kmeansoptpanel.java

一个用于排队系统仿真的开源软件,有非常形象的图象仿真过程!

package jmt.engine.jwat.workloadAnalysis.clustering.kMean.panels;

import java.awt.BorderLayout;
import java.awt.FlowLayout;
import java.awt.GridLayout;
import java.awt.event.ActionEvent;
import java.awt.event.KeyEvent;

import javax.swing.AbstractAction;
import javax.swing.Action;
import javax.swing.Box;
import javax.swing.ButtonGroup;
import javax.swing.JLabel;
import javax.swing.JOptionPane;
import javax.swing.JPanel;
import javax.swing.JRadioButton;
import javax.swing.JSpinner;
import javax.swing.KeyStroke;
import javax.swing.SpinnerNumberModel;
import javax.swing.border.EtchedBorder;
import javax.swing.border.TitledBorder;

import jmt.engine.jwat.ProgressStatusListener;
import jmt.engine.jwat.VariableNumber;
import jmt.engine.jwat.workloadAnalysis.clustering.kMean.MainKMean;
import jmt.engine.jwat.workloadAnalysis.panels.ClusterPanel;
import jmt.engine.jwat.workloadAnalysis.utils.ModelWorkloadAnalysis;
import jmt.engine.jwat.workloadAnalysis.wizard.WorkloadAnalysisWizard;
import jmt.gui.common.CommonConstants;
import jmt.gui.common.resources.ImageLoader;
import jmt.gui.jwat.JWatWizard;
import jmt.gui.jwat.input.ProgressMonitorShow;
import jmt.gui.wizard.WizardPanel;

//UPDATE 29/10/2006: + aggiunto pannello per selezione trasformazione da applicare alle variabili durange K-Means

public class KMeansOptPanel extends JPanel implements CommonConstants{
	
	private ClusterPanel parent;
	//private final String numClustDescription = HTML_START + HTML_FONT_NORM + 
	//"Specify the maximum number of clusters to be generated" + HTML_FONT_NOR_END + HTML_END;
	//private final String iterationDescription = HTML_START + HTML_FONT_NORM +
	//"Specify the maximum number of iterations for the algorithm" + HTML_FONT_NOR_END + HTML_END;
	private final String KMEANS_DESCRIPTION = HTML_START
		+ HTML_FONT_TITLE + "k-Means Clustering" + HTML_FONT_TIT_END + HTML_FONT_NORM
		+ "Specify maximum number of clusters and maximum number of iterations" + HTML_FONT_NOR_END
		+ HTML_END;
	private final String KMEANS_CLUSTER_OPT = HTML_START + HTML_FONT_NORM
	+ "Select maximum number of clusters that the algorithm has to produce." + HTML_FONT_NOR_END
	+ HTML_END;
	private final String KMEANS_CLUSTER_ITER = HTML_START + HTML_FONT_NORM
	+ "Select maximum number of iterations that the algorithm has to perform to find solutions." + HTML_FONT_NOR_END
	+ HTML_END;
	private final String KMEANS_OPTIONS_DESCRIPTION = HTML_START
	+ HTML_FONT_TITLE + "Transformations" + HTML_FONT_TIT_END + HTML_FONT_NORM
	+ "Specify the transformation to apply<p>" +
	  "to selected variables considered<p>" +
	  "by the algorithm." + HTML_FONT_NOR_END
	+ HTML_END;
	
	//private JSlider numClust;
	//private JSlider numIteration;
	private ProgressStatusListener lst;
	private ModelWorkloadAnalysis model;
	private JRadioButton noneT;
	private JRadioButton minmaxT;
	private JRadioButton stdDevT;
	private JSpinner numOfClust;
	private JSpinner numOfIter;
		
	public KMeansOptPanel(WizardPanel parent,ProgressStatusListener lst,ModelWorkloadAnalysis model)
	{
		this.model=model;
		this.lst=lst;
		this.parent=(ClusterPanel)parent;
		setSolveAction();
		initPanel();
	}
	
	private void initPanel(){
		this.setLayout(new BorderLayout(10,60));
		//UPDATE 13 Nov 2006
		JPanel northPanel = new JPanel(new BorderLayout());
		northPanel.add(Box.createHorizontalStrut(5),BorderLayout.EAST);
		northPanel.add(new JLabel(KMEANS_DESCRIPTION),BorderLayout.CENTER);
		this.add(northPanel,BorderLayout.NORTH);
		
		JPanel centralPanel = new JPanel(new BorderLayout());
		//NORTH Panel options
		JPanel northCPanel = new JPanel(new GridLayout(1,2));
		
		JPanel clusterOpt = new JPanel(new BorderLayout());
		clusterOpt.setBorder(new TitledBorder(new TitledBorder(
				new EtchedBorder(EtchedBorder.LOWERED), "Clusters")));
		
		JPanel clustNorth = new JPanel(new FlowLayout(FlowLayout.LEFT));
		clustNorth.add(new JLabel("number of clusters: "));
		numOfClust = new JSpinner(new SpinnerNumberModel(3,2,37,1));
		clustNorth.add(numOfClust);
		clusterOpt.add(clustNorth,BorderLayout.NORTH);
		clusterOpt.add(new JLabel(KMEANS_CLUSTER_OPT),BorderLayout.CENTER);
		
		JPanel iterationOpt = new JPanel(new BorderLayout());
		iterationOpt.setBorder(new TitledBorder(new TitledBorder(
				new EtchedBorder(EtchedBorder.LOWERED), "Clustering iteration")));
		JPanel iterNorth = new JPanel(new FlowLayout(FlowLayout.LEFT));
		iterNorth.add(new JLabel("number of iterations: "));
		numOfIter = new JSpinner(new SpinnerNumberModel(4,2,50,1));
		iterNorth.add(numOfIter);
		iterationOpt.add(iterNorth,BorderLayout.NORTH);
		iterationOpt.add(new JLabel(KMEANS_CLUSTER_ITER),BorderLayout.CENTER);
		
		northCPanel.add(clusterOpt);
		northCPanel.add(iterationOpt);
		
		centralPanel.add(northCPanel,BorderLayout.NORTH);
		
		JPanel transf = new JPanel(new BorderLayout());		//Terzo panel
		transf.setBorder(new TitledBorder(new TitledBorder(
				new EtchedBorder(EtchedBorder.LOWERED), "Transformations")));
		JPanel options = new JPanel(new GridLayout(3,1));	//Panel ratio buttons
		noneT = new JRadioButton("none");
		minmaxT = new JRadioButton("(value - min) / (max - min)");
		stdDevT = new JRadioButton("(value - mean) / std. dev.");
		ButtonGroup op = new ButtonGroup();
		op.add(noneT);
		minmaxT.setSelected(true);
		op.add(minmaxT);
		op.add(stdDevT);
		options.add(noneT);
		options.add(minmaxT);
		options.add(stdDevT);
		transf.add(options,BorderLayout.EAST);
		transf.add(new JLabel(KMEANS_OPTIONS_DESCRIPTION),BorderLayout.WEST);
		
		centralPanel.add(transf,BorderLayout.SOUTH);
		
		this.add(centralPanel,BorderLayout.CENTER);
		/*
		JPanel center = new JPanel(new GridLayout(3,1));
		
		JPanel numClustP=new JPanel();
		numClustP.setLayout(new BorderLayout(0,0));
		numClustP.setBorder(new TitledBorder(new TitledBorder(
				new EtchedBorder(EtchedBorder.LOWERED), "Cluster options")));
		numClustP.add(new JLabel(numClustDescription),BorderLayout.NORTH);
		numClust=new JSlider(2,20);
		numClust.setMajorTickSpacing(1);
		numClust.setPaintLabels(true);
		numClust.setPaintTicks(true);
		numClust.setSnapToTicks(true);
		numClustP.add(numClust,BorderLayout.CENTER);
		center.add(numClustP);
		
		JPanel iterationP=new JPanel();
		iterationP.setLayout(new BorderLayout(0,0));
		iterationP.add(new JLabel(iterationDescription),BorderLayout.NORTH);
		iterationP.setBorder(new TitledBorder(new TitledBorder(
				new EtchedBorder(EtchedBorder.LOWERED), "Clustering iteration options")));
		numIteration=new JSlider(1,20);
		numIteration.setMajorTickSpacing(1);
		numIteration.setPaintLabels(true);
		numIteration.setPaintTicks(true);
		numIteration.setSnapToTicks(true);
		iterationP.add(numIteration,BorderLayout.CENTER);
		center.add(iterationP);
		
		JPanel transf = new JPanel(new BorderLayout());	//Terzo panel
		JPanel options = new JPanel(new GridLayout(3,1));					//Panel ratio buttons
		noneT = new JRadioButton("none");
		minmaxT = new JRadioButton("(value - min) / (max - min)");
		stdDevT = new JRadioButton("(value - mean) / std. dev.");
		ButtonGroup op = new ButtonGroup();
		op.add(noneT);
		noneT.setSelected(true);
		op.add(minmaxT);
		op.add(stdDevT);
		options.add(noneT);
		options.add(minmaxT);
		options.add(stdDevT);
		transf.add(options,BorderLayout.WEST);
		transf.add(new JLabel(KMEANS_OPTIONS_DESCRIPTION),BorderLayout.EAST);

		center.add(transf);
		
		this.add(new JLabel(KMEANS_DESCRIPTION),BorderLayout.NORTH);
		this.add(center,BorderLayout.CENTER);
		*/
	}
	
	private void setSolveAction(){
		//if(model.getListOfClustering().size() > 0) ((JWatWizard)parent.getParentWizard()).setEnableButton("Next >",true);
		//else ((JWatWizard)parent.getParentWizard()).setEnableButton("Next >",false);
		((JWatWizard)parent.getParentWizard()).setEnableButton("Solve",true);	
		((WorkloadAnalysisWizard)parent.getParentWizard()).setActionButton("Solve",new AbstractAction(){
			public void actionPerformed(ActionEvent e) {
				if(parent.getVarSelected().length >= 2){
					short trasf = VariableNumber.NONE;
					if(minmaxT.isSelected()) trasf = VariableNumber.MINMAX;
					if(stdDevT.isSelected()) trasf = VariableNumber.STDEV;
					MainKMean eng=new MainKMean(
							new ProgressMonitorShow(parent,"Processing...",1),
							model.getMatrix(),
							parent.getVarSelected(),
							//numClust.getValue(),
							//numIteration.getValue(),
							((Integer)numOfClust.getValue()).intValue(),
							((Integer)numOfIter.getValue()).intValue(),
							trasf
					);
					eng.addStatusListener(lst);
					eng.start();
					((JWatWizard)parent.getParentWizard()).setEnableButton("Solve",false);
				}
				else{
					JOptionPane.showMessageDialog(parent,"Select at least two variables to proceed with clustering","Warning",JOptionPane.WARNING_MESSAGE);
				}
			}
		});
		((WorkloadAnalysisWizard)parent.getParentWizard()).setActionTool(new AbstractAction(){
			{
	            putValue(Action.SHORT_DESCRIPTION, "Clusterize");
	            putValue(Action.SMALL_ICON, ImageLoader.loadImage("Sim"));

	            putValue(Action.ACCELERATOR_KEY,
	                    KeyStroke.getKeyStroke(KeyEvent.VK_L,
	                            ActionEvent.CTRL_MASK));
	            putValue(Action.MNEMONIC_KEY, new Integer(KeyEvent.VK_L));				
			}
			public void actionPerformed(ActionEvent e) {
				if(parent.getVarSelected().length >= 2){
					short trasf = VariableNumber.NONE;
					if(minmaxT.isSelected()) trasf = VariableNumber.MINMAX;
					if(stdDevT.isSelected()) trasf = VariableNumber.STDEV;
					MainKMean eng=new MainKMean(
							new ProgressMonitorShow(parent,"Processing...",1),
							model.getMatrix(),
							parent.getVarSelected(),
							//numClust.getValue(),
							//numIteration.getValue(),
							((Integer)numOfClust.getValue()).intValue(),
							((Integer)numOfIter.getValue()).intValue(),
							trasf
					);
					eng.addStatusListener(lst);
					eng.start();
					((JWatWizard)parent.getParentWizard()).setEnableButton("Solve",false);
				}
				else{
					JOptionPane.showMessageDialog(parent,"Select at least two variables to proceed with clustering","Warning",JOptionPane.WARNING_MESSAGE);
				}
			}
		});
	}	
}