deepspeedai / deepspeedai/DeepSpeed

Two PipelineModule within a Model will block the deepspeed initialization

Open
#4,589 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug training
Dominant language
Python
Stars
43.1k
Forks
5k
Avg merge
4d 15h
Merged PRs (30d)
112

Description

Describe the bug
I am training a text-protein model, and I need to split both text (LLaMA) and protein (ESM2) models into multiple GPUs.
I add PipelineModule in the text model and protein model separately; then, the code is blocked at the deepspeed initialization.
Does deepspeed not support this manner? Any suggestions, please?
I already make sure that using PipelineModule either in text or protein models individually is feasible.

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 with the PipelineModule initialization path in DeepSpeed and reproduce the reported setup using separate pipeline modules for the text and protein models. The issue names no files or tests, so first determine where initialization blocks and whether multiple PipelineModule instances are supported. Done means identifying the cause and documenting or validating a supported resolution.

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
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.