Build Probabilistic Machine Learning Apps Using Streamlit: A Case Study of COVID-19 Reproduction Metric
Meetup.com: https://www.meetup.com/Bayesian-Data-Science-DC/events/270623940
Date: May 27, 2020
Place: Online Webinar
Speaker: Daniel Emaasit
- Slides HTML
for the code.
- Code Repository
Rapidly build and deploy probabilistic machine learning Apps, using Streamlit’s open-source app framework. Make your Apps easily accessible to end users who don't have Bayesian expertise & without having to go through long product-release cycles. As a case study, we shall use PyMC3 to build a probabilistic model that estimates the reproduction metric of COVID-19. Then we shall quickly build an App using Streamlit. All in pure Python.
Only a laptop with a modern web browser like Google chrome or Mozilla Firefox. Click this icon 
Daniel Emaasit is a Data Scientist at Haystax Technology. His interests involve developing principled probabilistic models for problems where training data are scarce by leveraging knowledge from subject-matter experts and context information. In particular, he interested in flexible probabilistic machine learning methods, such as Gaussian processes and Dirichlet processes, and data-efficient learning methods such as Bayesian optimization and model-based reinforcement learning.
Daniel is also a Ph.D. Candidate of Transportation Engineering at UNLV where his research in nonparametric Bayesian methods is focused on developing flexible-statistical models for traveler-behavior analytics.