Lightning-AI / Lightning-AI/lightning-thunder
Make ThunderModule.no_sync behave as PyTorch's Distributed one
Open
Nobody has claimed this yet.
distributed
lightning-l1
- Dominant language
- Python
- Stars
- 1.5k
- Forks
- 121
- PR merge metrics
- No merged PRs in 30d
Description
Currently, we force sync upon exiting the context manager.
This is different to the PyTorch behaviour which syncs at the next call without no-sync.
To enhance compatibility, we should keep this accounting on the ThunderModule (which was not there when we initially implemented no_sync).
cc @crcrpar @ali-alshaar7
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.