What makes a successful machine learning practitioner?

To be clear, the challenges of real-world machine learning are not due to the addition of more advanced or complex methods; after all, the first nine chapters of this book covered practical, real-world problems as diverse and challenging as identifying cancer cells, filtering spam messages, and predicting risky bank loans. Instead, the challenges of real-world machine learning have much to do with aspects of the field that are difficult to convey in a scripted setting, like a textbook or lecture. Machine learning is as much art as it is science, and just as it would be challenging to learn to paint, dance, or speak a foreign language without real-world practice, it is equally difficult to apply machine learning methods to new, uncharted domains.

Like pioneers exploring distant lands, you will encounter never-before-seen challenges requiring soft skills, including persistence and creativity. You will encounter large, messy...