Update app.py

app.py

@@ -2,7 +2,6 @@ import gradio as gr
 import tensorflow as tf
 import numpy as np
 
-# Assuming you have the model loaded correctly as 'model'
 def load_model():
     model_path = 'fruitquality-1.0.h5'
     model = tf.keras.models.load_model(model_path)
@@ -10,14 +9,14 @@ def load_model():
 
 model = load_model()
 
-# Define the prediction function
 def predict(image):
-    image = image.reshape((-1, 300, 300, 3))  # Adjust shape as needed
-    preds = model.predict(image)
-    labels = ['Fresh Peaches', 'Fresh Pomegranates', 'Fresh Strawberries', 'Rotten Peaches', 'Rotten Pomegranates', 'Rotten Strawberries']  # Update with your actual labels
-    return {labels[i]: float(preds[0][i]) for i in range(len(labels))}
+    image = image.reshape((-1, 300, 300, 3))  # Ensure image has the correct shape
+    preds = model.predict(image)[0]  # Assuming single image prediction
+    labels = ['Fresh Peaches', 'Fresh Pomegranates', 'Fresh Strawberries', 
+              'Rotten Peaches', 'Rotten Pomegranates', 'Rotten Strawberries']
+    # Format the predictions as a dictionary {'label': probability}
+    return {labels[i]: float(preds[i]) for i in range(len(labels))}
 
-# Update Gradio Interface creation to match Gradio 2.x syntax
 iface = gr.Interface(fn=predict, inputs="image", outputs="label")
 
 iface.launch()