Skip to content

when delaying optimizer creation only prepare the model #39152

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.

Sign up for GitHub

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
ArthurZucker merged 1 commit into from
Jul 3, 2025
+1 −1
Merged

when delaying optimizer creation only prepare the model #39152

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion src/transformers/trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2357,7 +2357,7 @@ def _inner_training_loop(
model = self.accelerator.prepare(self.model)
else:
if delay_optimizer_creation:
Copy link
Copy Markdown
Collaborator

@ArthurZucker ArthurZucker Jul 1, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

cc @IlyasMoutawwakil as you wanted to remove this 👀

Copy link
Copy Markdown
Member

@IlyasMoutawwakil IlyasMoutawwakil Jul 1, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

yeah this fixes it too ! I honestly don't understand delay_optimizer_creation, like delay until when and why ? 😅 might make sense to explain it somewhere in the trainer

Copy link
Copy Markdown
Member

@IlyasMoutawwakil IlyasMoutawwakil Jul 1, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

you see why I removed it, is because currently we do create the optimizer here, and we need to prepare the fsdp model as well (otherwise fsdp fails), so the two branches of the if statement become the same

self.optimizer = self.accelerator.prepare(self.optimizer)
model = self.accelerator.prepare(self.model)
else:
model, self.optimizer = self.accelerator.prepare(self.model, self.optimizer)
else:
Expand Down