Skip to content

[scheduler] fix some scheduler dtype error #2992

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.

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

Merged
merged 2 commits into from
Apr 6, 2023

Conversation

tenderness-git
Copy link
Contributor

when use nvidia gpu, the dtype of timesteps is float64 and the dtype of timesteps_interpol is float32, but they will be pushed into same torch.stack. This bug will make some optimization failed. For example, when I use oneflow to mock torch, torch.stack will push timesteps_interpol and timesteps which has different dtype, and then oneflow will clash.

@HuggingFaceDocBuilderDev
Copy link

HuggingFaceDocBuilderDev commented Apr 6, 2023

The documentation is not available anymore as the PR was closed or merged.

@patrickvonplaten
Copy link
Contributor

Test failures are unrelated - good to go for me

@patrickvonplaten patrickvonplaten merged commit e405264 into huggingface:main Apr 6, 2023
w4ffl35 pushed a commit to w4ffl35/diffusers that referenced this pull request Apr 14, 2023
dg845 pushed a commit to dg845/diffusers that referenced this pull request May 6, 2023
yoonseokjin pushed a commit to yoonseokjin/diffusers that referenced this pull request Dec 25, 2023
Co-authored-by: wangguan <[email protected]>
Co-authored-by: Patrick von Platen <[email protected]>
AmericanPresidentJimmyCarter pushed a commit to AmericanPresidentJimmyCarter/diffusers that referenced this pull request Apr 26, 2024
Co-authored-by: wangguan <[email protected]>
Co-authored-by: Patrick von Platen <[email protected]>
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.

3 participants