Skip to content

Conversation

nopperl
Copy link

@nopperl nopperl commented Jun 20, 2023

The readme instructs to use the old torch.distributed.launch command instead of torch.distributed.run, which is incompatible with finetune/finetune.py because it sets the local rank using --local-rank instead of --local_rank, leading to the following error:

finetune.py: error: unrecognized arguments: --local-rank=0

This PR updates the readme to use torch.distributed.run.

Also, it seems like the --local_rank argument is not necessary because it's not used in finetune/finetune.py at all, so I removed it.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant