Who this book is for

I had multiple audiences in mind as I wrote this book, but I most consistently thought about a dear friend of mine who bought a Transact-SQL book 30 years ago and instantly developed great confidence in her database work, ultimately building a career around those skills. I would love it if someone just starting their career as a data scientist or analyst worked through this book and had a similar experience as my friend. More than anything else, I want you to feel good and excited about what you can do as a result of reading this book.

I also hope this book will be a useful reference for folks who have been doing this kind of work for a while. Here, I imagine someone opening the book and wondering to themselves, what are good values to use in my grid search for my logistic regression model?

In keeping with the hands-on nature of this text, every bit of output is reproducible with code in this book. I also stuck to a rule throughout, even when it was challenging. Every section, except for the conceptual sections, starts with raw data largely unchanged from the original downloaded file. You go from data file to model in each section. If you have forgotten how a particular object was created, all you will ever need to do is turn back a page or two to see.

Readers who have some knowledge of pandas and NumPy will have an easier time with some code blocks, as will folks with some knowledge of Python and scikit-learn. None of that is essential though. There are just some sections you might want to pause over longer. If you need additional instruction on doing data work with Python, my Python Data Cleaning Cookbook is a good companion book I think.