-
Notifications
You must be signed in to change notification settings - Fork 6.3k
[Examples] Fix type-casting issue in the ControlNet training script #2994
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
Conversation
The documentation is not available anymore as the PR was closed or merged. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
nice!
a bit confused why it didn't happen before but should be safe to merge
Maybe the training runs were not using mixed precision at all? Can happen in the following scenario. We likely have to specify the accelerate mixed_precision flag after |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Cool!
I will wait for @williamberman to check out #2994 (comment) and ensure that we're on the same page regarding this. |
@williamberman a friendly ping ^ |
We pass the mixed_precision arg from the cli args to the Accelerator object
so I don't think it's an issue of whether or not accelerate launch was used. It's ok to merge without going super in depth here |
@williamberman, hmm interesting! I tried with an earlier |
…uggingface#2994) * fix: norm group test for UNet3D. * fix: type-casting issue in controlnet training.
…uggingface#2994) * fix: norm group test for UNet3D. * fix: type-casting issue in controlnet training.
…uggingface#2994) * fix: norm group test for UNet3D. * fix: type-casting issue in controlnet training.
Related issues:
I didn't want to use autocasting here to not mess with
accelerate
.