/*
* This program 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.
*
* This program 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 this program; if not, write to the Free Software
* Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*/
/*
* Clusterer.java
* Copyright (C) 2004 University of Waikato, Hamilton, New Zealand
*
*/
package weka.gui.beans;
import weka.clusterers.EM;
import weka.core.Instances;
import weka.filters.Filter;
import weka.filters.unsupervised.attribute.Remove;
import weka.gui.Logger;
import weka.gui.ExtensionFileFilter;
import java.awt.BorderLayout;
import java.beans.EventSetDescriptor;
import java.io.*;
import java.util.Enumeration;
import java.util.Hashtable;
import java.util.Vector;
import javax.swing.JPanel;
import javax.swing.JOptionPane;
import javax.swing.JFileChooser;
import javax.swing.JPanel;
/**
* Bean that wraps around weka.clusterers
*
* @author <a href="mailto:mutter@cs.waikato.ac.nz">Stefan Mutter</a>
* @version $Revision: 1.10 $
* @see JPanel
* @see BeanCommon
* @see Visible
* @see WekaWrapper
* @see Serializable
* @see UserRequestAcceptor
* @see TrainingSetListener
* @see TestSetListener
*/
public class Clusterer
extends JPanel
implements BeanCommon, Visible, WekaWrapper, EventConstraints,
UserRequestAcceptor, TrainingSetListener, TestSetListener{
/** for serialization */
private static final long serialVersionUID = 7729795159836843810L;
protected BeanVisual m_visual =
new BeanVisual("Clusterer",
BeanVisual.ICON_PATH+"EM.gif",
BeanVisual.ICON_PATH+"EM_animated.gif");
private static int IDLE = 0;
private static int BUILDING_MODEL = 1;
private static int CLUSTERING = 2;
private int m_state = IDLE;
private Thread m_buildThread = null;
/**
* Global info for the wrapped classifier (if it exists).
*/
protected String m_globalInfo;
/**
* Objects talking to us
*/
private Hashtable m_listenees = new Hashtable();
/**
* Objects listening for batch clusterer events
*/
private Vector m_batchClustererListeners = new Vector();
/**
* Objects listening for graph events
*/
private Vector m_graphListeners = new Vector();
/**
* Objects listening for text events
*/
private Vector m_textListeners = new Vector();
/**
* Holds training instances for batch training.
*/
private Instances m_trainingSet;
private transient Instances m_testingSet;
private weka.clusterers.Clusterer m_Clusterer = new EM();
private transient Logger m_log = null;
private Double m_dummy = new Double(0.0);
private transient JFileChooser m_fileChooser = null;
/**
* Global info (if it exists) for the wrapped classifier
*
* @return the global info
*/
public String globalInfo() {
return m_globalInfo;
}
/**
* Creates a new <code>Clusterer</code> instance.
*/
public Clusterer() {
setLayout(new BorderLayout());
add(m_visual, BorderLayout.CENTER);
setClusterer(m_Clusterer);
}
/**
* Set a custom (descriptive) name for this bean
*
* @param name the name to use
*/
public void setCustomName(String name) {
m_visual.setText(name);
}
/**
* Get the custom (descriptive) name for this bean (if one has been set)
*
* @return the custom name (or the default name)
*/
public String getCustomName() {
return m_visual.getText();
}
/**
* Set the clusterer for this wrapper
*
* @param c a <code>weka.clusterers.Clusterer</code> value
*/
public void setClusterer(weka.clusterers.Clusterer c) {
boolean loadImages = true;
if (c.getClass().getName().
compareTo(m_Clusterer.getClass().getName()) == 0) {
loadImages = false;
} else {
// clusterer has changed so any batch training status is now
// invalid
m_trainingSet = null;
}
m_Clusterer = c;
String clustererName = c.getClass().toString();
clustererName = clustererName.substring(clustererName.
lastIndexOf('.')+1,
clustererName.length());
if (loadImages) {
if (!m_visual.loadIcons(BeanVisual.ICON_PATH+clustererName+".gif",
BeanVisual.ICON_PATH+clustererName+"_animated.gif")) {
useDefaultVisual();
}
}
m_visual.setText(clustererName);
// get global info
m_globalInfo = KnowledgeFlowApp.getGlobalInfo(m_Clusterer);
}
/**
* Returns true if this clusterer has an incoming connection that is
* a batch set of instances
*
* @return a <code>boolean</code> value
*/
public boolean hasIncomingBatchInstances() {
if (m_listenees.size() == 0) {
return false;
}
if (m_listenees.containsKey("trainingSet") ||
m_listenees.containsKey("testSet") ||
m_listenees.containsKey("dataSet")) {
return true;
}
return false;
}
/**
* Get the clusterer currently set for this wrapper
*
* @return a <code>weka.clusterers.Clusterer</code> value
*/
public weka.clusterers.Clusterer getClusterer() {
return m_Clusterer;
}
/**
* Sets the algorithm (clusterer) for this bean
*
* @param algorithm an <code>Object</code> value
* @exception IllegalArgumentException if an error occurs
*/
public void setWrappedAlgorithm(Object algorithm)
{
if (!(algorithm instanceof weka.clusterers.Clusterer)) {
throw new IllegalArgumentException(algorithm.getClass()+" : incorrect "
+"type of algorithm (Clusterer)");
}
setClusterer((weka.clusterers.Clusterer)algorithm);
}
/**
* Returns the wrapped clusterer
*
* @return an <code>Object</code> value
*/
public Object getWrappedAlgorithm() {
return getClusterer();
}
/**
* Accepts a training set and builds batch clusterer
*
* @param e a <code>TrainingSetEvent</code> value
*/
public void acceptTrainingSet(final TrainingSetEvent e) {
if (e.isStructureOnly()) {
// no need to build a clusterer, instead just generate a dummy
// BatchClustererEvent in order to pass on instance structure to
// any listeners
BatchClustererEvent ce =
new BatchClustererEvent(this, m_Clusterer,
new DataSetEvent(this, e.getTrainingSet()),
e.getSetNumber(), e.getMaxSetNumber(),1);
notifyBatchClustererListeners(ce);
return;
}
if (m_buildThread == null) {
try {
if (m_state == IDLE) {
synchronized (this) {
m_state = BUILDING_MODEL;
}
m_trainingSet = e.getTrainingSet();
final String oldText = m_visual.getText();
m_buildThread = new Thread() {
public void run() {
try {
if (m_trainingSet != null) {
m_visual.setAnimated();
m_visual.setText("Building clusters...");
if (m_log != null) {
m_log.statusMessage("Clusterer : building clusters...");
}
buildClusterer();
if(m_batchClustererListeners.size() > 0){
BatchClustererEvent ce =
new BatchClustererEvent(this, m_Clusterer,
new DataSetEvent(this, e.getTrainingSet()),
e.getSetNumber(), e.getMaxSetNumber(),1);
notifyBatchClustererListeners(ce);
}
if (m_Clusterer instanceof weka.core.Drawable &&
m_graphListeners.size() > 0) {
String grphString =
((weka.core.Drawable)m_Clusterer).graph();
int grphType = ((weka.core.Drawable)m_Clusterer).graphType();
String grphTitle = m_Clusterer.getClass().getName();
grphTitle = grphTitle.substring(grphTitle.
lastIndexOf('.')+1,
grphTitle.length());
grphTitle = "Set " + e.getSetNumber() + " ("
+e.getTrainingSet().relationName() + ") "
+grphTitle;
GraphEvent ge = new GraphEvent(Clusterer.this,
grphString,
grphTitle,
grphType);
notifyGraphListeners(ge);
}
if (m_textListeners.size() > 0) {
String modelString = m_Clusterer.toString();
String titleString = m_Clusterer.getClass().getName();
titleString = titleString.
substring(titleString.lastIndexOf('.') + 1,
titleString.length());
modelString = "=== Clusterer model ===\n\n" +
"Scheme: " +titleString+"\n" +
"Relation: " + m_trainingSet.relationName() +
((e.getMaxSetNumber() > 1)
? "\nTraining Fold: "+e.getSetNumber()
:"")
+ "\n\n"
+ modelString;
titleString = "Model: " + titleString;
TextEvent nt = new TextEvent(Clusterer.this,
modelString,
titleString);
notifyTextListeners(nt);
}
}
} catch (Exception ex) {
ex.printStackTrace();
} finally {
m_visual.setText(oldText);
m_visual.setStatic();
m_state = IDLE;
if (isInterrupted()) {
// prevent any clusterer events from being fired
m_trainingSet = null;
if (m_log != null) {
m_log.logMessage("Build clusterer interrupted!");
m_log.statusMessage("OK");
}
} else {
// save header
m_trainingSet = new Instances(m_trainingSet, 0);
}
if (m_log != null) {
m_log.statusMessage("Interrupted");
}
block(false);
}
}
};
m_buildThread.setPriority(Thread.MIN_PRIORITY);
m_buildThread.start();
// make sure the thread is still running before we block
// if (m_buildThread.isAlive()) {
block(true);
// }
m_buildThread = null;
m_state = IDLE;
}
} catch (Exception ex) {
ex.printStackTrace();
}
}
}
/**
* Accepts a test set for a batch trained clusterer
*
* @param e a <code>TestSetEvent</code> value
*/
public void acceptTestSet(TestSetEvent e) {
if (m_trainingSet != null) {
try {
if (m_state == IDLE) {
synchronized(this) {
m_state = CLUSTERING;
}
m_testingSet = e.getTestSet();
if (m_trainingSet.equalHeaders(m_testingSet)) {
BatchClustererEvent ce =
new BatchClustererEvent(this, m_Clusterer,
new DataSetEvent(this, e.getTestSet()),
e.getSetNumber(), e.getMaxSetNumber(),0);
notifyBatchClustererListeners(ce);
}
m_state = IDLE;
}
} catch (Exception ex) {
ex.printStackTrace();
}
}
}
/**
* Builds the clusters
*/
private void buildClusterer() throws Exception {
if(m_trainingSet.classIndex() < 0)
m_Clusterer.buildClusterer(m_trainingSet);
else{ //class based evaluation if class attribute is set
Remove removeClass = new Remove();
removeClass.setAttributeIndices(""+(m_trainingSet.classIndex()+1));
removeClass.setInvertSelection(false);
removeClass.setInputFormat(m_trainingSet);
Instances clusterTrain = Filter.useFilter(m_trainingSet, removeClass);
m_Clusterer.buildClusterer(clusterTrain);
}
}
/**
* Sets the visual appearance of this wrapper bean
*
* @param newVisual a <code>BeanVisual</code> value
*/
public void setVisual(BeanVisual newVisual) {
m_visual = newVisual;
}
/**
* Gets the visual appearance of this wrapper bean
*/
public BeanVisual getVisual() {
return m_visual;
}
/**
* Use the default visual appearance for this bean
*/
public void useDefaultVisual() {
m_visual.loadIcons(BeanVisual.ICON_PATH+"DefaultClusterer.gif",
BeanVisual.ICON_PATH+"DefaultClusterer_animated.gif");
}
/**
* Add a batch clusterer listener
*
* @param cl a <code>BatchClustererListener</code> value
*/
public synchronized void
addBatchClustererListener(BatchClustererListener cl) {
m_batchClustererListeners.addElement(cl);
}
/**
* Remove a batch clusterer listener
*
* @param cl a <code>BatchClustererListener</code> value
*/
public synchronized void
removeBatchClustererListener(BatchClustererListener cl) {
m_batchClustererListeners.remove(cl);
}
/**
* Notify all batch clusterer listeners of a batch clusterer event
*
* @param ce a <code>BatchClustererEvent</code> value
*/
private void notifyBatchClustererListeners(BatchClustererEvent ce) {
Vector l;
synchronized (this) {
l = (Vector)m_batchClustererListeners.clone();
}
if (l.size() > 0) {
for(int i = 0; i < l.size(); i++) {
((BatchClustererListener)l.elementAt(i)).acceptClusterer(ce);
}
}
}
/**
* Add a graph listener
*
* @param cl a <code>GraphListener</code> value
*/
public synchronized void addGraphListener(GraphListener cl) {
m_graphListeners.addElement(cl);
}
/**
* Remove a graph listener
*
* @param cl a <code>GraphListener</code> value
*/
public synchronized void removeGraphListener(GraphListener cl) {
m_graphListeners.remove(cl);
}
/**
* Notify all graph listeners of a graph event
*
* @param ge a <code>GraphEvent</code> value
*/
private void notifyGraphListeners(GraphEvent ge) {
Vector l;
synchronized (this) {
l = (Vector)m_graphListeners.clone();
}
if (l.size() > 0) {
for(int i = 0; i < l.size(); i++) {
((GraphListener)l.elementAt(i)).acceptGraph(ge);
}
}
}
/**
* Add a text listener
*
* @param cl a <code>TextListener</code> value
*/
public synchronized void addTextListener(TextListener cl) {
m_textListeners.addElement(cl);
}
/**
* Remove a text listener
*
* @param cl a <code>TextListener</code> value
*/
public synchronized void removeTextListener(TextListener cl) {
m_textListeners.remove(cl);
}
/**
* Notify all text listeners of a text event
*
* @param ge a <code>TextEvent</code> value
*/
private void notifyTextListeners(TextEvent ge) {
Vector l;
synchronized (this) {
l = (Vector)m_textListeners.clone();
}
if (l.size() > 0) {
for(int i = 0; i < l.size(); i++) {
((TextListener)l.elementAt(i)).acceptText(ge);
}
}
}
/**
* Returns true if, at this time,
* the object will accept a connection with respect to the named event
*
* @param eventName the event
* @return true if the object will accept a connection
*/
public boolean connectionAllowed(String eventName) {
/* if (eventName.compareTo("instance") == 0) {
if (!(m_Clusterer instanceof weka.classifiers.UpdateableClassifier)) {
return false;
}
} */
if (m_listenees.containsKey(eventName)) {
return false;
}
return true;
}
/**
* Returns true if, at this time,
* the object will accept a connection according to the supplied
* EventSetDescriptor
*
* @param esd the EventSetDescriptor
* @return true if the object will accept a connection
*/
public boolean connectionAllowed(EventSetDescriptor esd) {
return connectionAllowed(esd.getName());
}
/**
* Notify this object that it has been registered as a listener with
* a source with respect to the named event
*
* @param eventName the event
* @param source the source with which this object has been registered as
* a listener
*/
public synchronized void connectionNotification(String eventName,
Object source) {
if (connectionAllowed(eventName)) {
m_listenees.put(eventName, source);
/* if (eventName.compareTo("instance") == 0) {
startIncrementalHandler();
} */
}
}
/**
* Notify this object that it has been deregistered as a listener with
* a source with respect to the supplied event name
*
* @param eventName the event
* @param source the source with which this object has been registered as
* a listener
*/
public synchronized void disconnectionNotification(String eventName,
Object source) {
m_listenees.remove(eventName);
}
/**
* Function used to stop code that calls acceptTrainingSet. This is
* needed as clusterer construction is performed inside a separate
* thread of execution.
*
* @param tf a <code>boolean</code> value
*/
private synchronized void block(boolean tf) {
if (tf) {
try {
// only block if thread is still doing something useful!
if (m_buildThread.isAlive() && m_state != IDLE) {
wait();
}
} catch (InterruptedException ex) {
}
} else {
notifyAll();
}
}
/**
* Stop any clusterer action
*/
public void stop() {
// tell all listenees (upstream beans) to stop
Enumeration en = m_listenees.keys();
while (en.hasMoreElements()) {
Object tempO = m_listenees.get(en.nextElement());
if (tempO instanceof BeanCommon) {
((BeanCommon)tempO).stop();
}
}
// stop the build thread
if (m_buildThread != null) {
m_buildThread.interrupt();
m_buildThread.stop();
m_buildThread = null;
m_visual.setStatic();
}
}
/**
* Set a logger
*
* @param logger a <code>Logger</code> value
*/
public void setLog(Logger logger) {
m_log = logger;
}
public void saveModel() {
try {
if (m_fileChooser == null) {
// i.e. after de-serialization
m_fileChooser =
new JFileChooser(new File(System.getProperty("user.dir")));
ExtensionFileFilter ef = new ExtensionFileFilter("model", "Serialized weka clusterer");
m_fileChooser.setFileFilter(ef);
}
int returnVal = m_fileChooser.showSaveDialog(this);
if (returnVal == JFileChooser.APPROVE_OPTION) {
File saveTo = m_fileChooser.getSelectedFile();
String fn = saveTo.getAbsolutePath();
if (!fn.endsWith(".model")) {
fn += ".model";
saveTo = new File(fn);
}
ObjectOutputStream os =
new ObjectOutputStream(new BufferedOutputStream(
new FileOutputStream(saveTo)));
os.writeObject(m_Clusterer);
if (m_trainingSet != null) {
Instances header = new Instances(m_trainingSet, 0);
os.writeObject(header);
}
os.close();
if (m_log != null) {
m_log.logMessage("Saved clusterer OK.");
}
}
} catch (Exception ex) {
JOptionPane.showMessageDialog(Clusterer.this,
"Problem saving clusterer.\n",
"Save Model",
JOptionPane.ERROR_MESSAGE);
if (m_log != null) {
m_log.logMessage("Problem saving clusterer. " + ex.getMessage());
}
}
}
public void loadModel() {
try {
if (m_fileChooser == null) {
// i.e. after de-serialization
m_fileChooser =
new JFileChooser(new File(System.getProperty("user.dir")));
ExtensionFileFilter ef = new ExtensionFileFilter("model", "Serialized weka clusterer");
m_fileChooser.setFileFilter(ef);
}
int returnVal = m_fileChooser.showOpenDialog(this);
if (returnVal == JFileChooser.APPROVE_OPTION) {
File loadFrom = m_fileChooser.getSelectedFile();
ObjectInputStream is =
new ObjectInputStream(new BufferedInputStream(
new FileInputStream(loadFrom)));
// try and read the model
weka.clusterers.Clusterer temp = (weka.clusterers.Clusterer)is.readObject();
// Update name and icon
setClusterer(temp);
// try and read the header (if present)
try {
m_trainingSet = (Instances)is.readObject();
} catch (Exception ex) {
// quietly ignore
}
is.close();
if (m_log != null) {
m_log.logMessage("Loaded clusterer: "
+ m_Clusterer.getClass().toString());
}
}
} catch (Exception ex) {
JOptionPane.showMessageDialog(Clusterer.this,
"Problem loading classifier.\n",
"Load Model",
JOptionPane.ERROR_MESSAGE);
if (m_log != null) {
m_log.logMessage("Problem loading classifier. " + ex.getMessage());
}
}
}
/**
* Return an enumeration of requests that can be made by the user
*
* @return an <code>Enumeration</code> value
*/
public Enumeration enumerateRequests() {
Vector newVector = new Vector(0);
if (m_buildThread != null) {
newVector.addElement("Stop");
}
if (m_buildThread == null &&
m_Clusterer != null) {
newVector.addElement("Save model");
}
if (m_buildThread == null) {
newVector.addElement("Load model");
}
return newVector.elements();
}
/**
* Perform a particular request
*
* @param request the request to perform
* @exception IllegalArgumentException if an error occurs
*/
public void performRequest(String request) {
if (request.compareTo("Stop") == 0) {
stop();
} else if (request.compareTo("Save model") == 0) {
saveModel();
} else if (request.compareTo("Load model") == 0) {
loadModel();
} else {
throw new IllegalArgumentException(request
+ " not supported (Clusterer)");
}
}
/**
* Returns true, if at the current time, the event described by the
* supplied event descriptor could be generated.
*
* @param esd an <code>EventSetDescriptor</code> value
* @return a <code>boolean</code> value
*/
public boolean eventGeneratable(EventSetDescriptor esd) {
String eventName = esd.getName();
return eventGeneratable(eventName);
}
/**
* Returns true, if at the current time, the named event could
* be generated. Assumes that the supplied event name is
* an event that could be generated by this bean
*
* @param eventName the name of the event in question
* @return true if the named event could be generated at this point in
* time
*/
public boolean eventGeneratable(String eventName) {
if (eventName.compareTo("graph") == 0) {
// can't generate a GraphEvent if clusterer is not drawable
if (!(m_Clusterer instanceof weka.core.Drawable)) {
return false;
}
// need to have a training set before the clusterer
// can generate a graph!
if (!m_listenees.containsKey("trainingSet")) {
return false;
}
// Source needs to be able to generate a trainingSet
// before we can generate a graph
Object source = m_listenees.get("trainingSet");
if (source instanceof EventConstraints) {
if (!((EventConstraints)source).eventGeneratable("trainingSet")) {
return false;
}
}
}
if (eventName.compareTo("batchClusterer") == 0) {
if (!m_listenees.containsKey("trainingSet")) {
return false;
}
Object source = m_listenees.get("trainingSet");
if (source != null && source instanceof EventConstraints) {
if (!((EventConstraints)source).eventGeneratable("trainingSet")) {
return false;
}
}
}
if (eventName.compareTo("text") == 0) {
if (!m_listenees.containsKey("trainingSet")){
return false;
}
Object source = m_listenees.get("trainingSet");
if (source != null && source instanceof EventConstraints) {
if (!((EventConstraints)source).eventGeneratable("trainingSet")) {
return false;
}
}
}
if (eventName.compareTo("batchClassifier") == 0)
return false;
if (eventName.compareTo("incrementalClassifier") == 0)
return false;
return true;
}
}