win11旗舰版安装WSL子系统和环境-4Cuda和torch

环境配置Cuda10.1配置和cudnn方便torch版本

https://developer.nvidia.com/cuda-toolkit-archive
https://developer.nvidia.com/cuda-10.1-download-archive-update2?target_os=Windows&target_arch=x86_64&target_version=10&target_type=exelocal


https://blog.csdn.net/weixin_40802676/article/details/94753951

Cuda配置和cudnn

在linux里面直接使用conda配置，首先cudatoolkit

conda install cudatoolkit=10.1 -c pytorch
首先配置底层环境，再对pytorch进行尝试

https://pytorch.org/get-started/previous-versions/

pip install -U torch1.5.1 torchvision0.6.1

Cuda配置和cudnn

这个配置的是10.0
不是太方便使用torch

https://zhuanlan.zhihu.com/p/142014944

具体方法如下：Ubuntu系统中，软件源文件地址为：/etc/apt/sources.list
1.备份原来的源，将以前的源备份一下，以防以后可以用的。sudo cp /etc/apt/sources.list /etc/apt/sources.list.bak
2.打开/etc/apt/sources.list文件，在前面添加如下条目，并保存。sudo vim /etc/apt/sources.list
#添加阿里源
deb http://mirrors.aliyun.com/ubuntu/ focal main restricted universe multiverse
deb-src http://mirrors.aliyun.com/ubuntu/ focal main restricted universe multiverse
deb http://mirrors.aliyun.com/ubuntu/ focal-security main restricted universe multiverse
deb-src http://mirrors.aliyun.com/ubuntu/ focal-security main restricted universe multiverse
deb http://mirrors.aliyun.com/ubuntu/ focal-updates main restricted universe multiverse
deb-src http://mirrors.aliyun.com/ubuntu/ focal-updates main restricted universe multiverse
deb http://mirrors.aliyun.com/ubuntu/ focal-proposed main restricted universe multiverse
deb-src http://mirrors.aliyun.com/ubuntu/ focal-proposed main restricted universe multiverse
deb http://mirrors.aliyun.com/ubuntu/ focal-backports main restricted universe multiverse
deb-src http://mirrors.aliyun.com/ubuntu/ focal-backports main restricted universe multiverse
#添加清华源 deb https://mirrors.tuna.tsinghua.edu.cn/ubuntu/ focal main restricted universe multiverse # deb-src https://mirrors.tuna.tsinghua.edu.cn/ubuntu/ focal main restricted universe multiverse deb https://mirrors.tuna.tsinghua.edu.cn/ubuntu/ focal-updates main restricted universe multiverse # deb-src https://mirrors.tuna.tsinghua.edu.cn/ubuntu/ focal-updates main restricted universe multiverse deb https://mirrors.tuna.tsinghua.edu.cn/ubuntu/ focal-backports main restricted universe multiverse # deb-src https://mirrors.tuna.tsinghua.edu.cn/ubuntu/ focal-backports main restricted universe multiverse deb https://mirrors.tuna.tsinghua.edu.cn/ubuntu/ focal-security main restricted universe multiverse # deb-src https://mirrors.tuna.tsinghua.edu.cn/ubuntu/ focal-security main restricted universe multiverse multiverse
3.更新
更新源sudo apt-get update
如出现依赖问题，解决方式如下：
sudo apt-get -f install
更新软件：
sudo apt-get upgrade

win10 Linux子系统WSL2+CUDA11.0+ubuntu-tensorflow-pytorch环境配置

1.2 安装自己的一些必须项
sudo apt install gcc g++ make

1.3 /mnt目录下挂载的文件系统默认权限为777的问题

https://blog.csdn.net/xautzxc/article/details/107610353?utm_medium=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-1.control&depth_1-utm_source=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-1.control

1.3 /mnt目录下挂载的文件系统默认权限为777的问题
把下面automount的选项添加到/etc/wsl.conf文件中
[automount]
enabled = true
root = /mnt/
options = “metadata,umask=22,fmask=11”
mountFsTab = false
在.profile、.bashrc、.zshrc或者其他shell配置文件中重新设置一下umask
#Fix mkdir command has wrong permissions
if grep -q Microsoft /proc/version; then
if [ “$(umask)” == ‘0000’ ]; then umask 0022
fi
fi

3 安装cuda toolkit 10.1

这里官方使用apt install的方式安装cuda-toolkit-11-0，在我这里很慢，我就使用已下载的cuda-10.1的toolkit进行安装（理论上其他toolkit版本的安装方式都一样）在终端执行sudo bash cuda_10.1.243_418.87.00_linux.run --no-opengl-libs在弹出的UI中，不选择nvidia-driver，其余都选安装

sudo bash cuda_10.0.130_410.48_linux.run --no-opengl-libs


https://blog.csdn.net/jacke121/article/details/106935408
查询是gcc出现问题
后面查看，第一个是刚安装的20.04
第二个是18.04




执行这个网页之后，版本还是没有变化
https://blog.csdn.net/snowdream86/article/details/106580471

执行
sudo apt-get remove gcc gcc-9sudo apt-get remove g++

sudo apt install gcc-8 g+±8
改为
sudo apt install gcc-7 g+±7

sudo ln -s /usr/bin/gcc-8 /usr/bin/gcc
sudo ln -s /usr/bin/g+±8 /usr/bin/g++
改为：
sudo ln -s /usr/bin/gcc-7 /usr/bin/gcc
sudo ln -s /usr/bin/g+±7 /usr/bin/g++
继续安装CUDA
sudo bash cuda_10.0.130_410.48_linux.run --no-opengl-libs

3.1添加环境，并验证

sudo nano .bashrc，在末尾添加以下内容：

GNU nano 4.8 .bashrc
export PATH=/usr/local/cuda-10.0/bin:/usr/local/cuda-10.0/nsight-compute-2019.4.0KaTeX parse error: Expected '}', got 'EOF' at end of input: {PATH:+:{PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda-10.0/lib64KaTeX parse error: Expected '}', got 'EOF' at end of input: …LIBRARY_PATH:+:{LD_LIBRARY_PATH}}

sudo nano .bashrc

https://blog.csdn.net/xautzxc/article/details/107610353?utm_medium=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-1.control&depth_1-utm_source=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-1.control

cd /mnt/y/backup/win10预览版need/linux-ubuntu/
source .bashrc
source .bashrc
重新打开终端，执行：nvcc -V

这里出现问题，发现是.bashrc问题，改为~/.bashrc就行

Windows 10 子系统Linux重启(不重启Win10)
Using CMD (Administrator)
net stop LxssManager
net start LxssManager

https://blog.csdn.net/weixin_38169562/article/details/99804798?utm_medium=distribute.pc_relevant_t0.none-task-blog-BlogCommendFromMachineLearnPai2-1.control&depth_1-utm_source=distribute.pc_relevant_t0.none-task-blog-BlogCommendFromMachineLearnPai2-1.control

打开终端，执行

cd ~/NVIDIA_CUDA-10.0_Samples/1_Utilities/deviceQuery
make
./deviceQuery



export PATH=/usr/local/cuda-10.0/bin:/usr/local/cuda-10.0/NsightCompute-1.0KaTeX parse error: Expected '}', got 'EOF' at end of input: {PATH:+:{PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda-10.0/lib64KaTeX parse error: Expected '}', got 'EOF' at end of input: …LIBRARY_PATH:+:{LD_LIBRARY_PATH}}

export LD_LIBRARY_PATH=/usr/lib/x86_64-linux-gnu:$L{D}_{L}IBRAR{Y}_{P}ATHexportL{D}_{L}IBRAR{Y}_{P}ATH=/lib/x86_{6}4-linux-gnu:$LD_LIBRARY_PATH

export CPLUS_INCLUDE_PATH=“path to python /home/yxy/anaconda3/include/python3.6”


cd /mnt/y/backup/win10预览版need/linux-ubuntu/
sudo tar zxvf cudnn-10.0-linux-x64-v7.4.2.24.tgz

4.安装cudnn7.6

下载对应cudnn版本，官网cudnn7.6
解压该文件，tar zxvf cudnn-10.1-linux-x64-v7.6.4.38.tgz
执行：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*
cat /usr/local/cuda/include/cudnn.h | grep CUDNN_MAJOR -A 2