Wednesday 10:15 AM–12:00 PM in Tutorial Room

GPU Data Science with RAPIDS

Joshua Patterson, Keith Kraus

Audience level:


Join us as we walk you through a detailed introduction to RAPIDS, including reading files in from disk, munging data and creating features, and fitting models all on the GPU. We will conclude by talking about future plans for RAPIDS and how the community can get involved.


Python has seen terrific progress as the data science language of choice. With the introduction of Pandas, users could interact with data in python in a way that fells intuitive. Open-source packages such as Scikit-Learn have democratized and accelerated data science. RAPIDS seeks to have a similar impact on the Python data science community by accelerating data science with GPUs. RAPIDS is an open-source suite of tools for GPU data science. Launched in October, RAPIDS includes cuDF, a library for reading data to the GPU and interacting with it via an API with a familiar look and feel; and cuML, a library for machine learning.

Also, please join us after the final talks and tutorials on Wednesday, from 6:00-9:00, for a Meetup and happy hour hosted by NVIDIA. Join the RAPIDS team from NVIDIA and learn about updates on the RAPIDS OSS project, and share your data science experiences. Be part of conversations on GPU-accelerated algorithms, scalable deployment, and more. Conveniently hosted in the CIC. Appetizers and drinks will be provided. Register your attendance here -

Subscribe to Receive PyData Updates



Get Now