Mastering machine learning is a desirable skill nowadays given its vast application everywhere, from business to academia. Nevertheless, just understanding the theory of it will only take you so far since practitioners also need to understand their tools to be self-sufficient and capable.
In this chapter, we started with a high-level introduction to machine learning and learned when to use each of the machine learning types; from classification and regression to clustering and reinforcement learning. We then learned about scikit-learn and why practitioners recommend it when solving both supervised and unsupervised learning problems. To keep this book self-sufficient, we also covered the basics of data manipulation for those who haven't used libraries such as pandas and Matplotlib before. In the following chapters, we will continue to combine our understanding of the underlying theory of machine learning with more practical examples using scikit-learn.
The first two parts of this book will cover supervised machine learning algorithms. The first part will cover basic algorithms, as well as some other machine learning basics, such as data splitting and preprocessing. Then, we will move on to more advanced topics in the second part. The third and final part will cover unsupervised learning and topics such as anomaly detection and recommendation engines.
So that this book remains a practical guide, I have made sure to provide examples in each chapter. I also did not want to separate the data preparation from model creation. Although topics such as data splitting, feature selection, data scaling, and model evaluation are key concepts to know about, we usually deal with them as part of an overall whole solution. I also feel that those concepts are best understood in their correct context. That's why, within each chapter, I will be covering one main algorithm but will use some examples to shed light on some other concepts along the way.
This means that it is up to you whether you read this book from cover to cover or use it as a reference and jump straight to the algorithms you want to know about when you need them. Nevertheless, I advise that you skim through all the chapters, even if you already know about the algorithm covered there or don't need to know about it at the moment.
I hope that you are now ready for the next chapter, where we will start by looking at decision trees and learn how to use them to solve different classification and regression problems.
