Skip to content

FS-LLM,拼写错误导致训练round数无法被正确赋值,怎么设置都只跑了10round #792

@MillionaireChen

Description

@MillionaireChen

Bug复现

standalone模式下,多个GPU训练,线程数大于1, 会生成StandaloneMultiGPURunner
if mode == 'standalone': if process_num <= 1: runner_cls = StandaloneRunner else: runner_cls = StandaloneMultiGPURunner

然后再runner初始化的时候,
total_round_num 被写成了totol_round_num

Image

server初始化这个参数是10。

Image

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