Example usage for weka.core Instance attributeSparse

Introduction

In this page you can find the example usage for weka.core Instance attributeSparse.

Prototype

public Attribute attributeSparse(int indexOfIndex);

Source Link

Document

Returns the attribute with the given index in the sparse representation.

Usage

From source file:MPCKMeans.java

License:Open Source License

/** Actual KMeans function */
protected void runKMeans() throws Exception {
    boolean converged = false;
    m_Iterations = 0;/*from w  ww.  ja v a2s . c o m*/
    m_numBlankIterations = 0;
    m_Objective = Double.POSITIVE_INFINITY;

    if (!m_isOfflineMetric) {
        if (m_useMultipleMetrics) {
            for (int i = 0; i < m_metrics.length; i++) {
                m_metrics[i].resetMetric();
                m_metricLearners[i].resetLearner();
            }
        } else {
            m_metric.resetMetric();
            m_metricLearner.resetLearner();
        }
        // initialize max CL penalties
        if (m_ConstraintsHash.size() > 0) {
            m_maxCLPenalties = calculateMaxCLPenalties();
        }
    }

    // initialize m_ClusterAssignments
    for (int i = 0; i < m_NumClusters; i++) {
        m_ClusterAssignments[i] = -1;
    }

    PrintStream fincoh = null;
    if (m_ConstraintIncoherenceFile != null) {
        fincoh = new PrintStream(new FileOutputStream(m_ConstraintIncoherenceFile));
    }

    while (!converged) {
        System.out.println("\n" + m_Iterations + ". Objective function: " + ((float) m_Objective));
        m_OldObjective = m_Objective;

        // E-step
        int numMovedPoints = findBestAssignments();

        m_numBlankIterations = (numMovedPoints == 0) ? m_numBlankIterations + 1 : 0;

        //      calculateObjectiveFunction(false);
        System.out.println((float) m_Objective + " - Objective function after point assignment(CALC)");
        System.out.println("\tvar=" + ((float) m_objVariance) + "\tC=" + ((float) m_objCannotLinks) + "\tM="
                + ((float) m_objMustLinks) + "\tLOG=" + ((float) m_objNormalizer) + "\tREG="
                + ((float) m_objRegularizer));

        // M-step
        updateClusterCentroids();

        //      calculateObjectiveFunction(false);
        System.out.println((float) m_Objective + " - Objective function after centroid estimation");
        System.out.println("\tvar=" + ((float) m_objVariance) + "\tC=" + ((float) m_objCannotLinks) + "\tM="
                + ((float) m_objMustLinks) + "\tLOG=" + ((float) m_objNormalizer) + "\tREG="
                + ((float) m_objRegularizer));

        if (m_Trainable == TRAINING_INTERNAL && !m_isOfflineMetric) {
            updateMetricWeights();
            if (m_verbose) {
                calculateObjectiveFunction(true);
                System.out.println((float) m_Objective + " - Objective function after metric update");
                System.out.println("\tvar=" + ((float) m_objVariance) + "\tC=" + ((float) m_objCannotLinks)
                        + "\tM=" + ((float) m_objMustLinks) + "\tLOG=" + ((float) m_objNormalizer) + "\tREG="
                        + ((float) m_objRegularizer));
            }

            if (m_ConstraintsHash.size() > 0) {
                m_maxCLPenalties = calculateMaxCLPenalties();
            }
        }

        if (fincoh != null) {
            printConstraintIncoherence(fincoh);
        }

        converged = convergenceCheck(m_OldObjective, m_Objective);
        m_Iterations++;
    }

    if (fincoh != null) {
        fincoh.close();
    }
    System.out.println("Converged!");
    System.err.print("Its\t" + m_Iterations + "\t");

    if (m_verbose) {
        System.out.println("Done clustering; top cluster features: ");
        for (int i = 0; i < m_NumClusters; i++) {
            System.out.println("Centroid " + i);
            TreeMap map = new TreeMap(Collections.reverseOrder());
            Instance centroid = m_ClusterCentroids.instance(i);
            for (int j = 0; j < centroid.numValues(); j++) {
                Attribute attr = centroid.attributeSparse(j);
                map.put(new Double(centroid.value(attr)), attr.name());
            }
            Iterator it = map.entrySet().iterator();
            for (int j = 0; j < 5 && it.hasNext(); j++) {
                Map.Entry entry = (Map.Entry) it.next();
                System.out.println("\t" + entry.getKey() + "\t" + entry.getValue());
            }
        }
    }
}

---

From source file:FeatureSelection.ReliefFAttributeEval.java

License:Open Source License

/**
 * Updates the minimum and maximum values for all the attributes based on a
 * new instance.//from  w  w  w  .jav  a 2s.com
 *
 * @param instance
 *            the new instance
 */
private void updateMinMax(Instance instance) {
    // for (int j = 0; j < m_numAttribs; j++) {
    try {
        for (int j = 0; j < instance.numValues(); j++) {
            if ((instance.attributeSparse(j).isNumeric()) && (!instance.isMissingSparse(j))) {
                if (Double.isNaN(m_minArray[instance.index(j)])) {
                    m_minArray[instance.index(j)] = instance.valueSparse(j);
                    m_maxArray[instance.index(j)] = instance.valueSparse(j);
                } else {
                    if (instance.valueSparse(j) < m_minArray[instance.index(j)]) {
                        m_minArray[instance.index(j)] = instance.valueSparse(j);
                    } else {
                        if (instance.valueSparse(j) > m_maxArray[instance.index(j)]) {
                            m_maxArray[instance.index(j)] = instance.valueSparse(j);
                        }
                    }
                }
            }
        }
    } catch (Exception ex) {
        System.err.println(ex);
        ex.printStackTrace();
    }
}

---