Elocutionist Evenings: Turning PyMC3 into sklearn

Talk Topic

PyMC3 is a powerful library for building probabilistic models. However, most of the examples online exist inside Jupyter notebooks so the step of making your model reusable and production-ready is a bit opaque. My goal is to show one example of creating a custom Bayesian Model class that mimics the sklearn API. Specifically, I’ll show you how I built my own hierarchical linear model class. After this talk, you should be able to build your own reusable PyMC3 models.

Speaker Biography

Nicole is a Data Scientist at Shiftgig. Prior to that, she has also worked at the University of Chicago, the Exploratorium, and Automatic. She received a PhD in Physics from the University of California Berkeley. On the weekends, Nicole teaches middle school and high school girls how to program at the Girls Who Code club of the University of Illinois Chicago.

Building Access

This meeting will be held at Civis Analytics' offices at 200 West Monroe on the 22nd floor.

• Please have a photo ID ready to show when you enter the office.

Event Schedule

• 6:00pm to 6:30pm: Networking and Noms

• 6:30pm to 7:15pm: Speaker Presentation

• 7:15pm to 7:45pm: Question and Answer and Chat with the Speaker

Code of Conduct

To ensure the safety and comfort of attendees, we will be strictly following the NumFOCUS Code of Conduct (http://www.numfocus.org/code-of-conduct.html) at this event.

Questions, Comments, or Concerns

If you have any questions, comments, or concerns, please contact the organizer, Safia, at safia (at) safia (dot) rocks.

Your Talk Here

Interested in presenting your own talk at an Elocutionist Evening? Fill out this form (https://docs.google.com/forms/d/1smCh7CrMbtA34aJ2KFac3qC3golD0DMZBDjpmSyxIVc/viewform) and reach out to the organizer, Safia, at safia (at) safia (dot) rocks if you need any help with your abstract or presentation. Feel free to use this as an opportunity to practice a talk for a larger conference or just to share an interesting topic with your fellow PyDataneers!

Sponsors

https://a248.e.akamai.net/secure.meetupstatic.com/photos/event/c/4/f/d/600_457310429.jpeg

https://a248.e.akamai.net/secure.meetupstatic.com/photos/event/4/7/8/6/600_446778310.jpeg