Details
Details
In this workshop, you get familiar with RAPIDS which runs on Python with pandas/scikit-learn-like API and you experience the speed-up it provides. We will also use GPU accelerated Xgboost to do Leave One Feature Out Importance in a reasonable time.
Prerequisites: Create a Kaggle account for accessing the data and notebooks
Prerequisites: Create a Kaggle account for accessing the data and notebooks
Prerequisites
Create a Kaggle account for using Kaggle Notebooks OR Install RAPIDS on a machine with GPU (https://rapids.ai/start.html)
Program
13:00: Introduction
13:05: Accelerate an end-to-end Machine Learning Pipeline using RAPIDS and GPU.
13:05: Accelerate an end-to-end Machine Learning Pipeline using RAPIDS and GPU.
13:50: Q&A
Schedule
11:30 am
Rapids - NVIDIA Workshop
In this workshop, you get familiar with RAPIDS which runs on Python with pandas/scikit-learn-like API and you experience the speed-up it provides. We will also use GPU accelerated Xgboost to do Leave One Feature Out Importance in a reasonable time.
Prerequisites: Create a Kaggle account for accessing the data and notebooks. Optional: Install RAPIDS on a conda environment.
PROGRAM
13:00: Introduction
13:05: Accelerate an end-to-end Machine Learning Pipeline using RAPIDS and GPU.
13:50: Q&A
Host
Ahmet Erdem
Machine Learning at NVIDIA