Skip to content
New issue

Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? # to your account

[Feature] The vae of mochi and hunyuan are different, so why their DIT blocks can be trained on the same dataset and the same code #177

Open
yomi1117 opened this issue Feb 7, 2025 · 1 comment

Comments

@yomi1117
Copy link

yomi1117 commented Feb 7, 2025

Motivation

The vae of mochi and hunyuan are different, so why their DIT blocks can be trained on the same dataset and the same code

Related resources

No response

@jzhang38
Copy link
Collaborator

Their DIT blocks can be trained on the same dataset and the same code

No, they cannot be trained on the same dataset. Technically they are not trained on the same code either. You should specify which model you are training with the --model_type flag.

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

No branches or pull requests

2 participants