Skip to content

[deepspeed] check whether model is NLP one instead of counting on input type #21800

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
sgugger merged 5 commits into from
Mar 1, 2023
Merged

[deepspeed] check whether model is NLP one instead of counting on input type #21800

Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
efc7e7f
trying to figure out whether model is NLP
izapolsk Feb 10, 2023
4586fd0
drop my changes and apply easier fix
izapolsk Feb 26, 2023
ff7fa0a
trying to handle all int input types
izapolsk Feb 27, 2023
7372a83
fix logic
stas00 Feb 28, 2023
99ef9b1
Merge remote-tracking branch 'origin/main' into fix/handle-inputs
stas00 Feb 28, 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
4 changes: 2 additions & 2 deletions src/transformers/trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2562,8 +2562,8 @@ def _prepare_input(self, data: Union[torch.Tensor, Any]) -> Union[torch.Tensor,
return type(data)(self._prepare_input(v) for v in data)
elif isinstance(data, torch.Tensor):
kwargs = {"device": self.args.device}
if self.deepspeed and data.dtype != torch.int64:
# NLP models inputs are int64 and those get adjusted to the right dtype of the
if self.deepspeed and (torch.is_floating_point(data) or torch.is_complex(data)):
# NLP models inputs are int/uint and those get adjusted to the right dtype of the
# embedding. Other models such as wav2vec2's inputs are already float and thus
# may need special handling to match the dtypes of the model
kwargs.update({"dtype": self.args.hf_deepspeed_config.dtype()})
Expand Down