Summary
The purpose of this chapter is to prepare you 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 in depth through 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 you have learned so far.
In the next chapter, we will discuss categorizing images of clothing using Convolutional Neural Networks.
