Package one.empty3.neuralnetwork

Class HiddenNeuron

java.lang.Object

one.empty3.neuralnetwork.Neuron

one.empty3.neuralnetwork.HiddenNeuron

All Implemented Interfaces:

Comparable

public class HiddenNeuron
extends Neuron

Field Summary

Fields inherited from class one.empty3.neuralnetwork.Neuron

bias, input, output

Constructor Summary

Constructors

Constructor	Description
HiddenNeuron(int length)

Method Summary

Modifier and Type	Method	Description
double[]	gradient(double dw)	Brut calculus
double[]	learn(double dw, double e, double n)
double[]	updateDescend(double e, double dw)

Methods inherited from class one.empty3.neuralnetwork.Neuron

activation, compareTo, compute, error, error, function, getActivationFunction, getActivationMethod, getBias, getInput, getLayer, getLength, getNetwork, getOutput, getW, initW, setActivationFunction, setActivationMethod, setBias, setInput, setInputImage, setLayer, setNetwork, setOutput, setW, sigmoid, updateW

Methods inherited from class java.lang.Object

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

Constructor Details

HiddenNeuron

public HiddenNeuron(int length)

Method Details

gradient

public double[] gradient(double dw)

Brut calculus

Parameters:

dw -

Returns:

updateDescend

public double[] updateDescend(double e,
 double dw)

learn

public double[] learn(double dw,
 double e,
 double n)