Navigate uncertainty with AI-powered forecasting
Customer behavior and needs have changed dramatically. As a result, businesses are becoming more agile to keep up with changes to identify new opportunities that meet customer needs. Identifying trends in data helps to anticipate, which is why companies rely more on forecasting. But forecasting remains complex and laborious. It requires manual updating of data and adjustments to forecast outputs. These steps can delay decisions, preventing businesses from responding immediately to new demand patterns and market changes.
AI-powered forecasting enables organizations to respond to changes faster and make the right decisions. A broader set of data scientists leverage DataRobot AI Cloud advances AI forecasting to combine automation with best-in-class modeling techniques to streamline forecasting. Now you can experiment faster, build models for new segments without sacrificing accuracy, and most importantly, operationalize models in a few clicks completely out-of-the-box.
Learn more about what’s new in DataRobot AI Cloud.
Time Series Clustering for Segmented Modeling
Discover new segments and build models much faster and with better accuracy.
Create a clustering model to define segments for segmented modeling. Enhance the efficiency of modeling a high volume of series without sacrificing the accuracy of your predictions. Now you can quickly explore new segments, build models, and shorten the time from data to value. Clustering for segmented modeling is available as a public preview feature.
Segmented Modeling Deployments
Operationalize segmented models with one-click deployment.
To fully leverage the value of segmented modeling, you can now deploy combined models the same way you deploy any other time series models. After selecting the champion model for each segment, deploy the combined model to bring predictions into production. Creating a deployment allows you to use DataRobot MLOps for accuracy monitoring, prediction intervals, and challenger models.
No more model factories and additional costs from modeling a high volume of series. Now, just one click deploys for segmented models in the DataRobot AI Cloud platform.
Multiclass Support for No-Code AI Apps
Enable your frontline decision-makers with powerful and more granular insights.
With multiclass support for No-Code AI Apps you can build an app on top of the multiclass model and expand the number of use cases you can tackle with AI. Predict which product the customer will purchase next or which customers are most likely to respond to new ads.
In multiclass classification, each record belongs to one or more classes. The algorithm’s goal is to construct a function that correctly identifies the class into which the latest data point falls, given a new data point. Combining the power of multiclass models with the No-Code AI Apps enables users, especially business analysts, to take their deployed models a step further. Your team can effortlessly explore how to leverage those insights via the Optimization for a preselected prediction class, dig into what-if scenarios, and share insights easily via customizable chart widgets.
MLOps Management Agent
Use the Management Agent to take advantage of automated deployment and monitoring of models, ensuring your machine learning pipeline is healthy and reliable. This release introduces usability improvements to the Management Agent, including deployment status reporting, deployment relaunch, and the option to force the deletion of a Management Agent deployment.
The MLOps Management Agent provides a standard mechanism for automating model deployments in any type of environment or infrastructure. The Management Agent supports models trained on DataRobot or models trained with open-source tools on external infrastructure. The agent, accessed from the DataRobot application, ships with an assortment of example plugins that support custom configurations.
DataRobot AI Cloud – June 2022 Release Full Feature list
For the full details of features included in the DataRobot AI Cloud June 2022 Release, visit the DataRobot Documentation Release Center.