Fix dtype error for StableDiffusionXL #9217
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
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. |
sayakpaul
reviewed
Aug 20, 2024
| # Add noise to the model input according to the noise magnitude at each timestep | ||
| # (this is the forward diffusion process) | ||
| noisy_model_input = noise_scheduler.add_noise(model_input, noise, timesteps) | ||
| noisy_model_input = noise_scheduler.add_noise(model_input, noise, timesteps).to(dtype=weight_dtype) |
Member
There was a problem hiding this comment.
Let's do this conditionally if the dtype of noisy_model_input doesn't match with vae.dtype?
Contributor
Author
There was a problem hiding this comment.
Thanks for your feedback, actually I tested several times, and I found the dtype error actually happens in:
model_pred = unet(
noisy_model_input,
timesteps,
prompt_embeds,
added_cond_kwargs=unet_added_conditions,
return_dict=False,
)[0]
So I think it may not affect the vae. Further, the model_pred will be model_pred.float() in the loss part.
sayakpaul
reviewed
Aug 20, 2024
Member
sayakpaul
left a comment
sayakpaul
left a comment
There was a problem hiding this comment.
Thank you! Just a single comment.
sayakpaul
added a commit
that referenced
this pull request
Dec 23, 2024
Fix dtype error Co-authored-by: 蒋硕 <[email protected]> Co-authored-by: Sayak Paul <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What does this PR do?
Fixes #6879, and #9200
I fixed the RuntimeError: expected mat1 and mat2 to have the same dtype, as well as the similar issues when processing model_pred by fixing the same dtype.
Before submitting
documentation guidelines, and
here are tips on formatting docstrings.
Who can review?
Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.