Skip to content
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

DMD: Inference After Training Completes #157

Open
kgmills opened this issue Nov 29, 2024 · 0 comments
Open

DMD: Inference After Training Completes #157

kgmills opened this issue Nov 29, 2024 · 0 comments

Comments

@kgmills
Copy link

kgmills commented Nov 29, 2024

Hello,

Thanks for providing the step distillation code PixArt-alpha using DMD. When I run the code, it saves FSDP checkpoints periodically, and then one more FSDP checkpoint when training is done, in ".bin" format.

Say I want to take any of these checkpoint files and use them for inference using the existing diffusers pipeline ("PixArt-alpha/PixArt-Alpha-DMD-XL-2-512x512", just replacing the provided DiT weights with the ones I trained), how do I accomplish this correctly? I've tried loading the state_dict from the .bin file and even using 'tools/convert_pixart_to_diffusers.py', but to no avail.

Any help/advice with this matter will be greatly appeciated.

Thanks.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant