Skip to content

Updating FusedAdamW epsilon to Habana default 1e-6 #22

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
regisss merged 1 commit into from
Apr 11, 2022
+2 −0
Merged

Updating FusedAdamW epsilon to Habana default 1e-6 #22

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: 2 additions & 0 deletions optimum/habana/training_args.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -66,6 +66,8 @@ class GaudiTrainingArguments(TrainingArguments):
metadata={"help": "Whether to use lazy mode for training the model."},
)

adam_epsilon: float = field(default=1e-6, metadata={"help": "Epsilon for AdamW optimizer."})

@cached_property
@torch_required
def _setup_devices(self) -> "torch.device":
Expand Down