Closed Decalogue closed 3 years ago
In our experiments, we found that it is difficult to reproduce the good results using the PyTorch version of baseline method U-GAT-IT (https://github.com/znxlwm/UGATIT-pytorch), as observed by others on Reddit as well. That's why we did not release this version.
Thank you!
Great work! Could you release the pytorch version of FQ-U-GAT-IT? Thank you so much!