GuoZF

icon company Institute of Computing Technology, Chinese Academ icon location China icon location Studying in Institute of Computing Technology, Chinese Academ. Focusing on Speech，Sound and NLP.

Results 2 comments of GuoZF

`Accelerate test` failing on an 8-GPU docker instance

I can run with a small amount of data, but large-scale data can cause errors: Some NCCL operations have failed or timed out. Due to the asynchronous nature of CUDA...

InitProcessGroupKwargs(timeout=timedelta(seconds=3600)) not work !!!!!

I also find this promblem.