CarperAI / CarperAI/trlx

Why train dataloader is not prepared by Accelerator

Open
#594 0 comments 0 reactions 0 assignees View on GitHub
bug
Dominant language
Python
Stars
4.8k
Forks
487
PR merge metrics
No merged PRs in 30d

Description

### 🐛 Describe the bug

https://github.com/CarperAI/trlx/blob/3340c2f3a56d1d14fdd5f13ad575121fa26b6d92/trlx/trainer/accelerate_ppo_trainer.py#L231

the train dataloader used in PPO is not prepared by Accelerator.

Then in the training phase (
(https://github.com/CarperAI/trlx/blob/3340c2f3a56d1d14fdd5f13ad575121fa26b6d92/trlx/trainer/accelerate_base_trainer.py#L563)), will the `minibatch_size` be used as the batch size per device or not?

### Which trlX version are you using?

_No response_

### Additional system and package information

_No response_

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.