deepspeedai / deepspeedai/DeepSpeed

pipe_parallel model continue eval_batch train_batch eval_batch.

Open
#3,649 0 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

https://github.com/microsoft/DeepSpeed/blob/52907a661c558044a428eaf9e8a54331d862b5c2/deepspeed/runtime/pipe/engine.py#LL663C23-L663C30

The mistake occurs when I run one pipe_parallel model, followed by the eval_batch train_batch eval_batch.

image

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 deepspeed/runtime/pipe/engine.py around line 663 and reproduce the reported sequence of eval_batch, train_batch, and eval_batch for one pipe-parallel model. Compare the observed behavior with the expected evaluation behavior and add a focused regression test if the relevant test location can be identified; done means both evaluation batches run correctly around training.

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
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.