deepspeedai / deepspeedai/DeepSpeed
About PipelineModule
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 43.1k
- Forks
- 5k
- Avg merge
- 4d 15h
- Merged PRs (30d)
- 112
Description
If my module contains many other modules,how set json set module_0 on GPU_0, set module_1 on GPU_1,like fairscale
it can set pipeline mode like this
pipe_model = Pipe(nn.Sequential(
Process_0(num_layers=15), Process_1(num_layers=15), Process_2(num_layers=15),
), balance=[1, 1, 1], chunks=3)
and if the input and output things not one , but it have 2 or 3 things, how to solve this matter
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
The issue names PipelineModule and a fairscale Pipe example but no file or test. Start by locating the PipelineModule entry point and determine how nested modules are assigned across GPUs and whether multiple inputs and outputs are supported. Done means the supported configuration is documented or the requested behavior has a clearly scoped implementation.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- 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