In the function definition helper, it is said that “cache – tuple of values (linear_cache, activation_cache) we store for computing backward propagation efficiently”. After looking into the relu_backward and sigmoid_backward functions, it seems I need to curve out the Z.
Can you help elaborate what are inside the “cache” argument? Thank you.
I think I figured it out. Thanks a lot.
It’s good to hear that you figured this out. For anyone else who has this question, note that we wrote the code that created the cache values as part of forward propagation. So if you want to know how they are structured, you can just look at that code again.