Week 2, ResNet, Exercise 2 - convolutional_block

I think everything is true but I got following error:

AssertionError: Wrong values when training=False.

That means that there must be some error in your code. Unfortunately this whole assignment is an excruciating exercise in proofreading. There are lots of details and they all need to be right. One common error is not using the correct input on the “shortcut” layers. The instructions are not very explicit about that, but the diagrams are the clearest description of what is intended to happen. Check all the parameters to your various layers versus the instructions. Let us know if this sheds any light or not!