lllyasviel / lllyasviel/ControlNet

Why `pl.Trainer` can not handle multi-gpu case?

Open
#319 6 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
34.1k
Forks
3k
PR merge metrics
No merged PRs in 30d

Description

I can run the original `tutorial_train.py` with single 3090Ti GPU (24G) with batch_size 3.

However, when upgrade to 2 or more gpus, it keep warning OOM.
```
trainer = pl.Trainer(gpus=2 precision=32, callbacks=[logger])
```

I am curious why? Why single GPU can handle batch 3 while multi-GPU can only handle 1?? The GPUS hold batches on their own parallelly, am I right?

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 with tutorial_train.py and the pl.Trainer configuration shown in the report. Reproduce the single-GPU and multi-GPU runs with the stated batch sizes, then inspect how the trainer distributes batches and allocates memory. Done means documenting a confirmed explanation for the differing OOM behavior.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.