AutoML: Binary and Multinomial Classification Models

  • Article

Important

This content is archived and is not being updated. For the latest documentation, see Microsoft Dynamics 365 product documentation. For the latest release plans, see Dynamics 365 and Microsoft Power Platform release plans.

Note

These release notes describe functionality that may not have been released yet. To see when this functionality is planned to release, please review What’s new and planned for business intelligence. Delivery timelines and projected functionality may change or may not ship (see Microsoft policy).

BI pros and data analysts can now build machine learning inside Power BI with just a few clicks. AutoML uses Azure Machine Learning technology to perform highly specialized tasks such as feature selection, algorithm selection or hyper-parameter optimization. These models can be customized via Power Query with filters or feature engineering.

Models are applied automatically during data ingestion and refresh. Rich Power BI reports are generated for each model, describing the performance of the model, as well as explaining the key influencers’ impact on the predictions generated by the model.

AutoML: Binary and Multinomial Classification Models

Tell us what you think

Help us improve Power BI by discussing ideas, providing suggestions, and giving feedback.

  • Use the Power BI Ideas site to provide suggestions for improving the product.
  • Use the feedback area (Feedback or Is this page helpful?) at the lower right of this page to provide suggestions for improving the content in release notes.