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

Jump to bottom

[Trainer] Fix sharding overlap bug #8333

Merged
merged 1 commit into from
Apr 28, 2024
Merged

[Trainer] Fix sharding overlap bug #8333

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
5 changes: 5 additions & 0 deletions paddlenlp/trainer/training_args.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1026,6 +1026,11 @@
enable_dp_comm_overlap and enable_sharding_comm_overlap
), "dp_comm_overlap and sharding_comm_overlap cannot be enabled at the same time"

if enable_sharding_comm_overlap and not self.amp_master_grad:
raise ValueError(

Check warning on line 1030 in paddlenlp/trainer/training_args.py

View check run for this annotation

Codecov / codecov/patch

paddlenlp/trainer/training_args.py#L1029-L1030 

Added lines #L1029 - L1030 were not covered by tests
"If `enable_sharding_comm_overlap` in pipeline_parallel_configs, `amp_master_grad` must be True."
)

dygraph_pp_configs = {
"delay_scale_loss": True if "enable_delay_scale_loss" in pipeline_parallel_config else False,
"dp_comm_overlap": enable_dp_comm_overlap,
Expand Down
Loading