Tutorial on The PyTorch Library for Deep Learning

Have you been interested in training and using Deep Neural Networks (DNNs)? In this two-part tutorial presentation, we will present on what PyTorch is and how to use it, with a focus on live examples using Jupyter Notebooks. Though not mandatory, attendees are encouraged to bring their laptops. For more advanced topics, we provide an overview and links for attendees to learn more.  Zoom:  https://arizona.zoom.us/j/92580666693

Day I – Pytorch: The Fundamentals

1) What is PyTorch? (15 min)

-What is PyTorch? When, who and why.

-Comparison with Tensorflow, Keras, and other deep learning frameworks.

-Tensors and Data types 

-Similarities and Differences with NumPy

-Understanding documentation & source code

-GPU Computing in python with PyTorch

2) Neural Networks in PyTorch: Architecture and Implementation (15 min)

-Basic structure of a NN in PyTorch

-Classes in PyTorch: Modules, parameters, and optimizers, Oh my!

-Structure of tensors and other common conventions in PyTorch.

-Backpropagation basics- Saving and loading models

3) Convolutional Neural Networks (CNNs) (15 min)

-Structure of and basics of CNNs

-Types and use cases of convolutions

-Example of CNN and training- Caveats and tips for practical use