小编opt*_*mus的帖子

我的操作系统是Ubuntu 18.04\xef\xbc\x8can并且我已经安装了nvidia驱动程序和nvidia-docker，但无法检测到CUDA。

$ docker run -it --gpus all nvidia/cuda:10.0-cudnn7-runtime-ubuntu18.04 /bin/bash\nroot@7c0be9bfaeec:/# nvcc --version\nbash: nvcc: command not found\n

我也无法在 Docker 容器中检测到 CUDNN：

root@694cefcad33a:/# cat /usr/local/cuda/include/cudnn.h | grep CUDNN_MAJOR -A 2\ncat: /usr/local/cuda/include/cudnn.h: No such file or directory\n

然而，当我跑步时nvidia-smi，它起作用了！

root@7c0be9bfaeec:/# nvidia-smi\nThu Dec 19 04:51:23 2019\n+-----------------------------------------------------------------------------+\n| NVIDIA-SMI 440.44 Driver Version: 440.44 CUDA Version: 10.2 |\n|-------------------------------+----------------------+----------------------+\n| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |\n| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |\n