Practical Automated Machine Learning Using H2O.ai

12

Working with H2O AutoML and Apache Spark

In Chapter 10, Working with Plain Old Java Objects (POJOs), and Chapter 11, Working with Model Object, Optimized (MOJO), we explored how to build and deploy our Machine Learning (ML) models as POJOs and MOJOs in production systems and use them to make predictions. In the majority of real-world problems, you will often need to deploy your entire ML pipeline in production so that you can deploy as well as train models on the fly. Your system will also be gathering and storing new data that you can later use to retrain your models. In such a scenario, you will eventually need to integrate your H2O server into your business product and coordinate the ML effort.

Apache Spark is one of the more commonly ...