-
Notifications
You must be signed in to change notification settings - Fork 101
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
[optimization] support 8bit optims from bistandbytes #163
Conversation
@@ -79,6 +79,7 @@ class Args: | |||
|
|||
# Optimizer arguments | |||
optimizer: str = "adamw" |
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.
Could we instead allow specifying adamw-8bit-bnb
here instead of use_8bit
? Reason being TorchAO also supports 8-bit implementations so we could then use adamw-8bit-torchao
in the future
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.
Done.
finetrainers/args.py
Outdated
@@ -79,6 +79,7 @@ class Args: | |||
|
|||
# Optimizer arguments | |||
optimizer: str = "adamw" | |||
use_8bit_bnb_bnb: bool = False |
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.
This is incorrect, no?
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.
Good to merge after the fix
Working fine with all the changes: https://wandb.ai/sayakpaul/finetrainers-ltxv/runs/kl9swsq8. I have added a note in the README too. |
Fixes #137.
Run: https://wandb.ai/sayakpaul/finetrainers-ltxv/runs/7kk3xhkh