Skip to content

多卡训练 bash scripts/finetune.sh报错 #245

@hdjghjb

Description

@hdjghjb

两张2080ti运行bash scripts/finetune.sh报错
修改的内容有MODEL_PATH改为已经下载到本地的模型路径,DATA_PATH修改为merge.json

错误信息如下:
WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 586 closing signal SIGTERM
ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: -9) local_rank: 1 (pid: 587) of binary: /data/anaconda3/envs/vicuna/bin/python

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions