A neighbor finder that specializes in dynamical systems. It determines neighbors by searching for the shortest path of actions between observations, and computes the distance as the number of time-steps in that path. This algorithm was published in Gashler, Michael S. and Martinez, Tony. Temporal nonlinear dimensionality reduction. In Proceedings of the International Joint Conference on Neural Networks IJCNN’11, pages 1959–1966, IEEE Press, 2011. More...

#include <GNeighborFinder.h>

Inheritance diagram for GClasses::GTemporalNeighborFinder:

Public Member Functions
	GTemporalNeighborFinder (GMatrix pObservations, GMatrix pActions, bool ownActionsData, size_t neighborCount, GRand *pRand, size_t maxDims=12)
	pObservations is typically a matrix of high-dimensional observations. pActions is a matrix of corresponding actions (peformed after the corresponding observation was observed). If ownActionsData is true, then this object will delete pActions when it is deleted. This neighbor-finder is somewhat slow in high-dimensional space. Consequently, if the data has more than maxDims dimensions, it will internally use PCA to reduce it to maxDims dimensions before computing neighbors. The default is 12. More...

virtual	~GTemporalNeighborFinder ()

virtual void	neighbors (size_t *pOutNeighbors, size_t index)
	Computes the neighbors of the specified vector. More...

virtual void	neighbors (size_t pOutNeighbors, double pOutDistances, size_t index)
	Computes the neighbors and distances of the specified vector. More...

Public Member Functions inherited from GClasses::GNeighborFinder
	GNeighborFinder (const GMatrix *pData, size_t neighborCount)

virtual	~GNeighborFinder ()

virtual bool	canGeneralize ()
	Returns true if this neighbor finder can operate on points that are not in the dataset passed to the constructor. More...

const GMatrix *	data ()
	Returns the data passed to the constructor of this object. More...

virtual bool	isCached ()
	Returns true iff the neighbors and distances are pre-computed. More...

size_t	neighborCount ()
	Returns the number of neighbors to find. More...

void	sortNeighbors (size_t pNeighbors, double pDistances)
	Uses Quick Sort to sort the neighbors from least to most dissimilar, followed by any slots for with INVALID_INDEX for the index. (Note: This method is pointless, since the neighors are already guaranteed to come in sorted order. Todo: figure out why it is still here) More...

Protected Member Functions
bool	findPath (size_t from, size_t to, double *path, double distCap)
	Returns false if distCap is exceeded, or if the results are too imprecise to be reliable. Otherwise, returns true, and path is set to contain the number of times that each action must be performed to travel from point "from" to point "to". More...

GMatrix *	preprocessObservations (GMatrix *pObs, size_t maxDims)
	This method uses PCA to reduce pObs to maxDims dimensions. (If pObs is already small enough, it just returns pObs.) More...

Protected Attributes
std::vector< GSupervisedLearner * >	m_consequenceMaps

size_t	m_maxDims

bool	m_ownActionsData

GMatrix *	m_pActions

GMatrix *	m_pPreprocessed

GRand *	m_pRand

Protected Attributes inherited from GClasses::GNeighborFinder
size_t	m_neighborCount

const GMatrix *	m_pData

Additional Inherited Members
Static Public Member Functions inherited from GClasses::GNeighborFinder
static void	sortNeighbors (size_t neighborCount, size_t pNeighbors, double pDistances)
	Uses Quick Sort to sort the neighbors from least to most dissimilar, followed by any slots for with INVALID_INDEX for the index. (Note: This method is pointless, since the neighors are already guaranteed to come in sorted order. Todo: figure out why it is still here) More...

Detailed Description

A neighbor finder that specializes in dynamical systems. It determines neighbors by searching for the shortest path of actions between observations, and computes the distance as the number of time-steps in that path. This algorithm was published in Gashler, Michael S. and Martinez, Tony. Temporal nonlinear dimensionality reduction. In Proceedings of the International Joint Conference on Neural Networks IJCNN’11, pages 1959–1966, IEEE Press, 2011.

Constructor & Destructor Documentation

GClasses::GTemporalNeighborFinder::GTemporalNeighborFinder	(	GMatrix *	pObservations,
		GMatrix *	pActions,
		bool	ownActionsData,
		size_t	neighborCount,
		GRand *	pRand,
		size_t	maxDims = `12`
	)

pObservations is typically a matrix of high-dimensional observations. pActions is a matrix of corresponding actions (peformed after the corresponding observation was observed). If ownActionsData is true, then this object will delete pActions when it is deleted. This neighbor-finder is somewhat slow in high-dimensional space. Consequently, if the data has more than maxDims dimensions, it will internally use PCA to reduce it to maxDims dimensions before computing neighbors. The default is 12.

virtual GClasses::GTemporalNeighborFinder::~GTemporalNeighborFinder ( )

virtual

Member Function Documentation

bool GClasses::GTemporalNeighborFinder::findPath	(	size_t	from,
		size_t	to,
		double *	path,
		double	distCap
	)

protected

Returns false if distCap is exceeded, or if the results are too imprecise to be reliable. Otherwise, returns true, and path is set to contain the number of times that each action must be performed to travel from point "from" to point "to".

virtual void GClasses::GTemporalNeighborFinder::neighbors	(	size_t *	pOutNeighbors,
		size_t	index
	)

virtual

Computes the neighbors of the specified vector.

Implements GClasses::GNeighborFinder.

virtual void GClasses::GTemporalNeighborFinder::neighbors	(	size_t *	pOutNeighbors,
		double *	pOutDistances,
		size_t	index
	)

virtual

Computes the neighbors and distances of the specified vector.

Implements GClasses::GNeighborFinder.

GMatrix* GClasses::GTemporalNeighborFinder::preprocessObservations	(	GMatrix *	pObs,
		size_t	maxDims
	)

protected

This method uses PCA to reduce pObs to maxDims dimensions. (If pObs is already small enough, it just returns pObs.)

Member Data Documentation

std::vector<GSupervisedLearner*> GClasses::GTemporalNeighborFinder::m_consequenceMaps

protected

size_t GClasses::GTemporalNeighborFinder::m_maxDims

protected

bool GClasses::GTemporalNeighborFinder::m_ownActionsData

protected

GMatrix* GClasses::GTemporalNeighborFinder::m_pActions

protected

GMatrix* GClasses::GTemporalNeighborFinder::m_pPreprocessed

protected

GRand* GClasses::GTemporalNeighborFinder::m_pRand

protected

Public Member Functions

Protected Member Functions

Protected Attributes

Additional Inherited Members

Detailed Description

Constructor & Destructor Documentation

Member Function Documentation

Member Data Documentation