You need to enable JavaScript to run this app.
优惠活动
大模型
产品
解决方案
定价
更多

使用Hydra Submitit Launcher插件提交Slurm集群任务时无法导入本地models模块的问题求助

使用Hydra Submitit Launcher插件提交Slurm集群任务时无法导入本地models模块的问题求助

我在实验中用Hydra做配置管理,现在尝试用hydra submitit launcher plugin自动把任务提交到slurm集群。

我的主配置文件是这样的:

defaults:
    - override hydra/launcher: slurm
foo: 1

对应的slurm配置文件如下:

defaults:
  - submitit_slurm

_target_: hydra_plugins.hydra_submitit_launcher.submitit_launcher.SlurmLauncher

submitit_folder: ${hydra.sweep.dir}/.submitit/%j
name: ${hydra.job.name}

项目结构是这样的:

|--project
|  |--src
|  |  |--main.py
|  |  |--models
|  |  |  |--__init__.py
|  |  |  |--file1.py
|  |  |  |--file2.py
|  |--scripts
|  |  |--run.sh

在__init__.py里我定义了要从models导出的内容,main.py里正常用from models import func导入。

问题所在:
当我把主配置文件里的submitit launcher注释掉时,一切运行正常,但取消注释后就会报错,错误信息如下:

submitit ERROR (2025-01-04 21:31:32,789) - Submitted job triggered an exception
Traceback (most recent call last):
  File "<frozen runpy>", line 198, in _run_module_as_main
  File "<frozen runpy>", line 88, in _run_code
  File "/home/a/.conda/envs/hinet/lib/python3.12/site-packages/submitit/core/_submit.py", line 11, in <module>
    submitit_main()
  File "/home/a/.conda/envs/hinet/lib/python3.12/site-packages/submitit/core/submission.py", line 76, in submitit_main
    process_job(args.folder)
  File "/home/a/.conda/envs/hinet/lib/python3.12/site-packages/submitit/core/submission.py", line 69, in process_job
    raise error
  File "/home/a/.conda/envs/hinet/lib/python3.12/site-packages/submitit/core/submission.py", line 52, in process_job
    delayed = utils.DelayedSubmission.load(paths.submitted_pickle)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/a/.conda/envs/hinet/lib/python3.12/site-packages/submitit/core/utils.py", line 153, in load
    obj = pickle_load(filepath)
          ^^^^^^^^^^^^^^^^^^^^^
  File "/home/a/.conda/envs/hinet/lib/python3.12/site-packages/submitit/core/utils.py", line 232, in pickle_load
    return pickle.load(ifile)
           ^^^^^^^^^^^^^^^^^^
ModuleNotFoundError: No module named 'models'

有没有人知道可能的原因和解决办法?

谢谢!

备注:内容来源于stack exchange,提问作者Ahmed

相关产品推荐
方舟 Agent Plan

超全模态模型 × Harness 升级,最新支持 Deepseek-V4.1-Flash、GLM-5.3 系列、Doubao-Seedream-5.0-pro、Kimi-K3 (部分), 限时 9.9 元起

最近更新时间:2026.04.14 12:43:08