cancel
Showing results for 
Search instead for 
Did you mean: 

What's New in DataRobot Release 5.3?

DataRobot Alumni
7 1 468

Release 5.3 banner-2.png

Here’s a highlight of some of the key features and changes delivered in Release 5.3. (Make sure to tell us what you think of this material.)

AI Catalog. ​Introduced in Release 5.2, AI Catalog can help you find, understand and use the assets you need for your AI projects. In this release, we’ve improved dataset versioning and feature list management so you always know you’re working with the latest information. We’ve also added the ability to store database credentials, import additional types of information, and improved the user experience throughout—including a revised SQL editor.

Automated Feature Engineering. ​DataRobot accelerates machine learning by automating feature engineering, often considered the most laborious and time-consuming step along the path from data to value. In this release, we’ve improved how primary and secondary datasets are used for feature discovery and predictions, as well as added a new feature to manage feature lineage.

MLOps. ​Also introduced in Release 5.2, MLOps provides monitoring and management for custom models built using leading frameworks and languages such as Python and R. Now you can use MLOps to monitor models trained and deployed by DataRobot alongside custom models in the same console for a view into ALL the models you have in production.

Automated Time Series enhancements.​ We already have the industry’s most powerful time series forecasting product, and we continue to innovate with every release. In Release 5.3 we’ve added several new modeling and scaling strategies, including support for up to 1M series.

Many Other New Features and Enhancements. Release 5.3 is full of powerful new capabilities you’ll want to take advantage of immediately, including enhancements to our feature association matrix and feature impact charts, more efficient user management tools, integration with SQL Server, and much more.

Early release features

Every DataRobot release includes some early release features. If you’re a customer and interested in helping test early release features, contact your CFDS or account executive for details on how you can participate.

Here are some early release features you can check out:

  • Feature Lineage
  • Feature Discovery for primary dataset
  • Switching secondary dataset for predictions
  • Data Quality Assessment
  • SHAP Prediction Explanations via API
  • MLOps enhancements

Resources for this release

To help you understand the changes and guide you while using this release, we’ve put together the following resources.

Feature overview

Introducing Microsoft SQL Server Integration

Demo videos for many of the new features and changes

Replacing CSV in AI Catalog
Feature graphs in AI Catalog
Dataset version management in AI Catalog
Feature Discovery overview
Feature graph editor
Viewing Feature Lineage
(early release feature)
Feature Discovery for the primary dataset
(early release feature)
Switching to alternative secondary datasets
(early release feature)
Eureqa Regressor for Forecast Distance Modeling
Flexible backtesting for time series and OTV projects
Multiseries calendars for time series projects
Prediction intervals

Tell us what you think

If this material is helpful, let us know. If you still have questions about the release, let us know. Other DataRobot Community members and our DataRobot experts will help fill in any blanks, and your feedback is critical to helping us get better. Click Comment and let us know.

1 Comment
Image Sensor

Hi,

Thanks for uploading all these videos. They are really usefull to fully understand how it works.

Carlos Acosta

Announcements
Welcome to DataRobot Community! Explore, learn, and engage with your peers around all things AI and ML.

To learn more about the community, check out About Community.


Join the Fight with DataRobot

Our Research Center centralizes information relating to COVID-19 research and ongoing activity.