Ryan
Recompiled a better trained model based on improved learning
6f76bb4