deepspeedai / deepspeedai/DeepSpeed
[REQUEST] Pipeline Parallelism support multi optimizer to train
Open
Nobody has claimed this yet.
enhancement
- Dominant language
- Python
- Stars
- 43.1k
- Forks
- 5k
- Avg merge
- 4d 15h
- Merged PRs (30d)
- 112
Description
Is your feature request related to a problem? Please describe.
i want to train big model gan, must shard to multi-gpu, but the pipeline_module seems not support multi-optimizer
Describe the solution you'd like
support
Describe alternatives you've considered
can control flow to which layer
Additional context
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 locating the pipeline_module entry point and reading how it currently handles optimizers for sharded multi-GPU training. Clarify the expected control flow for assigning layers and optimizers, then define tests showing that a GAN can train with multiple optimizers across pipeline stages.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- distributed-systems, machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100