mlpack::range::RangeSearchRules< MetricType, TreeType > Class Template Reference

Collaboration diagram for mlpack::range::RangeSearchRules< MetricType, TreeType >:
Collaboration graph
[legend]

List of all members.

Public Member Functions

 RangeSearchRules (const arma::mat &referenceSet, const arma::mat &querySet, const math::Range &range, std::vector< std::vector< size_t > > &neighbors, std::vector< std::vector< double > > &distances, MetricType &metric)
 Construct the RangeSearchRules object.
double BaseCase (const size_t queryIndex, const size_t referenceIndex)
 Compute the base case between the given query point and reference point.
double Rescore (TreeType &queryNode, TreeType &referenceNode, const double oldScore) const
 Re-evaluate the score for recursion order.
double Rescore (const size_t queryIndex, TreeType &referenceNode, const double oldScore) const
 Re-evaluate the score for recursion order.
double Score (TreeType &queryNode, TreeType &referenceNode)
 Get the score for recursion order.
double Score (const size_t queryIndex, TreeType &referenceNode)
 Get the score for recursion order.

Private Member Functions

void AddResult (const size_t queryIndex, TreeType &referenceNode)
 Add all the points in the given node to the results for the given query point.

Private Attributes

std::vector< std::vector
< double > > & 
distances
 The vector the resultant neighbor distances should be stored in.
size_t lastQueryIndex
 The last query index.
size_t lastReferenceIndex
 The last reference index.
MetricType & metric
 The instantiated metric.
std::vector< std::vector
< size_t > > & 
neighbors
 The vector the resultant neighbor indices should be stored in.
const arma::mat & querySet
 The query set.
const math::Rangerange
 The range of distances for which we are searching.
const arma::mat & referenceSet
 The reference set.

Detailed Description

template<typename MetricType, typename TreeType>
class mlpack::range::RangeSearchRules< MetricType, TreeType >

Definition at line 30 of file range_search_rules.hpp.


Constructor & Destructor Documentation

template<typename MetricType , typename TreeType >
mlpack::range::RangeSearchRules< MetricType, TreeType >::RangeSearchRules ( const arma::mat &  referenceSet,
const arma::mat &  querySet,
const math::Range range,
std::vector< std::vector< size_t > > &  neighbors,
std::vector< std::vector< double > > &  distances,
MetricType &  metric 
)

Construct the RangeSearchRules object.

This is usually done from within the RangeSearch class at search time.

Parameters:
referenceSet Set of reference data.
querySet Set of query data.
range Range to search for.
neighbors Vector to store resulting neighbors in.
distances Vector to store resulting distances in.
metric Instantiated metric.

Member Function Documentation

template<typename MetricType , typename TreeType >
void mlpack::range::RangeSearchRules< MetricType, TreeType >::AddResult ( const size_t  queryIndex,
TreeType &  referenceNode 
) [private]

Add all the points in the given node to the results for the given query point.

If the base case has already been calculated, we make sure to not add that to the results twice.

template<typename MetricType , typename TreeType >
double mlpack::range::RangeSearchRules< MetricType, TreeType >::BaseCase ( const size_t  queryIndex,
const size_t  referenceIndex 
)

Compute the base case between the given query point and reference point.

Parameters:
queryIndex Index of query point.
referenceIndex Index of reference point.
template<typename MetricType , typename TreeType >
double mlpack::range::RangeSearchRules< MetricType, TreeType >::Rescore ( TreeType &  queryNode,
TreeType &  referenceNode,
const double  oldScore 
) const

Re-evaluate the score for recursion order.

A low score indicates priority for recursion, while DBL_MAX indicates that the node should not be recursed into at all (it should be pruned). This is used when the score has already been calculated, but another recursion may have modified the bounds for pruning. So the old score is checked against the new pruning bound.

Parameters:
queryNode Candidate query node to recurse into.
referenceNode Candidate reference node to recurse into.
oldScore Old score produced by Score() (or Rescore()).
template<typename MetricType , typename TreeType >
double mlpack::range::RangeSearchRules< MetricType, TreeType >::Rescore ( const size_t  queryIndex,
TreeType &  referenceNode,
const double  oldScore 
) const

Re-evaluate the score for recursion order.

A low score indicates priority for recursion, while DBL_MAX indicates that the node should not be recursed into at all (it should be pruned). This is used when the score has already been calculated, but another recursion may have modified the bounds for pruning. So the old score is checked against the new pruning bound.

Parameters:
queryIndex Index of query point.
referenceNode Candidate node to be recursed into.
oldScore Old score produced by Score() (or Rescore()).
template<typename MetricType , typename TreeType >
double mlpack::range::RangeSearchRules< MetricType, TreeType >::Score ( TreeType &  queryNode,
TreeType &  referenceNode 
)

Get the score for recursion order.

A low score indicates priority for recursion, while DBL_MAX indicates that the node should not be recursed into at all (it should be pruned).

Parameters:
queryNode Candidate query node to recurse into.
referenceNode Candidate reference node to recurse into.
template<typename MetricType , typename TreeType >
double mlpack::range::RangeSearchRules< MetricType, TreeType >::Score ( const size_t  queryIndex,
TreeType &  referenceNode 
)

Get the score for recursion order.

A low score indicates priority for recursion, while DBL_MAX indicates that the node should not be recursed into at all (it should be pruned).

Parameters:
queryIndex Index of query point.
referenceNode Candidate node to be recursed into.

Member Data Documentation

template<typename MetricType , typename TreeType >
std::vector<std::vector<double> >& mlpack::range::RangeSearchRules< MetricType, TreeType >::distances [private]

The vector the resultant neighbor distances should be stored in.

Definition at line 123 of file range_search_rules.hpp.

template<typename MetricType , typename TreeType >
size_t mlpack::range::RangeSearchRules< MetricType, TreeType >::lastQueryIndex [private]

The last query index.

Definition at line 129 of file range_search_rules.hpp.

template<typename MetricType , typename TreeType >
size_t mlpack::range::RangeSearchRules< MetricType, TreeType >::lastReferenceIndex [private]

The last reference index.

Definition at line 131 of file range_search_rules.hpp.

template<typename MetricType , typename TreeType >
MetricType& mlpack::range::RangeSearchRules< MetricType, TreeType >::metric [private]

The instantiated metric.

Definition at line 126 of file range_search_rules.hpp.

template<typename MetricType , typename TreeType >
std::vector<std::vector<size_t> >& mlpack::range::RangeSearchRules< MetricType, TreeType >::neighbors [private]

The vector the resultant neighbor indices should be stored in.

Definition at line 120 of file range_search_rules.hpp.

template<typename MetricType , typename TreeType >
const arma::mat& mlpack::range::RangeSearchRules< MetricType, TreeType >::querySet [private]

The query set.

Definition at line 114 of file range_search_rules.hpp.

template<typename MetricType , typename TreeType >
const math::Range& mlpack::range::RangeSearchRules< MetricType, TreeType >::range [private]

The range of distances for which we are searching.

Definition at line 117 of file range_search_rules.hpp.

template<typename MetricType , typename TreeType >
const arma::mat& mlpack::range::RangeSearchRules< MetricType, TreeType >::referenceSet [private]

The reference set.

Definition at line 111 of file range_search_rules.hpp.


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

Generated on 29 Sep 2016 for MLPACK by  doxygen 1.6.1