In this tutorial we will review the basics of inference with probabilistic models, the more recent "deep" probabilistic models and how to implement them using the Pyro probabilistic programming library.
After that we will have a hands-on experience training probabilistic models to analyze time series from astronomical survey projects.
To install the dependencies I suggest to use conda:
conda env create -f environment.yml
And then launch
jupyter notebook
And navigate to tutorial.ipynb
You can also open this tutorial in google colab
For more on these topics see:
Author: Pablo Huijse, phuijse at inf dot uach dot cl
This tutorial was presented online at the IEEE Summer School on Computational Intelligence 2020, hosted by UFRO. For more activities organized by the IEEE Chile CIS chapter see: https://cis.ieeechile.cl/