ACES: AI/ML on Intel PVC GPUs
Overview
Instructor: Zhenhua He
Time: Tuesday, October 15, 2024 — 1:30PM-4:00PM CT
Location: online using Zoom
Prerequisites: Current ACCESS ID; basic Linux/Unix skills; basic understanding of machine learning concepts, neural networks, and deep learning; familiarity with deep learning frameworks TensorFlow and/or PyTorch
This course provides an overview of Intel PVC GPUs, guidance on accessing these GPUs on the ACES cluster at Texas A&M High Performance Research Computing, and demonstrations of running AI/ML models with the GPUs using PyTorch and Tensorflow.
Registration will open up on this webpage the week before the class.
Course Materials
Presentation slides
The presentation slides will be available when the course starts.
Participation
Attendees will log in to the cluster during the training session in order to follow examples and complete exercises with the instructor's guidance.
Learning Objectives and Agenda
In this class, participants will:
- Access the Intel PVC GPUs on ACES cluster
- Learn how to run PyTorch and TensorFlow models with PVC GPUs.
- Learn to migrate a TensorFlow image classification model to PVC.
- Learn to migrate a PyTorch image classification model to PVC.
Topics covered include:
- Intro to Intel PVC
We will introduce Intel's PVC, its architecture, and the PVC GPUs on the TAMU ACES platform. - Demo on ACES
Students will learn how to run models of different frameworks with PVC GPUs on the ACES system. - TensorFlow on PVC
Students will learn how to convert a TensorFlow image classification model to run on a PVC GPU. - PyTorch on PVC
Students will learn how to convert a PyTorch image classification model to run on a PVC GPU.