2019-01-16

I'm training Kaldi's Aishell mandarin model today, and there comes error " "

Failed to allocate a memory region of 5625610240 bytes. Possibly this is due to sharing the GPU. Try switching the GPUs to exclusive mode (nvidia-smi -c 3) and using the option --use-gpu=wait to scripts like steps/nnet3/chain/train.py. Memory info: free:10729M, used:449M, total:11178M, free/total:0.959811

i guess it may caused by using the remote Desktop to control server, which consume some GPU memories.

try to solve it by setting exlusive

compute-exclusive mode. Suggestion: use 'nvidia-smi -c 3' to set compute exclusive mode

你可能感兴趣的:(2019-01-16)