Dirichlet Process GSOC 2021

Dear PyMC developers,

My name is Larry and I am currently a first year PhD student in biostatistics at the Dalla Lana School of Public Health at the University of Toronto. My current research interests revolve around causal inference and Bayesian methods, and I intend to my second year learning more about (Bayesian) dynamic treatment regimes (DTRs) and Bayesian nonparametric (BNP) methods. Broadly speaking, my goals in this PhD program are to TA graduate mathematical statistics, contribute to medical research (especially under a Bayesian framework) and encourage open science (through the use of Python).

I would love to participate in the Google Summer of Code program working on the truncated Dirichlet process project proposed here. I was hoping to reach out to potential mentors @fonnesbeck, @AustinRochford and (at)ferrine* to hear your opinions and discuss further about this potential opportunity.


Here’s a bit more about me!

*Tag was not able to go through because I am a new user

1 Like

Hey Larry! Thanks for reaching out. Let us know if you have specific questions. The student application period has just begun, so I would start pulling together your application, as well as becoming familiar with the PyMC3 code base (which is particularly challenging right now, as things are changing very quickly). We always recommend looking through the issue tracker to find a simple bug or enhancement to work on–its a great way to get to know the library.


Hi Dr. Fonnesbeck,

Thanks for writing back! While I have used PyMC3 previously, my understanding is that, as you said, I should familiarize myself with the code base and write a first draft of my proposal in the coming days. If I have any questions, I intend to write to you (via email); I also intend to share my first draft when available so that I can improve it before the deadline.

Thanks again!