Update app.py

app.py

@@ -22,6 +22,20 @@ id2label = {
     "9": "Trou de balle"
 }
 
+# Define the image slider with examples of each disease
+image_slider = [
+    {"image": "aleurocanthus_spiniferus.jpg", "label": "Aleurocanthus spiniferus"},
+    {"image": "chancre_citrique.jpg", "label": "Chancre citrique"},
+    {"image": "cochenille_blanche.jpg", "label": "Cochenille blanche"},
+    {"image": "deperissement_des_agrumes.jpg", "label": "Dépérissement des agrumes"},
+    {"image": "feuille_saine.jpg", "label": "Feuille saine"},
+    {"image": "jaunissement_des_feuilles.jpg", "label": "Jaunissement des feuilles"},
+    {"image": "maladie_de_loidium.jpg", "label": "Maladie de l'oïdium"},
+    {"image": "maladie_du_dragon_jaune.jpg", "label": "Maladie du dragon jaune"},
+    {"image": "mineuse_des_agrumes.jpg", "label": "Mineuse des agrumes"},
+    {"image": "trou_de_balle.jpg", "label": "Trou de balle"}
+]
+
 def predict(image):
     # Preprocess the image
     inputs = feature_extractor(images=image, return_tensors="pt")
@@ -40,8 +54,9 @@ def predict(image):
     # Return the predicted label and confidence score
     return predicted_label, f"Confidence: {confidence_score:.2f}"
 
-gr.Markdown("""
-
+# Create the Gradio interface
+with gr.Blocks() as demo:
+    gr.Markdown("""
         <h1>Citrus Leaf Disease Classification</h1>
         <p>Upload an image of a citrus leaf to classify its disease.</p>
         <p>Supported diseases:</p>
@@ -57,11 +72,13 @@ gr.Markdown("""
             <li>Mineuse des agrumes</li>
             <li>Trou de balle</li>
         </ul>
-
+        <p>Example images:</p>
     """)
-# Create the Gradio interface
-image = gr.Image(type="pil")
-label = gr.Textbox(label="Prediction")
-
-gr.Interface(fn=predict, inputs=image, outputs=label, title="Citrus Disease Classification",
-             description="Upload an image of a citrus leaf to classify its disease.").launch(share=True)
+    image_slider_component = gr.Gallery(image_slider, label="Example Images")
+    image_input = gr.Image(type="pil", label="Upload Image")
+    label_output = gr.Textbox(label="Prediction")
+    btn = gr.Button("Classify")
+    
+    btn.click(fn=predict, inputs=image_input, outputs=label_output)
+    
+demo.launch()