Expectation Maximization
This repo contains the code associated with the project described on my Blog here: Expectation Maximization.
The project is a short tutorial designed to help build intuition for how the expectation maximization algorithm works using a simple problem based on real world data. The data I used for the project was from the national longitudinal survey. In this repo, I've replaced this data with some manufactured data.
Files:
- exp_max_sim.py - python file that runs a simulation and animation of the EM algorithm
- sample_data.csv - some manufactured gender/height data
- EM_algorithm.key - a few slides associated with this project (please see my blog post for more information)