62 lines
1.2 KiB
Python
Executable File
62 lines
1.2 KiB
Python
Executable File
#!/usr/bin/env python3
|
|
import os
|
|
import pandas as pd
|
|
import glob
|
|
from train_classifiers import perform_grid_search, load_dataset
|
|
|
|
|
|
DIR: str = os.path.dirname(os.path.realpath(__file__))
|
|
OUT_DIR: str = DIR + '/models'
|
|
|
|
RANDOM_STATES: list[int] = [
|
|
0xDEADB017,
|
|
0xDEADBEEF,
|
|
4, # chosen by a fair dice roll
|
|
0xDECAFBAD,
|
|
0x000FF1CE,
|
|
8451,
|
|
42,
|
|
2056,
|
|
25,
|
|
6 // 2,
|
|
91,
|
|
7, # Where is Cherapunji?
|
|
115,
|
|
65,
|
|
76,
|
|
85,
|
|
83,
|
|
111,
|
|
110,
|
|
101,
|
|
]
|
|
|
|
|
|
def clean_output():
|
|
filelist = glob.glob(OUT_DIR + '/models.csv')
|
|
for f in filelist:
|
|
os.remove(f)
|
|
|
|
|
|
def main():
|
|
if not os.path.exists(OUT_DIR + '/evaluation.csv'):
|
|
X, y = load_dataset()
|
|
|
|
dfs: list[pd.DataFrame] = []
|
|
|
|
for i, state in enumerate(RANDOM_STATES):
|
|
print("Iteration " + str(i + 1) + " of " + str(len(RANDOM_STATES)) + "...")
|
|
dfs.append(perform_grid_search(X, y, 5, state))
|
|
|
|
# concatenate all runs in single dataframe
|
|
df = pd.concat(dfs, ignore_index=True)
|
|
df.to_csv(OUT_DIR + '/evaluation.csv', index=False)
|
|
else:
|
|
df = pd.read_csv(OUT_DIR + '/evaluation.csv')
|
|
|
|
# TODO: statistical analysis
|
|
|
|
|
|
if __name__ == '__main__':
|
|
main()
|