【Linux】Ubuntu 深度学习环境配置（CUDA+CUDNN+Pycharm）

1. CUDA

查看 cuda 版本号

nvidia-smi

下载 CUDA 安装包 https://developer.nvidia.com/cuda-downloads

照着提示安装 cuda

wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/cuda-ubuntu2004.pin
sudo mv cuda-ubuntu2004.pin /etc/apt/preferences.d/cuda-repository-pin-600
wget https://developer.download.nvidia.com/compute/cuda/11.4.2/local_installers/cuda-repo-ubuntu2004-11-4-local_11.4.2-470.57.02-1_amd64.deb
sudo dpkg -i cuda-repo-ubuntu2004-11-4-local_11.4.2-470.57.02-1_amd64.deb
sudo apt-key add /var/cuda-repo-ubuntu2004-11-4-local/7fa2af80.pub
sudo apt-get update
sudo apt-get -y install cuda

添加 cuda 环境变量

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

更新环境变量

source ~/.bashrc

验证安装是否成功

nvcc -V

2. CUDNN

下载 cudnn https://developer.nvidia.com/zh-cn/cudnn （需要注册）

解压

tar zxvf cudnn-11.4-linux-x64-v8.2.4.15.tgz

把 CUDNN 文件复制到 CUDA 的对应文件夹（插入式设计）

sudo cp cuda/include/cudnn*.h /usr/local/cuda/include
sudo cp cuda/lib64/libcudnn* /usr/local/cuda/lib64/
sudo chmod a+r /usr/local/cuda/include/cudnn*.h
sudo chmod a+r /usr/local/cuda/lib64/libcudnn*

查看安装情况

cat /usr/local/cuda/include/cudnn_version.h | grep CUDNN_MAJOR -A 2

3. Miniconda

下载 https://docs.conda.io/en/latest/miniconda.html

安装

bash Miniconda3-latest-Linux-x86_64.sh

测试安装是否成功

conda list

如果不想看见 base

conda config --set auto_activate_base false

4. conda 使用

conda 创建使用教程

conda GPU测试教程

5. Pycharm

下载 https://www.jetbrains.com/zh-cn/pycharm/download/#section=linux

解压

tar zxvf pycharm-professional-2021.2.2.tar.gz

安装

sudo mkdir /opt/pycharm
sudo mv pycharm-2021.2.2/ /opt/pycharm/

启动

sh /opt/pycharm/pycharm-2021.2.2/bin/pycharm.sh

桌面图标放进应用程序

Reference

1. Linux系统CUDA10.2+CUDNN安装教程
2. Ubuntu 18.04安装CUDA和cuDNN
3. NVIDIA CUDA Installation Guide for Linux-3.8. Ubuntu
4. NVIDIA CUDA Installation Guide for Linux-12. Post-installation Actions
5. conda Docs » User guide » Installation » Installing on Linux
6. 在Ubuntu 20.04中安装Pycharm 2020.1