I think this assignment could use a bit of proofing. The provided code and explanations are occasionally misleading.
For example, at the end of section 5.4, the local variables
style _targets are defined, but they are never used. Why?
At the top of section 5.4, the explanation of the purpose of the function
get_layer_outputs is unclear: “loads the VGG19 model and returns a list of the outputs for the middle layers”. The function returns a tf.keras.Model object; it does not return a list of the outputs for the middle layers.
In section 5.5.1, I think there’s a missing word: “Later in the assignment, you will need to do the proper with the generated image…”
In section 5.5.2, this explanation is misleading: " The code below sets a_S to be the tensor giving the hidden layer activation for
STYLE_LAYERS using our style image." In fact, a_S is a list of six tensors, it is not a tensor.