RAPIDS: Open-Source GPU Data Science

JOSHUA PATTERSON | DIRECTOR AI INFRASTRUCTURE AT NVIDIA & KEITH KRAUS | MANAGER AI INFRASTRUCTURE AT NVIDIA

Traditional machine-learning workloads have yet to be GPU-accelerated the way deep learning and other neural net methods have. RAPIDS aims to change that, while maintaining the ease of use of the PyData ecosystem. The goal is to build a ridiculously fast open-source platform that allows practitioners to explore data, train ML algorithms and build applications while primarily staying in GPU memory. This talk covers where RAPIDS is today, where it’s going, how to get started, and how attendees can contribute to the movement.

Видео RAPIDS: Open-Source GPU Data Science канала Anaconda, Inc.