0 Docker容器!
看起来似乎和主机没什么区别...
1 进入Docker并打开Jupyter
sudo nvidia-docker run -p 8888:8888 --privileged=true --device /dev/nvidia-uvm:/dev/nvidia-uvm --device /dev/nvidia0:/dev/nvidia0 --device /dev/nvidia1:/dev/nvidia1 --device /dev/nvidiactl:/dev/nvidiactl -it -v /home/test:/home/ghy -v /usr/lib/x86_64-linux-gnu/:/usr/lib/x86_64-linux-gnu / tensorflow/tensorflow:latest-gpu
注:
-p 8888:8888 打开Jupyter(端口号为8888),前面的8888为指定的本地端口号
-v /usr/lib/x86_64-linux-gnu/:/usr/lib/x86_64-linux-gnu 不这么写的话,在使用Tensorflow时会出现‘ ImportError: libcuda.so.1: cannot open shared object file: No such file or directory’
/ tensorflow/tensorflow:latest-gpu 下载的TensorFlow