MinDistModel.h

Go to the documentation of this file.

#ifndef GRT_MINDISTMODEL_HEADER
#define GRT_MINDISTMODEL_HEADER

#include "../../Util/GRTCommon.h"
#include "../../ClusteringModules/KMeans/KMeans.h"

GRT_BEGIN_NAMESPACE

class GRT_API MinDistModel
{
public:
    
 MinDistModel();
    
    MinDistModel(const MinDistModel &rhs);
    
 ~MinDistModel(void);
    
 MinDistModel &operator=(const MinDistModel &rhs);
 
 bool train(UINT classLabel,MatrixFloat &trainingData,UINT numClusters,Float minChange,UINT maxNumEpochs);
 Float predict(const VectorFloat &observation);
 void recomputeThresholdValue();
 
 UINT getClassLabel() const;
 UINT getNumFeatures() const;
 UINT getNumClusters() const;
 UINT getDistanceMode() const;
    Float getRejectionThreshold() const;
    Float getGamma() const;
    Float getTrainingMu() const;
    Float getTrainingSigma() const;
    MatrixFloat getClusters() const;
    
    bool setClassLabel(UINT classLabel);
    bool setClusters(MatrixFloat &clusters);
    bool setGamma(Float gamma);
    bool setRejectionThreshold(Float rejectionThreshold);
    bool setTrainingSigma(Float trainingSigma);
    bool setTrainingMu(Float trainingMu);
    bool setTrainingLoggingEnabled(bool enabled);
    
private:
    Float SQR(const Float &x){ return (x*x); }
    bool trainingLoggingEnabled;
 UINT classLabel;
 UINT numFeatures;
 UINT numClusters;
 Float rejectionThreshold;   //The classification threshold value
 Float gamma;    //The number of standard deviations to use for the threshold
 Float trainingMu;   //The average confidence value in the training data
 Float trainingSigma;  //The simga confidence value in the training data
 MatrixFloat clusters;
};

GRT_END_NAMESPACE

#endif //GRT_MINDISTMODEL_HEADER