You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Nov 3, 2022. It is now read-only.
It seems like just running the WGAN-GP on MNIST dataset (without changing anything in the code) results in black samples produced by the generator after 9 epochs and the samples remain black in the upcoming epochs (the critic loss stops to decrease as well). The samples are visible before 9 epochs.
Epoch 0:
Epoch 9:
Epoch 41:
How can this problem be fixed?
Thanks
The text was updated successfully, but these errors were encountered:
It seems like just running the WGAN-GP on MNIST dataset (without changing anything in the code) results in black samples produced by the generator after 9 epochs and the samples remain black in the upcoming epochs (the critic loss stops to decrease as well). The samples are visible before 9 epochs.
Epoch 0:
Epoch 9:
Epoch 41:
How can this problem be fixed?
Thanks
The text was updated successfully, but these errors were encountered: