lm-sys / lm-sys/FastChat

Issue when training long context model on V100 with xformers with padding_mask error

Open
#2,623 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
39.5k
Forks
4.8k
PR merge metrics
No merged PRs in 30d

Description

I followed the training code in README.md, -- of course change train_mem.py with train_xformers.py.

I got the following error message:
```
TypeError: xformers_forward() got an unexpected keyword argument 'padding_mask'
```
Any hints?

Is there any requirement on the transformers versions of FastChat? I am using transformers==4.34.1

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 the training instructions in README.md, then compare train_mem.py with train_xformers.py around the xformers_forward call. Reproduce the reported padding_mask error using transformers==4.34.1 and determine whether the documented training path or a compatible dependency version is required; done means the long-context V100 training path no longer raises this error.

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
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.