双系统ubuntu下的深度学习

admin管理员组

文章数量:1130349

一、ubuntu安装

1. 加装SSD

2. ubuntu 20.04

分区参考http://t.csdnimg/fIIY0

（1）ubuntu和win时间不一致

timedatectl set-local-rtc 1

---

二、深度学习配置

深度学习配置参考https://blog.csdn/m0_37412775/article/details/109355044

1. 显卡驱动

• gcc 9.4.0
• g++ 已经是最新版 (4:9.3.0-1ubuntu2)
• make 已经是最新版 (4.2.1-1.2)
• 驱动安装

驱动版本https://www.nvidia/geforce/drivers/

2. anaconda 2024.06

anaconda安装https://www.anaconda/download

• base : python=3.12

cd ./下载
bash Anaconda3-2024.06-1-Linux-x86_64.sh

• Do you accept the license terms? [yes|no]                >>> yes
• 自动配置环境变量                                                      >>> yes
• ==> For changes to take effect, close and re-open your current shell. <==

取消终端自动进入base环境

conda config --set auto_activate_base false

3. pycharm

pycharm安装https://www.jetbrains/zh-cn/pycharm/download/?section=linux#section=linux

• pycharm-2024.2.1社区版 

cd ./下载

// 安装包解压
tar -zxvf pycharm-community-2024.2.1.tar.gz

// 在/opt/pycharm下安装
sudo mkdir /opt/pycharm
sudo mv pycharm-community-2024.2.1 /opt/pycharm/

sh /opt/pycharm/pycharm-community-2024.2.1/bin/pycharm.sh

4. cuda 11.3.1

cuda安装https://developer.nvidia/cuda-toolkit-archive

wget https://developer.download.nvidia/compute/cuda/11.3.1/local_installers/cuda_11.3.1_465.19.01_linux.run
sudo sh cuda_11.3.1_465.19.01_linux.run

环境配置：

sudo gedit ~/.bashrc
export PATH=$PATH:/usr/local/cuda-11.3/bin
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda-11.3/lib64
export CUDA_HOME=/usr/local/cuda-11.3
source ~/.bashrc

// 报错
No such file or directory: ':/usr/local/cuda-11.3/bin/nvcc': ':/usr/local/cuda-11.3/bin/nvcc'

// 修改
export CUDA_HOME=$CUDA_HOME:/usr/local/cuda-11.3
export CUDA_HOME=/usr/local/cuda-11.3

验证：

nvcc --version

5. cudnn 8.5.0

cudnn安装https://developer.nvidia/cudnn-archive

// 移动压缩包
mv cudnn-linux-x86_64-8.5.0.96_cuda11-archive.tar.xz ./XXX/

// 解压
tar -xvf cudnn-linux-x86_64-8.5.0.96_cuda11-archive.tar.xz

// 移动文件
sudo cp cudnn-*-archive/include/cudnn*.h /usr/local/cuda/include 
sudo cp -P cudnn-*-archive/lib/libcudnn* /usr/local/cuda/lib64 
sudo chmod a+r /usr/local/cuda/include/cudnn*.h

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

一、ubuntu安装

1. 加装SSD

2. ubuntu 20.04

分区参考http://t.csdnimg/fIIY0

（1）ubuntu和win时间不一致

timedatectl set-local-rtc 1

---

二、深度学习配置

深度学习配置参考https://blog.csdn/m0_37412775/article/details/109355044

1. 显卡驱动

• gcc 9.4.0
• g++ 已经是最新版 (4:9.3.0-1ubuntu2)
• make 已经是最新版 (4.2.1-1.2)
• 驱动安装

驱动版本https://www.nvidia/geforce/drivers/

2. anaconda 2024.06

anaconda安装https://www.anaconda/download

• base : python=3.12

cd ./下载
bash Anaconda3-2024.06-1-Linux-x86_64.sh

• Do you accept the license terms? [yes|no]                >>> yes
• 自动配置环境变量                                                      >>> yes
• ==> For changes to take effect, close and re-open your current shell. <==

取消终端自动进入base环境

conda config --set auto_activate_base false

3. pycharm

pycharm安装https://www.jetbrains/zh-cn/pycharm/download/?section=linux#section=linux

• pycharm-2024.2.1社区版 

cd ./下载

// 安装包解压
tar -zxvf pycharm-community-2024.2.1.tar.gz

// 在/opt/pycharm下安装
sudo mkdir /opt/pycharm
sudo mv pycharm-community-2024.2.1 /opt/pycharm/

sh /opt/pycharm/pycharm-community-2024.2.1/bin/pycharm.sh

4. cuda 11.3.1

cuda安装https://developer.nvidia/cuda-toolkit-archive

wget https://developer.download.nvidia/compute/cuda/11.3.1/local_installers/cuda_11.3.1_465.19.01_linux.run
sudo sh cuda_11.3.1_465.19.01_linux.run

环境配置：

sudo gedit ~/.bashrc
export PATH=$PATH:/usr/local/cuda-11.3/bin
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda-11.3/lib64
export CUDA_HOME=/usr/local/cuda-11.3
source ~/.bashrc

// 报错
No such file or directory: ':/usr/local/cuda-11.3/bin/nvcc': ':/usr/local/cuda-11.3/bin/nvcc'

// 修改
export CUDA_HOME=$CUDA_HOME:/usr/local/cuda-11.3
export CUDA_HOME=/usr/local/cuda-11.3

验证：

nvcc --version

5. cudnn 8.5.0

cudnn安装https://developer.nvidia/cudnn-archive

// 移动压缩包
mv cudnn-linux-x86_64-8.5.0.96_cuda11-archive.tar.xz ./XXX/

// 解压
tar -xvf cudnn-linux-x86_64-8.5.0.96_cuda11-archive.tar.xz

// 移动文件
sudo cp cudnn-*-archive/include/cudnn*.h /usr/local/cuda/include 
sudo cp -P cudnn-*-archive/lib/libcudnn* /usr/local/cuda/lib64 
sudo chmod a+r /usr/local/cuda/include/cudnn*.h

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

本文标签： 深度双系统Ubuntu