mlpack::neighbor::NSModel< SortPolicy > Class Template Reference

The NSModel class provides an easy way to serialize a model, abstracts away the different types of trees, and also reflects the NeighborSearch API. More...

#include <ns_model.hpp>

Public Types
enum	TreeTypes {   KD_TREE, COVER_TREE, R_TREE, R_STAR_TREE,   BALL_TREE, X_TREE, HILBERT_R_TREE, R_PLUS_TREE,   R_PLUS_PLUS_TREE, VP_TREE, RP_TREE, MAX_RP_TREE,   SPILL_TREE, UB_TREE, OCTREE }
	Enum type to identify each accepted tree type.

Public Member Functions
	NSModel (TreeTypes treeType=TreeTypes::KD_TREE, bool randomBasis=false)
	Initialize the NSModel with the given type and whether or not a random basis should be used. More...
	NSModel (const NSModel &other)
	Copy the given NSModel. More...
	NSModel (NSModel &&other)
	Take ownership of the given NSModel. More...
NSModel &	operator= (const NSModel &other)
	Copy the given NSModel. More...
NSModel &	operator= (NSModel &&other)
	Take ownership of the given NSModel. More...
	~NSModel ()
	Clean memory, if necessary.
template<typename Archive >
void	serialize (Archive &ar, const uint32_t)
	Serialize the neighbor search model. More...
const arma::mat &	Dataset () const
	Expose the dataset.
NeighborSearchMode	SearchMode () const
	Expose SearchMode. More...
NeighborSearchMode &	SearchMode ()
	Modify the search mode.
size_t	LeafSize () const
	Expose LeafSize.
size_t &	LeafSize ()
double	Tau () const
	Expose Tau.
double &	Tau ()
double	Rho () const
	Expose Rho.
double &	Rho ()
double	Epsilon () const
	Expose Epsilon.
double &	Epsilon ()
TreeTypes	TreeType () const
	Expose treeType.
TreeTypes &	TreeType ()
bool	RandomBasis () const
	Expose randomBasis.
bool &	RandomBasis ()
void	InitializeModel (const NeighborSearchMode searchMode, const double epsilon)
	Initialize the model type. (This does not perform any training.) More...
void	BuildModel (arma::mat &&referenceSet, const NeighborSearchMode searchMode, const double epsilon=0)
	Build the reference tree.
void	Search (arma::mat &&querySet, const size_t k, arma::Mat< size_t > &neighbors, arma::mat &distances)
	Perform neighbor search. The query set will be reordered.
void	Search (const size_t k, arma::Mat< size_t > &neighbors, arma::mat &distances)
	Perform monochromatic neighbor search. More...
std::string	TreeName () const
	Return a string representation of the current tree type. More...

Detailed Description

template<typename SortPolicy>
class mlpack::neighbor::NSModel< SortPolicy >

The NSModel class provides an easy way to serialize a model, abstracts away the different types of trees, and also reflects the NeighborSearch API.

This class is meant to be used by the command-line mlpack_knn and mlpack_kfn programs, and thus does not have the same complete functionality and flexibility as the NeighborSearch class. So if you are using it outside of mlpack_knn and mlpack_kfn, be aware that it is limited!

Template Parameters

SortPolicy

The sort policy for distances; see NearestNeighborSort.

Constructor & Destructor Documentation

NSModel() [1/3]

template<typename SortPolicy >

mlpack::neighbor::NSModel< SortPolicy >::NSModel	(	TreeTypes	treeType = TreeTypes::KD_TREE,
		bool	randomBasis = false
	)

Initialize the NSModel with the given type and whether or not a random basis should be used.

Parameters

treeType	Type of tree to use.
randomBasis	Whether or not to project the points onto a random basis before searching.

NSModel() [2/3]

template<typename SortPolicy >

mlpack::neighbor::NSModel< SortPolicy >::NSModel

(

const NSModel< SortPolicy > &

other

)

Copy the given NSModel.

Parameters

other

Model to copy.

NSModel() [3/3]

template<typename SortPolicy >

mlpack::neighbor::NSModel< SortPolicy >::NSModel

(

NSModel< SortPolicy > &&

other

)

Take ownership of the given NSModel.

Parameters

other

Model to take ownership of.

Member Function Documentation

InitializeModel()

template<typename SortPolicy >

void mlpack::neighbor::NSModel< SortPolicy >::InitializeModel	(	const NeighborSearchMode	searchMode,
		const double	epsilon
	)

Initialize the model type. (This does not perform any training.)

Initialize a model given the tree type. (No training happens here.)

operator=() [1/2]

template<typename SortPolicy >

NSModel< SortPolicy > & mlpack::neighbor::NSModel< SortPolicy >::operator=

(

const NSModel< SortPolicy > &

other

)

Copy the given NSModel.

Parameters

other

Model to copy.

operator=() [2/2]

template<typename SortPolicy >

NSModel< SortPolicy > & mlpack::neighbor::NSModel< SortPolicy >::operator=

(

NSModel< SortPolicy > &&

other

)

Take ownership of the given NSModel.

Parameters

other

Model to take ownership of.

Search()

template<typename SortPolicy >

void mlpack::neighbor::NSModel< SortPolicy >::Search	(	const size_t	k,
		arma::Mat< size_t > &	neighbors,
		arma::mat &	distances
	)

Perform monochromatic neighbor search.

Perform neighbor search.

SearchMode()

template<typename SortPolicy >

NeighborSearchMode mlpack::neighbor::NSModel< SortPolicy >::SearchMode

(

)

const

Expose SearchMode.

Access the search mode.

serialize()

template<typename SortPolicy >

template<typename Archive >

void mlpack::neighbor::NSModel< SortPolicy >::serialize	(	Archive &	ar,
		const uint32_t
	)

Serialize the neighbor search model.

Serialize the kNN model.

TreeName()

template<typename SortPolicy >

std::string mlpack::neighbor::NSModel< SortPolicy >::TreeName

(

)

const

Return a string representation of the current tree type.

Get the name of the tree type.

---

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

• src/mlpack/methods/neighbor_search/ns_model.hpp
• src/mlpack/methods/neighbor_search/ns_model_impl.hpp