deepspeedai / deepspeedai/DeepSpeed
[Question] Combining sync batchnorm with DeepSpeed?
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 43.1k
- Forks
- 5k
- Avg merge
- 4d 15h
- Merged PRs (30d)
- 112
Description
Hi,
I'm not sure how to combine synchronized batchnormalization with deepspeed.
Using model_engine.module = torch.nn.SyncBatchNorm.convert_sync_batchnorm(model_engine.module)
Gives me the error: AttributeError: SyncBatchNorm is only supported within torch.nn.parallel.DistributedDataParallel
Is there another way of making my batchnorms synchronized?
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
Start by reproducing the call to torch.nn.SyncBatchNorm.convert_sync_batchnorm on model_engine.module and reviewing the reported DistributedDataParallel restriction. Determine whether DeepSpeed supports synchronized batch normalization with this setup; done would be a confirmed supported approach or a clearly documented limitation.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- distributed-systems, machine-learning
- Issue type
- Bug
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100