Update app.py

app.py

@@ -5,8 +5,14 @@ import torchvision.models as models
 from torchvision import transforms
 from PIL import Image
 import os
+import sys
+
+print("Starting Rice Disease Classifier...")
+print(f"Python version: {sys.version}")
+print(f"Gradio version: {gr.__version__}")
 
 DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
+print(f"Using device: {DEVICE}")
 
 CLASS_NAMES = ["bacterial", "blast", "brownspot", "healthy", "tungro"]
 
@@ -46,6 +52,7 @@ TRANSFORMS = transforms.Compose([
 ])
 
 def load_model():
+    print("Loading model...")
     model = models.efficientnet_b0(weights=None)
     model.classifier = nn.Sequential(
         nn.Dropout(p=0.3),
@@ -53,37 +60,41 @@ def load_model():
     )
     
     model_path = os.getenv("MODEL_PATH", "efficientnet_b0_best.pth")
-    checkpoint = torch.load(
-        model_path,
-        map_location=DEVICE
-    )
+    print(f"Loading model from: {model_path}")
+    
+    if not os.path.exists(model_path):
+        print(f"ERROR: Model file {model_path} not found!")
+        return None
+    
+    checkpoint = torch.load(model_path, map_location=DEVICE)
     model.load_state_dict(checkpoint["model_state_dict"])
     model.to(DEVICE)
     model.eval()
+    print("Model loaded successfully!")
     return model
 
 model = load_model()
 
 def predict(image: Image.Image):
-    if image is None:
-        return "Please upload an image.", {}
-
+    if image is None or model is None:
+        return "Please upload an image or model not loaded.", {}
+    
     tensor = TRANSFORMS(image).unsqueeze(0).to(DEVICE)
-
+    
     with torch.no_grad():
         outputs = model(tensor)
         probs = torch.softmax(outputs, dim=1)[0]
         idx = probs.argmax().item()
-
+    
     predicted = CLASS_NAMES[idx]
     info = DISEASE_INFO[predicted]
     confidence = float(probs[idx])
-
+    
     all_probs = {
         CLASS_NAMES[i]: float(probs[i])
         for i in range(len(CLASS_NAMES))
     }
-
+    
     result = f"""
 ## {info['full_name']}
 | Field | Value |
@@ -95,6 +106,7 @@ def predict(image: Image.Image):
 """
     return result, all_probs
 
+# Gradio 3.50.2 compatible interface
 demo = gr.Interface(
     fn=predict,
     inputs=gr.Image(type="pil", label="Upload rice leaf image"),
@@ -107,22 +119,15 @@ demo = gr.Interface(
         "Upload a rice leaf image to detect disease using EfficientNet-B0. "
         "Detects: Bacterial Blight, Rice Blast, Brown Spot, Tungro, or Healthy."
     ),
-    theme=gr.themes.Soft(),
 )
 
 if __name__ == "__main__":
     port = int(os.getenv("PORT", 7860))
+    print(f"Starting server on port {port}...")
     
-    # For Hugging Face Spaces, use this configuration
-    try:
-        demo.launch(
-            server_name="0.0.0.0",
-            server_port=port,
-            share=False  # Set to False for HF Spaces
-        )
-    except:
-        # Fallback for newer Gradio versions
-        demo.launch(
-            server_port=port,
-            share=False
-        )
+    # Use this format for Gradio 3.50.2
+    demo.launch(
+        server_name="0.0.0.0",
+        server_port=port,
+        share=True
+    )