Skip to content

Scale loss before backward #35207

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 3 commits into from
Dec 23, 2024
Merged

Scale loss before backward #35207

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
3579d4d
Scale loss before backward
qgallouedec Dec 11, 2024
86998a7
Merge branch 'main' into fix_grad_norm_no_num_item_in_batch
qgallouedec Dec 11, 2024
e12c3b2
Merge branch 'main' into fix_grad_norm_no_num_item_in_batch
qgallouedec Dec 11, 2024
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
6 changes: 4 additions & 2 deletions src/transformers/trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3697,10 +3697,12 @@ def training_step(
with amp.scale_loss(loss, self.optimizer) as scaled_loss:
scaled_loss.backward()
else:
self.accelerator.backward(loss, **kwargs)
# Finally we need to normalize the loss for reporting
if num_items_in_batch is None:
return loss.detach() / self.args.gradient_accumulation_steps
loss /= self.args.gradient_accumulation_steps

self.accelerator.backward(loss, **kwargs)

return loss.detach()

def compute_loss(self, model, inputs, return_outputs=False, num_items_in_batch=None):
Expand Down