Rthur2003 commited on
Commit
edc551d
·
1 Parent(s): 4b87da0

feat: update excluded columns in feature statistics computation to prevent data leakage

Browse files
app/training/compute_feature_stats.py CHANGED
@@ -26,7 +26,7 @@ def compute_stats(csv_path: str | Path, output_path: str | Path) -> None:
26
  rows = list(reader)
27
  fieldnames = reader.fieldnames or []
28
 
29
- excluded = {"file_path", "label_int"}
30
  feature_cols = [c for c in fieldnames if c not in excluded]
31
 
32
  stats: dict[str, dict[str, float]] = {}
 
26
  rows = list(reader)
27
  fieldnames = reader.fieldnames or []
28
 
29
+ excluded = {"file_path", "label_int", "duration_sec", "sample_rate"}
30
  feature_cols = [c for c in fieldnames if c not in excluded]
31
 
32
  stats: dict[str, dict[str, float]] = {}