Feedback about Asynchronous Saving with Distributed Checkpoint (DCP)
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 9.3k
- Forks
- 4.4k
- Avg merge
- 1d 21h
- Merged PRs (30d)
- 4
Description
Hey there! Little nitpick about the last block of this docs page: https://docs.pytorch.org/tutorials/recipes/distributed_async_checkpoint_recipe.html
The checkpoint_future variable is never written to in the last block.
Perhaps the intent was to have this instead?
checkpoint_future = dcp.async_save(state_dict, storage_writer=writer, checkpoint_id=f"{CHECKPOINT_DIR}_step{step}")
cc @LucasLLC @MeetVadakkanchery @mhorowitz @pradeepfn @ekr0 @haochengsong @Saiteja64
Contributor guide
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.
Research direction
Open the linked asynchronous checkpoint recipe and inspect the last code block. Compare the checkpoint_future variable with the preceding dcp.async_save usage and confirm the intended assignment from the issue. Done means the example no longer leaves checkpoint_future unwritten and the documentation remains valid.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- distributed-systems, documentation
- Issue type
- Documentation
- Difficulty
- 1/5
- Estimated time
- Under an hour
- Activity status
- Quiet
- Clarity
- Clearly specified
- Newbie friendliness
- 78/100