Skip to content

Conversation

@yiyixuxu
Copy link
Collaborator

@yiyixuxu yiyixuxu requested a review from a-r-r-o-w August 25, 2024 11:57
Returns:
`torch.Tensor`: positional embedding with shape `(temporal_size * grid_size[0] * grid_size[1], embed_dim/2)`.
"""
if use_real is not True:
Copy link
Collaborator Author

Choose a reason for hiding this comment

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

I removed the support for use_real for now since it is not used in cogvideox (and the logic is somewhat different from what I would have expected based on our 2d rope implementation - let's add that support in the future when it is needed)

@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.

Copy link
Contributor

@a-r-r-o-w a-r-r-o-w left a comment

Choose a reason for hiding this comment

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

Thanks, looks great! I've verified that the outputs before and after this PR are the same

@yiyixuxu yiyixuxu merged commit 1ca0a75 into main Aug 25, 2024
@yiyixuxu yiyixuxu deleted the cog-3d-rope branch August 25, 2024 21:57
sayakpaul pushed a commit that referenced this pull request Dec 23, 2024
* refactor 3d rope

* repeat -> expand
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

Successfully merging this pull request may close these issues.

4 participants