famunir

Installing the NVIDIA driver, CUDA and cuDNN on Linux (Ubuntu 20.04)

This is a companion piece to my instructions on building TensorFlow from source. In particular, the aim is to install the following pieces of software

• NVIDIA graphics card driver (v450.57)
• CUDA (v11.0.2)
• cuDNN (v8.0.2.39)

on an Ubuntu Linux system, in particular Ubuntu 20.04.

famunir

Introduction

This document will not be providing any details, mathematics, algorithms or complex ideas on machine learning (By the way a series of blog posts by Chris McCormik provide details about lots of basic ideas). Instead, in this document I will try my best to provide a pathway to the beginners on how to approach the topic of machine learning (specifically in terms of NEURAL NETWROKS).

So basically, if you are new to the field, I will try to provide you a path way by pointing out the best sources (IN MY OPINION) that will give you the best understanding of a machine learing related topic. In addition, I will also try to provide the sources of prerequisite topics for machine learning, instead of saying, for instance, you should know the signal processing. Now signal processing is a vast field and includes a gazzilion important topics. One may not need to learn (THOUGH YOU SHOULD TRY TO) all the major concepts of signal processing for learning or atleast start learn

famunir

Updated 9/12/2018

NVIDIA CUDA kit version X.X.X installation guide on Ubuntu Desktop 16.04.4 LTS.

Table of Contents

• Install NVIDIA Graphics Driver via apt-get
• Install CUDA
• Install cuDNN

Table of contents generated with markdown-toc