[PT-D][Checkpoint] Move distributed checkpointing from torch.distributed._shard.checkpoint to torch.distributed.checkpoint #88698
Conversation
wz337
requested review from
mrshenli,
zhaojuanmao,
pritamdamania87,
rohan-varma,
H-Huang,
awgu and
kwen2501
as code owners
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/88698
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 9ef96b7: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
wz337
changed the title
wz337
changed the title
| ) | ||
| sys.modules['torch.distributed._shard.checkpoint'] = torch.distributed.checkpoint |
| from .api import CheckpointException | ||
|
|
||
|
|
||
| from .planner import ( |
There was a problem hiding this comment.
are these all of the public APIs we want to expose to the user? shall we also add a __all__ here to hide the ones we don't want to expose?
| @@ -4,6 +4,9 @@ | |||
| ShardMetadata, | |||
| ) | |||
|
|
|||
| __all__: List[str] = [] | |||
There was a problem hiding this comment.
do we need the __all__ on all python modules?
|
@pytorchmergebot rebase |
|
@pytorchbot successfully started a rebase job. Check the current status here |
|
Successfully rebased |
|
@pytorchmergebot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Merge failedReason: 2 additional jobs have failed, first few of them are: trunk ,trunk / cuda11.6-py3.10-gcc7-sm86 / test (default, 2, 4, linux.g5.4xlarge.nvidia.gpu) Details for Dev Infra teamRaised by workflow job |
|
@pytorchmergebot rebase |
|
@pytorchbot successfully started a rebase job. Check the current status here |
…eckpoint to torch.distributed.checkpoint
|
Successfully rebased |
|
@pytorchmergebot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
wz337
changed the title
Some distributed tests are moved to a new location after pytorch#88698
Some distributed tests are moved to a new location after #88698. This is currently failing periodic multigpu job: * https://github.com/pytorch/pytorch/actions/runs/3484486207/jobs/5829301159 * https://github.com/pytorch/pytorch/actions/runs/3484486207/jobs/5829301093 Pull Request resolved: #89225 Approved by: https://github.com/clee2000
Update import for spmd checkpoint files, as we have moved distributed checkpointing from torch.distributed._shard.checkpoint to torch.distributed.checkpoint in PyTorch (pytorch/pytorch#88698). Test: CI
… checkpoint (#89256) Update test import and docstring as we have moved distributed checkpointing from torch.distributed._shard.checkpoint to torch.distributed.checkpoint (#88698). Test: CI Pull Request resolved: #89256 Approved by: https://github.com/fduwjj
…ibuted._shard.checkpoint to torch.distributed.checkpoint (pytorch#88698) Context in RFC: pytorch#86620 .rst file will be finalized in subsequent PRs. Pull Request resolved: pytorch#88698 Approved by: https://github.com/wanchaol
…h#89225) Some distributed tests are moved to a new location after pytorch#88698. This is currently failing periodic multigpu job: * https://github.com/pytorch/pytorch/actions/runs/3484486207/jobs/5829301159 * https://github.com/pytorch/pytorch/actions/runs/3484486207/jobs/5829301093 Pull Request resolved: pytorch#89225 Approved by: https://github.com/clee2000
… checkpoint (pytorch#89256) Update test import and docstring as we have moved distributed checkpointing from torch.distributed._shard.checkpoint to torch.distributed.checkpoint (pytorch#88698). Test: CI Pull Request resolved: pytorch#89256 Approved by: https://github.com/fduwjj
Context in RFC: #86620
.rst file will be finalized in subsequent PRs.