An error when runned applications/ChatGPT/examples/train_sft.py #3224
Unanswered
guijuzhejiang
asked this question in
Community | Q&A
Replies: 1 comment
-
Hi @guijuzhejiang Thanks for your feedback. It's still under development, maybe you can try it again next week. Thanks. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Excuting:
python train_sft.py --pretrain facebook/opt-350m --model opt --strategy naive --log_interval 10
Error message:
RuntimeError: Default process group has not been initialized, please make sure to call init_process_group.
It seems that it cannot run on a single GPU. May I ask the parameter settings when running on a single GPU.
Also, is train_sft.py the training code of stage1 of chatgpt?
Beta Was this translation helpful? Give feedback.
All reactions