I feel like something wrong with agent's observations

[c0mb1ne]

I was inspecting what kind of observations agents gets, and found that in this line of code, we trying to normalize the number remaining enemies to add them to observation. And this variable never being used in this code anymore. https://github.com/Unity-Technologies/ml-agents-dodgeball-env/blob/b2915bd442f88bef391d8c380227f5be65bfae60/Assets/Dodgeball/Scripts/DodgeBallAgent.cs#L305

but here we just add the flat number of remaining enemies, which is kinda goes against ml-agents guidelines about normalizing observation for better perfomance. https://github.com/Unity-Technologies/ml-agents-dodgeball-env/blob/b2915bd442f88bef391d8c380227f5be65bfae60/Assets/Dodgeball/Scripts/DodgeBallAgent.cs#L314