Now Reading
All The Features Rolled Out In DataRobot 7.1

All The Features Rolled Out In DataRobot 7.1

  • The new version offers a no-code app builder to turn models into AI apps.

DataRobot is one of the most prominent players in Augmented Intelligence, which works to democratise data science with end-to-end automation to help build, deploy and manage ML models. The platform enables companies to leverage the benefits of utilising AI and seeks to maximise business value through delivering AI at scale and constantly optimising its performance levels. On that note, DataRobot recently released their second major release of 2021– the DataRobot version 7.1. 

What’s new?

A variety of new features have been made available with this update, the first of which is Automated AI Reports. These AI Reports have been designed to summarise key findings of a user’s modelling project for the perusal of shareholders in an easy-to-consume format. Through this, the tool provides a comprehensive summary of the project with accurate insights and cross-validation scores. Additionally, the Report also displays interpretability insights from a Feature Impact histogram and generates detailed explanations, performance metrics and ethics insights. 

Deep Learning DevCon 2021 | 23-24th Sep | Register>>

Second, DataRobot 7.1 will provide feature discovery push down integration for Snowflake. This will allow users to build features from their Snowflake data automatically. DataRobot’s Feature Discovery option will help streamline AI adoption across organisations by offering automated feature engineering and generating new features for ML models from multiple datasets. Integrating this with Snowflake will push this processing into the data cloud, making the all-new feature further cost-effective and accurate. 

Third, the new release offers a unique relationship quality assessment tool for feature discovery in DataRobot AutoML, allowing users to understand the quality and potential problems in relationships between primary and secondary datasets. Such issues include missing data and incorrect feature derivation windows. Working on this would mitigate the risk of bad relationships early on in the ML modelling process.

Fourth, the new DataRobot Automated Time Series will now provide automatic data preparation capabilities, letting DataRobot identify gaps in data and suggest ways in which the user could fill them. 

Looking for a job change? Let us help you.

Fifth, it will offer Nowcasting for time-aware models. Nowcasting is an approach in modelling that enables organisations to collect essential insights by estimating the present conditions of the target variable. Version 7.1 will allow users to access a more comprehensive range of blueprints and additional time-series settings for modelling processes.  

Sixth, DataRobot’s new version can also run forecasting models in Eureqa, a proprietary modelling engine created by Cornell’s AI Lab. These models revolve around the idea that a genetic algorithm can fit different analytic expressions in trained data and return a mathematical formula as an ML model. As per DataRobot, this is a fundamentally different approach to traditional supervised ML models but can reduce complexity and work well with small and large datasets. 

Seven, DataRobots has simplified creating, managing and monitoring prediction jobs as an MLOps administrator. Here, one can also see job histories and statuses directly on the MLOps user interface and access information on how their resources are being used and by whom. Also arriving in DataRobot 7.1 are MLOps management agents. These will help provide lifecycle management for remote AI and ML models and comprehend models’ state regardless of where they were created or where they might be running. These management agents can also automate tasks and can deploy or replace models directly in their environment.

Finally, the new version offers a no-code app builder to turn models into AI apps. This app builder can enter new data, perform what-if scenarios, run simulations, and automate various tasks using pre-built templates and drag-and-drop widgets. 

As per DataRobot, their objective is to drive better business outcomes and possibilities using AI. “We are in constant communication with our customers regarding the challenges they face when deploying AI,” said Nenshad Bardoliwalla, DataRobot Senior Vice President. He added, “Thus, (we) will tailor our updates based on their unique needs.” The AI-powered platform hopes to bring AI to business users, technical or not. Here’s hoping its new version provides enterprises with the tools they need to leverage real value from AI. 

What Do You Think?

Join Our Discord Server. Be part of an engaging online community. Join Here.

Subscribe to our Newsletter

Get the latest updates and relevant offers by sharing your email.

Copyright Analytics India Magazine Pvt Ltd

Scroll To Top