updated table in report

muttest.py

@@ -57,6 +57,8 @@ def compute_stats(df_gen: pd.DataFrame, df_fuz: pd.DataFrame, output_file: str,
     combined_df.columns = ['source', *combined_df.columns[1:]]
     del combined_df[combined_df.columns[1]]
 
+    combined_df = combined_df.sort_values(['source', 'file'])
+
     plt.figure(figsize=(10, 6))
     sns.set(style="whitegrid")
     sns.boxplot(data=combined_df, x="file", y="score", hue="source")
@@ -66,7 +68,7 @@ def compute_stats(df_gen: pd.DataFrame, df_fuz: pd.DataFrame, output_file: str,
     plt.savefig(output_file)
 
     plt.figure(figsize=(10, 6))
-    df_avg = combined_df.groupby(['file', 'source']).mean().reset_index()
+    df_avg = combined_df.groupby(['file', 'source']).mean().reset_index().sort_values(['source', 'file'])
     sns.set(style="whitegrid")
     sns.barplot(data=df_avg, x="file", y="score", hue="source")
     plt.yticks(range(0, 101, 10))

report/main.tex

@@ -112,20 +112,23 @@
         \centering
         \begin{tabular}{lrrp{3.5cm}r}
             \toprule
-            \textbf{File}          & \textbf{$E(\text{Fuzzer})$} & \textbf{$E(\text{Genetic})$} & \hfill \textbf{Cohen's $d$}   & \textbf{Wilcoxon $p$} \\ \midrule
-            anagram\_check         & 23.16                & 18.51                 & -0.5690 \hfill (Large)       & 0.0526                    \\
-            caesar\_cipher         & 60.18                & 62.39                 & 0.4672 \hfill (Medium)      & 0.3590                    \\
-            check\_armstrong       & 89.54                & 89.18                 & -0.1427 \hfill (Small)      & 0.6250                    \\
-            common\_divisor\_count & 71.21                & 72.26                 & 0.2596 \hfill (Medium)      & 0.5566                    \\
-            exponentiation         & 68.99                & 68.48                 & -0.0990 \hfill (Small)       & 0.7695                    \\
-            gcd                    & 50.80                & 44.68                 & -1.0306 \hfill (Very large) & 0.0665                    \\
-            longest\_substring     & 83.96                & 83.01                 & -0.1547 \hfill (Small)      & 0.8457                    \\
-            rabin\_karp            & 66.15                & 64.51                 & -0.4597 \hfill (Medium)     & 0.3081                    \\
-            railfence\_cipher      & 90.28                & 89.62                 & -0.3514 \hfill (Medium)     & 0.3750                    \\
-            zellers\_birthday      & 69.10                & 67.86                 & -0.5598 \hfill (Large)      & 0.1851                    \\ \bottomrule
+            \textbf{File} & \textbf{$E(\text{Fuzzer})$} & \textbf{$E(\text{Genetic})$} & \textbf{Cohen's $|d|$} & \textbf{Wilcoxon $p$} \\
+            \midrule
+            check\_armstrong       & 58.07 & 93.50 & 2.0757  \hfill Huge       & 0.0020  \\
+            railfence\_cipher      & 88.41 & 87.44 & 0.8844 \hfill Very large & 0.1011 \\
+            longest\_substring     & 77.41 & 76.98 & 0.0771 \hfill Small      & 0.7589 \\
+            common\_divisor\_count & 76.17 & 72.76 & 0.7471 \hfill Large      & 0.1258 \\
+            zellers\_birthday      & 68.09 & 71.75 & 1.4701  \hfill Huge       & 0.0039 \\
+            exponentiation         & 69.44 & 67.14 & 0.3342 \hfill Medium     & 0.7108 \\
+            caesar\_cipher         & 60.59 & 61.20 & 0.3549  \hfill Medium     & 0.2955 \\
+            gcd                    & 59.15 & 55.66 & 0.5016 \hfill Large      & 0.1627 \\
+            rabin\_karp            & 27.90 & 47.55 & 2.3688  \hfill Huge       & 0.0078 \\
+            anagram\_check         & 23.10 & 7.70  & $\infty$  \hfill Huge       & 0.0020  \\
+            \bottomrule
         \end{tabular}
         \caption{Statistical comparison between fuzzer and genetic algorithm test case generation in terms of mutation
-        score as reported by \textit{mut.py} over 10 runs. The table reports run means, the wilcoxon paired test p-value
-        and the Cohen's $d$ effect size for each file in the benchmark.}\label{tab:stats}
+        score as reported by \textit{mut.py} over 10 runs, sorted by genetic mutation score. The table reports run
+        means, the wilcoxon paired test p-value and the Cohen's $d$ effect size for each file in the
+        benchmark.}\label{tab:stats}
     \end{table}
 \end{document}