Introduction

This blog is about how to setup gpu drivers and nvidia-smi for Deep Learning in ubuntu servers.
This guide will be using conda install of python as it is easier to maintain and keep track of correct versions using it.

Dependencies :

No particular dependencies except this has been only tested on ubuntu 16.04 LTS and 18.04 LTS.
(Should work in other versions as well)

Steps :

• Download conda :
  wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh

• Install the latest version of conda :
  bash Miniconda3-latest-Linux-x86_64.sh
  Here during installing add the path to ~/bashrc, so that you can call conda using bash.

• Now create a new conda virtual environment for TF-GPU
  conda create -n TFGPU -c defaults tensorflow-gpu
  (this will install the tensorflow base gpu, cudnn and all required modules)

• Then after this start it by using
  conda activate TFGPU
  (But at this point there are no nvidia drivers so module wont use gpu)

• Now install nvidia-drivers ( this takes a little bit time)

#!/bin/bash                
	echo "Checking for CUDA and installing."          
	# Check for CUDA and try to install.             
	if ! dpkg-query -W cuda; then
	  # The 16.04 installer works with 16.10.
	  curl -O http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64/cuda-repo-ubuntu1604_8.0.61-1_amd64.deb
	  dpkg -i ./cuda-repo-ubuntu1604_8.0.61-1_amd64.deb
	  apt-get update
	  # apt-get install cuda -y
	  sudo apt-get install cuda-8-0
fi

• Then after it is complete you can check the status by
  nvidia-smi :
  Result should be something like this ————————————————