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.

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

Trainer: delegate default generation values to generation_config #25987

Merged
merged 3 commits into from
Sep 5, 2023
Merged
Changes from 1 commit
Commits
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: 0 additions & 5 deletions src/transformers/trainer_seq2seq.py
Original file line number Diff line number Diff line change
Expand Up @@ -261,11 +261,6 @@ def prediction_step(
if len(gen_kwargs) == 0 and hasattr(self, "_gen_kwargs"):
gen_kwargs = self._gen_kwargs.copy()

if gen_kwargs.get("max_length") is None and gen_kwargs.get("max_new_tokens") is None:
gen_kwargs["max_length"] = self.model.config.max_length
gen_kwargs["num_beams"] = (
gen_kwargs["num_beams"] if gen_kwargs.get("num_beams") is not None else self.model.config.num_beams
)
default_synced_gpus = True if is_deepspeed_zero3_enabled() else False
gen_kwargs["synced_gpus"] = (
gen_kwargs["synced_gpus"] if gen_kwargs.get("synced_gpus") is not None else default_synced_gpus
Expand Down