Update app.py

app.py

@@ -2,85 +2,84 @@ import gradio as gr
 from transformers import CLIPModel, CLIPProcessor
 from PIL import Image
 
-# Load Model and Processor
+# Step 1: Load Fine-Tuned Model from Hugging Face Model Hub
 model_name = "quadranttechnologies/retail-content-safety-clip-finetuned"
-print("Initializing the model and processor...")
+
+print("Initializing the application...")
 
 try:
+    print("Loading the model from Hugging Face Model Hub...")
     model = CLIPModel.from_pretrained(model_name, trust_remote_code=True)
     processor = CLIPProcessor.from_pretrained(model_name)
     print("Model and processor loaded successfully.")
 except Exception as e:
-    print(f"Error loading model or processor: {e}")
-    raise RuntimeError(f"Failed to load the model: {e}")
-
+    print(f"Error loading the model or processor: {e}")
+    raise RuntimeError(f"Failed to load model: {e}")
 
-# Inference Function
+# Step 2: Define the Inference Function
 def classify_image(image):
     """
-    Classifies an image as 'safe' or 'unsafe' using the CLIP model.
+    Classify an image as 'safe' or 'unsafe' and return probabilities.
 
     Args:
         image (PIL.Image.Image): Uploaded image.
-
+    
     Returns:
-        Tuple: Predicted category and probabilities for "safe" and "unsafe".
+        dict: Classification results or an error message.
     """
     try:
         print("Starting image classification...")
 
-        # Validate image input
+        # Validate input
         if image is None:
             raise ValueError("No image provided. Please upload a valid image.")
+
+        # Validate image format
         if not hasattr(image, "convert"):
-            raise ValueError("Uploaded file is not a valid image format (JPEG, PNG, etc.).")
+            raise ValueError("Invalid image format. Please upload a valid image (JPEG, PNG, etc.).")
 
-        # Define classification categories
+        # Define categories
         categories = ["safe", "unsafe"]
-        print(f"Using categories: {categories}")
 
-        # Process image using CLIPProcessor
+        # Process the image with the processor
+        print("Processing the image...")
         inputs = processor(text=categories, images=image, return_tensors="pt", padding=True)
-        print("Image processed successfully.")
+        print(f"Processed inputs: {inputs}")
 
-        # Perform inference
+        # Run inference with the model
+        print("Running model inference...")
         outputs = model(**inputs)
-        print("Inference completed successfully.")
+        print(f"Model outputs: {outputs}")
 
-        # Calculate probabilities
-        logits_per_image = outputs.logits_per_image
-        probs = logits_per_image.softmax(dim=1)
-        print(f"Probabilities: {probs}")
+        # Extract logits and probabilities
+        logits_per_image = outputs.logits_per_image  # Image-text similarity scores
+        probs = logits_per_image.softmax(dim=1)  # Convert logits to probabilities
+        print(f"Calculated probabilities: {probs}")
 
         # Extract probabilities for each category
-        safe_prob = probs[0][0].item() * 100
-        unsafe_prob = probs[0][1].item() * 100
-        print(f"Safe: {safe_prob:.2f}%, Unsafe: {unsafe_prob:.2f}%")
+        safe_prob = probs[0][0].item() * 100  # Safe percentage
+        unsafe_prob = probs[0][1].item() * 100  # Unsafe percentage
 
-        # Determine the predicted category
-        predicted_category = "safe" if safe_prob > unsafe_prob else "unsafe"
-        print(f"Predicted Category: {predicted_category}")
-
-        # Return category and probabilities
-        return predicted_category, {"safe": f"{safe_prob:.2f}%", "unsafe": f"{unsafe_prob:.2f}%"}
+        # Return results
+        return {
+            "safe": f"{safe_prob:.2f}%",
+            "unsafe": f"{unsafe_prob:.2f}%"
+        }
 
     except Exception as e:
         print(f"Error during classification: {e}")
-        return "Error", {"safe": "N/A", "unsafe": "N/A"}
-
+        return {"Error": str(e)}
 
-# Gradio Interface
+# Step 3: Set Up Gradio Interface
 iface = gr.Interface(
     fn=classify_image,
-    inputs=gr.Image(type="pil"),  # Accept image input
-    outputs=[
-        gr.Textbox(label="Predicted Category"),  # Predicted category
-        gr.Label(label="Probabilities"),        # Probabilities as progress bars
-    ],
+    inputs=gr.Image(type="pil"),
+    outputs=gr.Label(label="Output"),  # Display probabilities as a percentage scale
     title="Content Safety Classification",
-    description="Upload an image to classify it as 'safe' or 'unsafe' with probabilities.",
+    description="Upload an image to classify it as 'safe' or 'unsafe' with corresponding probabilities.",
 )
 
+# Step 4: Launch Gradio Interface
 if __name__ == "__main__":
     print("Launching the Gradio interface...")
     iface.launch()
@@ -107,6 +106,8 @@ if __name__ == "__main__":
 
 
 
+
+