Data Science in 30 Minutes: Accelerating Data Science Workflows with Bartley Richardson
GPUs constructed on CUDA have been used for deep-learning and different functions for a very long time. However, if you have a look at information scientists and so they work they’re doing, CUDA doesn’t actually match effectively into their workflow. As we speak’s scientists need fast exploration, fast outcomes and to have the ability to shift gears with out interrupting their prepare of thought. They wish to suppose on the velocity of knowledge.
On this webinar, Bartley Richardson, PhD, a former fellow of The Knowledge Incubator and a senior information scientist at Nvidia, addresses this problem. Richardson shares Nvidia RAPIDS mission, an open-source suite of knowledge processing and machine studying libraries that permits GPU acceleration for information science workflows. It additionally delivers a 50- to 100-times enchancment over conventional GPU processing, however nonetheless utilizing the identical code and following the APIs that information scientists are accustomed to (e.g., Pandas, SciKit).