For alpaca model, I passed inputs in data_augmentation and preprocess_input, and I set input_shape= input_shape in base_model and update shape=input_shape in inputs. For average pooling I also passed x = GlobalAveragePooling2D()(x) as well as for x = Dropout()(x), and I used linear function for prediction.

after running the code I recive
Test failed
Expected value

[‘Sequential’, (None, 160, 160, 3), 0]

does not match the input value:

[‘TensorFlowOpLayer’, [(None, 160, 160, 3)], 0]

I am not sure what is wrong? Please help me look at it.

Hi Yiming,

You are close to solve the exercise! Just by chance, did you forgot to add the 0.2 to the Dropout layer?
Besides could you share what did you write for prediction_layer = ?



thank you so much for reply. I just figured it out, I used inputs data twice on data_augmentation and preprocessor.