This repo contains my Python code written for the interview with Dataiku.
This excercise was based on a US census dataset containing 42 anonymised information about approx. 300,000 persons. The goal was to predict whether or not these people earns more than 50k$ per year.
The dataset was divided in two part: a training set containing approx. 200,000 rows and a testing set containing approx. 100,000 rows.