DataPredict™ [Release 2.6] (Mature + Maintenance Mode) - Machine Learning, Deep Learning And Reinforcement Learning Library - 45+ Models!


Hmm, ive been training him for quite a while using DoubleQLearningV2 however for some reason he just doesnt want to correct his mistakes…


This is his BuildModel function.


His environmentVector.

image
His reward and respawning logic(he is penalised and respawned for touching sidewalks)

What’s wrong with this agent??