modelscope / modelscope/DiffSynth-Studio

`RuntimeError: Sizes of tensors must match` during `torch.cat` when using `deepspeed_stage_3`

Open
#711 5 comments 4 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
13.1k
Forks
1.3k
Avg merge
13h 12m
Merged PRs (30d)
45

Description

When using deepspeed_stage_3 as the training strategy, I encounter a runtime error:

x = torch.cat([cache_x, x], dim=2)
RuntimeError: Sizes of tensors must match except in dimension 2. Expected size 384 but got size 96 for tensor number 1 in the list.

This does not happen with deepspeed_stage_2 or auto.

Contributor guide

No contributing guide indexed for this repository

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 at the torch.cat([cache_x, x], dim=2) call and reproduce the failure with the deepspeed_stage_3 training strategy, comparing the tensor shapes with deepspeed_stage_2 and auto. Done means the size mismatch is resolved for stage 3 and the reported runtime error no longer occurs.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
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.