Examples of weka.classifiers.functions.supportVector.SMOset

Package weka.classifiers.functions.supportVector

Examples of weka.classifiers.functions.supportVector.SMOset

weka.classifiers.functions.supportVector.SMOset
Stores a set of integer of a given size. @author Eibe Frank (eibe@cs.waikato.ac.nz) @version $Revision: 1.5 $

  if (m_KernelIsLinear) {
    m_sparseWeights = new double[0];
    m_sparseIndices = new int[0];
    m_class = null;
  } else {
    m_supportVectors = new SMOset(0);
    m_alpha = new double[0];
    m_class = new double[0];
  }


  // Fit sigmoid if requested
  if (fitLogistic) {
    fitLogistic(insts, cl1, cl2, numFolds, new Random(randomSeed));
  }
  return;
      }
      
      // Set the reference to the data
      m_data = insts;


      // If machine is linear, reserve space for weights
      if (m_KernelIsLinear) {
  m_weights = new double[m_data.numAttributes()];
      } else {
  m_weights = null;
      }
      
      // Initialize alpha array to zero
      m_alpha = new double[m_data.numInstances()];
      
      // Initialize sets
      m_supportVectors = new SMOset(m_data.numInstances());
      m_I0 = new SMOset(m_data.numInstances());
      m_I1 = new SMOset(m_data.numInstances());
      m_I2 = new SMOset(m_data.numInstances());
      m_I3 = new SMOset(m_data.numInstances());
      m_I4 = new SMOset(m_data.numInstances());


      // Clean out some instance variables
      m_sparseWeights = null;
      m_sparseIndices = null;

View Full Code Here

          m_b = 1;
        } else {
          m_class = null;
          return;
        }
        m_supportVectors = new SMOset(0);
        m_alpha = new double[0];
        m_class = new double[0];


        // Fit sigmoid if requested
        if (fitLogistic) {
          fitLogistic(insts, cl1, cl2, numFolds, new Random(randomSeed));
        }
        return;
      }


      // Set the reference to the data
      m_data = insts;
      m_weights = null;


      // Initialize alpha array to zero
      m_alpha = new double[m_data.numInstances()];


      // Initialize sets
      m_supportVectors = new SMOset(m_data.numInstances());
      m_I0 = new SMOset(m_data.numInstances());
      m_I1 = new SMOset(m_data.numInstances());
      m_I2 = new SMOset(m_data.numInstances());
      m_I3 = new SMOset(m_data.numInstances());
      m_I4 = new SMOset(m_data.numInstances());


      // Clean out some instance variables
      m_sparseWeights = null;
      m_sparseIndices = null;

View Full Code Here

  if (m_KernelIsLinear) {
    m_sparseWeights = new double[0];
    m_sparseIndices = new int[0];
    m_class = null;
  } else {
    m_supportVectors = new SMOset(0);
    m_alpha = new double[0];
    m_class = new double[0];
  }


  // Fit sigmoid if requested
  if (fitLogistic) {
    fitLogistic(insts, cl1, cl2, numFolds, new Random(randomSeed));
  }
  return;
      }
      
      // Set the reference to the data
      m_data = insts;


      // If machine is linear, reserve space for weights
      if (m_KernelIsLinear) {
  m_weights = new double[m_data.numAttributes()];
      } else {
  m_weights = null;
      }
      
      // Initialize alpha array to zero
      m_alpha = new double[m_data.numInstances()];
      
      // Initialize sets
      m_supportVectors = new SMOset(m_data.numInstances());
      m_I0 = new SMOset(m_data.numInstances());
      m_I1 = new SMOset(m_data.numInstances());
      m_I2 = new SMOset(m_data.numInstances());
      m_I3 = new SMOset(m_data.numInstances());
      m_I4 = new SMOset(m_data.numInstances());


      // Clean out some instance variables
      m_sparseWeights = null;
      m_sparseIndices = null;

View Full Code Here


    // Initialize fcache
    m_fcache = new double[m_data.numInstances()];


    // Initialize sets
    m_I0 = new SMOset(m_data.numInstances());
    m_I1 = new SMOset(m_data.numInstances());
    m_I2 = new SMOset(m_data.numInstances());
    m_I3 = new SMOset(m_data.numInstances());




    /* MAIN ROUTINE FOR MODIFICATION 1 */
    // Follows the specification of the first modification of Shevade's paper

View Full Code Here

          m_b = 1;
        } else {
          m_class = null;
          return;
        }
        m_supportVectors = new SMOset(0);
        m_alpha = new double[0];
        m_class = new double[0];


        // Fit sigmoid if requested
        if (fitLogistic) {
          fitLogistic(insts, cl1, cl2, numFolds, new Random(randomSeed));
        }
        return;
      }


      // Set the reference to the data
      m_data = insts;
      m_weights = null;


      // Initialize alpha array to zero
      m_alpha = new double[m_data.numInstances()];


      // Initialize sets
      m_supportVectors = new SMOset(m_data.numInstances());
      m_I0 = new SMOset(m_data.numInstances());
      m_I1 = new SMOset(m_data.numInstances());
      m_I2 = new SMOset(m_data.numInstances());
      m_I3 = new SMOset(m_data.numInstances());
      m_I4 = new SMOset(m_data.numInstances());


      // Clean out some instance variables
      m_sparseWeights = null;
      m_sparseIndices = null;

View Full Code Here

  if (m_KernelIsLinear) {
    m_sparseWeights = new double[0];
    m_sparseIndices = new int[0];
    m_class = null;
  } else {
    m_supportVectors = new SMOset(0);
    m_alpha = new double[0];
    m_class = new double[0];
  }


  // Fit sigmoid if requested
  if (fitLogistic) {
    fitLogistic(insts, cl1, cl2, numFolds, new Random(randomSeed));
  }
  return;
      }
      
      // Set the reference to the data
      m_data = insts;


      // If machine is linear, reserve space for weights
      if (m_KernelIsLinear) {
  m_weights = new double[m_data.numAttributes()];
      } else {
  m_weights = null;
      }
      
      // Initialize alpha array to zero
      m_alpha = new double[m_data.numInstances()];
      
      // Initialize sets
      m_supportVectors = new SMOset(m_data.numInstances());
      m_I0 = new SMOset(m_data.numInstances());
      m_I1 = new SMOset(m_data.numInstances());
      m_I2 = new SMOset(m_data.numInstances());
      m_I3 = new SMOset(m_data.numInstances());
      m_I4 = new SMOset(m_data.numInstances());


      // Clean out some instance variables
      m_sparseWeights = null;
      m_sparseIndices = null;

View Full Code Here

TOP

Related Classes of weka.classifiers.functions.supportVector.SMOset

weka.classifiers.functions.SMO$BinarySMO

weka.classifiers.functions.SMOreg

weka.classifiers.mi.MISMO$BinaryMISMO

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.