Add application file

app.py

@@ -3,7 +3,7 @@ import requests
 import gradio as gr
 
 # Load the InceptionNet model
-inception_net = tf.keras.applications.MobileNetV2()
+inception_net = tf.keras.applications.MobileNetV2(weights="imagenet")
 
 # Download human-readable labels for ImageNet
 response = requests.get("https://git.io/JJkYN")
@@ -12,28 +12,29 @@ labels = response.text.split("\n")
 # Define the function to classify an image
 def classify_image(image):
     # Preprocess the user-uploaded image
-    image = image.reshape((-1, 224, 224, 3))
+    image = tf.image.resize(image, [224, 224])
     image = tf.keras.applications.mobilenet_v2.preprocess_input(image)
-    
+    image = tf.expand_dims(image, axis=0)
+
     # Make predictions using the MobileNetV2 model
     prediction = inception_net.predict(image).flatten()
-    
+
     # Get the top 3 predicted labels with their confidence scores
-    top_classes = [labels[i] for i in prediction.argsort()[-3:][::-1]]
-    top_scores = [float(prediction[i]) for i in prediction.argsort()[-3:][::-1]]
-    
+    top_indices = prediction.argsort()[-3:][::-1]
+    top_classes = [labels[i] for i in top_indices]
+    top_scores = [float(prediction[i]) for i in top_indices]
+
     return {top_classes[i]: top_scores[i] for i in range(3)}
 
 # Create the Gradio interface
 iface = gr.Interface(
     fn=classify_image,
-    inputs=gr.Image(shape=(224, 224)),
-    outputs=gr.Label(num_top_classes=3),
+    inputs=gr.inputs.Image(shape=(224, 224)),
+    outputs=gr.outputs.Label(num_top_classes=3),
     live=True,
-    capture_session=True,  # This captures the user's uploaded image
     title="Image Classification",
     description="Upload an image, and the model will classify it into the top 3 categories.",
 )
 
 # Launch the Gradio interface
-iface.launch()
+iface.launch()