[115/196 (58.7%)] Ablation 16/48 (color_reverse)
Zeroed out 16 filters: [44, 42, 11, 4, 10, 46, 2, 29, 40, 6, 35, 3, 7, 13, 16, 31]
Making predictions on 1280 images...
/content/biomimetic-training/src/eval/evaluate_bias.py:244: PerformanceWarning: DataFrame is highly fragmented. This is usually the result of calling `frame.insert` many times, which has poor performance. Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`
res_pd[col_name] = [
✓ Predictions complete. Shape: (1280, 1000)
Counting shape-texture statistics...
✓ Statistics complete. Processed: 1200, Skipped: 80
Sample counts for 'knife': shape=0, other=70, texture=5
✓ Saved to column: color_reverse_biomimetic_alexnet_ablation_16
