Skip to content

Use model.from_pretrained for DataParallel also #8795

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 3 commits into from
Nov 30, 2020
Merged

Use model.from_pretrained for DataParallel also #8795

Changes from 2 commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
d4d79c6
Use model.from_pretrained for DataParallel also
shaie Nov 26, 2020
e1d99db
Fix silly bug
shaie Nov 26, 2020
b94712d
Address review comments
shaie Nov 30, 2020
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
7 changes: 7 additions & 0 deletions src/transformers/trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -808,8 +808,15 @@ def train(self, model_path: Optional[str] = None, trial: Union["optuna.Trial", D
logger.info(
f"Loading best model from {self.state.best_model_checkpoint} (score: {self.state.best_metric})."
)
is_data_parallel = False
if isinstance(model, torch.nn.DataParallel):
model = model.module
is_data_parallel = True
if isinstance(model, PreTrainedModel):
self.model = model.from_pretrained(self.state.best_model_checkpoint)
if is_data_parallel:
# re-wrap with DataParallel
self.model = torch.nn.DataParallel(self.model)

@sgugger sgugger Nov 30, 2020

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The model attribute of the Learner is never wrapped in DataParallel or the like, it stays a reference to the original model so these lines are not necessary.
With the same reasoning, I think the only change to make is on the line above this comment (L816) and replace

self.model = model.from_pretrained(self.state.best_model_checkpoint)

by

self.model = self.model.from_pretrained(self.state.best_model_checkpoint)

if not self.args.model_parallel:
self.model = self.model.to(self.args.device)
else:
Expand Down