C# Class numl.Reinforcement.ReinforcementModel

Reinforcement model.

Inheritance: Supervised.Model, IReinforcementModel

Afficher le fichier Open project: sethjuarez/numl

Méthodes publiques

Méthode	Description
Learn ( Vector x1, double y, Vector x2, double r ) : void	Reinforces the model from the new State, Action, StateP and Reward.
Learn ( Vector x, double y, double r ) : void	Reinforces the model from the new state, action and reward.
Learn ( object state, object stateP ) : void	Reinforces the model from the new item and reward.

Reinforces the model from the new State, Action, StateP and Reward.

public abstract Learn ( Vector x1, double y, Vector x2, double r ) : void
x1	numl.Math.LinearAlgebra.Vector	Item features, i.e. the State.
y	double	Action label.
x2	numl.Math.LinearAlgebra.Vector	State/action reward value.
r	double	Reward value.
Résultat	void

Reinforces the model from the new state, action and reward.

public abstract Learn ( Vector x, double y, double r ) : void
x	numl.Math.LinearAlgebra.Vector	Item features, i.e. the State.
y	double	Action label.
r	double	Reward value.
Résultat	void

Reinforces the model from the new item and reward.

public Learn ( object state, object stateP ) : void
state	object	Initial State object or features with action label.
stateP	object	New State object or features with reward label.
Résultat	void