Update app.py

app.py

@@ -4,38 +4,38 @@ from tensorflow.keras.preprocessing.image import img_to_array
 from PIL import Image
 import numpy as np
 
-# Load your trained .h5 model
+# Load the .h5 model
 model = load_model("model.h5")
 
-# Optional: Replace this list with your actual class labels if available
-class_labels = [f"Class_{i}" for i in range(model.output_shape[1])]
+# Class index to label map
+label_map = {
+    0: "Healthy",
+    1: "Powdery",
+    2: "Rust"
+}
 
 # Prediction function
-def predict_disease(image):
+def predict(image):
     try:
-        # Resize and preprocess image
         image = image.resize((225, 225))
         img_array = img_to_array(image) / 255.0
         img_array = np.expand_dims(img_array, axis=0)
 
-        # Predict
-        prediction = model.predict(img_array)
-        predicted_class_index = int(np.argmax(prediction, axis=1)[0])
-        confidence = float(np.max(prediction))
+        predictions = model.predict(img_array)
+        predicted_index = int(np.argmax(predictions, axis=1)[0])
+        predicted_class = label_map.get(predicted_index, "Unknown")
 
-        predicted_label = class_labels[predicted_class_index]
-
-        return f"Prediction: {predicted_label} (Confidence: {confidence:.2f})"
+        return predicted_class
     except Exception as e:
         return f"Error: {str(e)}"
 
 # Gradio UI
 interface = gr.Interface(
-    fn=predict_disease,
+    fn=predict,
     inputs=gr.Image(type="pil"),
     outputs="text",
     title="Plant Disease Classifier",
-    description="Upload a plant leaf image to detect disease using a trained ML model (.h5)."
+    description="Upload a plant image to detect whether it is Healthy, Powdery, or Rust."
 )
 
 interface.launch()