A machine learning tool for automated prediction engineering. It allows you to easily structure prediction problems and generate labels for supervised learning.
This is the official implementation of the bipartite matching experiment from the paper "Learning Randomly Perturbed Structured Predictors for Direct Loss Minimization".
In this repo, I attempt to understand systemic biases in gender and racial policing as well as train an sklearn ML model for predicting the ethnicity of NYPD officers from allegation data.