Important
As of April 2026, this repository is no longer maintained; some of the code developed for the data science blueprints program was ultimately released or reused in other contexts, while other NVIDIA projects (like NVIDIA AI Blueprints and DGX Spark Playbooks) are actively maintained and provide substantial examples of AI techniques and end-to-end applications.
This repository contains an example of modeling customer churn, from federating data and performing exploratory query-based analytics to feature engineering, model training, and model operationalization. The data engineering portions of the blueprint are accelerated with the RAPIDS Accelerator for Apache Spark and the machine learning portions are accelerated with the RAPIDS Python libraries. To learn more about this blueprint and these technologies, you can read our ebook.
Copyright © 2020–2026 NVIDIA Corporation.
This code is distributed under the Apache License, Version 2.0.