activedecorate.java

wekaUT是 university texas austin 开发的基于weka的半指导学习(semi supervised learning)的分类器

     * @return true if debugging output is on
     */
    public boolean getDebug() {
	return m_Debug;
    }
    
    /**
     * Set the base classifier for Decorate.
     *
     * @param newClassifier the Classifier to use.
     */
    public void setClassifier(Classifier newClassifier) {
	m_Classifier = newClassifier;
    }

    /**
     * Get the classifier used as the base classifier
     *
     * @return the classifier used as the classifier
     */
    public Classifier getClassifier() {
	return m_Classifier;
    }

    /**
     * Factor that determines number of artificial examples to generate.
     *
     * @return factor that determines number of artificial examples to generate
     */
    public double getArtificialSize() {
	return m_ArtSize;
    }
  
    /**
     * Sets factor that determines number of artificial examples to generate.
     *
     * @param newwArtSize factor that determines number of artificial examples to generate
     */
    public void setArtificialSize(double newArtSize) {
	m_ArtSize = newArtSize;
    }
    
    /**
     * Gets the desired size of the committee.
     *
     * @return the desired size of the committee
     */
    public int getDesiredSize() {
	return m_DesiredSize;
    }
    
    /**
     * Sets the desired size of the committee.
     *
     * @param newDesiredSize the desired size of the committee
     */
    public void setDesiredSize(int newDesiredSize) {
	m_DesiredSize = newDesiredSize;
    }
    
    /**
     * Sets the max number of Decorate iterations to run.
     *
     * @param numIterations  max number of Decorate iterations to run
     */
    public void setNumIterations(int numIterations) {
	m_NumIterations = numIterations;
    }

    /**
     * Gets the max number of Decorate iterations to run.
     *
     * @return the  max number of Decorate iterations to run
     */
    public int getNumIterations() {
        return m_NumIterations;
    }
    
    /**
     * Set the seed for random number generator.
     *
     * @param seed the random number seed 
     */
    public void setSeed(int seed) {
	m_Seed = seed;
	if(m_Seed==-1){
	    m_Random = new Random();
	}else{
	    m_Random = new Random(m_Seed);
	}
    }
    
    /**
     * Gets the seed for the random number generator.
     *
     * @return the seed for the random number generator
     */
    public int getSeed() {
        return m_Seed;
    }

    /**
     * Build Decorate classifier
     *
     * @param data the training data to be used for generating the classifier
     * @exception Exception if the classifier could not be built successfully
     */
    public void buildClassifier(Instances data) throws Exception {
	if(m_Classifier == null) {
	    throw new Exception("A base classifier has not been specified!");
	}
	if(data.checkForStringAttributes()) {
	    throw new UnsupportedAttributeTypeException("Cannot handle string attributes!");
	}
	if(data.classAttribute().isNumeric()) {
	    throw new UnsupportedClassTypeException("Decorate can't handle a numeric class!");
	}
	if(m_NumIterations < m_DesiredSize)
	    throw new Exception("Max number of iterations must be >= desired ensemble size!");
	
	trainSelectionCommittee(data);


	int i = 1;//current committee size
	int numTrials = 1;//number of Decorate iterations 
	Instances divData = new Instances(data);//local copy of data - diversity data
	Instances artData = null;//artificial data

	//compute number of artficial instances to add at each iteration
	int artSize = (int) (Math.abs(m_ArtSize)*data.numInstances());
	if(artSize==0) artSize=1;//atleast add one random example
	computeStats(data);//Compute training data stats for creating artificial examples
	
	//initialize new committee
	m_Committee = new Vector();
	Classifier newClassifier = m_Classifier;
	newClassifier.buildClassifier(divData);
	m_Committee.add(newClassifier);
	double eComm = computeError(divData);//compute ensemble error
	//if(m_Debug) System.out.println("Initialize:\tClassifier "+i+" added to ensemble. Ensemble error = "+eComm);
	
	//repeat till desired committee size is reached OR the max number of iterations is exceeded 
	while(i<m_DesiredSize && numTrials<m_NumIterations){
	    //Generate artificial training examples
	    artData = generateArtificialData(artSize, data);
	    
	    //Label artificial examples
	    labelData(artData);
	    
	    //Remove all the artificial data from the previous step (if any)
	    if(divData.numInstances() > data.numInstances()) {
		removeInstances(divData, artSize);
	    }
	    addInstances(divData, artData);//Add new artificial data
	    
	    //Build new classifier
	    Classifier tmp[] = Classifier.makeCopies(m_Classifier,1);
	    newClassifier = tmp[0]; 
	    newClassifier.buildClassifier(divData);
	    
	    //Test if the new classifier should be added to the ensemble
	    m_Committee.add(newClassifier);//add new classifier to current committee
	    double currError = computeError(data);
	    if(currError <= eComm){//adding the new member did not increase the error
		i++;
		eComm = currError;
		//if(m_Debug) System.out.println("Iteration: "+(1+numTrials)+"\tClassifier "+i+" added to ensemble. Ensemble error = "+eComm);
	    }else{//reject the current classifier because it increased the ensemble error 
		m_Committee.removeElementAt(m_Committee.size()-1);//pop the last member
	    }
	    numTrials++;
	}
    }

    //Train alternate ensemble method for use in selection
    protected void trainSelectionCommittee(Instances data) throws Exception{
    	if(m_SelectionScheme==BAGGING){
	    if(m_SelectionCommittee==null){//initialize Bagging
		System.out.println("Initializing Bagging...");
		m_SelectionCommittee = new Bagging();
		((Bagging)m_SelectionCommittee).setClassifier(getClassifier());
		((Bagging)m_SelectionCommittee).setSeed(getSeed());
		((Bagging)m_SelectionCommittee).setNumIterations(getDesiredSize());
		((Bagging)m_SelectionCommittee).setBagSizePercent(100);
	    }
	    m_SelectionCommittee.buildClassifier(data);
	}else if(m_SelectionScheme==BOOSTING){
	    if(m_SelectionCommittee==null){//initialize Boosting
		System.out.println("Initializing AdaBoost...");
		m_SelectionCommittee = new AdaBoostM1();
		((AdaBoostM1)m_SelectionCommittee).setClassifier(getClassifier());
		((AdaBoostM1)m_SelectionCommittee).setSeed(getSeed());
		((AdaBoostM1)m_SelectionCommittee).setMaxIterations(getDesiredSize());
	    }
	    m_SelectionCommittee.buildClassifier(data);
	}
    }

    
    /** 
     * Compute and store statistics required for generating artificial data.
     *
     * @param data training instances
     * @exception Exception if statistics could not be calculated successfully
     */
    protected void computeStats(Instances data) throws Exception{
	int numAttributes = data.numAttributes();
	m_AttributeStats = new Vector(numAttributes);//use to map attributes to their stats
	
	for(int j=0; j<numAttributes; j++){
	    if(data.attribute(j).isNominal()){
		//Compute the probability of occurence of each distinct value 
		int []nomCounts = (data.attributeStats(j)).nominalCounts;
		double []counts = new double[nomCounts.length];
		if(counts.length < 2) throw new Exception("Nominal attribute has less than two distinct values!"); 
		//Perform Laplace smoothing
		for(int i=0; i<counts.length; i++)
		    counts[i] = nomCounts[i] + 1;
		Utils.normalize(counts);
		double []stats = new double[counts.length - 1];
		stats[0] = counts[0];
		//Calculate cumulative probabilities
		for(int i=1; i<stats.length; i++)
		    stats[i] = stats[i-1] + counts[i];
		m_AttributeStats.add(j,stats);
	    }else if(data.attribute(j).isNumeric()){
		//Get mean and standard deviation from the training data
		double []stats = new double[2];
		stats[0] = data.meanOrMode(j);
		stats[1] = Math.sqrt(data.variance(j));
		m_AttributeStats.add(j,stats);
	    }else System.err.println("Decorate can only handle numeric and nominal values.");
	}
    }

    /**
     * Generate artificial training examples.
     * @param artSize size of examples set to create
     * @param data training data
     * @return the set of unlabeled artificial examples
     */
    protected Instances generateArtificialData(int artSize, Instances data){
	int numAttributes = data.numAttributes();
	Instances artData = new Instances(data, artSize);
	double []att; 
	Instance artInstance;
	
	for(int i=0; i<artSize; i++){
	    att = new double[numAttributes];
	    for(int j=0; j<numAttributes; j++){
		if(data.attribute(j).isNominal()){
		    //Select nominal value based on the frequency of occurence in the training data  
		    double []stats = (double [])m_AttributeStats.get(j);
		    att[j] =  (double) selectIndexProbabilistically(stats);
		}
		else if(data.attribute(j).isNumeric()){
		    //Generate numeric value from the Guassian distribution 
		    //defined by the mean and std dev of the attribute
		    double []stats = (double [])m_AttributeStats.get(j);
		    att[j] = (m_Random.nextGaussian()*stats[1])+stats[0];
		}else System.err.println("Decorate can only handle numeric and nominal values.");
	    }
	    artInstance = new Instance(1.0, att);
	    artData.add(artInstance);
	}
	return artData;
    }
    
    
    /** 
     * Labels the artificially generated data.
     *
     * @param artData the artificially generated instances
     * @exception Exception if instances cannot be labeled successfully 
     */
    protected void labelData(Instances artData) throws Exception {
	Instance curr;
	double []probs;
	
	for(int i=0; i<artData.numInstances(); i++){
	    curr = artData.instance(i);
	    //compute the class membership probs predicted by the current ensemble 
	    probs = distributionForInstance(curr);
	    //select class label inversely proportional to the ensemble predictions
	    curr.setClassValue(inverseLabel(probs));
	}	
    }
    

    /** 
     * Select class label such that the probability of selection is
     * inversely proportional to the ensemble's predictions.
     *
     * @param probs class membership probabilities of instance
     * @return index of class label selected
     * @exception Exception if instances cannot be labeled successfully 
     */
    protected int inverseLabel(double []probs) throws Exception{
	double []invProbs = new double[probs.length];
	//Produce probability distribution inversely proportional to the given
	for(int i=0; i<probs.length; i++){
	    if(probs[i]==0){
		invProbs[i] = Double.MAX_VALUE/probs.length; 
		//Account for probability values of 0 - to avoid divide-by-zero errors
		//Divide by probs.length to make sure normalizing works properly
	    }else{
		invProbs[i] = 1.0 / probs[i];
	    }
	}
	Utils.normalize(invProbs);
	double []cdf = new double[invProbs.length];