Nvidia is an artificial intelligence computing company. Its specialty is GPU. The graphics card it makes is called N card. Try to use N card for deep learning.
GPU refers to a graphics processing unit (English: graphics processing unit, abbreviation: GPU), which is a chip on a graphics card. Compared with a CPU, it has more cores, a smaller core cache, and more computing units. It can perform a large number of parallel calculations. Its specialty is processing Image-related calculations.
Graphics card driver: Obviously it means literally, usually refers to NVIDIA Driver, in fact, it is a driver software, and the previous graphics card is hardware, which realizes the communication between hardware and software
CUDA is a parallel computing platform and programming model (Compute Unified Device Architecture) invented by NVIDIA. It dramatically increases computing performance by harnessing the processing power of a graphics processing unit (GPU).
cudatoolkit is not equal to the graphics card driver, it provides a development environment for the development of high-performance GPU accelerated applications, the toolkit includes GPU accelerated libraries, debugging and optimization tools, a C/C++ compiler and a runtime library for x86, Arm and Build and deploy applications on major architectures such as POWER:
nvcc: nvcc is actually the CUDA compiler, which can be obtained from the /bin directory of CUDA Toolkit, similar to gcc is the c language compiler.
conda cudatoolkit and nvidia cudatoolkit: anaconda's cudatoolkit does not contain all the files for a complete cuda installation, but only the shared library files required for tensorflow, pytorch, xgboost and Cupy.
Use Linux for deep learning work, don't fix wsl, dual system and some moths