Lightning-AI / Lightning-AI/pytorch-lightning

Strategy for DeepSpeedRLHFEngine and DeepSpeed Hybrid Engine

Open
#18,455 0 comments 2 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

feature
Dominant language
Python
Stars
31.4k
Forks
3.8k
Avg merge
6d 7h
Merged PRs (30d)
6

Description

### Description & Motivation

I want to train RLHF model with DeepSpeedRLHFEngine or DeepSpeed Hybrid Engine. My understanding is current DeepSpeedStrategy doesn't work with them. Would like to know any plan for new strategy to support them? If not in the near future, It's great if you could show me simple work around to do it. Thanks for reply!

### Pitch

_No response_

### Alternatives

_No response_

### Additional context

_No response_

cc @borda

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reviewing the current DeepSpeedStrategy and the DeepSpeedRLHFEngine and DeepSpeed Hybrid Engine mentioned in the issue. Determine whether supporting these engines requires a new strategy or a documented workaround; done means a clear implementation plan or usable workaround is available.

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

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.