This tutorial introduces PyTorch on UVA’s High-Performance Computing (HPC) systems for researchers interested in deep learning and GPU-accelerated computing. Participants will learn the fundamentals of the PyTorch framework, including tensors, automatic differentiation, neural network construction, and model training workflows. The session also covers how to run PyTorch efficiently on UVA HPC resources using interactive GPU jobs, Slurm batch scripts, and containerized or module-based software environments. Attendees will gain hands-on experience building and training neural networks while learning best practices for scaling machine learning workloads on HPC infrastructure.

Familiarity with Python and basic Linux command-line usage is recommended. An active HPC account is required, though temporary access may be provided for limited instructional use.

The Introduction to HPC tutorial and familiarity with Python is recommended for attending.

Limited to UVA affiliates - must have NetBadge to register.