Discover groundbreaking advancements in accelerated computing for popular data science tools in this 21-minute video presentation. Explore how RAPIDS cuDF's new pandas acceleration mode offers up to 150x faster performance with GPUs, and learn about NetworkX graph algorithms achieving up to 600x faster speeds using RAPIDS cuGraph. Gain insights into accelerated data processing and vector search for large language model pipelines, and understand how enterprises can optimize machine learning workflows. Delve into topics such as GPU-acceleration for pandas, NetworkX, and Apache Spark MLlib, end-to-end accelerated data science, powering LLMs with accelerated data science, and production-grade software for AI. Access additional resources, including blog posts and Google Colab notebooks, to further enhance your understanding of these cutting-edge developments in the accelerated data science ecosystem.
GPU-Acceleration for Pandas, NetworkX, and Apache Spark MLlib - Accelerating Data Science Tools
Overview
Syllabus
- Introduction
– Accelerated Computing for Modern Data Demands
– End-to-end Accelerated Data Science
– Announcing GPU-acceleration for pandas, NetworkX, and Apache Spark MLlib with zero code changes
– Powering LLMs with accelerated data science
– Production-grade Software for AI
- End-to-end open data science platform
Taught by
NVIDIA Developer
