Update app.py

app.py

@@ -130,20 +130,20 @@ def generate_heatmap(image):
     return superimposed_img, prediction_text
 
 # Initialize models when the script loads
-print("Initializing models... This may take a moment.")
+print("Initializing models... this may take a moment.")
 initialize_models()
-print("Models initialized successfully!")
+print("Models initialized!")
 
 # Create Gradio interface
 with gr.Blocks(title="Class Activation Heatmap Visualizer") as demo:
     gr.Markdown(
         """
-        # 🔥 Class Activation Heatmap Visualizer
+        # Class Activation Heatmap Visualizer
         
-        Upload an image to see what parts of the image the neural network focuses on when making predictions.
+        Upload an image or choose one of the examples to see what parts of the image the neural network focuses on when making predictions.
         The heatmap shows which regions of the image are most important for the top predicted class.
 
-        Adapted from: https://deeplearningwithpython.io/chapters/chapter10_interpreting-what-convnets-learn/#visualizing-heatmaps-of-class-activation
+        Code adapted from: https://deeplearningwithpython.io/chapters/chapter10_interpreting-what-convnets-learn/#visualizing-heatmaps-of-class-activation
         
         **Model:** Xception trained on ImageNet (1,000 classes)
         """
@@ -157,6 +157,27 @@ with gr.Blocks(title="Class Activation Heatmap Visualizer") as demo:
                 height=400
             )
             submit_btn = gr.Button("Generate Heatmap", variant="primary", size="lg")
+
+            # Example images
+            gr.Examples(
+                examples=[
+                    ["images/elephant.jpg"],
+                    ["images/dog.jpg"],
+                    ["images/F1_car.jpg"],
+                    ["images/multiple_animals.jpg"],
+                    ["images/osprey.jpeg"]
+                ],
+                inputs=input_image,
+                label="Try an example:"
+            )
+
+            gr.Markdown(
+                """
+                ### How to interpret the heatmap:
+                - **Red/Yellow regions**: Areas the model focuses on most for its prediction
+                - **Blue/Purple regions**: Areas the model considers less important
+                """
+            )
             
         with gr.Column():
             output_image = gr.Image(
@@ -170,28 +191,6 @@ with gr.Blocks(title="Class Activation Heatmap Visualizer") as demo:
                 interactive=False
             )
     
-    gr.Markdown(
-        """
-        ### How to interpret the heatmap:
-        - **Red/Yellow regions**: Areas the model focuses on most for its prediction
-        - **Blue/Purple regions**: Areas the model considers less important
-        - The heatmap is overlaid at 40% opacity on your original image
-        """
-    )
-    
-    # Example images
-    gr.Examples(
-        examples=[
-            ["elephant.jpg"],
-            ["dog.jpg"],
-            ["F1_car.jpg"],
-            ["multiple_animals.jpg"],
-            ["osprey.jpeg"],
-        ],
-        inputs=input_image,
-        label="Try an example:"
-    )
-    
     # Connect the button to the function
     submit_btn.click(
         fn=generate_heatmap,