Update app.py

app.py

@@ -6,6 +6,8 @@ from torchvision import models, transforms
 from huggingface_hub import hf_hub_download
 from PIL import Image
 import requests
+import os
+import base64
 from io import BytesIO
 
 # Define the number of classes
@@ -44,46 +46,26 @@ transform = transforms.Compose([
     transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225]),
 ])
 
-def predict(input_data):
+def predict(image):
     try:
-        print(f"Input data received: {input_data}, Type: {type(input_data)}")
-        
-        # Check if the input is a URL or image
-        if isinstance(input_data, str):  # If it's a string, assume it's a URL
-            try:
-                response = requests.get(input_data)
-                response.raise_for_status()  # Raise error for HTTP issues
-                img = Image.open(BytesIO(response.content))
-                print("Image fetched successfully from URL.")
-            except Exception as e:
-                print(f"Error fetching image from URL: {e}")
-                return json.dumps({"error": f"Failed to fetch image from URL: {e}"})
-        else:  # If it's not a string, assume it's an image file
-            img = input_data
+        print(f"Received image input: {image}")
 
-        # Validate the image
-        if not isinstance(img, Image.Image):
-            print("Invalid image format received.")
+        # Check if the input is a PIL Image type (Gradio automatically provides a PIL image)
+        if not isinstance(image, Image.Image):
             return json.dumps({"error": "Invalid image format received. Please provide a valid image."})
-        else:
-            print(f"Image successfully loaded: {img}")
 
         # Apply transformations to the image
-        img = transform(img).unsqueeze(0)
-        print(f"Transformed image tensor shape: {img.shape}")
-
-        # Ensure model is loaded
-        if model is None:
-            return json.dumps({"error": "Model not loaded. Ensure the model file is available and correctly loaded."})
+        image = transform(image).unsqueeze(0)
+        print(f"Transformed image tensor: {image.shape}")
 
         # Move the image to the correct device
-        img = img.to(torch.device("cuda" if torch.cuda.is_available() else "cpu"))
+        image = image.to(torch.device("cuda" if torch.cuda.is_available() else "cpu"))
 
         # Make predictions
         with torch.no_grad():
-            outputs = model(img)
+            outputs = model(image)
             predicted_class = torch.argmax(outputs, dim=1).item()
-            print(f"Model prediction outputs: {outputs}, Predicted class: {predicted_class}")
+            print(f"Prediction output: {outputs}, Predicted class: {predicted_class}")
 
         # Return the result based on the predicted class
         if predicted_class == 0:
@@ -97,17 +79,15 @@ def predict(input_data):
         print(f"Error processing image: {e}")
         return json.dumps({"error": f"Error processing image: {e}"})
 
-
-# Create the Gradio interface with both local file upload and URL input
+# Create the Gradio interface
 iface = gr.Interface(
     fn=predict,
-    inputs=[gr.Image(type="pil", label="Upload an image or provide a local path"), 
-            gr.Textbox(label="Or enter image URL (if available)", placeholder="Enter a URL for the image")],
+    inputs=gr.Image(type="pil", label="Upload an image or provide a URL or local path"),
     outputs=gr.Textbox(label="Prediction Result"),
     live=True,
     title="Maize Anomaly Detection",
     description="Upload an image of maize to detect anomalies like disease or pest infestation. You can provide local paths, URLs, or base64-encoded images."
 )
 
-# Launch the Gradio interface
+# Launch the Gradio interface 
 iface.launch(share=True, show_error=True)