org.neuroph.nnet

Class RBFNetwork

java.lang.Object

org.neuroph.core.NeuralNetwork

org.neuroph.nnet.RBFNetwork

All Implemented Interfaces:

java.io.Serializable

public class RBFNetwork
extends NeuralNetwork

Radial basis function neural network. TODO: learning for rbf layer: k-means clustering weights between input and rbf layer are Ci vector each weight is a component of a Ci vector Ci are centroids of the clusters trained by k means clustering Each neuron in rbf layer corresponds to a single cluster neuronns in rbf layer are clusters

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
RBFNetwork(int inputNeuronsCount, int rbfNeuronsCount, int outputNeuronsCount) Creates new RBFNetwork with specified number of neurons in input, rbf and output layer

Method Summary

Methods inherited from class org.neuroph.core.NeuralNetwork

addLayer, addLayer, addListener, addPlugin, calculate, createConnection, createFromFile, createFromFile, fireNetworkEvent, getInputNeurons, getInputsCount, getLabel, getLayerAt, getLayers, getLayersCount, getLearningRule, getNetworkType, getOutput, getOutputNeurons, getOutputsCount, getPlugin, getWeights, indexOf, isEmpty, learn, learn, load, load, pauseLearning, randomizeWeights, randomizeWeights, randomizeWeights, randomizeWeights, removeLayer, removeLayerAt, removeListener, removePlugin, reset, resumeLearning, save, setInput, setInputNeurons, setLabel, setLearningRule, setNetworkType, setOutputLabels, setOutputNeurons, setWeights, stopLearning, toString

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, wait, wait, wait

Constructor Detail

RBFNetwork

public RBFNetwork(int inputNeuronsCount,
                  int rbfNeuronsCount,
                  int outputNeuronsCount)

Creates new RBFNetwork with specified number of neurons in input, rbf and output layer

Parameters:

inputNeuronsCount - number of neurons in input layer

rbfNeuronsCount - number of neurons in rbf layer

outputNeuronsCount - number of neurons in output layer