open-mmlab / mmskeleton

A OpenMMLAB toolbox for human pose estimation, skeleton-based action recognition, and action synthesis.
Apache License 2.0
2.93k stars 1.04k forks source link

AssertionError: Invalid device id #304

Closed 1584899550 closed 4 years ago

1584899550 commented 4 years ago

thanks for your works! when I run the command
mmskl configs/recognition/st_gcn_aaai18/ntu-rgbd-xview/test.yaml I've got an error AssertionError: Invalid device id can you tell me how to solve this problem? looking for your replay! thanks a lot.

yysijie commented 4 years ago

The default setting uses 4 GPUs. We have modified the default setting in the latest version. Please update the codebase and run the command again:

mmskl configs/recognition/st_gcn_aaai18/ntu-rgbd-xview/test.yaml

If you want to change the number of gpus, use --gpus. For example:

mmskl configs/recognition/st_gcn_aaai18/ntu-rgbd-xview/test.yaml --gpus 4
1584899550 commented 4 years ago

Thank you for your reply. I have solved this problem. Thank you very much! ------------------ 原始邮件 ------------------ 发件人: "sijie yan"<notifications@github.com>; 发送时间: 2020年5月24日(星期天) 晚上10:08 收件人: "open-mmlab/mmskeleton"<mmskeleton@noreply.github.com>; 抄送: "我在说"<1584899550@qq.com>;"Author"<author@noreply.github.com>; 主题: Re: [open-mmlab/mmskeleton] AssertionError: Invalid device id (#304)

The default setting uses 4 GPUs. We have modified the default setting in the latest version. So just update the codebase and run the command again: mmskl configs/recognition/st_gcn_aaai18/ntu-rgbd-xview/test.yaml
If you want to change the number of gpus, use --gpus. For example: mmskl configs/recognition/st_gcn_aaai18/ntu-rgbd-xview/test.yaml --gpus 4
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

MaarufB commented 3 years ago

I've got the same error but not for test.yaml, I got that error when running the train.yaml. This is the command i used ----- mmskl configs/recognition/st_gcn_aaai18/kinetics-skeleton/train.yaml --gpus 4