deepspeedai / deepspeedai/DeepSpeed

[BUG] RuntimeError: 'weight' must be 2-D when two models are used

Open
#4,292 1 comment 1 reaction 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
My training involves two models, in which model A will update parameters, and model B is only used to generate results. Due to resource problems, I have to enable the zero3 strategy, and the sharding of both models at the same time will cause some errors, so how can I make model B not participate in the sharding? I use the deepspeed launcher.

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 DeepSpeed launcher and reproduce the reported setup using two models with the zero3 strategy, recording where the 'weight' 2-D error occurs. Done means identifying a supported way for model B to avoid sharding while model A remains sharded, or documenting the missing configuration or limitation.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
distributed-systems, machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.