A small script utilising scikit-learn for predicting the size of a build of Congenica on-premise given the number of genome samples to process.