learningparametersdialog.java

java实现的隐马尔科夫模型

/* jahmm package - v0.3.1 */

/*
 *  Copyright (c) 2004, Jean-Marc Francois.
 *
 *  This file is part of Jahmm.
 *  Jahmm is free software; you can redistribute it and/or modify
 *  it under the terms of the GNU General Public License as published by
 *  the Free Software Foundation; either version 2 of the License, or
 *  (at your option) any later version.
 *
 *  Jahmm is distributed in the hope that it will be useful,
 *  but WITHOUT ANY WARRANTY; without even the implied warranty of
 *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *  GNU General Public License for more details.
 *
 *  You should have received a copy of the GNU General Public License
 *  along with Jahmm; if not, write to the Free Software
 *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
 */

package be.ac.ulg.montefiore.run.jahmm.apps;

import java.text.*;
import java.util.*;
import java.awt.*;
import java.awt.event.*;
import javax.swing.*;

import be.ac.ulg.montefiore.run.jahmm.*;


/*
 * Observation type (integer or vector) and dimension.
 */
public class LearningParametersDialog 
    extends JDialog implements ActionListener {

    final private JPanel mainPanel;
    final private JButton okButton;
    final private JButton cancelButton;
    final private JSpinner nbStatesSpinner;
    //final private JComboBox algorithmComboBox;
    private JComboBox opdfComboBox;
    
    private int nbStates;
    private int observationType = -1;
    private int observationDimension = -1;
    // Has the dialog been validated (OK button pressed)
    private boolean isValidated; 
    private String[] opdfFactoryDescription;
    private OpdfFactory[] opdfFactory;
    private int selectedFactory;
    
    private String[] opdfIntegerFactoryDescription = new String[] {
	"Discrete distribution"
    };
    private OpdfFactory[] opdfIntegerFactory = null;
    private String[] opdfVectorFactoryDescription = new String[] {
	"Multivariate gaussian distribution"
    };
    private OpdfFactory[] opdfVectorFactory = null;
    
    
    /*
     * Parameters for a learning algorithm
     */
    public class LearningParameters {
	public final int nbStates;
	public final OpdfFactory opdfFactory;
	
	public LearningParameters(int nbStates, OpdfFactory opdfFactory) {
	    this.nbStates = nbStates;
	    this.opdfFactory = opdfFactory;
	}
    }
    	
	
    public LearningParametersDialog() {
	super(new JFrame(), "Select learning parameters", true);
	
	nbStatesSpinner = new 
	    JSpinner(new SpinnerNumberModel(5, 1, Integer.MAX_VALUE, 1));
	
	JLabel messageLabel =
	    new JLabel("Choose a learning algorithm and its parameters.");
	
	opdfComboBox = new JComboBox(); // Dummy combo box

	JLabel spinnerLabel = new JLabel("Number of HMM states");
	spinnerLabel.setHorizontalAlignment(SwingConstants.RIGHT);
	spinnerLabel.setLabelFor(nbStatesSpinner);
	
	JLabel typesLabel = new JLabel("Observations type");
	typesLabel.setHorizontalAlignment(SwingConstants.RIGHT);
	typesLabel.setLabelFor(opdfComboBox);
	
	mainPanel = new JPanel(new GridLayout(2,2));
	mainPanel.add(typesLabel);
	mainPanel.add(opdfComboBox);
	mainPanel.add(spinnerLabel);
	mainPanel.add(nbStatesSpinner);
	
	okButton = new JButton("Ok");
	cancelButton = new JButton("Cancel");
	
	okButton.addActionListener(this);
	cancelButton.addActionListener(this);
	
	JPanel buttonsPanel = new JPanel(new GridLayout(1,1));
	buttonsPanel.add(okButton);
	buttonsPanel.add(cancelButton);
	
	getContentPane().add(messageLabel, BorderLayout.NORTH);
	getContentPane().add(mainPanel, BorderLayout.CENTER);
	getContentPane().add(buttonsPanel, BorderLayout.SOUTH);
	setDefaultCloseOperation(JDialog.HIDE_ON_CLOSE);
	setResizable(false);
    }

    
    public void setupObservationType(int observationType, 
				     int observationDimension) {
	if (this.observationType == observationType &&
	    this.observationDimension == observationDimension)
	    return; // Don't change the previous combo box selections
	
	this.observationType = observationType;
	this.observationDimension = observationDimension;	
	
	int i = Arrays.asList(mainPanel.getComponents()).indexOf(opdfComboBox);
	mainPanel.remove(i);
	
	if (observationType == ObservationSequences.INTEGER) {
	    opdfIntegerFactory = new OpdfFactory[] {
		new OpdfIntegerFactory(observationDimension)
	    };
	    opdfComboBox = new JComboBox(opdfIntegerFactoryDescription);
	    opdfFactoryDescription = opdfIntegerFactoryDescription;
	    opdfFactory = opdfIntegerFactory;
	} else { // Observations are real vectors
	    opdfVectorFactory = new OpdfFactory[] {
		new OpdfMultiGaussianFactory(observationDimension)
	    };
	    opdfComboBox = new JComboBox(opdfVectorFactoryDescription);
	    opdfFactoryDescription = opdfVectorFactoryDescription;
	    opdfFactory = opdfVectorFactory;
	}

	mainPanel.add(opdfComboBox, i);
	mainPanel.validate();
    }
    
    
    public LearningParameters parameters() {
	return new LearningParameters(nbStates, opdfFactory[selectedFactory]);
    }
    

    public boolean isValidated() {
	return isValidated;
    }
    
    
    /*
     * Ok/Cancel button pressed.
     */
    public void actionPerformed(ActionEvent e) {
	if (e.getSource().equals(okButton)) {
	    try {
		nbStatesSpinner.commitEdit();
	    }
	    catch(ParseException pe) {
		throw new RuntimeException("Internal error");
	    }
	    
	    isValidated = true;
	    nbStates = ((Integer) nbStatesSpinner.getValue()).intValue();
	    selectedFactory = Arrays.asList(opdfFactoryDescription).
		indexOf(opdfComboBox.getSelectedItem());
	}
	
	if (e.getSource().equals(cancelButton))
	    isValidated = false;
		
	setVisible(false);
    }
}