Chapter 9: Production Scoring and the H2O MOJO
We spent the entire previous section learning how to build world-class models against data at scale with H2O. In this chapter, we will learn how to deploy these models and make predictions from them. First, we will cover the background on putting models into production scoring systems. We will then learn how H2O makes this easy and flexible. At the center of this story is the H2O MOJO (short for Model Object, Optimized), a ready-to-deploy scoring artifact that you export from your model building environment. We will learn technically what a MOJO is and how to deploy it. We will then code a simple batch file scoring program and embed a MOJO in it. We will finish with some final notes on the MOJO. ...
Get Machine Learning at Scale with H2O now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
