Whats an ml function

Machine Learning (ML) models in Elara allow you to predict unknown values from known values. When you define an ML model, a special model is created that estimates an output value from 0-or-more input values (or features). With sufficient and suitable data, an ML model can be trained to learn what the output value will be, given feature values, and can subsequently be used to generation predictions in functions or scenarios. As you would expect, data to train an ML model is defined as a data stream.

ML Model has a range of configuration options, including the kind of ML model that you can use to model the unknown value. For ease-of-use, Elara automatically chooses a model kind depending on the type of output being predicted by an ML Model. The model kinds are described below:

Type	Description
Neural Network	A type of ML Model that can be used to infer a numeric or a categorical property from 1-or-more features.
Gaussian Process	A type of ML Model that can be used to infer a numeric property from 1-or-more features.
Boosted Tree	A type of ML Model that can be used to infer a categorical property from 1-or-more features.

Next Steps

In the next lessonIn the next lesson you will learn about how to use ml functions to build predictive insights.