net.sourceforge.cilib.games.agent.neural
Class NeuralStateInputStrategy

java.lang.Object
  net.sourceforge.cilib.games.agent.neural.NeuralStateInputStrategy

Direct Known Subclasses:

PredatorPreyNeuralInputStrategy, TetrisFeaturesInputStrategy, TTTStateInputStrategy

public abstract class NeuralStateInputStrategy
extends Object

Author:

leo This class is used by the NeuralAgent to get an input vector based on a given GameState

Constructor Summary

NeuralStateInputStrategy()

Method Summary

abstract int	amountInputs() get the amount of inputs that the Neural Network should have based on this input strategy
abstract Vector	getNeuralInputArray(NeuralAgent currentPlayer, Game<GameState> state) Get an input vector for the given game state

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

NeuralStateInputStrategy

public NeuralStateInputStrategy()

Method Detail

amountInputs

public abstract int amountInputs()

get the amount of inputs that the Neural Network should have based on this input strategy

Returns:

the input count

getNeuralInputArray

public abstract Vector getNeuralInputArray(NeuralAgent currentPlayer,
                                           Game<GameState> state)

Get an input vector for the given game state

Parameters:

currentPlayer - the agent represented by the Neural Network

state - the current game state

Returns:

the input vector