Skip to content

Instantly share code, notes, and snippets.

@arifsuhan
Last active November 27, 2019 07:12
Show Gist options
  • Save arifsuhan/f0574eb0fc6bc1cfbd734eb7351f15f5 to your computer and use it in GitHub Desktop.
Save arifsuhan/f0574eb0fc6bc1cfbd734eb7351f15f5 to your computer and use it in GitHub Desktop.
ubuntu 14.06 driver setup

Step 1: Add the Official Nvidia PPA to Ubuntu

To add the drivers repository to Ubuntu, run the commands below:

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

When you run the commands above, you should get a prompt to accept the repository signing key as well.. accept and continue installing the PPA…

Step 2: Update and Install Nvidia Drivers

Now that the PPA is installed, run the commands below to install the current latest drivers for your system.

sudo apt update
sudo apt install nvidia-387

Step 3: Check GPU:

  nvidia-smi

Step 4: Cuda

 $ cd Downloads
 $ chmod +x ./[file].run
 $ sudo ./[file].run

set path

 export PATH="/home/User/anaconda3/bin:/usr/local/cuda/bin:$PATH"
 export LD_LIBRARY_PATH="/usr/local/cuda/lib64:$LD_LIBRARY_PATH"
 source ~/.bashrc

Step 5: Cudnn

 sudo dpkg -i [x].deb (runtime)
 sudo dpkg -i [x].deb (developer)
 sudo dpkg -i [x].deb (sample code)

Step 6: Conda

 curl -o [filename].sh

Step 7 : Pytorch

Reference

https://websiteforstudents.com/install-proprietary-nvidia-gpu-drivers-on-ubuntu-16-04-17-10-18-04/ https://medium.com/hackernoon/deep-learning-software-setup-cuda-10-ubuntu-18-04-15548cefa30 https://medium.com/repro-repo/install-cuda-and-cudnn-for-tensorflow-gpu-on-ubuntu-79306e4ac04e

Lastest Working

https://medium.com/repro-repo/install-cuda-10-1-and-cudnn-7-5-0-for-pytorch-on-ubuntu-18-04-lts-9b6124c44cc

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment