GestureRecognitionToolkit  Version: 1.0 Revision: 04-03-15
The Gesture Recognition Toolkit (GRT) is a cross-platform, open-source, c++ machine learning library for real-time gesture recognition.
GRT::AdaBoost Class Reference
Inheritance diagram for GRT::AdaBoost:
GRT::Classifier GRT::MLBase GRT::GRTBase GRT::Observer< TrainingResult > GRT::Observer< TestInstanceResult >

Public Types

enum  PredictionMethods { MAX_POSITIVE_VALUE =0, MAX_VALUE }
 
- Public Types inherited from GRT::Classifier
typedef std::map< string, Classifier *(*)() > StringClassifierMap
 
- Public Types inherited from GRT::MLBase
enum  BaseTypes { BASE_TYPE_NOT_SET =0, CLASSIFIER, REGRESSIFIER, CLUSTERER }
 

Public Member Functions

 AdaBoost (const WeakClassifier &weakClassifier=DecisionStump(), bool useScaling=false, bool useNullRejection=false, double nullRejectionCoeff=10.0, UINT numBoostingIterations=20, UINT predictionMethod=MAX_VALUE)
 
 AdaBoost (const AdaBoost &rhs)
 
virtual ~AdaBoost ()
 
AdaBoostoperator= (const AdaBoost &rhs)
 
virtual bool deepCopyFrom (const Classifier *classifier)
 
virtual bool train_ (ClassificationData &trainingData)
 
virtual bool predict_ (VectorDouble &inputVector)
 
virtual bool clear ()
 
virtual bool saveModelToFile (fstream &file) const
 
virtual bool loadModelFromFile (fstream &file)
 
virtual bool recomputeNullRejectionThresholds ()
 
bool setNullRejectionCoeff (double nullRejectionCoeff)
 
bool setWeakClassifier (const WeakClassifier &weakClassifer)
 
bool addWeakClassifier (const WeakClassifier &weakClassifer)
 
bool clearWeakClassifiers ()
 
bool setNumBoostingIterations (UINT numBoostingIterations)
 
bool setPredictionMethod (UINT predictionMethod)
 
void printModel ()
 
vector< AdaBoostClassModelgetModels () const
 
- Public Member Functions inherited from GRT::Classifier
 Classifier (void)
 
virtual ~Classifier (void)
 
bool copyBaseVariables (const Classifier *classifier)
 
virtual bool reset ()
 
string getClassifierType () const
 
bool getSupportsNullRejection () const
 
bool getNullRejectionEnabled () const
 
double getNullRejectionCoeff () const
 
double getMaximumLikelihood () const
 
double getBestDistance () const
 
double getPhase () const
 
virtual UINT getNumClasses () const
 
UINT getClassLabelIndexValue (UINT classLabel) const
 
UINT getPredictedClassLabel () const
 
VectorDouble getClassLikelihoods () const
 
VectorDouble getClassDistances () const
 
VectorDouble getNullRejectionThresholds () const
 
vector< UINT > getClassLabels () const
 
vector< MinMaxgetRanges () const
 
bool enableNullRejection (bool useNullRejection)
 
virtual bool setNullRejectionThresholds (VectorDouble newRejectionThresholds)
 
bool getTimeseriesCompatible () const
 
ClassifiercreateNewInstance () const
 
ClassifierdeepCopy () const
 
const ClassifiergetClassifierPointer () const
 
const ClassifiergetBaseClassifier () const
 
- Public Member Functions inherited from GRT::MLBase
 MLBase (void)
 
virtual ~MLBase (void)
 
bool copyMLBaseVariables (const MLBase *mlBase)
 
virtual bool train (ClassificationData trainingData)
 
virtual bool train (RegressionData trainingData)
 
virtual bool train_ (RegressionData &trainingData)
 
virtual bool train (TimeSeriesClassificationData trainingData)
 
virtual bool train_ (TimeSeriesClassificationData &trainingData)
 
virtual bool train (TimeSeriesClassificationDataStream trainingData)
 
virtual bool train_ (TimeSeriesClassificationDataStream &trainingData)
 
virtual bool train (UnlabelledData trainingData)
 
virtual bool train_ (UnlabelledData &trainingData)
 
virtual bool train (MatrixDouble data)
 
virtual bool train_ (MatrixDouble &data)
 
virtual bool predict (VectorDouble inputVector)
 
virtual bool predict (MatrixDouble inputMatrix)
 
virtual bool predict_ (MatrixDouble &inputMatrix)
 
virtual bool map (VectorDouble inputVector)
 
virtual bool map_ (VectorDouble &inputVector)
 
virtual bool print () const
 
virtual bool save (const string filename) const
 
virtual bool load (const string filename)
 
virtual bool saveModelToFile (string filename) const
 
virtual bool loadModelFromFile (string filename)
 
virtual bool getModel (ostream &stream) const
 
double scale (const double &x, const double &minSource, const double &maxSource, const double &minTarget, const double &maxTarget, const bool constrain=false)
 
virtual string getModelAsString () const
 
UINT getBaseType () const
 
UINT getNumInputFeatures () const
 
UINT getNumInputDimensions () const
 
UINT getNumOutputDimensions () const
 
UINT getMinNumEpochs () const
 
UINT getMaxNumEpochs () const
 
UINT getValidationSetSize () const
 
UINT getNumTrainingIterationsToConverge () const
 
double getMinChange () const
 
double getLearningRate () const
 
double getRootMeanSquaredTrainingError () const
 
double getTotalSquaredTrainingError () const
 
bool getUseValidationSet () const
 
bool getRandomiseTrainingOrder () const
 
bool getTrained () const
 
bool getModelTrained () const
 
bool getScalingEnabled () const
 
bool getIsBaseTypeClassifier () const
 
bool getIsBaseTypeRegressifier () const
 
bool getIsBaseTypeClusterer () const
 
bool enableScaling (bool useScaling)
 
bool setMaxNumEpochs (const UINT maxNumEpochs)
 
bool setMinNumEpochs (const UINT minNumEpochs)
 
bool setMinChange (const double minChange)
 
bool setLearningRate (double learningRate)
 
bool setUseValidationSet (const bool useValidationSet)
 
bool setValidationSetSize (const UINT validationSetSize)
 
bool setRandomiseTrainingOrder (const bool randomiseTrainingOrder)
 
bool registerTrainingResultsObserver (Observer< TrainingResult > &observer)
 
bool registerTestResultsObserver (Observer< TestInstanceResult > &observer)
 
bool removeTrainingResultsObserver (const Observer< TrainingResult > &observer)
 
bool removeTestResultsObserver (const Observer< TestInstanceResult > &observer)
 
bool removeAllTrainingObservers ()
 
bool removeAllTestObservers ()
 
bool notifyTrainingResultsObservers (const TrainingResult &data)
 
bool notifyTestResultsObservers (const TestInstanceResult &data)
 
MLBasegetMLBasePointer ()
 
const MLBasegetMLBasePointer () const
 
vector< TrainingResultgetTrainingResults () const
 
- Public Member Functions inherited from GRT::GRTBase
 GRTBase (void)
 
virtual ~GRTBase (void)
 
bool copyGRTBaseVariables (const GRTBase *GRTBase)
 
string getClassType () const
 
string getLastWarningMessage () const
 
string getLastErrorMessage () const
 
string getLastInfoMessage () const
 
GRTBasegetGRTBasePointer ()
 
const GRTBasegetGRTBasePointer () const
 
- Public Member Functions inherited from GRT::Observer< TrainingResult >
virtual void notify (const TrainingResult &data)
 
- Public Member Functions inherited from GRT::Observer< TestInstanceResult >
virtual void notify (const TestInstanceResult &data)
 

Protected Member Functions

bool loadLegacyModelFromFile (fstream &file)
 
- Protected Member Functions inherited from GRT::Classifier
bool saveBaseSettingsToFile (fstream &file) const
 
bool loadBaseSettingsFromFile (fstream &file)
 
- Protected Member Functions inherited from GRT::MLBase
bool saveBaseSettingsToFile (fstream &file) const
 
bool loadBaseSettingsFromFile (fstream &file)
 
- Protected Member Functions inherited from GRT::GRTBase
double SQR (const double &x) const
 

Protected Attributes

UINT numBoostingIterations
 
UINT predictionMethod
 
vector< WeakClassifier * > weakClassifiers
 
vector< AdaBoostClassModelmodels
 
- Protected Attributes inherited from GRT::Classifier
string classifierType
 
bool supportsNullRejection
 
bool useNullRejection
 
UINT numClasses
 
UINT predictedClassLabel
 
UINT classifierMode
 
double nullRejectionCoeff
 
double maxLikelihood
 
double bestDistance
 
double phase
 
VectorDouble classLikelihoods
 
VectorDouble classDistances
 
VectorDouble nullRejectionThresholds
 
vector< UINT > classLabels
 
vector< MinMaxranges
 
- Protected Attributes inherited from GRT::MLBase
bool trained
 
bool useScaling
 
UINT baseType
 
UINT numInputDimensions
 
UINT numOutputDimensions
 
UINT numTrainingIterationsToConverge
 
UINT minNumEpochs
 
UINT maxNumEpochs
 
UINT validationSetSize
 
double learningRate
 
double minChange
 
double rootMeanSquaredTrainingError
 
double totalSquaredTrainingError
 
bool useValidationSet
 
bool randomiseTrainingOrder
 
Random random
 
vector< TrainingResulttrainingResults
 
TrainingResultsObserverManager trainingResultsObserverManager
 
TestResultsObserverManager testResultsObserverManager
 
- Protected Attributes inherited from GRT::GRTBase
string classType
 
DebugLog debugLog
 
ErrorLog errorLog
 
InfoLog infoLog
 
TrainingLog trainingLog
 
TestingLog testingLog
 
WarningLog warningLog
 

Static Protected Attributes

static RegisterClassifierModule< AdaBoostregisterModule
 

Additional Inherited Members

- Static Public Member Functions inherited from GRT::Classifier
static ClassifiercreateInstanceFromString (string const &classifierType)
 
static vector< string > getRegisteredClassifiers ()
 
- Static Public Member Functions inherited from GRT::GRTBase
static string getGRTVersion (bool returnRevision=true)
 
static string getGRTRevison ()
 
- Protected Types inherited from GRT::Classifier
enum  ClassifierModes { STANDARD_CLASSIFIER_MODE =0, TIMESERIES_CLASSIFIER_MODE }
 
- Static Protected Member Functions inherited from GRT::Classifier
static StringClassifierMapgetMap ()
 

Detailed Description

Definition at line 57 of file AdaBoost.h.

Member Enumeration Documentation

These are the two prediction methods that the GRT::AdaBoost classifier can use.

Definition at line 240 of file AdaBoost.h.

Constructor & Destructor Documentation

GRT::AdaBoost::AdaBoost ( const WeakClassifier weakClassifier = DecisionStump(),
bool  useScaling = false,
bool  useNullRejection = false,
double  nullRejectionCoeff = 10.0,
UINT  numBoostingIterations = 20,
UINT  predictionMethod = MAX_VALUE 
)

Default Constructor

Parameters
constWeakClassifier &weakClassifier: sets the initial weak classifier that is added to the vector of weak classifiers used to train the AdaBoost model
booluseScaling: sets if the training and prediction data should be scaled to a specific range. Default value is useScaling = false
booluseNullRejection: sets if null rejection will be used for the realtime prediction. If useNullRejection is set to true then the predictedClassLabel will be set to 0 (which is the default null label) if the distance between the inputVector and the top K datum is greater than the null rejection threshold for the top predicted class. The null rejection threshold is computed for each class during the training phase. Default value is useNullRejection = false
doublenullRejectionCoeff: sets the null rejection coefficient, this is a multipler controlling the null rejection threshold for each class. This will only be used if the useNullRejection parameter is set to true. Default value is nullRejectionCoeff = 10.0
UINTnumBoostingIterations: sets the number of boosting iterations to use during training. Default value = 20
UINTpredictionMethod: sets the prediction method for AdaBoost, this should be one of the PredictionMethods. Default value = MAX_VALUE

Definition at line 30 of file AdaBoost.cpp.

GRT::AdaBoost::AdaBoost ( const AdaBoost rhs)

Defines the copy constructor.

Parameters
constAdaBoost &rhs: the instance from which all the data will be copied into this instance

Definition at line 47 of file AdaBoost.cpp.

GRT::AdaBoost::~AdaBoost ( void  )
virtual

Default Destructor

Definition at line 57 of file AdaBoost.cpp.

Member Function Documentation

bool GRT::AdaBoost::addWeakClassifier ( const WeakClassifier weakClassifer)

Adds a WeakClassifier to the list of WeakClassifiers to use for boosting.

If this function is called, the new WeakClassifier will be added to the list of WeakClassifiers.

Returns
returns true if the WeakClassifier was added successfully, false otherwise

Definition at line 526 of file AdaBoost.cpp.

bool GRT::AdaBoost::clear ( )
virtual

This overrides the clear function in the Classifier base class. It will completely clear the ML module, removing any trained model and setting all the base variables to their default values.

Returns
returns true if the module was cleared succesfully, false otherwise

Reimplemented from GRT::Classifier.

Definition at line 503 of file AdaBoost.cpp.

bool GRT::AdaBoost::clearWeakClassifiers ( )

Clears all the current WeakClassifiers.

Returns
returns true if the WeakClassifiers was cleared successfully, false otherwise

Definition at line 534 of file AdaBoost.cpp.

bool GRT::AdaBoost::deepCopyFrom ( const Classifier classifier)
virtual

This is required for the Gesture Recognition Pipeline for when the pipeline.setClassifier(...) method is called. It clones the data from the Base Class Classifier pointer (which should be pointing to an AdaBoost instance) into this instance

Parameters
Classifier*classifier: a pointer to the Classifier Base Class, this should be pointing to another AdaBoost instance
Returns
returns true if the clone was successfull, false otherwise

Reimplemented from GRT::Classifier.

Definition at line 85 of file AdaBoost.cpp.

vector< AdaBoostClassModel > GRT::AdaBoost::getModels ( ) const
inline

Returns the current vector of AdaBoostClassModel models.

Returns
a vector containing the current AdaBoostClassModel models.

Definition at line 216 of file AdaBoost.h.

bool GRT::AdaBoost::loadModelFromFile ( fstream &  file)
virtual

This loads a trained AdaBoost model from a file. This overrides the loadModelFromFile function in the Classifier base class.

Parameters
fstream&file: a reference to the file the AdaBoost model will be loaded from
Returns
returns true if the model was loaded successfully, false otherwise

Reimplemented from GRT::MLBase.

Definition at line 436 of file AdaBoost.cpp.

AdaBoost & GRT::AdaBoost::operator= ( const AdaBoost rhs)

Defines how the data from the rhs AdaBoost should be copied to this AdaBoost

Parameters
constAdaBoost &rhs: another instance of a AdaBoost
Returns
returns a reference to this instance of the AdaBoost

Definition at line 63 of file AdaBoost.cpp.

bool GRT::AdaBoost::predict_ ( VectorDouble &  inputVector)
virtual

This predicts the class of the inputVector. This overrides the predict function in the Classifier base class.

Parameters
VectorDouble&inputVector: the input vector to classify
Returns
returns true if the prediction was performed, false otherwise

Reimplemented from GRT::MLBase.

Definition at line 293 of file AdaBoost.cpp.

void GRT::AdaBoost::printModel ( )

Prints the current model to the std::out.

Returns
void.

Definition at line 562 of file AdaBoost.cpp.

bool GRT::AdaBoost::recomputeNullRejectionThresholds ( )
virtual

This recomputes the null rejection thresholds for each of the classes in the AdaBoost model. This will be called automatically if the setGamma(double gamma) function is called. The AdaBoost model needs to be trained first before this function can be called.

Returns
returns true if the null rejection thresholds were updated successfully, false otherwise

Reimplemented from GRT::Classifier.

Definition at line 382 of file AdaBoost.cpp.

bool GRT::AdaBoost::saveModelToFile ( fstream &  file) const
virtual

This saves the trained AdaBoost model to a file. This overrides the saveModelToFile function in the Classifier base class.

Parameters
fstream&file: a reference to the file the AdaBoost model will be saved to
Returns
returns true if the model was saved successfully, false otherwise

Reimplemented from GRT::MLBase.

Definition at line 401 of file AdaBoost.cpp.

bool GRT::AdaBoost::setNullRejectionCoeff ( double  nullRejectionCoeff)
virtual

Sets the nullRejectionCoeff parameter. The nullRejectionCoeff parameter is a multipler controlling the null rejection threshold for each class. This function will also recompute the null rejection thresholds.

Returns
returns true if the gamma parameter was updated successfully, false otherwise

Reimplemented from GRT::Classifier.

Definition at line 391 of file AdaBoost.cpp.

bool GRT::AdaBoost::setNumBoostingIterations ( UINT  numBoostingIterations)

Sets the number of boosting iterations that should be used when training the AdaBoost model. The numBoostingIterations parameter must be greater than zero.

Returns
returns true if the numBoostingIterations was set successfully, false otherwise

Definition at line 546 of file AdaBoost.cpp.

bool GRT::AdaBoost::setPredictionMethod ( UINT  predictionMethod)

Sets the prediction method for AdaBoost, this should be one of the PredictionMethods enumerations.

Parameters
UINTpredictionMethod: the predictionMethod that should be used by AdaBoost, this should be one of the PredictionMethods enumerations
Returns
returns true if the predictionMethod was set successfully, false otherwise

Definition at line 554 of file AdaBoost.cpp.

bool GRT::AdaBoost::setWeakClassifier ( const WeakClassifier weakClassifer)

Sets the WeakClassifier to use for boosting.

If this function is called, any previously set WeakClassifiers will be removed.

Returns
returns true if the WeakClassifier was added successfully, false otherwise

Definition at line 514 of file AdaBoost.cpp.

bool GRT::AdaBoost::train_ ( ClassificationData trainingData)
virtual

This trains the AdaBoost model, using the labelled classification data. This overrides the train function in the Classifier base class.

Parameters
ClassificationData&trainingData: a reference to the training data
Returns
returns true if the AdaBoost model was trained, false otherwise

Reimplemented from GRT::MLBase.

Definition at line 116 of file AdaBoost.cpp.


The documentation for this class was generated from the following files: