Skip to content

加载100000模型,_load_zero_checkpoint失败,提示没有相关zero_pp_rank*文件 #46

@Tron1994

Description

@Tron1994

官方提供checkpoint只有四个mp_rank文件,但加载时提示找不到zero_pp_rank
环境是deepspeed==0.7,目前手动设置不让加载zero_checkpoint

是说提供的预训练模型没有用到zero_pp吗

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions