Analytics provides a simple, interactive, UI-driven approach to machine learning. It provides a seamless, automated interface for users to easily develop, train, test, evaluate and deploy their machine learning models. It reduces the need for ad-hoc custom tooling and promotes reusability and collaboration.

Go to Documentation

Analytics features

UI driven data wrangler and cleansing

Seamless, integrated experience from data preparation and cleansing to model development, evaluation and deployment.

Support for popular ML libraries

Out of the box support for common ML libraries such as SparkML

Scoring plugins for running predictions

Built in scoring plugins take you from model development to running predictions on data in a few seconds.

Model evaluation

Integrated metrics and visualization that provides rich summaries and graphs for evaluating model performance.

Automated training and test data split

Automated splitting into training and test datasets reduces the need for custom tooling.

Hyperparameter tuning

Switches and knobs for advanced users to tune model performance using hyperparameters

Model Lifecycle Management

Automated model lifecycle management from deployment to promotion and retiring.


Rapid Time to Value with CDAP: Enterprise-Ready Machine Learning in under Three Minutes

#BDAM: Machine Learning for the Masses, by Albert Shau, Cask