[Feature] disable distributed pytorch

Open bblumberg opened this issue 2 years ago • 2 comments

What is the feature?

The default pytorch implementation on the nvidia orin does not include torch.distributed. So to use mmengine on the orin requires a special build of pytorch which is non-trivial. So, it would be awesome if there was a way to disable distributed so the code doesn't look for functions which aren't present: ReduceOp() for example.

Any other context?

No response

Mar 24 '23 19:03 bblumberg

Thanks for your feedback. It would be better to support running on the NVIDIA orin but we do not have the device. If you are interested, we will provide much help for you to support it.

Mar 29 '23 09:03 zhouzaida

Thanks for your feedback. It would be better to support running on the NVIDIA orin but we do not have the device. If you are interested, we will provide much help for you to support it.

i met the problems too. Have u solved it?

Jul 13 '24 15:07 quanfeifan