Update app.py

app.py

@@ -26,29 +26,22 @@ def predict_image(image):
     # Make a prediction
     prediction = model.predict(image)
 
-    # Get the probability of being 'Good'
-    probability_good = prediction[0][0]  # Assuming it's a binary classification
-    probability_good = str(probability_good)
-    # Define the prediction result
-    result = {
-        "prediction": probability_good
-    }
-
-    return result
-    #predictions = tf.nn.softmax(outputs.logits, axis=-1)
-    #predicted_class = np.argmax(predictions)
-    #if predicted_class == 0:
-        #predict_label = "Clean"
-    #else:
-        #predict_label = "Carries"
-    #confidence = float(np.max(predictions))
-    #prediction_dict = {"prediction": predict_label, "confidence": confidence}
-    #return prediction_dict
-   
+    # Get the probability of being 'Clean' or 'Carries'
+    probabilities = tf.nn.softmax(prediction, axis=-1)
+    predicted_class_index = np.argmax(probabilities)
+    if predicted_class_index == 0:
+        predicted_label = "Clean"
+    else:
+        predicted_label = "Carries"
+    confidence = float(np.max(probabilities))
+    confidence = str(confidence)
+    # Print the predicted label and evaluation
+    # Return the prediction result as a dictionary
+    return {"Predicted Label:", predicted_label, f"Evaluate the topic according to {predicted_label} is: {confidence}"}
 
 
 # Create the interface
-input_interface = gr.Image(type = "pil")
+input_interface = gr.Image(type="pil")
 output_interface = "json"
 
 iface = gr.Interface(