ensemble Module

Contains functionality for creating ensembles from previous automated machine learning iterations.

Creating ensembles can improve machine learning results by combining multiple iterations that may provide better predictions compared to a single iteration. Configure an experiment to use ensembles with the AutoMLConfig object.

Classes

Ensemble	Represents a collection of previous AutoML iterations brought together into an ensemble. The ensemble pipeline is initialized from a collection of already fitted pipelines. Create an Ensemble pipeline out of a collection of already fitted pipelines.
VotingEnsemble	Defines an ensemble created from previous AutoML iterations that implements soft voting. You do not use the VotingEnsemble class directly. Rather, specify using VotingEnsemble with the AutoMLConfig object. Create an Ensemble pipeline out of a collection of already fitted pipelines.