关于ubuntu20.04的nvidia-docker安装及pycharm连接docker的配置

1.首先安装docker
docker安装参考链接

2.再安装nivida-docker
安装参考链接

注意：根据的cuda的版本进行测试
cuda查看方式：

user@master:/etc/apt$ nvidia-smi
Fri Nov  6 09:08:02 2020       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.95.01    Driver Version: 440.95.01    CUDA Version: 10.2     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce GTX 1650    Off  | 00000000:01:00.0  On |                  N/A |
| 42%   32C    P8    N/A /  75W |    425MiB /  3908MiB |     17%      Default |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|    0       897      G   /usr/lib/xorg/Xorg                            35MiB |
|    0      1552      G   /usr/lib/xorg/Xorg                           148MiB |
|    0      1686      G   /usr/bin/gnome-shell                         132MiB |
|    0      4532      G   ...AAAAAAAAAAAACAAAAAAAAAA= --shared-files    89MiB |
|    0      6007      G   /usr/lib/firefox/firefox                       2MiB |
|    0      6125      G   ...-token=B3707BBB5A0310CCD0309FAA9728D3B2     5MiB |
+-----------------------------------------------------------------------------+

我的电脑cuda是10.2,这个测试命令修改为

sudo docker run --rm --gpus all nvidia/cuda:11.0-base nvidia-smi

修改为

sudo docker run --rm --gpus all nvidia/cuda:10.0-base nvidia-smi

3.配置docker的远程登录

sudo vim /lib/systemd/system/docker.service

修改ExecStart

# ExecStart=/usr/bin/dockerd -H fd:// --containerd=/run/containerd/containerd.sock

ExecStart=/usr/bin/dockerd -H tcp://0.0.0.0:2375 -H unix://var/run/docker.sock

重启docker服务

systemctl daemon-reload 
systemctl restart docker.service

4.配置pycharm
File->Setting->BuildExecution,Deployment->Docker

5.编译器配置，点击设置-Add

6.配置远程登录

7.设置运行配置，默认是docker执行，我们需要修改为nvidia-docker,修改Edit-config

增加–runtime=nvidia

然后就可以运行程序了