The purpose of the last chapter of this book is to prepare ourselves for real-world machine learning problems. We started with the general workflow that a machine learning solution follows: data preparation, training sets generation, algorithm training, evaluation and selection, and finally, system deployment and monitoring. We then went through in depth the typical tasks, common challenges, and best practices for each of these four stages.
Practice makes perfect. The most important best practice is practice itself. Get started with a real-world project to deepen your understanding and apply what we have learned throughout the entire book.
