Update app.py

app.py

@@ -4,11 +4,12 @@ from transformers import pipeline
 # Load your Hugging Face model
 classifier = pipeline("image-classification", model="dima806/cat_breed_image_detection")
 
-# Function to predict from uploaded image
+# Function to predict top 3 cat breeds from uploaded image
 def predict(image):
-    result = classifier(image)
-    top_result = result[0]
-    return f"{top_result['label']} ({round(top_result['score'] * 100, 2)}%)"
+    results = classifier(image)
+    top3 = results[:3]
+    formatted = [f"{i+1}. {res['label']} ({round(res['score'] * 100, 2)}%)" for i, res in enumerate(top3)]
+    return "\n".join(formatted)
 
 # All cat breeds detected by the model
 cat_breeds = [
@@ -29,14 +30,15 @@ breed_list = "\n".join(cat_breeds)
 with gr.Blocks(theme="soft") as demo:
     gr.Markdown("""
     # 🐾 Cat Breed Detector
-    Upload a picture of your cat, and let AI tell you its breed!  
-    Powered by 🤗 Hugging Face Transformers and a custom fine-tuned model.
+    Upload a picture of your cat, and let AI tell you its top 3 possible breeds!  
+    Powered by 🤗 Hugging Face Transformers and a fine-tuned model.
     """)
     
     with gr.Row(equal_height=True):
         with gr.Column(scale=1):
-            image_input = gr.Image(type="pil", label="📸 Upload a Cat Image", shape=(300, 300))
-            output = gr.Textbox(label="🎯 Predicted Breed", interactive=False)
+            image_input = gr.Image(type="pil", label="📸 Upload a Cat Image")
+            predict_button = gr.Button("🔍 Detect Breed")
+            output = gr.Textbox(label="🎯 Top 3 Predicted Breeds", interactive=False)
         with gr.Column(scale=1):
             gr.Markdown("### 🐱 All Supported Cat Breeds")
             gr.Textbox(value=breed_list, label="", lines=25, interactive=False, max_lines=25, show_copy_button=True)
@@ -44,6 +46,7 @@ with gr.Blocks(theme="soft") as demo:
     gr.Markdown("---")
     gr.Markdown("Made with ❤️ by Sumon Banerjee • Model: `dima806/cat_breed_image_detection`")
 
-    image_input.change(fn=predict, inputs=image_input, outputs=output)
+    # Loading spinner while processing
+    predict_button.click(fn=predict, inputs=image_input, outputs=output)
 
 demo.launch()