keel.Algorithms.Symbolic_Regression.Shared

Class ParseFileRegSym

java.lang.Object

keel.Algorithms.Symbolic_Regression.Shared.ParseFileRegSym

public class ParseFileRegSym
extends java.lang.Object

This class obtains a symbolic model of training input data using algorithms: -GAP (Genetic Algorithm Programming). method symbolicRegressionFuzzyGAP. -SAP (Simulated Anneling Programming). method symbolicRegressionFuzzySAP.

Since:

JDK1.4

Version:

1.0

Author:

Written by Luciano Sánchez (University of Oviedo) 27/02/2007, Modified by Enrique A. de la Cal (University of Oviedo) 13/12/2008

Constructor Summary

Constructors

Constructor and Description
ParseFileRegSym()

Method Summary

Modifier and Type	Method and Description
static void	symbolicRegressionFuzzyGAP(int constType, boolean tty, ProcessConfig pc, Randomize rand) This method learns a symbolic model for training input data given in pc using the Genetic Algorithm Programming (GAP) paradigm.
static void	symbolicRegressionFuzzySAP(int constType, boolean tty, ProcessConfig pc, Randomize rand) This method learns a symbolic model for training input data given in pc using the Simulated Anneling Programming (SAP) paradigm.

Methods inherited from class java.lang.Object

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

Constructor Detail

ParseFileRegSym

public ParseFileRegSym()

Method Detail

symbolicRegressionFuzzyGAP

public static void symbolicRegressionFuzzyGAP(int constType,
                                              boolean tty,
                                              ProcessConfig pc,
                                              Randomize rand)

This method learns a symbolic model for training input data given in pc using the Genetic Algorithm Programming (GAP) paradigm. The method enables selecting the type of constants used in learning algorithm (Crisp or Fuzzy). Finally, it prints out the statistical results for train and test datasets.

Parameters:

constType - type of variables (FuzzyRegressor.Crisp or FuzzyRegressor.Fuzzy)

tty - unused boolean parameter, kept for compatibility

pc - ProcessConfig object to obtain the train and test datasets and the method's parameters.

rand - Random number generator

symbolicRegressionFuzzySAP

public static void symbolicRegressionFuzzySAP(int constType,
                                              boolean tty,
                                              ProcessConfig pc,
                                              Randomize rand)

This method learns a symbolic model for training input data given in pc using the Simulated Anneling Programming (SAP) paradigm. The method enables selecting the type of constants used in learning algorithm (Crisp or Fuzzy). Finally, it prints out the statistical results for train and test datasets.

Parameters:

constType - type of variables (FuzzyRegressor.Crisp or FuzzyRegressor.Fuzzy)

tty - unused boolean parameter, kept for compatibility

pc - ProcessConfig object to obtain the train and test datasets and the method's parameters.

rand - Random number generator