Update app.py

app.py

@@ -207,14 +207,14 @@ def evaluate_model(labels, preds):
 def process_zip(zip_file):
     extracted_dir = extract_zip(zip_file.name)
 
-    # Initialize model pipelines separately to avoid reuse issues
-    model_pipelines = [pipe0, pipe1, pipe2]
-
     # Run classification for each model
     results = {}
-    for idx, pipe in enumerate(model_pipelines):
+    for idx in range(len(models)):
         print(f"Processing with model {idx}")
-        
+
+        # Create a new pipeline for each model within the loop
+        pipe = pipeline("image-classification", f"{models[idx]}")  
+
         # Classify images with the correct pipeline per model
         labels, preds, images = classify_images(extracted_dir, pipe, idx)
         accuracy, roc_score, report, cm_fig, roc_fig = evaluate_model(labels, preds)
@@ -231,12 +231,13 @@ def process_zip(zip_file):
     # Return results for all three models
     return (results['Model_0_accuracy'], results['Model_0_roc_score'], results['Model_0_report'], 
             results['Model_0_cm_fig'], results['Model_0_roc_fig'],
-            results['Model_1_accuracy'], results['Model_1_roc_score'], results['Model_1_report'], 
+           results['Model_1_accuracy'], results['Model_1_roc_score'], results['Model_1_report'], 
             results['Model_1_cm_fig'], results['Model_1_roc_fig'],
-            results['Model_2_accuracy'], results['Model_2_roc_score'], results['Model_2_report'], 
+           results['Model_2_accuracy'], results['Model_2_roc_score'], results['Model_2_report'], 
             results['Model_2_cm_fig'], results['Model_2_roc_fig'])
 
 
+
 # Single image section
 def load_url(url):
     try: