dox/Infovis/vtkKMeansStatistics.h

Go to the documentation of this file.

/*=========================================================================

Program:   Visualization Toolkit
Module:    vtkKMeansStatistics.h

Copyright (c) Ken Martin, Will Schroeder, Bill Lorensen
All rights reserved.
See Copyright.txt or http://www.kitware.com/Copyright.htm for details.

This software is distributed WITHOUT ANY WARRANTY; without even
the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
PURPOSE.  See the above copyright notice for more information.

=========================================================================*/
/*-------------------------------------------------------------------------
  Copyright 2008 Sandia Corporation.
  Under the terms of Contract DE-AC04-94AL85000 with Sandia Corporation,
  the U.S. Government retains certain rights in this software.
  -------------------------------------------------------------------------*/
#ifndef __vtkKMeansStatistics_h
#define __vtkKMeansStatistics_h

#include "vtkStatisticsAlgorithm.h"

class vtkIdTypeArray;
class vtkIntArray;
class vtkDoubleArray;
class vtkKMeansDistanceFunctor;

class VTK_INFOVIS_EXPORT vtkKMeansStatistics : public vtkStatisticsAlgorithm
{
public:
  vtkTypeMacro(vtkKMeansStatistics, vtkStatisticsAlgorithm);
  virtual void PrintSelf( ostream& os, vtkIndent indent );
  static vtkKMeansStatistics* New();
 

  virtual void SetDistanceFunctor( vtkKMeansDistanceFunctor* );
  vtkGetObjectMacro(DistanceFunctor,vtkKMeansDistanceFunctor);


  vtkSetMacro(DefaultNumberOfClusters, int);
  vtkGetMacro(DefaultNumberOfClusters, int);


  vtkSetStringMacro(KValuesArrayName);
  vtkGetStringMacro(KValuesArrayName);


  vtkSetMacro( MaxNumIterations, int );
  vtkGetMacro( MaxNumIterations, int );


  vtkSetMacro( Tolerance, double );
  vtkGetMacro( Tolerance, double );


  virtual void Aggregate( vtkDataObjectCollection*,
                          vtkDataObject* ) { return; };

protected:
  vtkKMeansStatistics();
  ~vtkKMeansStatistics();


  virtual int FillOutputPortInformation( int port, vtkInformation* info );
  virtual int FillInputPortInformation( int port, vtkInformation* info );


  virtual void Learn( vtkTable* inData,
                      vtkTable* inParameters,
                      vtkDataObject* outMeta );

  virtual void Derive( vtkDataObject* );


  virtual void Assess( vtkTable*, 
                       vtkDataObject*, 
                       vtkTable* );


  virtual void Test( vtkTable*,
                     vtkDataObject*,
                     vtkDataObject* ) { return; };

  //BTX  

  virtual void SelectAssessFunctor( vtkTable* inData, 
                                    vtkDataObject* inMeta,
                                    vtkStringArray* rowNames,
                                    AssessFunctor*& dfunc );
  //ETX  
  // Description:
  // Subroutine to update new cluster centers from the old centers.
  // Called from within Learn (and will be overridden by vtkPKMeansStatistics
  // to handle distributed datasets).
  virtual void UpdateClusterCenters( vtkTable* newClusterElements, 
                                     vtkTable* curClusterElements, 
                                     vtkIdTypeArray* numMembershipChanges,
                                     vtkIdTypeArray* numElementsInCluster, 
                                     vtkDoubleArray* error,
                                     vtkIdTypeArray* startRunID, 
                                     vtkIdTypeArray* endRunID, 
                                     vtkIntArray *computeRun );

  virtual vtkIdType GetTotalNumberOfObservations( vtkIdType numObservations );


  int InitializeDataAndClusterCenters(vtkTable* inParameters,
                                      vtkTable* inData,
                                      vtkTable*  dataElements,
                                      vtkIdTypeArray*  numberOfClusters,
                                      vtkTable*  curClusterElements,
                                      vtkTable*  newClusterElements,
                                      vtkIdTypeArray*  startRunID,
                                      vtkIdTypeArray*  endRunID);


  virtual void CreateInitialClusterCenters(vtkIdType numToAllocate, 
                                           vtkIdTypeArray* numberOfClusters, 
                                           vtkTable* inData, 
                                           vtkTable* curClusterElements, 
                                           vtkTable* newClusterElements);



  int DefaultNumberOfClusters;
  // Description:
  // This is the name of the column that specifies the number of clusters in each run. 
  // This is only used if the user has not specified initial clusters.
  char* KValuesArrayName;
  // Description:
  // This is the maximum number of iterations allowed if the new cluster centers have not yet converged. 
  int MaxNumIterations;
  // Description:
  // This is the percentage of data elements that swap cluster IDs 
  double Tolerance;
  // Description:
  // This is the Distance functor.  The default is Euclidean distance, however this can be overridden.
  vtkKMeansDistanceFunctor* DistanceFunctor;
  
private:
  vtkKMeansStatistics( const vtkKMeansStatistics& ); // Not implemented
  void operator=( const vtkKMeansStatistics& );  // Not implemented
};

#endif

---