deepspeedai / deepspeedai/DeepSpeed

Zero level 3 fails when using a nn.Parameter, because Deepspeed can only partition nn.Linear

Open
#956 6 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

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

Description

Zero level 3 fails when using a nn.Parameter, because Deepspeed can only partition nn.Linear

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 reproducing the reported ZeRO level 3 failure with an nn.Parameter, then trace the partitioning path that handles nn.Linear. Compare how the two parameter types are treated and identify the existing tests covering ZeRO-3. Done means nn.Parameter works under ZeRO-3 without the reported failure and the behavior is covered by a regression test.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
distributed-systems, machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.