#include <WeightedSum.h>

Inheritance diagram for ProbabilisticSceneRecognition::WeightedSum:

Public Member Functions
double	calculateCost (boost::shared_ptr< SceneModel::Topology > instance)

	WeightedSum (unsigned int minFalsePositives, unsigned int maxFalsePositives, unsigned int minFalseNegatives, unsigned int maxFalseNegatives, double minAverageRecognitionRuntime, double maxAverageRecognitionRuntime, double alpha, double beta, double gamma)

	~WeightedSum ()

Private Member Functions
double	getNormalisedAverageRecognitionRuntime (double averageRecognitionRuntime)

double	getNormalisedFalseNegatives (unsigned int falseNegatives)

double	getNormalisedFalsePositives (unsigned int falsePositives)

Private Attributes
double	mAlpha

double	mBeta

double	mGamma

double	mMaxAverageRecognitionRuntime

unsigned int	mMaxFalseNegatives

unsigned int	mMaxFalsePositives

double	mMinAverageRecognitionRuntime

unsigned int	mMinFalseNegatives

unsigned int	mMinFalsePositives

Detailed Description

Calculates the cost of a topology as a weighted sum of its number of false positives and average recognition runtime on test sets.

Definition at line 30 of file WeightedSum.h.

Constructor & Destructor Documentation

ProbabilisticSceneRecognition::WeightedSum::WeightedSum	(	unsigned int	minFalsePositives,
		unsigned int	maxFalsePositives,
		unsigned int	minFalseNegatives,
		unsigned int	maxFalseNegatives,
		double	minAverageRecognitionRuntime,
		double	maxAverageRecognitionRuntime,
		double	alpha,
		double	beta,
		double	gamma
	)

Constructor.

Parameters

minFalsePositives	Minimum number of false positives for normalization.
maxFalsePositives	Maximum number of false positives for normalization.
minAverageRecognitionRuntime	Minimum average recognition runtime for normalization.
maxAverageRecognitionRuntime	Maximum average recognition runtime for normalization.
alpha	Weight of the normalized number of false positives.
beta	Weight of the normalized average recognition runtime.
gamma	Weight of the normalized number of false negatives.

Definition at line 22 of file WeightedSum.cpp.

ProbabilisticSceneRecognition::WeightedSum::~WeightedSum ( )

Destructor.

Definition at line 37 of file WeightedSum.cpp.

Member Function Documentation

double ProbabilisticSceneRecognition::WeightedSum::calculateCost ( boost::shared_ptr< SceneModel::Topology > instance )

Calculate the cost of the topology as weighted sum of its number of false positives and average recognition runtime on test sets. Stores the cost in the topology.

Parameters

instance the topology to calculate the cost for.

Returns: the cost of the topology.

Definition at line 41 of file WeightedSum.cpp.

double ProbabilisticSceneRecognition::WeightedSum::getNormalisedAverageRecognitionRuntime ( double averageRecognitionRuntime )

private

Normalize average recognition runtime.

Parameters

averageRecognitionRuntime the average recognition runtime.

Returns: the normalized average recognition runtime.

Definition at line 100 of file WeightedSum.cpp.

double ProbabilisticSceneRecognition::WeightedSum::getNormalisedFalseNegatives ( unsigned int falseNegatives )

private

Normalize number of false negatives.

Parameters

falsePositives the number of false negatives.

Returns: the normalized number of false negatives.

Definition at line 84 of file WeightedSum.cpp.

double ProbabilisticSceneRecognition::WeightedSum::getNormalisedFalsePositives ( unsigned int falsePositives )

private

Normalize number of false positives.

Parameters

falsePositives the number of false positives.

Returns: the normalized number of false positives.

Definition at line 68 of file WeightedSum.cpp.

Member Data Documentation

double ProbabilisticSceneRecognition::WeightedSum::mAlpha

private

Weight of the normalized number of false positives.

Definition at line 94 of file WeightedSum.h.

double ProbabilisticSceneRecognition::WeightedSum::mBeta

private

Weight of the normalized average recognition runtime.

Definition at line 98 of file WeightedSum.h.

double ProbabilisticSceneRecognition::WeightedSum::mGamma

private

Weight of the normalized number of false negatives.

Definition at line 102 of file WeightedSum.h.

double ProbabilisticSceneRecognition::WeightedSum::mMaxAverageRecognitionRuntime

private

Maximum average recognition runtime for normalization.

Definition at line 89 of file WeightedSum.h.

unsigned int ProbabilisticSceneRecognition::WeightedSum::mMaxFalseNegatives

private

Maximum number of false negatives for normalization.

Definition at line 80 of file WeightedSum.h.

unsigned int ProbabilisticSceneRecognition::WeightedSum::mMaxFalsePositives

private

Maximum number of false positives for normalization.

Definition at line 70 of file WeightedSum.h.

double ProbabilisticSceneRecognition::WeightedSum::mMinAverageRecognitionRuntime

private

Minimum average recognition runtime for normalization.

Definition at line 85 of file WeightedSum.h.

unsigned int ProbabilisticSceneRecognition::WeightedSum::mMinFalseNegatives

private

Minimum number of false negatives for normalization.

Definition at line 75 of file WeightedSum.h.

unsigned int ProbabilisticSceneRecognition::WeightedSum::mMinFalsePositives

private

Minimum number of false positives for normalization.

Definition at line 66 of file WeightedSum.h.

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

Public Member Functions

Private Member Functions

Private Attributes

Detailed Description

Constructor & Destructor Documentation

Member Function Documentation

Member Data Documentation