-
Notifications
You must be signed in to change notification settings - Fork 5.6k
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
[Examples] Multiple enhancements to the ControlNet training scripts #7096
Conversation
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. |
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!
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>
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.
I like the added section in the readme -
I am not sure if the rest of the change is needed here since there isn't any actual bug in #7080; they just did not resize the control_image
, and I think it complicates codes quite a bit
But I'm ok with it if you think it is better this way
Why do you think the code becomes more complicated? The change is also in line with what we do in other training scripts, i.e., run a final validation inference with the saved checkpoint. |
ok! |
What does this PR do?
This PR adds another call to the
log_validation()
function after the serialization of the trained ControlNet checkpoint. This helps to validate if the training was effective and catch issues like #7080.