A big data assignment, testing the feasibility of predicting alcohol use. Data used is from Riksmaten 2010-11 by Swedish food agency.
More on the data format here.
Column of interest is 'Alko' representing mean daily intake of alcohol in grams for observed inviduals.
Code available in Jupyter Notebook