Custom data training - Githubissues

zsyzzsoft / co-mod-gan

[ICLR 2021, Spotlight] Large Scale Image Completion via Co-Modulated Generative Adversarial Networks

Other

444 stars 67 forks source link

Custom data training #42

Open vivekbharadhwajsa opened 2 years ago

vivekbharadhwajsa commented 2 years ago

Hello,

Is it possible to do a custom training on my dataset (not FFHQ)? Also, if it is possible, can you add a small tutorial section in the README,md so that it would be helpful for others?

Thanks

zsyzzsoft commented 2 years ago

Yes, it was roughly explained in the Datasets section. The basic usage is to run create_from_images.py with --train-image-dir specified.