I have put the datasets for each Dichotomy model together, so that two models aren't needed to create the effect. This model has also been trained with more steps.