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

Allow from_transformer in SD3ControlNetModel #8749

Merged
merged 4 commits into from
Jul 1, 2024

Conversation

haofanwang
Copy link
Contributor

Solve #8723.

Currently, in SD3ControlNetModel.from_transformer(), the num_layers is None by default. If we don't set num_layers manually, it will lead to size mismatch error at the last block. Some explanations can be found here.

Copy link
Collaborator

@yiyixuxu yiyixuxu left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

thanks!
I think only the transformer_blocks need strict=False though - other layers should be exact matches so that's remove the strict=False where it is not needed

haofanwang and others added 3 commits July 1, 2024 20:01
Co-authored-by: YiYi Xu <yixu310@gmail.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
Co-authored-by: YiYi Xu <yixu310@gmail.com>
@haofanwang
Copy link
Contributor Author

@yiyixuxu Done.

@HuggingFaceDocBuilderDev

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

@yiyixuxu yiyixuxu merged commit 0bae6e4 into huggingface:main Jul 1, 2024
15 checks passed
sayakpaul pushed a commit that referenced this pull request Dec 23, 2024
* Update controlnet_sd3.py

---------

Co-authored-by: YiYi Xu <yixu310@gmail.com>
# for free to join this conversation on GitHub. Already have an account? # to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants