This repository will look at using a Artificial Neural Network to help determine the phase space variables to be used in a Q-Factor analysis. We will develop a classifier to separate b1 events and a0/a2 events. At the last layer, say the softmax layer, there should be strong separation of the two types of events. We can pass this directly to q-factors to be used as part of the phase space to calculate distances under. This will do our feature extraction for us