debug: print the prediction

app.py

@@ -10,15 +10,17 @@ from fastai.vision.all import load_learner, PILImage, PILMask
 MODEL_PATH = Path('.') / 'models'
 TEST_IMAGES_PATH = Path('.') / 'test'
 
+
 def preprocess_mask(file_name):
     """Ensures masks are in grayscale format and removes transparency."""
-    mask_path = Path('/kaggle/input/car-segmentation/car-segmentation/masks') / file_name.name
+    mask_path = Path(
+        '/kaggle/input/car-segmentation/car-segmentation/masks') / file_name.name
     mask = Image.open(mask_path)
 
     # Convert palette-based images to RGBA first to ensure proper color interpretation
     if mask.mode == 'P':
-        mask = mask.convert('RGBA')  
-    
+        mask = mask.convert('RGBA')
+
     # Convert any non-RGBA images to RGBA
     if mask.mode != 'RGBA':
         mask = mask.convert('RGBA')
@@ -27,23 +29,30 @@ def preprocess_mask(file_name):
 
     # Replace fully transparent pixels with black (or another valid label)
     new_mask_data = [
-        (r, g, b, 255) if a > 0 else (0, 0, 0, 255)  # Ensure full opacity in new mask
+        # Ensure full opacity in new mask
+        (r, g, b, 255) if a > 0 else (0, 0, 0, 255)
         for r, g, b, a in mask_data
     ]
-    
+
     mask.putdata(new_mask_data)
 
     # Convert to grayscale after handling transparency
     return PILMask.create(mask.convert('L'))
 
+
 LEARNER = load_learner(MODEL_PATH / 'car-segmentation_v1.pkl')
 
 
 def segment_image(image):
     image = PILImage.create(image)
     prediction, _, _ = LEARNER.predict(image)
+
+    print("Prediction shape:", prediction.shape)
+    print("Unique values:", numpy.unique(prediction))
+
     return numpy.array(prediction, dtype=numpy.uint8)
 
+
 demo = gradio.Interface(
     segment_image,
     inputs=gradio.Image(type='pil'),