A simple guideline for installing CUDA 11.3 and CUDNN 8.2.1.32 in Ubuntu 20.04 LTS. CUDA 11.3 supports maximum frameworks or libraries like Pytorch & TensorRT with easy installation guide.

Install CUDA 11.3 and CUDNN 8.2.1.32 in Ubuntu 20.04 LTS

#!/bin/bash

### Step 1: Verify your gpu is cuda enable ###

    lspci | grep -i nvidia


### Step 2: If you have a previous installation remove it first ###

    sudo apt-get purge nvidia*

    sudo apt remove nvidia-*

    sudo rm /etc/apt/sources.list.d/cuda*

    sudo apt-get autoremove && sudo apt-get autoclean

    sudo rm -rf /usr/local/cuda*


### Step 3: System update ###

    sudo apt-get update

    sudo apt-get upgrade


### Step 4: Install other import packages ###

    sudo apt-get install g++ freeglut3-dev build-essential libx11-dev libxmu-dev libxi-dev libglu1-mesa libglu1-mesa-dev


### Step 5: Get the PPA repository driver ###

    sudo add-apt-repository ppa:graphics-drivers/ppa

    sudo apt update


### Step 6: Install nvidia driver with dependencies ###

    sudo apt install libnvidia-common-470

    sudo apt install libnvidia-gl-470

    sudo apt install nvidia-driver-470


### Step 7: Reboot the system ###

    sudo reboot


### Step 8: Set up CUDA ###

    wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/cuda-ubuntu2004.pin
    
    sudo mv cuda-ubuntu2004.pin /etc/apt/preferences.d/cuda-repository-pin-600
    
    sudo apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/3bf863cc.pub
    
    sudo add-apt-repository "deb https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/ /"
    
    sudo apt-get update


### Step 9: Installing CUDA-11.3 ###

    sudo apt install cuda-11–3


### Step 10: Setup your CUDA paths ###

    echo 'export PATH=/usr/local/cuda-11.3/bin:$PATH' >> ~/.bashrc
    
    echo 'export LD_LIBRARY_PATH=/usr/local/cuda-11.3/lib64:$LD_LIBRARY_PATH' >> ~/.bashrc
    
    source ~/.bashrc
    
    sudo ldconfig


### Step 11: Install cuDNN v11.3 ###

    # First register here: https://developer.nvidia.com/developer-program/signup

    wget https://developer.nvidia.com/compute/machine-learning/cudnn/secure/8.2.1.32/11.3_06072021/cudnn-11.3-linux-x64-v8.2.1.32.tgz

    tar -xzvf ${CUDNN_TAR_FILE}


### Step 12: Copy the following files into the CUDA toolkit directory ###

    sudo cp -P cuda/include/cudnn.h /usr/local/cuda-11.3/include

    sudo cp -P cuda/lib64/libcudnn* /usr/local/cuda-11.3/lib64/

    sudo chmod a+r /usr/local/cuda-11.3/lib64/libcudnn*
 
 
### Step 13: Again, reboot the system ###

    sudo reboot


### Step 14: Finally, verify the installation ###

    nvidia-smi

    nvcc -V