reference:
Install WSL2 (Ubuntu20.04) in Windows11 and configure the data analysis environment (anaconda, pyspark) - Bilibili introduces the installation of the linux subsystem on win10/win11, and builds an environment on it for data analysis. Installing WSL2 before installation requires Windows 10 2004 or above or Windows 11, and it needs to be Professional Edition or above. If you don’t meet the conditions, please give up or upgrade the system yourself. In addition, you need to check whether CPU virtualization and virtual machine platform are turned on. Open the task manager (Ctrl+Alt+Esc), switch to performance, you can see the status of CPU virtualization as shown in the figure, the default is disabled, you need to enter the Bios to enable it, the Bios of different manufacturers' motherboards are different, please check it yourself, generally speaking , it will be https://www.bilibili.com/read/cv16825907/ in CPU Configuration
windw11 system:
choose
win11+rtx3060ti+tensorflow installation
win11+rtx3060ti+ tensorflow 
installation
If you forget the cuda installation path, you can use cmd, set cuda
(base) C:\Users\12974>set cuda
CUDA_PATH=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6
CUDA_PATH_V11_6=C:\Program Files \NVIDIA GPU Computing Toolkit\CUDA\v11.6
1. Check Nvidia graphics card and cuda version in win10
One: The installation process
Two: related issues
1. Error code: Wsl/Service/CreateVm/ConfigureNetworking/ERROR_FILE_NOT_FOUND
solve:
Open the cmd command window of window11
First look at what other subsystems
wsl --list --all
unregister subsystem
wsl --unregister Ubuntu-22.04
Reference: How to solve "The system cannot find the file specified" when upgrading from wsl1 to wsl2? - Zhihu (zhihu.com)