The book is similar in many ways to my old blog series. It's a no-nonsense tour of twenty or so machine learning algorithms, covering many typical use-cases. You learn to write many of the algorithms from scratch, though now that JS has a bit of an ML ecosystem we also use popular libraries (like TensorFlow.js) for the more complex algorithms. I show you the math where necessary, though I focus more on the implementations and the practical advice you'll need when starting to work with ML algorithms.
Teaching is a passion of mine, and I'm always looking for new ways to do it. While I love what I've been able to do with this blog, I think that ML in JS in book form is the right way to present the right material to the right people. I hope you agree, and thanks for being a reader!