【Ubuntu】ubuntu20.04 nv驱动+cuda+cudnn+anaconda3安装教程

wL魔法师

已于 2022-05-14 23:30:54 修改

阅读量778

点赞数

分类专栏： linux 运维与安装文章标签： ubuntu

于 2021-01-21 08:35:54 首次发布

本文链接：https://blog.csdn.net/qq_33000453/article/details/112760350

版权

运维与安装同时被 2 个专栏收录

4 篇文章 0 订阅

订阅专栏

linux

2 篇文章 0 订阅

订阅专栏

1.安装nvida 驱动

1.1. 换阿里源

1. 备份source.list

sudo cp /etc/apt/sources.list /etc/apt/sources.list.save

2. 修改source.list

sudo gedit /etc/apt/sources.list

阿里源

3. 更新 apt

sudo apt-get update

1.2. 安装驱动

使用终端命令查看Ubuntu推荐的驱动版本：

这里推荐了460版本

PC:~$  ubuntu-drivers devices
== /sys/devices/pci0000:00/0000:00:01.0/0000:01:00.0 ==
modalias : pci:v000010DEd00002206sv00007377sd0000140Abc03sc00i00
vendor   : NVIDIA Corporation
driver   : nvidia-driver-460 - distro non-free recommended
driver   : xserver-xorg-video-nouveau - distro free builtin

以nvidia-460版本为例

sudo add-apt-repository ppa:graphics-drivers/ppa   # 添加软件源
sudo apt-get update
sudo apt-get install nvidia-460

自动安装

sudo ubuntu-drivers autoinstall

1.3. 成功

MS-7B17:~$ nvidia-smi
Wed Jan 20 18:29:55 2021       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 460.32.03    Driver Version: 460.32.03    CUDA Version: 11.2     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  GeForce RTX 3080    Off  | 00000000:01:00.0  On |                  N/A |
| 31%   24C    P5    32W / 320W |    164MiB / 10010MiB |      1%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|    0   N/A  N/A       911      G   /usr/lib/xorg/Xorg                 63MiB |
|    0   N/A  N/A      1214      G   /usr/bin/gnome-shell               89MiB |
|    0   N/A  N/A      2136      G   /usr/lib/firefox/firefox            3MiB |
|    0   N/A  N/A      2370      G   /usr/lib/firefox/firefox            5MiB |
+-----------------------------------------------------------------------------+

2.cuda

2.1首先安装gcc

sudo apt install gcc

2.2安装cuda

sudo chmod+x ./cuda_11.1.0_455.23.05_linux.run
sudo sh ./cuda_11.1.0_455.23.05_linux.run

2.3设置环境变量

gedit ~/.bashrc

export PATH=/usr/local/cuda-11.1/bin${PATH:+:${PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda-11.1/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}

source ~/.bashrc

2.4验证

MS-7B17:~$ nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2020 NVIDIA Corporation
Built on Tue_Sep_15_19:10:02_PDT_2020
Cuda compilation tools, release 11.1, V11.1.74
Build cuda_11.1.TC455_06.29069683_0

3.cudnn

运行命令 sudo nautilus,可以直接粘贴

4.anaconda3

环境配置

sudo gedit ~/.basrc

export PATH=/home/{username}/anaconda3/bin:$PATH

source ~/.bashrc

镜像加速

conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/
conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main/
conda config --set show_channel_urls yes