Or more broadly: how do you train a machine learning model to capture the nuances of entirely different, conflicting rule sets? I built a XGBoost classification pipeline to answer that. To stress-test the architecture across heterogeneous environments, I applied it to a highly debated and popular hy
