New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
stylegan2 produces color splats #27
Comments
Which parameters did you use? What is the resolution of your dataset? How many Kimg have passed to produce those results? |
default parameters from Colab: I tried different resolutions from 128 (examples above) to 512. All making same results Examples above are 30k images and 60k images passed But I got same results even after 140k images past. Splats are just in different places |
above are 128^2 and with 10 images dataset |
Depending on the dataset, StyleGAN2 can sometimes (but rarely) still become unstable. What dataset are you using here? |
@xl-sr I'm using my own Pokemon dataset based on original pokemons + ~3500 fanart pokemons from deviantart + augmentations of all above (hue+mirror) so in the end its like 35k images of pokemons |
Main question is: if I'll continue to teach this network. Will it produce more sane results in future or is everything above looks like sort of bug? |
Nope, continued training will not result in anything useful, the optimization already spiraled out of control. You can try different configurations for the SG backbone, e.g G.synthesis_kwargs.architecture = 'orig' or 'resnet' (default is 'skip'). Another thing to try is lowering the learning rates. Lastly, you can still resort to the FastGAN backbone which should work fine in your case. |
@xl-sr yeah, FastGAN works just fine! I was just curious about difference in these configs outputs :) I'll try different architectures and write back here about any differences. Thank you for advice :) |
I'm getting the same color splats with every dataset I'm trying on stylegan2 but fastgan seems to be working fine. I guess I'll stick with stylegan2-ada for the ones I want trained using stylegan and only use fastgan with this implementation. |
Training projected StyleGAN2 on large datasets eg. LSUN will result in better and faster performance than StyleGAN2-ADA. On small datasets (which I suspect you tried), I did not do much testing, so you might need to do some tuning yourself eg. adapt the learning rate, add R1 regularization, etc. |
Just wanted to say I can reproduce it. Haven't got stylegan2 to run. My dataset works with all other implementations, but not PG. Training options: Output directory: /content/drive/MyDrive/projected_gan/output/00001-stylegan2-dataset-1024x1024-gpus1-batch64 Creating output directory... Num images: 1267 |
I had this issue too for a diverse dataset on 512px stylegan2 and setting the glr and dlr to .0007 fixed it for me. |
I'm trying to run stylegan2 configuration, but it produces almost random color splats. What can it be?
Same results appeared on google collar pro (p100)
and on paper space gradient (nvcr.io/nvidia/pytorch:21.10-py3 docker + quadro m4000)
The text was updated successfully, but these errors were encountered: