Update app.py

app.py

@@ -82,10 +82,8 @@ def run_pipeline(image: np.ndarray, subset: str) -> tuple:
     arr_model = (arr_model - mean) / std
     img_tensor = torch.from_numpy(arr_model).permute(2, 0, 1).float()
 
-    print(f"DEBUG display_arr: min={display_arr.min()} max={display_arr.max()}")
     # ── Step 2: Segment ───────────────────────────────────────────────────────
     mask = predict_mask(model, img_tensor, IMAGE_SIZE)
-    print(f"DEBUG mask: unique={np.unique(mask).tolist()} defect_px={( mask==1).sum()}")
 
     # ── Step 3: Extract features ──────────────────────────────────────────────
     features = extract_features(mask, IMAGE_SIZE)
@@ -150,7 +148,6 @@ def run_pipeline(image: np.ndarray, subset: str) -> tuple:
     # Ensure output is exactly what Gradio expects
     overlay = overlay.astype(np.uint8)
     assert overlay.ndim == 3 and overlay.shape[2] == 3
-    print(f"DEBUG overlay: shape={overlay.shape} dtype={overlay.dtype} min={overlay.min()} max={overlay.max()}")
     return overlay, features_text, diagnosis_text, risk_label
 # ── Gradio UI ─────────────────────────────────────────────────────────────────