Machine Learning Systems: Designs that scale

by Jeff Smith

Machine learning applications autonomously reason about data at massive scale. It's important that they remain responsive in the face of failure and changes in load. But machine learning systems are different than other applications when it comes to testing, building, deploying, and monitoring.

Reactive Machine Learning Systems teaches readers how to implement reactive design solutions in their machine learning systems to make them as reliable as a well-built web app. Using Scala and powerful frameworks such as Spark, MLlib, and Akka, they'll learn to quickly and reliably move from a single machine to a massive cluster.

Purchase of the print book includes a free eBook in PDF, Kindle, and ePub formats from Manning Publications.

Genres: Computer Science, Technology

224 pages, Paperback

Published July 8, 2018

About the author

Librarian Note: There is more than one author in the GoodReads database with this name. See authors with similar names.

Reviews

[Zdeněk Tisoň · Rating 3 out of 5]

This book describes all steps which needs to be done to have machine learning feature in your system. Steps are described briefly and simple implementation on Scala stack is provided.
I think that book should provide more details also some real world issues and solutions could be provided.
According to reactivity its not only about Spark, Akka. I miss reactive ml models learning, data processing by streams and so on.

[Mikhail Filatov · Rating 1 out of 5]

This book has very little relation to "machine learning" : except a couple of chapters it's really about software engineering topics the author likes: functional programming with Scala, actors in Akka, etc., etc. Can they be used in "machine learning system" - of course. Are they really needed to build one? Not at all.
It seems to be an extension of the blog post where the social app for dogs was used as an example, so in every chapter the "case study" involves some animals: lions, turtles, etc. You feel like reading a book for children at times.

[Raja Pakrashi · Rating 4 out of 5]

Loved this short book. Rather than focusing on the working of Machine learning it focuses a lot on the system design around Machine Learning applications which in the real world determines the success of an organization ML strategy. Highlight of the book was the discussion around distributed databases to effectively collect data using the animal kingdom analogy.

[Bart Frenk]

Haven't read it in full, but this seems to be a fairly shallow book. I also thought the supposedly humorous example detract from the message. To me, they were more distracting than useful.