DataRobot features an in-depth API that allows data scientists to produce fully automated workflows in their coding environment of choice. This accelerator shows how to pair the power of DataRobot with the Spark-backed notebook environment provided by Databricks.
In this notebook you'll see how data acquired and prepared in a Databricks notebook can be used to train a collection of models on DataRobot. You'll then deploy a recommended model and use DataRobot's exportable Scoring Code to generate predictions on the Databricks Spark cluster.
This accelerator notebook covers the following activities:
- Acquiring a training dataset.
- Building a new DataRobot project.
- Deploying a recommended model.
- Scoring via Spark using DataRobot's exportable Java Scoring Code.
- Scoring via DataRobot's Prediction API.
- Reporting monitoring data to DataRobot's MLOps agent framework.
- Writing results back to a new table.