Create app.py

app.py

@@ -0,0 +1,35 @@
+model = tf.keras.models.load_model('densenet.h5')
+model.load_weights("pretrained_model.h5")
+class_names = ['Cardiomegaly', 'Emphysema', 'Effusion', 'Hernia', 'Infiltration', 'Mass', 'Nodule', 'Atelectasis', 'Pneumothorax', 'Pleural_Thickening', 'Pneumonia', 'Fibrosis', 'Edema', 'Consolidation']
+def custom_decode_predictions(predictions, class_labels):
+    
+    decoded_predictions = []
+    for pred in predictions:
+        # Get indices of top predicted classes
+        top_indices = pred.argsort()[-3:][::-1]  # Change 5 to the number of top classes you want to retrieve
+        # Decode each top predicted class
+        decoded_pred = [(class_labels[i], pred[i]) for i in top_indices]
+        decoded_predictions.append(decoded_pred)
+    return decoded_predictions
+
+def classify_image(img):
+    img_array = image.img_to_array(img)
+    img_array = np.expand_dims(img_array, axis=0)
+    img_array = preprocess_input(img_array)
+
+
+    predictions1 = model.predict(img_array)
+    decoded_predictions = custom_decode_predictions(predictions1, class_names)
+    return decoded_predictions
+
+# Gradio interface
+iface = gr.Interface(
+    fn=classify_image, 
+    inputs="image", 
+    outputs="text", 
+    title="Image Classification",
+    description="Classify images using your pre-trained model."
+)
+
+# Launch the interface
+iface.launch(inline = False)