rosinality/alias-free-gan-pytorch

Training Speed

nicolas-dufour opened this issue · 1 comments

nicolas-dufour commented

Hi!
The authors from the paper said that they needed to use a custom conda kernel to get acceptable performance (They say it's 20x faster than with pytorch primitives. I noticed u didn't use a custom cuda object. How is your performance compared to StyleGAN2?

Thanks!

rosinality commented

Config-T (3x3 conv) is quite similar to stylegan2, and Config-R (1x1 conv) is about 1.5x slower.

  • 👍2