Package libai.nn.supervised

Class Perceptron

java.lang.Object

libai.nn.NeuralNetwork

libai.nn.supervised.SupervisedLearning

libai.nn.supervised.Perceptron

All Implemented Interfaces:

java.io.Serializable

Direct Known Subclasses:

Adaline

public class Perceptron
extends SupervisedLearning

Perceptron is the first trainable neural network proposed. The network is formed by one matrix (Weights) and one vector (Bias). The output for the network is calculated by O = sign(W * pattern + b).

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field	Description
protected Column	b
protected int	ins
protected int	outs
protected static Sign	signum
protected Matrix	W

Fields inherited from class libai.nn.NeuralNetwork

plotter, progress, random

Constructor Summary

Constructors

Constructor	Description
Perceptron(int in, int out)	Constructor.
Perceptron(int in, int out, java.util.Random rand)	Constructor.

Method Summary

Modifier and Type	Method	Description
Matrix	getWeights()
Column	simulate(Column p)	Calculates the output for the pattern.
void	simulate(Column pattern, Column result)	Calculate the output for the pattern and left the result on result.
void	train(Column[] patterns, Column[] answers, double alpha, int epochs, int offset, int length, double minerror)	Train the perceptron using the standard update rule: W = W + alpha.e.pattern^t b = b + alpha.e

Methods inherited from class libai.nn.NeuralNetwork

error, error, euclideanDistance2, euclideanDistance2, gaussian, getDefaultRandomGenerator, getPlotter, getProgressBar, initializeProgressBar, open, open, open, save, setPlotter, setProgressBar, train, train

Methods inherited from class java.lang.Object

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

Methods inherited from class libai.nn.supervised.SupervisedLearning

validatePreconditions

Field Detail

signum

protected static final Sign signum

W

protected Matrix W

b

protected Column b

ins

protected final int ins

outs

protected final int outs

Constructor Detail

Perceptron

public Perceptron(int in,
                  int out)

Constructor.

Parameters:

in - Number of inputs for the network = number of elements in the patterns.

out - Number of outputs for the network.

Perceptron

public Perceptron(int in,
                  int out,
                  java.util.Random rand)

Constructor.

Parameters:

in - Number of inputs for the network = number of elements in the patterns.

out - Number of outputs for the network.

rand - Random generator used for creating matrices

Method Detail

train

public void train(Column[] patterns,
                  Column[] answers,
                  double alpha,
                  int epochs,
                  int offset,
                  int length,
                  double minerror)

Train the perceptron using the standard update rule:
W = W + alpha.e.pattern^t
b = b + alpha.e

Specified by:

train in class NeuralNetwork

Parameters:

patterns - The patterns to be learned.

answers - The expected answers.

alpha - The learning rate.

epochs - The maximum number of iterations

offset - The first pattern position

length - How many patterns will be used.

minerror - The minimal error expected.

simulate

public Column simulate(Column p)

Description copied from class: NeuralNetwork

Calculates the output for the pattern.

Specified by:

simulate in class NeuralNetwork

Parameters:

p - Pattern to use as input.

Returns:

The output for the neural network.

simulate

public void simulate(Column pattern,
                     Column result)

Calculate the output for the pattern and left the result on result. result = signum(W * pattern + b)

Specified by:

simulate in class NeuralNetwork

Parameters:

pattern - The input pattern

result - The output result.

getWeights

public Matrix getWeights()