aima.core.learning.neural

Class Perceptron

java.lang.Object

aima.core.learning.neural.Perceptron

All Implemented Interfaces:

FunctionApproximator

public class Perceptron
extends java.lang.Object
implements FunctionApproximator

Author:

Ravi Mohan, Mike Stampone

Constructor Summary

Constructors

Constructor and Description
Perceptron(int numberOfNeurons, int numberOfInputs)

Method Summary

Modifier and Type	Method and Description
Vector	predict(NNExample nne) Returns the outcome predicted for the specified example
void	processError(Vector error) Accept an error and change the parameters to accommodate it
Vector	processInput(Vector input) Returns the output values for the specified input values
int[]	testOnDataSet(NNDataSet nnds) Returns the accuracy of the hypothesis on the specified set of examples
void	trainOn(NNDataSet innds, int numberofEpochs) Induces the layer of this perceptron from the specified set of examples

Methods inherited from class java.lang.Object

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

Constructor Detail

Perceptron

public Perceptron(int numberOfNeurons,
                  int numberOfInputs)

Method Detail

processInput

public Vector processInput(Vector input)

Description copied from interface: FunctionApproximator

Returns the output values for the specified input values

Specified by:

processInput in interface FunctionApproximator

Parameters:

input - the input values

Returns:

the output values for the specified input values

processError

public void processError(Vector error)

Description copied from interface: FunctionApproximator

Accept an error and change the parameters to accommodate it

Specified by:

processError in interface FunctionApproximator

Parameters:

error - an error vector

trainOn

public void trainOn(NNDataSet innds,
                    int numberofEpochs)

Induces the layer of this perceptron from the specified set of examples

Parameters:

innds - a set of training examples for constructing the layer of this perceptron.

numberofEpochs - the number of training epochs to be used.

predict

public Vector predict(NNExample nne)

Returns the outcome predicted for the specified example

Parameters:

nne - an example

Returns:

the outcome predicted for the specified example

testOnDataSet

public int[] testOnDataSet(NNDataSet nnds)

Returns the accuracy of the hypothesis on the specified set of examples

Parameters:

nnds - the neural network data set to be tested on.

Returns:

the accuracy of the hypothesis on the specified set of examples