Last Updated on 2021-01-20 by Clay
Introduction
If you are familiar with the deep learning and its environment, you may often hear "CUDA". Before we starting to train our model, maybe we need to install GPU driver, CUDA and cuDNN toolkits. They can make our training have the best effect.
But today I want to record how to remove CUDA –– it is a parallel computing architecture.
Nowadays, removing CUDA is very easy.
Removing CUDA version > 10.0
If your CUDA version is higher than 10.0（not include 10.0）, you can use the uninstall program to do it.
sudo /usr/local/cuda-10.2/bin/cuda-uninstallercuda-10.x you need to write your version.
This method is suitable for CUDA 10.1 and CUDA 10.2.
Removing CUDA version <= 10.0
If your CUDA version is lower than 10.0, you also can use the following script to remove it.
sudo /usr/local/cuda-10.0/bin/uninstall_cuda_10.0.plSince I have not tested a lower version, it is difficult to be sure which version the uninstall program can be applied to. Maybe you can study it.