Skip to content
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

Do not prepare lr scheduler as it as the right number of steps #24088

Merged
merged 7 commits into from
Jun 7, 2023
Merged

Do not prepare lr scheduler as it as the right number of steps #24088

Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
1aa6a53
Do not prepare lr scheduler as it as the right number of steps
sgugger Jun 7, 2023
dfc1b7d
Trigger CI
sgugger Jun 7, 2023
569d555
Trigger CI
sgugger Jun 7, 2023
3d049f3
Trigger CI
sgugger Jun 7, 2023
7ab67db
Add fake comment
sgugger Jun 7, 2023
0f3f8ba
Remove fake comment
sgugger Jun 7, 2023
03a093c
Trigger CI please!
sgugger Jun 7, 2023
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
5 changes: 2 additions & 3 deletions src/transformers/trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1747,9 +1747,7 @@ def _inner_training_loop(

# prepare using `accelerator` prepare
if use_accelerator_prepare:
model, self.optimizer, self.lr_scheduler = self.accelerator.prepare(
self.model, self.optimizer, self.lr_scheduler
)
model, self.optimizer = self.accelerator.prepare(self.model, self.optimizer)

if self.is_fsdp_enabled:
self.model = model
Expand Down Expand Up @@ -1996,6 +1994,7 @@ def _inner_training_loop(
optimizer_was_run = scale_before <= scale_after
else:
self.optimizer.step()
optimizer_was_run = not self.accelerator.optimizer_step_was_skipped

if optimizer_was_run:
# Delay optimizer scheduling until metrics are generated
Expand Down