Скачать или смотреть Install WSL2 on Windows 11 with NVIDIA CUDA 11.8

I explain how to install Ubuntu in WSL2. Afterwards I install Nvidia Cuda 11.8 and the nvidia cuda toolkit.
Then I install pip3 and pytorch. I verify the successful installation by compiling a cuda program.
I show how to fix an issue related to the accessibility of the nvidia profiler output. I have listed the exact comamnds down below the links and timestamps.
The links to the official documentation are here:
NVIDIA CUDA Installation Guide for Linux: https://docs.nvidia.com/cuda/cuda-ins...
CUDA Toolkit 11.8 Downloads: https://developer.nvidia.com/cuda-11-...
Pytorch Download: https://pytorch.org/get-started/locally/
_________________________________________________________________
Timestamps:
0:00 Activate Windows Features to Install WSL2 on Windows 11/10
0:32 Setting default WSL Version to WSL2
0:44 How to Install WSL2 on Windows 11/10
1:22 basics about WSL2 commands
2:24 how to install GCC for NVIDIA CUDA
2:39 how to install NVIDIA CUDA for WSL2
4:16 how to set path variable for NVIDIA CUDA on linux
5:42 how to install nvidia cuda toolkit on linux
6:15 how to install pytorch on linux
7:45 check pytorch installation
8:15 compiling cuda program to see if cuda works
8:40 fix nvidia profiler error
9:22 how to uninstall nvidia cuda
9:57 optional: incase WSL2 does not work: install kernel update
_________________________________________________________________
Commands without comments:
cd ~
sudo apt install gcc --fix-missing
wget https://developer.download.nvidia.com...
mv cuda-wsl-ubuntu.pin /etc/apt/preferences.d/cuda-repository-pin-600
wget https://developer.download.nvidia.com...
dpkg -i cuda-repo-wsl-ubuntu-11-8-local_11.8.0-1_amd64.debsudo
cp /var/cuda-repo-wsl-ubuntu-11-8-local/cuda-*-keyring.gpg /usr/share/keyrings/
sudo apt-get updatesudo apt-get -y install cuda
_________________________________________________________________
cd ~
nano .bashrc
export PATH=/usr/local/cuda-11.8/bin${PATH:+:${PATH}}
source ~/.bashrc
echo $PATH
sudo apt install nvidia-cuda-toolkit
nvcc -V
nvidia-smi

sudo apt-get install python3-pip
pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118
python3
torch.cuda.is_available()
torch.__version__

cd ~
nano test.cu
nvcc test.cu -o test
./test
nvprof ./test
_________________________________________________________________
#pytorch #nvidia #wsl2
_________________________________________________________________
Danke!