About this book
This book serves two slightly different audiences. First, it serves software engineers who are interested in machine learning but haven’t built many real-world machine learning systems. I presume such readers want to put their skills into practice by actually building something with machine learning. The book is different from other books you may have picked up on machine learning. In it, you’ll find techniques applicable to building whole production-grade systems, not just naive scripts. We’ll explore the entire range of possible components you might need to implement in a machine learning system, with lots of hard-won tips about common design pitfalls. Along the way, you’ll learn about the various jobs of a machine learning system, in the context of implementing systems that fulfill those needs. So, if you don’t have a lot of background in machine learning, don’t worry that you’ll have to wade through pages of math before you get to build things. The book will have you coding all the way through, often relying on libraries to handle the more complex implementation concerns like model learning algorithms and distributed data processing.