Hi everyone,
I’m fairly new in this group but I’ve been an avid Julia user since Ver. 0.2. About a year ago I decided to take the next step and start using Julia professionally, namely for data science projects (even if at that time I was a PM in Microsoft). Shortly afterwards, I started writing a book about it, focusing on how we can use this wonderful tool for data science projects. My aim was to make it easy for everyone to learn to make use of their Julia know-how for tackling data science project, but also to help more experienced data scientists to do what they usually do but instead of Python / R / Scala, use a more elegant tool (Julia). I’m writing this post because this book is finally a reality. In this book, which is titled Julia for Data Science and published by Technics Publications, I cover various data science topics. These include data engineering, supervised and unsupervised machine learning, statistics, and some graph analysis. Also, since the focus is on applications rather than digging into the deeper layers of the language, I make use of IJulia instead of Juno, while I also refrain from delving into custom types and meta-programming. Yet, in this book I make use of tools and metrics that are rarely, if ever, mentioned in other data science books (e.g. the T-SNE method, some variants of Jaccard Similarity, some alternative error averages for regression, and more). Also, I try to keep assumptions about the reader’s knowledge to a minimum, so there are plenty of links to references for the various concepts used in the book, from reliable sources. Finally, the book includes plenty of reference sections at the end, so you don’t need to remember all the packages introduced, or all the places where you can learn more about the language. Each chapter is accompanied by a series of questions and some exercises, to help you make sure you comprehend everything you’ve read, while at the end I include a full project for you to practice on (answers to all the exercises and the project itself are at an appendix). All the code used in the book is available on Jupyter files, while the data files are also available in .csv and text format. The book is available in both paperback and eBook format (PDF, Kindle, and Safari) at the publisher’s website: https://technicspub.com/analytics Please note that for some reason the Packt publishing house, which has had the monopoly on Julia books up until now, decided to follow suit, which is why it is releasing a book with the same title next month (clearly, imagination is not their strongest suit!). So, please make sure that you don’t confuse the two books. My goal is not to make a quick buck through this book (which is why I’m not publishing it via Packt); instead, I aim to make Julia more well-known to the data science world while at the same time make data science more accessible to all Julia users. Thanks, Zack
