Create app.py

app.py

@@ -0,0 +1,31 @@
+import gradio as gr
+import numpy as np
+from keras.applications import MobileNetV2
+from keras.applications.mobilenet_v2 import preprocess_input, decode_predictions
+from keras.utils import img_to_array
+from PIL import Image
+
+# Load pretrained MobileNetV2 (ImageNet weights = transfer learning)
+model = MobileNetV2(weights="imagenet")
+
+def predict(image):
+    # Resize to 224x224 as required by MobileNetV2
+    img = image.resize((224, 224))
+    arr = img_to_array(img)
+    arr = np.expand_dims(arr, axis=0)       # shape: (1, 224, 224, 3)
+    arr = preprocess_input(arr)             # normalize for MobileNetV2
+
+    preds = model.predict(arr)
+    top5 = decode_predictions(preds, top=5)[0]  # [(id, label, prob), ...]
+
+    return {label: float(prob) for (_, label, prob) in top5}
+
+demo = gr.Interface(
+    fn=predict,
+    inputs=gr.Image(type="pil"),
+    outputs=gr.Label(num_top_classes=5),
+    title="Image Classifier (MobileNetV2 Transfer Learning)",
+    description="Upload an image and the model predicts what it is using Keras MobileNetV2 pretrained on ImageNet.",
+)
+
+demo.launch()