/pyro-may2023

Meet Uber's Pyro - popular framework for probabilistic programming. Learn how to introduce regularization and prior assumptions into a model, at first for a simple use case of Bayesian Linear Regression and later in an introduction to deep generative models with Pyro.

Primary LanguageJupyter NotebookMIT LicenseMIT

An Introduction to Pyro

Level: Beginner

Presentation: Introduction to Pyro

Workshop description

In this workshop we will go through an introduction of the popular framework for probabilistic programming that is Uber's Pyro. Participants will learn how to introduce regularization and prior assumptions into a model, at first for a simple use case of Bayesian Linear Regression and later in an introduction to deep generative models with Pyro.

As Pyro is built on PyTorch, some prior knowledge of PyTorch can be useful. Feel free to check out the PyLadies' previous introduction to the topic: https://github.com/pyladiesams/deepLearningPyTorch-beginner-nov2022

Requirements

  • Python 3.8 or higher
  • Jupyter notebook or jupyter-lab
  • [Optional] graphviz for visualization of models
    • Can be installed e.g. on Ubuntu with sudo apt install graphviz

Usage

  • Clone the repository
  • Install the required dependencies with pip3 install -r requirements.txt

Video record

Re-watch this YouTube stream

Credits

This workshop was set up by @pyladiesams and GiuliaCaglia.