deepspeedai / deepspeedai/DeepSpeed
[REQUEST] How to do 3D Parallelism for HuggingFace Models?
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 43.1k
- Forks
- 5k
- Avg merge
- 4d 15h
- Merged PRs (30d)
- 112
Description
Overview
Does DeepSpeed leverage the 3D parallelism (i.e. data parallelism + pipeline parallelism + tensor parallelism) for huggingface models (e.g., GPT-J, LLaMA) fine-tuning?
May I ask anybody know how to simply implement this using DeepSpeed? Thanks!
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 no files, tests, or entry points. Start by reviewing DeepSpeed's documentation for HuggingFace fine-tuning and the combination of data, pipeline, and tensor parallelism for GPT-J or LLaMA; done means a documented, reproducible procedure or a clear statement of support.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- distributed-systems, machine-learning
- Issue type
- Documentation
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100