Skip to content

Invoke parent DDP configuration for torch>1.10.2 #12912

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
merged 3 commits into from
Apr 28, 2022
Merged

Invoke parent DDP configuration for torch>1.10.2 #12912

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
5861aeb
Invoke parent DDP configuration for torch>1.10.2
jerome-habana Apr 28, 2022
172cd23
update change log
jerome-habana Apr 28, 2022
02b80b1
Merge branch 'master' into hpu_parallel
carmocca Apr 28, 2022
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: 3 additions & 2 deletions CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -150,10 +150,11 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
- Fixed an issue to ensure all the checkpoint states are saved in a common filepath with `DeepspeedStrategy` ([#12887](https://github.com/PyTorchLightning/pytorch-lightning/pull/12887))


- Fixed an issue where sharded grad scaler is passed in when using BF16 with the `ShardedStrategy` ([#12915](https://github.com/PyTorchLightning/pytorch-lightning/pull/12915))
- Fixed an issue wrt recursive invocation of DDP configuration in hpu parallel plugin ([#12912](https://github.com/PyTorchLightning/pytorch-lightning/pull/12912))


-
- Fixed an issue where sharded grad scaler is passed in when using BF16 with the `ShardedStrategy` ([#12915](https://github.com/PyTorchLightning/pytorch-lightning/pull/12915))


- Fixed printing of ragged dictionaries in `Trainer.validate` and `Trainer.test` ([#12857](https://github.com/PyTorchLightning/pytorch-lightning/pull/12857))

Expand Down
2 changes: 1 addition & 1 deletion pytorch_lightning/strategies/hpu_parallel.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -103,7 +103,7 @@ def configure_ddp(self) -> None:
self._model._set_static_graph() # type: ignore
self._register_ddp_hooks()
else:
self.configure_ddp()
super().configure_ddp()

def broadcast(self, obj: object, src: int = 0) -> object: # type: ignore
obj = [obj]
Expand Down