FeedForwardWeightedLayer

Overview

Package

Class

Use

Deprecated

Index

Home >> All >> AI >> [ NeuralNetworks overview ]

PREV CLASS NEXT CLASS

SUMMARY:

JAVADOC |

Sourc code of AI.NeuralNetworks.FeedForwardWeightedLayer

SOURCE |

Download AI.NeuralNetworks.FeedForwardWeightedLayer

DOWNLOAD | NESTED | FIELD | CONSTR | METHOD

DETAIL: FIELD | CONSTR | METHOD

AI.NeuralNetworks
Class FeedForwardWeightedLayer

java.lang.Object
  AI.NeuralNetworks.FeedForwardLayer
      AI.NeuralNetworks.FeedForwardWeightedLayer

Direct Known Subclasses:: FeedForwardWeightedTrainingLayer

public class FeedForwardWeightedLayer
extends FeedForwardLayer

This class is used to represent the inner and output layers of a multilayer feed forward neural network

Field Summary

private static float SEED_WIDTH


protected float[][] weights


Fields inherited from class AI.NeuralNetworks.FeedForwardLayer

activation

Constructor Summary

FeedForwardWeightedLayer(FeedForwardWeightedLayer layer)
          creates a new insatnce of FeedForwardWeightedLayer

FeedForwardWeightedLayer(float[][] weights)
          creates a new insatnce of FeedForwardWeightedLayer

FeedForwardWeightedLayer(int size, int previousSize)
          creates a new insatnce of FeedForwardWeightedLayer

Method Summary

protected float[][] getWeights()


void runLayer(FeedForwardLayer previousLayer)
          Calculates the new activation values of this layer based on the previous layer

protected static float sigmoid(float x)
          returns the value of the sigmoid function 1/(1+exp(-x))

java.lang.String toString()
          returns a String representation of the layer

Methods inherited from class AI.NeuralNetworks.FeedForwardLayer

getActivation, size

Methods inherited from class java.lang.Object

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

Field Detail

weights

protected float[][] weights

SEED_WIDTH

private static float SEED_WIDTH

Constructor Detail