Chris Myers
Cornell Center for Advanced Computing

6/2023 (original)

Running PyTorch-based code on TACC systems requires building Python virtual environments containing the required packages, and configuring either interactive or batch-queue access to suitable TACC resources. We describe here what is required to run PyTorch codes on both the Frontera GPU nodes and the Frontera CPU-only CLX nodes.

Objectives

After you complete this segment, you should be able to:

  • Run a simple PyTorch program on TACC systems (assuming you have an allocation)
  • Understand how to configure your workflows to run your own PyTorch programs at TACC
Prerequisites

In order to run programs on TACC systems, you need an allocation for service units on whichever system(s) you choose to run on. In addition, if you are unfamiliar with some of the mechanics of working on TACC systems, you might want to consult the topic on "Deep Learning at TACC" and the links contained therein.

 
©  |   Cornell University    |   Center for Advanced Computing    |   Copyright Statement    |   Inclusivity Statement