Top Books to get started in Machine Learning

Published Jan. 24, 2020

The internet is filled with technical resources that help beginners pick up a new topic, and Machine Learning is no exception there. It’s filled so much, it’s hard to distinguish good resources from bad, and create a good roadmap of resources. We’ve put together a list of resources for your first step in ML together with our mentors, and are excited to share it.

Top Books to get started in Machine Learning

Preparation

At the start of your journey into ML, you should prepare and build all the skills needed to make full use of the tools needed in Machine Learning itself. While the journeys differ, it’s fair to say that Python is a stable in this area. To make the most use of Machine Learning frameworks, it’s also crucial to learn data processing and data analysis skills. So where can you learn all of this?

Python Cookbook by David Beazley & Brian K. Jones

The cookbook series of O’Reilly is always a great starting point for any language. In Machine Learning, you can really profit from a big universal understand of Python, even if it’s not about the data domain specifically.

The Python Cookbook takes you through all those basics by providing practical examples. Unlike other language learning books, it’s not just a collection of syntax examples, but shows you how to achieve things in different ways, showing you many faces of the Python Programming Language

Python Cookbook by David Beazley & Brian K. Jones

Python for Data Analysis: Data Wrangling with Pandas, NumPy, and IPython By Wes McKinney

Before data can be fed into a Machine Learning model, it’s usually a good thing to analyze it, process it and learn from it. The tools needed for that are usually Pandas, Numpy and Matplotlib.

The Data Wrangling book by Wes McKinney leads you through all of that and makes you comfortable working with a variety of datasets.

Python for Data Analysis: Data Wrangling with Pandas, NumPy, and IPython By Wes McKinney

Learning the Basics

Let’s get to the core! These books can help you learn ML theory and practice.

Machine Learning for Absolute Begineers by Oliver Theobald

Machine Learning can be a mind-boggling concept. While this may not be a technical introduction, this book may be a great intro to all thise terminologies and concepts around Machine Learning.

Again, this is not a technical book at all, but it can help understand the further materials better.

Machine Learning For Absolute Beginners by Oliver Theobald

Hands-On Machine Learning with Scikit-Learn, Keras and TensorFlow by Aurélien Géron

The Hands-On ML book by Aurélien Géron is the all-in-one guide to solving Machine Learning problems. From solving easy predictions with Scikit-Learn, to building complex Neural Networks with TensorFlow.

Theory is important, that’s why this book does its best to combine ML theory with useful applications.

Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow by Aurélien Géron

Grokking Deep Learning by Andrew Trask

Let’s take it right from the website: Grokking Deep Learning teaches you to build deep learning neural networks from scratch! Using only Python and its math-supporting library, NumPy, you’ll train your own neural networks to see and understand images, translate text into different languages, and even write like Shakespeare!

If you are looking to understand the theory behind Neural Networks more without all the theoretical load, then doing it grokking style can be a great idea!

Grokking Deep Learning by Andrew Trask

Tips & Tricks

So, now that you have the basics done, what can you do to really get that ‘wow’ effect when applying or trying out something new?

Strengthening Deep Neural Networks by Katy Warr

Machine Learning algorithms can be fooled by a variety of methods, and it will be a rising skill to know how to guard against that! This book shows you through the process of attacking and defending neural networks.

Strengthening Deep Neural Networks by Katy Warr

TinyML by Pete Warden

Machine Learning is entering the production and with that, there are new requirements. The common ‘state of the art’ methods are usually done on large, complex algorithms that take a long time to load and execute.

So what if you need to run on a tiny device with a few kilobytes of memory and not a ton of power! This book can show you some way to make these nets more efficient for low power environments.

TinyML by Pete Warden