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.

Sign up for GitHub

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

Jump to bottom

Fine tuning guidance #99

Open

Otsutsukii opened this issue Dec 18, 2024 · 1 comment

Labels

Answered

Otsutsukii commented Dec 18, 2024

Hi, I would like to finetune the model on an additional 10k images, it consists of different objects.

Should I fine tune with dreambooth or full fine tuning?

Also if I fine tune with this repo's code how to then convert it to a huggingface model?

Collaborator

lawrence-cj commented Dec 19, 2024

10K images is too much for dreambooth and enough for full fine-tuning.
We have script to convert Sana's pth file into diffusers safetensor format:
https://github.com/NVlabs/Sana/blob/main/tools/convert_sana_to_diffusers.py

lawrence-cj added the Answered label

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment