UF Research Computing

Training module on Pytorch for image classification

This training module is based off of the Deep Learning with Pytorch: A 60 minute blitz tutorial by Soumith Chintala.

It has been updated and customized for running on HiPerGator by Matt Gitzendanner.

Contents:

This training module has four notebooks:

• Introduction to PyTorch and Tensors
• Automatic Differentiation, Backpropagation and Gradients with autograd
• Neural Networks and Loss functions
• Training a CNN Classifier for image recognition

You can run these notebooks in JupyterHub on HiPerGator or open them in the Google Colaboratory.