My First Book
I have spent my last 6 professional years centered around machine learning and today marks a significant milestone in this journey. 4 months ago I was offered to co-write a book on supervised learning. I immediately accepted the offer. It was not easy to write a technical book in such a short amount of time especially alongside a full-time job. But, it has certainly been worth the time and effort spent. This blog post is to share what the book is about and my journey in writing it.
The Supervised Learning Workshop
The book, titled - The Supervised Learning Workshop is meant for professionals as well as students who have some knowledge of general programming (preferably in Python) and who want to get some hands on experience with the kind of machine learning that uses labelled datasets i.e. supervised learning. In my practical experience, almost 80-90 % of machine learning work I have done has been about supervised learning, and this book helps readers to both understands its concepts as well as gain the ability to train and validate a range of supervised machine learning models.
About the publisher
Packt is a publisher based out of Birmingham, focused on publishing tech-related books, ebooks, training videos, etc. This particular book, I am told, is a part of a Machine Learning series that Packt is producing primarily for another IT training company Global Knowledge. Apparently, Global Knowledge is the biggest of its kind in the USA. Packt publishes such books related to programming and technology regularly and has a huge resource pool of authors. Some of its books are quite popular in the tech-community such as this one. Overall, my experience of working with them was good.
About the journey
Many of you might wonder what is it like to write a technical book. I will try to summarize my experience here. With the caveat that it might depend a lot on the editor and the publishing team that is working with you, the process is usually straight-foward. In my case, I was lucky to be working with a great team of proactive people. I have written the second edition of this book which means, my work was split into writing a few pieces from scratch and editing some pre-written articles.
I have written chapters 5, 6 and 7 in this book which were focused around classification models, ensemble models and model evaluation techniques. For the material, I mostly drew inspiration from all that I have learnt in my formal education and even more so from the day-to-day work experience. For some topics, I did go back to read up from some of the foundational books - Bishop and HTF.
Writing activity was done on google docs with 5 rounds of review in the form of google-docs comments. Once the writing was finalised, it was taken over by the production team for the final touch-ups. Because this book is interactive and hands-on, I spent more than half the time writing actual python tutorials which were embedded into the book as exercises and activities. All of that code exists on Packt’s github repo. I also wrote unit tests for all the tutorials, and also wrote all the skeleton text equivalent of the code for the textbook. The production team created videos of the jupyter notebooks written for tutorials.
All in all, co-authoring the book was a good learning experience for me. I got to see the internals of how a publishing team works as well as I was forced to re-visit a bunch of supervised learning concepts in order to pen them down appropriately for the readers. If you are eager to write a technical book on an area you have expertise in, I can only encourage you to do that. Besides learning and making an accomplishment out of it, this is also a good way to give back to the community. This might sound rhetorical but, for instance, my career perhaps wouldn’t exist if Andrew Ng didn’t do what he did for the ML community.
The book is on Amazon
You can find the book both as paperback as well as kindle on Amazon. The book is also available on Packt’s own website. If you do read the book, please leave a review based on how you find it. I sincerely hope you find it useful in getting up to speed with working on your Machine Learning projects.
Besides the book, Packt is for the first time also trying an interactive version. In the interactive version, readers can not just read the material but also run and test their code as they read. The interactive version also includes video tutorials on how to run the exercises and activities. You can almost imagine the interactive version to be like a coursera course, just replace the tutor’s videos with the textbook’s text.
Acknowledgements
Big thanks to Packt for providing me this co-authoring opportunity. Special thanks to Anindya Sil for reaching out on LinkedIn and Snehal Tambe for her continued assistance throughout the project. I also thank and congratulate my co-author Blaine Bateman for helping in and sharing this journey together.
Thank you for reading this post. Hope it was insightful.
Sneak Peek
Here is a glimpse of what the book is about:
Download this pdf here