Last active
April 3, 2021 03:12
Wasserstein ACGAN in Keras
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
I assume that your weight initialization is based on the fact that you want your critic to not be almost completely clipped at first iteration. Why then are the weights for the generator also initialized in the same way? Leaving the default initialization should be better for it in my opinion