Chapter 8 Fine-tuning a pretrained model, conv_base.trainable = True missing

[r1git]

At the end of Chapter 8, "Fine-tuning a pretrained model", it seems that conv_base.trainable = True is missing from the code (though it is mentionned in the book in the Partial fine-tuning note).
So it's basically the same as "Feature extraction together with data augmentation"
This is consistent with the epoch duration which are the same but should be longer if the conv_base was trainable.