_paths: If `True`, will write relative paths to the checkpoint
state file.
all_model_checkpoint_timestamps: Optional list of timestamps (floats,
seconds since the Epoch) indicating when the checkpoints in
`all_model_checkpoint_paths` were created.
last_preserved_timestamp: A float, indicating the number of seconds since
the Epoch when the last preserved checkpoint was written, e.g. due to a
`keep_checkpoint_every_n_hours` parameter (see
`tf.train.CheckpointManager` for an implementation).
Raises:
RuntimeError: If any of the model checkpoint paths conflict with the file
containing CheckpointSate."
11295,get_checkpoint_state,tensorflow/tensorflow/python/training/checkpoint_management.py,252,function,"Returns CheckpointState proto from the ""checkpoint"" file.
If the ""checkpoint"" file contains a valid CheckpointState
proto, returns it.
Args:
checkpoint_dir: The directory of checkpoints.
latest_filename: Optional name of the checkpoint file. Default to
'checkpoint'.
Returns:
A CheckpointState if the state was available, None
otherwise.
Raises:
ValueError: if the checkpoint read doesn't have model_checkpoint_path set."
11296,checkpoint_exists_internal,tensorflow/tensorflow/python/training/checkpoint_management.py,367,function,"Checks whether a V1 or V2 checkpoint exists with the specified prefix.
This is an internal function to check if a checkpoint exists,
since it takes into account the naming difference between V1 and V2 formats.
Args:
checkpoint_prefix: the prefix of a V1 or V2 checkpoint, with V2 taking
priority. Typically the result of `Saver.save()` or that of
`tf.train.latest_checkpoint()`, regardless of sharded/non-sharded or
V1/V2.
Returns:
A bool, true if a checkpoint referred to by `checkpoint_prefix` exists."
11297,checkpoint_exists,tensorflow/tensorflow/python/training/checkpoint_management.py,395,function,"Checks whether a V1 or V2 checkpoint exists with the specified prefix.
This is the recommended way to check if a checkpoint exists, since it takes
into account the naming difference between V1 and V2 formats.
Args:
checkpoint_prefix: the prefix of a V1 or V2 checkpoint, with V2 taking
priority. Typically the result of `Saver.save()` or that of
`tf.train.latest_checkpoint()`, regardless of sharded/non-sharded or
V1/V2.
Returns:
A bool, true if a checkpoint referred to by `checkpoint_prefix` exists."
11298,get_checkpoint_mtimes,tensorflow/tensorflow/python/training/checkpoint_management.py,417,function,"Returns the mtimes (modification timestamps) of the checkpoints.
Globs for the checkpoints pointed to by `checkpoint_prefixes`. If the files
exist, collect their mtime. Both V2 and V1 checkpoints are considered, in
that priority.
This is the recommended way to get the mtimes, since it takes into account
the naming difference between V1 and V2 formats.
Note: If not all checkpoints exist, the length of the returned mtimes list
will be smaller than the length of `checkpoint_prefixes` list, so mapping
checkpoints to corresponding mtimes will not be possible.
Args:
checkpoint_prefixes: a list of checkpoint paths, typically the results of
`Saver.save()` or those of `tf.train.latest_checkpoint()`, regardless of
sharded/non-sharded or V1/V2.
Returns:
A list of mtimes (in microseconds) of the found checkpoints."
11299,remove_checkpoint,tensorflow/tensorflow/python/training/checkpoint_management.py,463,function,"Removes a checkpoint given by `checkpoint_prefix`.
Args:
checkpoint_prefix: The prefix of a V1 or V2 checkpoint. Typically the result
of `Saver.save()` or that of `tf.train.latest_checkpoint()`, regardless of
sharded/non-sharded or V1/V2.
checkpoint_format_version: `SaverDef.CheckpointFormatVersion`, defaults to
`SaverDef.V2`.
meta_graph_suffix: Suffix for `MetaGraphDef` file. Defaults to 'meta'."
11300,meta_graph_filename,tensorflow/tensorflow/python/training/checkpoint_management.py,493,function,"Returns the meta graph filename.
Args:
checkpoint_filename: Name of the checkpoint file.
meta_graph_suffix: Suffix for `MetaGraphDef` file. Defaults to 'meta'.
Returns:
MetaGraph file name."
11301,CheckpointManager,tensorflow/tensorflow/python/training/checkpoint_management.py,513,class,"Manages multiple checkpoints by keeping some and deleting unneeded ones.
Example usage:
```python
import tensorflow as tf
checkpoint = tf.train.Checkpoint(optimizer=optimizer, model=model)
manager = tf.train.CheckpointManager(
checkpoint, directory=""/tmp/model"", max_to_keep=5)
status = checkpoint.restore(manager.latest_checkpoint)
while True:
# train
manager.save()
```
`CheckpointManager` preserves its own state across instantiations (see the
`__init__` documentation for details). Only one should be active in a
particular directory at a time."
11302,directory,tensorflow/tensorflow/python/training/checkpoint_management.py,667,method,
11303,checkpoint_interval,tensorflow/tensorflow/python/training/checkpoint_management.py,671,method,
11304,checkpoints,tensorflow/tensorflow/python/training/checkpoint_management.py,689,method,"A list of managed checkpoints.
Note that checkpoints saved due to `keep_checkpoint_every_n_hours` will not
show up in this list (to avoid ever-growing filename lists).
Returns:
A list of filenames, sorted from oldest to newest."
11305,checkpoint,tensorflow/tensorflow/python/training/checkpoint_management.py,746,method,Returns the `tf.train.Checkpoint` object.
11306,save,tensorflow/tensorflow/python/training/checkpoint_management.py,750,method,"Creates a new checkpoint and manages it.
Args:
checkpoint_number: An optional integer, or an integer-dtype `Variable` or
`Tensor`, used to number the checkpoint. If `None` (default),
checkpoints are numbered using `checkpoint.save_counter`. Even if
`checkpoint_number` is provided, `save_counter` is still incremented. A
user-provided `checkpoint_number` is not incremented even if it is a
`Variable`.
check_interval: An optional boolean. The argument is only effective when
`checkpoint_interval` is passed into the manager. If `True`, the manager
will only save the checkpoint if the interval between checkpoints is
larger than `checkpoint_interval`. Otherwise it will always save the
checkpoint unless a checkpoint has already been saved for the current
step.
Returns:
The path to the new checkpoint. It is also recorded in the `checkpoints`
and `latest_checkpoint` properties. `None` if no checkpoint is saved."
11307,restore_or_initialize,tensorflow/tensorflow/python/training/checkpoint_management.py,826,method,"Restore items in `checkpoint` from the latest checkpoint file.
This method will first try to restore from the most recent checkpoint in
`directory`. If no checkpoints exist in `directory`, and `init_fn` is
specified, this method will call `init_fn` to do customized
initialization. This can be used to support initialization from pretrained
models.
Note that unlike `tf.train.Checkpoint.restore()`, this method doesn't return
a load status object that users can run assertions on
(e.g. assert_consumed()). Thus to run assertions, users should directly use
`tf.train.Checkpoint.restore()` method.
Returns:
The restored checkpoint path if the lastest checkpoint is found and
restored. Otherwise None."
11308,load_checkpoint,tensorflow/tensorflow/python/training/checkpoint_utils.py,46,function,"Returns `CheckpointReader` for checkpoint found in `ckpt_dir_or_file`.
If `ckpt_dir_or_file` resolves to a directory with multiple checkpoints,
reader for the latest checkpoint is returned.
Args:
ckpt_dir_or_file: Directory with checkpoints file or path to checkpoint
file.
Returns:
`CheckpointReader` object.
Raises:
ValueError: If `ckpt_dir_or_file` resolves to a directory with no
checkpoints."
11309,load_variable,tensorflow/tensorflow/python/training/checkpoint_utils.py,71,function,"Returns the tensor value of the given variable in the checkpoint.
Args:
ckpt_dir_or_file: Directory with checkpoints file or path to checkpoint.
name: Name of the variable to return.
Returns:
A numpy `ndarray` with a copy of the value of this variable."
11310,list_variables,tensorflow/tensorflow/python/training/checkpoint_utils.py,89,function,"Lists the checkpoint keys and shapes of variables in a checkpoint.
Checkpoint keys are paths in a checkpoint graph.
Example usage:
```python
import tensorflow as tf
import os
ckpt_directory = ""/tmp/training_checkpoints/ckpt""
ckpt = tf.train.Checkpoint(optimizer=optimizer, model=model)
manager = tf.train.CheckpointManager(ckpt, ckpt_directory, max_to_keep=3)
train_and_checkpoint(model, manager)
tf.train.list_variables(manager.latest_checkpoint)
```
Args:
ckpt_dir_or_file: Directory with checkpoints file or path to checkpoint.
Returns:
List of tuples `(key, shape)`."
11311,wait_for_new_checkpoint,tensorflow/tensorflow/python/training/checkpoint_utils.py,121,function,"Waits until a new checkpoint file is found.
Args:
checkpoint_dir: The directory in which checkpoints are saved.
last_checkpoint: The last checkpoint path used or `None` if we're expecting
a checkpoint for the first time.
seconds_to_sleep: The number of seconds to sleep for before looking for a
new checkpoint.
timeout: The maximum number of seconds to wait. If left as `None`, then the
process will wait indefinitely.
Returns:
a new checkpoint path, or None if the timeout was reached."
11312,checkpoints_iterator,tensorflow/tensorflow/python/training/checkpoint_utils.py,153,function,"Continuously yield new checkpoint files as they appear.
The iterator only checks for new checkpoints when control flow has been
reverted to it. This means it can miss checkpoints if your code takes longer
to run between iterations than `min_interval_secs` or the interval at which
new checkpoints are written.
The `timeout` argument is the maximum number of seconds to block waiting for
a new checkpoint. It is used in combination with the `timeout_fn` as
follows:
* If the timeout expires and no `timeout_fn` was specified, the iterator
stops yielding.
* If a `timeout_fn` was specified, that function is called and if it returns
a true boolean value the iterator stops yielding.
* If the function returns a false boolean value then the iterator resumes the
wait for new checkpoints. At this point the timeout logic applies again.
This behavior gives control to callers on what to do if checkpoints do not
come fast enough or stop being generated. For example, if callers have a way
to detect that the training has stopped and know that no new checkpoints
will be generated, they can provide a `timeout_fn` that returns `True` when
the training has stopped. If they know that the training is still going on
they return `False` instead.
Args:
checkpoint_dir: The directory in which checkpoints are saved.
min_interval_secs: The minimum number of seconds between yielding
checkpoints.
timeout: The maximum number of seconds to wait between checkpoints. If left
as `None`, then the process will wait indefinitely.
timeout_fn: Optional function to call after a timeout. If the function
returns True, then it means that no new checkpoints will be generated and
the iterator will exit. The function is called with no arguments.
Yields:
String paths to latest checkpoint files as they arrive."
11313,init_from_checkpoint,tensorflow/tensorflow/python/training/checkpoint_utils.py,219,function,"Replaces `tf.Variable` initializers so they load from a checkpoint file.
Values are not loaded immediately, but when the initializer is run
(typically by running a `tf.compat.v1.global_variables_initializer` op).
Note: This overrides default initialization ops of specified variables and
redefines dtype.
Assignment map supports following syntax:
* `'checkpoint_scope_name/': 'scope_name/'` - will load all variables in
current `scope_name` from `checkpoint_scope_name` with matching tensor
names.
* `'checkpoint_scope_name/some_other_variable': 'scope_name/variable_name'` -
will initialize `scope_name/variable_name` variable
from `checkpoint_scope_name/some_other_variable`.
* `'scope_variable_name': variable` - will initialize given `tf.Variable`
object with tensor 'scope_variable_name' from the checkpoint.
* `'scope_variable_name': list(variable)` - will initialize list of
partitioned variables with tensor 'scope_variable_name' from the checkpoint.
* `'/': 'scope_name/'` - will load all variables in current `scope_name` from
checkpoint's root (e.g. no scope).
Supports loading into partitioned variables, which are represented as
`'<variable>/part_<part #>'`.
Example:
```python
# Say, '/tmp/model.ckpt' has the following tensors:
# -- name='old_scope_1/var1', shape=[20, 2]
# -- name='old_scope_1/var2', shape=[50, 4]
# -- name='old_scope_2/var3', shape=[100, 100]
# Create new model's variables
with tf.compat.v1.variable_scope('new_scope_1'):
var1 = tf.compat.v1.get_variable('var1', shape=[20, 2],
initializer=tf.compat.v1.zeros_initializer())
with tf.compat.v1.variable_scope('new_scope_2'):
var2 = tf.compat.v1.get_variable('var2', shape=[50, 4],
initializer=tf.compat.v1.zeros_initializer())
# Partition into 5 variables along the first axis.
var3 = tf.compat.v1.get_variable(name='var3', shape=[100, 100],
initializer=tf.compat.v1.zeros_initializer(),
partitioner=lambda shape, dtype: [5, 1])
# Initialize all variables in `new_scope_1` from `old_scope_1`.
init_from_checkpoint('/tmp/model.ckpt', {'old_scope_1/': 'new_scope_1'})
# Use names to specify which variables to initialize from checkpoint.
init_from_checkpoint('/tmp/model.ckpt',
{'old_scope_1/var1': 'new_scope_1/var1',
'old_scope_1/var2': 'new_scope_2/var2'})
# Or use tf.Variable objects to identify what to initialize.
init_from_checkpoint('/tmp/model.ckpt',
{'old_scope_1/var1': var1,
'old_scope_1/var2': var2})
# Initialize partitioned variables using variable's name
init_from_checkpoint('/tmp/model.ckpt',
{'old_scope_2/var3': 'new_scope_2/var3'})
# Or specify the list of tf.Variable objects.
init_from_checkpoint('/tmp/model.ckpt',
{'old_scope_2/var3': var3._get_variable_list()})
```
Args:
ckpt_dir_or_file: Directory with checkpoints file or path to checkpoint.
assignment_map: Dict, where keys are names of the variables in the
checkpoint and values are current variables or names of current variables
(in default graph).
Raises:
ValueError: If missing variables in current graph, or if missing
checkpoints or tensors in checkpoints."
11314,Coordinator,tensorflow/tensorflow/python/training/coordinator.py,34,class,"A coordinator for threads.
This class implements a simple mechanism to coordinate the termination of a
set of threads.
#### Usage:
```python
# Create a coordinator.
coord = Coordinator()
# Start a number of threads, passing the coordinator to each of them.
...start thread 1...(coord, ...)
...start thread N...(coord, ...)
# Wait for all the threads to terminate.
coord.join(threads)
```
Any of the threads can call `coord.request_stop()` to ask for all the threads
to stop. To cooperate with the requests, each thread must check for
`coord.should_stop()` on a regular basis. `coord.should_stop()` returns
`True` as soon as `coord.request_stop()` has been called.
A typical thread running with a coordinator will do something like:
```python
while not coord.should_stop():
...do some work...
```
#### Exception handling:
A thread can report an exception to the coordinator as part of the
`request_stop()` call. The exception will be re-raised from the
`coord.join()` call.
Thread code:
```python
try:
while not coord.should_stop():
...do some work...
except Exception as e:
coord.request_stop(e)
```
Main code:
```python
try:
...
coord = Coordinator()
# Start a number of threads, passing the coordinator to each of them.
...start thread 1...(coord, ...)
...start thread N...(coord, ...)
# Wait for all the threads to terminate.
coord.join(threads)
except Exception as e:
...exception that was passed to coord.request_stop()
```
To simplify the thread implementation, the Coordinator provides a
context handler `stop_on_exception()` that automatically requests a stop if
an exception is raised. Using the context handler the thread code above
can be written as:
```python
with coord.stop_on_exception():
while not coord.should_stop():
...do some work...
```
#### Grace period for stopping:
After a thread has called `coord.request_stop()` the other threads have a
fixed time to stop, this is called the 'stop grace period' and defaults to 2
minutes. If any of the threads is still alive after the grace period expires
`coord.join()` raises a RuntimeError reporting the laggards.
```python
try:
...
coord = Coordinator()
# Start a number of threads, passing the coordinator to each of them.
...start thread 1...(coord, ...)
...start thread N...(coord, ...)
# Wait for all the threads to terminate, give them 10s grace period
coord.join(threads, stop_grace_period_secs=10)
except RuntimeError:
...one of the threads took more than 10s to stop after request_stop()
...was called.
except Exception:
...exception that was passed to coord.request_stop()
```"
11315,request_stop,tensorflow/tensorflow/python/training/coordinator.py,187,method,"Request that the threads stop.
After this is called, calls to `should_stop()` will return `True`.
Note: If an exception is being passed in, in must be in the context of
handling the exception (i.e. `try: ... except Exception as ex: ...`) and not
a newly created one.
Args:
ex: Optional `Exception`, or Python `exc_info` tuple as returned by
`sys.exc_info()`. If this is the first call to `request_stop()` the
corresponding exception is recorded and re-raised from `join()`."
11316,clear_stop,tensorflow/tensorflow/python/training/coordinator.py,246,method,"Clears the stop flag.
After this is called, calls to `should_stop()` will return `False`."
11317,should_stop,tensorflow/tensorflow/python/training/coordinator.py,257,method,"Check if stop was requested.
Returns:
True if a stop was requested."
11318,stop_on_exception,tensorflow/tensorflow/python/training/coordinator.py,266,method,"Context manager to request stop when an Exception is raised.
Code that uses a coordinator must catch exceptions and pass
them to the `request_stop()` method to stop the other threads
managed by the coordinator.
This context handler simplifies the exception handling.
Use it as follows:
```python
with coord.stop_on_exception():
# Any exception raised in the body of the with
# clause is reported to the coordinator before terminating
# the execution of the body.
...body...
```
This is completely equivalent to the slightly longer code:
```python
try:
...body...
except:
coord.request_stop(sys.exc_info())
```
Yields:
nothing."
11319,wait_for_stop,tensorflow/tensorflow/python/training/coordinator.py,301,method,"Wait till the Coordinator is told to stop.
Args:
timeout: Float. Sleep for up to that many seconds waiting for
should_stop() to become True.
Returns:
True if the Coordinator is told stop, False if the timeout expired."
11320,register_thread,tensorflow/tensorflow/python/training/coordinator.py,313,method,"Register a thread to join.
Args:
thread: A Python thread to join."
11321,join,tensorflow/tensorflow/python/training/coordinator.py,322,method,"Wait for threads to terminate.
This call blocks until a set of threads have terminated. The set of thread
is the union of the threads passed in the `threads` argument and the list
of threads that registered with the coordinator by calling
`Coordinator.register_thread()`.
After the threads stop, if an `exc_info` was passed to `request_stop`, that
exception is re-raised.
Grace period handling: When `request_stop()` is called, threads are given
'stop_grace_period_secs' seconds to terminate. If any of them is still
alive after that period expires, a `RuntimeError` is raised. Note that if
an `exc_info` was passed to `request_stop()` then it is raised instead of
that `RuntimeError`.
Args:
threads: List of `threading.Threads`. The started threads to join in
addition to the registered threads.
stop_grace_period_secs: Number of seconds given to threads to stop after
`request_stop()` has been called.
ignore_live_threads: If `False`, raises an error if any of the threads are
still alive after `stop_grace_period_secs`.
Raises:
RuntimeError: If any thread is still alive after `request_stop()`
is called and the grace period expires."
11322,joined,tensorflow/tensorflow/python/training/coordinator.py,400,method,
11323,raise_requested_exception,tensorflow/tensorflow/python/training/coordinator.py,403,method,"If an exception has been passed to `request_stop`, this raises it."
11324,LooperThread,tensorflow/tensorflow/python/training/coordinator.py,412,class,"A thread that runs code repeatedly, optionally on a timer.
This thread class is intended to be used with a `Coordinator`. It repeatedly
runs code specified either as `target` and `args` or by the `run_loop()`
method.
Before each run the thread checks if the coordinator has requested stop. In
that case the looper thread terminates immediately.
If the code being run raises an exception, that exception is reported to the
coordinator and the thread terminates. The coordinator will then request all
the other threads it coordinates to stop.
You typically pass looper threads to the supervisor `Join()` method."
11325,loop,tensorflow/tensorflow/python/training/coordinator.py,460,method,"Start a LooperThread that calls a function periodically.
If `timer_interval_secs` is None the thread calls `target(args)`
repeatedly. Otherwise `target(args)` is called every `timer_interval_secs`
seconds. The thread terminates when a stop of the coordinator is
requested.
Args:
coord: A Coordinator.
timer_interval_secs: Number. Time boundaries at which to call `target`.
target: A callable object.
args: Optional arguments to pass to `target` when calling it.
kwargs: Optional keyword arguments to pass to `target` when calling it.
Returns:
The started thread."
11326,run,tensorflow/tensorflow/python/training/coordinator.py,483,method,
11327,start_loop,tensorflow/tensorflow/python/training/coordinator.py,498,method,Called when the thread starts.
11328,stop_loop,tensorflow/tensorflow/python/training/coordinator.py,502,method,Called when the thread stops.
11329,run_loop,tensorflow/tensorflow/python/training/coordinator.py,506,method,Called at 'timer_interval_secs' boundaries.
11330,StopOnEvent,tensorflow/tensorflow/python/training/coordinator_test.py,30,function,
11331,RaiseOnEvent,tensorflow/tensorflow/python/training/coordinator_test.py,36,function,
11332,RaiseOnEventUsingContextHandler,tensorflow/tensorflow/python/training/coordinator_test.py,50,function,
11333,SleepABit,tensorflow/tensorflow/python/training/coordinator_test.py,58,function,
11334,WaitForThreadsToRegister,tensorflow/tensorflow/python/training/coordinator_test.py,64,function,
11335,replica_device_setter,tensorflow/tensorflow/python/training/device_setter.py,137,function,"Return a `device function` to use when building a Graph for replicas.
Device Functions are used in `with tf.device(device_function):` statement to
automatically assign devices to `Operation` objects as they are constructed,
Device constraints are added from the inner-most context first, working
outwards. The merging behavior adds constraints to fields that are yet unset
by a more inner context. Currently the fields are (job, task, cpu/gpu).
If `cluster` is `None`, and `ps_tasks` is 0, the returned function is a no-op.
Otherwise, the value of `ps_tasks` is derived from `cluster`.
By default, only Variable ops are placed on ps tasks, and the placement
strategy is round-robin over all ps tasks. A custom `ps_strategy` may be used
to do more intelligent placement, such as
`tf.contrib.training.GreedyLoadBalancingStrategy`.
For example,
```python
# To build a cluster with two ps jobs on hosts ps0 and ps1, and 3 worker
# jobs on hosts worker0, worker1 and worker2.
cluster_spec = {
""ps"": [""ps0:2222"", ""ps1:2222""],
""worker"": [""worker0:2222"", ""worker1:2222"", ""worker2:2222""]}
with
tf.device(tf.compat.v1.train.replica_device_setter(cluster=cluster_spec)):
# Build your graph
v1 = tf.Variable(...) # assigned to /job:ps/task:0
v2 = tf.Variable(...) # assigned to /job:ps/task:1
v3 = tf.Variable(...) # assigned to /job:ps/task:0
# Run compute
```
Args:
ps_tasks: Number of tasks in the `ps` job. Ignored if `cluster` is
provided.
ps_device: String. Device of the `ps` job. If empty no `ps` job is used.
Defaults to `ps`.
worker_device: String. Device of the `worker` job. If empty no `worker`
job is used.
merge_devices: `Boolean`. If `True`, merges or only sets a device if the
device constraint is completely unset. merges device specification rather
than overriding them.
cluster: `ClusterDef` proto or `ClusterSpec`.
ps_ops: List of strings representing `Operation` types that need to be
placed on `ps` devices. If `None`, defaults to `STANDARD_PS_OPS`.
ps_strategy: A callable invoked for every ps `Operation` (i.e. matched by
`ps_ops`), that takes the `Operation` and returns the ps task index to
use. If `None`, defaults to a round-robin strategy across all `ps`
devices.
Returns:
A function to pass to `tf.device()`.
Raises:
TypeError if `cluster` is not a dictionary or `ClusterDef` protocol buffer,
or if `ps_strategy` is provided but not a callable."
11336,logistic_classifier,tensorflow/tensorflow/python/training/evaluation_test.py,47,function,
11337,local_variable,tensorflow/tensorflow/python/training/evaluation_test.py,51,function,
11338,FtrlOptimizer,tensorflow/tensorflow/python/training/ftrl.py,29,class,"Optimizer that implements the FTRL algorithm.
This version has support for both online L2 (McMahan et al., 2013) and
shrinkage-type L2, which is the addition of an L2 penalty
to the loss function.
References:
Ad-click prediction:
[McMahan et al., 2013](https://dl.acm.org/citation.cfm?id=2488200)
([pdf](https://dl.acm.org/ft_gateway.cfm?id=2488200&ftid=1388399&dwn=1&CFID=32233078&CFTOKEN=d60fe57a294c056a-CB75C374-F915-E7A6-1573FBBC7BF7D526))"
11339,GradientDescentOptimizer,tensorflow/tensorflow/python/training/gradient_descent.py,30,class,"Optimizer that implements the gradient descent algorithm.
"
11340,match_filenames_once,tensorflow/tensorflow/python/training/input.py,62,function,"Save the list of files matching pattern, so it is only computed once.
NOTE: The order of the files returned is deterministic.
Args:
pattern: A file pattern (glob), or 1D tensor of file patterns.
name: A name for the operations (optional).
Returns:
A variable that is initialized to the list of files matching the pattern(s)."
11341,limit_epochs,tensorflow/tensorflow/python/training/input.py,85,function,"Returns tensor `num_epochs` times and then raises an `OutOfRange` error.
Note: creates local counter `epochs`. Use `local_variables_initializer()` to
initialize local variables.
Args:
tensor: Any `Tensor`.
num_epochs: A positive integer (optional). If specified, limits the number
of steps the output tensor may be evaluated.
name: A name for the operations (optional).
Returns:
tensor or `OutOfRange`.
Raises:
ValueError: if `num_epochs` is invalid."
11342,input_producer,tensorflow/tensorflow/python/training/input.py,123,function,"Output the rows of `input_tensor` to a queue for an input pipeline.
Note: if `num_epochs` is not `None`, this function creates local counter
`epochs`. Use `local_variables_initializer()` to initialize local variables.
Args:
input_tensor: A tensor with the rows to produce. Must be at least
one-dimensional. Must either have a fully-defined shape, or
`element_shape` must be defined.
element_shape: (Optional.) A `TensorShape` representing the shape of a
row of `input_tensor`, if it cannot be inferred.
num_epochs: (Optional.) An integer. If specified `input_producer` produces
each row of `input_tensor` `num_epochs` times before generating an
`OutOfRange` error. If not specified, `input_producer` can cycle through
the rows of `input_tensor` an unlimited number of times.
shuffle: (Optional.) A boolean. If true, the rows are randomly shuffled
within each epoch.
seed: (Optional.) An integer. The seed to use if `shuffle` is true.
capacity: (Optional.) The capacity of the queue to be used for buffering
the input.
shared_name: (Optional.) If set, this queue will be shared under the given
name across multiple sessions.
summary_name: (Optional.) If set, a scalar summary for the current queue
size will be generated, using this name as part of the tag.
name: (Optional.) A name for queue.
cancel_op: (Optional.) Cancel op for the queue
Returns:
A queue with the output rows. A `QueueRunner` for the queue is
added to the current `QUEUE_RUNNER` collection of the current
graph.
Raises:
ValueError: If the shape of the input cannot be inferred from the arguments.
RuntimeError: If called with eager execution enabled.
@compatibility(eager)
Input pipelines based on Queues are not supported when eager execution is
enabled. Please use the `tf.data` API to ingest data under eager execution.
@end_compatibility"
11343,string_input_producer,tensorflow/tensorflow/python/training/input.py,211,function,"Output strings (e.g. filenames) to a queue for an input pipeline.
Note: if `num_epochs` is not `None`, this function creates local counter
`epochs`. Use `local_variables_initializer()` to initialize local variables.
Args:
string_tensor: A 1-D string tensor with the strings to produce.
num_epochs: An integer (optional). If specified, `string_input_producer`
produces each string from `string_tensor` `num_epochs` times before
generating an `OutOfRange` error. If not specified,
`string_input_producer` can cycle through the strings in `string_tensor`
an unlimited number of times.
shuffle: Boolean. If true, the strings are randomly shuffled within each
epoch.
seed: An integer (optional). Seed used if shuffle == True.
capacity: An integer. Sets the queue capacity.
shared_name: (optional). If set, this queue will be shared under the given
name across multiple sessions. All sessions open to the device which has
this queue will be able to access it via the shared_name. Using this in
a distributed setting means each name will only be seen by one of the
sessions which has access to this operation.
name: A name for the operations (optional).
cancel_op: Cancel op for the queue (optional).
Returns:
A queue with the output strings. A `QueueRunner` for the Queue
is added to the current `Graph`'s `QUEUE_RUNNER` collection.
Raises:
ValueError: If the string_tensor is a null Python list. At runtime,
will fail with an assertion if string_tensor becomes a null tensor.
@compatibility(eager)
Input pipelines based on Queues are not supported when eager execution is
enabled. Please use the `tf.data` API to ingest data under eager execution.
@end_compatibility"
11344,range_input_producer,tensorflow/tensorflow/python/training/input.py,285,function,"Produces the integers from 0 to limit-1 in a queue.
Note: if `num_epochs` is not `None`, this function creates local counter
`epochs`. Use `local_variables_initializer()` to initialize local variables.
Args:
limit: An int32 scalar tensor.
num_epochs: An integer (optional). If specified, `range_input_producer`
produces each integer `num_epochs` times before generating an
OutOfRange error. If not specified, `range_input_producer` can cycle
through the integers an unlimited number of times.
shuffle: Boolean. If true, the integers are randomly shuffled within each
epoch.
seed: An integer (optional). Seed used if shuffle == True.
capacity: An integer. Sets the queue capacity.
shared_name: (optional). If set, this queue will be shared under the given
name across multiple sessions.
name: A name for the operations (optional).
Returns:
A Queue with the output integers. A `QueueRunner` for the Queue
is added to the current `Graph`'s `QUEUE_RUNNER` collection.
@compatibility(eager)
Input pipelines based on Queues are not supported when eager execution is
enabled. Please use the `tf.data` API to ingest data under eager execution.
@end_compatibility"
11345,slice_input_producer,tensorflow/tensorflow/python/training/input.py,328,function,"Produces a slice of each `Tensor` in `tensor_list`.
Implemented using a Queue -- a `QueueRunner` for the Queue
is added to the current `Graph`'s `QUEUE_RUNNER` collection.
Args:
tensor_list: A list of `Tensor` objects. Every `Tensor` in
`tensor_list` must have the same size in the first dimension.
num_epochs: An integer (optional). If specified, `slice_input_producer`
produces each slice `num_epochs` times before generating
an `OutOfRange` error. If not specified, `slice_input_producer` can cycle
through the slices an unlimited number of times.
shuffle: Boolean. If true, the integers are randomly shuffled within each
epoch.
seed: An integer (optional). Seed used if shuffle == True.
capacity: An integer. Sets the queue capacity.
shared_name: (optional). If set, this queue will be shared under the given
name across multiple sessions.
name: A name for the operations (optional).
Returns:
A list of tensors, one for each element of `tensor_list`. If the tensor
in `tensor_list` has shape `[N, a, b, .., z]`, then the corresponding output
tensor will have shape `[a, b, ..., z]`.
Raises:
ValueError: if `slice_input_producer` produces nothing from `tensor_list`.
@compatibility(eager)
Input pipelines based on Queues are not supported when eager execution is
enabled. Please use the `tf.data` API to ingest data under eager execution.
@end_compatibility"
11346,batch,tensorflow/tensorflow/python/training/input.py,929,function,"Creates batches of tensors in `tensors`.
The argument `tensors` can be a list or a dictionary of tensors.
The value returned by the function will be of the same type
as `tensors`.
This function is implemented using a queue. A `QueueRunner` for the
queue is added to the current `Graph`'s `QUEUE_RUNNER` collection.
If `enqueue_many` is `False`, `tensors` is assumed to represent a single
example. An input tensor with shape `[x, y, z]` will be output as a tensor
with shape `[batch_size, x, y, z]`.
If `enqueue_many` is `True`, `tensors` is assumed to represent a batch of
examples, where the first dimension is indexed by example, and all members of
`tensors` should have the same size in the first dimension. If an input
tensor has shape `[*, x, y, z]`, the output will have shape `[batch_size, x,
y, z]`. The `capacity` argument controls the how long the prefetching is
allowed to grow the queues.
The returned operation is a dequeue operation and will throw
`tf.errors.OutOfRangeError` if the input queue is exhausted. If this
operation is feeding another input queue, its queue runner will catch
this exception, however, if this operation is used in your main thread
you are responsible for catching this yourself.
*N.B.:* If `dynamic_pad` is `False`, you must ensure that either
(i) the `shapes` argument is passed, or (ii) all of the tensors in
`tensors` must have fully-defined shapes. `ValueError` will be
raised if neither of these conditions holds.
If `dynamic_pad` is `True`, it is sufficient that the *rank* of the
tensors is known, but individual dimensions may have shape `None`.
In this case, for each enqueue the dimensions with value `None`
may have a variable length; upon dequeue, the output tensors will be padded
on the right to the maximum shape of the tensors in the current minibatch.
For numbers, this padding takes value 0. For strings, this padding is
the empty string. See `PaddingFIFOQueue` for more info.
If `allow_smaller_final_batch` is `True`, a smaller batch value than
`batch_size` is returned when the queue is closed and there are not enough
elements to fill the batch, otherwise the pending elements are discarded.
In addition, all output tensors' static shapes, as accessed via the
`shape` property will have a first `Dimension` value of `None`, and
operations that depend on fixed batch_size would fail.
Args:
tensors: The list or dictionary of tensors to enqueue.
batch_size: The new batch size pulled from the queue.
num_threads: The number of threads enqueuing `tensors`. The batching will
be nondeterministic if `num_threads > 1`.
capacity: An integer. The maximum number of elements in the queue.
enqueue_many: Whether each tensor in `tensors` is a single example.
shapes: (Optional) The shapes for each example. Defaults to the
inferred shapes for `tensors`.
dynamic_pad: Boolean. Allow variable dimensions in input shapes.
The given dimensions are padded upon dequeue so that tensors within a
batch have the same shapes.
allow_smaller_final_batch: (Optional) Boolean. If `True`, allow the final
batch to be smaller if there are insufficient items left in the queue.
shared_name: (Optional). If set, this queue will be shared under the given
name across multiple sessions.
name: (Optional) A name for the operations.
Returns:
A list or dictionary of tensors with the same types as `tensors` (except if
the input is a list of one element, then it returns a tensor, not a list).
Raises:
ValueError: If the `shapes` are not specified, and cannot be
inferred from the elements of `tensors`.
@compatibility(eager)
Input pipelines based on Queues are not supported when eager execution is
enabled. Please use the `tf.data` API to ingest data under eager execution.
@end_compatibility"
11347,maybe_batch,tensorflow/tensorflow/python/training/input.py,1028,function,"Conditionally creates batches of tensors based on `keep_input`.
See docstring in `batch` for more details.
Args:
tensors: The list or dictionary of tensors to enqueue.
keep_input: A `bool` Tensor. This tensor controls whether the input is
added to the queue or not. If it is a scalar and evaluates `True`, then
`tensors` are all added to the queue. If it is a vector and `enqueue_many`
is `True`, then each example is added to the queue only if the
corresponding value in `keep_input` is `True`. This tensor essentially
acts as a filtering mechanism.
batch_size: The new batch size pulled from the queue.
num_threads: The number of threads enqueuing `tensors`. The batching will
be nondeterministic if `num_threads > 1`.
capacity: An integer. The maximum number of elements in the queue.
enqueue_many: Whether each tensor in `tensors` is a single example.
shapes: (Optional) The shapes for each example. Defaults to the
inferred shapes for `tensors`.
dynamic_pad: Boolean. Allow variable dimensions in input shapes.
The given dimensions are padded upon dequeue so that tensors within a
batch have the same shapes.
allow_smaller_final_batch: (Optional) Boolean. If `True`, allow the final
batch to be smaller if there are insufficient items left in the queue.
shared_name: (Optional). If set, this queue will be shared under the given
name across multiple sessions.
name: (Optional) A name for the operations.
Returns:
A list or dictionary of tensors with the same types as `tensors`.
Raises:
ValueError: If the `shapes` are not specified, and cannot be
inferred from the elements of `tensors`."
11348,batch_join,tensorflow/tensorflow/python/training/input.py,1085,function,"Runs a list of tensors to fill a queue to create batches of examples.
The `tensors_list` argument is a list of tuples of tensors, or a list of
dictionaries of tensors. Each element in the list is treated similarly
to the `tensors` argument of `tf.compat.v1.train.batch()`.
WARNING: This function is nondeterministic, since it starts a separate thread
for each tensor.
Enqueues a different list of tensors in different threads.
Implemented using a queue -- a `QueueRunner` for the queue
is added to the current `Graph`'s `QUEUE_RUNNER` collection.
`len(tensors_list)` threads will be started,
with thread `i` enqueuing the tensors from
`tensors_list[i]`. `tensors_list[i1][j]` must match
`tensors_list[i2][j]` in type and shape, except in the first
dimension if `enqueue_many` is true.
If `enqueue_many` is `False`, each `tensors_list[i]` is assumed
to represent a single example. An input tensor `x` will be output as a
tensor with shape `[batch_size] + x.shape`.
If `enqueue_many` is `True`, `tensors_list[i]` is assumed to
represent a batch of examples, where the first dimension is indexed
by example, and all members of `tensors_list[i]` should have the
same size in the first dimension. The slices of any input tensor
`x` are treated as examples, and the output tensors will have shape
`[batch_size] + x.shape[1:]`.
The `capacity` argument controls the how long the prefetching is allowed to
grow the queues.
The returned operation is a dequeue operation and will throw
`tf.errors.OutOfRangeError` if the input queue is exhausted. If this
operation is feeding another input queue, its queue runner will catch
this exception, however, if this operation is used in your main thread
you are responsible for catching this yourself.
*N.B.:* If `dynamic_pad` is `False`, you must ensure that either
(i) the `shapes` argument is passed, or (ii) all of the tensors in
`tensors_list` must have fully-defined shapes. `ValueError` will be
raised if neither of these conditions holds.
If `dynamic_pad` is `True`, it is sufficient that the *rank* of the
tensors is known, but individual dimensions may have value `None`.
In this case, for each enqueue the dimensions with value `None`
may have a variable length; upon dequeue, the output tensors will be padded
on the right to the maximum shape of the tensors in the current minibatch.
For numbers, this padding takes value 0. For strings, this padding is
the empty string. See `PaddingFIFOQueue` for more info.
If `allow_smaller_final_batch` is `True`, a smaller batch value than
`batch_size` is returned when the queue is closed and there are not enough
elements to fill the batch, otherwise the pending elements are discarded.
In addition, all output tensors' static shapes, as accessed via the
`shape` property will have a first `Dimension` value of `None`, and
operations that depend on fixed batch_size would fail.
Args:
tensors_list: A list of tuples or dictionaries of tensors to enqueue.
batch_size: An integer. The new batch size pulled from the queue.
capacity: An integer. The maximum number of elements in the queue.
enqueue_many: Whether each tensor in `tensor_list_list` is a single
example.
shapes: (Optional) The shapes for each example. Defaults to the
inferred shapes for `tensor_list_list[i]`.
dynamic_pad: Boolean. Allow variable dimensions in input shapes.
The given dimensions are padded upon dequeue so that tensors within a
batch have the same shapes.
allow_smaller_final_batch: (Optional) Boolean. If `True`, allow the final
batch to be smaller if there are insufficient items left in the queue.
shared_name: (Optional) If set, this queue will be shared under the given
name across multiple sessions.
name: (Optional) A name for the operations.
Returns:
A list or dictionary of tensors with the same number and types as
`tensors_list[i]`.
Raises:
ValueError: If the `shapes` are not specified, and cannot be
inferred from the elements of `tensor_list_list`.
@compatibility(eager)
Input pipelines based on Queues are not supported when eager execution is
enabled. Please use the `tf.data` API to ingest data under eager execution.
@end_compatibility"
11349,maybe_batch_join,tensorflow/tensorflow/python/training/input.py,1195,function,"Runs a list of tensors to conditionally fill a queue to create batches.
See docstring in `batch_join` for more details.
Args:
tensors_list: A list of tuples or dictionaries of tensors to enqueue.
keep_input: A `bool` Tensor. This tensor controls whether the input is
added to the queue or not. If it is a scalar and evaluates `True`, then
`tensors` are all added to the queue. If it is a vector and `enqueue_many`
is `True`, then each example is added to the queue only if the
corresponding value in `keep_input` is `True`. This tensor essentially
acts as a filtering mechanism.
batch_size: An integer. The new batch size pulled from the queue.
capacity: An integer. The maximum number of elements in the queue.
enqueue_many: Whether each tensor in `tensor_list_list` is a single
example.
shapes: (Optional) The shapes for each example. Defaults to the
inferred shapes for `tensor_list_list[i]`.
dynamic_pad: Boolean. Allow variable dimensions in input shapes.
The given dimensions are padded upon dequeue so that tensors within a
batch have the same shapes.
allow_smaller_final_batch: (Optional) Boolean. If `True`, allow the final
batch to be smaller if there are insufficient items left in the queue.
shared_name: (Optional) If set, this queue will be shared under the given
name across multiple sessions.
name: (Optional) A name for the operations.
Returns:
A list or dictionary of tensors with the same number and types as
`tensors_list[i]`.
Raises:
ValueError: If the `shapes` are not specified, and cannot be
inferred from the elements of `tensor_list_list`."
11350,shuffle_batch,tensorflow/tensorflow/python/training/input.py,1251,function,"Creates batches by randomly shuffling tensors.
This function adds the following to the current `Graph`:
* A shuffling queue into which tensors from `tensors` are enqueued.
* A `dequeue_many` operation to create batches from the queue.
* A `QueueRunner` to `QUEUE_RUNNER` collection, to enqueue the tensors
from `tensors`.
If `enqueue_many` is `False`, `tensors` is assumed to represent a
single example. An input tensor with shape `[x, y, z]` will be output
as a tensor with shape `[batch_size, x, y, z]`.
If `enqueue_many` is `True`, `tensors` is assumed to represent a
batch of examples, where the first dimension is indexed by example,
and all members of `tensors` should have the same size in the
first dimension. If an input tensor has shape `[*, x, y, z]`, the
output will have shape `[batch_size, x, y, z]`.
The `capacity` argument controls the how long the prefetching is allowed to
grow the queues.
The returned operation is a dequeue operation and will throw
`tf.errors.OutOfRangeError` if the input queue is exhausted. If this
operation is feeding another input queue, its queue runner will catch
this exception, however, if this operation is used in your main thread
you are responsible for catching this yourself.
For example:
```python
# Creates batches of 32 images and 32 labels.
image_batch, label_batch = tf.compat.v1.train.shuffle_batch(
[single_image, single_label],
batch_size=32,
num_threads=4,
capacity=50000,
min_after_dequeue=10000)
```
*N.B.:* You must ensure that either (i) the `shapes` argument is
passed, or (ii) all of the tensors in `tensors` must have
fully-defined shapes. `ValueError` will be raised if neither of
these conditions holds.
If `allow_smaller_final_batch` is `True`, a smaller batch value than
`batch_size` is returned when the queue is closed and there are not enough
elements to fill the batch, otherwise the pending elements are discarded.
In addition, all output tensors' static shapes, as accessed via the
`shape` property will have a first `Dimension` value of `None`, and
operations that depend on fixed batch_size would fail.
Args:
tensors: The list or dictionary of tensors to enqueue.
batch_size: The new batch size pulled from the queue.
capacity: An integer. The maximum number of elements in the queue.
min_after_dequeue: Minimum number elements in the queue after a
dequeue, used to ensure a level of mixing of elements.
num_threads: The number of threads enqueuing `tensor_list`.
seed: Seed for the random shuffling within the queue.
enqueue_many: Whether each tensor in `tensor_list` is a single example.
shapes: (Optional) The shapes for each example. Defaults to the
inferred shapes for `tensor_list`.
allow_smaller_final_batch: (Optional) Boolean. If `True`, allow the final
batch to be smaller if there are insufficient items left in the queue.
shared_name: (Optional) If set, this queue will be shared under the given
name across multiple sessions.
name: (Optional) A name for the operations.
Returns:
A list or dictionary of tensors with the types as `tensors`.
Raises:
ValueError: If the `shapes` are not specified, and cannot be
inferred from the elements of `tensors`.
@compatibility(eager)
Input pipelines based on Queues are not supported when eager execution is
enabled. Please use the `tf.data` API to ingest data under eager execution.
@end_compatibility"
11351,maybe_shuffle_batch,tensorflow/tensorflow/python/training/input.py,1355,function,"Creates batches by randomly shuffling conditionally-enqueued tensors.
See docstring in `shuffle_batch` for more details.
Args:
tensors: The list or dictionary of tensors to enqueue.
batch_size: The new batch size pulled from the queue.
capacity: An integer. The maximum number of elements in the queue.
min_after_dequeue: Minimum number elements in the queue after a
dequeue, used to ensure a level of mixing of elements.
keep_input: A `bool` Tensor. This tensor controls whether the input is
added to the queue or not. If it is a scalar and evaluates `True`, then
`tensors` are all added to the queue. If it is a vector and `enqueue_many`
is `True`, then each example is added to the queue only if the
corresponding value in `keep_input` is `True`. This tensor essentially
acts as a filtering mechanism.
num_threads: The number of threads enqueuing `tensor_list`.
seed: Seed for the random shuffling within the queue.
enqueue_many: Whether each tensor in `tensor_list` is a single example.
shapes: (Optional) The shapes for each example. Defaults to the
inferred shapes for `tensor_list`.
allow_smaller_final_batch: (Optional) Boolean. If `True`, allow the final
batch to be smaller if there are insufficient items left in the queue.
shared_name: (Optional) If set, this queue will be shared under the given
name across multiple sessions.
name: (Optional) A name for the operations.
Returns:
A list or dictionary of tensors with the types as `tensors`.
Raises:
ValueError: If the `shapes` are not specified, and cannot be
inferred from the elements of `tensors`.
@compatibility(eager)
Input pipelines based on Queues are not supported when eager execution is
enabled. Please use the `tf.data` API to ingest data under eager execution.
@end_compatibility"
11352,shuffle_batch_join,tensorflow/tensorflow/python/training/input.py,1419,function,"Create batches by randomly shuffling tensors.
The `tensors_list` argument is a list of tuples of tensors, or a list of
dictionaries of tensors. Each element in the list is treated similarly
to the `tensors` argument of `tf.compat.v1.train.shuffle_batch()`.
This version enqueues a different list of tensors in different threads.
It adds the following to the current `Graph`:
* A shuffling queue into which tensors from `tensors_list` are enqueued.
* A `dequeue_many` operation to create batches from the queue.
* A `QueueRunner` to `QUEUE_RUNNER` collection, to enqueue the tensors
from `tensors_list`.
`len(tensors_list)` threads will be started, with thread `i` enqueuing
the tensors from `tensors_list[i]`. `tensors_list[i1][j]` must match
`tensors_list[i2][j]` in type and shape, except in the first dimension if
`enqueue_many` is true.
If `enqueue_many` is `False`, each `tensors_list[i]` is assumed
to represent a single example. An input tensor with shape `[x, y, z]`
will be output as a tensor with shape `[batch_size, x, y, z]`.
If `enqueue_many` is `True`, `tensors_list[i]` is assumed to
represent a batch of examples, where the first dimension is indexed
by example, and all members of `tensors_list[i]` should have the
same size in the first dimension. If an input tensor has shape `[*, x,
y, z]`, the output will have shape `[batch_size, x, y, z]`.
The `capacity` argument controls the how long the prefetching is allowed to
grow the queues.
The returned operation is a dequeue operation and will throw
`tf.errors.OutOfRangeError` if the input queue is exhausted. If this
operation is feeding another input queue, its queue runner will catch
this exception, however, if this operation is used in your main thread
you are responsible for catching this yourself.
If `allow_smaller_final_batch` is `True`, a smaller batch value than
`batch_size` is returned when the queue is closed and there are not enough
elements to fill the batch, otherwise the pending elements are discarded.
In addition, all output tensors' static shapes, as accessed via the
`shape` property will have a first `Dimension` value of `None`, and
operations that depend on fixed batch_size would fail.
Args:
tensors_list: A list of tuples or dictionaries of tensors to enqueue.
batch_size: An integer. The new batch size pulled from the queue.
capacity: An integer. The maximum number of elements in the queue.
min_after_dequeue: Minimum number elements in the queue after a
dequeue, used to ensure a level of mixing of elements.
seed: Seed for the random shuffling within the queue.
enqueue_many: Whether each tensor in `tensor_list_list` is a single
example.
shapes: (Optional) The shapes for each example. Defaults to the
inferred shapes for `tensors_list[i]`.
allow_smaller_final_batch: (Optional) Boolean. If `True`, allow the final
batch to be smaller if there are insufficient items left in the queue.
shared_name: (optional). If set, this queue will be shared under the given
name across multiple sessions.
name: (Optional) A name for the operations.
Returns:
A list or dictionary of tensors with the same number and types as
`tensors_list[i]`.
Raises:
ValueError: If the `shapes` are not specified, and cannot be
inferred from the elements of `tensors_list`.
@compatibility(eager)
Input pipelines based on Queues are not supported when eager execution is
enabled. Please use the `tf.data` API to ingest data under eager execution.
@end_compatibility"
11353,maybe_shuffle_batch_join,tensorflow/tensorflow/python/training/input.py,1517,function,"Create batches by randomly shuffling conditionally-enqueued tensors.
See docstring in `shuffle_batch_join` for more details.
Args:
tensors_list: A list of tuples or dictionaries of tensors to enqueue.
batch_size: An integer. The new batch size pulled from the queue.
capacity: An integer. The maximum number of elements in the queue.
min_after_dequeue: Minimum number elements in the queue after a
dequeue, used to ensure a level of mixing of elements.
keep_input: A `bool` Tensor. This tensor controls whether the input is
added to the queue or not. If it is a scalar and evaluates `True`, then
`tensors` are all added to the queue. If it is a vector and `enqueue_many`
is `True`, then each example is added to the queue only if the
corresponding value in `keep_input` is `True`. This tensor essentially
acts as a filtering mechanism.
seed: Seed for the random shuffling within the queue.
enqueue_many: Whether each tensor in `tensor_list_list` is a single
example.
shapes: (Optional) The shapes for each example. Defaults to the
inferred shapes for `tensors_list[i]`.
allow_smaller_final_batch: (Optional) Boolean. If `True`, allow the final
batch to be smaller if there are insufficient items left in the queue.
shared_name: (optional). If set, this queue will be shared under the given
name across multiple sessions.
name: (Optional) A name for the operations.
Returns:
A list or dictionary of tensors with the same number and types as
`tensors_list[i]`.
Raises:
ValueError: If the `shapes` are not specified, and cannot be
inferred from the elements of `tensors_list`.
@compatibility(eager)
Input pipelines based on Queues are not supported when eager execution is
enabled. Please use the `tf.data` API to ingest data under eager execution.
@end_compatibility"
11354,CreateLocalClusterBenchmark,tensorflow/tensorflow/python/training/localhost_cluster_performance_test.py,60,class,
11355,benchmarkCreateLocalCluster,tensorflow/tensorflow/python/training/localhost_cluster_performance_test.py,62,method,
11356,PartitionedVariablesBenchmark,tensorflow/tensorflow/python/training/localhost_cluster_performance_test.py,80,class,
11357,benchmark_create_1000_partitions_with_100_parameter_servers,tensorflow/tensorflow/python/training/localhost_cluster_performance_test.py,82,method,
11358,MomentumOptimizer,tensorflow/tensorflow/python/training/momentum.py,29,class,"Optimizer that implements the Momentum algorithm.
Computes (if `use_nesterov = False`):
```
accumulation = momentum * accumulation + gradient
variable -= learning_rate * accumulation
```
Note that in the dense version of this algorithm, `accumulation` is updated
and applied regardless of a gradient's value, whereas the sparse version (when
the gradient is an `IndexedSlices`, typically because of `tf.gather` or an
embedding) only updates variable slices and corresponding `accumulation` terms
when that part of the variable was used in the forward pass."
11359,Scaffold,tensorflow/tensorflow/python/training/monitored_session.py,59,class,"Structure to create or gather pieces commonly needed to train a model.
When you build a model for training you usually need ops to initialize
variables, a `Saver` to checkpoint them, an op to collect summaries for
the visualizer, and so on.
Various libraries built on top of the core TensorFlow library take care of
creating some or all of these pieces and storing them in well known
collections in the graph. The `Scaffold` class helps pick these pieces from
the graph collections, creating and adding them to the collections if needed.
If you call the scaffold constructor without any arguments, it will pick
pieces from the collections, creating default ones if needed when
`scaffold.finalize()` is called. You can pass arguments to the constructor to
provide your own pieces. Pieces that you pass to the constructor are not
added to the graph collections.
The following pieces are directly accessible as attributes of the `Scaffold`
object:
* `saver`: A `tf.compat.v1.train.Saver` object taking care of saving the
variables.
Picked from and stored into the `SAVERS` collection in the graph by default.
* `init_op`: An op to run to initialize the variables. Picked from and
stored into the `INIT_OP` collection in the graph by default.
* `ready_op`: An op to verify that the variables are initialized. Picked
from and stored into the `READY_OP` collection in the graph by default.
* `ready_for_local_init_op`: An op to verify that global state has been
initialized and it is alright to run `local_init_op`. Picked from and
stored into the `READY_FOR_LOCAL_INIT_OP` collection in the graph by
default. This is needed when the initialization of local variables depends
on the values of global variables.
* `local_init_op`: An op to initialize the local variables. Picked
from and stored into the `LOCAL_INIT_OP` collection in the graph by default.
* `summary_op`: An op to run and merge the summaries in the graph. Picked
from and stored into the `SUMMARY_OP` collection in the graph by default.
You can also pass the following additional pieces to the constructor:
* `init_feed_dict`: A session feed dictionary that should be used when
running the init op.
* `init_fn`: A callable to run after the init op to perform additional
initializations. The callable will be called as
`init_fn(scaffold, session)`."
11360,finalize,tensorflow/tensorflow/python/training/monitored_session.py,190,method,Creates operations if needed and finalizes the graph.
11361,init_fn,tensorflow/tensorflow/python/training/monitored_session.py,250,method,
11362,init_op,tensorflow/tensorflow/python/training/monitored_session.py,254,method,
11363,ready_op,tensorflow/tensorflow/python/training/monitored_session.py,258,method,
11364,ready_for_local_init_op,tensorflow/tensorflow/python/training/monitored_session.py,262,method,
11365,local_init_op,tensorflow/tensorflow/python/training/monitored_session.py,266,method,
11366,local_init_feed_dict,tensorflow/tensorflow/python/training/monitored_session.py,270,method,
11367,summary_op,tensorflow/tensorflow/python/training/monitored_session.py,274,method,
11368,saver,tensorflow/tensorflow/python/training/monitored_session.py,278,method,
11369,init_feed_dict,tensorflow/tensorflow/python/training/monitored_session.py,282,method,
11370,get_or_default,tensorflow/tensorflow/python/training/monitored_session.py,286,method,Get from cache or create a default operation.
11371,default_local_init_op,tensorflow/tensorflow/python/training/monitored_session.py,303,method,"Returns an op that groups the default local init ops.
This op is used during session initialization when a Scaffold is
initialized without specifying the local_init_op arg. It includes
`tf.compat.v1.local_variables_initializer`,
`tf.compat.v1.tables_initializer`, and also
initializes local session resources.
Returns:
The default Scaffold local init op."
11372,default_init_op,tensorflow/tensorflow/python/training/monitored_session.py,194,method,
11373,default_ready_op,tensorflow/tensorflow/python/training/monitored_session.py,203,method,
11374,default_ready_for_local_init_op,tensorflow/tensorflow/python/training/monitored_session.py,214,method,
11375,MonitoredTrainingSession,tensorflow/tensorflow/python/training/monitored_session.py,434,function,"Creates a `MonitoredSession` for training.
For a chief, this utility sets proper session initializer/restorer. It also
creates hooks related to checkpoint and summary saving. For workers, this
utility sets proper session creator which waits for the chief to
initialize/restore. Please check `tf.compat.v1.train.MonitoredSession` for
more
information.
Args:
master: `String` the TensorFlow master to use.
is_chief: If `True`, it will take care of initialization and recovery the
underlying TensorFlow session. If `False`, it will wait on a chief to
initialize or recover the TensorFlow session.
checkpoint_dir: A string. Optional path to a directory where to restore
variables.
scaffold: A `Scaffold` used for gathering or building supportive ops. If not
specified, a default one is created. It's used to finalize the graph.
hooks: Optional list of `SessionRunHook` objects.
chief_only_hooks: list of `SessionRunHook` objects. Activate these hooks if
`is_chief==True`, ignore otherwise.
save_checkpoint_secs: The frequency, in seconds, that a checkpoint is saved
using a default checkpoint saver. If both `save_checkpoint_steps` and
`save_checkpoint_secs` are set to `None`, then the default checkpoint
saver isn't used. If both are provided, then only `save_checkpoint_secs`
is used. Default 600.
save_summaries_steps: The frequency, in number of global steps, that the
summaries are written to disk using a default summary saver. If both
`save_summaries_steps` and `save_summaries_secs` are set to `None`, then
the default summary saver isn't used. Default 100.
save_summaries_secs: The frequency, in secs, that the summaries are written
to disk using a default summary saver. If both `save_summaries_steps` and
`save_summaries_secs` are set to `None`, then the default summary saver
isn't used. Default not enabled.
config: an instance of `tf.compat.v1.ConfigProto` proto used to configure
the session. It's the `config` argument of constructor of
`tf.compat.v1.Session`.
stop_grace_period_secs: Number of seconds given to threads to stop after
`close()` has been called.
log_step_count_steps: The frequency, in number of global steps, that the
global step/sec is logged.
max_wait_secs: Maximum time workers should wait for the session to become
available. This should be kept relatively short to help detect incorrect
code, but sometimes may need to be increased if the chief takes a while to
start up.
save_checkpoint_steps: The frequency, in number of global steps, that a
checkpoint is saved using a default checkpoint saver. If both
`save_checkpoint_steps` and `save_checkpoint_secs` are set to `None`, then
the default checkpoint saver isn't used. If both are provided, then only
`save_checkpoint_secs` is used. Default not enabled.
summary_dir: A string. Optional path to a directory where to save
summaries. If None, checkpoint_dir is used instead.
save_graph_def: Whether to save the GraphDef and MetaGraphDef to
`checkpoint_dir`. The GraphDef is saved after the session is created as
`graph.pbtxt`. MetaGraphDefs are saved out for every checkpoint as
`model.ckpt-*.meta`.
Returns:
A `MonitoredSession` object."
11376,SessionCreator,tensorflow/tensorflow/python/training/monitored_session.py,609,class,A factory for tf.Session.
11377,create_session,tensorflow/tensorflow/python/training/monitored_session.py,613,method,
11378,ChiefSessionCreator,tensorflow/tensorflow/python/training/monitored_session.py,619,class,Creates a tf.compat.v1.Session for a chief.
11379,create_session,tensorflow/tensorflow/python/training/monitored_session.py,659,method,
11380,WorkerSessionCreator,tensorflow/tensorflow/python/training/monitored_session.py,673,class,Creates a tf.compat.v1.Session for a worker.
11381,create_session,tensorflow/tensorflow/python/training/monitored_session.py,709,method,
11382,MonitoredSession,tensorflow/tensorflow/python/training/monitored_session.py,954,class,"Session-like object that handles initialization, recovery and hooks.
Example usage:
```python
saver_hook = CheckpointSaverHook(...)
summary_hook = SummarySaverHook(...)
with MonitoredSession(session_creator=ChiefSessionCreator(...),
hooks=[saver_hook, summary_hook]) as sess:
while not sess.should_stop():
sess.run(train_op)
```
Initialization: At creation time the monitored session does following things
in given order:
* calls `hook.begin()` for each given hook
* finalizes the graph via `scaffold.finalize()`
* create session
* initializes the model via initialization ops provided by `Scaffold`
* restores variables if a checkpoint exists
* launches queue runners
* calls `hook.after_create_session()`
Run: When `run()` is called, the monitored session does following things:
* calls `hook.before_run()`
* calls TensorFlow `session.run()` with merged fetches and feed_dict
* calls `hook.after_run()`
* returns result of `session.run()` asked by user
* if `AbortedError` or `UnavailableError` occurs, it recovers or
reinitializes the session before executing the run() call again
Exit: At the `close()`, the monitored session does following things in order:
* calls `hook.end()`
* closes the queue runners and the session
* suppresses `OutOfRange` error which indicates that all inputs have been
processed if the monitored_session is used as a context
How to set `tf.compat.v1.Session` arguments:
* In most cases you can set session arguments as follows:
```python
MonitoredSession(
session_creator=ChiefSessionCreator(master=..., config=...))
```
* In distributed setting for a non-chief worker, you can use following:
```python
MonitoredSession(
session_creator=WorkerSessionCreator(master=..., config=...))
```
See `MonitoredTrainingSession` for an example usage based on chief or worker.
Note: This is not a `tf.compat.v1.Session`. For example, it cannot do
following:
* it cannot be set as default session.
* it cannot be sent to saver.save.
* it cannot be sent to tf.train.start_queue_runners.
Args:
session_creator: A factory object to create session. Typically a
`ChiefSessionCreator` which is the default one.
hooks: An iterable of `SessionRunHook' objects.
Returns:
A MonitoredSession object."
11383,SingularMonitoredSession,tensorflow/tensorflow/python/training/monitored_session.py,1042,class,"Session-like object that handles initialization, restoring, and hooks.
Please note that this utility is not recommended for distributed settings.
For distributed settings, please use `tf.compat.v1.train.MonitoredSession`.
The
differences between `MonitoredSession` and `SingularMonitoredSession` are:
* `MonitoredSession` handles `AbortedError` and `UnavailableError` for
distributed settings, but `SingularMonitoredSession` does not.
* `MonitoredSession` can be created in `chief` or `worker` modes.
`SingularMonitoredSession` is always created as `chief`.
* You can access the raw `tf.compat.v1.Session` object used by
`SingularMonitoredSession`, whereas in MonitoredSession the raw session is
private. This can be used:
- To `run` without hooks.
- To save and restore.
* All other functionality is identical.
Example usage:
```python
saver_hook = CheckpointSaverHook(...)
summary_hook = SummarySaverHook(...)
with SingularMonitoredSession(hooks=[saver_hook, summary_hook]) as sess:
while not sess.should_stop():
sess.run(train_op)
```
Initialization: At creation time the hooked session does following things
in given order:
* calls `hook.begin()` for each given hook
* finalizes the graph via `scaffold.finalize()`
* create session
* initializes the model via initialization ops provided by `Scaffold`
* restores variables if a checkpoint exists
* launches queue runners
Run: When `run()` is called, the hooked session does following things:
* calls `hook.before_run()`
* calls TensorFlow `session.run()` with merged fetches and feed_dict
* calls `hook.after_run()`
* returns result of `session.run()` asked by user
Exit: At the `close()`, the hooked session does following things in order:
* calls `hook.end()`
* closes the queue runners and the session
* suppresses `OutOfRange` error which indicates that all inputs have been
processed if the `SingularMonitoredSession` is used as a context."
11384,raw_session,tensorflow/tensorflow/python/training/monitored_session.py,1130,method,Returns underlying `TensorFlow.Session` object.
11385,FakeHook,tensorflow/tensorflow/python/training/monitored_session_test.py,257,class,
11386,begin,tensorflow/tensorflow/python/training/monitored_session_test.py,266,method,
11387,after_create_session,tensorflow/tensorflow/python/training/monitored_session_test.py,269,method,
11388,before_run,tensorflow/tensorflow/python/training/monitored_session_test.py,272,method,
11389,after_run,tensorflow/tensorflow/python/training/monitored_session_test.py,277,method,
11390,end,tensorflow/tensorflow/python/training/monitored_session_test.py,283,method,
11391,MockExtended,tensorflow/tensorflow/python/training/monitored_session_test.py,433,class,
11392,MockStrategy,tensorflow/tensorflow/python/training/monitored_session_test.py,443,class,
11393,StopAtNSession,tensorflow/tensorflow/python/training/monitored_session_test.py,564,class,A wrapped session that stops at the N-th call to _check_stop.
11394,busy_wait_for_coord_stop,tensorflow/tensorflow/python/training/monitored_session_test.py,636,function,
11395,AbortAtNSession,tensorflow/tensorflow/python/training/monitored_session_test.py,770,class,A mock session that aborts at the N-th run call.
11396,close,tensorflow/tensorflow/python/training/monitored_session_test.py,777,method,
11397,run,tensorflow/tensorflow/python/training/monitored_session_test.py,780,method,
11398,StopCoordinatorWithException,tensorflow/tensorflow/python/training/monitored_session_test.py,787,class,With this hook Coordinator throws an exception after N-runs.
11399,after_create_session,tensorflow/tensorflow/python/training/monitored_session_test.py,809,method,
11400,after_run,tensorflow/tensorflow/python/training/monitored_session_test.py,821,method,
11401,FailTrainingAfterCoordinatorStopped,tensorflow/tensorflow/python/training/monitored_session_test.py,832,class,With this hook training encounters an exception after N-runs.
11402,after_create_session,tensorflow/tensorflow/python/training/monitored_session_test.py,839,method,
11403,after_run,tensorflow/tensorflow/python/training/monitored_session_test.py,844,method,
11404,CountingSessionCreator,tensorflow/tensorflow/python/training/monitored_session_test.py,861,class,A creator that counts the number of created sessions.
11405,number_of_sessions_created,tensorflow/tensorflow/python/training/monitored_session_test.py,872,method,
11406,create_session,tensorflow/tensorflow/python/training/monitored_session_test.py,875,method,
11407,FakeSession,tensorflow/tensorflow/python/training/monitored_session_test.py,1252,class,
11408,run,tensorflow/tensorflow/python/training/monitored_session_test.py,1258,method,
11409,RaiseOnceAtCountN,tensorflow/tensorflow/python/training/monitored_session_test.py,1419,class,Hook that raises an Exception at step N.
11410,before_run,tensorflow/tensorflow/python/training/monitored_session_test.py,1427,method,
11411,RunOptionsMetadataHook,tensorflow/tensorflow/python/training/monitored_session_test.py,1436,class,A hook that observes & optionally modifies RunOptions and RunMetadata.
11412,before_run,tensorflow/tensorflow/python/training/monitored_session_test.py,1451,method,
11413,after_run,tensorflow/tensorflow/python/training/monitored_session_test.py,1462,method,
11414,assign_moving_average,tensorflow/tensorflow/python/training/moving_averages.py,36,function,"Compute the moving average of a variable.
The moving average of 'variable' updated with 'value' is:
variable * decay + value * (1 - decay)
The returned Operation sets 'variable' to the newly computed moving average,
by performing this subtraction:
variable -= (1 - decay) * (variable - value)
Since variables that are initialized to a `0` value will be `0` biased,
`zero_debias` optionally enables scaling by the mathematically correct
debiasing factor of
1 - decay ** num_updates
See Section 3 of (Kingma et al., 2015) for more details.
The names of the debias shadow variables, by default, include both the scope
they were created in and the scope of the variables they debias. They are also
given a uniquifying-suffix.
E.g.:
```
with tf.compat.v1.variable_scope('scope1'):
with tf.compat.v1.variable_scope('scope2'):
var = tf.compat.v1.get_variable('foo')
update_1 = tf.assign_moving_average(var, 0.0, 1.0)
update_2 = tf.assign_moving_average(var, 0.0, 0.9)
# var.name: 'scope1/scope2/foo'
# shadow var names: 'scope1/scope2/scope1/scope2/foo/biased'
# 'scope1/scope2/scope1/scope2/foo/biased_1'
```
Args:
variable: A Variable.
value: A tensor with the same shape as 'variable'.
decay: A float Tensor or float value. The moving average decay.
zero_debias: A python bool. If true, assume the variable is 0-initialized
and unbias it, as in (Kingma et al., 2015). See docstring in
`_zero_debias` for more details.
name: Optional name of the returned operation.
Returns:
A tensor which if evaluated will compute and return the new moving average.
References:
Adam - A Method for Stochastic Optimization:
[Kingma et al., 2015](https://arxiv.org/abs/1412.6980)
([pdf](https://arxiv.org/pdf/1412.6980.pdf))"
11415,weighted_moving_average,tensorflow/tensorflow/python/training/moving_averages.py,117,function,"Compute the weighted moving average of `value`.
Conceptually, the weighted moving average is:
`moving_average(value * weight) / moving_average(weight)`,
where a moving average updates by the rule
`new_value = decay * old_value + (1 - decay) * update`
Internally, this Op keeps moving average variables of both `value * weight`
and `weight`.
Args:
value: A numeric `Tensor`.
decay: A float `Tensor` or float value. The moving average decay.
weight: `Tensor` that keeps the current value of a weight. Shape should be
able to multiply `value`.
truediv: Boolean, if `True`, dividing by `moving_average(weight)` is
floating point division. If `False`, use division implied by dtypes.
collections: List of graph collections keys to add the internal variables
`value * weight` and `weight` to. Defaults to
`[GraphKeys.GLOBAL_VARIABLES]`.
name: Optional name of the returned operation. Defaults to
""WeightedMovingAvg"".
Returns:
An Operation that updates and returns the weighted moving average."
11416,ExponentialMovingAverage,tensorflow/tensorflow/python/training/moving_averages.py,285,class,"Maintains moving averages of variables by employing an exponential decay.
When training a model, it is often beneficial to maintain moving averages of
the trained parameters. Evaluations that use averaged parameters sometimes
produce significantly better results than the final trained values.
The `apply()` method adds shadow copies of trained variables and add ops that
maintain a moving average of the trained variables in their shadow copies.
It is used when building the training model. The ops that maintain moving
averages are typically run after each training step.
The `average()` and `average_name()` methods give access to the shadow
variables and their names. They are useful when building an evaluation
model, or when restoring a model from a checkpoint file. They help use the
moving averages in place of the last trained values for evaluations.
The moving averages are computed using exponential decay. You specify the
decay value when creating the `ExponentialMovingAverage` object. The shadow
variables are initialized with the same initial values as the trained
variables. When you run the ops to maintain the moving averages, each
shadow variable is updated with the formula:
`shadow_variable -= (1 - decay) * (shadow_variable - variable)`
This is mathematically equivalent to the classic formula below, but the use
of an `assign_sub` op (the `""-=""` in the formula) allows concurrent lockless
updates to the variables:
`shadow_variable = decay * shadow_variable + (1 - decay) * variable`
Reasonable values for `decay` are close to 1.0, typically in the
multiple-nines range: 0.999, 0.9999, etc.
Example usage when creating a training model:
```python
# Create variables.
var0 = tf.Variable(...)
var1 = tf.Variable(...)
# ... use the variables to build a training model...
...
# Create an op that applies the optimizer. This is what we usually
# would use as a training op.
opt_op = opt.minimize(my_loss, [var0, var1])
# Create an ExponentialMovingAverage object
ema = tf.train.ExponentialMovingAverage(decay=0.9999)
with tf.control_dependencies([opt_op]):
# Create the shadow variables, and add ops to maintain moving averages
# of var0 and var1. This also creates an op that will update the moving
# averages after each training step. This is what we will use in place
# of the usual training op.
training_op = ema.apply([var0, var1])
...train the model by running training_op...
```
There are two ways to use the moving averages for evaluations:
* Build a model that uses the shadow variables instead of the variables.
For this, use the `average()` method which returns the shadow variable
for a given variable.
* Build a model normally but load the checkpoint files to evaluate by using
the shadow variable names. For this use the `average_name()` method. See
the `tf.compat.v1.train.Saver` for more
information on restoring saved variables.
Example of restoring the shadow variable values:
```python
# Create a Saver that loads variables from their saved shadow values.
shadow_var0_name = ema.average_name(var0)
shadow_var1_name = ema.average_name(var1)
saver = tf.compat.v1.train.Saver({shadow_var0_name: var0, shadow_var1_name:
var1})
saver.restore(...checkpoint filename...)
# var0 and var1 now hold the moving average values
```"
11417,name,tensorflow/tensorflow/python/training/moving_averages.py,399,method,The name of this ExponentialMovingAverage object.
11418,apply,tensorflow/tensorflow/python/training/moving_averages.py,403,method,"Maintains moving averages of variables.
`var_list` must be a list of `Variable` or `Tensor` objects. This method
creates shadow variables for all elements of `var_list`. Shadow variables
for `Variable` objects are initialized to the variable's initial value.
They will be added to the `GraphKeys.MOVING_AVERAGE_VARIABLES` collection.
For `Tensor` objects, the shadow variables are initialized to 0 and zero
debiased (see docstring in `assign_moving_average` for more details).
shadow variables are created with `trainable=False` and added to the
`GraphKeys.ALL_VARIABLES` collection. They will be returned by calls to
`tf.compat.v1.global_variables()`.
Returns an op that updates all shadow variables from the current value of
their associated variables.
Note that `apply()` can be called multiple times. When eager execution is
enabled each call to apply will update the variables once, so this needs to
be called in a loop.
Args:
var_list: A list of Variable or Tensor objects. The variables and Tensors
must be of types bfloat16, float16, float32, or float64.
Returns:
An Operation that updates the moving averages.
Raises:
TypeError: If the arguments are not an allowed type."
11419,average,tensorflow/tensorflow/python/training/moving_averages.py,489,method,"Returns the `Variable` holding the average of `var`.
Args:
var: A `Variable` object.
Returns:
A `Variable` object or `None` if the moving average of `var`
is not maintained."
11420,average_name,tensorflow/tensorflow/python/training/moving_averages.py,501,method,"Returns the name of the `Variable` holding the average for `var`.
The typical scenario for `ExponentialMovingAverage` is to compute moving
averages of variables during training, and restore the variables from the
computed moving averages during evaluations.
To restore variables, you have to know the name of the shadow variables.
That name and the original variable can then be passed to a `Saver()` object
to restore the variable from the moving average value with:
`saver = tf.compat.v1.train.Saver({ema.average_name(var): var})`
`average_name()` can be called whether or not `apply()` has been called.
Args:
var: A `Variable` object.
Returns:
A string: The name of the variable that will be used or was used
by the `ExponentialMovingAverage class` to hold the moving average of
`var`."
11421,variables_to_restore,tensorflow/tensorflow/python/training/moving_averages.py,528,method,"Returns a map of names to `Variables` to restore.
If a variable has a moving average, use the moving average variable name as
the restore name; otherwise, use the variable name.
For example,
```python
variables_to_restore = ema.variables_to_restore()
saver = tf.compat.v1.train.Saver(variables_to_restore)
```
Below is an example of such mapping:
```
conv/batchnorm/gamma/ExponentialMovingAverage: conv/batchnorm/gamma,
conv_4/conv2d_params/ExponentialMovingAverage: conv_4/conv2d_params,
global_step: global_step
```
Args:
moving_avg_variables: a list of variables that require to use of the
moving average variable name to be restored. If None, it will default to
variables.moving_average_variables() + variables.trainable_variables()
Returns:
A map from restore_names to variables. The restore_name is either the
original or the moving average version of the variable name, depending
on whether the variable name is in the `moving_avg_variables`."
11422,get_filtered_grad_fn,tensorflow/tensorflow/python/training/optimizer.py,48,function,
11423,Optimizer,tensorflow/tensorflow/python/training/optimizer.py,217,class,"Base class for optimizers.
This class defines the API to add Ops to train a model. You never use this
class directly, but instead instantiate one of its subclasses such as
`GradientDescentOptimizer`, `AdagradOptimizer`, or `MomentumOptimizer`.
### Usage
```python
# Create an optimizer with the desired parameters.
opt = GradientDescentOptimizer(learning_rate=0.1)
# Add Ops to the graph to minimize a cost by updating a list of variables.
# ""cost"" is a Tensor, and the list of variables contains tf.Variable
# objects.
opt_op = opt.minimize(cost, var_list=<list of variables>)
```
In the training program you will just have to run the returned Op.
```python
# Execute opt_op to do one step of training:
opt_op.run()
```
### Processing gradients before applying them.
Calling `minimize()` takes care of both computing the gradients and
applying them to the variables. If you want to process the gradients
before applying them you can instead use the optimizer in three steps:
1. Compute the gradients with `compute_gradients()`.
2. Process the gradients as you wish.
3. Apply the processed gradients with `apply_gradients()`.
Example:
```python
# Create an optimizer.
opt = GradientDescentOptimizer(learning_rate=0.1)
# Compute the gradients for a list of variables.
grads_and_vars = opt.compute_gradients(loss, <list of variables>)
# grads_and_vars is a list of tuples (gradient, variable). Do whatever you
# need to the 'gradient' part, for example cap them, etc.
capped_grads_and_vars = [(MyCapper(gv[0]), gv[1]) for gv in grads_and_vars]
# Ask the optimizer to apply the capped gradients.
opt.apply_gradients(capped_grads_and_vars)
```
### Gating Gradients
Both `minimize()` and `compute_gradients()` accept a `gate_gradients`
argument that controls the degree of parallelism during the application of
the gradients.
The possible values are: `GATE_NONE`, `GATE_OP`, and `GATE_GRAPH`.
<b>`GATE_NONE`</b>: Compute and apply gradients in parallel. This provides
the maximum parallelism in execution, at the cost of some non-reproducibility
in the results. For example the two gradients of `matmul` depend on the input
values: With `GATE_NONE` one of the gradients could be applied to one of the
inputs _before_ the other gradient is computed resulting in non-reproducible
results.
<b>`GATE_OP`</b>: For each Op, make sure all gradients are computed before
they are used. This prevents race conditions for Ops that generate gradients
for multiple inputs where the gradients depend on the inputs.
<b>`GATE_GRAPH`</b>: Make sure all gradients for all variables are computed
before any one of them is used. This provides the least parallelism but can
be useful if you want to process all gradients before applying any of them.
### Slots
Some optimizer subclasses, such as `MomentumOptimizer` and `AdagradOptimizer`
allocate and manage additional variables associated with the variables to
train. These are called <i>Slots</i>. Slots have names and you can ask the
optimizer for the names of the slots that it uses. Once you have a slot name
you can ask the optimizer for the variable it created to hold the slot value.
This can be useful if you want to log debug a training algorithm, report stats
about the slots, etc."
11424,get_name,tensorflow/tensorflow/python/training/optimizer.py,352,method,
11425,minimize,tensorflow/tensorflow/python/training/optimizer.py,355,method,"Add operations to minimize `loss` by updating `var_list`.
This method simply combines calls `compute_gradients()` and
`apply_gradients()`. If you want to process the gradient before applying
them call `compute_gradients()` and `apply_gradients()` explicitly instead
of using this function.
Args:
loss: A `Tensor` containing the value to minimize.
global_step: Optional `Variable` to increment by one after the
variables have been updated.
var_list: Optional list or tuple of `Variable` objects to update to
minimize `loss`. Defaults to the list of variables collected in
the graph under the key `GraphKeys.TRAINABLE_VARIABLES`.
gate_gradients: How to gate the computation of gradients. Can be
`GATE_NONE`, `GATE_OP`, or `GATE_GRAPH`.
aggregation_method: Specifies the method used to combine gradient terms.
Valid values are defined in the class `AggregationMethod`.
colocate_gradients_with_ops: If True, try colocating gradients with
the corresponding op.
name: Optional name for the returned operation.
grad_loss: Optional. A `Tensor` holding the gradient computed for `loss`.
Returns:
An Operation that updates the variables in `var_list`. If `global_step`
was not `None`, that operation also increments `global_step`.
Raises:
ValueError: If some of the variables are not `Variable` objects.
@compatibility(eager)
When eager execution is enabled, `loss` should be a Python function that
takes no arguments and computes the value to be minimized. Minimization (and
gradient computation) is done with respect to the elements of `var_list` if
not None, else with respect to any trainable variables created during the
execution of the `loss` function. `gate_gradients`, `aggregation_method`,
`colocate_gradients_with_ops` and `grad_loss` are ignored when eager
execution is enabled.
@end_compatibility"
11426,compute_gradients,tensorflow/tensorflow/python/training/optimizer.py,415,method,"Compute gradients of `loss` for the variables in `var_list`.
This is the first part of `minimize()`. It returns a list
of (gradient, variable) pairs where ""gradient"" is the gradient
for ""variable"". Note that ""gradient"" can be a `Tensor`, an
`IndexedSlices`, or `None` if there is no gradient for the
given variable.
Args:
loss: A Tensor containing the value to minimize or a callable taking
no arguments which returns the value to minimize. When eager execution
is enabled it must be a callable.
var_list: Optional list or tuple of `tf.Variable` to update to minimize
`loss`. Defaults to the list of variables collected in the graph
under the key `GraphKeys.TRAINABLE_VARIABLES`.
gate_gradients: How to gate the computation of gradients. Can be
`GATE_NONE`, `GATE_OP`, or `GATE_GRAPH`.
aggregation_method: Specifies the method used to combine gradient terms.
Valid values are defined in the class `AggregationMethod`.
colocate_gradients_with_ops: If True, try colocating gradients with
the corresponding op.
grad_loss: Optional. A `Tensor` holding the gradient computed for `loss`.
Returns:
A list of (gradient, variable) pairs. Variable is always present, but
gradient can be `None`.
Raises:
TypeError: If `var_list` contains anything else than `Variable` objects.
ValueError: If some arguments are invalid.
RuntimeError: If called with eager execution enabled and `loss` is
not callable.
@compatibility(eager)
When eager execution is enabled, `gate_gradients`, `aggregation_method`,
and `colocate_gradients_with_ops` are ignored.
@end_compatibility"
11427,apply_gradients,tensorflow/tensorflow/python/training/optimizer.py,531,method,"Apply gradients to variables.
This is the second part of `minimize()`. It returns an `Operation` that
applies gradients.
Args:
grads_and_vars: List of (gradient, variable) pairs as returned by
`compute_gradients()`.
global_step: Optional `Variable` to increment by one after the
variables have been updated.
name: Optional name for the returned operation. Default to the
name passed to the `Optimizer` constructor.
Returns:
An `Operation` that applies the specified gradients. If `global_step`
was not None, that operation also increments `global_step`.
Raises:
TypeError: If `grads_and_vars` is malformed.
ValueError: If none of the variables have gradients.
RuntimeError: If you should use `_distributed_apply()` instead."
11428,get_slot,tensorflow/tensorflow/python/training/optimizer.py,737,method,"Return a slot named `name` created for `var` by the Optimizer.
Some `Optimizer` subclasses use additional variables. For example
`Momentum` and `Adagrad` use variables to accumulate updates. This method
gives access to these `Variable` objects if for some reason you need them.
Use `get_slot_names()` to get the list of slot names created by the
`Optimizer`.
Args:
var: A variable passed to `minimize()` or `apply_gradients()`.
name: A string.
Returns:
The `Variable` for the slot if it was created, `None` otherwise."
11429,get_slot_names,tensorflow/tensorflow/python/training/optimizer.py,775,method,"Return a list of the names of slots created by the `Optimizer`.
See `get_slot()`.
Returns:
A list of strings."
11430,variables,tensorflow/tensorflow/python/training/optimizer.py,785,method,"A list of variables which encode the current state of `Optimizer`.
Includes slot variables and additional global variables created by the
optimizer in the current default graph.
Returns:
A list of variables."
11431,update,tensorflow/tensorflow/python/training/optimizer.py,676,method,Apply gradients to a replica variable.
11432,finish,tensorflow/tensorflow/python/training/optimizer.py,714,method,
11433,ProximalAdagradOptimizer,tensorflow/tensorflow/python/training/proximal_adagrad.py,30,class,"Optimizer that implements the Proximal Adagrad algorithm.
References:
Adaptive Subgradient Methods for Online Learning and Stochastic Optimization:
[Duchi et al., 2011](http://jmlr.org/papers/v12/duchi11a.html)
([pdf](http://www.jmlr.org/papers/volume12/duchi11a/duchi11a.pdf))
Efficient Learning using Forward-Backward Splitting:
[Duchi et al., 2009](http://papers.nips.cc/paper/3793-efficient-learning-using-forward-backward-splitting)
([pdf](http://papers.nips.cc/paper/3793-efficient-learning-using-forward-backward-splitting.pdf))"
11434,ProximalGradientDescentOptimizer,tensorflow/tensorflow/python/training/proximal_gradient_descent.py,31,class,"Optimizer that implements the proximal gradient descent algorithm.
References:
Efficient Learning using Forward-Backward Splitting:
[Duchi et al., 2009](http://papers.nips.cc/paper/3793-efficient-learning-using-forward-backward-splitting)
([pdf](http://papers.nips.cc/paper/3793-efficient-learning-using-forward-backward-splitting.pdf))"
11435,error_translator,tensorflow/tensorflow/python/training/py_checkpoint_reader.py,27,function,Translate the tensor_slice_reader.cc errors.
11436,get_variable_to_dtype_map,tensorflow/tensorflow/python/training/py_checkpoint_reader.py,51,function,
11437,has_tensor,tensorflow/tensorflow/python/training/py_checkpoint_reader.py,60,function,
11438,get_tensor,tensorflow/tensorflow/python/training/py_checkpoint_reader.py,66,function,Get the tensor from the Checkpoint object.
11439,NewCheckpointReader,tensorflow/tensorflow/python/training/py_checkpoint_reader.py,85,function,"A function that returns a CheckPointReader.
Args:
filepattern: The filename.
Returns:
A CheckpointReader object."
11440,do_quantize_training_on_graphdef,tensorflow/tensorflow/python/training/quantize_training.py,31,function,"A general quantization scheme is being developed in `tf.contrib.quantize`.
Consider using that instead, though since it is in the tf.contrib namespace,
it is not subject to backward compatibility guarantees.
Args:
input_graph: A `GraphDef`.
num_bits: The number of bits for quantize training.
Returns:
The graph with quantize training done."
11441,QueueRunner,tensorflow/tensorflow/python/training/queue_runner_impl.py,38,class,"Holds a list of enqueue operations for a queue, each to be run in a thread.
Queues are a convenient TensorFlow mechanism to compute tensors
asynchronously using multiple threads. For example in the canonical 'Input
Reader' setup one set of threads generates filenames in a queue; a second set
of threads read records from the files, processes them, and enqueues tensors
on a second queue; a third set of threads dequeues these input records to
construct batches and runs them through training operations.
There are several delicate issues when running multiple threads that way:
closing the queues in sequence as the input is exhausted, correctly catching
and reporting exceptions, etc.
The `QueueRunner`, combined with the `Coordinator`, helps handle these issues.
@compatibility(eager)
QueueRunners are not compatible with eager execution. Instead, please
use `tf.data` to get data into your model.
@end_compatibility"
11442,queue,tensorflow/tensorflow/python/training/queue_runner_impl.py,195,method,
11443,enqueue_ops,tensorflow/tensorflow/python/training/queue_runner_impl.py,199,method,
11444,close_op,tensorflow/tensorflow/python/training/queue_runner_impl.py,203,method,
11445,cancel_op,tensorflow/tensorflow/python/training/queue_runner_impl.py,207,method,
11446,queue_closed_exception_types,tensorflow/tensorflow/python/training/queue_runner_impl.py,211,method,
11447,exceptions_raised,tensorflow/tensorflow/python/training/queue_runner_impl.py,215,method,"Exceptions raised but not handled by the `QueueRunner` threads.
Exceptions raised in queue runner threads are handled in one of two ways
depending on whether or not a `Coordinator` was passed to
`create_threads()`:
* With a `Coordinator`, exceptions are reported to the coordinator and
forgotten by the `QueueRunner`.
* Without a `Coordinator`, exceptions are captured by the `QueueRunner` and
made available in this `exceptions_raised` property.
Returns:
A list of Python `Exception` objects. The list is empty if no exception
was captured. (No exceptions are captured when using a Coordinator.)"
11448,name,tensorflow/tensorflow/python/training/queue_runner_impl.py,234,method,The string name of the underlying Queue.
11449,create_threads,tensorflow/tensorflow/python/training/queue_runner_impl.py,301,method,"Create threads to run the enqueue ops for the given session.
This method requires a session in which the graph was launched. It creates
a list of threads, optionally starting them. There is one thread for each
op passed in `enqueue_ops`.
The `coord` argument is an optional coordinator that the threads will use
to terminate together and report exceptions. If a coordinator is given,
this method starts an additional thread to close the queue when the
coordinator requests a stop.
If previously created threads for the given session are still running, no
new threads will be created.
Args:
sess: A `Session`.
coord: Optional `Coordinator` object for reporting errors and checking
stop conditions.
daemon: Boolean. If `True` make the threads daemon threads.
start: Boolean. If `True` starts the threads. If `False` the
caller must call the `start()` method of the returned threads.
Returns:
A list of threads."
11450,to_proto,tensorflow/tensorflow/python/training/queue_runner_impl.py,358,method,"Converts this `QueueRunner` to a `QueueRunnerDef` protocol buffer.
Args:
export_scope: Optional `string`. Name scope to remove.
Returns:
A `QueueRunnerDef` protocol buffer, or `None` if the `Variable` is not in
the specified name scope."
11451,from_proto,tensorflow/tensorflow/python/training/queue_runner_impl.py,388,method,Returns a `QueueRunner` object created from `queue_runner_def`.
11452,add_queue_runner,tensorflow/tensorflow/python/training/queue_runner_impl.py,396,function,"Adds a `QueueRunner` to a collection in the graph.
When building a complex model that uses many queues it is often difficult to
gather all the queue runners that need to be run. This convenience function
allows you to add a queue runner to a well known collection in the graph.
The companion method `start_queue_runners()` can be used to start threads for
all the collected queue runners.
Args:
qr: A `QueueRunner`.
collection: A `GraphKey` specifying the graph collection to add
the queue runner to. Defaults to `GraphKeys.QUEUE_RUNNERS`."
11453,start_queue_runners,tensorflow/tensorflow/python/training/queue_runner_impl.py,417,function,"Starts all queue runners collected in the graph.
This is a companion method to `add_queue_runner()`. It just starts
threads for all queue runners collected in the graph. It returns
the list of all threads.
Args:
sess: `Session` used to run the queue ops. Defaults to the
default session.
coord: Optional `Coordinator` for coordinating the started threads.
daemon: Whether the threads should be marked as `daemons`, meaning
they don't block program exit.
start: Set to `False` to only create the threads, not start them.
collection: A `GraphKey` specifying the graph collection to
get the queue runners from. Defaults to `GraphKeys.QUEUE_RUNNERS`.
Raises:
ValueError: if `sess` is None and there isn't any default session.
TypeError: if `sess` is not a `tf.compat.v1.Session` object.
Returns:
A list of threads.
Raises:
RuntimeError: If called with eager execution enabled.
ValueError: If called without a default `tf.compat.v1.Session` registered.
@compatibility(eager)
Not compatible with eager execution. To ingest data under eager execution,
use the `tf.data` API instead.
@end_compatibility"
11454,RMSPropOptimizer,tensorflow/tensorflow/python/training/rmsprop.py,54,class,"Optimizer that implements the RMSProp algorithm (Tielemans et al.
2012).
References:
Coursera slide 29:
Hinton, 2012
([pdf](http://www.cs.toronto.edu/~tijmen/csc321/slides/lecture_slides_lec6.pdf))"
11455,BaseSaverBuilder,tensorflow/tensorflow/python/training/saver.py,70,class,"Base class for Savers.
Can be extended to create different Ops."
11456,save_op,tensorflow/tensorflow/python/training/saver.py,86,method,"Create an Op to save 'saveables'.
This is intended to be overridden by subclasses that want to generate
different Ops.
Args:
filename_tensor: String Tensor.
saveables: A list of BaseSaverBuilder.SaveableObject objects.
Returns:
An Operation that save the variables.
Raises:
RuntimeError: (implementation detail) if ""self._write_version"" is an
unexpected value."
11457,bulk_restore,tensorflow/tensorflow/python/training/saver.py,126,method,"Restore all tensors contained in saveables.
By default, this issues separate calls to `restore_op` for each saveable.
Subclasses may override to load multiple saveables in a single call.
Args:
filename_tensor: String Tensor.
saveables: List of BaseSaverBuilder.SaveableObject objects.
preferred_shard: Int. Shard to open first when loading a sharded file.
restore_sequentially: Unused. Bool. If true, each restore is sequential.
Returns:
A list of Tensors resulting from reading 'saveable' from
'filename'."
11458,restore_op,tensorflow/tensorflow/python/training/saver.py,157,method,"Create ops to restore 'saveable'.
This is intended to be overridden by subclasses that want to generate
different Ops.
Args:
filename_tensor: String Tensor.
saveable: A BaseSaverBuilder.SaveableObject object.
preferred_shard: Int. Shard to open first when loading a sharded file.
Returns:
A list of Tensors resulting from reading 'saveable' from
'filename'."
11459,sharded_filename,tensorflow/tensorflow/python/training/saver.py,183,method,"Append sharding information to a filename.
Args:
filename_tensor: A string tensor.
shard: Integer. The shard for the filename.
num_shards: An int Tensor for the number of shards.
Returns:
A string tensor."
11460,build,tensorflow/tensorflow/python/training/saver.py,419,method,"Builds save/restore graph nodes or runs save/restore in eager mode.
Args:
names_to_saveables: A dictionary mapping name to a Variable or
SaveableObject. Each name will be associated with the corresponding
variable in the checkpoint.
reshape: If True, allow restoring parameters from a checkpoint that where
the parameters have a different shape. This is only needed when you try
to restore from a Dist-Belief checkpoint, and only some times.
sharded: If True, shard the checkpoints, one per device that has Variable
nodes.
max_to_keep: Maximum number of checkpoints to keep. As new checkpoints
are created, old ones are deleted. If None or 0, no checkpoints are
deleted from the filesystem but only the last one is kept in the
`checkpoint` file. Presently the number is only roughly enforced. For
example in case of restarts more than max_to_keep checkpoints may be
kept.
keep_checkpoint_every_n_hours: How often checkpoints should be kept.
Defaults to 10,000 hours.
name: String. Optional name to use as a prefix when adding operations.
restore_sequentially: A Bool, which if true, causes restore of different
variables to happen sequentially within each device.
filename: If known at graph construction time, filename used for variable
loading/saving. If None, then the default name ""model"" will be used.
Returns:
A SaverDef proto.
Raises:
TypeError: If 'names_to_saveables' is not a dictionary mapping string
keys to variable Tensors.
ValueError: If any of the keys or values in 'names_to_saveables' is not
unique."
11461,BulkSaverBuilder,tensorflow/tensorflow/python/training/saver.py,567,class,SaverBuilder with support for bulk restoring multiple saveables.
11462,bulk_restore,tensorflow/tensorflow/python/training/saver.py,570,method,
11463,Saver,tensorflow/tensorflow/python/training/saver.py,614,class,"Saves and restores variables.
See [Variables](https://tensorflow.org/guide/variables)
for an overview of variables, saving and restoring.
The `Saver` class adds ops to save and restore variables to and from
*checkpoints*. It also provides convenience methods to run these ops.
Checkpoints are binary files in a proprietary format which map variable names
to tensor values. The best way to examine the contents of a checkpoint is to
load it using a `Saver`.
Savers can automatically number checkpoint filenames with a provided counter.
This lets you keep multiple checkpoints at different steps while training a
model. For example you can number the checkpoint filenames with the training
step number. To avoid filling up disks, savers manage checkpoint files
automatically. For example, they can keep only the N most recent files, or
one checkpoint for every N hours of training.
You number checkpoint filenames by passing a value to the optional
`global_step` argument to `save()`:
```python
saver.save(sess, 'my-model', global_step=0) ==> filename: 'my-model-0'
...
saver.save(sess, 'my-model', global_step=1000) ==> filename: 'my-model-1000'
```
Additionally, optional arguments to the `Saver()` constructor let you control
the proliferation of checkpoint files on disk:
* `max_to_keep` indicates the maximum number of recent checkpoint files to
keep. As new files are created, older files are deleted. If None or 0,
no checkpoints are deleted from the filesystem but only the last one is
kept in the `checkpoint` file. Defaults to 5 (that is, the 5 most recent
checkpoint files are kept.)
* `keep_checkpoint_every_n_hours`: In addition to keeping the most recent
`max_to_keep` checkpoint files, you might want to keep one checkpoint file
for every N hours of training. This can be useful if you want to later
analyze how a model progressed during a long training session. For
example, passing `keep_checkpoint_every_n_hours=2` ensures that you keep
one checkpoint file for every 2 hours of training. The default value of
10,000 hours effectively disables the feature.
Note that you still have to call the `save()` method to save the model.
Passing these arguments to the constructor will not save variables
automatically for you.
A training program that saves regularly looks like:
```python
...
# Create a saver.
saver = tf.compat.v1.train.Saver(...variables...)
# Launch the graph and train, saving the model every 1,000 steps.
sess = tf.compat.v1.Session()
for step in xrange(1000000):
sess.run(..training_op..)
if step % 1000 == 0:
# Append the step number to the checkpoint name:
saver.save(sess, 'my-model', global_step=step)
```
In addition to checkpoint files, savers keep a protocol buffer on disk with
the list of recent checkpoints. This is used to manage numbered checkpoint
files and by `latest_checkpoint()`, which makes it easy to discover the path
to the most recent checkpoint. That protocol buffer is stored in a file named
'checkpoint' next to the checkpoint files.
If you create several savers, you can specify a different filename for the
protocol buffer file in the call to `save()`."
11464,build,tensorflow/tensorflow/python/training/saver.py,845,method,
11465,as_saver_def,tensorflow/tensorflow/python/training/saver.py,975,method,"Generates a `SaverDef` representation of this saver.
Returns:
A `SaverDef` proto."
11466,to_proto,tensorflow/tensorflow/python/training/saver.py,983,method,"Converts this `Saver` to a `SaverDef` protocol buffer.
Args:
export_scope: Optional `string`. Name scope to remove.
Returns:
A `SaverDef` protocol buffer."
11467,from_proto,tensorflow/tensorflow/python/training/saver.py,1011,method,"Returns a `Saver` object created from `saver_def`.
Args:
saver_def: a `SaverDef` protocol buffer.
import_scope: Optional `string`. Name scope to use.
Returns:
A `Saver` built from saver_def."
11468,last_checkpoints,tensorflow/tensorflow/python/training/saver.py,1024,method,"List of not-yet-deleted checkpoint filenames.
You can pass any of the returned values to `restore()`.
Returns:
A list of checkpoint filenames, sorted from oldest to newest."
11469,set_last_checkpoints,tensorflow/tensorflow/python/training/saver.py,1034,method,"DEPRECATED: Use set_last_checkpoints_with_time.
Sets the list of old checkpoint filenames.
Args:
last_checkpoints: A list of checkpoint filenames.
Raises:
AssertionError: If last_checkpoints is not a list."
11470,set_last_checkpoints_with_time,tensorflow/tensorflow/python/training/saver.py,1051,method,"Sets the list of old checkpoint filenames and timestamps.
Args:
last_checkpoints_with_time: A list of tuples of checkpoint filenames and
timestamps.
Raises:
AssertionError: If last_checkpoints_with_time is not a list."
11471,recover_last_checkpoints,tensorflow/tensorflow/python/training/saver.py,1064,method,"Recovers the internal saver state after a crash.
This method is useful for recovering the ""self._last_checkpoints"" state.
Globs for the checkpoints pointed to by `checkpoint_paths`. If the files
exist, use their mtime as the checkpoint timestamp.
Args:
checkpoint_paths: a list of checkpoint paths."
11472,save,tensorflow/tensorflow/python/training/saver.py,1082,method,"Saves variables.
This method runs the ops added by the constructor for saving variables.
It requires a session in which the graph was launched. The variables to
save must also have been initialized.
The method returns the path prefix of the newly created checkpoint files.
This string can be passed directly to a call to `restore()`.
Args:
sess: A Session to use to save the variables.
save_path: String. Prefix of filenames created for the checkpoint.
global_step: If provided the global step number is appended to `save_path`
to create the checkpoint filenames. The optional argument can be a
`Tensor`, a `Tensor` name or an integer.
latest_filename: Optional name for the protocol buffer file that will
contains the list of most recent checkpoints. That file, kept in the
same directory as the checkpoint files, is automatically managed by the
saver to keep track of recent checkpoints. Defaults to 'checkpoint'.
meta_graph_suffix: Suffix for `MetaGraphDef` file. Defaults to 'meta'.
write_meta_graph: `Boolean` indicating whether or not to write the meta
graph file.
write_state: `Boolean` indicating whether or not to write the
`CheckpointStateProto`.
strip_default_attrs: Boolean. If `True`, default-valued attributes will be
removed from the NodeDefs. For a detailed guide, see
[Stripping Default-Valued
Attributes](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/saved_model/README.md#stripping-default-valued-attributes).
save_debug_info: If `True`, save the GraphDebugInfo to a separate file,
which in the same directory of save_path and with `_debug` added before
the file extension. This is only enabled when `write_meta_graph` is
`True`
Returns:
A string: path prefix used for the checkpoint files. If the saver is
sharded, this string ends with: '-?????-of-nnnnn' where 'nnnnn'
is the number of shards created.
If the saver is empty, returns None.
Raises:
TypeError: If `sess` is not a `Session`.
ValueError: If `latest_filename` contains path components, or if it
collides with `save_path`.
RuntimeError: If save and restore ops weren't built."
11473,export_meta_graph,tensorflow/tensorflow/python/training/saver.py,1219,method,"Writes `MetaGraphDef` to save_path/filename.
Args:
filename: Optional meta_graph filename including the path.
collection_list: List of string keys to collect.
as_text: If `True`, writes the meta_graph as an ASCII proto.
export_scope: Optional `string`. Name scope to remove.
clear_devices: Whether or not to clear the device field for an `Operation`
or `Tensor` during export.
clear_extraneous_savers: Remove any Saver-related information from the
graph (both Save/Restore ops and SaverDefs) that are not associated with
this Saver.
strip_default_attrs: Boolean. If `True`, default-valued attributes will be
removed from the NodeDefs. For a detailed guide, see
[Stripping Default-Valued
Attributes](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/saved_model/README.md#stripping-default-valued-attributes).
save_debug_info: If `True`, save the GraphDebugInfo to a separate file,
which in the same directory of filename and with `_debug` added before
the file extension.
Returns:
A `MetaGraphDef` proto."
11474,restore,tensorflow/tensorflow/python/training/saver.py,1265,method,"Restores previously saved variables.
This method runs the ops added by the constructor for restoring variables.
It requires a session in which the graph was launched. The variables to
restore do not have to have been initialized, as restoring is itself a way
to initialize variables.
The `save_path` argument is typically a value previously returned from a
`save()` call, or a call to `latest_checkpoint()`.
Args:
sess: A `Session` to use to restore the parameters. None in eager mode.
save_path: Path where parameters were previously saved.
Raises:
ValueError: If save_path is None or not a valid checkpoint."
11475,import_meta_graph,tensorflow/tensorflow/python/training/saver.py,1351,function,"Recreates a Graph saved in a `MetaGraphDef` proto.
This function takes a `MetaGraphDef` protocol buffer as input. If
the argument is a file containing a `MetaGraphDef` protocol buffer ,
it constructs a protocol buffer from the file content. The function
then adds all the nodes from the `graph_def` field to the
current graph, recreates all the collections, and returns a saver
constructed from the `saver_def` field.
In combination with `export_meta_graph()`, this function can be used to
* Serialize a graph along with other Python objects such as `QueueRunner`,
`Variable` into a `MetaGraphDef`.
* Restart training from a saved graph and checkpoints.
* Run inference from a saved graph and checkpoints.
```Python
...
# Create a saver.
saver = tf.compat.v1.train.Saver(...variables...)
# Remember the training_op we want to run by adding it to a collection.
tf.compat.v1.add_to_collection('train_op', train_op)
sess = tf.compat.v1.Session()
for step in xrange(1000000):
sess.run(train_op)
if step % 1000 == 0:
# Saves checkpoint, which by default also exports a meta_graph
# named 'my-model-global_step.meta'.
saver.save(sess, 'my-model', global_step=step)
```
Later we can continue training from this saved `meta_graph` without building
the model from scratch.
```Python
with tf.Session() as sess:
new_saver =
tf.train.import_meta_graph('my-save-dir/my-model-10000.meta')
new_saver.restore(sess, 'my-save-dir/my-model-10000')
# tf.get_collection() returns a list. In this example we only want
# the first one.
train_op = tf.get_collection('train_op')[0]
for step in xrange(1000000):
sess.run(train_op)
```
NOTE: Restarting training from saved `meta_graph` only works if the
device assignments have not changed.
Example:
Variables, placeholders, and independent operations can also be stored, as
shown in the following example.
```Python
# Saving contents and operations.
v1 = tf.placeholder(tf.float32, name=""v1"")
v2 = tf.placeholder(tf.float32, name=""v2"")
v3 = tf.math.multiply(v1, v2)
vx = tf.Variable(10.0, name=""vx"")
v4 = tf.add(v3, vx, name=""v4"")
saver = tf.train.Saver([vx])
sess = tf.Session()
sess.run(tf.global_variables_initializer())
sess.run(vx.assign(tf.add(vx, vx)))
result = sess.run(v4, feed_dict={v1:12.0, v2:3.3})
print(result)
saver.save(sess, ""./model_ex1"")
```
Later this model can be restored and contents loaded.
```Python
# Restoring variables and running operations.
saver = tf.train.import_meta_graph(""./model_ex1.meta"")
sess = tf.Session()
saver.restore(sess, ""./model_ex1"")
result = sess.run(""v4:0"", feed_dict={""v1:0"": 12.0, ""v2:0"": 3.3})
print(result)
```
Args:
meta_graph_or_file: `MetaGraphDef` protocol buffer or filename (including
the path) containing a `MetaGraphDef`.
clear_devices: Whether or not to clear the device field for an `Operation`
or `Tensor` during import.
import_scope: Optional `string`. Name scope to add. Only used when
initializing from protocol buffer.
**kwargs: Optional keyed arguments.
Returns:
A saver constructed from `saver_def` in `MetaGraphDef` or None.
A None value is returned if no variables exist in the `MetaGraphDef`
(i.e., there are no variables to restore).
Raises:
RuntimeError: If called with eager execution enabled.
@compatibility(eager)
Exporting/importing meta graphs is not supported. No graph exists when eager
execution is enabled.
@end_compatibility"
11476,export_meta_graph,tensorflow/tensorflow/python/training/saver.py,1518,function,"Returns `MetaGraphDef` proto.
Optionally writes it to filename.
This function exports the graph, saver, and collection objects into
`MetaGraphDef` protocol buffer with the intention of it being imported
at a later time or location to restart training, run inference, or be
a subgraph.
Args:
filename: Optional filename including the path for writing the generated
`MetaGraphDef` protocol buffer.
meta_info_def: `MetaInfoDef` protocol buffer.
graph_def: `GraphDef` protocol buffer.
saver_def: `SaverDef` protocol buffer.
collection_list: List of string keys to collect.
as_text: If `True`, writes the `MetaGraphDef` as an ASCII proto.
graph: The `Graph` to export. If `None`, use the default graph.
export_scope: Optional `string`. Name scope under which to extract the
subgraph. The scope name will be striped from the node definitions for
easy import later into new name scopes. If `None`, the whole graph is
exported. graph_def and export_scope cannot both be specified.
clear_devices: Whether or not to clear the device field for an `Operation`
or `Tensor` during export.
clear_extraneous_savers: Remove any Saver-related information from the graph
(both Save/Restore ops and SaverDefs) that are not associated with the
provided SaverDef.
strip_default_attrs: Boolean. If `True`, default-valued attributes will be
removed from the NodeDefs. For a detailed guide, see
[Stripping Default-Valued Attributes](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/saved_model/README.md#stripping-default-valued-attributes).
save_debug_info: If `True`, save the GraphDebugInfo to a separate file,
which in the same directory of filename and with `_debug` added before the
file extend.
**kwargs: Optional keyed arguments.
Returns:
A `MetaGraphDef` proto.
Raises:
ValueError: When the `GraphDef` is larger than 2GB.
RuntimeError: If called with eager execution enabled.
@compatibility(eager)
Exporting/importing meta graphs is not supported unless both `graph_def` and
`graph` are provided. No graph exists when eager execution is enabled.
@end_compatibility"
11477,object_graph_key_mapping,tensorflow/tensorflow/python/training/saver.py,1617,function,"Return name to key mappings from the checkpoint.
Args:
checkpoint_path: string, path to object-based checkpoint
Returns:
Dictionary mapping tensor names to checkpoint keys."
11478,saver_from_object_based_checkpoint,tensorflow/tensorflow/python/training/saver.py,1637,function,"Return a `Saver` which reads from an object-based checkpoint.
This function validates that all variables in the variables list are remapped
in the object-based checkpoint (or `names_to_keys` dict if provided). A
saver will be created with the list of remapped variables.
The `cached_saver` argument allows the user to pass in a previously created
saver, so multiple `saver.restore()` calls don't pollute the graph when graph
building. This assumes that keys are consistent, meaning that the
1) `checkpoint_path` checkpoint, and
2) checkpoint used to create the `cached_saver`
are the same type of object-based checkpoint. If this argument is set, this
function will simply validate that all variables have been remapped by the
checkpoint at `checkpoint_path`.
Note that in general, `tf.train.Checkpoint` should be used to restore/save an
object-based checkpoint.
Args:
checkpoint_path: string, path to object-based checkpoint
var_list: list of `Variables` that appear in the checkpoint. If `None`,
`var_list` will be set to all saveable objects.
builder: a `BaseSaverBuilder` instance. If `None`, a new `BulkSaverBuilder`
will be created.
names_to_keys: dict mapping string tensor names to checkpoint keys. If
`None`, this dict will be generated from the checkpoint file.
cached_saver: Cached `Saver` object with remapped variables.
Returns:
`Saver` with remapped variables for reading from an object-based checkpoint.
Raises:
ValueError if the checkpoint provided is not an object-based checkpoint.
NotFoundError: If one of the variables in `var_list` can not be found in the
checkpoint. This could mean the checkpoint or `names_to_keys` mapping is
missing the variable."
11479,SaveRestoreWithVariableNameMap,tensorflow/tensorflow/python/training/saver_test.py,1727,class,
11480,CheckpointedOp,tensorflow/tensorflow/python/training/saver_test_utils.py,28,class,"Op with a custom checkpointing implementation.
Defined as part of the test because the MutableHashTable Python code is
currently in contrib."
11481,name,tensorflow/tensorflow/python/training/saver_test_utils.py,49,method,
11482,saveable,tensorflow/tensorflow/python/training/saver_test_utils.py,53,method,
11483,insert,tensorflow/tensorflow/python/training/saver_test_utils.py,59,method,
11484,lookup,tensorflow/tensorflow/python/training/saver_test_utils.py,62,method,
11485,keys,tensorflow/tensorflow/python/training/saver_test_utils.py,65,method,
11486,values,tensorflow/tensorflow/python/training/saver_test_utils.py,68,method,
11487,restore,tensorflow/tensorflow/python/training/saver_test_utils.py,88,method,
11488,Server,tensorflow/tensorflow/python/training/server_lib.py,100,class,"An in-process TensorFlow server, for use in distributed training.
A `tf.distribute.Server` instance encapsulates a set of devices and a
`tf.compat.v1.Session` target that
can participate in distributed training. A server belongs to a
cluster (specified by a `tf.train.ClusterSpec`), and
corresponds to a particular task in a named job. The server can
communicate with any other server in the same cluster."
11489,start,tensorflow/tensorflow/python/training/server_lib.py,171,method,"Starts this server.
Raises:
tf.errors.OpError: Or one of its subclasses if an error occurs while
starting the TensorFlow server."
11490,join,tensorflow/tensorflow/python/training/server_lib.py,180,method,"Blocks until the server has shut down.
This method currently blocks forever.
Raises:
tf.errors.OpError: Or one of its subclasses if an error occurs while
joining the TensorFlow server."
11491,server_def,tensorflow/tensorflow/python/training/server_lib.py,192,method,"Returns the `tf.train.ServerDef` for this server.
Returns:
A `tf.train.ServerDef` protocol buffer that describes the configuration
of this server."
11492,target,tensorflow/tensorflow/python/training/server_lib.py,202,method,"Returns the target for a `tf.compat.v1.Session` to connect to this server.
To create a
`tf.compat.v1.Session` that
connects to this server, use the following snippet:
```python
server = tf.distribute.Server(...)
with tf.compat.v1.Session(server.target):
# ...
```
Returns:
A string containing a session target for this server."
11493,create_local_server,tensorflow/tensorflow/python/training/server_lib.py,221,method,"Creates a new single-process cluster running on the local host.
This method is a convenience wrapper for creating a
`tf.distribute.Server` with a `tf.train.ServerDef` that specifies a
single-process cluster containing a single task in a job called
`""local""`.
Args:
config: (Options.) A `tf.compat.v1.ConfigProto` that specifies default
configuration options for all sessions that run on this server.
start: (Optional.) Boolean, indicating whether to start the server after
creating it. Defaults to `True`.
Returns:
A local `tf.distribute.Server`."
11494,ClusterSpec,tensorflow/tensorflow/python/training/server_lib.py,247,class,"Represents a cluster as a set of ""tasks"", organized into ""jobs"".
A `tf.train.ClusterSpec` represents the set of processes that
participate in a distributed TensorFlow computation. Every
`tf.distribute.Server` is constructed in a particular cluster.
To create a cluster with two jobs and five tasks, you specify the
mapping from job names to lists of network addresses (typically
hostname-port pairs).
```python
cluster = tf.train.ClusterSpec({""worker"": [""worker0.example.com:2222"",
""worker1.example.com:2222"",
""worker2.example.com:2222""],
""ps"": [""ps0.example.com:2222"",
""ps1.example.com:2222""]})
```
Each job may also be specified as a sparse mapping from task indices
to network addresses. This enables a server to be configured without
needing to know the identity of (for example) all other worker
tasks:
```python
cluster = tf.train.ClusterSpec({""worker"": {1: ""worker1.example.com:2222""},
""ps"": [""ps0.example.com:2222"",
""ps1.example.com:2222""]})
```"
11495,as_dict,tensorflow/tensorflow/python/training/server_lib.py,341,method,"Returns a dictionary from job names to their tasks.
For each job, if the task index space is dense, the corresponding
value will be a list of network addresses; otherwise it will be a
dictionary mapping (sparse) task indices to the corresponding
addresses.
Returns:
A dictionary mapping job names to lists or dictionaries
describing the tasks in those jobs."
11496,as_cluster_def,tensorflow/tensorflow/python/training/server_lib.py,368,method,Returns a `tf.train.ClusterDef` protocol buffer based on this cluster.
11497,jobs,tensorflow/tensorflow/python/training/server_lib.py,373,method,"Returns a list of job names in this cluster.
Returns:
A list of strings, corresponding to the names of jobs in this cluster."
11498,num_tasks,tensorflow/tensorflow/python/training/server_lib.py,381,method,"Returns the number of tasks defined in the given job.
Args:
job_name: The string name of a job in this cluster.
Returns:
The number of tasks defined in the given job.
Raises:
ValueError: If `job_name` does not name a job in this cluster."
11499,task_indices,tensorflow/tensorflow/python/training/server_lib.py,399,method,"Returns a list of valid task indices in the given job.
Args:
job_name: The string name of a job in this cluster.
Returns:
A list of valid task indices in the given job.
Raises:
ValueError: If `job_name` does not name a job in this cluster,
or no task with index `task_index` is defined in that job."
11500,task_address,tensorflow/tensorflow/python/training/server_lib.py,418,method,"Returns the address of the given task in the given job.
Args:
job_name: The string name of a job in this cluster.
task_index: A non-negative integer.
Returns:
The address of the given task in the given job.
Raises:
ValueError: If `job_name` does not name a job in this cluster,
or no task with index `task_index` is defined in that job."
11501,job_tasks,tensorflow/tensorflow/python/training/server_lib.py,442,method,"Returns a mapping from task ID to address in the given job.
NOTE: For backwards compatibility, this method returns a list. If
the given job was defined with a sparse set of task indices, the
length of this list may not reflect the number of tasks defined in
this job. Use the `tf.train.ClusterSpec.num_tasks` method
to find the number of tasks defined in a particular job.
Args:
job_name: The string name of a job in this cluster.
Returns:
A list of task addresses, where the index in the list
corresponds to the task index of each task. The list may contain
`None` if the job was defined with a sparse set of task indices.
Raises:
ValueError: If `job_name` does not name a job in this cluster."
11502,ClusterDeviceFilters,tensorflow/tensorflow/python/training/server_lib.py,500,class,"Represent a collection of device filters for the remote workers in cluster.
NOTE: this is an experimental API and subject to changes.
Set device filters for selective jobs and tasks. For each remote worker, the
device filters are a list of strings. When any filters are present, the remote
worker will ignore all devices which do not match any of its filters. Each
filter can be partially specified, e.g. ""/job:ps"", ""/job:worker/replica:3"",
etc. Note that a device is always visible to the worker it is located on.
For example, to set the device filters for a parameter server cluster:
```python
cdf = tf.config.experimental.ClusterDeviceFilters()
for i in range(num_workers):
cdf.set_device_filters('worker', i, ['/job:ps'])
for i in range(num_ps):
cdf.set_device_filters('ps', i, ['/job:worker'])
tf.config.experimental_connect_to_cluster(cluster_def,
cluster_device_filters=cdf)
```
The device filters can be partically specified. For remote tasks that do not
have device filters specified, all devices will be visible to them."
11503,set_device_filters,tensorflow/tensorflow/python/training/server_lib.py,537,method,Set the device filters for given job name and task id.
11504,SessionManager,tensorflow/tensorflow/python/training/session_manager.py,51,class,"Training helper that restores from checkpoint and creates session.
This class is a small wrapper that takes care of session creation and
checkpoint recovery. It also provides functions that to facilitate
coordination among multiple training threads or processes.
* Checkpointing trained variables as the training progresses.
* Initializing variables on startup, restoring them from the most recent
checkpoint after a crash, or wait for checkpoints to become available.
### Usage:
```python
with tf.Graph().as_default():
...add operations to the graph...
# Create a SessionManager that will checkpoint the model in '/tmp/mydir'.
sm = SessionManager()
sess = sm.prepare_session(master, init_op, saver, checkpoint_dir)
# Use the session to train the graph.
while True:
sess.run(<my_train_op>)
```
`prepare_session()` initializes or restores a model. It requires `init_op`
and `saver` as an argument.
A second process could wait for the model to be ready by doing the following:
```python
with tf.Graph().as_default():
...add operations to the graph...
# Create a SessionManager that will wait for the model to become ready.
sm = SessionManager()
sess = sm.wait_for_session(master)
# Use the session to train the graph.
while True:
sess.run(<my_train_op>)
```
`wait_for_session()` waits for a model to be initialized by other processes."
11505,prepare_session,tensorflow/tensorflow/python/training/session_manager.py,229,method,"Creates a `Session`. Makes sure the model is ready to be used.
Creates a `Session` on 'master'. If a `saver` object is passed in, and
`checkpoint_dir` points to a directory containing valid checkpoint
files, then it will try to recover the model from checkpoint. If
no checkpoint files are available, and `wait_for_checkpoint` is
`True`, then the process would check every `recovery_wait_secs`,
up to `max_wait_secs`, for recovery to succeed.
If the model cannot be recovered successfully then it is initialized by
running the `init_op` and calling `init_fn` if they are provided.
The `local_init_op` is also run after init_op and init_fn, regardless of
whether the model was recovered successfully, but only if
`ready_for_local_init_op` passes.
If the model is recovered from a checkpoint it is assumed that all
global variables have been initialized, in particular neither `init_op`
nor `init_fn` will be executed.
It is an error if the model cannot be recovered and no `init_op`
or `init_fn` or `local_init_op` are passed.
Args:
master: `String` representation of the TensorFlow master to use.
init_op: Optional `Operation` used to initialize the model.
saver: A `Saver` object used to restore a model.
checkpoint_dir: Path to the checkpoint files. The latest checkpoint in the
dir will be used to restore.
checkpoint_filename_with_path: Full file name path to the checkpoint file.
wait_for_checkpoint: Whether to wait for checkpoint to become available.
max_wait_secs: Maximum time to wait for checkpoints to become available.
config: Optional `ConfigProto` proto used to configure the session.
init_feed_dict: Optional dictionary that maps `Tensor` objects to feed
values. This feed dictionary is passed to the session `run()` call when
running the init op.
init_fn: Optional callable used to initialize the model. Called after the
optional `init_op` is called. The callable must accept one argument,
the session being initialized.
Returns:
A `Session` object that can be used to drive the model.
Raises:
RuntimeError: If the model cannot be initialized or recovered.
ValueError: If both checkpoint_dir and checkpoint_filename_with_path are
set."
11506,recover_session,tensorflow/tensorflow/python/training/session_manager.py,321,method,"Creates a `Session`, recovering if possible.
Creates a new session on 'master'. If the session is not initialized
and can be recovered from a checkpoint, recover it.
Args:
master: `String` representation of the TensorFlow master to use.
saver: A `Saver` object used to restore a model.
checkpoint_dir: Path to the checkpoint files. The latest checkpoint in the
dir will be used to restore.
checkpoint_filename_with_path: Full file name path to the checkpoint file.
wait_for_checkpoint: Whether to wait for checkpoint to become available.
max_wait_secs: Maximum time to wait for checkpoints to become available.
config: Optional `ConfigProto` proto used to configure the session.
Returns:
A pair (sess, initialized) where 'initialized' is `True` if
the session could be recovered and initialized, `False` otherwise.
Raises:
ValueError: If both checkpoint_dir and checkpoint_filename_with_path are
set."
11507,wait_for_session,tensorflow/tensorflow/python/training/session_manager.py,385,method,"Creates a new `Session` and waits for model to be ready.
Creates a new `Session` on 'master'. Waits for the model to be
initialized or recovered from a checkpoint. It's expected that
another thread or process will make the model ready, and that this
is intended to be used by threads/processes that participate in a
distributed training configuration where a different thread/process
is responsible for initializing or recovering the model being trained.
NB: The amount of time this method waits for the session is bounded
by max_wait_secs. By default, this function will wait indefinitely.
Args:
master: `String` representation of the TensorFlow master to use.
config: Optional ConfigProto proto used to configure the session.
max_wait_secs: Maximum time to wait for the session to become available.
Returns:
A `Session`. May be None if the operation exceeds the timeout
specified by config.operation_timeout_in_ms.
Raises:
tf.DeadlineExceededError: if the session is not available after
max_wait_secs."
11508,SessionRunHook,tensorflow/tensorflow/python/training/session_run_hook.py,98,class,Hook to extend calls to MonitoredSession.run().
11509,begin,tensorflow/tensorflow/python/training/session_run_hook.py,101,method,"Called once before using the session.
When called, the default graph is the one that will be launched in the
session. The hook can modify the graph by adding new operations to it.
After the `begin()` call the graph will be finalized and the other callbacks
can not modify the graph anymore. Second call of `begin()` on the same
graph, should not change the graph."
11510,after_create_session,tensorflow/tensorflow/python/training/session_run_hook.py,112,method,"Called when new TensorFlow session is created.
This is called to signal the hooks that a new session has been created. This
has two essential differences with the situation in which `begin` is called:
* When this is called, the graph is finalized and ops can no longer be added
to the graph.
* This method will also be called as a result of recovering a wrapped
session, not only at the beginning of the overall session.
Args:
session: A TensorFlow Session that has been created.
coord: A Coordinator object which keeps track of all threads."
11511,before_run,tensorflow/tensorflow/python/training/session_run_hook.py,129,method,"Called before each call to run().
You can return from this call a `SessionRunArgs` object indicating ops or
tensors to add to the upcoming `run()` call. These ops/tensors will be run
together with the ops/tensors originally passed to the original run() call.
The run args you return can also contain feeds to be added to the run()
call.
The `run_context` argument is a `SessionRunContext` that provides
information about the upcoming `run()` call: the originally requested
op/tensors, the TensorFlow Session.
At this point graph is finalized and you can not add ops.
Args:
run_context: A `SessionRunContext` object.
Returns:
None or a `SessionRunArgs` object."
11512,after_run,tensorflow/tensorflow/python/training/session_run_hook.py,152,method,"Called after each call to run().
The `run_values` argument contains results of requested ops/tensors by
`before_run()`.
The `run_context` argument is the same one send to `before_run` call.
`run_context.request_stop()` can be called to stop the iteration.
If `session.run()` raises any exceptions then `after_run()` is not called.
Args:
run_context: A `SessionRunContext` object.
run_values: A SessionRunValues object."
11513,end,tensorflow/tensorflow/python/training/session_run_hook.py,171,method,"Called at the end of session.
The `session` argument can be used in case the hook wants to run final ops,
such as saving a last checkpoint.
If `session.run()` raises exception other than OutOfRangeError or
StopIteration then `end()` is not called.
Note the difference between `end()` and `after_run()` behavior when
`session.run()` raises OutOfRangeError or StopIteration. In that case
`end()` is called but `after_run()` is not called.
Args:
session: A TensorFlow Session that will be soon closed."
11514,SessionRunArgs,tensorflow/tensorflow/python/training/session_run_hook.py,190,class,"Represents arguments to be added to a `Session.run()` call.
Args:
fetches: Exactly like the 'fetches' argument to Session.Run().
Can be a single tensor or op, a list of 'fetches' or a dictionary
of fetches. For example:
fetches = global_step_tensor
fetches = [train_op, summary_op, global_step_tensor]
fetches = {'step': global_step_tensor, 'summ': summary_op}
Note that this can recurse as expected:
fetches = {'step': global_step_tensor,
'ops': [train_op, check_nan_op]}
feed_dict: Exactly like the `feed_dict` argument to `Session.Run()`
options: Exactly like the `options` argument to `Session.run()`, i.e., a
config_pb2.RunOptions proto."
11515,SessionRunContext,tensorflow/tensorflow/python/training/session_run_hook.py,215,class,"Provides information about the `session.run()` call being made.
Provides information about original request to `Session.Run()` function.
SessionRunHook objects can stop the loop by calling `request_stop()` of
`run_context`. In the future we may use this object to add more information
about run without changing the Hook API."
11516,original_args,tensorflow/tensorflow/python/training/session_run_hook.py,231,method,"A `SessionRunArgs` object holding the original arguments of `run()`.
If user called `MonitoredSession.run(fetches=a, feed_dict=b)`, then this
field is equal to SessionRunArgs(a, b).
Returns:
A `SessionRunArgs` object"
11517,session,tensorflow/tensorflow/python/training/session_run_hook.py,243,method,A TensorFlow session object which will execute the `run`.
11518,stop_requested,tensorflow/tensorflow/python/training/session_run_hook.py,248,method,"Returns whether a stop is requested or not.
If true, `MonitoredSession` stops iterations.
Returns:
A `bool`"
11519,request_stop,tensorflow/tensorflow/python/training/session_run_hook.py,257,method,"Sets stop requested field.
Hooks can use this function to request stop of iterations.
`MonitoredSession` checks whether this is called or not."
11520,SessionRunValues,tensorflow/tensorflow/python/training/session_run_hook.py,267,class,"Contains the results of `Session.run()`.
In the future we may use this object to add more information about result of
run without changing the Hook API.
Args:
results: The return values from `Session.run()` corresponding to the fetches
attribute returned in the RunArgs. Note that this has the same shape as
the RunArgs fetches. For example:
fetches = global_step_tensor
=> results = nparray(int)
fetches = [train_op, summary_op, global_step_tensor]
=> results = [None, nparray(string), nparray(int)]
fetches = {'step': global_step_tensor, 'summ': summary_op}
=> results = {'step': nparray(int), 'summ': nparray(string)}
options: `RunOptions` from the `Session.run()` call.
run_metadata: `RunMetadata` from the `Session.run()` call."
11521,create_slot,tensorflow/tensorflow/python/training/slot_creator.py,104,function,"Create a slot initialized to the given value.
The type of the slot is determined by the given value.
Args:
primary: The primary `Variable` or `Tensor`.
val: A `Tensor` specifying the initial value of the slot.
name: Name to use for the slot variable.
colocate_with_primary: Boolean. If True the slot is located
on the same device as `primary`.
Returns:
A `Variable` object."
11522,create_slot_with_initializer,tensorflow/tensorflow/python/training/slot_creator.py,138,function,"Creates a slot initialized using an `Initializer`.
The type of the slot is determined by the given value.
Args:
primary: The primary `Variable` or `Tensor`.
initializer: An `Initializer`. The initial value of the slot.
shape: Shape of the initial value of the slot.
dtype: Type of the value of the slot.
name: Name to use for the slot variable.
colocate_with_primary: Boolean. If True the slot is located
on the same device as `primary`.
Returns:
A `Variable` object."
11523,create_zeros_slot,tensorflow/tensorflow/python/training/slot_creator.py,177,function,"Create a slot initialized to 0 with same shape as the primary object.
Args:
primary: The primary `Variable` or `Tensor`.
name: Name to use for the slot variable.
dtype: Type of the slot variable. Defaults to the type of `primary`.
colocate_with_primary: Boolean. If True the slot is located
on the same device as `primary`.
Returns:
A `Variable` object."
11524,SummaryWriter,tensorflow/tensorflow/python/training/summary_io.py,29,class,
11525,Supervisor,tensorflow/tensorflow/python/training/supervisor.py,44,class,"A training helper that checkpoints models and computes summaries.
This class is deprecated. Please use
`tf.compat.v1.train.MonitoredTrainingSession` instead.
The Supervisor is a small wrapper around a `Coordinator`, a `Saver`,
and a `SessionManager` that takes care of common needs of TensorFlow
training programs.
#### Use for a single program
```python
with tf.Graph().as_default():
...add operations to the graph...
# Create a Supervisor that will checkpoint the model in '/tmp/mydir'.
sv = Supervisor(logdir='/tmp/mydir')
# Get a TensorFlow session managed by the supervisor.
with sv.managed_session(FLAGS.master) as sess:
# Use the session to train the graph.
while not sv.should_stop():
sess.run(<my_train_op>)
```
Within the `with sv.managed_session()` block all variables in the graph have
been initialized. In addition, a few services have been started to
checkpoint the model and add summaries to the event log.
If the program crashes and is restarted, the managed session automatically
reinitialize variables from the most recent checkpoint.
The supervisor is notified of any exception raised by one of the services.
After an exception is raised, `should_stop()` returns `True`. In that case
the training loop should also stop. This is why the training loop has to
check for `sv.should_stop()`.
Exceptions that indicate that the training inputs have been exhausted,
`tf.errors.OutOfRangeError`, also cause `sv.should_stop()` to return `True`
but are not re-raised from the `with` block: they indicate a normal
termination.
#### Use for multiple replicas
To train with replicas you deploy the same program in a `Cluster`.
One of the tasks must be identified as the *chief*: the task that handles
initialization, checkpoints, summaries, and recovery. The other tasks
depend on the *chief* for these services.
The only change you have to do to the single program code is to indicate
if the program is running as the *chief*.
```python
# Choose a task as the chief. This could be based on server_def.task_index,
# or job_def.name, or job_def.tasks. It's entirely up to the end user.
# But there can be only one *chief*.
is_chief = (server_def.task_index == 0)
server = tf.distribute.Server(server_def)
with tf.Graph().as_default():
...add operations to the graph...
# Create a Supervisor that uses log directory on a shared file system.
# Indicate if you are the 'chief'
sv = Supervisor(logdir='/shared_directory/...', is_chief=is_chief)
# Get a Session in a TensorFlow server on the cluster.
with sv.managed_session(server.target) as sess:
# Use the session to train the graph.
while not sv.should_stop():
sess.run(<my_train_op>)
```
In the *chief* task, the `Supervisor` works exactly as in the first example
above. In the other tasks `sv.managed_session()` waits for the Model to have
been initialized before returning a session to the training code. The
non-chief tasks depend on the chief task for initializing the model.
If one of the tasks crashes and restarts, `managed_session()`
checks if the Model is initialized. If yes, it just creates a session and
returns it to the training code that proceeds normally. If the model needs
to be initialized, the chief task takes care of reinitializing it; the other
tasks just wait for the model to have been initialized.
NOTE: This modified program still works fine as a single program.
The single program marks itself as the chief.
#### What `master` string to use
Whether you are running on your machine or in the cluster you can use the
following values for the --master flag:
* Specifying `''` requests an in-process session that does not use RPC.
* Specifying `'local'` requests a session that uses the RPC-based
""Master interface"" to run TensorFlow programs. See
`tf.train.Server.create_local_server` for
details.
* Specifying `'grpc://hostname:port'` requests a session that uses
the RPC interface to a specific host, and also allows the in-process
master to access remote tensorflow workers. Often, it is
appropriate to pass `server.target` (for some `tf.distribute.Server`
named `server).
#### Advanced use
##### Launching additional services
`managed_session()` launches the Checkpoint and Summary services (threads).
If you need more services to run you can simply launch them in the block
controlled by `managed_session()`.
Example: Start a thread to print losses. We want this thread to run
every 60 seconds, so we launch it with `sv.loop()`.
```python
...
sv = Supervisor(logdir='/tmp/mydir')
with sv.managed_session(FLAGS.master) as sess:
sv.loop(60, print_loss, (sess, ))
while not sv.should_stop():
sess.run(my_train_op)
```
##### Launching fewer services
`managed_session()` launches the ""summary"" and ""checkpoint"" threads which use
either the optionally `summary_op` and `saver` passed to the constructor, or
default ones created automatically by the supervisor. If you want to run
your own summary and checkpointing logic, disable these services by passing
`None` to the `summary_op` and `saver` parameters.
Example: Create summaries manually every 100 steps in the chief.
```python
# Create a Supervisor with no automatic summaries.
sv = Supervisor(logdir='/tmp/mydir', is_chief=is_chief, summary_op=None)
# As summary_op was None, managed_session() does not start the
# summary thread.
with sv.managed_session(FLAGS.master) as sess:
for step in xrange(1000000):
if sv.should_stop():
break
if is_chief and step % 100 == 0:
# Create the summary every 100 chief steps.
sv.summary_computed(sess, sess.run(my_summary_op))
else:
# Train normally
sess.run(my_train_op)
```
##### Custom model initialization
`managed_session()` only supports initializing the model by running an
`init_op` or restoring from the latest checkpoint. If you have special
initialization needs, see how to specify a `local_init_op` when creating the
supervisor. You can also use the `SessionManager` directly to create a
session and check if it could be initialized automatically."
11526,is_chief,tensorflow/tensorflow/python/training/supervisor.py,503,method,"Return True if this is a chief supervisor.
Returns:
A bool."
11527,session_manager,tensorflow/tensorflow/python/training/supervisor.py,512,method,"Return the SessionManager used by the Supervisor.
Returns:
A SessionManager object."
11528,coord,tensorflow/tensorflow/python/training/supervisor.py,521,method,"Return the Coordinator used by the Supervisor.
The Coordinator can be useful if you want to run multiple threads
during your training.
Returns:
A Coordinator object."
11529,init_op,tensorflow/tensorflow/python/training/supervisor.py,533,method,"Return the Init Op used by the supervisor.
Returns:
An Op or `None`."
11530,init_feed_dict,tensorflow/tensorflow/python/training/supervisor.py,542,method,"Return the feed dictionary used when evaluating the `init_op`.
Returns:
A feed dictionary or `None`."
11531,ready_op,tensorflow/tensorflow/python/training/supervisor.py,551,method,"Return the Ready Op used by the supervisor.
Returns:
An Op or `None`."
11532,ready_for_local_init_op,tensorflow/tensorflow/python/training/supervisor.py,560,method,
11533,summary_writer,tensorflow/tensorflow/python/training/supervisor.py,564,method,"Return the SummaryWriter used by the chief supervisor.
Returns:
A SummaryWriter."
11534,summary_op,tensorflow/tensorflow/python/training/supervisor.py,573,method,"Return the Summary Tensor used by the chief supervisor.
Returns:
A string Tensor for the summary or `None`."
11535,save_summaries_secs,tensorflow/tensorflow/python/training/supervisor.py,582,method,"Return the delay between summary computations.
Returns:
A timestamp."
11536,global_step,tensorflow/tensorflow/python/training/supervisor.py,591,method,"Return the global_step Tensor used by the supervisor.
Returns:
An integer Tensor for the global_step."
11537,saver,tensorflow/tensorflow/python/training/supervisor.py,600,method,"Return the Saver used by the supervisor.
Returns:
A Saver object."
11538,save_model_secs,tensorflow/tensorflow/python/training/supervisor.py,609,method,"Return the delay between checkpoints.
Returns:
A timestamp."
11539,save_path,tensorflow/tensorflow/python/training/supervisor.py,618,method,"Return the save path used by the supervisor.
Returns:
A string."
11540,start_standard_services,tensorflow/tensorflow/python/training/supervisor.py,638,method,"Start the standard services for 'sess'.
This starts services in the background. The services started depend
on the parameters to the constructor and may include:
- A Summary thread computing summaries every save_summaries_secs.
- A Checkpoint thread saving the model every save_model_secs.
- A StepCounter thread measure step time.
Args:
sess: A Session.
Returns:
A list of threads that are running the standard services. You can use
the Supervisor's Coordinator to join these threads with:
sv.coord.Join(<list of threads>)
Raises:
RuntimeError: If called with a non-chief Supervisor.
ValueError: If not `logdir` was passed to the constructor as the
services need a log directory."
11541,prepare_or_wait_for_session,tensorflow/tensorflow/python/training/supervisor.py,690,method,"Make sure the model is ready to be used.
Create a session on 'master', recovering or initializing the model as
needed, or wait for a session to be ready. If running as the chief
and `start_standard_service` is set to True, also call the session
manager to start the standard services.
Args:
master: name of the TensorFlow master to use. See the
`tf.compat.v1.Session` constructor for how this is interpreted.
config: Optional ConfigProto proto used to configure the session, which is
passed as-is to create the session.
wait_for_checkpoint: Whether we should wait for the availability of a
checkpoint before creating Session. Defaults to False.
max_wait_secs: Maximum time to wait for the session to become available.
start_standard_services: Whether to start the standard services and the
queue runners.
Returns:
A Session object that can be used to drive the model."
11542,start_queue_runners,tensorflow/tensorflow/python/training/supervisor.py,747,method,"Start threads for `QueueRunners`.
Note that the queue runners collected in the graph key `QUEUE_RUNNERS`
are already started automatically when you create a session with the
supervisor, so unless you have non-collected queue runners to start
you do not need to call this explicitly.
Args:
sess: A `Session`.
queue_runners: A list of `QueueRunners`. If not specified, we'll use the
list of queue runners gathered in the graph under the key
`GraphKeys.QUEUE_RUNNERS`.
Returns:
The list of threads started for the `QueueRunners`.
Raises:
RuntimeError: If called with eager execution enabled.
@compatibility(eager)
Queues are not compatible with eager execution. To ingest data when eager
execution is enabled, use the `tf.data` API.
@end_compatibility"
11543,loop,tensorflow/tensorflow/python/training/supervisor.py,782,method,"Start a LooperThread that calls a function periodically.
If `timer_interval_secs` is None the thread calls `target(*args, **kwargs)`
repeatedly. Otherwise it calls it every `timer_interval_secs`
seconds. The thread terminates when a stop is requested.
The started thread is added to the list of threads managed by the supervisor
so it does not need to be passed to the `stop()` method.
Args:
timer_interval_secs: Number. Time boundaries at which to call `target`.
target: A callable object.
args: Optional arguments to pass to `target` when calling it.
kwargs: Optional keyword arguments to pass to `target` when calling it.
Returns:
The started thread."
11544,stop,tensorflow/tensorflow/python/training/supervisor.py,810,method,"Stop the services and the coordinator.
This does not close the session.
Args:
threads: Optional list of threads to join with the coordinator. If
`None`, defaults to the threads running the standard services, the
threads started for `QueueRunners`, and the threads started by the
`loop()` method. To wait on additional threads, pass the list in this
parameter.
close_summary_writer: Whether to close the `summary_writer`. Defaults to
`True` if the summary writer was created by the supervisor, `False`
otherwise.
ignore_live_threads: If `True` ignores threads that remain running after a
grace period when joining threads via the coordinator, instead of
raising a RuntimeError."
11545,request_stop,tensorflow/tensorflow/python/training/supervisor.py,849,method,"Request that the coordinator stop the threads.
See `Coordinator.request_stop()`.
Args:
ex: Optional `Exception`, or Python `exc_info` tuple as returned by
`sys.exc_info()`. If this is the first call to `request_stop()` the
corresponding exception is recorded and re-raised from `join()`."
11546,should_stop,tensorflow/tensorflow/python/training/supervisor.py,861,method,"Check if the coordinator was told to stop.
See `Coordinator.should_stop()`.
Returns:
True if the coordinator was told to stop, False otherwise."
11547,stop_on_exception,tensorflow/tensorflow/python/training/supervisor.py,871,method,"Context handler to stop the supervisor when an exception is raised.
See `Coordinator.stop_on_exception()`.
Returns:
A context handler."
11548,wait_for_stop,tensorflow/tensorflow/python/training/supervisor.py,881,method,Block waiting for the coordinator to stop.
11549,summary_computed,tensorflow/tensorflow/python/training/supervisor.py,885,method,"Indicate that a summary was computed.
Args:
sess: A `Session` object.
summary: A Summary proto, or a string holding a serialized summary proto.
global_step: Int. global step this summary is associated with. If `None`,
it will try to fetch the current step.
Raises:
TypeError: if 'summary' is not a Summary proto or a string.
RuntimeError: if the Supervisor was created without a `logdir`."
11550,managed_session,tensorflow/tensorflow/python/training/supervisor.py,936,method,"Returns a context manager for a managed session.
This context manager creates and automatically recovers a session. It
optionally starts the standard services that handle checkpoints and
summaries. It monitors exceptions raised from the `with` block or from the
services and stops the supervisor as needed.
The context manager is typically used as follows:
```python
def train():
sv = tf.compat.v1.train.Supervisor(...)
with sv.managed_session(<master>) as sess:
for step in xrange(..):
if sv.should_stop():
break
sess.run(<my training op>)
...do other things needed at each training step...
```
An exception raised from the `with` block or one of the service threads is
raised again when the block exits. This is done after stopping all threads
and closing the session. For example, an `AbortedError` exception, raised
in case of preemption of one of the workers in a distributed model, is
raised again when the block exits.
If you want to retry the training loop in case of preemption you can do it
as follows:
```python
def main(...):
while True
try:
train()
except tf.errors.Aborted:
pass
```
As a special case, exceptions used for control flow, such as
`OutOfRangeError` which reports that input queues are exhausted, are not
raised again from the `with` block: they indicate a clean termination of
the training loop and are considered normal termination.
Args:
master: name of the TensorFlow master to use. See the
`tf.compat.v1.Session` constructor for how this is interpreted.
config: Optional `ConfigProto` proto used to configure the session. Passed
as-is to create the session.
start_standard_services: Whether to start the standard services, such as
checkpoint, summary and step counter.
close_summary_writer: Whether to close the summary writer when closing the
session. Defaults to True.
Returns:
A context manager that yields a `Session` restored from the latest
checkpoint or initialized from scratch if not checkpoint exists. The
session is closed when the `with` block exits."
11551,SVSummaryThread,tensorflow/tensorflow/python/training/supervisor.py,1028,class,A thread to save summaries on a timer.
11552,run_loop,tensorflow/tensorflow/python/training/supervisor.py,1042,method,
11553,SVStepCounterThread,tensorflow/tensorflow/python/training/supervisor.py,1054,class,Threads to count steps and measure their duration.
11554,start_loop,tensorflow/tensorflow/python/training/supervisor.py,1075,method,
11555,run_loop,tensorflow/tensorflow/python/training/supervisor.py,1079,method,
11556,SVTimerCheckpointThread,tensorflow/tensorflow/python/training/supervisor.py,1102,class,A thread to checkpoint on a timer.
11557,run_loop,tensorflow/tensorflow/python/training/supervisor.py,1116,method,
11558,SyncReplicasOptimizer,tensorflow/tensorflow/python/training/sync_replicas_optimizer.py,45,class,"Class to synchronize, aggregate gradients and pass them to the optimizer.
This class is deprecated. For synchronous training, please use [Distribution
Strategies](https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/distribute).
In a typical asynchronous training environment, it's common to have some
stale gradients. For example, with a N-replica asynchronous training,
gradients will be applied to the variables N times independently. Depending
on each replica's training speed, some gradients might be calculated from
copies of the variable from several steps back (N-1 steps on average). This
optimizer avoids stale gradients by collecting gradients from all replicas,
averaging them, then applying them to the variables in one shot, after
which replicas can fetch the new variables and continue.
The following accumulators/queue are created:
* N `gradient accumulators`, one per variable to train. Gradients are pushed
to them and the chief worker will wait until enough gradients are collected
and then average them before applying to variables. The accumulator will
drop all stale gradients (more details in the accumulator op).
* 1 `token` queue where the optimizer pushes the new global_step value after
all variables are updated.
The following local variable is created:
* `sync_rep_local_step`, one per replica. Compared against the global_step in
each accumulator to check for staleness of the gradients.
The optimizer adds nodes to the graph to collect gradients and pause the
trainers until variables are updated.
For the Parameter Server job:
1. An accumulator is created for each variable, and each replica pushes the
gradients into the accumulators instead of directly applying them to the
variables.
2. Each accumulator averages once enough gradients (replicas_to_aggregate)
have been accumulated.
3. Apply the averaged gradients to the variables.
4. Only after all variables have been updated, increment the global step.
5. Only after step 4, pushes `global_step` in the `token_queue`, once for
each worker replica. The workers can now fetch the global step, use it to
update its local_step variable and start the next batch. Please note that
some workers can consume multiple minibatches, while some may not consume
even one. This is because each worker fetches minibatches as long as
a token exists. If one worker is stuck for some reason and does not
consume a token, another worker can use it.
For the replicas:
1. Start a step: fetch variables and compute gradients.
2. Once the gradients have been computed, push them into gradient
accumulators. Each accumulator will check the staleness and drop the stale.
3. After pushing all the gradients, dequeue an updated value of global_step
from the token queue and record that step to its local_step variable. Note
that this is effectively a barrier.
4. Start the next batch.
### Usage
```python
# Create any optimizer to update the variables, say a simple SGD:
opt = GradientDescentOptimizer(learning_rate=0.1)
# Wrap the optimizer with sync_replicas_optimizer with 50 replicas: at each
# step the optimizer collects 50 gradients before applying to variables.
# Note that if you want to have 2 backup replicas, you can change
# total_num_replicas=52 and make sure this number matches how many physical
# replicas you started in your job.
opt = tf.compat.v1.train.SyncReplicasOptimizer(opt, replicas_to_aggregate=50,
total_num_replicas=50)
# Some models have startup_delays to help stabilize the model but when using
# sync_replicas training, set it to 0.
# Now you can call `minimize()` or `compute_gradients()` and
# `apply_gradients()` normally
training_op = opt.minimize(total_loss, global_step=self.global_step)
# You can create the hook which handles initialization and queues.
sync_replicas_hook = opt.make_session_run_hook(is_chief)
```
In the training program, every worker will run the train_op as if not
synchronized.
```python
with training.MonitoredTrainingSession(
master=workers[worker_id].target, is_chief=is_chief,
hooks=[sync_replicas_hook]) as mon_sess:
while not mon_sess.should_stop():
mon_sess.run(training_op)
```
To use SyncReplicasOptimizer with an `Estimator`, you need to send
sync_replicas_hook while calling the fit.
```python
my_estimator = DNNClassifier(..., optimizer=opt)
my_estimator.fit(..., hooks=[sync_replicas_hook])
```"
11559,compute_gradients,tensorflow/tensorflow/python/training/sync_replicas_optimizer.py,207,method,"Compute gradients of ""loss"" for the variables in ""var_list"".
This simply wraps the compute_gradients() from the real optimizer. The
gradients will be aggregated in the apply_gradients() so that user can
modify the gradients like clipping with per replica global norm if needed.
The global norm with aggregated gradients can be bad as one replica's huge
gradients can hurt the gradients from other replicas.
Args:
*args: Arguments for compute_gradients().
**kwargs: Keyword arguments for compute_gradients().
Returns:
A list of (gradient, variable) pairs."
11560,apply_gradients,tensorflow/tensorflow/python/training/sync_replicas_optimizer.py,225,method,"Apply gradients to variables.
This contains most of the synchronization implementation and also wraps the
apply_gradients() from the real optimizer.
Args:
grads_and_vars: List of (gradient, variable) pairs as returned by
compute_gradients().
global_step: Optional Variable to increment by one after the
variables have been updated.
name: Optional name for the returned operation. Default to the
name passed to the Optimizer constructor.
Returns:
train_op: The op to dequeue a token so the replicas can exit this batch
and start the next one. This is executed by each replica.
Raises:
ValueError: If the grads_and_vars is empty.
ValueError: If global step is not provided, the staleness cannot be
checked."
11561,get_chief_queue_runner,tensorflow/tensorflow/python/training/sync_replicas_optimizer.py,360,method,"Returns the QueueRunner for the chief to execute.
This includes the operations to synchronize replicas: aggregate gradients,
apply to variables, increment global step, insert tokens to token queue.
Note that this can only be called after calling apply_gradients() which
actually generates this queuerunner.
Returns:
A `QueueRunner` for chief to execute.
Raises:
ValueError: If this is called before apply_gradients()."
11562,get_slot,tensorflow/tensorflow/python/training/sync_replicas_optimizer.py,380,method,"Return a slot named ""name"" created for ""var"" by the Optimizer.
This simply wraps the get_slot() from the actual optimizer.
Args:
*args: Arguments for get_slot().
**kwargs: Keyword arguments for get_slot().
Returns:
The `Variable` for the slot if it was created, `None` otherwise."
11563,variables,tensorflow/tensorflow/python/training/sync_replicas_optimizer.py,394,method,"Fetches a list of optimizer variables in the default graph.
This wraps `variables()` from the actual optimizer. It does not include
the `SyncReplicasOptimizer`'s local step.
Returns:
A list of variables."
11564,get_slot_names,tensorflow/tensorflow/python/training/sync_replicas_optimizer.py,405,method,"Return a list of the names of slots created by the `Optimizer`.
This simply wraps the get_slot_names() from the actual optimizer.
Args:
*args: Arguments for get_slot().
**kwargs: Keyword arguments for get_slot().
Returns:
A list of strings."
11565,get_init_tokens_op,tensorflow/tensorflow/python/training/sync_replicas_optimizer.py,419,method,"Returns the op to fill the sync_token_queue with the tokens.
This is supposed to be executed in the beginning of the chief/sync thread
so that even if the total_num_replicas is less than replicas_to_aggregate,
the model can still proceed as the replicas can compute multiple steps per
variable update. Make sure:
`num_tokens >= replicas_to_aggregate - total_num_replicas`.
Args:
num_tokens: Number of tokens to add to the queue.
Returns:
An op for the chief/sync replica to fill the token queue.
Raises:
ValueError: If this is called before apply_gradients().
ValueError: If num_tokens are smaller than replicas_to_aggregate -
total_num_replicas."
11566,make_session_run_hook,tensorflow/tensorflow/python/training/sync_replicas_optimizer.py,460,method,Creates a hook to handle SyncReplicasHook ops such as initialization.
11567,get_workers,tensorflow/tensorflow/python/training/sync_replicas_optimizer_test.py,35,function,
11568,get_verbosity,tensorflow/tensorflow/python/training/tensorboard_logging.py,78,function,
11569,set_verbosity,tensorflow/tensorflow/python/training/tensorboard_logging.py,82,function,
11570,set_summary_writer,tensorflow/tensorflow/python/training/tensorboard_logging.py,94,function,"Sets the summary writer that events will be logged to.
Calling any logging methods inside this module without calling this method
will fail. If you don't want to log, call `set_summary_writer(None)`.
Args:
summary_writer: Either a SummaryWriter or None. None will cause messages not
to be logged to any SummaryWriter, but they will still be passed to the
platform logging module."
11571,log,tensorflow/tensorflow/python/training/tensorboard_logging.py,119,function,"Conditionally logs `message % args` at the level `level`.
Note that tensorboard_logging verbosity and logging verbosity are separate;
the message will always be passed through to the logging module regardless of
whether it passes the tensorboard_logging verbosity check.
Args:
level: The verbosity level to use. Must be one of
tensorboard_logging.{DEBUG, INFO, WARN, ERROR, FATAL}.
message: The message template to use.
*args: Arguments to interpolate to the message template, if any.
Raises:
ValueError: If `level` is not a valid logging level.
RuntimeError: If the `SummaryWriter` to use has not been set."
11572,debug,tensorflow/tensorflow/python/training/tensorboard_logging.py,152,function,
11573,info,tensorflow/tensorflow/python/training/tensorboard_logging.py,156,function,
11574,warn,tensorflow/tensorflow/python/training/tensorboard_logging.py,160,function,
11575,error,tensorflow/tensorflow/python/training/tensorboard_logging.py,164,function,
11576,fatal,tensorflow/tensorflow/python/training/tensorboard_logging.py,168,function,
11577,global_step,tensorflow/tensorflow/python/training/training_util.py,41,function,"Small helper to get the global step.
```python
# Create a variable to hold the global_step.
global_step_tensor = tf.Variable(10, trainable=False, name='global_step')
# Create a session.
sess = tf.compat.v1.Session()
# Initialize the variable
sess.run(global_step_tensor.initializer)
# Get the variable value.
print('global_step: %s' % tf.compat.v1.train.global_step(sess,
global_step_tensor))
global_step: 10
```
Args:
sess: A TensorFlow `Session` object.
global_step_tensor: `Tensor` or the `name` of the operation that contains
the global step.
Returns:
The global step value."
11578,get_global_step,tensorflow/tensorflow/python/training/training_util.py,72,function,"Get the global step tensor.
The global step tensor must be an integer variable. We first try to find it
in the collection `GLOBAL_STEP`, or by name `global_step:0`.
Args:
graph: The graph to find the global step in. If missing, use default graph.
Returns:
The global step variable, or `None` if none was found.
Raises:
TypeError: If the global step tensor has a non-integer type, or if it is not
a `Variable`."
11579,create_global_step,tensorflow/tensorflow/python/training/training_util.py,107,function,"Create global step tensor in graph.
Args:
graph: The graph in which to create the global step tensor. If missing, use
default graph.
Returns:
Global step tensor.
Raises:
ValueError: if global step tensor is already defined."
11580,get_or_create_global_step,tensorflow/tensorflow/python/training/training_util.py,148,function,"Returns and create (if necessary) the global step tensor.
Args:
graph: The graph in which to create the global step tensor. If missing, use
default graph.
Returns:
The global step tensor."
11581,assert_global_step,tensorflow/tensorflow/python/training/training_util.py,166,function,"Asserts `global_step_tensor` is a scalar int `Variable` or `Tensor`.
Args:
global_step_tensor: `Tensor` to test."
11582,VocabInfo,tensorflow/tensorflow/python/training/warm_starting_util.py,39,class,"Vocabulary information for warm-starting.
See `tf.estimator.WarmStartSettings` for examples of using
VocabInfo to warm-start.
Args:
new_vocab: [Required] A path to the new vocabulary file (used with the model
to be trained).
new_vocab_size: [Required] An integer indicating how many entries of the new
vocabulary will used in training.
num_oov_buckets: [Required] An integer indicating how many OOV buckets are
associated with the vocabulary.
old_vocab: [Required] A path to the old vocabulary file (used with the
checkpoint to be warm-started from).
old_vocab_size: [Optional] An integer indicating how many entries of the old
vocabulary were used in the creation of the checkpoint. If not provided,
the entire old vocabulary will be used.
backup_initializer: [Optional] A variable initializer used for variables
corresponding to new vocabulary entries and OOV. If not provided, these
entries will be zero-initialized.
axis: [Optional] Denotes what axis the vocabulary corresponds to. The
default, 0, corresponds to the most common use case (embeddings or
linear weights for binary classification / regression). An axis of 1
could be used for warm-starting output layers with class vocabularies.
Returns:
A `VocabInfo` which represents the vocabulary information for warm-starting.
Raises:
ValueError: `axis` is neither 0 or 1.
Example Usage:
```python
embeddings_vocab_info = tf.VocabInfo(
new_vocab='embeddings_vocab',
new_vocab_size=100,
num_oov_buckets=1,
old_vocab='pretrained_embeddings_vocab',
old_vocab_size=10000,
backup_initializer=tf.compat.v1.truncated_normal_initializer(
mean=0.0, stddev=(1 / math.sqrt(embedding_dim))),
axis=0)
softmax_output_layer_kernel_vocab_info = tf.VocabInfo(
new_vocab='class_vocab',
new_vocab_size=5,
num_oov_buckets=0, # No OOV for classes.
old_vocab='old_class_vocab',
old_vocab_size=8,
backup_initializer=tf.compat.v1.glorot_uniform_initializer(),
axis=1)
softmax_output_layer_bias_vocab_info = tf.VocabInfo(
new_vocab='class_vocab',
new_vocab_size=5,
num_oov_buckets=0, # No OOV for classes.
old_vocab='old_class_vocab',
old_vocab_size=8,
backup_initializer=tf.compat.v1.zeros_initializer(),
axis=0)
#Currently, only axis=0 and axis=1 are supported.
```
"
11583,warm_start,tensorflow/tensorflow/python/training/warm_starting_util.py,413,function,"Warm-starts a model using the given settings.
If you are using a tf.estimator.Estimator, this will automatically be called
during training.
Args:
ckpt_to_initialize_from: [Required] A string specifying the directory with
checkpoint file(s) or path to checkpoint from which to warm-start the
model parameters.
vars_to_warm_start: [Optional] One of the following:
- A regular expression (string) that captures which variables to
warm-start (see tf.compat.v1.get_collection). This expression will only
consider variables in the TRAINABLE_VARIABLES collection -- if you need
to warm-start non_TRAINABLE vars (such as optimizer accumulators or
batch norm statistics), please use the below option.
- A list of strings, each a regex scope provided to
tf.compat.v1.get_collection with GLOBAL_VARIABLES (please see
tf.compat.v1.get_collection). For backwards compatibility reasons,
this is separate from the single-string argument type.
- A list of Variables to warm-start. If you do not have access to the
`Variable` objects at the call site, please use the above option.
- `None`, in which case only TRAINABLE variables specified in
`var_name_to_vocab_info` will be warm-started.
Defaults to `'.*'`, which warm-starts all variables in the
TRAINABLE_VARIABLES collection. Note that this excludes variables such
as accumulators and moving statistics from batch norm.
var_name_to_vocab_info: [Optional] Dict of variable names (strings) to
`tf.estimator.VocabInfo`. The variable names should be ""full"" variables,
not the names of the partitions. If not explicitly provided, the variable
is assumed to have no (changes to) vocabulary.
var_name_to_prev_var_name: [Optional] Dict of variable names (strings) to
name of the previously-trained variable in `ckpt_to_initialize_from`. If
not explicitly provided, the name of the variable is assumed to be same
between previous checkpoint and current model. Note that this has no
effect on the set of variables that is warm-started, and only controls
name mapping (use `vars_to_warm_start` for controlling what variables to
warm-start).
Raises:
ValueError: If the WarmStartSettings contains prev_var_name or VocabInfo
configuration for variable names that are not used. This is to ensure
a stronger check for variable configuration than relying on users to
examine the logs."
11584,LossScale,tensorflow/tensorflow/python/training/experimental/loss_scale.py,43,class,"Base class for all loss scales.
This is an abstract base class, so you cannot instantiate it directly.
Instead, use one of its concrete subclasses:
* `tf.mixed_precision.experimental.DynamicLossScale` (recommended)
* `tf.mixed_precision.experimental.FixedLossScale`
It's recommended to use a loss scale with a
`tf.keras.mixed_precision.experimental.LossScaleOptimizer`, as its easier than
using a loss scale directly.
Loss scaling is a process that multiplies the loss by a multiplier called the
loss scale, and divides each gradient by the same multiplier. The pseudocode
for this process is:
```
loss = ...
loss *= loss_scale
grads = gradients(loss, vars)
grads /= loss_scale
```
Mathematically, loss scaling has no effect, but can help avoid numerical
underflow in intermediate gradients when float16 tensors are used for mixed
precision training. By multiplying the loss, each intermediate gradient will
have the same multiplier applied.
Instances of this class represent a loss scale. Calling instances of this
class returns the loss scale as a scalar float32 tensor, while method
`update()` updates the loss scale depending on the values of the gradients.
Optimizers use instances of this class to scale loss and gradients.
In most functions that accept a LossScale, you can also pass an int (such as
8) to create a `FixedLossScale` or the string `""dynamic""` to create a dynamic
loss scale."
11585,update,tensorflow/tensorflow/python/training/experimental/loss_scale.py,91,method,"Updates the value of the loss scale.
The loss scale will be potentially updated, based on the value of `grads`.
The tensor returned by calling this class is only updated when this function
is evaluated.
In eager mode, this directly updates the loss scale, so that calling
`__call__` will return the newly updated loss scale. In graph mode,
this returns an op that, when evaluated, updates the loss scale.
This function also returns a `should_apply_gradients` bool. If False,
gradients should not be applied to the variables that step, as nonfinite
gradients were found, and the loss scale has been be updated to reduce the
chance of finding nonfinite gradients in the next step. Some loss scale
classes will always return True, as they cannot adjust themselves in
response to nonfinite gradients.
When a DistributionStrategy is used, this function may only be called in a
cross-replica context.
Args:
grads: A nested structure of unscaled gradients, each which is the
gradient of the loss with respect to a weight. The gradients should have
already been divided by the loss scale being before passed to this
function. 'None' gradients are accepted, and are ignored.
Returns:
update_op: In eager mode, None. In graph mode, an op to update the loss
scale.
should_apply_gradients: Either a bool or a scalar boolean tensor. If
False, the caller should skip applying `grads` to the variables this
step."
11586,get_config,tensorflow/tensorflow/python/training/experimental/loss_scale.py,191,method,Returns the config of this loss scale.
11587,from_config,tensorflow/tensorflow/python/training/experimental/loss_scale.py,196,method,Creates the LossScale from its config.
11588,FixedLossScale,tensorflow/tensorflow/python/training/experimental/loss_scale.py,204,class,"Loss scale with a fixed value.
The loss scale is not updated for the lifetime of instances of this class.
A given instance of this class always returns the same number when called."
11589,update,tensorflow/tensorflow/python/training/experimental/loss_scale.py,240,method,
11590,get_config,tensorflow/tensorflow/python/training/experimental/loss_scale.py,247,method,
11591,DynamicLossScale,tensorflow/tensorflow/python/training/experimental/loss_scale.py,286,class,"Loss scale that dynamically adjusts itself.
Dynamic loss scaling works by adjusting the loss scale as training progresses.
The goal is to keep the loss scale as high as possible without overflowing the
gradients. As long as the gradients do not overflow, raising the loss scale
never hurts.
The algorithm starts by setting the loss scale to an initial value. Every N
steps that the gradients are finite, the loss scale is increased by some
factor. However, if a NaN or Inf gradient is found, the gradients for that
step are not applied, and the loss scale is decreased by the factor. This
process tends to keep the loss scale as high as possible without gradients
overflowing."
11592,initial_loss_scale,tensorflow/tensorflow/python/training/experimental/loss_scale.py,335,method,
11593,increment_period,tensorflow/tensorflow/python/training/experimental/loss_scale.py,339,method,
11594,multiplier,tensorflow/tensorflow/python/training/experimental/loss_scale.py,343,method,
11595,update,tensorflow/tensorflow/python/training/experimental/loss_scale.py,349,method,Updates loss scale based on if gradients are finite in current step.
11596,get_config,tensorflow/tensorflow/python/training/experimental/loss_scale.py,409,method,
11597,update_if_finite_grads,tensorflow/tensorflow/python/training/experimental/loss_scale.py,370,method,Update assuming the gradients are finite.
11598,update_if_not_finite_grads,tensorflow/tensorflow/python/training/experimental/loss_scale.py,384,method,Update assuming the gradients are nonfinite.
11599,get_is_finite,tensorflow/tensorflow/python/training/experimental/loss_scale.py,355,method,
11600,incr_loss_scale,tensorflow/tensorflow/python/training/experimental/loss_scale.py,373,method,
11601,get,tensorflow/tensorflow/python/training/experimental/loss_scale.py,417,function,Get a loss scale object.
11602,MixedPrecisionLossScaleOptimizer,tensorflow/tensorflow/python/training/experimental/loss_scale_optimizer.py,31,class,"An optimizer that applies loss scaling.
Loss scaling is a process that multiplies the loss by a multiplier called the
loss scale, and divides each gradient by the same multiplier. The pseudocode
for this process is:
```
loss = ...
loss *= loss_scale
grads = gradients(loss, vars)
grads /= loss_scale
```
Mathematically, loss scaling has no effect, but can help avoid numerical
underflow in intermediate gradients when float16 tensors are used for mixed
precision training. By multiplying the loss, each intermediate gradient will
have the same multiplier applied.
The loss scale can either be a fixed constant, chosen by the user, or be
dynamically determined. Dynamically determining the loss scale is convenient
as a loss scale does not have to be explicitly chosen. However it reduces
performance.
This optimizer wraps another optimizer and applies loss scaling to it via a
`LossScale`. Loss scaling is applied whenever gradients are
computed, such as through `minimize()`."
11603,compute_gradients,tensorflow/tensorflow/python/training/experimental/loss_scale_optimizer.py,80,method,"Compute gradients of `loss` for the variables in `var_list`.
This adjusts the dynamic range of the gradient evaluation by scaling up
the `loss` value. The gradient values are then scaled back down by the
reciprocal of the loss scale. This is useful in reduced precision training
where small gradient values would otherwise underflow the representable
range.
Args:
loss: A Tensor containing the value to minimize or a callable taking no
arguments which returns the value to minimize. When eager execution is
enabled it must be a callable.
var_list: Optional list or tuple of `tf.Variable` to update to minimize
`loss`. Defaults to the list of variables collected in the graph under
the key `GraphKeys.TRAINABLE_VARIABLES`.
gate_gradients: How to gate the computation of gradients. Can be
`GATE_NONE`, `GATE_OP`, or `GATE_GRAPH`.
aggregation_method: Specifies the method used to combine gradient terms.
Valid values are defined in the class `AggregationMethod`.
colocate_gradients_with_ops: If True, try colocating gradients with the
corresponding op.
grad_loss: Optional. A `Tensor` holding the gradient computed for `loss`.
Returns:
A list of (gradient, variable) pairs. Variable is always present, but
gradient can be `None`."
11604,apply_gradients,tensorflow/tensorflow/python/training/experimental/loss_scale_optimizer.py,152,method,"Apply gradients to variables.
This is the second part of `minimize()`. It returns an `Operation` that
conditionally applies gradients if all gradient values are finite.
Otherwise no update is performed (nor is `global_step` incremented).
Args:
grads_and_vars: List of (gradient, variable) pairs as returned by
`compute_gradients()`.
global_step: Optional `Variable` to increment by one after the variables
have been updated.
name: Optional name for the returned operation. Default to the name
passed to the `Optimizer` constructor.
Returns:
An `Operation` that conditionally applies the specified gradients. If
`global_step` was not None, that operation also increments `global_step`.
Raises:
RuntimeError: If you should use `_distributed_apply()` instead."
11605,variables,tensorflow/tensorflow/python/training/experimental/loss_scale_optimizer.py,247,method,Returns the variables of the Optimizer.
11606,apply_fn,tensorflow/tensorflow/python/training/experimental/loss_scale_optimizer.py,215,method,
11607,new_loss,tensorflow/tensorflow/python/training/experimental/loss_scale_optimizer.py,131,method,
11608,create_mirrored_strategy,tensorflow/tensorflow/python/training/experimental/loss_scale_optimizer_test.py,47,function,
11609,get_gradients,tensorflow/tensorflow/python/training/experimental/loss_scale_optimizer_test.py,63,function,
11610,create_identity_with_grad_check_fn,tensorflow/tensorflow/python/training/experimental/loss_scale_optimizer_test.py,69,function,"Returns a function that asserts it's gradient has a certain value.
This serves as a hook to assert intermediate gradients have a certain value.
This returns an identity function. The identity's gradient function is also
the identity function, except it asserts that the gradient equals
`expected_gradient` and has dtype `expected_dtype`.
Args:
expected_gradient: The gradient function asserts that the gradient is this
value.
expected_dtype: The gradient function asserts the gradient has this dtype.
Returns:
An identity function whose gradient function asserts the gradient has a
certain value."
11611,create_mirrored_strategy,tensorflow/tensorflow/python/training/experimental/loss_scale_test.py,45,function,
11612,LossScaleGradientTape,tensorflow/tensorflow/python/training/experimental/loss_scaling_gradient_tape.py,50,class,"A gradient tape that scales losses and unscales resulting gradients.
Operates as a normal gradient tape, but takes in a
`tf.mixed_precision.experimental.LossScale` object. Losses are scaled up by
some amount before the gradients are calculated and the resulting gradients
are scaled down by the same amount.
This has no net mathematical effect, but can be used to prevent vanishing
gradients, for example in the case of mixed precision training.
If a DynamicLossScale object is used and non-finite gradients are encountered,
the loss scale will be updated and the gradients recomputed until either
finite gradients are encountered or the loss scale becomes 1.
This class should *not* be used with a LossScaleOptimizer, as both classes
update the LossScale object. Use a non-loss scaling optimizer instead.
Usage:
```
opt = tf.keras.optimizers.SGD(1.0)
model_loss_scale = tf.mixed_precision.experimental.DynamicLossScale()
for step in training_steps:
with LossScaleGradientTape(model_loss_scale) as tape:
logits = ... # Run model and get logits
loss = tf.nn.softmax_cross_entropy_with_logits(logits=logits,
labels=labels)
loss = tf.reduce_mean(loss)
vars = tape.watched_variables()
grads = tape.gradient(loss, vars)
opt.apply_gradients(zip(grads, vars))
```
WARNING: Computing second-order (or higher) gradients with a
`LossScaleGradientTape` does not yet work properly when a
`tf.distribute.Strategy` is used. Computing second-order gradients will return
None instead of the gradient tensors. This only occurs when you nest multiple
gradient tapes under each other; if you do not nest them, this issue will not
occur."
11613,gradient,tensorflow/tensorflow/python/training/experimental/loss_scaling_gradient_tape.py,127,method,"Computes the gradient using operations recorded in context of this tape.
Uses the `LossScale` object provided in the constructor to scale `target`
and then to unscale the resulting gradients.
Args:
target: a list or nested structure of Tensors or Variables to be
differentiated.
sources: a list or nested structure of Tensors or Variables. `target` will
be differentiated against elements in `sources`.
output_gradients: a list of gradients, one for each element of target.
Defaults to None.
unconnected_gradients: a value which can either hold 'none' or 'zero' and
alters the value which will be returned if the target and sources are
unconnected. The possible values and effects are detailed in
'UnconnectedGradients' and it defaults to 'none'.
Returns:
a list or nested structure of Tensors (or IndexedSlices, or None),
one for each element in `sources`. Returned structure is the same as
the structure of `sources`. If non-finite gradients are encountered
after dynamic scaling, the loss scale will be updated and the gradients
recomputed until either finite gradients are encountered or the loss scale
becomes 1.
Raises:
RuntimeError: if called inside the context of the tape, or if called more
than once on a non-persistent tape.
ValueError: if the target is a variable or if unconnected gradients is
called with an unknown value."
11614,jacobian,tensorflow/tensorflow/python/training/experimental/loss_scaling_gradient_tape.py,183,method,
11615,batch_jacobian,tensorflow/tensorflow/python/training/experimental/loss_scaling_gradient_tape.py,193,method,
11616,create_mirrored_strategy,tensorflow/tensorflow/python/training/experimental/loss_scaling_gradient_tape_test.py,44,function,
11617,enable_mixed_precision_graph_rewrite,tensorflow/tensorflow/python/training/experimental/mixed_precision.py,65,function,"Enable mixed precision via a graph rewrite.
Mixed precision is the use of both float32 and float16 data types when
training a model to improve performance. This is achieved via a graph rewrite
operation and a loss-scale optimizer.
Performing arithmetic operations in float16 takes advantage of specialized
processing units, such as NVIDIA Tensor Cores, for much higher arithmetic
throughput. However, due to the smaller representable range, performing the
entire training with float16 can result in gradient underflow, that is, small
gradient values becoming zeroes. Instead, performing only select arithmetic
operations in float16 results in higher throughput and decreased training
time when using compatible hardware accelerators while also reducing memory
usage, typically without sacrificing model accuracy.
Note: While the mixed precision rewrite changes the datatype of various
layers throughout the model, the same accuracy reached in float32 is
expected. If a `NaN` gradient occurs with dynamic loss scaling, the model
update for that batch is skipped. In this case, the global step count is not
incremented, and the `LossScaleOptimizer` attempts to decrease the loss
scaling value to avoid `NaN` values in subsequent iterations. This approach
has been shown to achieve the same accuracy as float32 and, in most cases,
better training throughput.
Example:
```python
model = tf.keras.models.Sequential([
tf.keras.layers.Dense(64, activation='relu'),
tf.keras.layers.Dense(64, activation='softmax'),
])
opt = tf.keras.optimizers.SGD()
opt = tf.train.experimental.enable_mixed_precision_graph_rewrite(opt)
model.compile(loss=""mse"", optimizer=opt)
x_train = np.random.random((1024, 64))
y_train = np.random.random((1024, 64))
model.fit(x_train, y_train)
```
Calling `enable_mixed_precision_graph_rewrite(opt)` enables the graph rewrite
operation before computing gradients. The function additionally returns an
`Optimizer` (`opt`) wrapped with a `LossScaleOptimizer`. This prevents
underflow in the float16 tensors during the backward pass. An optimizer of
type `tf.keras.optimizers.Optimizer` or `tf.compat.v1.train.Optimizer` must be
passed to this function, which will then be wrapped to use loss scaling.
The graph rewrite operation changes the dtype of certain operations in the
graph from float32 to float16. There are several categories of operations
that are either included or excluded by this rewrite operation. The following
categories of Ops are defined inside corresponding functions under the class
`AutoMixedPrecisionLists` in
<a href=""https://github.com/tensorflow/tensorflow/blob/master/tensorflow/
core/grappler/optimizers/auto_mixed_precision_lists.h"">
auto_mixed_precision_lists.h</a>:
* `ClearList`: Ops that do not have numerically significant adverse effects.
E.g. `ArgMax` and `Floor`.
* `AllowList`: Ops that are considered numerically safe for execution in
float16, and thus are always converted. E.g. `Conv2D`.
* `DenyList`: Ops that are numerically unsafe to execute in float16 and
can negatively affect downstream nodes. E.g. `Softmax`.
* `GrayList`: Ops that are considered numerically safe for execution in
float16 unless downstream from a DenyList Op. E.g. `Add` and `AvgPool`.
When this function is used, gradients should be computed and applied with the
returned optimizer, either by calling `opt.minimize()` or
`opt.compute_gradients()` followed by `opt.apply_gradients()`. If gradients
are instead computed with `tf.gradients` or `tf.GradientTape`, loss scaling
will not be applied, which will likely cause your model not to converge due to
float16 underflow problems. To apply lossing scaling with `tf.gradients` or
`tf.GradientTape`, `LossScaleOptimizer.get_scaled_loss` and
`LossScaleOptimizer.get_unscaled_gradients`. See
`keras.mixed_precision.experimental.LossScaleOptimizer` for details how to do
this.
When eager execution is enabled, the mixed precision graph rewrite is only
enabled within `tf.function`s, as outside `tf.function`s, there is no graph.
For NVIDIA GPUs with Tensor cores, as a general performance guide, dimensions
(such as batch size, input size, output size, and channel counts)
should be powers of two if under 256, or otherwise divisible by 8 if above
256. For more information, check out the
[NVIDIA Deep Learning Performance Guide](
https://docs.nvidia.com/deeplearning/sdk/dl-performance-guide/index.html).
Currently, mixed precision is only enabled on NVIDIA Tensor Core GPUs with
Compute Capability 7.0 and above (Volta, Turing, or newer architectures). The
parts of the graph on CPUs and TPUs are untouched by the graph rewrite.
## Comparison with the Keras mixed precision API
Both this function and the [Keras mixed precision
API](https://www.tensorflow.org/guide/keras/mixed_precision) enable the use of
mixed precision in a model. Therefore, only one of the two APIs can be used.
We recommend using the Keras mixed precision API, as it is more customizable
and supports Eager execution. However, it only supports models which use Keras
layers, while the graph rewrite works in any model that uses `tf.function`s.
The core difference between the two APIs is that this function is a graph
rewrite, and so it changes the graph to use mixed precision under the hood.
You still build your graph in float32, and the graph rewrite will change
certain ops to float16. The Keras mixed precision API directly builds the
Keras Model using a mix of float16 and float32.
One core advantage of the Keras API is it supports mixed precision with Eager
execution, i.e. mixed precision outside `tf.function`s. The graph rewrite will
only affect ops within `tf.function`s, making it harder to debug if issues
occur with mixed precision. The Keras API is also more customizable, as you
can override any layer to run in float32 by passing `dtype=""float32""` to the
layer constructor. Additionally, you can query the dtype of tensors in the
model by checking `tensor.dtype`. With the graph rewrite, all tensors appear
to be float32 since the dtype is only changed under the hood.
The main advantage of the graph rewrite (this function) is that it works even
if you do not use Keras layers or any other part of Keras. The Keras mixed
precision API requires models which use Keras layers, as it only inserts casts
inside Keras layers and models. Another advantage is that the graph rewrite
never results in a TypeError, which the Keras API may introduce if you do
certain operations outside Keras. For example, the following will result in a
TypeError if the Keras mixed precision API is enabled, as a float16 and
float32 tensor will be added:
`tf.keras.layers.Dense(2)(x) + tf.keras.layers.Dense(2, dtype=""float32"")(x)`
Raises:
`ValueError`, if the `tf.keras.mixed_precision` API is also used by calling
`tf.keras.mixed_precision.experimental.set_policy`. Only one mixed precision
API can be used.
Args:
opt: An instance of a `tf.keras.optimizers.Optimizer`.
loss_scale: Either an int/float, the string `""dynamic""`, or an instance of a
`tf.mixed_precision.experimental.LossScale`. The loss scale to use. It is
recommended to keep this as its default value of `""dynamic""`, which will
adjust the scaling automatically to prevent `Inf` or `NaN` values.
Returns:
A version of `opt` that will use loss scaling to prevent underflow."
11618,enable_mixed_precision_graph_rewrite_v1,tensorflow/tensorflow/python/training/experimental/mixed_precision.py,210,function,"Enable mixed precision via a graph rewrite.
Mixed precision is the use of both float32 and float16 data types when
training a model to improve performance. This is achieved via a graph rewrite
operation and a loss-scale optimizer.
Performing arithmetic operations in float16 takes advantage of specialized
processing units, such as NVIDIA Tensor Cores, for much higher arithmetic
throughput. However, due to the smaller representable range, performing the
entire training with float16 can result in gradient underflow, that is, small
gradient values becoming zeroes. Instead, performing only select arithmetic
operations in float16 results in higher throughput and decreased training
time when using compatible hardware accelerators while also reducing memory
usage, typically without sacrificing model accuracy.
Note: While the mixed precision rewrite changes the datatype of various
layers throughout the model, the same accuracy reached in float32 is
expected. If a `NaN` gradient occurs with dynamic loss scaling, the model
update for that batch is skipped. In this case, the global step count is not
incremented, and the `LossScaleOptimizer` attempts to decrease the loss
scaling value to avoid `NaN` values in subsequent iterations. This approach
has been shown to achieve the same accuracy as float32 and, in most cases,
better training throughput.
Example:
```python
model = tf.keras.models.Sequential([
tf.keras.layers.Dense(64, activation='relu'),
tf.keras.layers.Dense(64, activation='softmax'),
])
opt = tf.keras.optimizers.SGD()
opt = tf.train.experimental.enable_mixed_precision_graph_rewrite(opt)
model.compile(loss=""mse"", optimizer=opt)
x_train = np.random.random((1024, 64))
y_train = np.random.random((1024, 64))
model.fit(x_train, y_train)
```
Calling `enable_mixed_precision_graph_rewrite(opt)` enables the graph rewrite
operation before computing gradients. The function additionally returns an
`Optimizer` (`opt`) wrapped with a `LossScaleOptimizer`. This prevents
underflow in the float16 tensors during the backward pass. An optimizer of
type `tf.train.Optimizer` or `tf.keras.optimizers.Optimizer` must be passed
to this function, which will then be wrapped to use loss scaling.
The graph rewrite operation changes the `dtype` of certain operations in the
graph from float32 to float16. There are several categories of operations
that are either included or excluded by this rewrite operation. The following
categories of Ops are defined inside corresponding functions under the class
`AutoMixedPrecisionLists` in
<a href=""https://github.com/tensorflow/tensorflow/blob/master/tensorflow/
core/grappler/optimizers/auto_mixed_precision_lists.h"">
auto_mixed_precision_lists.h</a>:
* `ClearList`: Ops that do not have numerically significant adverse effects.
E.g. `ArgMax` and `Floor`.
* `AllowList`: Ops that are considered numerically safe for execution in
float16, and thus are always converted. E.g. `Conv2D`.
* `DenyList`: Ops that are numerically unsafe to execute in float16 and
can negatively affect downstream nodes. E.g. `Softmax`.
* `GrayList`: Ops that are considered numerically safe for execution in
float16 unless downstream from a DenyList Op. E.g. `Add` and `AvgPool`.
When this function is used, gradients should only be computed and applied
with the returned optimizer, either by calling `opt.minimize()` or
`opt.compute_gradients()` followed by `opt.apply_gradients()`.
Gradients should not be computed with `tf.gradients` or `tf.GradientTape`.
This is because the returned optimizer will apply loss scaling, and
`tf.gradients` or `tf.GradientTape` will not. If you do directly use
`tf.gradients` or `tf.GradientTape`, your model may not converge due to
float16 underflow problems.
When eager execution is enabled, the mixed precision graph rewrite is only
enabled within `tf.function`s, as outside `tf.function`s, there is no graph.
For NVIDIA GPUs with Tensor cores, as a general performance guide, dimensions
(such as batch size, input size, output size, and channel counts)
should be powers of two if under 256, or otherwise divisible by 8 if above
256. For more information, check out the
[NVIDIA Deep Learning Performance Guide](
https://docs.nvidia.com/deeplearning/sdk/dl-performance-guide/index.html).
Currently, mixed precision is only enabled on NVIDIA Tensor Core GPUs with
Compute Capability 7.0 and above (Volta, Turing, or newer architectures). The
parts of the graph on CPUs and TPUs are untouched by the graph rewrite.
Raises:
`ValueError`, if the `tf.keras.mixed_precision` API is also used by calling
`tf.keras.mixed_precision.experimental.set_policy`. Only one mixed precision
API can be used.
Args:
opt: An instance of a `tf.keras.optimizers.Optimizer` or a
`tf.train.Optimizer`.
loss_scale: Either an int/float, the string `""dynamic""`, or an instance of
a `tf.mixed_precision.experimental.LossScale`. The loss scale to use. It
is recommended to keep this as its default value of `""dynamic""`, which
will adjust the scaling automatically to prevent `Inf` or `NaN` values.
Returns:
A version of `opt` that will use loss scaling to prevent underflow."
11619,disable_mixed_precision_graph_rewrite,tensorflow/tensorflow/python/training/experimental/mixed_precision.py,352,function,"Disables the mixed precision graph rewrite.
After this is called, the mixed precision graph rewrite will no longer run for
tf.functions, and so float32 operations will no longer be converted to
float16.
This does not undo the effects of loss scaling. Any optimizers wrapped with a
LossScaleOptimizer will continue to do loss scaling, although this loss
scaling will no longer be useful, as the graph rewrite no longer converts
tf.functions to use float16.
This function is useful for unit testing. A unit test can test using the mixed
precision graph rewrite, then disable it so future unit tests continue using
float32."
11620,disable_mixed_precision_graph_rewrite_v1,tensorflow/tensorflow/python/training/experimental/mixed_precision.py,376,function,"Disables the mixed precision graph rewrite.
After this is called, the mixed precision graph rewrite will no longer run for
new Sessions, and so float32 operations will no longer be converted to float16
in such Sessions. However, any existing Sessions will continue to have the
graph rewrite enabled if they were created after
`enable_mixed_precision_graph_rewrite` was called but before
`disable_mixed_precision_graph_rewrite` was called.
This does not undo the effects of loss scaling. Any optimizers wrapped with a
LossScaleOptimizer will continue to do loss scaling, although this loss
scaling will no longer be useful if the optimizer is used in new Sessions, as
the graph rewrite no longer converts the graph to use float16.
This function is useful for unit testing. A unit tests can test using the
mixed precision graph rewrite, then disable it so future unit tests continue
using float32. If this is done, unit tests should not share a single session,
as `enable_mixed_precision_graph_rewrite` and
`disable_mixed_precision_graph_rewrite` have no effect on existing sessions."
11621,CheckpointOptions,tensorflow/tensorflow/python/training/saving/checkpoint_options.py,25,class,"Options for constructing a Checkpoint.
Used as the `_options` argument to the `tf.Checkpoint` constructor to adjust
how variables are saved.
Example: Run IO ops on ""localhost"" while saving a checkpoint:
```
step = tf.Variable(0, name=""step"")
checkpoint = tf.Checkpoint(step=step)
options = tf.CheckpointOptions(experimental_io_device=""/job:localhost"")
checkpoint.save(""/tmp/ckpt"", options=options)
```"
11622,sharded_filename,tensorflow/tensorflow/python/training/saving/functional_saver.py,117,function,"Append sharding information to a filename.
Args:
filename_tensor: A string tensor.
shard: Integer. The shard for the filename.
num_shards: An int Tensor for the number of shards.
Returns:
A string tensor."
11623,MultiDeviceSaver,tensorflow/tensorflow/python/training/saving/functional_saver.py,131,class,"Saves checkpoints directly from multiple devices.
Note that this is a low-level utility which stores Tensors in the keys
specified by `SaveableObject`s. Higher-level utilities for object-based
checkpointing are built on top of it."
11624,to_proto,tensorflow/tensorflow/python/training/saving/functional_saver.py,174,method,Serializes to a SaverDef referencing the current graph.
11625,save,tensorflow/tensorflow/python/training/saving/functional_saver.py,204,method,"Save the saveable objects to a checkpoint with `file_prefix`.
Args:
file_prefix: A string or scalar string Tensor containing the prefix to
save under.
options: Optional `CheckpointOptions` object.
Returns:
An `Operation`, or None when executing eagerly."
11626,restore,tensorflow/tensorflow/python/training/saving/functional_saver.py,299,method,"Restore the saveable objects from a checkpoint with `file_prefix`.
Args:
file_prefix: A string or scalar string Tensor containing the prefix for
files to read from.
options: Optional `CheckpointOptions` object.
Returns:
A dictionary mapping from SaveableObject names to restore operations."
11627,save_fn,tensorflow/tensorflow/python/training/saving/functional_saver.py,254,method,
11628,restore_fn,tensorflow/tensorflow/python/training/saving/functional_saver.py,312,method,
11629,tf_function_save,tensorflow/tensorflow/python/training/saving/functional_saver.py,293,method,
11630,tf_function_restore,tensorflow/tensorflow/python/training/saving/functional_saver.py,329,method,
11631,SaveableHook,tensorflow/tensorflow/python/training/saving/saveable_hook.py,24,class,"Base class for running callbacks at Save/Restore time.
Subclasses should override one or both methods to modify or read variables
during the saving process. No guarantees are made regarding the precedence
of execution between multiple `SaveableHook` objects, but execution is
guaranteed to occur before or after the respective event.
Users should emit the SaveableHook alongside other SaveableObjects, such as
in Trackable._gather_saveables_for_checkpoint().
Saves a single constant in order to be compliant with the SaveableObject API."
11632,device,tensorflow/tensorflow/python/training/saving/saveable_hook.py,50,method,
11633,before_save,tensorflow/tensorflow/python/training/saving/saveable_hook.py,53,method,This method will be called before iterating devices for saving.
11634,after_restore,tensorflow/tensorflow/python/training/saving/saveable_hook.py,57,method,This method will be called after each device is restored.
11635,SaveSpec,tensorflow/tensorflow/python/training/saving/saveable_object.py,21,class,Class used to describe tensor slices that need to be saved.
11636,tensor,tensorflow/tensorflow/python/training/saving/saveable_object.py,54,method,
11637,SaveableObject,tensorflow/tensorflow/python/training/saving/saveable_object.py,58,class,Base class for saving and restoring saveable objects.
11638,optional_restore,tensorflow/tensorflow/python/training/saving/saveable_object.py,76,method,A hint to restore assertions that this object is optional.
11639,device,tensorflow/tensorflow/python/training/saving/saveable_object.py,81,method,The device for SaveSpec Tensors.
11640,restore,tensorflow/tensorflow/python/training/saving/saveable_object.py,85,method,"Restores this object from 'restored_tensors'.
Args:
restored_tensors: the tensors that were loaded from a checkpoint
restored_shapes: the shapes this object should conform to after
restore, or None.
Returns:
An operation that restores the state of the object.
Raises:
ValueError: If the object cannot be restored using the provided
parameters."
11641,set_cpu0,tensorflow/tensorflow/python/training/saving/saveable_object_util.py,54,function,"Creates a new device string based on `device_string` but using /CPU:0.
If the device is already on /CPU:0, this is a no-op.
Args:
device_string: A device string.
Returns:
A device string."
11642,ReferenceVariableSaveable,tensorflow/tensorflow/python/training/saving/saveable_object_util.py,70,class,SaveableObject implementation that handles reference variables.
11643,restore,tensorflow/tensorflow/python/training/saving/saveable_object_util.py,77,method,
11644,ResourceVariableSaveable,tensorflow/tensorflow/python/training/saving/saveable_object_util.py,88,class,SaveableObject implementation that handles ResourceVariables.
11645,restore,tensorflow/tensorflow/python/training/saving/saveable_object_util.py,119,method,
11646,f,tensorflow/tensorflow/python/training/saving/saveable_object_util.py,100,method,
11647,saveable_objects_for_op,tensorflow/tensorflow/python/training/saving/saveable_object_util.py,134,function,"Create `SaveableObject`s from an operation.
Args:
op: A variable, operation, or SaveableObject to coerce into a
SaveableObject.
name: A string name for the SaveableObject.
Yields:
`SaveableObject`s which together save/restore `op`.
Raises:
TypeError: If `name` is not a string.
ValueError: For operations with no known conversion to SaveableObject."
11648,op_list_to_dict,tensorflow/tensorflow/python/training/saving/saveable_object_util.py,224,function,"Create a dictionary of names to operation lists.
Args:
op_list: A (nested) list, tuple, or set of Variables or SaveableObjects.
convert_variable_to_tensor: Whether or not to convert single Variables
with no slice info into Tensors.
Returns:
A dictionary of names to the operations that must be saved under
that name. Variables with save_slice_info are grouped together under the
same key in no particular order.
Raises:
TypeError: If the type of op_list or its elements is not supported.
ValueError: If at least two saveables share the same name."
11649,validate_and_slice_inputs,tensorflow/tensorflow/python/training/saving/saveable_object_util.py,336,function,"Returns the variables and names that will be used for a Saver.
Args:
names_to_saveables: A dict (k, v) where k is the name of an operation and
v is an operation to save or a BaseSaverBuilder.Saver.
Returns:
A list of SaveableObjects.
Raises:
TypeError: If any of the keys are not strings or any of the
values are not one of Tensor or Variable or a trackable operation.
ValueError: If the same operation is given in more than one value
(this also applies to slices of SlicedVariables)."
11650,trace_save_restore_functions,tensorflow/tensorflow/python/training/saving/saveable_object_util.py,365,function,Gathers all SaveableObjects and traces the save and restore ops.
11651,RestoredSaveableObject,tensorflow/tensorflow/python/training/saving/saveable_object_util.py,444,class,SaveableObject restored from SavedModel using the traced save/restore.
11652,restore,tensorflow/tensorflow/python/training/saving/saveable_object_util.py,461,method,
11653,restored_saved_object_factory,tensorflow/tensorflow/python/training/saving/saveable_object_util.py,467,function,
11654,create_saveable_object,tensorflow/tensorflow/python/training/saving/saveable_object_util.py,473,function,"Creates a SaveableObject while potentially in a different graph.
When creating the frozen saver for SavedModel, the save and restore ops are
placed in a separate graph. Since RestoredSaveableObject uses tf.functions to
save and restore, the function captures must be mapped to the new graph.
Args:
factory: Factory method for creating the SaveableObject.
name: Checkpoint key of this SaveableObject.
call_with_mapped_captures: Helper that calls a tf.function while remapping
the captures.
Returns:
a SaveableObject."
11655,is_factory_for_restored_saveable_object,tensorflow/tensorflow/python/training/saving/saveable_object_util.py,504,function,
11656,CheckpointInitialValue,tensorflow/tensorflow/python/training/tracking/base.py,57,class,"Tensor wrapper for managing update UIDs in `Variables`.
When supplied as an initial value, objects of this type let a `Variable`
(`Variable`, `ResourceVariable`, etc.) know the UID of the restore the initial
value came from. This allows deferred restorations to be sequenced in the
order the user specified them, and lets us fall back on assignment if an
initial value is not set (e.g. due to a custom getter interfering).
See comments in _add_variable_with_custom_getter for more information about
how `CheckpointInitialValue` is used."
11657,checkpoint_position,tensorflow/tensorflow/python/training/tracking/base.py,85,method,
11658,NoRestoreSaveable,tensorflow/tensorflow/python/training/tracking/base.py,89,class,Embeds a tensor in a checkpoint with no restore ops.
11659,restore,tensorflow/tensorflow/python/training/tracking/base.py,97,method,
11660,PythonStateSaveable,tensorflow/tensorflow/python/training/tracking/base.py,102,class,An interface for saving/restoring volatile Python state.
11661,feed_dict_additions,tensorflow/tensorflow/python/training/tracking/base.py,106,method,"When running a graph, indicates fresh state to feed.
Returns:
A dictionary mapping `Tensor`s to current Python state."
11662,freeze,tensorflow/tensorflow/python/training/tracking/base.py,115,method,"Create a new `SaveableObject` which freezes current state as a constant.
Used when executing eagerly to embed the current state as a constant, or
when creating a static tf.compat.v1.train.Saver with the frozen current
Python state.
Returns:
A `SaveableObject` which is not a `PythonStateSaveable` instance (i.e. has
no Python state associated with it)."
11663,PythonStringStateSaveable,tensorflow/tensorflow/python/training/tracking/base.py,129,class,Saves Python state in a checkpoint.
11664,optional_restore,tensorflow/tensorflow/python/training/tracking/base.py,159,method,"For values with no restore, relaxes assert_consumed()."
11665,feed_dict_additions,tensorflow/tensorflow/python/training/tracking/base.py,163,method,"When running a graph, indicates fresh state to feed."
11666,freeze,tensorflow/tensorflow/python/training/tracking/base.py,167,method,Create a frozen `SaveableObject` which saves the current state.
11667,python_restore,tensorflow/tensorflow/python/training/tracking/base.py,179,method,Called to restore Python state.
11668,restore,tensorflow/tensorflow/python/training/tracking/base.py,185,method,Called to restore TensorFlow state (nothing to do).
11669,CheckpointPosition,tensorflow/tensorflow/python/training/tracking/base.py,190,class,Indicates a position within a `_CheckpointRestoreCoordinator`.
11670,restore,tensorflow/tensorflow/python/training/tracking/base.py,205,method,Restore this value into `trackable`.
11671,bind_object,tensorflow/tensorflow/python/training/tracking/base.py,215,method,"Set a checkpoint<->object correspondence and process slot variables.
Args:
trackable: The object to record a correspondence for.
Returns:
True if this is a new assignment, False if this object has already been
mapped to a checkpointed `Object` proto.
Raises:
AssertionError: If another object is already bound to the `Object` proto."
11672,is_simple_variable,tensorflow/tensorflow/python/training/tracking/base.py,282,method,Determine whether this value is restorable with a Tensor initializer.
11673,value_tensors,tensorflow/tensorflow/python/training/tracking/base.py,289,method,"Create value `Tensor`s for this object's attributes.
Does not require that the Python object has been created. Used for
restore-on-create when executing eagerly.
Returns:
A dictionary mapping from object attribute names to `Tensor`s."
11674,gather_ops_or_named_saveables,tensorflow/tensorflow/python/training/tracking/base.py,317,method,Looks up or creates SaveableObjects which don't have cached ops.
11675,restore_ops,tensorflow/tensorflow/python/training/tracking/base.py,389,method,"Create or fetch restore ops for this object's attributes.
Requires that the `Trackable` Python object has been bound to an object
ID in the checkpoint.
Returns:
A list of operations when graph building, or an empty list when executing
eagerly."
11676,checkpoint,tensorflow/tensorflow/python/training/tracking/base.py,406,method,
11677,trackable,tensorflow/tensorflow/python/training/tracking/base.py,410,method,
11678,object_proto,tensorflow/tensorflow/python/training/tracking/base.py,414,method,
11679,restore_uid,tensorflow/tensorflow/python/training/tracking/base.py,418,method,
11680,no_automatic_dependency_tracking,tensorflow/tensorflow/python/training/tracking/base.py,443,function,"Disables automatic dependency tracking on attribute assignment.
Use to decorate any method of a Trackable object. Attribute assignment in
that method will not add dependencies (also respected in Model). Harmless if
used in a class which does not do automatic dependency tracking (which means
it's safe to use in base classes which may have subclasses which also inherit
from Trackable).
Args:
method: The method to decorate.
Returns:
A decorated method which sets and un-sets automatic dependency tracking for
the object the method is called on (not thread safe)."
11681,no_manual_dependency_tracking_scope,tensorflow/tensorflow/python/training/tracking/base.py,474,function,"A context that disables manual dependency tracking for the given `obj`.
Sometimes library methods might track objects on their own and we might want
to disable that and do the tracking on our own. One can then use this context
manager to disable the tracking the library method does and do your own
tracking.
For example:
class TestLayer(tf.keras.Layer):
def build():
with no_manual_dependency_tracking_scope(self):
var = self.add_variable(""name1"") # Creates a var and doesn't track it
self._track_trackable(""name2"", var) # We track variable with name `name2`
Args:
obj: A trackable object.
Yields:
a scope in which the object doesn't track dependencies manually."
11682,no_automatic_dependency_tracking_scope,tensorflow/tensorflow/python/training/tracking/base.py,506,function,"A context that disables automatic dependency tracking when assigning attrs.
Objects that inherit from Autotrackable automatically creates dependencies
to trackable objects through attribute assignments, and wraps data structures
(lists or dicts) with trackable classes. This scope may be used to temporarily
disable this behavior. This works similar to the decorator
`no_automatic_dependency_tracking`.
Example usage:
```
model = tf.keras.Model()
model.arr1 = [] # Creates a ListWrapper object
with no_automatic_dependency_tracking_scope(model):
model.arr2 = [] # Creates a regular, untracked python list
```
Args:
obj: A trackable object.
Yields:
a scope in which the object doesn't track dependencies."
11683,Trackable,tensorflow/tensorflow/python/training/tracking/base.py,537,class,"Base class for `Trackable` objects without automatic dependencies.
This class has no __setattr__ override for performance reasons. Dependencies
must be added explicitly. Unless attribute assignment is performance-critical,
use `AutoTrackable` instead. Use `Trackable` for `isinstance`
checks."
11684,SavingBenchmarks,tensorflow/tensorflow/python/training/tracking/benchmarks_test.py,75,class,
11685,benchmark_baseline_no_restore,tensorflow/tensorflow/python/training/tracking/benchmarks_test.py,89,method,
11686,benchmark_batch_restore,tensorflow/tensorflow/python/training/tracking/benchmarks_test.py,97,method,
11687,benchmark_restore_on_create,tensorflow/tensorflow/python/training/tracking/benchmarks_test.py,107,method,
11688,benchmark_raw_restore,tensorflow/tensorflow/python/training/tracking/benchmarks_test.py,117,method,
11689,NoDependency,tensorflow/tensorflow/python/training/tracking/data_structures.py,41,class,"Allows attribute assignment to `Trackable` objects with no dependency.
Example usage:
```python
obj = Trackable()
obj.has_dependency = tf.Variable(0., name=""dep"")
obj.no_dependency = NoDependency(tf.Variable(1., name=""nodep""))
assert obj.no_dependency.name == ""nodep:0""
```
`obj` in this example has a dependency on the variable ""dep"", and both
attributes contain un-wrapped `Variable` objects.
`NoDependency` also works with `tf.keras.Model`, but only for checkpoint
dependencies: wrapping a `Layer` in `NoDependency` will assign the (unwrapped)
`Layer` to the attribute without a checkpoint dependency, but the `Model` will
still track the `Layer` (so it will appear in `Model.layers`, and its
variables will appear in `Model.variables`)."
11690,wrap_or_unwrap,tensorflow/tensorflow/python/training/tracking/data_structures.py,82,function,"Wraps basic data structures, unwraps NoDependency objects."
11691,sticky_attribute_assignment,tensorflow/tensorflow/python/training/tracking/data_structures.py,105,function,"Adds dependencies, generally called from __setattr__.
This behavior is shared between Trackable and Model.
Respects NoDependency indicators, but otherwise makes trackable objects
out of common data structures and tracks objects by their attribute names.
Args:
trackable: The object to add dependencies to (generally the one having
an attribute assigned).
name: The attribute name being assigned.
value: The value being assigned. Not necessarily a trackable object.
Returns:
The value which should be stored in the attribute (unwrapped from a
NoDependency object if necessary)."
11692,TrackableDataStructure,tensorflow/tensorflow/python/training/tracking/data_structures.py,151,class,Base class for data structures which contain trackable objects.
11693,trainable,tensorflow/tensorflow/python/training/tracking/data_structures.py,168,method,
11694,trainable,tensorflow/tensorflow/python/training/tracking/data_structures.py,172,method,
11695,layers,tensorflow/tensorflow/python/training/tracking/data_structures.py,211,method,
11696,trainable_weights,tensorflow/tensorflow/python/training/tracking/data_structures.py,215,method,
11697,non_trainable_weights,tensorflow/tensorflow/python/training/tracking/data_structures.py,222,method,
11698,weights,tensorflow/tensorflow/python/training/tracking/data_structures.py,229,method,
11699,trainable_variables,tensorflow/tensorflow/python/training/tracking/data_structures.py,233,method,
11700,non_trainable_variables,tensorflow/tensorflow/python/training/tracking/data_structures.py,237,method,
11701,variables,tensorflow/tensorflow/python/training/tracking/data_structures.py,241,method,
11702,updates,tensorflow/tensorflow/python/training/tracking/data_structures.py,245,method,Aggregate updates from any `Layer` instances.
11703,losses,tensorflow/tensorflow/python/training/tracking/data_structures.py,257,method,Aggregate losses from any `Layer` instances.
11704,List,tensorflow/tensorflow/python/training/tracking/data_structures.py,276,class,"An append-only sequence type which is trackable.
Maintains checkpoint dependencies on its contents (which must also be
trackable), and forwards any `Layer` metadata such as updates and losses.
Note that `List` is purely a container. It lets a `tf.keras.Model` or
other trackable object know about its contents, but does not call any
`Layer` instances which are added to it. To indicate a sequence of `Layer`
instances which should be called sequentially, use `tf.keras.Sequential`.
Example usage:
```python
class HasList(tf.keras.Model):
def __init__(self):
super(HasList, self).__init__()
self.layer_list = List([layers.Dense(3)])
self.layer_list.append(layers.Dense(4))
def call(self, x):
aggregation = 0.
for l in self.layer_list:
x = l(x)
aggregation += tf.reduce_sum(x)
return aggregation
```
This kind of wrapping is necessary because `Trackable` objects do not
(yet) deeply inspect regular Python data structures, so for example assigning
a regular list (`self.layer_list = [layers.Dense(3)]`) does not create a
checkpoint dependency and does not add the `Layer` instance's weights to its
parent `Model`."
11705,copy,tensorflow/tensorflow/python/training/tracking/data_structures.py,319,method,
11706,append,tensorflow/tensorflow/python/training/tracking/data_structures.py,340,method,Add a new trackable value.
11707,extend,tensorflow/tensorflow/python/training/tracking/data_structures.py,345,method,Add a sequence of trackable values.
11708,ListWrapper,tensorflow/tensorflow/python/training/tracking/data_structures.py,398,class,"Wraps the built-in `list` to support restore-on-create for variables.
Unlike `List`, this sequence type is mutable in the same ways built-in lists
are. Instead of throwing an error immediately like `List`, it records
problematic mutations (e.g. assigning a new element to a position already
occupied, meaning both elements get the same names at different times) and
refuses to save.
On assignment to an attribute of a Model or Trackable object, Python
lists are replaced with ListWrapper. Wrapping a list in a
`NoDependency` object prevents this."
11709,append,tensorflow/tensorflow/python/training/tracking/data_structures.py,566,method,Add a new trackable value.
11710,extend,tensorflow/tensorflow/python/training/tracking/data_structures.py,572,method,Add a sequence of trackable values.
11711,insert,tensorflow/tensorflow/python/training/tracking/data_structures.py,613,method,
11712,sort,tensorflow/tensorflow/python/training/tracking/data_structures.py,620,method,
11713,Mapping,tensorflow/tensorflow/python/training/tracking/data_structures.py,658,class,"An append-only trackable mapping data structure with string keys.
Maintains checkpoint dependencies on its contents (which must also be
trackable), named based on its keys.
Note that once a key has been added, it may not be deleted or replaced."
11714,update,tensorflow/tensorflow/python/training/tracking/data_structures.py,711,method,
11715,ObjectGraphView,tensorflow/tensorflow/python/training/tracking/graph_view.py,142,class,Gathers and serializes an object graph.
11716,list_dependencies,tensorflow/tensorflow/python/training/tracking/graph_view.py,158,method,
11717,saveables_cache,tensorflow/tensorflow/python/training/tracking/graph_view.py,165,method,"Maps Trackable objects -> attribute names -> list(SaveableObjects).
Used to avoid re-creating SaveableObjects when graph building. None when
executing eagerly.
Returns:
The cache (an object-identity dictionary), or None if caching is disabled."
11718,root,tensorflow/tensorflow/python/training/tracking/graph_view.py,177,method,
11719,serialize_object_graph,tensorflow/tensorflow/python/training/tracking/graph_view.py,364,method,"Determine checkpoint keys for variables and build a serialized graph.
Non-slot variables are keyed based on a shortest path from the root saveable
to the object which owns the variable (i.e. the one which called
`Trackable._add_variable` to create it).
Slot variables are keyed based on a shortest path to the variable being
slotted for, a shortest path to their optimizer, and the slot name.
Returns:
A tuple of (named_variables, object_graph_proto, feed_additions):
named_variables: A dictionary mapping names to variable objects.
object_graph_proto: A TrackableObjectGraph protocol buffer
containing the serialized object graph and variable references.
feed_additions: A dictionary mapping from Tensors to values which should
be fed when saving.
Raises:
ValueError: If there are invalid characters in an optimizer's slot names."
11720,frozen_saveable_objects,tensorflow/tensorflow/python/training/tracking/graph_view.py,389,method,Creates SaveableObjects with the current object graph frozen.
11721,objects_ids_and_slot_variables,tensorflow/tensorflow/python/training/tracking/graph_view.py,412,method,"Traverse the object graph and list all accessible objects.
Looks for `Trackable` objects which are dependencies of
`root_trackable`. Includes slot variables only if the variable they are
slotting for and the optimizer are dependencies of `root_trackable`
(i.e. if they would be saved with a checkpoint).
Returns:
A tuple of (trackable objects, object -> node id, slot variables)"
11722,list_objects,tensorflow/tensorflow/python/training/tracking/graph_view.py,436,method,Traverse the object graph and list all accessible objects.
11723,is_layer,tensorflow/tensorflow/python/training/tracking/layer_utils.py,37,function,Implicit check for Layer-like objects.
11724,has_weights,tensorflow/tensorflow/python/training/tracking/layer_utils.py,43,function,Implicit check for Layer-like objects.
11725,invalidate_recursive_cache,tensorflow/tensorflow/python/training/tracking/layer_utils.py,52,function,Convenience decorator to invalidate the cache when setting attributes.
11726,MutationSentinel,tensorflow/tensorflow/python/training/tracking/layer_utils.py,64,class,Container for tracking whether a property is in a cached state.
11727,mark_as,tensorflow/tensorflow/python/training/tracking/layer_utils.py,68,method,
11728,in_cached_state,tensorflow/tensorflow/python/training/tracking/layer_utils.py,74,method,
11729,AttributeSentinel,tensorflow/tensorflow/python/training/tracking/layer_utils.py,78,class,"Container for managing attribute cache state within a Layer.
The cache can be invalidated either on an individual basis (for instance when
an attribute is mutated) or a layer-wide basis (such as when a new dependency
is added)."
11730,add_parent,tensorflow/tensorflow/python/training/tracking/layer_utils.py,101,method,
11731,get,tensorflow/tensorflow/python/training/tracking/layer_utils.py,112,method,
11732,mark_cached,tensorflow/tensorflow/python/training/tracking/layer_utils.py,123,method,
11733,invalidate,tensorflow/tensorflow/python/training/tracking/layer_utils.py,127,method,
11734,invalidate_all,tensorflow/tensorflow/python/training/tracking/layer_utils.py,131,method,
11735,filter_empty_layer_containers,tensorflow/tensorflow/python/training/tracking/layer_utils.py,141,function,Filter out empty Layer-like containers and uniquify.
11736,gather_trainable_weights,tensorflow/tensorflow/python/training/tracking/layer_utils.py,161,function,"Lists the trainable weights for an object with sub-layers.
Args:
trainable: Whether the object collecting the variables is trainable.
sub_layers: A flat list of Layer objects owned by this object, to collect
variables from.
extra_variables: Any extra variables to include. Their `.trainable` property
is used to categorize them.
Returns:
A list of collected trainable weights/variables."
11737,gather_non_trainable_weights,tensorflow/tensorflow/python/training/tracking/layer_utils.py,184,function,"Lists the non-trainable weights for an object with sub-layers.
Args:
trainable: Whether the object collecting the variables is trainable.
sub_layers: A flat list of Layer objects owned by this object, to collect
variables from.
extra_variables: Any extra variables to include. Their `.trainable` property
is used to categorize them.
Returns:
A list of collected non-trainable weights/variables."
11738,PythonState,tensorflow/tensorflow/python/training/tracking/python_state.py,31,class,"A mixin for putting Python state in an object-based checkpoint.
This is an abstract class which allows extensions to TensorFlow's object-based
checkpointing (see `tf.train.Checkpoint`). For example a wrapper for NumPy
arrays:
```python
import io
import numpy
class NumpyWrapper(tf.train.experimental.PythonState):
def __init__(self, array):
self.array = array
def serialize(self):
string_file = io.BytesIO()
try:
numpy.save(string_file, self.array, allow_pickle=False)
serialized = string_file.getvalue()
finally:
string_file.close()
return serialized
def deserialize(self, string_value):
string_file = io.BytesIO(string_value)
try:
self.array = numpy.load(string_file, allow_pickle=False)
finally:
string_file.close()
```
Instances of `NumpyWrapper` are checkpointable objects, and will be saved and
restored from checkpoints along with TensorFlow state like variables.
```python
root = tf.train.Checkpoint(numpy=NumpyWrapper(numpy.array([1.])))
save_path = root.save(prefix)
root.numpy.array *= 2.
assert [2.] == root.numpy.array
root.restore(save_path)
assert [1.] == root.numpy.array
```"
11739,serialize,tensorflow/tensorflow/python/training/tracking/python_state.py,78,method,Callback to serialize the object. Returns a string.
11740,deserialize,tensorflow/tensorflow/python/training/tracking/python_state.py,82,method,Callback to deserialize the object.
11741,NotTrackable,tensorflow/tensorflow/python/training/tracking/tracking.py,41,class,"Marks instances of child classes as unsaveable using an object-based API.
Useful for marking objects which would otherwise look trackable because
of inheritance (e.g. through `Layer`) as not trackable. Inheriting from
`NotTrackable` does not prevent an object from being assigned to any
attributes, but will throw an error on save/restore."
11742,AutoTrackable,tensorflow/tensorflow/python/training/tracking/tracking.py,52,class,"Manages dependencies on other objects.
`Trackable` objects may have dependencies: other `Trackable` objects
which should be saved if the object declaring the dependency is saved. A
correctly saveable program has a dependency graph such that if changing a
global variable affects an object (e.g. changes the behavior of any of its
methods) then there is a chain of dependencies from the influenced object to
the variable.
Dependency edges have names, and are created implicitly when a
`Trackable` object is assigned to an attribute of another
`Trackable` object. For example:
```
obj = Trackable()
obj.v = ResourceVariable(0.)
```
The `Trackable` object `obj` now has a dependency named ""v"" on a
variable.
`Trackable` objects may specify `Tensor`s to be saved and restored
directly (e.g. a `Variable` indicating how to save itself) rather than through
dependencies on other objects. See
`Trackable._gather_saveables_for_checkpoint` for details."
11743,delete_tracking,tensorflow/tensorflow/python/training/tracking/tracking.py,127,function,Removes the tracking of name from object.
11744,ResourceTracker,tensorflow/tensorflow/python/training/tracking/tracking.py,140,class,An object that tracks a list of resources.
11745,resources,tensorflow/tensorflow/python/training/tracking/tracking.py,149,method,
11746,add_resource,tensorflow/tensorflow/python/training/tracking/tracking.py,152,method,
11747,resource_tracker_scope,tensorflow/tensorflow/python/training/tracking/tracking.py,157,function,"A context to manage resource trackers.
Use this in order to collect up all resources created within a block of code.
Example usage:
```python
resource_tracker = ResourceTracker()
with resource_tracker_scope(resource_tracker):
resource = TrackableResource()
assert resource_tracker.resources == [resource]
Args:
resource_tracker: The passed in ResourceTracker object
Yields:
A scope in which the resource_tracker is active."
11748,CapturableResourceDeleter,tensorflow/tensorflow/python/training/tracking/tracking.py,185,class,Deleter to destroy CapturableResource without overriding its __del__().
11749,destroy_resource,tensorflow/tensorflow/python/training/tracking/tracking.py,199,method,
11750,CapturableResource,tensorflow/tensorflow/python/training/tracking/tracking.py,209,class,"Holds a Tensor which a tf.function can capture.
`CapturableResource`s are discovered by traversing the graph of object
attributes, e.g. during `tf.saved_model.save`. They are excluded from the
scope-based tracking of `TrackableResource`; generally things that require
initialization should inherit from `TrackableResource` instead of
`CapturableResource` directly."
11751,resource_handle,tensorflow/tensorflow/python/training/tracking/tracking.py,244,method,Returns the resource handle associated with this Resource.
11752,TrackableResource,tensorflow/tensorflow/python/training/tracking/tracking.py,286,class,Adds scope tracking to CapturableResource.
11753,Asset,tensorflow/tensorflow/python/training/tracking/tracking.py,307,class,"Represents a file asset to hermetically include in a SavedModel.
A SavedModel can include arbitrary files, called assets, that are needed
for its use. For example a vocabulary file used initialize a lookup table.
When a trackable object is exported via `tf.saved_model.save()`, all the
`Asset`s reachable from it are copied into the SavedModel assets directory.
Upon loading, the assets and the serialized functions that depend on them
will refer to the correct filepaths inside the SavedModel directory.
Example:
```
filename = tf.saved_model.Asset(""file.txt"")
@tf.function(input_signature=[])
def func():
return tf.io.read_file(filename)
trackable_obj = tf.train.Checkpoint()
trackable_obj.func = func
trackable_obj.filename = filename
tf.saved_model.save(trackable_obj, ""/tmp/saved_model"")
# The created SavedModel is hermetic, it does not depend on
# the original file and can be moved to another path.
tf.io.gfile.remove(""file.txt"")
tf.io.gfile.rename(""/tmp/saved_model"", ""/tmp/new_location"")
reloaded_obj = tf.saved_model.load(""/tmp/new_location"")
print(reloaded_obj.func())
```
Attributes:
asset_path: A 0-D `tf.string` tensor with path to the asset."
11754,asset_path,tensorflow/tensorflow/python/training/tracking/tracking.py,355,method,Fetch the current asset path.
11755,cached_per_instance,tensorflow/tensorflow/python/training/tracking/tracking.py,360,function,"Lightweight decorator for caching lazily constructed properties.
When to use:
This decorator provides simple caching with minimal overhead. It is designed
for properties which are expensive to compute and static over the life of a
class instance, and provides no mechanism for cache invalidation. Thus it is
best suited for lazily exposing derived properties of other static data.
For classes with custom getattr / setattr behavior (such as trackable
objects), storing cache results as object attributes is not performant.
Instead, a specialized cache can significantly reduce property lookup
overhead. (While still allowing the decorated property to be lazily computed.)
Consider the following class:
```
class MyClass(object):
def __setattr__(self, key, value):
# Some expensive class specific code
# ...
# ...
super(MyClass, self).__setattr__(key, value)
@property
def thing(self):
# `thing` is expensive to compute (and may not even be requested), so we
# want to lazily compute it and then cache it.
output = getattr(self, '_thing', None)
if output is None:
self._thing = output = compute_thing(self)
return output
```
It's also worth noting that ANY overriding of __setattr__, even something as
simple as:
```
def __setattr__(self, key, value):
super(MyClass, self).__setattr__(key, value)
```
Slows down attribute assignment by nearly 10x.
By contrast, replacing the definition of `thing` with the following sidesteps
the expensive __setattr__ altogether:
'''
@property
@tracking.cached_per_instance
def thing(self):
# `thing` is expensive to compute (and may not even be requested), so we
# want to lazily compute it and then cache it.
return compute_thing(self)
'''
Performance:
The overhead for this decorator is ~0.4 us / call. A much lower overhead
implementation (~0.085 us / call) can be achieved by using a custom dict type:
```
def dict_based_cache(f):
class Cache(dict):
__slots__ = ()
def __missing__(self, key):
self[key] = output = f(key)
return output
return property(Cache().__getitem__)
```
However, that implementation holds class instances as keys, and as a result
blocks garbage collection. (And modifying it to use weakref's as keys raises
the lookup overhead to ~0.4 us) As a result, the WeakKeyDictionary
implementation below turns out to be more prudent.
Args:
f: The function to cache.
Returns:
f decorated with simple caching behavior."
11756,MyPickleableObject,tensorflow/tensorflow/python/training/tracking/tracking_test.py,41,class,"Needed for InterfaceTests.test_property_cache_serialization.
This class must be at the top level. This is a constraint of pickle,
unrelated to `cached_per_instance`."
11757,my_id,tensorflow/tensorflow/python/training/tracking/tracking_test.py,50,method,
11758,register_session_provider,tensorflow/tensorflow/python/training/tracking/util.py,65,function,
11759,get_session,tensorflow/tensorflow/python/training/tracking/util.py,71,function,
11760,add_variable,tensorflow/tensorflow/python/training/tracking/util.py,441,function,Add a variable to a Trackable with no scope influence.
11761,object_metadata,tensorflow/tensorflow/python/training/tracking/util.py,457,function,"Retrieves information about the objects in a checkpoint.
Example usage:
```python
object_graph = tf.contrib.checkpoint.object_metadata(
tf.train.latest_checkpoint(checkpoint_directory))
ckpt_variable_names = set()
for node in object_graph.nodes:
for attribute in node.attributes:
ckpt_variable_names.add(attribute.full_name)
```
Args:
save_path: The path to the checkpoint, as returned by `save` or
`tf.train.latest_checkpoint`.
Returns:
A parsed `tf.contrib.checkpoint.TrackableObjectGraph` protocol buffer.
Raises:
ValueError: If an object graph was not found in the checkpoint."
11762,list_objects,tensorflow/tensorflow/python/training/tracking/util.py,494,function,"Traverse the object graph and list all accessible objects.
Looks for `Trackable` objects which are dependencies of
`root_trackable`. Includes slot variables only if the variable they are
slotting for and the optimizer are dependencies of `root_trackable`
(i.e. if they would be saved with a checkpoint).
Args:
root_trackable: A `Trackable` object whose dependencies should be flattened.
Returns:
A flat list of objects."
11763,gather_initializers,tensorflow/tensorflow/python/training/tracking/util.py,511,function,"Traverse the object graph and find initialization ops.
Looks for `Trackable` objects which are dependencies of
`root_trackable` and which have an `initializer` property. Includes
initializers for slot variables only if the variable they are slotting for and
the optimizer are dependencies of `root_trackable` (i.e. if they would be
saved with a checkpoint).
Args:
root_trackable: A `Trackable` object to gather initializers for.
Returns:
A list of initialization ops."
11764,capture_dependencies,tensorflow/tensorflow/python/training/tracking/util.py,535,function,"Capture variables created within this scope as `Template` dependencies.
Requires that `template.variable_scope` is active.
This scope is intended as a compatibility measure, allowing a trackable
object to add dependencies on variables created in a block of code which is
not aware of object-based saving (and instead uses variable names
heavily). This is how `Template` objects add dependencies on variables and
sub-`Template`s. Where possible, use `tf.compat.v1.make_template` directly.
Args:
template: The `Template` object to register dependencies with.
Yields:
None (when used as a context manager)."
11765,streaming_restore,tensorflow/tensorflow/python/training/tracking/util.py,658,function,"When graph building, runs restore ops as soon as they come in.
Args:
status: A _LoadStatus objects from an object-based saver's restore().
Streaming restore from name-based checkpoints is not currently supported.
session: A session to run new restore ops in."
11766,CheckpointLoadStatus,tensorflow/tensorflow/python/training/tracking/util.py,689,class,"Checks the status of checkpoint loading and manages restore ops.
Returned from `Saver.restore`. Since `restore` may defer the loading of values
in the checkpoint which don't yet have corresponding Python objects,
`CheckpointLoadStatus` provides a callback to verify that checkpoint loading
is complete (`assert_consumed`).
When graph building, `restore` does not run restore ops itself since their
creation may be deferred. The `run_restore_ops` method must be called once all
Python objects with values to restore have been created and added to the
dependency graph (this does not necessarily have to be the whole checkpoint;
calling `run_restore_ops` while `assert_consumed` fails is supported and will
partially restore the checkpoint).
See `Saver.restore` for usage examples."
11767,assert_consumed,tensorflow/tensorflow/python/training/tracking/util.py,714,method,"Asserts that all objects in the checkpoint have been created/matched.
Returns:
`self` for chaining.
Raises:
AssertionError: If there are any Python objects in the dependency graph
which have not been restored from this checkpoint or a later `restore`,
or if there are any checkpointed values which have not been matched to
Python objects."
11768,assert_existing_objects_matched,tensorflow/tensorflow/python/training/tracking/util.py,757,method,"Asserts that trackable Python objects have been matched.
Note that this is a weaker assertion than `assert_consumed`. It will only
fail for existing Python objects which are (transitive) dependencies of the
root object and which do not have an entry in the checkpoint.
It will not fail, for example, if a `tf.keras.Layer` object has not yet been
built and so has not created any `tf.Variable` objects.
Returns:
`self` for chaining.
Raises:
AssertionError: If a Python object exists in the transitive dependencies
of the root object but does not have a value in the checkpoint."
11769,assert_nontrivial_match,tensorflow/tensorflow/python/training/tracking/util.py,801,method,Raises an exception if only the root object matched.
11770,run_restore_ops,tensorflow/tensorflow/python/training/tracking/util.py,823,method,Run operations to restore objects in the dependency graph.
11771,initialize_or_restore,tensorflow/tensorflow/python/training/tracking/util.py,831,method,"Run operations to initialize or restore objects in the dependency graph.
Any objects in the dependency graph which have initializers but are not in
the checkpoint will have those initializers run, unless those variables are
being restored by a later call to `tf.train.Checkpoint.restore()`.
This method has a sibling in `InitializationOnlyStatus` which instead
initializes variables. That type is returned if no checkpoint is specified
in `Saver.restore`.
Args:
session: The session to run init/restore ops in. If `None`, uses the
default session."
11772,expect_partial,tensorflow/tensorflow/python/training/tracking/util.py,862,method,Silence warnings about incomplete checkpoint restores.
11773,InitializationOnlyStatus,tensorflow/tensorflow/python/training/tracking/util.py,868,class,"Returned from `Saver.restore` when no checkpoint has been specified.
Objects of this type have the same `assert_consumed` method as
`CheckpointLoadStatus`, but it always fails. However,
`initialize_or_restore` works on objects of both types, and will
initialize variables in `InitializationOnlyStatus` objects or restore them
otherwise."
11774,assert_consumed,tensorflow/tensorflow/python/training/tracking/util.py,884,method,Assertion for consistency with `CheckpointLoadStatus`. Always fails.
11775,assert_existing_objects_matched,tensorflow/tensorflow/python/training/tracking/util.py,889,method,Assertion for consistency with `CheckpointLoadStatus`. Always fails.
11776,assert_nontrivial_match,tensorflow/tensorflow/python/training/tracking/util.py,894,method,Assertion for consistency with `CheckpointLoadStatus`. Always fails.
11777,run_restore_ops,tensorflow/tensorflow/python/training/tracking/util.py,899,method,"For consistency with `CheckpointLoadStatus`.
Use `initialize_or_restore` for initializing if no checkpoint was passed
to `Saver.restore` and restoring otherwise.
Args:
session: Not used."
11778,initialize_or_restore,tensorflow/tensorflow/python/training/tracking/util.py,912,method,"Runs initialization ops for variables.
Objects which would be saved by `Saver.save` will be initialized, unless
those variables are being restored by a later call to
`tf.train.Checkpoint.restore()`.
This method does nothing when executing eagerly (initializers get run
eagerly).
Args:
session: The session to run initialization ops in. If `None`, uses the
default session."
11779,NameBasedSaverStatus,tensorflow/tensorflow/python/training/tracking/util.py,948,class,Status for loading a name-based training checkpoint.
11780,add_to_optionally_restored,tensorflow/tensorflow/python/training/tracking/util.py,962,method,"Add a variable to the list of optionally restored variables.
There are situations where certain variables should be ignored in assertions
such as assert_existing_objects_matched(). One example is that of a
checkpoint saved with train.Saver(), and restored with train.Checkpoint():
it is possible for the train.Saver() checkpoint to be missing the internal
`save_counter` variable, which we want to ignore on restore.
Args:
var: The variable to treat as optionally restored."
11781,assert_consumed,tensorflow/tensorflow/python/training/tracking/util.py,976,method,Raises an exception if any variables are unmatched.
11782,assert_existing_objects_matched,tensorflow/tensorflow/python/training/tracking/util.py,999,method,Raises an exception if currently created objects are unmatched.
11783,assert_nontrivial_match,tensorflow/tensorflow/python/training/tracking/util.py,1007,method,Raises an exception if currently created objects are unmatched.
11784,run_restore_ops,tensorflow/tensorflow/python/training/tracking/util.py,1031,method,Load the name-based checkpoint using a new `tf.compat.v1.train.Saver`.
11785,initialize_or_restore,tensorflow/tensorflow/python/training/tracking/util.py,1042,method,Alias for `run_restore_ops`.
11786,TrackableSaver,tensorflow/tensorflow/python/training/tracking/util.py,1064,class,"Saves and restores a `Trackable` object and its dependencies.
See `Trackable` for details of dependency management. `Saver` wraps
`tf.compat.v1.train.Saver` for saving, including extra information about the
graph of
dependencies between Python objects. When restoring, it uses this information
about the save-time dependency graph to more robustly match objects with their
checkpointed values. When executing eagerly, it supports restoring variables
on object creation (see `Saver.restore`).
Values in a checkpoint are mapped to `Trackable` Python objects
(`Variable`s, `Optimizer`s, `Layer`s) based on the names provided when the
checkpoint was written. To avoid breaking existing checkpoints when modifying
a class, dependency names (the names of attributes to which `Trackable`
objects are assigned) may not change. These names are local to objects, in
contrast to the `Variable.name`-based save/restore from
`tf.compat.v1.train.Saver`, and
so allow additional program transformations."
11787,save,tensorflow/tensorflow/python/training/tracking/util.py,1159,method,"Save a training checkpoint.
The saved checkpoint includes variables created by this object and any
Trackable objects it depends on at the time `Saver.save()` is called.
Args:
file_prefix: A prefix to use for the checkpoint filenames
(/path/to/directory/and_a_prefix). Names are generated based on this
prefix and `checkpoint_number`, if provided.
checkpoint_number: An integer variable or Tensor, used to number
checkpoints. Typically this value is saved along with other variables in
training checkpoints, which will happen automatically if it was created
by `root_trackable` or one of its dependencies (via
`Trackable._add_variable`).
session: The session to evaluate variables in. Ignored when executing
eagerly. If not provided when graph building, the default session is
used.
options: Optional `tf.train.CheckpointOptions` object.
Returns:
The full path to the checkpoint."
11788,restore,tensorflow/tensorflow/python/training/tracking/util.py,1220,method,"Restore a training checkpoint.
Restores `root_trackable` and any objects that it tracks
(transitive). Either assigns values immediately if variables to restore have
been created already, or defers restoration until the variables are
created. Dependencies added to the `root_trackable` passed to the
constructor after this call will be matched if they have a corresponding
object in the checkpoint.
When building a graph, restorations are added to the graph but not run.
To disallow deferred loading, assert immediately that all checkpointed
variables have been matched to variable objects:
```python
saver = Saver(root)
saver.restore(path).assert_consumed()
```
An exception will be raised unless every object was matched and its
variables already exist.
When graph building, `assert_consumed()` indicates that all of the restore
ops which will be created for this checkpoint have been created. They can be
run via the `run_restore_ops()` function of the status object:
```python
saver.restore(path).assert_consumed().run_restore_ops()
```
If the checkpoint has not been consumed completely, then the list of restore
ops will grow as more objects are added to the dependency graph.
Name-based `tf.compat.v1.train.Saver` checkpoints can be loaded using this
method. There is no deferred loading, and names are used to match
variables. No restore ops are created/run until `run_restore_ops()` or
`initialize_or_restore()` are called on the returned status object, even
when executing eagerly. Re-encode name-based checkpoints using this
object-based `Saver.save` as soon as possible.
Args:
save_path: The path to the checkpoint, as returned by `save` or
`tf.train.latest_checkpoint`. If None (as when there is no latest
checkpoint for `tf.train.latest_checkpoint` to return), returns an
object which may run initializers for objects in the dependency graph.
If the checkpoint was written by the name-based
`tf.compat.v1.train.Saver`, names are used to match variables.
options: Optional `tf.train.CheckpointOptions` object.
Returns:
A load status object, which can be used to make assertions about the
status of checkpoint restoration and run initialization/restore ops
(of type `CheckpointLoadStatus`, or `InitializationOnlyStatus` if
`save_path` is `None`).
If `save_path` points to a name-based checkpoint, a `NameBasedSaverStatus`
object is returned which runs restore ops from a name-based saver."
11789,frozen_saver,tensorflow/tensorflow/python/training/tracking/util.py,1335,function,"Creates a static `tf.compat.v1.train.Saver` from a trackable object.
The returned `Saver` saves object-based checkpoints, but these checkpoints
will no longer reflect structural changes to the object graph, only changes to
the values of `Variable`s added as dependencies of the root object before
`freeze` was called.
`restore` works on the returned `Saver`, but requires that the object graph of
the checkpoint being loaded exactly matches the object graph when `freeze` was
called. This is in contrast the object-based restore performed by
`tf.train.Checkpoint` which attempts a fuzzy matching between a checkpoint's
object graph and the current Python object graph.
Args:
root_trackable: A trackable object to save.
Returns:
A saver which saves object-based checkpoints for the object graph frozen at
the time `frozen_saver` was called."
11790,saver_with_op_caching,tensorflow/tensorflow/python/training/tracking/util.py,1361,function,A TrackableSaver with a SaveableObject cache when graph building.
11791,CheckpointV1,tensorflow/tensorflow/python/training/tracking/util.py,1374,class,"Groups trackable objects, saving and restoring them.
`Checkpoint`'s constructor accepts keyword arguments whose values are types
that contain trackable state, such as `tf.compat.v1.train.Optimizer`
implementations, `tf.Variable`, `tf.keras.Layer` implementations, or
`tf.keras.Model` implementations. It saves these values with a checkpoint, and
maintains a `save_counter` for numbering checkpoints.
Example usage when graph building:
```python
import tensorflow as tf
import os
checkpoint_directory = ""/tmp/training_checkpoints""
checkpoint_prefix = os.path.join(checkpoint_directory, ""ckpt"")
checkpoint = tf.train.Checkpoint(optimizer=optimizer, model=model)
status = checkpoint.restore(tf.train.latest_checkpoint(checkpoint_directory))
train_op = optimizer.minimize( ... )
status.assert_consumed() # Optional sanity checks.
with tf.compat.v1.Session() as session:
# Use the Session to restore variables, or initialize them if
# tf.train.latest_checkpoint returned None.
status.initialize_or_restore(session)
for _ in range(num_training_steps):
session.run(train_op)
checkpoint.save(file_prefix=checkpoint_prefix)
```
Example usage with eager execution enabled:
```python
import tensorflow as tf
import os
tf.compat.v1.enable_eager_execution()
checkpoint_directory = ""/tmp/training_checkpoints""
checkpoint_prefix = os.path.join(checkpoint_directory, ""ckpt"")
checkpoint = tf.train.Checkpoint(optimizer=optimizer, model=model)
status = checkpoint.restore(tf.train.latest_checkpoint(checkpoint_directory))
for _ in range(num_training_steps):
optimizer.minimize( ... ) # Variables will be restored on creation.
status.assert_consumed() # Optional sanity checks.
checkpoint.save(file_prefix=checkpoint_prefix)
```
`Checkpoint.save` and `Checkpoint.restore` write and read object-based
checkpoints, in contrast to `tf.compat.v1.train.Saver` which writes and reads
`variable.name` based checkpoints. Object-based checkpointing saves a graph of
dependencies between Python objects (`Layer`s, `Optimizer`s, `Variable`s,
etc.) with named edges, and this graph is used to match variables when
restoring a checkpoint. It can be more robust to changes in the Python
program, and helps to support restore-on-create for variables when executing
eagerly. Prefer `tf.train.Checkpoint` over `tf.compat.v1.train.Saver` for new
code.
`Checkpoint` objects have dependencies on the objects passed as keyword
arguments to their constructors, and each dependency is given a name that is
identical to the name of the keyword argument for which it was created.
TensorFlow classes like `Layer`s and `Optimizer`s will automatically add
dependencies on their variables (e.g. ""kernel"" and ""bias"" for
`tf.keras.layers.Dense`). Inheriting from `tf.keras.Model` makes managing
dependencies easy in user-defined classes, since `Model` hooks into attribute
assignment. For example:
```python
class Regress(tf.keras.Model):
def __init__(self):
super(Regress, self).__init__()
self.input_transform = tf.keras.layers.Dense(10)
# ...
def call(self, inputs):
x = self.input_transform(inputs)
# ...
```
This `Model` has a dependency named ""input_transform"" on its `Dense` layer,
which in turn depends on its variables. As a result, saving an instance of
`Regress` using `tf.train.Checkpoint` will also save all the variables created
by the `Dense` layer.
When variables are assigned to multiple workers, each worker writes its own
section of the checkpoint. These sections are then merged/re-indexed to behave
as a single checkpoint. This avoids copying all variables to one worker, but
does require that all workers see a common filesystem.
While `tf.keras.Model.save_weights` and `tf.train.Checkpoint.save` save in the
same format, note that the root of the resulting checkpoint is the object the
save method is attached to. This means saving a `tf.keras.Model` using
`save_weights` and loading into a `tf.train.Checkpoint` with a `Model`
attached (or vice versa) will not match the `Model`'s variables. See the
[guide to training
checkpoints](https://www.tensorflow.org/guide/checkpoint) for
details. Prefer `tf.train.Checkpoint` over `tf.keras.Model.save_weights` for
training checkpoints.
Attributes:
save_counter: Incremented when `save()` is called. Used to number
checkpoints."
11792,write,tensorflow/tensorflow/python/training/tracking/util.py,1521,method,"Writes a training checkpoint.
The checkpoint includes variables created by this object and any
trackable objects it depends on at the time `Checkpoint.write()` is
called.
`write` does not number checkpoints, increment `save_counter`, or update the
metadata used by `tf.train.latest_checkpoint`. It is primarily intended for
use by higher level checkpoint management utilities. `save` provides a very
basic implementation of these features.
Args:
file_prefix: A prefix to use for the checkpoint filenames
(/path/to/directory/and_a_prefix).
session: The session to evaluate variables in. Ignored when executing
eagerly. If not provided when graph building, the default session is
used.
Returns:
The full path to the checkpoint (i.e. `file_prefix`)."
11793,save_counter,tensorflow/tensorflow/python/training/tracking/util.py,1555,method,"An integer variable which starts at zero and is incremented on save.
Used to number checkpoints.
Returns:
The save counter variable."
11794,save,tensorflow/tensorflow/python/training/tracking/util.py,1566,method,"Saves a training checkpoint and provides basic checkpoint management.
The saved checkpoint includes variables created by this object and any
trackable objects it depends on at the time `Checkpoint.save()` is
called.
`save` is a basic convenience wrapper around the `write` method,
sequentially numbering checkpoints using `save_counter` and updating the
metadata used by `tf.train.latest_checkpoint`. More advanced checkpoint
management, for example garbage collection and custom numbering, may be
provided by other utilities which also wrap `write`
(`tf.train.CheckpointManager` for example).
Args:
file_prefix: A prefix to use for the checkpoint filenames
(/path/to/directory/and_a_prefix). Names are generated based on this
prefix and `Checkpoint.save_counter`.
session: The session to evaluate variables in. Ignored when executing
eagerly. If not provided when graph building, the default session is
used.
Returns:
The full path to the checkpoint."
11795,restore,tensorflow/tensorflow/python/training/tracking/util.py,1626,method,"Restore a training checkpoint.
Restores this `Checkpoint` and any objects it depends on.
When executing eagerly, either assigns values immediately if variables to
restore have been created already, or defers restoration until the variables
are created. Dependencies added after this call will be matched if they have
a corresponding object in the checkpoint (the restore request will queue in
any trackable object waiting for the expected dependency to be added).
When graph building, restoration ops are added to the graph but not run
immediately.
To ensure that loading is complete and no more assignments will take place,
use the `assert_consumed()` method of the status object returned by
`restore`:
```python
checkpoint = tf.train.Checkpoint( ... )
checkpoint.restore(path).assert_consumed()
```
An exception will be raised if any Python objects in the dependency graph
were not found in the checkpoint, or if any checkpointed values do not have
a matching Python object.
When graph building, `assert_consumed()` indicates that all of the restore
ops that will be created for this checkpoint have been created. They can be
run via the `run_restore_ops()` method of the status object:
```python
checkpoint.restore(path).assert_consumed().run_restore_ops()
```
If the checkpoint has not been consumed completely, then the list of restore
ops will grow as more objects are added to the dependency graph.
Name-based `tf.compat.v1.train.Saver` checkpoints can be loaded using this
method. Names are used to match variables. No restore ops are created/run
until `run_restore_ops()` or `initialize_or_restore()` are called on the
returned status object when graph building, but there is restore-on-creation
when executing eagerly. Re-encode name-based checkpoints using
`tf.train.Checkpoint.save` as soon as possible.
Args:
save_path: The path to the checkpoint, as returned by `save` or
`tf.train.latest_checkpoint`. If None (as when there is no latest
checkpoint for `tf.train.latest_checkpoint` to return), returns an
object which may run initializers for objects in the dependency graph.
If the checkpoint was written by the name-based
`tf.compat.v1.train.Saver`, names are used to match variables.
Returns:
A load status object, which can be used to make assertions about the
status of a checkpoint restoration and run initialization/restore ops.
The returned status object has the following methods:
* `assert_consumed()`:
Raises an exception if any variables are unmatched: either
checkpointed values which don't have a matching Python object or
Python objects in the dependency graph with no values in the
checkpoint. This method returns the status object, and so may be
chained with `initialize_or_restore` or `run_restore_ops`.
* `assert_existing_objects_matched()`:
Raises an exception if any existing Python objects in the dependency
graph are unmatched. Unlike `assert_consumed`, this assertion will
pass if values in the checkpoint have no corresponding Python
objects. For example a `tf.keras.Layer` object which has not yet been
built, and so has not created any variables, will pass this assertion
but fail `assert_consumed`. Useful when loading part of a larger
checkpoint into a new Python program, e.g. a training checkpoint with
a `tf.compat.v1.train.Optimizer` was saved but only the state required
for
inference is being loaded. This method returns the status object, and
so may be chained with `initialize_or_restore` or `run_restore_ops`.
* `assert_nontrivial_match()`: Asserts that something aside from the root
object was matched. This is a very weak assertion, but is useful for
sanity checking in library code where objects may exist in the
checkpoint which haven't been created in Python and some Python
objects may not have a checkpointed value.
* `expect_partial()`: Silence warnings about incomplete checkpoint
restores. Warnings are otherwise printed for unused parts of the
checkpoint file or object when the `Checkpoint` object is deleted
(often at program shutdown).
* `initialize_or_restore(session=None)`:
When graph building, runs variable initializers if `save_path` is
`None`, but otherwise runs restore operations. If no `session` is
explicitly specified, the default session is used. No effect when
executing eagerly (variables are initialized or restored eagerly).
* `run_restore_ops(session=None)`:
When graph building, runs restore operations. If no `session` is
explicitly specified, the default session is used. No effect when
executing eagerly (restore operations are run eagerly). May only be
called when `save_path` is not `None`."
11796,Checkpoint,tensorflow/tensorflow/python/training/tracking/util.py,1739,class,"Groups trackable objects, saving and restoring them.
`Checkpoint`'s constructor accepts keyword arguments whose values are types
that contain trackable state, such as `tf.keras.optimizers.Optimizer`
implementations, `tf.Variable`s, `tf.data.Dataset` iterators, `tf.keras.Layer`
implementations, or `tf.keras.Model` implementations. It saves these values
with a checkpoint, and maintains a `save_counter` for numbering checkpoints.
Example usage:
```python
import tensorflow as tf
import os
checkpoint_directory = ""/tmp/training_checkpoints""
checkpoint_prefix = os.path.join(checkpoint_directory, ""ckpt"")
# Create a Checkpoint that will manage two objects with trackable state,
# one we name ""optimizer"" and the other we name ""model"".
checkpoint = tf.train.Checkpoint(optimizer=optimizer, model=model)
status = checkpoint.restore(tf.train.latest_checkpoint(checkpoint_directory))
for _ in range(num_training_steps):
optimizer.minimize( ... ) # Variables will be restored on creation.
status.assert_consumed() # Optional sanity checks.
checkpoint.save(file_prefix=checkpoint_prefix)
```
`Checkpoint.save()` and `Checkpoint.restore()` write and read object-based
checkpoints, in contrast to TensorFlow 1.x's `tf.compat.v1.train.Saver` which
writes and
reads `variable.name` based checkpoints. Object-based checkpointing saves a
graph of dependencies between Python objects (`Layer`s, `Optimizer`s,
`Variable`s, etc.) with named edges, and this graph is used to match variables
when restoring a checkpoint. It can be more robust to changes in the Python
program, and helps to support restore-on-create for variables.
`Checkpoint` objects have dependencies on the objects passed as keyword
arguments to their constructors, and each dependency is given a name that is
identical to the name of the keyword argument for which it was created.
TensorFlow classes like `Layer`s and `Optimizer`s will automatically add
dependencies on their own variables (e.g. ""kernel"" and ""bias"" for
`tf.keras.layers.Dense`). Inheriting from `tf.keras.Model` makes managing
dependencies easy in user-defined classes, since `Model` hooks into attribute
assignment. For example:
```python
class Regress(tf.keras.Model):
def __init__(self):
super(Regress, self).__init__()
self.input_transform = tf.keras.layers.Dense(10)
# ...
def call(self, inputs):
x = self.input_transform(inputs)
# ...
```
This `Model` has a dependency named ""input_transform"" on its `Dense` layer,
which in turn depends on its variables. As a result, saving an instance of
`Regress` using `tf.train.Checkpoint` will also save all the variables created
by the `Dense` layer.
When variables are assigned to multiple workers, each worker writes its own
section of the checkpoint. These sections are then merged/re-indexed to behave
as a single checkpoint. This avoids copying all variables to one worker, but
does require that all workers see a common filesystem.
While `tf.keras.Model.save_weights` and `tf.train.Checkpoint.save` save in the
same format, note that the root of the resulting checkpoint is the object the
save method is attached to. This means saving a `tf.keras.Model` using
`save_weights` and loading into a `tf.train.Checkpoint` with a `Model`
attached (or vice versa) will not match the `Model`'s variables. See the
[guide to training
checkpoints](https://www.tensorflow.org/guide/checkpoint) for
details. Prefer `tf.train.Checkpoint` over `tf.keras.Model.save_weights` for
training checkpoints.
Attributes:
save_counter: Incremented when `save()` is called. Used to number
checkpoints."
11797,write,tensorflow/tensorflow/python/training/tracking/util.py,1863,method,"Writes a training checkpoint.
The checkpoint includes variables created by this object and any
trackable objects it depends on at the time `Checkpoint.write()` is
called.
`write` does not number checkpoints, increment `save_counter`, or update the
metadata used by `tf.train.latest_checkpoint`. It is primarily intended for
use by higher level checkpoint management utilities. `save` provides a very
basic implementation of these features.
Checkpoints written with `write` must be read with `read`.
Example usage:
```
step = tf.Variable(0, name=""step"")
checkpoint = tf.Checkpoint(step=step)
checkpoint.write(""/tmp/ckpt"")
# Later, read the checkpoint with read()
checkpoint.read(""/tmp/ckpt"").assert_consumed()
# You can also pass options to write() and read(). For example this
# runs the IO ops on the localhost:
options = tf.CheckpointOptions(experimental_io_device=""/job:localhost"")
checkpoint.write(""/tmp/ckpt"", options=options)
# Later, read the checkpoint with read()
checkpoint.read(""/tmp/ckpt"", options=options).assert_consumed()
```
Args:
file_prefix: A prefix to use for the checkpoint filenames
(/path/to/directory/and_a_prefix).
options: Optional `tf.train.CheckpointOptions` object.
Returns:
The full path to the checkpoint (i.e. `file_prefix`)."
11798,save_counter,tensorflow/tensorflow/python/training/tracking/util.py,1917,method,"An integer variable which starts at zero and is incremented on save.
Used to number checkpoints.
Returns:
The save counter variable."
11799,save,tensorflow/tensorflow/python/training/tracking/util.py,1928,method,"Saves a training checkpoint and provides basic checkpoint management.
The saved checkpoint includes variables created by this object and any
trackable objects it depends on at the time `Checkpoint.save()` is
called.
`save` is a basic convenience wrapper around the `write` method,
sequentially numbering checkpoints using `save_counter` and updating the
metadata used by `tf.train.latest_checkpoint`. More advanced checkpoint
management, for example garbage collection and custom numbering, may be
provided by other utilities which also wrap `write` and `read`.
(`tf.train.CheckpointManager` for example).
```
step = tf.Variable(0, name=""step"")
checkpoint = tf.Checkpoint(step=step)
checkpoint.save(""/tmp/ckpt"")
# Later, read the checkpoint with restore()
checkpoint.restore(""/tmp/ckpt"").assert_consumed()
# You can also pass options to save() and restore(). For example this
# runs the IO ops on the localhost:
options = tf.CheckpointOptions(experimental_io_device=""/job:localhost"")
checkpoint.save(""/tmp/ckpt"", options=options)
# Later, read the checkpoint with restore()
checkpoint.restore(""/tmp/ckpt"", options=options).assert_consumed()
```
Args:
file_prefix: A prefix to use for the checkpoint filenames
(/path/to/directory/and_a_prefix). Names are generated based on this
prefix and `Checkpoint.save_counter`.
options: Optional `tf.train.CheckpointOptions` object.
Returns:
The full path to the checkpoint."
11800,read,tensorflow/tensorflow/python/training/tracking/util.py,2003,method,"Reads a training checkpoint written with `write`.
Reads this `Checkpoint` and any objects it depends on.
This method is just like `restore()` but does not expect the `save_counter`
variable in the checkpoint. It only restores the objects that the checkpoint
already depends on.
The method is primarily intended for use by higher level checkpoint
management utilities that use `write()` instead of `save()` and have their
own mechanisms to number and track checkpoints.
Example usage:
```python
# Create a checkpoint with write()
ckpt = tf.train.Checkpoint(v=tf.Variable(1.))
path = ckpt.write('/tmp/my_checkpoint')
# Later, load the checkpoint with read()
# With restore() assert_consumed() would have failed.
checkpoint.read(path).assert_consumed()
# You can also pass options to read(). For example this
# runs the IO ops on the localhost:
options = tf.CheckpointOptions(experimental_io_device=""/job:localhost"")
checkpoint.read(path, options=options)
```
Args:
save_path: The path to the checkpoint as returned by `write`.
options: Optional `tf.train.CheckpointOptions` object.
Returns:
A load status object, which can be used to make assertions about the
status of a checkpoint restoration. See `restore` for details."
11801,restore,tensorflow/tensorflow/python/training/tracking/util.py,2044,method,"Restores a training checkpoint.
Restores this `Checkpoint` and any objects it depends on.
This method is intended to be used to load checkpoints created by `save()`.
For checkpoints created by `write()` use the `read()` method which does not
expect the `save_counter` variable added by `save()`.
`restore()` either assigns values immediately if variables to restore have
been created already, or defers restoration until the variables are
created. Dependencies added after this call will be matched if they have a
corresponding object in the checkpoint (the restore request will queue in
any trackable object waiting for the expected dependency to be added).
To ensure that loading is complete and no more assignments will take place,
use the `assert_consumed()` method of the status object returned by
`restore()`:
```python
checkpoint = tf.train.Checkpoint( ... )
checkpoint.restore(path).assert_consumed()
# You can additionally pass options to restore():
options = tf.CheckpointOptions(experimental_io_device=""/job:localhost"")
checkpoint.restore(path, options=options).assert_consumed()
```
An exception will be raised if any Python objects in the dependency graph
were not found in the checkpoint, or if any checkpointed values do not have
a matching Python object.
Name-based `tf.compat.v1.train.Saver` checkpoints from TensorFlow 1.x can be
loaded
using this method. Names are used to match variables. Re-encode name-based
checkpoints using `tf.train.Checkpoint.save` as soon as possible.
Args:
save_path: The path to the checkpoint, as returned by `save` or
`tf.train.latest_checkpoint`. If the checkpoint was written by the
name-based `tf.compat.v1.train.Saver`, names are used to match
variables.
options: Optional `tf.train.CheckpointOptions` object.
Returns:
A load status object, which can be used to make assertions about the
status of a checkpoint restoration.
The returned status object has the following methods:
* `assert_consumed()`:
Raises an exception if any variables are unmatched: either
checkpointed values which don't have a matching Python object or
Python objects in the dependency graph with no values in the
checkpoint. This method returns the status object, and so may be
chained with other assertions.
* `assert_existing_objects_matched()`:
Raises an exception if any existing Python objects in the dependency
graph are unmatched. Unlike `assert_consumed`, this assertion will
pass if values in the checkpoint have no corresponding Python
objects. For example a `tf.keras.Layer` object which has not yet been
built, and so has not created any variables, will pass this assertion
but fail `assert_consumed`. Useful when loading part of a larger
checkpoint into a new Python program, e.g. a training checkpoint with
a `tf.compat.v1.train.Optimizer` was saved but only the state required
for
inference is being loaded. This method returns the status object, and
so may be chained with other assertions.
* `assert_nontrivial_match()`: Asserts that something aside from the root
object was matched. This is a very weak assertion, but is useful for
sanity checking in library code where objects may exist in the
checkpoint which haven't been created in Python and some Python
objects may not have a checkpointed value.
* `expect_partial()`: Silence warnings about incomplete checkpoint
restores. Warnings are otherwise printed for unused parts of the
checkpoint file or object when the `Checkpoint` object is deleted
(often at program shutdown)."
11802,NonLayerTrackable,tensorflow/tensorflow/python/training/tracking/util_test.py,49,class,
11803,Tensor,tensorflow/tensorflow/python/types/core.py,33,class,"The base class of all dense Tensor objects.
A dense tensor has a static data type (dtype), and may have a static rank and
shape. Tensor objects are immutable. Mutable objects may be backed by a Tensor
which holds the unique handle that identifies the mutable object."
11804,dtype,tensorflow/tensorflow/python/types/core.py,42,method,
11805,shape,tensorflow/tensorflow/python/types/core.py,46,method,
11806,Symbol,tensorflow/tensorflow/python/types/core.py,50,class,"Symbolic ""graph"" Tensor.
These objects represent the output of an op definition and do not carry a
value."
11807,Value,tensorflow/tensorflow/python/types/core.py,59,class,"Tensor that can be associated with a value (aka ""eager tensor"").
These objects represent the (usually future) output of executing an op
immediately."
11808,numpy,tensorflow/tensorflow/python/types/core.py,66,method,
11809,Iterable,tensorflow/tensorflow/python/types/distribute.py,25,class,Interface for distributed objects that admit iteration/reduction.
11810,reduce,tensorflow/tensorflow/python/types/distribute.py,32,method,"Reduces this iterable object to a single element.
The transformation calls `reduce_func` successively on each element.
The `initial_state` argument is used for the initial state and the final
state is returned as the result.
Args:
initial_state: An element representing the initial state of the
reduction.
reduce_func: A function that maps `(old_state, input_element)` to
`new_state`. The structure of `new_state` must match the structure of
`old_state`. For the first element, `old_state` is `initial_state`.
Returns:
The final state of the transformation."
11811,Iterator,tensorflow/tensorflow/python/types/distribute.py,51,class,Interface for distributed iterators.
11812,get_next,tensorflow/tensorflow/python/types/distribute.py,54,method,"Unlike __next__, this may use a non-raising mechanism."
11813,document,tensorflow/tensorflow/python/types/doc_typealias.py,24,function,"Adds a docstring to typealias by overriding the `__doc__` attribute.
Note: Overriding `__doc__` is only possible after python 3.7.
Args:
obj: Typealias object that needs to be documented.
doc: Docstring of the typealias. It should follow the standard pystyle
docstring rules."
11814,NativeObject,tensorflow/tensorflow/python/types/internal.py,26,class,"Types natively supported by various TF operations.
The most notable example of NativeObject is Tensor."
11815,my_fact,tensorflow/tensorflow/python/user_ops/user_ops.py,30,function,Example of overriding the generated code for an Op.
11816,make_all,tensorflow/tensorflow/python/util/all_util.py,30,function,"Generates `__all__` from the docstring of one or more modules.
Usage: `make_all(__name__)` or
`make_all(__name__, [sys.modules(__name__), other_module])`. The doc string
modules must each a docstring, and `__all__` will contain all symbols with
`@@` references, where that symbol currently exists in the module named
`module_name`.
Args:
module_name: The name of the module (usually `__name__`).
doc_string_modules: a list of modules from which to take docstring.
If None, then a list containing only the module named `module_name` is used.
Returns:
A list suitable for use as `__all__`."
11817,reveal_undocumented,tensorflow/tensorflow/python/util/all_util.py,66,function,"Reveals a symbol that was previously removed by `remove_undocumented`.
This should be used by tensorflow internal tests only. It explicitly
defeats the encapsulation afforded by `remove_undocumented`.
It throws an exception when the symbol was not hidden in the first place.
Args:
symbol_name: a string representing the full absolute path of the symbol.
target_module: if specified, the module in which to restore the symbol."
11818,remove_undocumented,tensorflow/tensorflow/python/util/all_util.py,86,function,"Removes symbols in a module that are not referenced by a docstring.
Args:
module_name: the name of the module (usually `__name__`).
allowed_exception_list: a list of names that should not be removed.
doc_string_modules: a list of modules from which to take the docstrings.
If None, then a list containing only the module named `module_name` is used.
Furthermore, if a symbol previously added with `add_to_global_allowlist`,
then it will always be allowed. This is useful for internal tests.
Returns:
None"
11819,as_bytes,tensorflow/tensorflow/python/util/compat.py,64,function,"Converts `bytearray`, `bytes`, or unicode python input types to `bytes`.
Uses utf-8 encoding for text by default.
Args:
bytes_or_text: A `bytearray`, `bytes`, `str`, or `unicode` object.
encoding: A string indicating the charset for encoding unicode.
Returns:
A `bytes` object.
Raises:
TypeError: If `bytes_or_text` is not a binary or unicode string."
11820,as_text,tensorflow/tensorflow/python/util/compat.py,90,function,"Converts any string-like python input types to unicode.
Returns the input as a unicode string. Uses utf-8 encoding for text
by default.
Args:
bytes_or_text: A `bytes`, `str`, or `unicode` object.
encoding: A string indicating the charset for decoding unicode.
Returns:
A `unicode` (Python 2) or `str` (Python 3) object.
Raises:
TypeError: If `bytes_or_text` is not a binary or unicode string."
11821,as_str,tensorflow/tensorflow/python/util/compat.py,114,function,
11822,as_str_any,tensorflow/tensorflow/python/util/compat.py,126,function,"Converts input to `str` type.
Uses `str(value)`, except for `bytes` typed inputs, which are converted
using `as_str`.
Args:
value: A object that can be converted to `str`.
Returns:
A `str` object."
11823,path_to_str,tensorflow/tensorflow/python/util/compat.py,145,function,"Converts input which is a `PathLike` object to `str` type.
Converts from any python constant representation of a `PathLike` object to
a string. If the input is not a `PathLike` object, simply returns the input.
Args:
path: An object that can be converted to path representation.
Returns:
A `str` object.
Usage:
In case a simplified `str` version of the path is needed from an
`os.PathLike` object
Examples:
```python
$ tf.compat.path_to_str('C:\XYZ\tensorflow\./.././tensorflow')
'C:\XYZ\tensorflow\./.././tensorflow' # Windows OS
$ tf.compat.path_to_str(Path('C:\XYZ\tensorflow\./.././tensorflow'))
'C:\XYZ\tensorflow\..\tensorflow' # Windows OS
$ tf.compat.path_to_str(Path('./corpus'))
'corpus' # Linux OS
$ tf.compat.path_to_str('./.././Corpus')
'./.././Corpus' # Linux OS
$ tf.compat.path_to_str(Path('./.././Corpus'))
'../Corpus' # Linux OS
$ tf.compat.path_to_str(Path('./..////../'))
'../..' # Linux OS
```"
11824,path_to_bytes,tensorflow/tensorflow/python/util/compat.py,183,function,"Converts input which is a `PathLike` object to `bytes`.
Converts from any python constant representation of a `PathLike` object
or `str` to bytes.
Args:
path: An object that can be converted to path representation.
Returns:
A `bytes` object.
Usage:
In case a simplified `bytes` version of the path is needed from an
`os.PathLike` object"
11825,path_to_str,tensorflow/tensorflow/python/util/compat_internal.py,24,function,"Returns the file system path representation of a `PathLike` object,
else as it is.
Args:
path: An object that can be converted to path representation.
Returns:
A `str` object."
11826,get_qualified_name,tensorflow/tensorflow/python/util/decorator_utils.py,24,function,
11827,add_notice_to_docstring,tensorflow/tensorflow/python/util/decorator_utils.py,76,function,"Adds a deprecation notice to a docstring.
Args:
doc: The original docstring.
instructions: A string, describing how to fix the problem.
no_doc_str: The default value to use for `doc` if `doc` is empty.
suffix_str: Is added to the end of the first line.
notice: A list of strings. The main notice warning body.
Returns:
A new docstring, with the notice attached.
Raises:
ValueError: If `notice` is empty."
11828,validate_callable,tensorflow/tensorflow/python/util/decorator_utils.py,117,function,
11829,classproperty,tensorflow/tensorflow/python/util/decorator_utils.py,126,class,"Class property decorator.
Example usage:
class MyClass(object):
@classproperty
def value(cls):
return '123'
> print MyClass.value
123"
11830,DeprecatedNamesAlreadySet,tensorflow/tensorflow/python/util/deprecation.py,41,class,Raised when setting deprecated names multiple times for the same symbol.
11831,deprecated_alias,tensorflow/tensorflow/python/util/deprecation.py,134,function,"Deprecate a symbol in favor of a new name with identical semantics.
This function is meant to be used when defining a backwards-compatibility
alias for a symbol which has been moved. For example:
module1.py:
```python
class NewNameForClass: pass
```
module2.py:
```python
import module1
DeprecatedNameForClass = deprecated_alias(
deprecated_name='module2.DeprecatedNameForClass',
name='module1.NewNameForClass',
func_or_class=module1.NewNameForClass)
```
This function works for classes and functions.
For classes, it creates a new class which is functionally identical (it
inherits from the original, and overrides its constructor), but which prints
a deprecation warning when an instance is created. It also adds a deprecation
notice to the class' docstring.
For functions, it returns a function wrapped by `tf_decorator.make_decorator`.
That function prints a warning when used, and has a deprecation notice in its
docstring. This is more or less equivalent (the deprecation warning has
slightly different text) to writing:
```python
@deprecated
def deprecated_alias(original_args):
real_function(original_args)
```
Args:
deprecated_name: The name of the symbol that is being deprecated, to be used
in the warning message. This should be its fully qualified name to avoid
confusion.
name: The name of the symbol that is to be used instead of the deprecated
name. This should be a fully qualified name to avoid confusion.
func_or_class: The (non-deprecated) class or function for which a deprecated
alias should be created.
warn_once: If True (the default), only print a deprecation warning the first
time this function is used, or the class is instantiated.
Returns:
A wrapped version of `func_or_class` which prints a deprecation warning on
use and has a modified docstring."
11832,deprecated_endpoints,tensorflow/tensorflow/python/util/deprecation.py,245,function,"Decorator for marking endpoints deprecated.
This decorator does not print deprecation messages.
TODO(annarev): eventually start printing deprecation warnings when
@deprecation_endpoints decorator is added.
Args:
*args: Deprecated endpoint names.
Returns:
A function that takes symbol as an argument and adds
_tf_deprecated_api_names to that symbol.
_tf_deprecated_api_names would be set to a list of deprecated
endpoint names for the symbol."
11833,deprecated,tensorflow/tensorflow/python/util/deprecation.py,274,function,"Decorator for marking functions or methods deprecated.
This decorator logs a deprecation warning whenever the decorated function is
called. It has the following format:
<function> (from <module>) is deprecated and will be removed after <date>.
Instructions for updating:
<instructions>
If `date` is None, 'after <date>' is replaced with 'in a future version'.
<function> will include the class name if it is a method.
It also edits the docstring of the function: ' (deprecated)' is appended
to the first line of the docstring and a deprecation notice is prepended
to the rest of the docstring.
Args:
date: String or None. The date the function is scheduled to be removed.
Must be ISO 8601 (YYYY-MM-DD), or None.
instructions: String. Instructions on how to update code using the
deprecated function.
warn_once: Boolean. Set to `True` to warn only the first time the decorated
function is called. Otherwise, every call will log a warning.
Returns:
Decorated function or method.
Raises:
ValueError: If date is not None or in ISO 8601 format, or instructions are
empty."
11834,deprecated_args,tensorflow/tensorflow/python/util/deprecation.py,336,function,"Decorator for marking specific function arguments as deprecated.
This decorator logs a deprecation warning whenever the decorated function is
called with the deprecated argument. It has the following format:
Calling <function> (from <module>) with <arg> is deprecated and will be
removed after <date>. Instructions for updating:
<instructions>
If `date` is None, 'after <date>' is replaced with 'in a future version'.
<function> includes the class name if it is a method.
It also edits the docstring of the function: ' (deprecated arguments)' is
appended to the first line of the docstring and a deprecation notice is
prepended to the rest of the docstring.
Args:
date: String or None. The date the function is scheduled to be removed.
Must be ISO 8601 (YYYY-MM-DD), or None.
instructions: String. Instructions on how to update code using the
deprecated function.
*deprecated_arg_names_or_tuples: String or 2-Tuple(String,
[ok_vals]). The string is the deprecated argument name.
Optionally, an ok-value may be provided. If the user provided
argument equals this value, the warning is suppressed.
**kwargs: If `warn_once=False` is passed, every call with a deprecated
argument will log a warning. The default behavior is to only warn the
first time the function is called with any given deprecated argument.
All other kwargs raise `ValueError`.
Returns:
Decorated function or method.
Raises:
ValueError: If date is not None or in ISO 8601 format, instructions are
empty, the deprecated arguments are not present in the function
signature, the second element of a deprecated_tuple is not a
list, or if a kwarg other than `warn_once` is passed."
11835,deprecated_arg_values,tensorflow/tensorflow/python/util/deprecation.py,516,function,"Decorator for marking specific function argument values as deprecated.
This decorator logs a deprecation warning whenever the decorated function is
called with the deprecated argument values. It has the following format:
Calling <function> (from <module>) with <arg>=<value> is deprecated and
will be removed after <date>. Instructions for updating:
<instructions>
If `date` is None, 'after <date>' is replaced with 'in a future version'.
<function> will include the class name if it is a method.
It also edits the docstring of the function: ' (deprecated arguments)' is
appended to the first line of the docstring and a deprecation notice is
prepended to the rest of the docstring.
Args:
date: String or None. The date the function is scheduled to be removed.
Must be ISO 8601 (YYYY-MM-DD), or None
instructions: String. Instructions on how to update code using the
deprecated function.
warn_once: If `True`, warn only the first time this function is called with
deprecated argument values. Otherwise, every call (with a deprecated
argument value) will log a warning.
**deprecated_kwargs: The deprecated argument values.
Returns:
Decorated function or method.
Raises:
ValueError: If date is not None or in ISO 8601 format, or instructions are
empty."
11836,deprecated_argument_lookup,tensorflow/tensorflow/python/util/deprecation.py,583,function,"Looks up deprecated argument name and ensures both are not used.
Args:
new_name: new name of argument
new_value: value of new argument (or None if not used)
old_name: old name of argument
old_value: value of old argument (or None if not used)
Returns:
The effective argument that should be used.
Raises:
ValueError: if new_value and old_value are both non-null"
11837,rewrite_argument_docstring,tensorflow/tensorflow/python/util/deprecation.py,604,function,
11838,silence,tensorflow/tensorflow/python/util/deprecation.py,610,function,Temporarily silence deprecation warnings.
11839,HiddenTfApiAttribute,tensorflow/tensorflow/python/util/deprecation.py,619,class,"Hides a class attribute from the public API.
Attributes in public classes can be hidden from the API by having an '_' in
front of the name (e.g. ClassName._variables). This doesn't work when
attributes or methods are inherited from a parent class. To hide inherited
attributes, set their values to be `deprecation.hide_attribute_from_api`.
For example, this is used in V2 Estimator to hide the deprecated
export_savedmodel method:
class EstimatorV2(Estimator):
export_savedmodel = deprecation.hide_attribute_from_api('...')"
11840,raise_error,tensorflow/tensorflow/python/util/deprecation.py,634,method,
11841,OpDispatcher,tensorflow/tensorflow/python/util/dispatch.py,46,class,"Abstract base class for TensorFlow operator dispatchers.
Each operation dispatcher acts as an override handler for a single
TensorFlow operation, and its results are used when the handler indicates
that it can handle the operation's arguments (by returning any value other
than `OpDispatcher.NOT_SUPPORTED`)."
11842,handle,tensorflow/tensorflow/python/util/dispatch.py,59,method,"Handle this dispatcher's operation with the specified arguments.
If this operation dispatcher can handle the given arguments, then
return an appropriate value (or raise an appropriate exception).
Args:
args: The arguments to the operation.
kwargs: They keyword arguments to the operation.
Returns:
The result of the operation, or `OpDispatcher.NOT_SUPPORTED` if this
dispatcher can not handle the given arguments."
11843,register,tensorflow/tensorflow/python/util/dispatch.py,75,method,"Register this dispatcher as a handler for `op`.
Args:
op: Python function: the TensorFlow operation that should be handled. Must
have a dispatch list (which is added automatically for generated ops,
and can be added to Python ops using the `add_dispatch_support`
decorator)."
11844,GlobalOpDispatcher,tensorflow/tensorflow/python/util/dispatch.py,89,class,Abstract base class for TensorFlow global operator dispatchers.
11845,handle,tensorflow/tensorflow/python/util/dispatch.py,94,method,Handle the specified operation with the specified arguments.
11846,register,tensorflow/tensorflow/python/util/dispatch.py,97,method,Register this dispatcher as a handler for all ops.
11847,dispatch,tensorflow/tensorflow/python/util/dispatch.py,102,function,"Returns the result from the first successful dispatcher for a given op.
Calls the `handle` method of each `OpDispatcher` that has been registered
to handle `op`, and returns the value from the first successful handler.
Args:
op: Python function: the operation to dispatch for.
args: The arguments to the operation.
kwargs: They keyword arguments to the operation.
Returns:
The result of the operation, or `NOT_SUPPORTED` if no registered
dispatcher can handle the given arguments."
11848,dispatch_for_types,tensorflow/tensorflow/python/util/dispatch.py,156,function,"Decorator to declare that a Python function overrides an op for a type.
The decorated function is used to override `op` if any of the arguments or
keyword arguments (including elements of lists or tuples) have one of the
specified types.
Example:
```python
@dispatch_for_types(math_ops.add, RaggedTensor, RaggedTensorValue)
def ragged_add(x, y, name=None): ...
```
Args:
op: Python function: the operation that should be overridden.
*types: The argument types for which this function should be used."
11849,add_dispatch_list,tensorflow/tensorflow/python/util/dispatch.py,188,function,Decorator that adds a dispatch_list attribute to an op.
11850,add_dispatch_support,tensorflow/tensorflow/python/util/dispatch.py,196,function,Decorator that adds a dispatch handling wrapper to an op.
11851,CustomTensor,tensorflow/tensorflow/python/util/dispatch_test.py,37,class,"A fake composite tensor class, for testing type-based dispatching."
11852,TensorTracer,tensorflow/tensorflow/python/util/dispatch_test.py,52,class,"An object used to trace TensorFlow graphs.
This is an example class that is used to test global op dispatchers. The
global op dispatcher for TensorTracers is defined below."
11853,TensorTracerOpDispatcher,tensorflow/tensorflow/python/util/dispatch_test.py,93,class,Global op dispatcher for TensorTracer.
11854,handle,tensorflow/tensorflow/python/util/dispatch_test.py,105,method,
11855,is_tensor_tracer_arg,tensorflow/tensorflow/python/util/dispatch_test.py,114,method,
11856,extract_example_parser_configuration,tensorflow/tensorflow/python/util/example_parser_configuration.py,26,function,"Returns an ExampleParserConfig proto.
Args:
parse_example_op: A ParseExample or ParseExampleV2 `Operation`
sess: A tf.compat.v1.Session needed to obtain some configuration values.
Returns:
A ExampleParserConfig proto.
Raises:
ValueError: If attributes are inconsistent."
11857,fn_args,tensorflow/tensorflow/python/util/function_utils.py,39,function,"Get argument names for function-like object.
Args:
fn: Function, or function-like object (e.g., result of `functools.partial`).
Returns:
`tuple` of string argument names.
Raises:
ValueError: if partial function has positionally bound arguments"
11858,has_kwargs,tensorflow/tensorflow/python/util/function_utils.py,66,function,"Returns whether the passed callable has **kwargs in its signature.
Args:
fn: Function, or function-like object (e.g., result of `functools.partial`).
Returns:
`bool`: if `fn` has **kwargs in its signature.
Raises:
`TypeError`: If fn is not a Function, or function-like object."
11859,get_func_name,tensorflow/tensorflow/python/util/function_utils.py,89,function,Returns name of passed callable.
11860,get_func_code,tensorflow/tensorflow/python/util/function_utils.py,104,function,"Returns func_code of passed callable, or None if not available."
11861,get_disabled_rewriter_config,tensorflow/tensorflow/python/util/function_utils.py,125,function,
11862,silly_example_function,tensorflow/tensorflow/python/util/function_utils_test.py,27,function,
11863,SillyCallableClass,tensorflow/tensorflow/python/util/function_utils_test.py,31,class,
11864,keyword_args_only,tensorflow/tensorflow/python/util/keyword_args.py,27,function,"Decorator for marking specific function accepting keyword args only.
This decorator raises a `ValueError` if the input `func` is called with any
non-keyword args. This prevents the caller from providing the arguments in
wrong order.
Args:
func: The function or method needed to be decorated.
Returns:
Decorated function or method.
Raises:
ValueError: If `func` is not callable."
11865,LazyLoader,tensorflow/tensorflow/python/util/lazy_loader.py,27,class,"Lazily import a module, mainly to avoid pulling in large dependencies.
`contrib`, and `ffmpeg` are examples of modules that are large and not always
needed, and this allows them to only be loaded when they are used."
11866,GroupLock,tensorflow/tensorflow/python/util/lock_util.py,24,class,"A lock to allow many members of a group to access a resource exclusively.
This lock provides a way to allow access to a resource by multiple threads
belonging to a logical group at the same time, while restricting access to
threads from all other groups. You can think of this as an extension of a
reader-writer lock, where you allow multiple writers at the same time. We
made it generic to support multiple groups instead of just two - readers and
writers.
Simple usage example with two groups accessing the same resource:
```python
lock = GroupLock(num_groups=2)
# In a member of group 0:
with lock.group(0):
# do stuff, access the resource
# ...
# In a member of group 1:
with lock.group(1):
# do stuff, access the resource
# ...
```
Using as a context manager with `.group(group_id)` is the easiest way. You
can also use the `acquire` and `release` method directly."
11867,group,tensorflow/tensorflow/python/util/lock_util.py,76,method,"Enter a context where the lock is with group `group_id`.
Args:
group_id: The group for which to acquire and release the lock.
Returns:
A context manager which will acquire the lock for `group_id`."
11868,acquire,tensorflow/tensorflow/python/util/lock_util.py,88,method,Acquire the group lock for a specific group `group_id`.
11869,release,tensorflow/tensorflow/python/util/lock_util.py,98,method,Release the group lock for a specific group `group_id`.
11870,dismantle_ordered_dict,tensorflow/tensorflow/python/util/memory.py,24,function,"Remove reference cycle in OrderedDict `ordered_dict`.
Helpful for making sure the garbage collector doesn't need to run after
using an OrderedDict.
Args:
ordered_dict: A `OrderedDict` object to destroy. This object is unusable
after this function runs."
11871,get_rename_v2,tensorflow/tensorflow/python/util/module_wrapper.py,34,function,
11872,contains_deprecation_decorator,tensorflow/tensorflow/python/util/module_wrapper.py,51,function,
11873,has_deprecation_decorator,tensorflow/tensorflow/python/util/module_wrapper.py,56,function,"Checks if given object has a deprecation decorator.
We check if deprecation decorator is in decorators as well as
whether symbol is a class whose __init__ method has a deprecation
decorator.
Args:
symbol: Python object.
Returns:
True if symbol has deprecation decorator."
11874,TFModuleWrapper,tensorflow/tensorflow/python/util/module_wrapper.py,81,class,Wrapper for TF modules to support deprecation messages and lazyloading.
11875,MockModule,tensorflow/tensorflow/python/util/module_wrapper_test.py,34,class,
11876,is_nested,tensorflow/tensorflow/python/util/nest.py,261,function,"Returns true if its input is a collections.abc.Sequence (except strings).
Args:
seq: an input sequence.
Returns:
True if the sequence is a not a string and is a collections.abc.Sequence
or a dict."
11877,flatten,tensorflow/tensorflow/python/util/nest.py,275,function,"Returns a flat list from a given nested structure.
If nest is not a structure , tuple (or a namedtuple), dict, or an attrs class,
then returns a single-element list:
[nest].
In the case of dict instances, the sequence consists of the values, sorted by
key to ensure deterministic behavior. This is true also for OrderedDict
instances: their sequence order is ignored, the sorting order of keys is used
instead. The same convention is followed in pack_sequence_as. This correctly
repacks dicts and OrderedDicts after they have been flattened, and also allows
flattening an OrderedDict and then repacking it back using a corresponding
plain dict, or vice-versa. Dictionaries with non-sortable keys cannot be
flattened.
Users must not modify any collections used in nest while this function is
running.
Examples:
1. Python dict (ordered by key):
>>> dict = { ""key3"": ""value3"", ""key1"": ""value1"", ""key2"": ""value2"" }
>>> tf.nest.flatten(dict)
['value1', 'value2', 'value3']
2. For a nested python tuple:
>>> tuple = ((1.0, 2.0), (3.0, 4.0, 5.0), (6.0))
>>> tf.nest.flatten(tuple)
[1.0, 2.0, 3.0, 4.0, 5.0, 6.0]
3. Numpy array (will not flatten):
>>> array = np.array([[1, 2], [3, 4]])
>>> tf.nest.flatten(array)
[array([[1, 2],
[3, 4]])]
4. `tf.Tensor` (will not flatten):
>>> tensor = tf.constant([[1., 2., 3.], [4., 5., 6.], [7., 8., 9.]])
>>> tf.nest.flatten(tensor)
[<tf.Tensor: shape=(3, 3), dtype=float32, numpy=
array([[1., 2., 3.],
[4., 5., 6.],
[7., 8., 9.]], dtype=float32)>]
Args:
structure: an arbitrarily nested structure. Note, numpy arrays are
considered atoms and are not flattened.
expand_composites: If true, then composite tensors such as
`tf.sparse.SparseTensor` and `tf.RaggedTensor` are expanded into their
component tensors.
Returns:
A Python list, the flattened version of the input.
Raises:
TypeError: The nest is or contains a dict with non-sortable keys."
11878,assert_same_structure,tensorflow/tensorflow/python/util/nest.py,360,function,"Asserts that two structures are nested in the same way.
Note that namedtuples with identical name and fields are always considered
to have the same shallow structure (even with `check_types=True`).
For instance, this code will print `True`:
```python
def nt(a, b):
return collections.namedtuple('foo', 'a b')(a, b)
print(assert_same_structure(nt(0, 1), nt(2, 3)))
```
Args:
nest1: an arbitrarily nested structure.
nest2: an arbitrarily nested structure.
check_types: if `True` (default) types of sequences are checked as well,
including the keys of dictionaries. If set to `False`, for example a
list and a tuple of objects will look the same if they have the same
size. Note that namedtuples with identical name and fields are always
considered to have the same shallow structure. Two types will also be
considered the same if they are both list subtypes (which allows ""list""
and ""_ListWrapper"" from trackable dependency tracking to compare
equal).
expand_composites: If true, then composite tensors such as
`tf.sparse.SparseTensor` and `tf.RaggedTensor` are expanded into their
component tensors.
Raises:
ValueError: If the two structures do not have the same number of elements or
if the two structures are not nested in the same way.
TypeError: If the two structures differ in the type of sequence in any of
their substructures. Only possible if `check_types` is `True`."
11879,flatten_dict_items,tensorflow/tensorflow/python/util/nest.py,407,function,"Returns a dictionary with flattened keys and values.
This function flattens the keys and values of a dictionary, which can be
arbitrarily nested structures, and returns the flattened version of such
structures:
```python
example_dictionary = {(4, 5, (6, 8)): (""a"", ""b"", (""c"", ""d""))}
result = {4: ""a"", 5: ""b"", 6: ""c"", 8: ""d""}
flatten_dict_items(example_dictionary) == result
```
The input dictionary must satisfy two properties:
1. Its keys and values should have the same exact nested structure.
2. The set of all flattened keys of the dictionary must not contain repeated
keys.
Args:
dictionary: the dictionary to zip
Returns:
The zipped dictionary.
Raises:
TypeError: If the input is not a dictionary.
ValueError: If any key and value do not have the same structure layout, or
if keys are not unique."
11880,pack_sequence_as,tensorflow/tensorflow/python/util/nest.py,539,function,"Returns a given flattened sequence packed into a given structure.
If `structure` is a scalar, `flat_sequence` must be a single-element list;
in this case the return value is `flat_sequence[0]`.
If `structure` is or contains a dict instance, the keys will be sorted to
pack the flat sequence in deterministic order. This is true also for
`OrderedDict` instances: their sequence order is ignored, the sorting order of
keys is used instead. The same convention is followed in `flatten`.
This correctly repacks dicts and `OrderedDict`s after they have been
flattened, and also allows flattening an `OrderedDict` and then repacking it
back using a corresponding plain dict, or vice-versa.
Dictionaries with non-sortable keys cannot be flattened.
Args:
structure: Nested structure, whose structure is given by nested lists,
tuples, and dicts. Note: numpy arrays and strings are considered
scalars.
flat_sequence: flat sequence to pack.
expand_composites: If true, then composite tensors such as
`tf.sparse.SparseTensor` and `tf.RaggedTensor` are expanded into their
component tensors.
Returns:
packed: `flat_sequence` converted to have the same recursive structure as
`structure`.
Raises:
ValueError: If `flat_sequence` and `structure` have different
element counts.
TypeError: `structure` is or contains a dict with non-sortable keys."
11881,map_structure,tensorflow/tensorflow/python/util/nest.py,576,function,"Applies `func` to each entry in `structure` and returns a new structure.
Applies `func(x[0], x[1], ...)` where x[i] is an entry in
`structure[i]`. All structures in `structure` must have the same arity,
and the return value will contain results with the same structure layout.
Args:
func: A callable that accepts as many arguments as there are structures.
*structure: scalar, or tuple or dict or list of constructed scalars and/or
other tuples/lists, or scalars. Note: numpy arrays are considered as
scalars.
**kwargs: Valid keyword args are:
* `check_types`: If set to `True` (default) the types of
iterables within the structures have to be same (e.g.
`map_structure(func, [1], (1,))` raises a `TypeError`
exception). To allow this set this argument to `False`.
Note that namedtuples with identical name and fields are always
considered to have the same shallow structure.
* `expand_composites`: If set to `True`, then composite tensors such
as `tf.sparse.SparseTensor` and `tf.RaggedTensor` are expanded into
their component tensors. If `False` (the default), then composite
tensors are not expanded.
Returns:
A new structure with the same arity as `structure`, whose values correspond
to `func(x[0], x[1], ...)` where `x[i]` is a value in the corresponding
location in `structure[i]`. If there are different sequence types and
`check_types` is `False` the sequence types of the first structure will be
used.
Raises:
TypeError: If `func` is not callable or if the structures do not match
each other by depth tree.
ValueError: If no structure is provided or if the structures do not match
each other by type.
ValueError: If wrong keyword arguments are provided."
11882,map_structure_with_paths,tensorflow/tensorflow/python/util/nest.py,641,function,"Applies `func` to each entry in `structure` and returns a new structure.
Applies `func(path, x[0], x[1], ..., **kwargs)` where x[i] is an entry in
`structure[i]` and `path` is the common path to x[i] in the structures. All
structures in `structure` must have the same arity, and the return value will
contain the results with the same structure layout. Special kwarg
`check_types` determines whether the types of iterables within the structure
must be the same-- see **kwargs definition below.
Args:
func: A callable with the signature func(path, *values, **kwargs) that is
evaluated on the leaves of the structure.
*structure: A variable number of compatible structures to process.
**kwargs: Optional kwargs to be passed through to func. Special kwarg
`check_types` is not passed to func, but instead determines whether the
types of iterables within the structures have to be same (e.g.,
`map_structure(func, [1], (1,))` raises a `TypeError` exception). By
default, the types must match. To allow iteration over structures of
different types (but common arity), set this kwarg to `False`.
Returns:
A structure of the same form as the input structures whose leaves are the
result of evaluating func on corresponding leaves of the input structures.
Raises:
TypeError: If `func` is not callable or if the structures do not match
each other by depth tree.
TypeError: If `check_types` is not `False` and the two structures differ in
the type of sequence in any of their substructures.
ValueError: If no structures are provided."
11883,map_structure_with_tuple_paths,tensorflow/tensorflow/python/util/nest.py,683,function,"Applies `func` to each entry in `structure` and returns a new structure.
Applies `func(tuple_path, x[0], x[1], ..., **kwargs)` where `x[i]` is an entry
in `structure[i]` and `tuple_path` is a tuple of indices and/or dictionary
keys (as returned by `nest.yield_flat_paths`), which uniquely specifies the
common path to x[i] in the structures. All structures in `structure` must have
the same arity, and the return value will contain the results in the same
structure. Special kwarg `check_types` determines whether the types of
iterables within the structure must be the same-- see **kwargs definition
below.
Args:
func: A callable with the signature `func(tuple_path, *values, **kwargs)`
that is evaluated on the leaves of the structure.
*structure: A variable number of compatible structures to process.
**kwargs: Optional kwargs to be passed through to func. Special kwarg
`check_types` is not passed to func, but instead determines whether the
types of iterables within the structures have to be same (e.g.
`map_structure(func, [1], (1,))` raises a `TypeError` exception). To allow
this set this argument to `False`.
Returns:
A structure of the same form as the input structures whose leaves are the
result of evaluating func on corresponding leaves of the input structures.
Raises:
TypeError: If `func` is not callable or if the structures do not match
each other by depth tree.
TypeError: If `check_types` is not `False` and the two structures differ in
the type of sequence in any of their substructures.
ValueError: If no structures are provided."
11884,assert_shallow_structure,tensorflow/tensorflow/python/util/nest.py,752,function,"Asserts that `shallow_tree` is a shallow structure of `input_tree`.
That is, this function tests if the `input_tree` structure can be created from
the `shallow_tree` structure by replacing its leaf nodes with deeper
tree structures.
Examples:
The following code will raise an exception:
```python
shallow_tree = {""a"": ""A"", ""b"": ""B""}
input_tree = {""a"": 1, ""c"": 2}
assert_shallow_structure(shallow_tree, input_tree)
```
The following code will raise an exception:
```python
shallow_tree = [""a"", ""b""]
input_tree = [""c"", [""d"", ""e""], ""f""]
assert_shallow_structure(shallow_tree, input_tree)
```
Args:
shallow_tree: an arbitrarily nested structure.
input_tree: an arbitrarily nested structure.
check_types: if `True` (default) the sequence types of `shallow_tree` and
`input_tree` have to be the same. Note that even with check_types==True,
this function will consider two different namedtuple classes with the same
name and _fields attribute to be the same class.
expand_composites: If true, then composite tensors such as
`tf.sparse.SparseTensor` and `tf.RaggedTensor` are expanded into their
component tensors.
Raises:
TypeError: If `shallow_tree` is a sequence but `input_tree` is not.
TypeError: If the sequence types of `shallow_tree` are different from
`input_tree`. Only raised if `check_types` is `True`.
ValueError: If the sequence lengths of `shallow_tree` are different from
`input_tree`."
11885,flatten_up_to,tensorflow/tensorflow/python/util/nest.py,875,function,"Flattens `input_tree` up to `shallow_tree`.
Any further depth in structure in `input_tree` is retained as elements in the
partially flatten output.
If `shallow_tree` and `input_tree` are not sequences, this returns a
single-element list: `[input_tree]`.
Use Case:
Sometimes we may wish to partially flatten a nested sequence, retaining some
of the nested structure. We achieve this by specifying a shallow structure,
`shallow_tree`, we wish to flatten up to.
The input, `input_tree`, can be thought of as having the same structure layout
as `shallow_tree`, but with leaf nodes that are themselves tree structures.
Examples:
```python
input_tree = [[[2, 2], [3, 3]], [[4, 9], [5, 5]]]
shallow_tree = [[True, True], [False, True]]
flattened_input_tree = flatten_up_to(shallow_tree, input_tree)
flattened_shallow_tree = flatten_up_to(shallow_tree, shallow_tree)
# Output is:
# [[2, 2], [3, 3], [4, 9], [5, 5]]
# [True, True, False, True]
```
```python
input_tree = [[('a', 1), [('b', 2), [('c', 3), [('d', 4)]]]]]
shallow_tree = [['level_1', ['level_2', ['level_3', ['level_4']]]]]
input_tree_flattened_as_shallow_tree = flatten_up_to(shallow_tree, input_tree)
input_tree_flattened = flatten(input_tree)
# Output is:
# [('a', 1), ('b', 2), ('c', 3), ('d', 4)]
# ['a', 1, 'b', 2, 'c', 3, 'd', 4]
```
Non-Sequence Edge Cases:
```python
flatten_up_to(0, 0) # Output: [0]
flatten_up_to(0, [0, 1, 2]) # Output: [[0, 1, 2]]
flatten_up_to([0, 1, 2], 0) # Output: TypeError
flatten_up_to([0, 1, 2], [0, 1, 2]) # Output: [0, 1, 2]
```
Args:
shallow_tree: a possibly pruned structure of input_tree.
input_tree: an arbitrarily nested structure or a scalar object.
Note, numpy arrays are considered scalars.
check_types: bool. If True, check that each node in shallow_tree has the
same type as the corresponding node in input_tree.
expand_composites: If true, then composite tensors such as
`tf.sparse.SparseTensor` and `tf.RaggedTensor` are expanded into their
component tensors.
Returns:
A Python list, the partially flattened version of `input_tree` according to
the structure of `shallow_tree`.
Raises:
TypeError: If `shallow_tree` is a sequence but `input_tree` is not.
TypeError: If the sequence types of `shallow_tree` are different from
`input_tree`.
ValueError: If the sequence lengths of `shallow_tree` are different from
`input_tree`."
11886,flatten_with_tuple_paths_up_to,tensorflow/tensorflow/python/util/nest.py,959,function,"Flattens `input_tree` up to `shallow_tree`.
Any further depth in structure in `input_tree` is retained as elements in the
partially flattened output.
Returns a list of (path, value) pairs, where value a leaf node in the
flattened tree, and path is the tuple path of that leaf in input_tree.
If `shallow_tree` and `input_tree` are not sequences, this returns a
single-element list: `[((), input_tree)]`.
Use Case:
Sometimes we may wish to partially flatten a nested sequence, retaining some
of the nested structure. We achieve this by specifying a shallow structure,
`shallow_tree`, we wish to flatten up to.
The input, `input_tree`, can be thought of as having the same structure layout
as `shallow_tree`, but with leaf nodes that are themselves tree structures.
Examples:
```python
input_tree = [[[2, 2], [3, 3]], [[4, 9], [5, 5]]]
shallow_tree = [[True, True], [False, True]]
flattened_input_tree = flatten_with_tuple_paths_up_to(shallow_tree,
input_tree)
flattened_shallow_tree = flatten_with_tuple_paths_up_to(shallow_tree,
shallow_tree)
# Output is:
# [((0, 0), [2, 2]),
# ((0, 1), [3, 3]),
# ((1, 0), [4, 9]),
# ((1, 1), [5, 5])]
#
# [((0, 0), True),
# ((0, 1), True),
# ((1, 0), False),
# ((1, 1), True)]
```
```python
input_tree = [[('a', 1), [('b', 2), [('c', 3), [('d', 4)]]]]]
shallow_tree = [['level_1', ['level_2', ['level_3', ['level_4']]]]]
input_tree_flattened_as_shallow_tree = flatten_up_to(shallow_tree, input_tree)
input_tree_flattened = flatten(input_tree)
# Output is:
# [((0, 0), ('a', 1)),
# ((0, 1, 0), ('b', 2)),
# ((0, 1, 1, 0), ('c', 3)),
# ((0, 1, 1, 1), ('d', 4))]
# ['a', 1, 'b', 2, 'c', 3, 'd', 4]
```
Non-Sequence Edge Cases:
```python
flatten_with_tuple_paths_up_to(0, 0) # Output: [(), 0]
flatten_with_tuple_paths_up_to(0, [0, 1, 2]) # Output: [(), [0, 1, 2]]
flatten_with_tuple_paths_up_to([0, 1, 2], 0) # Output: TypeError
flatten_with_tuple_paths_up_to([0, 1, 2], [0, 1, 2])
# Output: [((0,) 0), ((1,), 1), ((2,), 2)]
```
Args:
shallow_tree: a possibly pruned structure of input_tree.
input_tree: an arbitrarily nested structure or a scalar object.
Note, numpy arrays are considered scalars.
check_types: bool. If True, check that each node in shallow_tree has the
same type as the corresponding node in input_tree.
expand_composites: If true, then composite tensors such as
`tf.sparse.SparseTensor` and `tf.RaggedTensor` are expanded into their
component tensors.
Returns:
A Python list, the partially flattened version of `input_tree` according to
the structure of `shallow_tree`.
Raises:
TypeError: If `shallow_tree` is a sequence but `input_tree` is not.
TypeError: If the sequence types of `shallow_tree` are different from
`input_tree`.
ValueError: If the sequence lengths of `shallow_tree` are different from
`input_tree`."
11887,map_structure_up_to,tensorflow/tensorflow/python/util/nest.py,1063,function,"Applies a function or op to a number of partially flattened inputs.
The `inputs` are flattened up to `shallow_tree` before being mapped.
Use Case:
Sometimes we wish to apply a function to a partially flattened
sequence (for example when the function itself takes sequence inputs). We
achieve this by specifying a shallow structure, `shallow_tree` we wish to
flatten up to.
The `inputs`, can be thought of as having the same structure layout as
`shallow_tree`, but with leaf nodes that are themselves tree structures.
This function therefore will return something with the same base structure as
`shallow_tree`.
Examples:
```python
shallow_tree = [None, None]
inp_val = [1, 2, 3]
out = map_structure_up_to(shallow_tree, lambda x: 2 * x, inp_val)
# Output is: [2, 4]
```
```python
ab_tuple = collections.namedtuple(""ab_tuple"", ""a, b"")
op_tuple = collections.namedtuple(""op_tuple"", ""add, mul"")
inp_val = ab_tuple(a=2, b=3)
inp_ops = ab_tuple(a=op_tuple(add=1, mul=2), b=op_tuple(add=2, mul=3))
out = map_structure_up_to(inp_val, lambda val, ops: (val + ops.add) * ops.mul,
inp_val, inp_ops)
# Output is: ab_tuple(a=6, b=15)
```
```python
data_list = [[2, 4, 6, 8], [[1, 3, 5, 7, 9], [3, 5, 7]]]
name_list = ['evens', ['odds', 'primes']]
out = map_structure_up_to(
name_list,
lambda name, sec: ""first_{}_{}"".format(len(sec), name),
name_list, data_list)
# Output is: ['first_4_evens', ['first_5_odds', 'first_3_primes']]
```
Args:
shallow_tree: a shallow tree, common to all the inputs.
func: callable which will be applied to each input individually.
*inputs: arbitrarily nested combination of objects that are compatible with
shallow_tree. The function `func` is applied to corresponding
partially flattened elements of each input, so the function must support
arity of `len(inputs)`.
**kwargs: kwargs to feed to func(). Special kwarg
`check_types` is not passed to func, but instead determines whether the
types of iterables within the structures have to be same (e.g.
`map_structure(func, [1], (1,))` raises a `TypeError` exception). To allow
this set this argument to `False`.
Raises:
TypeError: If `shallow_tree` is a sequence but `input_tree` is not.
TypeError: If the sequence types of `shallow_tree` are different from
`input_tree`.
ValueError: If the sequence lengths of `shallow_tree` are different from
`input_tree`.
Returns:
result of repeatedly applying `func`, with the same structure layout as
`shallow_tree`."
11888,map_structure_with_tuple_paths_up_to,tensorflow/tensorflow/python/util/nest.py,1144,function,"Applies a function or op to a number of partially flattened inputs.
Like map_structure_up_to(), except that the 'func' argument takes a path
tuple as its first argument, followed by the corresponding values from
*inputs.
Example:
```python
lowercase = {'a': 'a', 'b': ('b0', 'b1')}
uppercase = {'a': 'A', 'b': ('B0', 'B1')}
def print_path_and_values(path, *values):
print(""path: {}, values: {}"".format(path, values))
shallow_tree = {'a': None}
map_structure_with_tuple_paths_up_to(shallow_tree,
print_path_and_values,
lowercase,
uppercase)
path: ('a',), values: ('a', 'A')
path: ('b', 0), values: ('b0', 'B0')
path: ('b', 1), values: ('b1', 'B1')
shallow_tree = {'b': None}
map_structure_with_tuple_paths_up_to(shallow_tree,
print_path_and_values,
lowercase,
uppercase,
check_types=False)
path: ('b', 1), values: (('bo', 'b1'), ('B0', 'B1'))
shallow_tree = {'a': None, 'b': {1: None}}
map_structure_with_tuple_paths_up_to(shallow_tree,
print_path_and_values,
lowercase,
uppercase,
check_types=False)
path: ('a',), values: ('a', 'A')
path: ('b', 1), values: ('b1', B1')
```
Args:
shallow_tree: a shallow tree, common to all the inputs.
func: callable that takes args (path, inputs_0_value, ... , inputs_N_value),
where path is a tuple path to a leaf node in shallow_tree, and
inputs_i_value is the corresponding value from inputs[i].
*inputs: nested structures that are all structurally compatible with
shallow_tree.
**kwargs: kwargs to feed to func(). Special kwarg
`check_types` is not passed to func, but instead determines whether the
types of iterables within the structures have to be same (e.g.
`map_structure(func, [1], (1,))` raises a `TypeError` exception). To allow
this set this argument to `False`.
Raises:
TypeError: If `shallow_tree` is a sequence but one of `*inputs` is not.
TypeError: If the sequence types of `shallow_tree` are different from
`input_tree`.
ValueError: If the sequence lengths of `shallow_tree` are different from
`input_tree`.
Returns:
Result of repeatedly applying `func`. Has the same structure layout as
`shallow_tree`."
11889,get_traverse_shallow_structure,tensorflow/tensorflow/python/util/nest.py,1242,function,"Generates a shallow structure from a `traverse_fn` and `structure`.
`traverse_fn` must accept any possible subtree of `structure` and return
a depth=1 structure containing `True` or `False` values, describing which
of the top-level subtrees may be traversed. It may also
return scalar `True` or `False` ""traversal is OK / not OK for all subtrees.""
Examples are available in the unit tests (nest_test.py).
Args:
traverse_fn: Function taking a substructure and returning either a scalar
`bool` (whether to traverse that substructure or not) or a depth=1
shallow structure of the same type, describing which parts of the
substructure to traverse.
structure: The structure to traverse.
expand_composites: If true, then composite tensors such as
`tf.sparse.SparseTensor` and `tf.RaggedTensor` are expanded into their
component tensors.
Returns:
A shallow structure containing python bools, which can be passed to
`map_structure_up_to` and `flatten_up_to`.
Raises:
TypeError: if `traverse_fn` returns a sequence for a non-sequence input,
or a structure with depth higher than 1 for a sequence input,
or if any leaf values in the returned structure or scalar are not type
`bool`."
11890,yield_flat_paths,tensorflow/tensorflow/python/util/nest.py,1312,function,"Yields paths for some nested structure.
Paths are lists of objects which can be str-converted, which may include
integers or other types which are used as indices in a dict.
The flat list will be in the corresponding order as if you called
`nest.flatten` on the structure. This is handy for naming Tensors such
the TF scope structure matches the tuple structure.
E.g. if we have a tuple `value = Foo(a=3, b=Bar(c=23, d=42))`
```shell
nest.flatten(value)
[3, 23, 42]
list(nest.yield_flat_paths(value))
[('a',), ('b', 'c'), ('b', 'd')]
```
```shell
list(nest.yield_flat_paths({'a': [3]}))
[('a', 0)]
list(nest.yield_flat_paths({'a': 3}))
[('a',)]
```
Args:
nest: the value to produce a flattened paths list for.
expand_composites: If true, then composite tensors such as
`tf.sparse.SparseTensor` and `tf.RaggedTensor` are expanded into their
component tensors.
Yields:
Tuples containing index or key values which form the path to a specific
leaf value in the nested structure."
11891,flatten_with_joined_string_paths,tensorflow/tensorflow/python/util/nest.py,1353,function,"Returns a list of (string path, data element) tuples.
The order of tuples produced matches that of `nest.flatten`. This allows you
to flatten a nested structure while keeping information about where in the
structure each data element was located. See `nest.yield_flat_paths`
for more information.
Args:
structure: the nested structure to flatten.
separator: string to separate levels of hierarchy in the results, defaults
to '/'.
expand_composites: If true, then composite tensors such as
`tf.sparse.SparseTensor` and `tf.RaggedTensor` are expanded into their
component tensors.
Returns:
A list of (string, data element) tuples."
11892,flatten_with_tuple_paths,tensorflow/tensorflow/python/util/nest.py,1381,function,"Returns a list of `(tuple_path, leaf_element)` tuples.
The order of pairs produced matches that of `nest.flatten`. This allows you
to flatten a nested structure while keeping information about where in the
structure each data element was located. See `nest.yield_flat_paths`
for more information about tuple paths.
Args:
structure: the nested structure to flatten.
expand_composites: If true, then composite tensors such as
`tf.sparse.SparseTensor` and `tf.RaggedTensor` are expanded into their
component tensors.
Returns:
A list of `(tuple_path, leaf_element)` tuples. Each `tuple_path` is a tuple
of indices and/or dictionary keys that uniquely specify the path to
`leaf_element` within `structure`."
11893,list_to_tuple,tensorflow/tensorflow/python/util/nest.py,1405,function,"Replace all lists with tuples.
The fork of nest that tf.data uses treats lists as single elements, while
tf.nest treats them as structures to recurse into. Keras has chosen to adopt
the latter convention, and must therefore deeply replace all lists with tuples
before passing structures to Dataset.from_generator.
Args:
structure: A nested structure to be remapped.
Returns:
structure mapped to replace all lists with tuples."
11894,NestBenchmark,tensorflow/tensorflow/python/util/nest_test.py,1220,class,
11895,run_and_report,tensorflow/tensorflow/python/util/nest_test.py,1222,method,
11896,benchmark_assert_structure,tensorflow/tensorflow/python/util/nest_test.py,1236,method,
11897,Reference,tensorflow/tensorflow/python/util/object_identity.py,83,class,"Reference that refers an object.
```python
x = [1]
y = [1]
x_ref1 = Reference(x)
x_ref2 = Reference(x)
y_ref2 = Reference(y)
print(x_ref1 == x_ref2)
==> True
print(x_ref1 == y)
==> False
```"
11898,deref,tensorflow/tensorflow/python/util/object_identity.py,105,method,"Returns the referenced object.
```python
x_ref = Reference(x)
print(x is x_ref.deref())
==> True
```"
11899,ObjectIdentityDictionary,tensorflow/tensorflow/python/util/object_identity.py,117,class,"A mutable mapping data structure which compares using ""is"".
This is necessary because we have trackable objects (_ListWrapper) which
have behavior identical to built-in Python lists (including being unhashable
and comparing based on the equality of their contents by default)."
11900,ObjectIdentityWeakKeyDictionary,tensorflow/tensorflow/python/util/object_identity.py,153,class,"Like weakref.WeakKeyDictionary, but compares objects with ""is""."
11901,ObjectIdentitySet,tensorflow/tensorflow/python/util/object_identity.py,173,class,"Like the built-in set, but compares objects with ""is""."
11902,discard,tensorflow/tensorflow/python/util/object_identity.py,193,method,
11903,add,tensorflow/tensorflow/python/util/object_identity.py,196,method,
11904,update,tensorflow/tensorflow/python/util/object_identity.py,199,method,
11905,clear,tensorflow/tensorflow/python/util/object_identity.py,202,method,
11906,intersection,tensorflow/tensorflow/python/util/object_identity.py,205,method,
11907,difference,tensorflow/tensorflow/python/util/object_identity.py,208,method,
11908,ObjectIdentityWeakSet,tensorflow/tensorflow/python/util/object_identity.py,221,class,"Like weakref.WeakSet, but compares objects with ""is""."
11909,get_json_type,tensorflow/tensorflow/python/util/serialization.py,29,function,"Serializes any object to a JSON-serializable structure.
Arguments:
obj: the object to serialize
Returns:
JSON-serializable structure representing `obj`.
Raises:
TypeError: if `obj` cannot be serialized."
11910,contextmanager,tensorflow/tensorflow/python/util/tf_contextlib.py,25,function,"A tf_decorator-aware wrapper for `contextlib.contextmanager`.
Usage is identical to `contextlib.contextmanager`.
Args:
target: A callable to be wrapped in a contextmanager.
Returns:
A callable that can be used inside of a `with` statement."
11911,make_decorator,tensorflow/tensorflow/python/util/tf_decorator.py,67,function,"Make a decorator from a wrapper and a target.
Args:
target: The final callable to be wrapped.
decorator_func: The wrapper function.
decorator_name: The name of the decorator. If `None`, the name of the
function calling make_decorator.
decorator_doc: Documentation specific to this application of
`decorator_func` to `target`.
decorator_argspec: The new callable signature of this decorator.
Returns:
The `decorator_func` argument with new metadata attached."
11912,rewrap,tensorflow/tensorflow/python/util/tf_decorator.py,128,function,"Injects a new target into a function built by make_decorator.
This function allows replacing a function wrapped by `decorator_func`,
assuming the decorator that wraps the function is written as described below.
The decorator function must use `<decorator name>.__wrapped__` instead of the
wrapped function that is normally used:
Example:
# Instead of this:
def simple_parametrized_wrapper(*args, **kwds):
return wrapped_fn(*args, **kwds)
tf_decorator.make_decorator(simple_parametrized_wrapper, wrapped_fn)
# Write this:
def simple_parametrized_wrapper(*args, **kwds):
return simple_parametrized_wrapper.__wrapped__(*args, **kwds)
tf_decorator.make_decorator(simple_parametrized_wrapper, wrapped_fn)
Note that this process modifies decorator_func.
Args:
decorator_func: Callable returned by `wrap`.
previous_target: Callable that needs to be replaced.
new_target: Callable to replace previous_target with.
Returns:
The updated decorator. If decorator_func is not a tf_decorator, new_target
is returned."
11913,unwrap,tensorflow/tensorflow/python/util/tf_decorator.py,200,function,"Unwraps an object into a list of TFDecorators and a final target.
Args:
maybe_tf_decorator: Any callable object.
Returns:
A tuple whose first element is an list of TFDecorator-derived objects that
were applied to the final callable target, and whose second element is the
final undecorated callable target. If the `maybe_tf_decorator` parameter is
not decorated by any TFDecorators, the first tuple element will be an empty
list. The `TFDecorator` list is ordered from outermost to innermost
decorators."
11914,TFDecorator,tensorflow/tensorflow/python/util/tf_decorator.py,229,class,"Base class for all TensorFlow decorators.
TFDecorator captures and exposes the wrapped target, and provides details
about the current decorator."
11915,decorated_target,tensorflow/tensorflow/python/util/tf_decorator.py,263,method,
11916,decorated_target,tensorflow/tensorflow/python/util/tf_decorator.py,267,method,
11917,decorator_name,tensorflow/tensorflow/python/util/tf_decorator.py,271,method,
11918,decorator_doc,tensorflow/tensorflow/python/util/tf_decorator.py,275,method,
11919,decorator_argspec,tensorflow/tensorflow/python/util/tf_decorator.py,279,method,
11920,SymbolAlreadyExposedError,tensorflow/tensorflow/python/util/tf_export.py,88,class,Raised when adding API names to symbol that already has API names.
11921,InvalidSymbolNameError,tensorflow/tensorflow/python/util/tf_export.py,93,class,Raised when trying to export symbol as an invalid or unallowed name.
11922,get_symbol_from_name,tensorflow/tensorflow/python/util/tf_export.py,100,function,
11923,get_canonical_name_for_symbol,tensorflow/tensorflow/python/util/tf_export.py,104,function,"Get canonical name for the API symbol.
Args:
symbol: API function or class.
api_name: API name (tensorflow or estimator).
add_prefix_to_v1_names: Specifies whether a name available only in V1
should be prefixed with compat.v1.
Returns:
Canonical name for the API symbol (for e.g. initializers.zeros) if
canonical name could be determined. Otherwise, returns None."
11924,get_canonical_name,tensorflow/tensorflow/python/util/tf_export.py,142,function,"Get preferred endpoint name.
Args:
api_names: API names iterable.
deprecated_api_names: Deprecated API names iterable.
Returns:
Returns one of the following in decreasing preference:
- first non-deprecated endpoint
- first endpoint
- None"
11925,get_v1_names,tensorflow/tensorflow/python/util/tf_export.py,164,function,"Get a list of TF 1.* names for this symbol.
Args:
symbol: symbol to get API names for.
Returns:
List of all API names for this symbol including TensorFlow and
Estimator names."
11926,get_v2_names,tensorflow/tensorflow/python/util/tf_export.py,190,function,"Get a list of TF 2.0 names for this symbol.
Args:
symbol: symbol to get API names for.
Returns:
List of all API names for this symbol including TensorFlow and
Estimator names."
11927,get_v1_constants,tensorflow/tensorflow/python/util/tf_export.py,216,function,"Get a list of TF 1.* constants in this module.
Args:
module: TensorFlow module.
Returns:
List of all API constants under the given module including TensorFlow and
Estimator constants."
11928,get_v2_constants,tensorflow/tensorflow/python/util/tf_export.py,237,function,"Get a list of TF 2.0 constants in this module.
Args:
module: TensorFlow module.
Returns:
List of all API constants under the given module including TensorFlow and
Estimator constants."
11929,api_export,tensorflow/tensorflow/python/util/tf_export.py,258,class,Provides ways to export symbols to the TensorFlow API.
11930,set_attr,tensorflow/tensorflow/python/util/tf_export.py,349,method,
11931,export_constant,tensorflow/tensorflow/python/util/tf_export.py,360,method,"Store export information for constants/string literals.
Export information is stored in the module where constants/string literals
are defined.
e.g.
```python
foo = 1
bar = 2
tf_export(""consts.foo"").export_constant(__name__, 'foo')
tf_export(""consts.bar"").export_constant(__name__, 'bar')
```
Args:
module_name: (string) Name of the module to store constant at.
name: (string) Current constant name."
11932,kwarg_only,tensorflow/tensorflow/python/util/tf_export.py,394,function,A wrapper that throws away all non-kwarg arguments.
11933,currentframe,tensorflow/tensorflow/python/util/tf_inspect.py,93,function,TFDecorator-aware replacement for inspect.currentframe.
11934,getargspec,tensorflow/tensorflow/python/util/tf_inspect.py,98,function,"TFDecorator-aware replacement for `inspect.getargspec`.
Note: `getfullargspec` is recommended as the python 2/3 compatible
replacement for this function.
Args:
obj: A function, partial function, or callable object, possibly decorated.
Returns:
The `ArgSpec` that describes the signature of the outermost decorator that
changes the callable's signature, or the `ArgSpec` that describes
the object if not decorated.
Raises:
ValueError: When callable's signature can not be expressed with
ArgSpec.
TypeError: For objects of unsupported types."
11935,getfullargspec,tensorflow/tensorflow/python/util/tf_inspect.py,238,function,"TFDecorator-aware replacement for `inspect.getfullargspec`.
This wrapper emulates `inspect.getfullargspec` in[^)]* Python2.
Args:
obj: A callable, possibly decorated.
Returns:
The `FullArgSpec` that describes the signature of
the outermost decorator that changes the callable's signature. If the
callable is not decorated, `inspect.getfullargspec()` will be called
directly on the callable."
11936,getcallargs,tensorflow/tensorflow/python/util/tf_inspect.py,260,function,"TFDecorator-aware replacement for inspect.getcallargs.
Args:
*func_and_positional: A callable, possibly decorated, followed by any
positional arguments that would be passed to `func`.
**named: The named argument dictionary that would be passed to `func`.
Returns:
A dictionary mapping `func`'s named arguments to the values they would
receive if `func(*positional, **named)` were called.
`getcallargs` will use the argspec from the outermost decorator that provides
it. If no attached decorators modify argspec, the final unwrapped target's
argspec will be used."
11937,getframeinfo,tensorflow/tensorflow/python/util/tf_inspect.py,297,function,
11938,getdoc,tensorflow/tensorflow/python/util/tf_inspect.py,301,function,"TFDecorator-aware replacement for inspect.getdoc.
Args:
object: An object, possibly decorated.
Returns:
The docstring associated with the object.
The outermost-decorated object is intended to have the most complete
documentation, so the decorated parameter is not unwrapped."
11939,getfile,tensorflow/tensorflow/python/util/tf_inspect.py,316,function,TFDecorator-aware replacement for inspect.getfile.
11940,getmembers,tensorflow/tensorflow/python/util/tf_inspect.py,330,function,TFDecorator-aware replacement for inspect.getmembers.
11941,getmodule,tensorflow/tensorflow/python/util/tf_inspect.py,335,function,TFDecorator-aware replacement for inspect.getmodule.
11942,getmro,tensorflow/tensorflow/python/util/tf_inspect.py,340,function,TFDecorator-aware replacement for inspect.getmro.
11943,getsource,tensorflow/tensorflow/python/util/tf_inspect.py,345,function,TFDecorator-aware replacement for inspect.getsource.
11944,getsourcefile,tensorflow/tensorflow/python/util/tf_inspect.py,350,function,TFDecorator-aware replacement for inspect.getsourcefile.
11945,getsourcelines,tensorflow/tensorflow/python/util/tf_inspect.py,355,function,TFDecorator-aware replacement for inspect.getsourcelines.
11946,isbuiltin,tensorflow/tensorflow/python/util/tf_inspect.py,360,function,TFDecorator-aware replacement for inspect.isbuiltin.
11947,isclass,tensorflow/tensorflow/python/util/tf_inspect.py,365,function,TFDecorator-aware replacement for inspect.isclass.
11948,isfunction,tensorflow/tensorflow/python/util/tf_inspect.py,370,function,TFDecorator-aware replacement for inspect.isfunction.
11949,isframe,tensorflow/tensorflow/python/util/tf_inspect.py,375,function,TFDecorator-aware replacement for inspect.ismodule.
11950,isgenerator,tensorflow/tensorflow/python/util/tf_inspect.py,380,function,TFDecorator-aware replacement for inspect.isgenerator.
11951,isgeneratorfunction,tensorflow/tensorflow/python/util/tf_inspect.py,385,function,TFDecorator-aware replacement for inspect.isgeneratorfunction.
11952,ismethod,tensorflow/tensorflow/python/util/tf_inspect.py,390,function,TFDecorator-aware replacement for inspect.ismethod.
11953,ismodule,tensorflow/tensorflow/python/util/tf_inspect.py,395,function,TFDecorator-aware replacement for inspect.ismodule.
11954,isroutine,tensorflow/tensorflow/python/util/tf_inspect.py,400,function,TFDecorator-aware replacement for inspect.isroutine.
11955,stack,tensorflow/tensorflow/python/util/tf_inspect.py,405,function,TFDecorator-aware replacement for inspect.stack.
11956,should_use_result,tensorflow/tensorflow/python/util/tf_should_use.py,216,function,"Function wrapper that ensures the function's output is used.
If the output is not used, a `logging.error` is logged. If
`error_in_function` is set, then a `RuntimeError` will be raised at the
end of function tracing if the output is not used by that point.
An output is marked as used if any of its attributes are read, modified, or
updated. Examples when the output is a `Tensor` include:
- Using it in any capacity (e.g. `y = t + 0`, `sess.run(t)`)
- Accessing a property (e.g. getting `t.name` or `t.op`).
- Calling `t.mark_used()`.
Note, certain behaviors cannot be tracked - for these the object may not
be marked as used. Examples include:
- `t != 0`. In this case, comparison is done on types / ids.
- `isinstance(t, tf.Tensor)`. Similar to above.
Args:
fn: The function to wrap.
warn_in_eager: Whether to create warnings in Eager as well.
error_in_function: Whether to raise an error when creating a tf.function.
Returns:
The wrapped function."
11957,reroute_error,tensorflow/tensorflow/python/util/tf_should_use_test.py,36,function,Temporarily reroute errors written to tf_logging.error into `captured`.
11958,StackTraceTransform,tensorflow/tensorflow/python/util/tf_stack.py,47,class,Base class for stack trace transformation functions.
11959,reset,tensorflow/tensorflow/python/util/tf_stack.py,75,method,
11960,StackTraceMapper,tensorflow/tensorflow/python/util/tf_stack.py,79,class,Allows remapping traceback information to different source code.
11961,reset,tensorflow/tensorflow/python/util/tf_stack.py,83,method,
11962,get_effective_source_map,tensorflow/tensorflow/python/util/tf_stack.py,86,method,"Returns a map (filename, lineno) -> (filename, lineno, function_name)."
11963,StackTraceFilter,tensorflow/tensorflow/python/util/tf_stack.py,91,class,Allows filtering traceback information by removing superfluous frames.
11964,reset,tensorflow/tensorflow/python/util/tf_stack.py,95,method,
11965,get_filtered_filenames,tensorflow/tensorflow/python/util/tf_stack.py,98,method,
11966,CurrentModuleFilter,tensorflow/tensorflow/python/util/tf_stack.py,102,class,Filters stack frames from the module where this is used (best effort).
11967,get_filtered_filenames,tensorflow/tensorflow/python/util/tf_stack.py,123,method,
11968,extract_stack,tensorflow/tensorflow/python/util/tf_stack.py,131,function,"A lightweight, extensible re-implementation of traceback.extract_stack.
NOTE(mrry): traceback.extract_stack eagerly retrieves the line of code for
each stack frame using linecache, which results in an abundance of stat()
calls. This implementation does not retrieve the code, and any consumer
should apply _convert_stack to the result to obtain a traceback that can
be formatted etc. using traceback methods.
Args:
limit: A limit on the number of frames to return.
Returns:
A sequence of FrameSummary objects (filename, lineno, name, line)
corresponding to the call stack of the current thread."
11969,extract_stack,tensorflow/tensorflow/python/util/tf_stack_test.py,56,function,
11970,convert_stack_frame,tensorflow/tensorflow/python/util/tf_stack_test.py,61,function,Converts a TF stack frame into Python's.
11971,assertProtoEqual,tensorflow/tensorflow/python/util/protobuf/compare.py,77,function,"Fails with a useful error if a and b aren't equal.
Comparison of repeated fields matches the semantics of
unittest.TestCase.assertEqual(), ie order and extra duplicates fields matter.
Args:
self: googletest.TestCase
a: proto2 PB instance, or text string representing one.
b: proto2 PB instance -- message.Message or subclass thereof.
check_initialized: boolean, whether to fail if either a or b isn't
initialized.
normalize_numbers: boolean, whether to normalize types and precision of
numbers before comparison.
msg: if specified, is used as the error message on failure."
11972,NormalizeNumberFields,tensorflow/tensorflow/python/util/protobuf/compare.py,121,function,"Normalizes types and precisions of number fields in a protocol buffer.
Due to subtleties in the python protocol buffer implementation, it is possible
for values to have different types and precision depending on whether they
were set and retrieved directly or deserialized from a protobuf. This function
normalizes integer values to ints and longs based on width, 32-bit floats to
five digits of precision to account for python always storing them as 64-bit,
and ensures doubles are floating point for when they're set to integers.
Modifies pb in place. Recurses into nested objects.
Args:
pb: proto2 message.
Returns:
the given pb, modified in place."
11973,ProtoEq,tensorflow/tensorflow/python/util/protobuf/compare.py,203,function,"Compares two proto2 objects for equality.
Recurses into nested messages. Uses list (not set) semantics for comparing
repeated fields, ie duplicates and order matter.
Args:
a: A proto2 message or a primitive.
b: A proto2 message or a primitive.
Returns:
`True` if the messages are equal."
11974,ProtoAssertions,tensorflow/tensorflow/python/util/protobuf/compare.py,258,class,"Mix this into a googletest.TestCase class to get proto2 assertions.
Usage:
class SomeTestCase(compare.ProtoAssertions, googletest.TestCase):
...
def testSomething(self):
...
self.assertProtoEqual(a, b)
See module-level definitions for method documentation."
11975,assertProtoEqual,tensorflow/tensorflow/python/util/protobuf/compare.py,273,method,
11976,LargePbs,tensorflow/tensorflow/python/util/protobuf/compare_test.py,34,function,Converts ASCII string Large PBs to messages.
11977,PythonObjectToProtoVisitor,tensorflow/tensorflow/tools/api/lib/python_object_to_proto_visitor.py,196,class,A visitor that summarizes given python objects as protobufs.
11978,GetProtos,tensorflow/tensorflow/tools/api/lib/python_object_to_proto_visitor.py,204,method,Return the list of protos stored.
11979,write_build_info,tensorflow/tensorflow/tools/build_info/gen_build_info.py,32,function,"Writes a Python that describes the build.
Args:
filename: filename to write to.
key_value_list: A list of ""key=value"" strings that will be added to the
module's ""build_info"" dictionary as additional entries."
11980,check_existence,tensorflow/tensorflow/tools/ci_build/copy_binary.py,40,function,Check the existence of file or dir.
11981,copy_binary,tensorflow/tensorflow/tools/ci_build/copy_binary.py,46,function,"Rename and copy binaries for different python versions.
Arguments:
directory: string of directory
origin_tag: str of the old python version tag
new_tag: str of the new tag
version: the version of the package
package: str, name of the package"
11982,check_existence,tensorflow/tensorflow/tools/ci_build/update_version.py,46,function,Check the existence of file or dir.
11983,check_all_files,tensorflow/tensorflow/tools/ci_build/update_version.py,53,function,Check all relevant files necessary for upgrade.
11984,replace_string_in_line,tensorflow/tensorflow/tools/ci_build/update_version.py,59,function,Replace with sed when regex is required.
11985,Version,tensorflow/tensorflow/tools/ci_build/update_version.py,67,class,Version class object that stores SemVer version information.
11986,set_identifier_string,tensorflow/tensorflow/tools/ci_build/update_version.py,96,method,
11987,pep_440_str,tensorflow/tensorflow/tools/ci_build/update_version.py,101,method,
11988,parse_from_string,tensorflow/tensorflow/tools/ci_build/update_version.py,115,method,"Returns version object from Semver string.
Args:
string: version string
version_type: version parameter
Raises:
RuntimeError: If the version string is not valid."
11989,get_current_semver_version,tensorflow/tensorflow/tools/ci_build/update_version.py,146,function,"Returns a Version object of current version.
Returns:
version: Version object of current SemVer string based on information from
core/public/version.h"
11990,update_version_h,tensorflow/tensorflow/tools/ci_build/update_version.py,183,function,Update tensorflow/core/public/version.h.
11991,update_setup_dot_py,tensorflow/tensorflow/tools/ci_build/update_version.py,200,function,Update setup.py.
11992,update_readme,tensorflow/tensorflow/tools/ci_build/update_version.py,206,function,Update README.
11993,update_tensorflow_bzl,tensorflow/tensorflow/tools/ci_build/update_version.py,214,function,Update tensorflow.bzl.
11994,major_minor_change,tensorflow/tensorflow/tools/ci_build/update_version.py,224,function,Check if a major or minor change occurred.
11995,check_for_lingering_string,tensorflow/tensorflow/tools/ci_build/update_version.py,233,function,Check for given lingering strings.
11996,check_for_old_version,tensorflow/tensorflow/tools/ci_build/update_version.py,255,function,Check for old version references.
11997,IntelPlatform,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,34,class,
11998,set_host_gcc_version,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,50,method,
11999,get_bazel_gcc_flags,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,73,method,
12000,use_old_arch_names,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,79,method,
12001,NehalemPlatform,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,89,class,
12002,get_bazel_gcc_flags,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,94,method,
12003,SandyBridgePlatform,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,105,class,
12004,get_bazel_gcc_flags,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,110,method,
12005,HaswellPlatform,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,121,class,
12006,get_bazel_gcc_flags,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,126,method,
12007,SkylakePlatform,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,140,class,
12008,get_bazel_gcc_flags,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,145,method,
12009,CascadelakePlatform,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,163,class,
12010,get_bazel_gcc_flags,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,168,method,
12011,BuildEnvSetter,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,183,class,Prepares the proper environment settings for various Intel platforms.
12012,get_gcc_version,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,201,method,
12013,parse_args,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,232,method,"Set up argument parser, and parse CLI args."
12014,validate_args,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,287,method,
12015,set_build_args,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,307,method,Generate Bazel build flags.
12016,write_build_args,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,325,method,
12017,go,tensorflow/tensorflow/tools/ci_build/linux/mkl/set-build-env.py,333,method,
12018,PublicAPIVisitor,tensorflow/tensorflow/tools/common/public_api.py,29,class,Visitor to use with `traverse` to visit exactly the public TF API.
12019,private_map,tensorflow/tensorflow/tools/common/public_api.py,81,method,"A map from parents to symbols that should not be included at all.
This map can be edited, but it should not be edited once traversal has
begun.
Returns:
The map marking symbols to not include."
12020,do_not_descend_map,tensorflow/tensorflow/tools/common/public_api.py,93,method,"A map from parents to symbols that should not be descended into.
This map can be edited, but it should not be edited once traversal has
begun.
Returns:
The map marking symbols to not explore."
12021,set_root_name,tensorflow/tensorflow/tools/common/public_api.py,104,method,Override the default root name of 'tf'.
12022,ModuleClass1,tensorflow/tensorflow/tools/common/test_module1.py,24,class,
12023,ModuleClass2,tensorflow/tensorflow/tools/common/test_module2.py,22,class,
12024,traverse,tensorflow/tensorflow/tools/common/traverse.py,77,function,"Recursively enumerate all members of `root`.
Similar to the Python library function `os.path.walk`.
Traverses the tree of Python objects starting with `root`, depth first.
Parent-child relationships in the tree are defined by membership in modules or
classes. The function `visit` is called with arguments
`(path, parent, children)` for each module or class `parent` found in the tree
of python objects starting with `root`. `path` is a string containing the name
with which `parent` is reachable from the current context. For example, if
`root` is a local class called `X` which contains a class `Y`, `visit` will be
called with `('Y', X.Y, children)`).
If `root` is not a module or class, `visit` is never called. `traverse`
never descends into built-in modules.
`children`, a list of `(name, object)` pairs are determined by
`tf_inspect.getmembers`. To avoid visiting parts of the tree, `children` can
be modified in place, using `del` or slice assignment.
Cycles (determined by reference equality, `is`) stop the traversal. A stack of
objects is kept to find cycles. Objects forming cycles may appear in
`children`, but `visit` will not be called with any object as `parent` which
is already in the stack.
Traversing system modules can take a long time, it is advisable to pass a
`visit` callable which denylists such modules.
Args:
root: A python object with which to start the traversal.
visit: A function taking arguments `(path, parent, children)`. Will be
called for each object found in the traversal."
12025,add_contrib_direct_import_support,tensorflow/tensorflow/tools/compatibility/all_renames_v2.py,560,function,Add support for `tf.contrib.*` alias `contrib_*.` Updates dict in place.
12026,full_name_node,tensorflow/tensorflow/tools/compatibility/ast_edits.py,49,function,"Make an Attribute or Name node for name.
Translate a qualified name into nested Attribute nodes (and a Name node).
Args:
name: The name to translate to a node.
ctx: What context this name is used in. Defaults to Load()
Returns:
A Name or Attribute node."
12027,get_arg_value,tensorflow/tensorflow/tools/compatibility/ast_edits.py,72,function,"Get the value of an argument from a ast.Call node.
This function goes through the positional and keyword arguments to check
whether a given argument was used, and if so, returns its value (the node
representing its value).
This cannot introspect *args or **args, but it safely handles *args in
Python3.5+.
Args:
node: The ast.Call node to extract arg values from.
arg_name: The name of the argument to extract.
arg_pos: The position of the argument (in case it's passed as a positional
argument).
Returns:
A tuple (arg_present, arg_value) containing a boolean indicating whether
the argument is present, and its value in case it is."
12028,uses_star_args_in_call,tensorflow/tensorflow/tools/compatibility/ast_edits.py,111,function,"Check if an ast.Call node uses arbitrary-length positional *args.
This function works with the AST call node format of Python3.5+
as well as the different AST format of earlier versions of Python.
Args:
node: The ast.Call node to check arg values for.
Returns:
True if the node uses starred variadic positional args or keyword args.
False if it does not."
12029,uses_star_kwargs_in_call,tensorflow/tensorflow/tools/compatibility/ast_edits.py,135,function,"Check if an ast.Call node uses arbitrary-length **kwargs.
This function works with the AST call node format of Python3.5+
as well as the different AST format of earlier versions of Python.
Args:
node: The ast.Call node to check arg values for.
Returns:
True if the node uses starred variadic positional args or keyword args.
False if it does not."
12030,uses_star_args_or_kwargs_in_call,tensorflow/tensorflow/tools/compatibility/ast_edits.py,159,function,"Check if an ast.Call node uses arbitrary-length *args or **kwargs.
This function works with the AST call node format of Python3.5+
as well as the different AST format of earlier versions of Python.
Args:
node: The ast.Call node to check arg values for.
Returns:
True if the node uses starred variadic positional args or keyword args.
False if it does not."
12031,excluded_from_module_rename,tensorflow/tensorflow/tools/compatibility/ast_edits.py,175,function,"Check if this module import should not be renamed.
Args:
module: (string) module name.
import_rename_spec: ImportRename instance.
Returns:
True if this import should not be renamed according to the
import_rename_spec."
12032,APIChangeSpec,tensorflow/tensorflow/tools/compatibility/ast_edits.py,192,class,"This class defines the transformations that need to happen.
This class must provide the following fields:
* `function_keyword_renames`: maps function names to a map of old -> new
argument names
* `symbol_renames`: maps function names to new function names
* `change_to_function`: a set of function names that have changed (for
notifications)
* `function_reorders`: maps functions whose argument order has changed to the
list of arguments in the new order
* `function_warnings`: maps full names of functions to warnings that will be
printed out if the function is used. (e.g. tf.nn.convolution())
* `function_transformers`: maps function names to custom handlers
* `module_deprecations`: maps module names to warnings that will be printed
if the module is still used after all other transformations have run
* `import_renames`: maps import name (must be a short name without '.')
to ImportRename instance.
For an example, see `TFAPIChangeSpec`."
12033,preprocess,tensorflow/tensorflow/tools/compatibility/ast_edits.py,215,method,"Preprocess a parse tree. Return a preprocessed node, logs and errors."
12034,clear_preprocessing,tensorflow/tensorflow/tools/compatibility/ast_edits.py,219,method,"Restore this APIChangeSpec to before it preprocessed a file.
This is needed if preprocessing a file changed any rewriting rules."
12035,NoUpdateSpec,tensorflow/tensorflow/tools/compatibility/ast_edits.py,227,class,A specification of an API change which doesn't change anything.
12036,AnalysisResult,tensorflow/tensorflow/tools/compatibility/ast_edits.py,787,class,"This class represents an analysis result and how it should be logged.
This class must provide the following fields:
* `log_level`: The log level to which this detection should be logged
* `log_message`: The message that should be logged for this detection
For an example, see `VersionedTFImport`."
12037,APIAnalysisSpec,tensorflow/tensorflow/tools/compatibility/ast_edits.py,799,class,"This class defines how `AnalysisResult`s should be generated.
It specifies how to map imports and symbols to `AnalysisResult`s.
This class must provide the following fields:
* `symbols_to_detect`: maps function names to `AnalysisResult`s
* `imports_to_detect`: maps imports represented as (full module name, alias)
tuples to `AnalysisResult`s
notifications)
For an example, see `TFAPIImportAnalysisSpec`."
12038,PastaAnalyzeVisitor,tensorflow/tensorflow/tools/compatibility/ast_edits.py,815,class,"AST Visitor that looks for specific API usage without editing anything.
This is used before any rewriting is done to detect if any symbols are used
that require changing imports or disabling rewriting altogether."
12039,results,tensorflow/tensorflow/tools/compatibility/ast_edits.py,828,method,
12040,add_result,tensorflow/tensorflow/tools/compatibility/ast_edits.py,831,method,
12041,visit_Attribute,tensorflow/tensorflow/tools/compatibility/ast_edits.py,834,method,"Handle bare Attributes i.e. [tf.foo, tf.bar]."
12042,visit_Import,tensorflow/tensorflow/tools/compatibility/ast_edits.py,847,method,"Handle visiting an import node in the AST.
Args:
node: Current Node"
12043,visit_ImportFrom,tensorflow/tensorflow/tools/compatibility/ast_edits.py,866,method,"Handle visiting an import-from node in the AST.
Args:
node: Current Node"
12044,ASTCodeUpgrader,tensorflow/tensorflow/tools/compatibility/ast_edits.py,893,class,Handles upgrading a set of Python files using a given API change spec.
12045,process_file,tensorflow/tensorflow/tools/compatibility/ast_edits.py,902,method,"Process the given python file for incompatible changes.
Args:
in_filename: filename to parse
out_filename: output file to write to
no_change_to_outfile_on_error: not modify the output file on errors
Returns:
A tuple representing number of files processed, log of actions, errors"
12046,format_log,tensorflow/tensorflow/tools/compatibility/ast_edits.py,930,method,
12047,update_string_pasta,tensorflow/tensorflow/tools/compatibility/ast_edits.py,937,method,Updates a file using pasta.
12048,process_opened_file,tensorflow/tensorflow/tools/compatibility/ast_edits.py,968,method,"Process the given python file for incompatible changes.
This function is split out to facilitate StringIO testing from
tf_upgrade_test.py.
Args:
in_filename: filename to parse
in_file: opened file (or StringIO)
out_filename: output file to write to
out_file: opened file (or StringIO)
Returns:
A tuple representing number of files processed, log of actions, errors"
12049,process_tree,tensorflow/tensorflow/tools/compatibility/ast_edits.py,993,method,"Processes upgrades on an entire tree of python files in place.
Note that only Python files. If you have custom code in other languages,
you will need to manually upgrade those.
Args:
root_directory: Directory to walk and process.
output_root_directory: Directory to use as base.
copy_other_files: Copy files that are not touched by this converter.
Returns:
A tuple of files processed, the report string for all files, and a dict
mapping filenames to errors encountered in that file."
12050,process_tree_inplace,tensorflow/tensorflow/tools/compatibility/ast_edits.py,1087,method,Process a directory of python files in place.
12051,ModuleDeprecationSpec,tensorflow/tensorflow/tools/compatibility/ast_edits_test.py,57,class,A specification which deprecates 'a.b'.
12052,RenameKeywordSpec,tensorflow/tensorflow/tools/compatibility/ast_edits_test.py,65,class,"A specification where kw2 gets renamed to kw3.
The new API is
def f(a, b, kw1, kw3): ..."
12053,update_renames,tensorflow/tensorflow/tools/compatibility/ast_edits_test.py,78,method,
12054,ReorderKeywordSpec,tensorflow/tensorflow/tools/compatibility/ast_edits_test.py,82,class,"A specification where kw2 gets moved in front of kw1.
The new API is
def f(a, b, kw2, kw1): ..."
12055,update_reorders,tensorflow/tensorflow/tools/compatibility/ast_edits_test.py,95,method,
12056,ReorderAndRenameKeywordSpec,tensorflow/tensorflow/tools/compatibility/ast_edits_test.py,100,class,"A specification where kw2 gets moved in front of kw1 and is changed to kw3.
The new API is
def f(a, b, kw3, kw1): ..."
12057,RemoveDeprecatedAliasKeyword,tensorflow/tensorflow/tools/compatibility/ast_edits_test.py,116,class,"A specification where kw1_alias is removed in g.
The new API is
def g(a, b, kw1, c): ...
def g2(a, b, kw1, c, d): ..."
12058,RemoveDeprecatedAliasAndReorderRest,tensorflow/tensorflow/tools/compatibility/ast_edits_test.py,132,class,"A specification where kw1_alias is removed in g.
The new API is
def g(a, b, c, kw1): ...
def g2(a, b, c, d, kw1): ..."
12059,RemoveMultipleKeywordArguments,tensorflow/tensorflow/tools/compatibility/ast_edits_test.py,149,class,"A specification where both keyword aliases are removed from h.
The new API is
def h(a, kw1, kw2): ..."
12060,RenameImports,tensorflow/tensorflow/tools/compatibility/ast_edits_test.py,166,class,Specification for renaming imports.
12061,is_python,tensorflow/tensorflow/tools/compatibility/ipynb.py,33,function,Checks if the cell consists of Python code.
12062,process_file,tensorflow/tensorflow/tools/compatibility/ipynb.py,41,function,The function where we inject the support for ipynb upgrade.
12063,skip_magic,tensorflow/tensorflow/tools/compatibility/ipynb.py,71,function,"Checks if the cell has magic, that is not Python-based.
Args:
code_line: A line of Python code
magic_list: A list of jupyter ""magic"" exceptions
Returns:
If the line jupyter ""magic"" line, not Python line
>>> skip_magic('!ls -laF', ['%', '!', '?'])
True"
12064,check_line_split,tensorflow/tensorflow/tools/compatibility/ipynb.py,92,function,"Checks if a line was split with `\`.
Args:
code_line: A line of Python code
Returns:
If the line was split with `\`
>>> skip_magic(""!gcloud ml-engine models create ${MODEL} \\\n"")
True"
12065,TFAPIChangeSpec,tensorflow/tensorflow/tools/compatibility/tf_upgrade.py,29,class,List of maps that describe what changed in the API.
12066,UnaliasedTFImport,tensorflow/tensorflow/tools/compatibility/tf_upgrade_v2.py,39,class,
12067,VersionedTFImport,tensorflow/tensorflow/tools/compatibility/tf_upgrade_v2.py,48,class,
12068,TFAPIImportAnalysisSpec,tensorflow/tensorflow/tools/compatibility/tf_upgrade_v2.py,61,class,
12069,CompatV1ImportReplacer,tensorflow/tensorflow/tools/compatibility/tf_upgrade_v2.py,72,class,"AST Visitor that replaces `import tensorflow.compat.v1 as tf`.
Converts `import tensorflow.compat.v1 as tf` to `import tensorflow as tf`"
12070,visit_Import,tensorflow/tensorflow/tools/compatibility/tf_upgrade_v2.py,78,method,"Handle visiting an import node in the AST.
Args:
node: Current Node"
12071,TFAPIChangeSpec,tensorflow/tensorflow/tools/compatibility/tf_upgrade_v2.py,92,class,List of maps that describe what changed in the API.
12072,preprocess,tensorflow/tensorflow/tools/compatibility/tf_upgrade_v2.py,1644,method,
12073,clear_preprocessing,tensorflow/tensorflow/tools/compatibility/tf_upgrade_v2.py,1674,method,
12074,process_file,tensorflow/tensorflow/tools/compatibility/tf_upgrade_v2_main.py,42,function,Process a file of type `.py` or `.ipynb`.
12075,TFAPIChangeSpec,tensorflow/tensorflow/tools/compatibility/tf_upgrade_v2_safety.py,26,class,List of maps that describe what changed in the API.
12076,get_symbol_for_name,tensorflow/tensorflow/tools/compatibility/tf_upgrade_v2_test.py,42,function,
12077,get_args,tensorflow/tensorflow/tools/compatibility/tf_upgrade_v2_test.py,51,function,
12078,get_func_and_args_from_str,tensorflow/tensorflow/tools/compatibility/tf_upgrade_v2_test.py,60,function,"Parse call string to get function and argument names.
Args:
call_str: Call string must be in the form:
`tf.foo(arg1=val1, arg2=val2, ...)`.
Returns:
(function_name, list of arg names) tuple."
12079,get_canonical_name,tensorflow/tensorflow/tools/compatibility/update/generate_v2_renames_map.py,72,function,
12080,get_all_v2_names,tensorflow/tensorflow/tools/compatibility/update/generate_v2_renames_map.py,78,function,Get a set of function/class names available in TensorFlow 2.0.
12081,collect_constant_renames,tensorflow/tensorflow/tools/compatibility/update/generate_v2_renames_map.py,97,function,"Looks for constants that need to be renamed in TF 2.0.
Returns:
Set of tuples of the form (current name, new name)."
12082,collect_function_renames,tensorflow/tensorflow/tools/compatibility/update/generate_v2_renames_map.py,128,function,"Looks for functions/classes that need to be renamed in TF 2.0.
Returns:
Set of tuples of the form (current name, new name)."
12083,get_rename_line,tensorflow/tensorflow/tools/compatibility/update/generate_v2_renames_map.py,162,function,
12084,update_renames_v2,tensorflow/tensorflow/tools/compatibility/update/generate_v2_renames_map.py,166,function,"Writes a Python dictionary mapping deprecated to canonical API names.
Args:
output_file_path: File path to write output to. Any existing contents
would be replaced."
12085,collect_function_arg_names,tensorflow/tensorflow/tools/compatibility/update/generate_v2_reorders_map.py,69,function,"Determines argument names for reordered function signatures.
Args:
function_names: Functions to collect arguments for.
Returns:
Dictionary mapping function name to its arguments."
12086,get_reorder_line,tensorflow/tensorflow/tools/compatibility/update/generate_v2_reorders_map.py,111,function,
12087,update_reorders_v2,tensorflow/tensorflow/tools/compatibility/update/generate_v2_reorders_map.py,115,function,"Writes a Python dictionary mapping function name to argument order.
Args:
output_file_path: File path to write output to. Any existing contents
would be replaced."
12088,TfDockerTagValidator,tensorflow/tensorflow/tools/dockerfiles/assembler.py,228,class,"Custom Cerberus validator for TF tag spec.
Note: Each _validate_foo function's docstring must end with a segment
describing its own validation schema, e.g. ""The rule's arguments are..."". If
you add a new validator, you can copy/paste that section."
12089,eprint,tensorflow/tensorflow/tools/dockerfiles/assembler.py,272,function,
12090,aggregate_all_slice_combinations,tensorflow/tensorflow/tools/dockerfiles/assembler.py,276,function,Figure out all of the possible slice groupings for a tag spec.
12091,build_name_from_slices,tensorflow/tensorflow/tools/dockerfiles/assembler.py,289,function,Build the tag name (cpu-devel...) from a list of slices.
12092,update_args_dict,tensorflow/tensorflow/tools/dockerfiles/assembler.py,302,function,Update a dict of arg values with more values from a list or dict.
12093,get_slice_sets_and_required_args,tensorflow/tensorflow/tools/dockerfiles/assembler.py,315,function,"Extract used-slice-sets and required CLI arguments from a spec string.
For example, {FOO}{bar}{bat} finds FOO, bar, and bat. Assuming bar and bat
are both named slice sets, FOO must be specified on the command line.
Args:
slice_sets: Dict of named slice sets
tag_spec: The tag spec string, e.g. {_FOO}{blep}
Returns:
(used_slice_sets, required_args), a tuple of lists"
12094,gather_tag_args,tensorflow/tensorflow/tools/dockerfiles/assembler.py,342,function,Build a dictionary of all the CLI and slice-specified args for a tag.
12095,gather_slice_list_items,tensorflow/tensorflow/tools/dockerfiles/assembler.py,360,function,"For a list of slices, get the flattened list of all of a certain key."
12096,find_first_slice_value,tensorflow/tensorflow/tools/dockerfiles/assembler.py,365,function,"For a list of slices, get the first value for a certain key."
12097,assemble_tags,tensorflow/tensorflow/tools/dockerfiles/assembler.py,373,function,"Gather all the tags based on our spec.
Args:
spec: Nested dict containing full Tag spec
cli_args: List of ARG=foo arguments to pass along to Docker build
enabled_releases: List of releases to parse. Empty list = all
all_partials: Dict of every partial, for reference
Returns:
Dict of tags and how to build them"
12098,merge_partials,tensorflow/tensorflow/tools/dockerfiles/assembler.py,426,function,Merge all partial contents with their header.
12099,upload_in_background,tensorflow/tensorflow/tools/dockerfiles/assembler.py,432,function,Upload a docker image (to be used by multiprocessing).
12100,mkdir_p,tensorflow/tensorflow/tools/dockerfiles/assembler.py,438,function,"Create a directory and its parents, even if it already exists."
12101,gather_existing_partials,tensorflow/tensorflow/tools/dockerfiles/assembler.py,447,function,"Find and read all available partials.
Args:
partial_path (string): read partials from this directory.
Returns:
Dict[string, string] of partial short names (like ""ubuntu/python"" or
""bazel"") to the full contents of that partial."
12102,get_base_dirs_and_prefixes,tensorflow/tensorflow/tools/docs/base_dir.py,29,function,Returns the base_dirs and code_prefixes for OSS TensorFlow api gen.
12103,do_not_generate_docs,tensorflow/tensorflow/tools/docs/doc_controls.py,24,function,"A decorator: Do not generate docs for this object.
For example the following classes:
```
class Parent(object):
def method1(self):
pass
def method2(self):
pass
class Child(Parent):
def method1(self):
pass
def method2(self):
pass
```
Produce the following api_docs:
```
/Parent.md
# method1
# method2
/Child.md
# method1
# method2
```
This decorator allows you to skip classes or methods:
```
@do_not_generate_docs
class Parent(object):
def method1(self):
pass
def method2(self):
pass
class Child(Parent):
@do_not_generate_docs
def method1(self):
pass
def method2(self):
pass
```
This will only produce the following docs:
```
/Child.md
# method2
```
Note: This is implemented by adding a hidden attribute on the object, so it
cannot be used on objects which do not allow new attributes to be added. So
this decorator must go *below* `@property`, `@classmethod`,
or `@staticmethod`:
```
class Example(object):
@property
@do_not_generate_docs
def x(self):
return self._x
```
Args:
obj: The object to hide from the generated docs.
Returns:
obj"
12104,do_not_doc_inheritable,tensorflow/tensorflow/tools/docs/doc_controls.py,105,function,"A decorator: Do not generate docs for this method.
This version of the decorator is ""inherited"" by subclasses. No docs will be
generated for the decorated method in any subclass. Even if the sub-class
overrides the method.
For example, to ensure that `method1` is **never documented** use this
decorator on the base-class:
```
class Parent(object):
@do_not_doc_inheritable
def method1(self):
pass
def method2(self):
pass
class Child(Parent):
def method1(self):
pass
def method2(self):
pass
```
This will produce the following docs:
```
/Parent.md
# method2
/Child.md
# method2
```
When generating docs for a class's arributes, the `__mro__` is searched and
the attribute will be skipped if this decorator is detected on the attribute
on any class in the `__mro__`.
Note: This is implemented by adding a hidden attribute on the object, so it
cannot be used on objects which do not allow new attributes to be added. So
this decorator must go *below* `@property`, `@classmethod`,
or `@staticmethod`:
```
class Example(object):
@property
@do_not_doc_inheritable
def x(self):
return self._x
```
Args:
obj: The class-attribute to hide from the generated docs.
Returns:
obj"
12105,for_subclass_implementers,tensorflow/tensorflow/tools/docs/doc_controls.py,168,function,"A decorator: Only generate docs for this method in the defining class.
Also group this method's docs with and `@abstractmethod` in the class's docs.
No docs will generated for this class attribute in sub-classes.
The canonical use case for this is `tf.keras.layers.Layer.call`: It's a
public method, essential for anyone implementing a subclass, but it should
never be called directly.
Works on method, or other class-attributes.
When generating docs for a class's arributes, the `__mro__` is searched and
the attribute will be skipped if this decorator is detected on the attribute
on any **parent** class in the `__mro__`.
For example:
```
class Parent(object):
@for_subclass_implementers
def method1(self):
pass
def method2(self):
pass
class Child1(Parent):
def method1(self):
pass
def method2(self):
pass
class Child2(Parent):
def method1(self):
pass
def method2(self):
pass
```
This will produce the following docs:
```
/Parent.md
# method1
# method2
/Child1.md
# method2
/Child2.md
# method2
```
Note: This is implemented by adding a hidden attribute on the object, so it
cannot be used on objects which do not allow new attributes to be added. So
this decorator must go *below* `@property`, `@classmethod`,
or `@staticmethod`:
```
class Example(object):
@property
@for_subclass_implementers
def x(self):
return self._x
```
Args:
obj: The class-attribute to hide from the generated docs.
Returns:
obj"
12106,should_skip,tensorflow/tensorflow/tools/docs/doc_controls.py,246,function,"Returns true if docs generation should be skipped for this object.
checks for the `do_not_generate_docs` or `do_not_doc_inheritable` decorators.
Args:
obj: The object to document, or skip.
Returns:
True if the object should be skipped"
12107,should_skip_class_attr,tensorflow/tensorflow/tools/docs/doc_controls.py,264,function,"Returns true if docs should be skipped for this class attribute.
Args:
cls: The class the attribute belongs to.
name: The name of the attribute.
Returns:
True if the attribute should be skipped."
12108,DocGeneratorVisitor,tensorflow/tensorflow/tools/docs/doc_generator_visitor.py,28,class,A visitor that generates docs for a python object when __call__ed.
12109,set_root_name,tensorflow/tensorflow/tools/docs/doc_generator_visitor.py,49,method,Sets the root name for subsequent __call__s.
12110,index,tensorflow/tensorflow/tools/docs/doc_generator_visitor.py,55,method,"A map from fully qualified names to objects to be documented.
The index is filled when the visitor is passed to `traverse`.
Returns:
The index filled by traversal."
12111,tree,tensorflow/tensorflow/tools/docs/doc_generator_visitor.py,66,method,"A map from fully qualified names to all its child names for traversal.
The full name to member names map is filled when the visitor is passed to
`traverse`.
Returns:
The full name to member name map filled by traversal."
12112,reverse_index,tensorflow/tensorflow/tools/docs/doc_generator_visitor.py,78,method,"A map from `id(object)` to the preferred fully qualified name.
This map only contains non-primitive objects (no numbers or strings) present
in `index` (for primitive objects, `id()` doesn't quite do the right thing).
It is computed when it, `duplicate_of`, or `duplicates` are first accessed.
Returns:
The `id(object)` to full name map."
12113,duplicate_of,tensorflow/tensorflow/tools/docs/doc_generator_visitor.py,93,method,"A map from duplicate full names to a preferred fully qualified name.
This map only contains names that are not themself a preferred name.
It is computed when it, `reverse_index`, or `duplicates` are first accessed.
Returns:
The map from duplicate name to preferred name."
12114,duplicates,tensorflow/tensorflow/tools/docs/doc_generator_visitor.py,107,method,"A map from preferred full names to a list of all names for this symbol.
This function returns a map from preferred (master) name for a symbol to a
lexicographically sorted list of all aliases for that name (incl. the master
name). Symbols without duplicate names do not appear in this map.
It is computed when it, `reverse_index`, or `duplicate_of` are first
accessed.
Returns:
The map from master name to list of all duplicate names."
12115,generate_raw_ops_doc,tensorflow/tensorflow/tools/docs/generate2.py,97,function,Generates docs for `tf.raw_ops`.
12116,TfExportAwareVisitor,tensorflow/tensorflow/tools/docs/generate2.py,134,class,"A `tf_export`, `keras_export` and `estimator_export` aware doc_visitor."
12117,build_docs,tensorflow/tensorflow/tools/docs/generate2.py,176,function,"Build api docs for tensorflow v2.
Args:
output_dir: A string path, where to put the files.
code_url_prefix: prefix for ""Defined in"" links.
search_hints: Bool. Include meta-data search hints at the top of each file."
12118,write_docs,tensorflow/tensorflow/tools/docs/generate_lib.py,40,function,"Write previously extracted docs to disk.
Write a docs page for each symbol included in the indices of parser_config to
a tree of docs at `output_dir`.
Symbols with multiple aliases will have only one page written about
them, which is referenced for all aliases.
Args:
output_dir: Directory to write documentation markdown files to. Will be
created if it doesn't exist.
parser_config: A `parser.ParserConfig` object, containing all the necessary
indices.
yaml_toc: Set to `True` to generate a ""_toc.yaml"" file.
root_title: The title name for the root level index.md.
search_hints: (bool) include meta-data search hints at the top of each
output file.
site_api_path: The output path relative to the site root. Used in the
`_toc.yaml` and `_redirects.yaml` files.
Raises:
ValueError: if `output_dir` is not an absolute path"
12119,add_dict_to_dict,tensorflow/tensorflow/tools/docs/generate_lib.py,221,function,
12120,DocControlsAwareCrawler,tensorflow/tensorflow/tools/docs/generate_lib.py,246,class,A `docs_controls` aware API-crawler.
12121,extract,tensorflow/tensorflow/tools/docs/generate_lib.py,255,function,Extract docs from tf namespace and write them to disk.
12122,build_doc_index,tensorflow/tensorflow/tools/docs/generate_lib.py,298,function,Build an index from a keyword designating a doc to _DocInfo objects.
12123,update_id_tags_inplace,tensorflow/tensorflow/tools/docs/generate_lib.py,397,function,"Set explicit ids on all second-level headings to ensure back-links work.
Args:
src_dir: The directory of md-files to convert (inplace)."
12124,replace_refs,tensorflow/tensorflow/tools/docs/generate_lib.py,421,function,"Fix @{} references in all files under `src_dir` matching `file_pattern`.
A matching directory structure, with the modified files is
written to `output_dir`.
`{""__init__.py"",""OWNERS"",""README.txt""}` are skipped.
Files not matching `file_pattern` (using `fnmatch`) are copied with no change.
Also, files in the `api_guides/python` directory get explicit ids set on all
heading-2s to ensure back-links work.
Args:
src_dir: The directory to convert files from.
output_dir: The root directory to write the resulting files to.
reference_resolver: A `parser.ReferenceResolver` to make the replacements.
file_pattern: Only replace references in files matching file_patters,
using fnmatch. Non-matching files are copied unchanged.
api_docs_relpath: Relative-path string to the api_docs, from the src_dir."
12125,DocGenerator,tensorflow/tensorflow/tools/docs/generate_lib.py,483,class,Main entry point for generating docs.
12126,add_output_dir_argument,tensorflow/tensorflow/tools/docs/generate_lib.py,512,method,
12127,add_src_dir_argument,tensorflow/tensorflow/tools/docs/generate_lib.py,520,method,
12128,add_base_dir_argument,tensorflow/tensorflow/tools/docs/generate_lib.py,528,method,
12129,parse_known_args,tensorflow/tensorflow/tools/docs/generate_lib.py,535,method,
12130,add_to_private_map,tensorflow/tensorflow/tools/docs/generate_lib.py,539,method,
12131,add_to_do_not_descend_map,tensorflow/tensorflow/tools/docs/generate_lib.py,542,method,
12132,set_private_map,tensorflow/tensorflow/tools/docs/generate_lib.py,545,method,
12133,set_do_not_descend_map,tensorflow/tensorflow/tools/docs/generate_lib.py,548,method,
12134,set_py_modules,tensorflow/tensorflow/tools/docs/generate_lib.py,551,method,
12135,py_module_names,tensorflow/tensorflow/tools/docs/generate_lib.py,554,method,
12136,make_reference_resolver,tensorflow/tensorflow/tools/docs/generate_lib.py,560,method,
12137,make_parser_config,tensorflow/tensorflow/tools/docs/generate_lib.py,564,method,
12138,run_extraction,tensorflow/tensorflow/tools/docs/generate_lib.py,576,method,
12139,build,tensorflow/tensorflow/tools/docs/generate_lib.py,580,method,"Build all the docs.
This produces two outputs
python api docs:
* generated from modules set with `set_py_modules`.
* written to '{FLAGS.output_dir}/api_docs/python/'
non-api docs:
* Everything in '{FLAGS.src_dir}' is copied to '{FLAGS.output_dir}'.
* '@{}' references in '.md' files are replaced with links.
* '.md' files under 'api_guides/python' have explicit ids set for their
second level headings.
Args:
flags:
* src_dir: Where to fetch the non-api-docs.
* base_dir: Base of the docs directory (Used to build correct
relative links).
* output_dir: Where to write the resulting docs.
Returns:
The number of errors encountered while processing."
12140,is_free_function,tensorflow/tensorflow/tools/docs/parser.py,40,function,"Check if input is a free function (and not a class- or static method).
Args:
py_object: The the object in question.
full_name: The full name of the object, like `tf.module.symbol`.
index: The {full_name:py_object} dictionary for the public API.
Returns:
True if the obeject is a stand-alone function, and not part of a class
definition."
12141,TFDocsError,tensorflow/tensorflow/tools/docs/parser.py,66,class,
12142,documentation_path,tensorflow/tensorflow/tools/docs/parser.py,101,function,"Returns the file path for the documentation for the given API symbol.
Given the fully qualified name of a library symbol, compute the path to which
to write the documentation for that symbol (relative to a base directory).
Documentation files are organized into directories that mirror the python
module/class structure.
Args:
full_name: Fully qualified name of a library symbol.
is_fragment: If `False` produce a direct markdown link (`tf.a.b.c` -->
`tf/a/b/c.md`). If `True` produce fragment link, `tf.a.b.c` -->
`tf/a/b.md#c`
Returns:
The file path to which to write the documentation for `full_name`."
12143,ReferenceResolver,tensorflow/tensorflow/tools/docs/parser.py,164,class,"Class for replacing @{...} references with Markdown links.
Attributes:
current_doc_full_name: A string (or None) indicating the name of the
document currently being processed, so errors can reference the broken
doc."
12144,add_error,tensorflow/tensorflow/tools/docs/parser.py,195,method,
12145,log_errors,tensorflow/tensorflow/tools/docs/parser.py,198,method,
12146,num_errors,tensorflow/tensorflow/tools/docs/parser.py,201,method,
12147,from_visitor,tensorflow/tensorflow/tools/docs/parser.py,205,method,"A factory function for building a ReferenceResolver from a visitor.
Args:
visitor: an instance of `DocGeneratorVisitor`
doc_index: a dictionary mapping document names to references objects with
""title"" and ""url"" fields
**kwargs: all remaining args are passed to the constructor
Returns:
an instance of `ReferenceResolver` ()"
12148,from_json_file,tensorflow/tensorflow/tools/docs/parser.py,231,method,
12149,to_json_file,tensorflow/tensorflow/tools/docs/parser.py,237,method,"Converts the RefenceResolver to json and writes it to the specified file.
Args:
filepath: The file path to write the json to."
12150,replace_references,tensorflow/tensorflow/tools/docs/parser.py,262,method,"Replace ""@{symbol}"" references with links to symbol's documentation page.
This functions finds all occurrences of ""@{symbol}"" in `string`
and replaces them with markdown links to the documentation page
for ""symbol"".
`relative_path_to_root` is the relative path from the document
that contains the ""@{symbol}"" reference to the root of the API
documentation that is linked to. If the containing page is part of
the same API docset, `relative_path_to_root` can be set to
`os.path.dirname(documentation_path(name))`, where `name` is the
python name of the object whose documentation page the reference
lives on.
Args:
string: A string in which ""@{symbol}"" references should be replaced.
relative_path_to_root: The relative path from the containing document to
the root of the API documentation that is being linked to.
Returns:
`string`, with ""@{symbol}"" references replaced by Markdown links."
12151,python_link,tensorflow/tensorflow/tools/docs/parser.py,305,method,"Resolve a ""@{python symbol}"" reference to a Markdown link.
This will pick the canonical location for duplicate symbols. The
input to this function should already be stripped of the '@' and
'{}'. This function returns a Markdown link. If `code_ref` is
true, it is assumed that this is a code reference, so the link
text will be rendered as code (using backticks).
`link_text` should refer to a library symbol, starting with 'tf.'.
Args:
link_text: The text of the Markdown link.
ref_full_name: The fully qualified name of the symbol to link to.
relative_path_to_root: The relative path from the location of the current
document to the root of the API documentation.
code_ref: If true (the default), put `link_text` in `...`.
Returns:
A markdown link to the documentation page of `ref_full_name`."
12152,py_master_name,tensorflow/tensorflow/tools/docs/parser.py,340,method,Return the master name for a Python symbol name.
12153,reference_to_url,tensorflow/tensorflow/tools/docs/parser.py,344,method,"Resolve a ""@{python symbol}"" reference to a relative path.
The input to this function should already be stripped of the '@'
and '{}', and its output is only the link, not the full Markdown.
If `ref_full_name` is the name of a class member, method, or property, the
link will point to the page of the containing class, and it will include the
method name as an anchor. For example, `tf.module.MyClass.my_method` will be
translated into a link to
`os.join.path(relative_path_to_root, 'tf/module/MyClass.md#my_method')`.
Args:
ref_full_name: The fully qualified name of the symbol to link to.
relative_path_to_root: The relative path from the location of the current
document to the root of the API documentation.
Returns:
A relative path that links from the documentation page of `from_full_name`
to the documentation page of `ref_full_name`.
Raises:
RuntimeError: If `ref_full_name` is not documented.
TFDocsError: If the @{} syntax cannot be decoded."
12154,strict_one_ref,tensorflow/tensorflow/tools/docs/parser.py,286,method,
12155,sloppy_one_ref,tensorflow/tensorflow/tools/docs/parser.py,295,method,
12156,ParserConfig,tensorflow/tensorflow/tools/docs/parser.py,1459,class,Stores all indexes required to parse the docs.
12157,py_name_to_object,tensorflow/tensorflow/tools/docs/parser.py,1496,method,Return the Python object for a Python symbol name.
12158,docs_for_object,tensorflow/tensorflow/tools/docs/parser.py,1501,function,"Return a PageInfo object describing a given object from the TF API.
This function uses _parse_md_docstring to parse the docs pertaining to
`object`.
This function resolves '@{symbol}' references in the docstrings into links to
the appropriate location. It also adds a list of alternative names for the
symbol automatically.
It assumes that the docs for each object live in a file given by
`documentation_path`, and that relative links to files within the
documentation are resolvable.
Args:
full_name: The fully qualified name of the symbol to be
documented.
py_object: The Python object to be documented. Its documentation is sourced
from `py_object`'s docstring.
parser_config: A ParserConfig object.
Returns:
Either a `_FunctionPageInfo`, `_ClassPageInfo`, or a `_ModulePageInfo`
depending on the type of the python object being documented.
Raises:
RuntimeError: If an object is encountered for which we don't know how
to make docs."
12159,generate_global_index,tensorflow/tensorflow/tools/docs/parser.py,1711,function,"Given a dict of full names to python objects, generate an index page.
The index page generated contains a list of links for all symbols in `index`
that have their own documentation page.
Args:
library_name: The name for the documented library to use in the title.
index: A dict mapping full names to python objects.
reference_resolver: An instance of ReferenceResolver.
Returns:
A string containing an index page as Markdown."
12160,build_md_page,tensorflow/tensorflow/tools/docs/pretty_docs.py,36,function,"Given a PageInfo object, return markdown for the page.
Args:
page_info: must be a `parser.FunctionPageInfo`, `parser.ClassPageInfo`, or
`parser.ModulePageInfo`
Returns:
Markdown for the page
Raises:
ValueError: if `page_info` is an instance of an unrecognized class"
12161,md_files_in_dir,tensorflow/tensorflow/tools/docs/py_guide_parser.py,29,function,"Returns a list of filename (full_path, base) pairs for guide files."
12162,PyGuideParser,tensorflow/tensorflow/tools/docs/py_guide_parser.py,38,class,"Simple parsing of a guide .md file.
Descendants can override the process_*() functions (called by process())
to either record information from the guide, or call replace_line()
to affect the return value of process()."
12163,process,tensorflow/tensorflow/tools/docs/py_guide_parser.py,49,method,Read and process the file at `full_path`.
12164,replace_line,tensorflow/tensorflow/tools/docs/py_guide_parser.py,89,method,Replace the contents of line numbered `line_number` with `line`.
12165,process_title,tensorflow/tensorflow/tools/docs/py_guide_parser.py,93,method,
12166,process_section,tensorflow/tensorflow/tools/docs/py_guide_parser.py,96,method,
12167,process_in_blockquote,tensorflow/tensorflow/tools/docs/py_guide_parser.py,99,method,
12168,process_line,tensorflow/tensorflow/tools/docs/py_guide_parser.py,102,method,
12169,recursive_import,tensorflow/tensorflow/tools/docs/tf_doctest.py,58,function,"Recursively imports all the sub-modules under a root package.
Args:
root: A python package."
12170,find_modules,tensorflow/tensorflow/tools/docs/tf_doctest.py,72,function,"Finds all the modules in the core package imported.
Returns:
A list containing all the modules in tensorflow.python."
12171,filter_on_submodules,tensorflow/tensorflow/tools/docs/tf_doctest.py,87,function,"Filters all the modules based on the modules flag.
The module flag has to be relative to the core package imported.
For example, if `module=keras.layers` then, this function will return
all the modules in the submodule.
Args:
all_modules: All the modules in the core package.
submodules: Submodules to filter from all the modules.
Returns:
All the modules in the submodule."
12172,get_module_and_inject_docstring,tensorflow/tensorflow/tools/docs/tf_doctest.py,109,function,"Replaces the docstring of the module with the changed file's content.
Args:
file_path: Path to the file
Returns:
A list containing the module changed by the file."
12173,setup_gpu,tensorflow/tensorflow/tools/docs/tf_doctest.py,132,function,"Sets up the GPU devices.
If there're more available GPUs than needed, it hides the additional ones. If
there're less, it creates logical devices. This is to make sure the tests see
a fixed number of GPUs regardless of the environment.
Args:
required_gpus: an integer. The number of GPUs required.
Raises:
ValueError: if num_gpus is larger than zero but no GPU is available."
12174,setUpModule,tensorflow/tensorflow/tools/docs/tf_doctest.py,219,function,
12175,create_examples,tensorflow/tensorflow/tools/gcs_test/python/gcs_smoke.py,39,function,Create ExampleProto's containing data.
12176,parse_branch_ref,tensorflow/tensorflow/tools/git/gen_git_source.py,39,function,"Given a filename of a .git/HEAD file return ref path.
In particular, if git is in detached head state, this will
return None. If git is in attached head, it will return
the branch reference. E.g. if on 'master', the HEAD will
contain 'ref: refs/heads/master' so 'refs/heads/master'
will be returned.
Example: parse_branch_ref("".git/HEAD"")
Args:
filename: file to treat as a git HEAD file
Returns:
None if detached head, otherwise ref subpath
Raises:
RuntimeError: if the HEAD file is unparseable."
12177,configure,tensorflow/tensorflow/tools/git/gen_git_source.py,67,function,Configure `src_base_path` to embed git hashes if available.
12178,get_git_version,tensorflow/tensorflow/tools/git/gen_git_source.py,144,function,"Get the git version from the repository.
This function runs `git describe ...` in the path given as `git_base_path`.
This will return a string of the form:
<base-tag>-<number of commits since tag>-<shortened sha hash>
For example, 'v0.10.0-1585-gbb717a6' means v0.10.0 was the last tag when
compiled. 1585 commits are after that commit tag, and we can get back to this
version by running `git checkout gbb717a6`.
Args:
git_base_path: where the .git directory is located
git_tag_override: Override the value for the git tag. This is useful for
releases where we want to build the release before the git tag is
created.
Returns:
A bytestring representing the git version"
12179,write_version_info,tensorflow/tensorflow/tools/git/gen_git_source.py,190,function,"Write a c file that defines the version functions.
Args:
filename: filename to write to.
git_version: the result of a git describe."
12180,generate,tensorflow/tensorflow/tools/git/gen_git_source.py,229,function,"Generate version_info.cc as given `destination_file`.
Args:
arglist: should be a sequence that contains
spec, head_symlink, ref_symlink, destination_file.
`destination_file` is the filename where version_info.cc will be written
`spec` is a filename where the file contains a JSON dictionary
'git' bool that is true if the source is in a git repo
'path' base path of the source code
'branch' the name of the ref specification of the current branch/tag
`head_symlink` is a filename to HEAD that is cross-referenced against
what is contained in the json branch designation.
`ref_symlink` is unused in this script but passed, because the build
system uses that file to detect when commits happen.
git_tag_override: Override the value for the git tag. This is useful for
releases where we want to build the release before the git tag is
created.
Raises:
RuntimeError: If ./configure needs to be run, RuntimeError will be raised."
12181,raw_generate,tensorflow/tensorflow/tools/git/gen_git_source.py,274,function,"Simple generator used for cmake/make build systems.
This does not create any symlinks. It requires the build system
to build unconditionally.
Args:
output_file: Output filename for the version info cc
source_dir: Base path of the source code
git_tag_override: Override the value for the git tag. This is useful for
releases where we want to build the release before the git tag is
created."
12182,TransformGraph,tensorflow/tensorflow/tools/graph_transforms/__init__.py,26,function,"Python wrapper for the Graph Transform Tool.
Gives access to all graph transforms available through the command line tool.
See documentation at https://github.com/tensorflow/tensorflow/blob/master/tensorflow/tools/graph_transforms/README.md
for full details of the options available.
Args:
input_graph_def: GraphDef object containing a model to be transformed.
inputs: List of node names for the model inputs.
outputs: List of node names for the model outputs.
transforms: List of strings containing transform names and parameters.
Returns:
New GraphDef with transforms applied."
12183,check_output_despite_error,tensorflow/tensorflow/tools/pip_package/check_load_py_test.py,28,function,"Get output of args from command line, even if there are errors.
Args:
args: a list of command line args.
Returns:
output as string."
12184,GetBuild,tensorflow/tensorflow/tools/pip_package/pip_smoke_test.py,44,function,Get the list of BUILD file all targets recursively startind at dir_base.
12185,BinaryDistribution,tensorflow/tensorflow/tools/pip_package/setup.py,132,class,
12186,has_ext_modules,tensorflow/tensorflow/tools/pip_package/setup.py,134,method,
12187,InstallCommand,tensorflow/tensorflow/tools/pip_package/setup.py,138,class,Override the dir where the headers go.
12188,finalize_options,tensorflow/tensorflow/tools/pip_package/setup.py,141,method,
12189,InstallHeaders,tensorflow/tensorflow/tools/pip_package/setup.py,149,class,"Override how headers are copied.
The install_headers that comes with setuptools copies all files to
the same directory. But we need the files to be in a specific directory
hierarchy for -I <include_dir> to work correctly."
12190,initialize_options,tensorflow/tensorflow/tools/pip_package/setup.py,165,method,
12191,finalize_options,tensorflow/tensorflow/tools/pip_package/setup.py,170,method,
12192,mkdir_and_copy_file,tensorflow/tensorflow/tools/pip_package/setup.py,174,method,
12193,run,tensorflow/tensorflow/tools/pip_package/setup.py,200,method,
12194,get_inputs,tensorflow/tensorflow/tools/pip_package/setup.py,210,method,
12195,get_outputs,tensorflow/tensorflow/tools/pip_package/setup.py,213,method,
12196,find_files,tensorflow/tensorflow/tools/pip_package/setup.py,217,function,Return all the files matching pattern below root dir.
12197,ConfigCompatChecker,tensorflow/tensorflow/tools/tensorflow_builder/compat_checker/compat_checker.py,119,class,"Class that checks configuration versions and dependency compatibilities.
`ConfigCompatChecker` checks a given set of configurations and their versions
against supported versions and dependency rules defined in `.ini` config file.
For project `TensorFlow Builder`, it functions as a sub-module for the builder
service that validates requested build configurations from a client prior to
initiating a TensorFlow build."
12198,get_all_reqs,tensorflow/tensorflow/tools/tensorflow_builder/compat_checker/compat_checker.py,352,method,"Parses all compatibility specifications listed in the `.ini` config file.
Reads and parses each and all compatibility specifications from the `.ini`
config file by sections. It then populates appropriate dicts that represent
each section (e.g. `self.required`) and returns a tuple of the populated
dicts.
Returns:
Dict of dict
{ `required`: Dict of `Required` configs and supported versions,
`optional`: Dict of `Optional` configs and supported versions,
`unsupported`: Dict of `Unsupported` configs and supported versions,
`dependency`: Dict of `Dependency` configs and supported versions }"
12199,filter_dependency,tensorflow/tensorflow/tools/tensorflow_builder/compat_checker/compat_checker.py,535,method,"Filters dependency compatibility rules defined in the `.ini` config file.
Dependency specifications are defined as the following:
`<config> <config_version> requires <dependency> <dependency_version>`
e.g.
`python 3.7 requires tensorflow 1.13`
`tensorflow range(1.0.0, 1.13.1) requires gcc range(4.8, )`
Args:
line: String that is a dependency specification defined under `Dependency`
section in the `.ini` config file.
Returns:
Dict with configuration and its dependency information.
e.g. {`cfg`: `python`, # configuration name
`cfg_spec`: `3.7`, # configuration version
`cfgd`: `tensorflow`, # dependency name
`cfgd_spec`: `4.8`} # dependency version"
12200,convert_to_list,tensorflow/tensorflow/tools/tensorflow_builder/compat_checker/compat_checker.py,565,method,"Converts a string into a list with a separator.
Args:
item: String that needs to be separated into a list by a given separator.
List item is also accepted but will take no effect.
separator: String with which the `item` will be splited.
Returns:
List that is a splited version of a given input string.
e.g. Input: `1.0, 2.0, 3.0` with `, ` separator
Output: [1.0, 2.0, 3.0]"
12201,filter_line,tensorflow/tensorflow/tools/tensorflow_builder/compat_checker/compat_checker.py,600,method,"Removes `[` or `]` from the input line.
Args:
line: String that is a compatibility specification line from the `.ini`
config file.
Returns:
String that is a compatibility specification line without `[` and `]`."
12202,in_range,tensorflow/tensorflow/tools/tensorflow_builder/compat_checker/compat_checker.py,653,method,"Checks if a version satisfies a version and/or compatibility requirement.
Args:
ver: List whose first item is a config version that needs to be checked
for support status and version compatibility.
e.g. ver = [`1.0`]
req: `_Reqs` class instance that represents a configuration version and
compatibility specifications.
Returns:
Boolean output of checking if version `ver` meets the requirement
stored in `req` (or a `_Reqs` requirements class instance)."
12203,check_compatibility,tensorflow/tensorflow/tools/tensorflow_builder/compat_checker/compat_checker.py,780,method,"Checks version and dependency compatibility for a given configuration.
`check_compatibility` immediately returns with `False` (or failure status)
if any child process or checks fail. For error and warning messages, either
print `self.(error_msg|warning_msg)` or call `_print` function.
Returns:
Boolean that is a status of the compatibility check result."
12204,get_status,tensorflow/tensorflow/tools/tensorflow_builder/compat_checker/compat_checker.py,167,method,"Get status of `_Reqs` initialization.
Returns:
Tuple
(Boolean indicating initialization status,
List of error messages, if any)"
12205,parse_single_req,tensorflow/tensorflow/tools/tensorflow_builder/compat_checker/compat_checker.py,207,method,"Parses a requirement and stores information.
`self.req` _initialized in `__init__` is called for retrieving the
requirement.
A requirement can come in two forms:
[1] String that includes `range` indicating range syntax for defining
a requirement.
e.g. `range(1.0, 2.0) include(3.0) exclude(1.5)`
[2] List that includes individual supported versions or items.
e.g. [`1.0`, `3.0`, `7.1`]
For a list type requirement, it directly stores the list to
`self.include`.
Call `get_status` for checking the status of the parsing. This function
sets `self._initialized` to `False` and immediately returns with an error
message upon encountering a failure. It sets `self._initialized` to `True`
and returns without an error message upon success."
12206,run_shell_cmd,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,135,function,"Executes shell commands and returns output.
Args:
args: String of shell commands to run.
Returns:
Tuple output (stdoutdata, stderrdata) from running the shell commands."
12207,get_platform,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,153,function,"Retrieves platform information.
Currently the script only support linux. If other platoforms such as Windows
or MacOS is detected, it throws an error and terminates.
Returns:
String that is platform type.
e.g. 'linux'"
12208,get_cpu_type,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,179,function,"Retrieves CPU (type) information.
Returns:
String that is name of the CPU.
e.g. 'GenuineIntel'"
12209,get_cpu_arch,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,195,function,"Retrieves processor architecture type (32-bit or 64-bit).
Returns:
String that is CPU architecture.
e.g. 'x86_64'"
12210,get_distrib,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,210,function,"Retrieves distribution name of the operating system.
Returns:
String that is the name of distribution.
e.g. 'Ubuntu'"
12211,get_distrib_version,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,225,function,"Retrieves distribution version of the operating system.
Returns:
String that is the distribution version.
e.g. '14.04'"
12212,get_gpu_type,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,242,function,"Retrieves GPU type.
Returns:
String that is the name of the detected NVIDIA GPU.
e.g. 'Tesla K80'
'unknown' will be returned if detected GPU type is an unknown name.
Unknown name refers to any GPU name that is not specified in this page:
https://developer.nvidia.com/cuda-gpus"
12213,get_gpu_count,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,280,function,"Retrieves total number of GPU's available in the system.
Returns:
Integer that is the total # of GPU's found."
12214,get_cuda_version_all,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,294,function,"Retrieves all additional CUDA versions available (other than default).
For retrieving default CUDA version, use `get_cuda_version` function.
stderr is silenced by default. Setting FLAGS.debug mode will not enable it.
Remove `2> /dev/null` command from `cmds_linux['cuda_ver_dflt']` to enable
stderr.
Returns:
List of all CUDA versions found (except default version).
e.g. ['10.1', '10.2']"
12215,get_cuda_version_default,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,327,function,"Retrieves default CUDA version.
Default version is the version found in `/usr/local/cuda/` installation.
stderr is silenced by default. Setting FLAGS.debug mode will not enable it.
Remove `2> /dev/null` command from `cmds_linux['cuda_ver_dflt']` to enable
stderr.
It iterates through two types of version retrieval method:
1) Using `nvcc`: If `nvcc` is not available, then it uses next method.
2) Read version file (`version.txt`) found in CUDA install directory.
Returns:
String that is the default CUDA version.
e.g. '10.1'"
12216,get_cuda_compute_capability,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,368,function,"Retrieves CUDA compute capability based on the detected GPU type.
This function uses the `cuda_compute_capability` module to retrieve the
corresponding CUDA compute capability for the given GPU type.
Args:
source_from_url: Boolean deciding whether to source compute capability
from NVIDIA website or from a local golden file.
Returns:
List of all supported CUDA compute capabilities for the given GPU type.
e.g. ['3.5', '3.7']"
12217,get_cudnn_version,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,402,function,"Retrieves the version of cuDNN library detected.
Returns:
String that is the version of cuDNN library detected.
e.g. '7.5.0'"
12218,get_gcc_version,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,426,function,"Retrieves version of GCC detected.
Returns:
String that is the version of GCC.
e.g. '7.3.0'"
12219,get_glibc_version,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,441,function,"Retrieves version of GLIBC detected.
Returns:
String that is the version of GLIBC.
e.g. '2.24'"
12220,get_libstdcpp_version,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,456,function,"Retrieves version of libstdc++ detected.
Returns:
String that is the version of libstdc++.
e.g. '3.4.25'"
12221,get_cpu_isa_version,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,472,function,"Retrieves all Instruction Set Architecture(ISA) available.
Required ISA(s): 'avx', 'avx2', 'avx512f', 'sse4', 'sse4_1'
Returns:
Tuple
(list of available ISA, list of missing ISA)"
12222,get_python_version,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,500,function,"Retrieves default Python version.
Returns:
String that is the version of default Python.
e.g. '2.7.4'"
12223,get_all_configs,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,512,function,"Runs all functions for detecting user machine configurations.
Returns:
Tuple
(List of all configurations found,
List of all missing configurations,
List of all configurations found with warnings,
Dict of all configurations)"
12224,print_all_configs,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,581,function,"Prints the status and info on all configurations in a table format.
Args:
configs: List of all configurations found.
missing: List of all configurations that are missing.
warning: List of all configurations found with warnings."
12225,save_to_file,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,624,function,"Saves all detected configuration(s) into a JSON file.
Args:
json_data: Dict of all configurations found.
filename: String that is the name of the output JSON file."
12226,manage_all_configs,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/config_detector.py,641,function,"Manages configuration detection and retrieval based on user input.
Args:
save_results: Boolean indicating whether to save the results to a file.
filename: String that is the name of the output JSON file."
12227,retrieve_from_web,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/data/cuda_compute_capability.py,55,function,"Retrieves list of all CUDA compute capability from NVIDIA webpage.
Args:
generate_csv: Boolean for generating an output file containing
the results.
Returns:
OrderedDict that is a list of all CUDA compute capability listed on the
NVIDIA page. Order goes from top to bottom of the webpage content (.html)."
12228,retrieve_from_golden,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/data/cuda_compute_capability.py,91,function,"Retrieves list of all CUDA compute capability from a golden file.
The following file is set as default:
`./golden/compute_capability_golden.csv`
Returns:
Dictionary that lists of all CUDA compute capability in the following
format:
{'<GPU name>': ['<version major>.<version minor>', ...], ...}
If there are multiple versions available for a given GPU, then it
appends all supported versions in the value list (in the key-value
pair.)"
12229,create_gpu_capa_map,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/data/cuda_compute_capability.py,118,function,"Generates a map between GPU types and corresponding compute capability.
This method is used for retrieving CUDA compute capability from the web only.
Args:
match_list: List of all CUDA compute capability detected from the webpage.
generate_csv: Boolean for creating csv file to store results.
filename: String that is the name of the csv file (without `.csv` ending).
Returns:
OrderedDict that lists in the incoming order of all CUDA compute capability
provided as `match_list`."
12230,write_csv_from_dict,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/data/cuda_compute_capability.py,172,function,"Writes out a `.csv` file from an input dictionary.
After writing out the file, it checks the new list against the golden
to make sure golden file is up-to-date.
Args:
filename: String that is the output file name.
input_dict: Dictionary that is to be written out to a `.csv` file."
12231,check_with_golden,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/data/cuda_compute_capability.py,195,function,"Checks the newly created CUDA compute capability file with the golden.
If differences are found, then it prints a list of all mismatches as
a `WARNING`.
Golden file must reside in `golden/` directory.
Args:
filename: String that is the name of the newly created file."
12232,print_dict,tensorflow/tensorflow/tools/tensorflow_builder/config_detector/data/cuda_compute_capability.py,227,function,"Prints dictionary with formatting (2 column table).
Args:
py_dict: Dictionary that is to be printed out in a table format."
12233,check_file,tensorflow/tensorflow/tools/test/check_futures_test.py,57,function,
12234,CUDADeviceProperties,tensorflow/tensorflow/tools/test/gpu_info_lib.py,51,class,
12235,gather_gpu_devices,tensorflow/tensorflow/tools/test/gpu_info_lib.py,167,function,"Gather gpu device info.
Returns:
A list of test_log_pb2.GPUInfo messages."
12236,gather_build_configuration,tensorflow/tensorflow/tools/test/run_and_gather_logs.py,58,function,
12237,MissingLogsError,tensorflow/tensorflow/tools/test/run_and_gather_logs_lib.py,37,class,
12238,get_git_commit_sha,tensorflow/tensorflow/tools/test/run_and_gather_logs_lib.py,41,function,"Get git commit SHA for this build.
Attempt to get the SHA from environment variable GIT_COMMIT, which should
be available on Jenkins build agents.
Returns:
SHA hash of the git commit used for the build, if available"
12239,process_benchmarks,tensorflow/tensorflow/tools/test/run_and_gather_logs_lib.py,92,function,
12240,run_and_gather_logs,tensorflow/tensorflow/tools/test/run_and_gather_logs_lib.py,101,function,"Run the bazel test given by test_name. Gather and return the logs.
Args:
name: Benchmark target identifier.
test_name: A unique bazel target, e.g. ""//path/to:test""
test_args: A string containing all arguments to run the target with.
benchmark_type: A string representing the BenchmarkType enum; the
benchmark type for this target.
Returns:
A tuple (test_results, mangled_test_name), where
test_results: A test_log_pb2.TestResults proto
test_adjusted_name: Unique benchmark name that consists of
benchmark name optionally followed by GPU type.
Raises:
ValueError: If the test_name is not a valid target.
subprocess.CalledProcessError: If the target itself fails.
IOError: If there are problems gathering test log output from the test.
MissingLogsError: If we couldn't find benchmark logs."
12241,gather_machine_configuration,tensorflow/tensorflow/tools/test/system_info_lib.py,44,function,Gather Machine Configuration. This is the top level fn of this library.
12242,gather_hostname,tensorflow/tensorflow/tools/test/system_info_lib.py,66,function,
12243,gather_memory_info,tensorflow/tensorflow/tools/test/system_info_lib.py,70,function,Gather memory info.
12244,gather_cpu_info,tensorflow/tensorflow/tools/test/system_info_lib.py,79,function,Gather CPU Information. Assumes all CPUs are the same.
12245,gather_available_device_info,tensorflow/tensorflow/tools/test/system_info_lib.py,126,function,"Gather list of devices available to TensorFlow.
Returns:
A list of test_log_pb2.AvailableDeviceInfo messages."
12246,gather_platform_info,tensorflow/tensorflow/tools/test/system_info_lib.py,146,function,Gather platform info.
12247,is_real_file,tensorflow/tensorflow/tools/test/upload_test_benchmarks.py,94,function,
12248,get_mtime,tensorflow/tensorflow/tools/test/upload_test_benchmarks.py,99,function,
12249,list_files_by_mtime,tensorflow/tensorflow/tools/test/upload_test_benchmarks.py,104,function,"Return a list of files in the directory, sorted in increasing ""mtime"".
Return a list of files in the given directory, sorted from older to newer file
according to their modification times. Only return actual files, skipping
directories, symbolic links, pipes, etc.
Args:
dirpath: directory pathname
Returns:
A list of file names relative to the given directory path."
12250,lock,tensorflow/tensorflow/tools/test/upload_test_benchmarks.py,125,function,
12251,unlock,tensorflow/tensorflow/tools/test/upload_test_benchmarks.py,129,function,
12252,trylock,tensorflow/tensorflow/tools/test/upload_test_benchmarks.py,133,function,
12253,upload_benchmark_data,tensorflow/tensorflow/tools/test/upload_test_benchmarks.py,141,function,"Parse benchmark data and use the client to upload it to the datastore.
Parse the given benchmark data from the serialized JSON-format used to write
the test results file. Create the different datastore Entities from that data
and upload them to the datastore in a batch using the client connection.
Args:
client: datastore client connection
data: JSON-encoded benchmark data"
12254,upload_benchmark_files,tensorflow/tensorflow/tools/test/upload_test_benchmarks.py,190,function,"Find benchmark files, process them, and upload their data to the datastore.
Locate benchmark files in the data directory, process them, and upload their
data to the datastore. After processing each file, move it to the archive
directory for safe-keeping. Each file is locked for processing, which allows
multiple uploader instances to run concurrently if needed, each one handling
different benchmark files, skipping those already locked by another.
Args:
opts: command line options object
Note: To use locking, the file is first opened, then its descriptor is used to
lock and read it. The lock is released when the file is closed. Do not open
that same file a 2nd time while the lock is already held, because when that
2nd file descriptor is closed, the lock will be released prematurely."
12255,parse_cmd_line,tensorflow/tensorflow/tools/test/upload_test_benchmarks.py,221,function,"Parse command line options.
Returns:
The parsed arguments object."
12256,ConfigError,tensorflow/third_party/gpus/check_cuda_libs.py,38,class,
12257,check_cuda_lib,tensorflow/third_party/gpus/check_cuda_libs.py,46,function,"Tests if a library exists on disk and whether its soname matches the filename.
Args:
path: the path to the library.
check_soname: whether to check the soname as well.
Raises:
ConfigError: If the library does not exist or if its soname does not match
the filename."
12258,ConfigError,tensorflow/third_party/gpus/find_cuda_config.py,72,class,
12259,find_cuda_config,tensorflow/third_party/gpus/find_cuda_config.py,567,function,Returns a dictionary of CUDA library and header file paths.
12260,Log,tensorflow/third_party/toolchains/preconfig/ubuntu16.04/gcc7_manylinux2010-nvcc-cuda10.0/windows/msvc_wrapper_for_nvcc.py,41,function,
12261,GetOptionValue,tensorflow/third_party/toolchains/preconfig/ubuntu16.04/gcc7_manylinux2010-nvcc-cuda10.0/windows/msvc_wrapper_for_nvcc.py,45,function,"Extract the list of values for option from options.
Args:
option: The option whose value to extract.
Returns:
1. A list of values, either directly following the option,
(eg., /opt val1 val2) or values collected from multiple occurrences of
the option (eg., /opt val1 /opt val2).
2. The leftover options."
12262,GetNvccOptions,tensorflow/third_party/toolchains/preconfig/ubuntu16.04/gcc7_manylinux2010-nvcc-cuda10.0/windows/msvc_wrapper_for_nvcc.py,74,function,"Collect the -nvcc_options values from argv.
Args:
argv: A list of strings, possibly the argv passed to main().
Returns:
1. The string that can be passed directly to nvcc.
2. The leftover options."
12263,InvokeNvcc,tensorflow/third_party/toolchains/preconfig/ubuntu16.04/gcc7_manylinux2010-nvcc-cuda10.0/windows/msvc_wrapper_for_nvcc.py,96,function,"Call nvcc with arguments assembled from argv.
Args:
argv: A list of strings, possibly the argv passed to main().
log: True if logging is requested.
Returns:
The return value of calling os.system('nvcc ' + args)"