Getting Started

There are many ways to leverage machine learning with Ignition such as Python3 libraries, R, or other cloud services like AWS SageMaker. These external resources require additional work to be able to use this data inside of Ignition through the use of some kind of communication protocol to send information back and forth between the model and Ignition.

The Machine Learning Manager resource provides several tools that allow Ignition users to leverage machine learning inside the Ignition platform without the use of external resources or communication methods. This guide will cover how to build, train, and deploy models using the Machine Learning Manager resource. The Machine Learning Manager resource is built using the included Apache Commons Math libraries.

note

This guide is not a resource to learn about machine learning more broadly. While we provide a few examples of algorithms that may be helpful for specific use cases, there are many resources available online where you can learn more about each of the included algorithms in more depth. Towards Data Science is a particularly good source for articles that we recommend for further learning.

In order to start using the Machine Learning Manager resource first you will need to download the project from the Exchange. Once the project is downloaded, import it into a new or existing project and set the default database you will be using to save the model configuration to (MySQL, MSSQL, Postgres, and Oracle are supported).