Update app.py

app.py

@@ -68,43 +68,37 @@ resnet50.load_state_dict(
 
 def predict(img):
     """Transforms and performs a prediction on img and returns prediction and time taken."""
+    start_time = timer()
+    
     try:
-        # Start the timer
-        start_time = timer()
-        
-        # Ensure img is valid
-        if img is None:
-            return {"Error": "No image provided"}, 0.0
-        
-        # Transform the target image and add a batch dimension
-        img_tensor = resnet50_transforms(img).unsqueeze(0)
-        
-        # Put model into evaluation mode and turn on inference mode
+        img = resnet50_transforms(img).unsqueeze(0)
         resnet50.eval()
+        
         with torch.inference_mode():
-            # Pass the transformed image through the model
-            pred_probs = torch.softmax(resnet50(img_tensor), dim=1)
+            pred_probs = torch.softmax(resnet50(img), dim=1)
             
             # Calculate entropy for OOD detection
             entropy = -torch.sum(pred_probs * torch.log(pred_probs + 1e-8)).item()
             max_prob = torch.max(pred_probs).item()
             
-            # Create prediction dictionary
+            # Create base prediction dictionary
             pred_labels_and_probs = {class_names[i]: float(pred_probs[0][i]) for i in range(len(class_names))}
             
-            # OOD Detection
+            # OOD Detection - modify existing probabilities instead of adding new keys
             if (max_prob > 0.95 and entropy < 0.2) or entropy > 2.0:
-                pred_labels_and_probs["May not be retina scan"] = 0.99
+                # Boost the probability of the first class and add a marker
+                pred_labels_and_probs[class_names[0]] = 0.99  # Use existing class
+                # You could also just print a warning or log it
+                print("May not be retina scan")
         
-        # Calculate prediction time
         pred_time = round(timer() - start_time, 5)
-        
         return pred_labels_and_probs, pred_time
         
     except Exception as e:
-        # Return error information in expected format
-        error_dict = {"Processing Error": 1.0}
-        return error_dict, 0.0
+        # Return dictionary with same structure as normal case
+        pred_labels_and_probs = {class_names[i]: 0.0 for i in range(len(class_names))}
+        pred_labels_and_probs[class_names[0]] = 1.0  # Show error in first class
+        return pred_labels_and_probs, 0.0
 
 ### 4. Gradio app ###