KMeans.h

Go to the documentation of this file.

#ifndef GRT_KMEANS_HEADER
#define GRT_KMEANS_HEADER

#include "../../Util/GRTCommon.h"
#include "../../CoreModules/Clusterer.h"
#include "../../DataStructures/ClassificationData.h"
#include "../../DataStructures/UnlabelledData.h"

GRT_BEGIN_NAMESPACE

class KMeans : public Clusterer{

public:
    KMeans(const UINT numClusters=10,const UINT minNumEpochs=5,const UINT maxNumEpochs=1000,const Float minChange=1.0e-5,const bool computeTheta=true);

    KMeans(const KMeans &rhs);
    
    virtual ~KMeans();
    
    KMeans &operator=(const KMeans &rhs);
    
    virtual bool deepCopyFrom(const Clusterer *clusterer);
    
    virtual bool reset();
    
    virtual bool clear();
    
    bool trainModel(MatrixFloat &data);
    
    virtual bool train_(MatrixFloat &data);
    
    virtual bool train_(ClassificationData &trainingData);
    
     virtual bool train_(UnlabelledData &trainingData);
    
    virtual bool predict_(VectorFloat &inputVector);
    
    virtual bool saveModelToFile( std::fstream &file ) const;
    
    virtual bool loadModelFromFile( std::fstream &file );
    
    //bool predict(VectorFloat inputVector,UINT &predictedClusterLabel,Float &maxLikelihood,VectorFloat &clusterLikelihoods);

    //Getters
    Float getTheta(){ return finalTheta; }
    bool getModelTrained(){ return trained; }

    const VectorFloat& getTrainingThetaLog() const { return thetaTracker; }
    const MatrixFloat& getClusters() const { return clusters; }
    const Vector< UINT >& getClassLabelsVector() const { return assign; }
    const Vector< UINT >& getClassCountVector() const { return count; }
    
    //Setters
    bool setComputeTheta(const bool computeTheta);
    
    bool setClusters(const MatrixFloat &clusters);
    
    //Tell the compiler we are using the following functions from the MLBase class to stop hidden virtual function warnings
    using MLBase::saveModelToFile;
    using MLBase::loadModelFromFile;
    using MLBase::train;
    using MLBase::train_;
    using MLBase::predict;
    using MLBase::predict_;

protected:
    UINT estep(const MatrixFloat &data);
    void mstep(const MatrixFloat &data);
    Float calculateTheta(const MatrixFloat &data);
    inline Float SQR(const Float a) {return a*a;};

    bool computeTheta;
    UINT numTrainingSamples;            
    UINT nchg;                          
    Float finalTheta;
    MatrixFloat clusters;
    Vector< UINT > assign, count;
    VectorFloat thetaTracker;
    
private:
    static RegisterClustererModule< KMeans > registerModule;
};
    
GRT_END_NAMESPACE

#endif //GRT_KMEANS_HEADER