Introduction

I discovered Julia a few years back, and I’ve been intrigued by its potential and its power ever since. Julia’s user-friendly Integrated Development Environment (IDE) made it very accessible, and its high level logic (very similar to Matlab and other high level languages) and performance made it powerful. However, I was more involved in other, more established platforms such as R and Java, and unable to give Julia my full attention.

As such, I didn’t delve much further beyond the basics, including the applications of the tutorial that was available at the time. Besides, I knew that there are constantly new and interesting languages being developed, most of which never become mainstream.

So, why am I bothering with Julia now? Well, for one, it remained relevant as the years went by; the number of attendees at the Julia conference has been growing considerably every year. Even though I had been familiar with its basics, when I revisited Julia I discovered that I still had a lot to learn, as it had evolved considerably since I’d first encountered it.

Most importantly, it had crossed the pond and made its presence known to people in Europe, one of whom had created a series of exercises and videos about this fairly new language.

After playing around with Version 0.2, I began to wonder if I could actually do something useful with it, beyond factoring numbers quickly or calculating the nth Fibonacci number. However, the few packages that were available with Version 0.2 were poorly documented. There were only a handful of videos introducing the language, most of which were talks from a Python conference. Still, I kept Julia installed on my machine and would use it to write a script from time to time, usually tackling a programming challenge from Project Euler, Programming Praxis, or some similar site. I was a program manager at the time, so I didn’t have much incentive to master a new programming language. Everything I did was a labor of love.

A few months later, though, I started working with data science again; I became more seriously involved with Julia programming. I quickly found that it was easier to code with than Python, for instance, which required a bunch of packages to complete basic data engineering tasks.

After enough casual use, I decided to tackle an entire data science project solely with Julia. Despite the inevitable learning curve and growing pains, I managed to complete it. It wasn’t my best work, but it was proof that with a little practice and trial-and-error, Julia could handle serious data science problems and do so efficiently.

In writing this book I will share this and many other subsequent experiences, exploring how Julia can be integrated into various parts of the data science pipeline. Although there are other books that touch on Julia, there is not a single volume that comprehensively illustrates how Julia can be useful in the realm of data science. I considered waiting for the arrival of such a book, but given the experience I had built in years of experimenting with Julia, I decided to simply write it myself.

I understand that it is a big risk to write a book on a language that’s still in its relative infancy, but it very well may be the case that Julia never stops evolving. If I waited for the language to reach homeostasis, this book would never get written.

I do not expect you to know anything about Julia or to be an established data scientist. If you have the desire to expand your skill-set, the drive to learn new ways of solving old problems, and the discipline to apply yourself throughout this book, you can make Julia an effective part of your data analytics ecosystem.