Instructions for CUDA v11.7 and cuDNN 8.5 installation on Ubuntu 22.04 for PyTorch 1.12.1

cuda_11.7_installation_on_Ubuntu_22.04

#!/bin/bash

### steps ####
# verify the system has a cuda-capable gpu
# download and install the nvidia cuda toolkit and cudnn
# setup environmental variables
# verify the installation
###

### to verify your gpu is cuda enable check
lspci | grep -i nvidia

### If you have 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*

# system update
sudo apt-get update
sudo apt-get upgrade

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

# first get the PPA repository driver
sudo add-apt-repository ppa:graphics-drivers/ppa
sudo apt update

# install nvidia driver with dependencies
sudo apt install libnvidia-common-515
sudo apt install libnvidia-gl-515
sudo apt install nvidia-driver-515

wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-ubuntu2204.pin
sudo mv cuda-ubuntu2204.pin /etc/apt/preferences.d/cuda-repository-pin-600
sudo apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/3bf863cc.pub
sudo add-apt-repository "deb https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/ /"
sudo apt-get update

 # installing CUDA-11.7
sudo apt install cuda-11-7 

# setup your paths
echo 'export PATH=/usr/local/cuda-11.7/bin:$PATH' >> ~/.bashrc
echo 'export LD_LIBRARY_PATH=/usr/local/cuda-11.7/lib64:$LD_LIBRARY_PATH' >> ~/.bashrc
source ~/.bashrc
sudo ldconfig

# install cuDNN v11.7
# First register here: https://developer.nvidia.com/developer-program/signup

CUDNN_TAR_FILE="cudnn-linux-x86_64-8.5.0.96_cuda11-archive.tar.xz"
wget https://developer.nvidia.com/compute/cudnn/secure/8.5.0/local_installers/11.7/cudnn-linux-x86_64-8.5.0.96_cuda11-archive.tar.xz
tar -xzvf ${CUDNN_TAR_FILE}

# copy the following files into the cuda toolkit directory.
sudo cp -P cuda/include/cudnn.h /usr/local/cuda-11.7/include
sudo cp -P cuda/lib/libcudnn* /usr/local/cuda-11.7/lib64/
sudo chmod a+r /usr/local/cuda-11.7/lib64/libcudnn*

# Finally, to verify the installation, check
nvidia-smi
nvcc -V

# install Pytorch (an open source machine learning framework)
pip install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu117

Reinstalling cuda after some update messed up the previous installation. These instructions worked for me when installing driver-535

but now i get this bunch of errors

sudo apt install nvidia-driver-535
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:

The following packages have unmet dependencies:
 nvidia-driver-535 : Depends: libnvidia-compute-535 (= 535.104.12-0ubuntu1) but 535.113.01-0ubuntu0.22.04.3 is to be installed
                     Depends: libnvidia-decode-535 (= 535.104.12-0ubuntu1) but it is not going to be installed
                     Depends: libnvidia-encode-535 (= 535.104.12-0ubuntu1) but it is not going to be installed
                     Recommends: nvidia-settings but it is not going to be installed
                     Recommends: nvidia-prime (>= 0.8) but it is not going to be installed
                     Recommends: libnvidia-compute-535:i386 (= 535.104.12-0ubuntu1)
                     Recommends: libnvidia-decode-535:i386 (= 535.104.12-0ubuntu1)
                     Recommends: libnvidia-encode-535:i386 (= 535.104.12-0ubuntu1)
                     Recommends: libnvidia-fbc1-535:i386 (= 535.104.12-0ubuntu1)
                     Recommends: libnvidia-gl-535:i386 (= 535.104.12-0ubuntu1)
E: Unable to correct problems, you have held broken packages.

Purging also didnt fix it. What options do i have?

You may want to look at this solution: https://github.com/ichxw/cuda_11.7_installation_on_Ubuntu_22.04

@ichxw hey thanks for sharing ur answer. I still get stuck at the step sudo apt install nvidia-driver-515, or for my case it is sudo apt install nvidia-driver-535 since im going for that version.

 sudo apt-get install nvidia-driver-535
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:

The following packages have unmet dependencies:
 nvidia-driver-535 : Depends: libnvidia-compute-535 (= 535.104.12-0ubuntu1) but 535.113.01-0ubuntu0.22.04.3 is to be installed
                     Depends: libnvidia-decode-535 (= 535.104.12-0ubuntu1) but it is not going to be installed
                     Depends: libnvidia-encode-535 (= 535.104.12-0ubuntu1) but it is not going to be installed
                     Recommends: nvidia-settings but it is not going to be installed
                     Recommends: nvidia-prime (>= 0.8) but it is not going to be installed
                     Recommends: libnvidia-compute-535:i386 (= 535.104.12-0ubuntu1)
                     Recommends: libnvidia-decode-535:i386 (= 535.104.12-0ubuntu1)
                     Recommends: libnvidia-encode-535:i386 (= 535.104.12-0ubuntu1)
                     Recommends: libnvidia-fbc1-535:i386 (= 535.104.12-0ubuntu1)
                     Recommends: libnvidia-gl-535:i386 (= 535.104.12-0ubuntu1)
E: Unable to correct problems, you have held broken packages.:
~$ sudo apt install -f libnvidia-decode-535 libnvidia-encode-535
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:

The following packages have unmet dependencies:
 libnvidia-decode-535 : Depends: libnvidia-compute-535 (= 535.104.12-0ubuntu1) but 535.113.01-0ubuntu0.22.04.3 is to be installed
E: Unable to correct problems, you have held broken packages.
~$ sudo apt install -f libnvidia-compute-535
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
libnvidia-compute-535 is already the newest version (535.113.01-0ubuntu0.22.04.3).
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.

Tried installing the dependencies manually too but it didnt work
If i trace the error message, would it be helpful if i tried to install libnvidia-compute-535 version 535.104.12-0ubuntu1 ?
but i am not sure how do i specify in my terminal that i want specifically that version only.

Reinstalling cuda after some update messed up the previous installation. These instructions worked for me when installing driver-535

but now i get this bunch of errors

sudo apt install nvidia-driver-535
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:

The following packages have unmet dependencies:
 nvidia-driver-535 : Depends: libnvidia-compute-535 (= 535.104.12-0ubuntu1) but 535.113.01-0ubuntu0.22.04.3 is to be installed
                     Depends: libnvidia-decode-535 (= 535.104.12-0ubuntu1) but it is not going to be installed
                     Depends: libnvidia-encode-535 (= 535.104.12-0ubuntu1) but it is not going to be installed
                     Recommends: nvidia-settings but it is not going to be installed
                     Recommends: nvidia-prime (>= 0.8) but it is not going to be installed
                     Recommends: libnvidia-compute-535:i386 (= 535.104.12-0ubuntu1)
                     Recommends: libnvidia-decode-535:i386 (= 535.104.12-0ubuntu1)
                     Recommends: libnvidia-encode-535:i386 (= 535.104.12-0ubuntu1)
                     Recommends: libnvidia-fbc1-535:i386 (= 535.104.12-0ubuntu1)
                     Recommends: libnvidia-gl-535:i386 (= 535.104.12-0ubuntu1)
E: Unable to correct problems, you have held broken packages.

Purging also didnt fix it. What options do i have?

do this sudo apt install libnvidia-compute-535

Thanks a lot for making the shell script. I am also running into a dependency issue, however. Any fix for this? sudo apt install libnvidia-compute-515 is already part of the script, so that doesn't help. Tried 535 as well
The following packages have unmet dependencies:
cuda-11-7 : Depends: cuda-runtime-11-7 (>= 11.7.1) but it is not going to be installed
Depends: cuda-demo-suite-11-7 (>= 11.7.91) but it is not going to be installed

@primus852 @fatbringer @ichxw @qburst-fidha Did you guys get this figured out?

Although the above suggestions did not work for me, I found that the runfile installation from the nvidia website, rather than the apt-get installation does work. Maybe this helps someone else!

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*

Reboot