在训练环境中设置环境变量

问题描述

在训练环境中通过os.environ["CUBLAS_WORKSPACE_CONFIG"]="4096:8"或是os.system("export ...")方式都失败，
先通过上述方面设置环境变量，后通过os.system("python train.py"), 出现下述提示
you must set an environment variable before running your PyTorch application: CUBLAS_WORKSPACE_CONFIG=:4096:8 or CUBLAS_WORKSPACE_CONFIG=:16:8.

任务类型（调试/训练/推理）

训练

任务名

日志说明或问题截图

you must set an environment variable before running your PyTorch application: CUBLAS_WORKSPACE_CONFIG=:4096:8 or CUBLAS_WORKSPACE_CONFIG=:16:8.

期望的解决方案或建议

image.png

4.9 KiB

平台不对设置环境变量做限制，正常都是可生效的，请尝试更换镜像。对于日志中的报错，可参考https://discuss.pytorch.org/t/random-seed-with-external-gpu/102260/19

非常感谢

谢谢

Deleting a branch is permanent. It CANNOT be undone. Continue?

Dear OpenI User

Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.

For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》

#1061 在训练环境中设置环境变量

问题描述

相关环境（GPU/NPU）

相关集群（启智/智算）

任务类型（调试/训练/推理）

任务名

日志说明或问题截图

期望的解决方案或建议