Examples of weka.core.SerializedObject

weka.core.SerializedObject
Class for storing an object in serialized form in memory. It can be used to make deep copies of objects, and also allows compression to conserve memory.
@author Richard Kirkby (rbk1@cs.waikato.ac.nz) @version $Revision: 1.12 $

  public static Kernel[] makeCopies(Kernel model, int num) throws Exception {
    if (model == null)
      throw new Exception("No model kernel set");


    Kernel[] kernels = new Kernel[num];
    SerializedObject so = new SerializedObject(model);
    for (int i = 0; i < kernels.length; i++)
      kernels[i] = (Kernel) so.getObject();


    return kernels;
  }

View Full Code Here

      m_Instances.stratify(numFolds);


      for (int k = 0; k < list.length; k++) {
        for (int i = 0; i < numFolds; i++) {
          Instances train = m_Instances.trainCV(numFolds, i, random);
          SerializedObject so = new SerializedObject(this);
          BayesianLogisticRegression blr = (BayesianLogisticRegression) so.getObject();
          //          blr.setHyperparameterSelection(3);
          blr.setHyperparameterSelection(new SelectedTag(SPECIFIC_VALUE, 
                                                         TAGS_HYPER_METHOD));
          blr.setHyperparameterValue(list[k]);
          //          blr.setPriorClass(PriorClass);

View Full Code Here

    }


    try {
      m_marginCalculator = new MarginCalculator();
      m_marginCalculator.calcMargins(m_BayesNet);
      SerializedObject so = new SerializedObject(m_marginCalculator);
      m_marginCalculatorWithEvidence = (MarginCalculator) so.getObject();
      for (int iNode = 0; iNode < m_BayesNet.getNrOfNodes(); iNode++) {
        if (m_BayesNet.getEvidence(iNode) >= 0) {
          m_marginCalculatorWithEvidence.setEvidence(iNode, m_BayesNet.getEvidence(iNode));
        }
      }

View Full Code Here

                if (m_BayesNet.getEvidence(m_nCurrentNode) < 0 && iValue >= 0) {
                  m_BayesNet.setEvidence(m_nCurrentNode, iValue);
                  m_marginCalculatorWithEvidence.setEvidence(m_nCurrentNode, iValue);
                } else {
                  m_BayesNet.setEvidence(m_nCurrentNode, iValue);
                  SerializedObject so = new SerializedObject(m_marginCalculator);
                  m_marginCalculatorWithEvidence = (MarginCalculator) so.getObject();
                  for (int iNode = 0; iNode < m_BayesNet.getNrOfNodes(); iNode++) {
                    if (m_BayesNet.getEvidence(iNode) >= 0) {
                      m_marginCalculatorWithEvidence.setEvidence(iNode, m_BayesNet.getEvidence(iNode));
                    }
                  }

View Full Code Here

   * @param model   the filter to copy
   * @return     a deep copy of the filter
   * @throws Exception   if an error occurs
   */
  public static Filter makeCopy(Filter model) throws Exception {
    return (Filter)new SerializedObject(model).getObject();
  }

View Full Code Here


    if (model == null) {
      throw new Exception("No model filter set");
    }
    Filter[] filters = new Filter[num];
    SerializedObject so = new SerializedObject(model);
    for (int i = 0; i < filters.length; i++) {
      filters[i] = (Filter) so.getObject();
    }
    return filters;
  }

View Full Code Here

        // unbiased predictions
        insts.randomize(random);
        insts.stratify(numFolds);
        for (int i = 0; i < numFolds; i++) {
          Instances train = insts.trainCV(numFolds, i, random);
          SerializedObject so = new SerializedObject(this);
          BinaryMISMO smo = (BinaryMISMO)so.getObject();
          smo.buildClassifier(train, cl1, cl2, false, -1, -1);
          Instances test = insts.testCV(numFolds, i);
          for (int j = 0; j < test.numInstances(); j++) {
            double[] vals = new double[2];
            vals[0] = smo.SVMOutput(-1, test.instance(j));

View Full Code Here

   * @param model the associator to copy
   * @return a deep copy of the associator
   * @exception Exception if an error occurs
   */
  public static Associator makeCopy(Associator model) throws Exception {
    return (Associator) new SerializedObject(model).getObject();
  }

View Full Code Here


    if (model == null) {
      throw new Exception("No model associator set");
    }
    Associator [] associators = new Associator [num];
    SerializedObject so = new SerializedObject(model);
    for(int i = 0; i < associators.length; i++) {
      associators[i] = (Associator) so.getObject();
    }
    return associators;
  }

View Full Code Here

      grph = ((Drawable)classifier).graph();
    } catch (Exception ex) {
    }
        }
        // copy full model for output
        SerializedObject so = new SerializedObject(classifier);
        fullClassifier = (Classifier) so.getObject();
      }
      
      switch (testMode) {
        case 3: // Test on training
        m_Log.statusMessage("Evaluating on training data...");

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

Related Classes of weka.core.SerializedObject

trackerModule.weka.WekaInterface

weka.associations.AbstractAssociator

weka.attributeSelection.ASEvaluation

weka.attributeSelection.ASSearch

weka.attributeSelection.CheckAttributeSelection

weka.classifiers.AbstractClassifier

weka.classifiers.bayes.BayesianLogisticRegression

weka.classifiers.bayes.net.EditableBayesNet$AddArcAction

weka.classifiers.bayes.net.EditableBayesNet$DeleteArcAction

weka.classifiers.bayes.net.EditableBayesNet$DeleteNodeAction

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.