RuntimeError: CUDA error: all CUDA-capable devices are busy or unavailable

在使用pytorch构建网络时,  执行model.cuda()时, 出现标题上的错误, 还会报如下错误 :

RuntimeError: cuda runtime error (30) : unknown error at /opt/conda/conda-bld/pytorch_1544199946412/work/aten/src/THC/THCGeneral.cpp:51

 

目前网上的解决办法是重新安装显卡(不一定行), 短期办法是重启电脑(以后仍会出现该问题).

 

 

在网上找了很多方法, 下面是一些解决方案 : 

https://stackoverflow.com/questions/6966496/nvidia-cuda-error-all-cuda-capable-devices-are-busy-or-unavailable-on-osx

https://devtalk.nvidia.com/default/topic/470566/cuda-programming-and-performance/why-quot-all-cuda-capable-devices-are-busy-or-unavailable-quot-/

https://discuss.pytorch.org/t/resolved-cuda-runtime-error-30/1116

https://github.com/milesial/Pytorch-UNet/issues/8

https://discuss.pytorch.org/t/runtimeerror-cuda-runtime-error-30-unknown-error/10100/3

https://discuss.pytorch.org/t/resolved-cuda-runtime-error-30/1116

https://github.com/pytorch/pytorch/issues/4293 (Zombie process when use GPU)