keel.Algorithms.Preprocess.NoiseFilters.ANR

Class ClassifierMLPerceptron

java.lang.Object

keel.Algorithms.Preprocess.NoiseFilters.ANR.ClassifierMLPerceptron

public class ClassifierMLPerceptron
extends java.lang.Object

Classification model by means of a multi-layered perceptron. This class is a wrapper for classification problems to solve using conjugated gradient algorithm.

Since:

JDK1.4

Version:

1.0

Author:

Written by Luciano Sánchez (University of Oviedo) 21/07/2005, Modified by Enrique A. de la Cal (University of Oviedo) 13/12/2008

Constructor Summary

Constructors

Constructor and Description
ClassifierMLPerceptron()

Method Summary

Modifier and Type	Method and Description
static void	createDatasets(java.lang.String trainIN, java.lang.String trainOUT, java.lang.String testIN, java.lang.String testOUT) It apllies the changes to remove the noise
static void	main(java.lang.String[] args) Method that calls the private wrapper method "neuralClassificationLS" that creates and runs a neural network for solving a classification problem using the Conjugated Gradient algorithm.

Methods inherited from class java.lang.Object

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

Constructor Detail

ClassifierMLPerceptron

public ClassifierMLPerceptron()

Method Detail

main

public static void main(java.lang.String[] args)

Method that calls the private wrapper method "neuralClassificationLS" that creates and runs a neural network for solving a classification problem using the Conjugated Gradient algorithm.

Parameters:

args - command line parameters with the name of configuration file with the information for classification process in position arg[0].

createDatasets

public static void createDatasets(java.lang.String trainIN,
                                  java.lang.String trainOUT,
                                  java.lang.String testIN,
                                  java.lang.String testOUT)

It apllies the changes to remove the noise

Parameters:

trainIN - Original training data file.

trainOUT - Modified training data file.

testIN - Original test data file.

testOUT - Modified test data file.