mlpack
master
|
A cover tree is a tree specifically designed to speed up nearest-neighbor computation in high-dimensional spaces. More...
Classes | |
class | DualTreeTraverser |
A dual-tree cover tree traverser; see dual_tree_traverser.hpp. More... | |
class | SingleTreeTraverser |
A single-tree cover tree traverser; see single_tree_traverser.hpp for implementation. More... | |
Public Types | |
template<typename RuleType > | |
using | BreadthFirstDualTreeTraverser = DualTreeTraverser< RuleType > |
typedef MatType::elem_type | ElemType |
The type held by the matrix type. More... | |
typedef MatType | Mat |
So that other classes can access the matrix type. More... | |
Public Member Functions | |
CoverTree (const MatType &dataset, const ElemType base=2.0, MetricType *metric=NULL) | |
Create the cover tree with the given dataset and given base. More... | |
CoverTree (const MatType &dataset, MetricType &metric, const ElemType base=2.0) | |
Create the cover tree with the given dataset and the given instantiated metric. More... | |
CoverTree (MatType &&dataset, const ElemType base=2.0) | |
Create the cover tree with the given dataset, taking ownership of the dataset. More... | |
CoverTree (MatType &&dataset, MetricType &metric, const ElemType base=2.0) | |
Create the cover tree with the given dataset and the given instantiated metric, taking ownership of the dataset. More... | |
CoverTree (const MatType &dataset, const ElemType base, const size_t pointIndex, const int scale, CoverTree *parent, const ElemType parentDistance, arma::Col< size_t > &indices, arma::vec &distances, size_t nearSetSize, size_t &farSetSize, size_t &usedSetSize, MetricType &metric=NULL) | |
Construct a child cover tree node. More... | |
CoverTree (const MatType &dataset, const ElemType base, const size_t pointIndex, const int scale, CoverTree *parent, const ElemType parentDistance, const ElemType furthestDescendantDistance, MetricType *metric=NULL) | |
Manually construct a cover tree node; no tree assembly is done in this constructor, and children must be added manually (use Children()). More... | |
CoverTree (const CoverTree &other) | |
Create a cover tree from another tree. More... | |
CoverTree (CoverTree &&other) | |
Move constructor for a Cover Tree, possess all the members of the given tree. More... | |
template<typename Archive > | |
CoverTree (Archive &ar, const typename std::enable_if_t< Archive::is_loading::value > *=0) | |
Create a cover tree from a boost::serialization archive. More... | |
~CoverTree () | |
Delete this cover tree node and its children. More... | |
ElemType | Base () const |
Get the base. More... | |
ElemType & | Base () |
Modify the base; don't do this, you'll break everything. More... | |
void | Center (arma::vec ¢er) const |
Get the center of the node and store it in the given vector. More... | |
const CoverTree & | Child (const size_t index) const |
Get a particular child node. More... | |
CoverTree & | Child (const size_t index) |
Modify a particular child node. More... | |
CoverTree *& | ChildPtr (const size_t index) |
const std::vector< CoverTree * > & | Children () const |
Get the children. More... | |
std::vector< CoverTree * > & | Children () |
Modify the children manually (maybe not a great idea). More... | |
const MatType & | Dataset () const |
Get a reference to the dataset. More... | |
size_t | Descendant (const size_t index) const |
Get the index of a particular descendant point. More... | |
size_t | DistanceComps () const |
size_t & | DistanceComps () |
ElemType | FurthestDescendantDistance () const |
Get the distance from the center of the node to the furthest descendant. More... | |
ElemType & | FurthestDescendantDistance () |
Modify the distance from the center of the node to the furthest descendant. More... | |
ElemType | FurthestPointDistance () const |
Get the distance to the furthest point. This is always 0 for cover trees. More... | |
template<typename VecType > | |
size_t | GetFurthestChild (const VecType &point, typename std::enable_if_t< IsVector< VecType >::value > *=0) |
Return the index of the furthest child node to the given query point. More... | |
size_t | GetFurthestChild (const CoverTree &queryNode) |
Return the index of the furthest child node to the given query node. More... | |
template<typename VecType > | |
size_t | GetNearestChild (const VecType &point, typename std::enable_if_t< IsVector< VecType >::value > *=0) |
Return the index of the nearest child node to the given query point. More... | |
size_t | GetNearestChild (const CoverTree &queryNode) |
Return the index of the nearest child node to the given query node. More... | |
bool | IsLeaf () const |
ElemType | MaxDistance (const CoverTree &other) const |
Return the maximum distance to another node. More... | |
ElemType | MaxDistance (const CoverTree &other, const ElemType distance) const |
Return the maximum distance to another node given that the point-to-point distance has already been calculated. More... | |
ElemType | MaxDistance (const arma::vec &other) const |
Return the maximum distance to another point. More... | |
ElemType | MaxDistance (const arma::vec &other, const ElemType distance) const |
Return the maximum distance to another point given that the distance from the center to the point has already been calculated. More... | |
MetricType & | Metric () const |
Get the instantiated metric. More... | |
ElemType | MinDistance (const CoverTree &other) const |
Return the minimum distance to another node. More... | |
ElemType | MinDistance (const CoverTree &other, const ElemType distance) const |
Return the minimum distance to another node given that the point-to-point distance has already been calculated. More... | |
ElemType | MinDistance (const arma::vec &other) const |
Return the minimum distance to another point. More... | |
ElemType | MinDistance (const arma::vec &other, const ElemType distance) const |
Return the minimum distance to another point given that the distance from the center to the point has already been calculated. More... | |
ElemType | MinimumBoundDistance () const |
Get the minimum distance from the center to any bound edge (this is the same as furthestDescendantDistance). More... | |
size_t | NumChildren () const |
Get the number of children. More... | |
size_t | NumDescendants () const |
Get the number of descendant points. More... | |
size_t | NumPoints () const |
CoverTree * | Parent () const |
Get the parent node. More... | |
CoverTree *& | Parent () |
Modify the parent node. More... | |
ElemType | ParentDistance () const |
Get the distance to the parent. More... | |
ElemType & | ParentDistance () |
Modify the distance to the parent. More... | |
size_t | Point () const |
Get the index of the point which this node represents. More... | |
size_t | Point (const size_t) const |
For compatibility with other trees; the argument is ignored. More... | |
math::RangeType< ElemType > | RangeDistance (const CoverTree &other) const |
Return the minimum and maximum distance to another node. More... | |
math::RangeType< ElemType > | RangeDistance (const CoverTree &other, const ElemType distance) const |
Return the minimum and maximum distance to another node given that the point-to-point distance has already been calculated. More... | |
math::RangeType< ElemType > | RangeDistance (const arma::vec &other) const |
Return the minimum and maximum distance to another point. More... | |
math::RangeType< ElemType > | RangeDistance (const arma::vec &other, const ElemType distance) const |
Return the minimum and maximum distance to another point given that the point-to-point distance has already been calculated. More... | |
int | Scale () const |
Get the scale of this node. More... | |
int & | Scale () |
Modify the scale of this node. Be careful... More... | |
template<typename Archive > | |
void | Serialize (Archive &ar, const unsigned int) |
Serialize the tree. More... | |
const StatisticType & | Stat () const |
Get the statistic for this node. More... | |
StatisticType & | Stat () |
Modify the statistic for this node. More... | |
Protected Member Functions | |
CoverTree () | |
A default constructor. More... | |
Private Member Functions | |
void | ComputeDistances (const size_t pointIndex, const arma::Col< size_t > &indices, arma::vec &distances, const size_t pointSetSize) |
Fill the vector of distances with the distances between the point specified by pointIndex and each point in the indices array. More... | |
void | CreateChildren (arma::Col< size_t > &indices, arma::vec &distances, size_t nearSetSize, size_t &farSetSize, size_t &usedSetSize) |
Create the children for this node. More... | |
void | MoveToUsedSet (arma::Col< size_t > &indices, arma::vec &distances, size_t &nearSetSize, size_t &farSetSize, size_t &usedSetSize, arma::Col< size_t > &childIndices, const size_t childFarSetSize, const size_t childUsedSetSize) |
size_t | PruneFarSet (arma::Col< size_t > &indices, arma::vec &distances, const ElemType bound, const size_t nearSetSize, const size_t pointSetSize) |
void | RemoveNewImplicitNodes () |
Take a look at the last child (the most recently created one) and remove any implicit nodes that have been created. More... | |
size_t | SortPointSet (arma::Col< size_t > &indices, arma::vec &distances, const size_t childFarSetSize, const size_t childUsedSetSize, const size_t farSetSize) |
Assuming that the list of indices and distances is sorted as [ childFarSet | childUsedSet | farSet | usedSet ], resort the sets so the organization is [ childFarSet | farSet | childUsedSet | usedSet ]. More... | |
size_t | SplitNearFar (arma::Col< size_t > &indices, arma::vec &distances, const ElemType bound, const size_t pointSetSize) |
Split the given indices and distances into a near and a far set, returning the number of points in the near set. More... | |
Private Attributes | |
ElemType | base |
The base used to construct the tree. More... | |
std::vector< CoverTree * > | children |
The list of children; the first is the self-child. More... | |
const MatType * | dataset |
Reference to the matrix which this tree is built on. More... | |
size_t | distanceComps |
ElemType | furthestDescendantDistance |
Distance to the furthest descendant. More... | |
bool | localDataset |
If true, we own the dataset and need to destroy it in the destructor. More... | |
bool | localMetric |
Whether or not we need to destroy the metric in the destructor. More... | |
MetricType * | metric |
The metric used for this tree. More... | |
size_t | numDescendants |
The number of descendant points. More... | |
CoverTree * | parent |
The parent node (NULL if this is the root of the tree). More... | |
ElemType | parentDistance |
Distance to the parent. More... | |
size_t | point |
Index of the point in the matrix which this node represents. More... | |
int | scale |
Scale level of the node. More... | |
StatisticType | stat |
The instantiated statistic. More... | |
A cover tree is a tree specifically designed to speed up nearest-neighbor computation in high-dimensional spaces.
Each non-leaf node references a point and has a nonzero number of children, including a "self-child" which references the same point. A leaf node represents only one point.
The tree can be thought of as a hierarchy with the root node at the top level and the leaf nodes at the bottom level. Each level in the tree has an assigned 'scale' i. The tree follows these two invariants:
Note that in the cover tree paper, there is a third invariant (the 'separation invariant'), but that does not apply to our implementation, because we have relaxed the invariant.
The value 'b' refers to the base, which is a parameter of the tree. These three properties make the cover tree very good for fast, high-dimensional nearest-neighbor search.
The theoretical structure of the tree contains many 'implicit' nodes which only have a "self-child" (a child referencing the same point, but at a lower scale level). This practical implementation only constructs explicit nodes – non-leaf nodes with more than one child. A leaf node has no children, and its scale level is INT_MIN.
For more information on cover trees, see
For information on runtime bounds of the nearest-neighbor computation using cover trees, see the following paper, presented at NIPS 2009:
The CoverTree class offers three template parameters; a custom metric type can be used with MetricType (this class defaults to the L2-squared metric). The root node's point can be chosen with the RootPointPolicy; by default, the FirstPointIsRoot policy is used, meaning the first point in the dataset is used. The StatisticType policy allows you to define statistics which can be gathered during the creation of the tree.
MetricType | Metric type to use during tree construction. |
RootPointPolicy | Determines which point to use as the root node. |
StatisticType | Statistic to be used during tree creation. |
MatType | Type of matrix to build the tree on (generally mat or sp_mat). |
Definition at line 99 of file cover_tree.hpp.
using mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::BreadthFirstDualTreeTraverser = DualTreeTraverser<RuleType> |
Definition at line 264 of file cover_tree.hpp.
typedef MatType::elem_type mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::ElemType |
The type held by the matrix type.
Definition at line 105 of file cover_tree.hpp.
typedef MatType mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Mat |
So that other classes can access the matrix type.
Definition at line 103 of file cover_tree.hpp.
mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::CoverTree | ( | const MatType & | dataset, |
const ElemType | base = 2.0 , |
||
MetricType * | metric = NULL |
||
) |
Create the cover tree with the given dataset and given base.
The dataset will not be modified during the building procedure (unlike BinarySpaceTree).
The last argument will be removed in mlpack 1.1.0 (see #274 and #273).
dataset | Reference to the dataset to build a tree on. |
base | Base to use during tree building (default 2.0). |
mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::CoverTree | ( | const MatType & | dataset, |
MetricType & | metric, | ||
const ElemType | base = 2.0 |
||
) |
Create the cover tree with the given dataset and the given instantiated metric.
Optionally, set the base. The dataset will not be modified during the building procedure (unlike BinarySpaceTree).
dataset | Reference to the dataset to build a tree on. |
metric | Instantiated metric to use during tree building. |
base | Base to use during tree building (default 2.0). |
mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::CoverTree | ( | MatType && | dataset, |
const ElemType | base = 2.0 |
||
) |
Create the cover tree with the given dataset, taking ownership of the dataset.
Optionally, set the base.
dataset | Reference to the dataset to build a tree on. |
base | Base to use during tree building (default 2.0). |
mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::CoverTree | ( | MatType && | dataset, |
MetricType & | metric, | ||
const ElemType | base = 2.0 |
||
) |
Create the cover tree with the given dataset and the given instantiated metric, taking ownership of the dataset.
Optionally, set the base.
dataset | Reference to the dataset to build a tree on. |
metric | Instantiated metric to use during tree building. |
base | Base to use during tree building (default 2.0). |
mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::CoverTree | ( | const MatType & | dataset, |
const ElemType | base, | ||
const size_t | pointIndex, | ||
const int | scale, | ||
CoverTree< MetricType, StatisticType, MatType, RootPointPolicy > * | parent, | ||
const ElemType | parentDistance, | ||
arma::Col< size_t > & | indices, | ||
arma::vec & | distances, | ||
size_t | nearSetSize, | ||
size_t & | farSetSize, | ||
size_t & | usedSetSize, | ||
MetricType & | metric = NULL |
||
) |
Construct a child cover tree node.
This constructor is not meant to be used externally, but it could be used to insert another node into a tree. This procedure uses only one vector for the near set, the far set, and the used set (this is to prevent unnecessary memory allocation in recursive calls to this constructor). Therefore, the size of the near set, far set, and used set must be passed in. The near set will be entirely used up, and some of the far set may be used. The value of usedSetSize will be set to the number of points used in the construction of this node, and the value of farSetSize will be modified to reflect the number of points in the far set after the construction of this node.
If you are calling this manually, be careful that the given scale is as small as possible, or you may be creating an implicit node in your tree.
dataset | Reference to the dataset to build a tree on. |
base | Base to use during tree building. |
pointIndex | Index of the point this node references. |
scale | Scale of this level in the tree. |
parent | Parent of this node (NULL indicates no parent). |
parentDistance | Distance to the parent node. |
indices | Array of indices, ordered [ nearSet | farSet | usedSet ]; will be modified to [ farSet | usedSet ]. |
distances | Array of distances, ordered the same way as the indices. These represent the distances between the point specified by pointIndex and each point in the indices array. |
nearSetSize | Size of the near set; if 0, this will be a leaf. |
farSetSize | Size of the far set; may be modified (if this node uses any points in the far set). |
usedSetSize | The number of points used will be added to this number. |
mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::CoverTree | ( | const MatType & | dataset, |
const ElemType | base, | ||
const size_t | pointIndex, | ||
const int | scale, | ||
CoverTree< MetricType, StatisticType, MatType, RootPointPolicy > * | parent, | ||
const ElemType | parentDistance, | ||
const ElemType | furthestDescendantDistance, | ||
MetricType * | metric = NULL |
||
) |
Manually construct a cover tree node; no tree assembly is done in this constructor, and children must be added manually (use Children()).
This constructor is useful when the tree is being "imported" into the CoverTree class after being created in some other manner.
dataset | Reference to the dataset this node is a part of. |
base | Base that was used for tree building. |
pointIndex | Index of the point in the dataset which this node refers to. |
scale | Scale of this node's level in the tree. |
parent | Parent node (NULL indicates no parent). |
parentDistance | Distance to parent node point. |
furthestDescendantDistance | Distance to furthest descendant point. |
metric | Instantiated metric (optional). |
mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::CoverTree | ( | const CoverTree< MetricType, StatisticType, MatType, RootPointPolicy > & | other | ) |
Create a cover tree from another tree.
Be careful! This may use a lot of memory and take a lot of time. This will also make a copy of the dataset.
other | Cover tree to copy from. |
mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::CoverTree | ( | CoverTree< MetricType, StatisticType, MatType, RootPointPolicy > && | other | ) |
Move constructor for a Cover Tree, possess all the members of the given tree.
other | Cover Tree to move. |
mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::CoverTree | ( | Archive & | ar, |
const typename std::enable_if_t< Archive::is_loading::value > * | = 0 |
||
) |
Create a cover tree from a boost::serialization archive.
mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::~CoverTree | ( | ) |
Delete this cover tree node and its children.
|
protected |
A default constructor.
This is meant to only be used with boost::serialization, which is allowed with the friend declaration below. This does not return a valid tree! This method must be protected, so that the serialization shim can work with the default constructor.
|
inline |
Get the base.
Definition at line 304 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::base.
|
inline |
Modify the base; don't do this, you'll break everything.
Definition at line 306 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::base.
|
inline |
Get the center of the node and store it in the given vector.
Definition at line 412 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::dataset.
|
inline |
Get a particular child node.
Definition at line 278 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::children.
|
inline |
Modify a particular child node.
Definition at line 280 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::children.
|
inline |
Definition at line 282 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::children.
|
inline |
Get the children.
Definition at line 288 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::children.
|
inline |
Modify the children manually (maybe not a great idea).
Definition at line 290 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::children, mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Descendant(), and mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::NumDescendants().
|
private |
Fill the vector of distances with the distances between the point specified by pointIndex and each point in the indices array.
The distances of the first pointSetSize points in indices are calculated (so, this does not necessarily need to use all of the points in the arrays).
pointIndex | Point to build the distances for. |
indices | List of indices to compute distances for. |
distances | Vector to store calculated distances in. |
pointSetSize | Number of points in arrays to calculate distances for. |
|
private |
Create the children for this node.
|
inline |
Get a reference to the dataset.
Definition at line 267 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::dataset.
size_t mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Descendant | ( | const size_t | index | ) | const |
Get the index of a particular descendant point.
Referenced by mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Children().
|
inline |
Definition at line 555 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::distanceComps.
|
inline |
Definition at line 556 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::distanceComps.
|
inline |
Get the distance from the center of the node to the furthest descendant.
Definition at line 401 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::furthestDescendantDistance.
|
inline |
Modify the distance from the center of the node to the furthest descendant.
Definition at line 405 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::furthestDescendantDistance.
|
inline |
Get the distance to the furthest point. This is always 0 for cover trees.
Definition at line 398 of file cover_tree.hpp.
size_t mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::GetFurthestChild | ( | const VecType & | point, |
typename std::enable_if_t< IsVector< VecType >::value > * | = 0 |
||
) |
Return the index of the furthest child node to the given query point.
If this is a leaf node, it will return NumChildren() (invalid index).
Referenced by mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Stat().
size_t mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::GetFurthestChild | ( | const CoverTree< MetricType, StatisticType, MatType, RootPointPolicy > & | queryNode | ) |
Return the index of the furthest child node to the given query node.
If it can't decide, it will return NumChildren() (invalid index).
size_t mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::GetNearestChild | ( | const VecType & | point, |
typename std::enable_if_t< IsVector< VecType >::value > * | = 0 |
||
) |
Return the index of the nearest child node to the given query point.
If this is a leaf node, it will return NumChildren() (invalid index).
Referenced by mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Stat().
size_t mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::GetNearestChild | ( | const CoverTree< MetricType, StatisticType, MatType, RootPointPolicy > & | queryNode | ) |
Return the index of the nearest child node to the given query node.
If it can't decide, it will return NumChildren() (invalid index).
|
inline |
Definition at line 274 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::children.
ElemType mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::MaxDistance | ( | const CoverTree< MetricType, StatisticType, MatType, RootPointPolicy > & | other | ) | const |
Return the maximum distance to another node.
Referenced by mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Stat().
ElemType mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::MaxDistance | ( | const CoverTree< MetricType, StatisticType, MatType, RootPointPolicy > & | other, |
const ElemType | distance | ||
) | const |
Return the maximum distance to another node given that the point-to-point distance has already been calculated.
ElemType mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::MaxDistance | ( | const arma::vec & | other | ) | const |
Return the maximum distance to another point.
ElemType mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::MaxDistance | ( | const arma::vec & | other, |
const ElemType | distance | ||
) | const |
Return the maximum distance to another point given that the distance from the center to the point has already been calculated.
|
inline |
Get the instantiated metric.
Definition at line 418 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::metric.
ElemType mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::MinDistance | ( | const CoverTree< MetricType, StatisticType, MatType, RootPointPolicy > & | other | ) | const |
Return the minimum distance to another node.
Referenced by mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Stat().
ElemType mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::MinDistance | ( | const CoverTree< MetricType, StatisticType, MatType, RootPointPolicy > & | other, |
const ElemType | distance | ||
) | const |
Return the minimum distance to another node given that the point-to-point distance has already been calculated.
ElemType mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::MinDistance | ( | const arma::vec & | other | ) | const |
Return the minimum distance to another point.
ElemType mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::MinDistance | ( | const arma::vec & | other, |
const ElemType | distance | ||
) | const |
Return the minimum distance to another point given that the distance from the center to the point has already been calculated.
|
inline |
Get the minimum distance from the center to any bound edge (this is the same as furthestDescendantDistance).
Definition at line 409 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::furthestDescendantDistance.
|
private |
|
inline |
Get the number of children.
Definition at line 285 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::children.
size_t mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::NumDescendants | ( | ) | const |
Get the number of descendant points.
Referenced by mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Children().
|
inline |
Definition at line 275 of file cover_tree.hpp.
|
inline |
Get the parent node.
Definition at line 388 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::parent.
|
inline |
Modify the parent node.
Definition at line 390 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::parent.
|
inline |
Get the distance to the parent.
Definition at line 393 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::parentDistance.
|
inline |
Modify the distance to the parent.
Definition at line 395 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::parentDistance.
|
inline |
Get the index of the point which this node represents.
Definition at line 270 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::point.
|
inline |
For compatibility with other trees; the argument is ignored.
Definition at line 272 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::point.
|
private |
math::RangeType<ElemType> mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::RangeDistance | ( | const CoverTree< MetricType, StatisticType, MatType, RootPointPolicy > & | other | ) | const |
Return the minimum and maximum distance to another node.
Referenced by mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Stat().
math::RangeType<ElemType> mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::RangeDistance | ( | const CoverTree< MetricType, StatisticType, MatType, RootPointPolicy > & | other, |
const ElemType | distance | ||
) | const |
Return the minimum and maximum distance to another node given that the point-to-point distance has already been calculated.
math::RangeType<ElemType> mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::RangeDistance | ( | const arma::vec & | other | ) | const |
Return the minimum and maximum distance to another point.
math::RangeType<ElemType> mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::RangeDistance | ( | const arma::vec & | other, |
const ElemType | distance | ||
) | const |
Return the minimum and maximum distance to another point given that the point-to-point distance has already been calculated.
|
private |
Take a look at the last child (the most recently created one) and remove any implicit nodes that have been created.
|
inline |
Get the scale of this node.
Definition at line 299 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::scale.
|
inline |
Modify the scale of this node. Be careful...
Definition at line 301 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::scale.
void mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Serialize | ( | Archive & | ar, |
const unsigned | int | ||
) |
Serialize the tree.
|
private |
Assuming that the list of indices and distances is sorted as [ childFarSet | childUsedSet | farSet | usedSet ], resort the sets so the organization is [ childFarSet | farSet | childUsedSet | usedSet ].
The size_t parameters specify the sizes of each set in the array. Only the ordering of the indices and distances arrays will be modified (not their actual contents).
The size of any of the four sets can be zero and this method will handle that case accordingly.
indices | List of indices to sort. |
distances | List of distances to sort. |
childFarSetSize | Number of points in child far set (childFarSet). |
childUsedSetSize | Number of points in child used set (childUsedSet). |
farSetSize | Number of points in far set (farSet). |
|
private |
Split the given indices and distances into a near and a far set, returning the number of points in the near set.
The distances must already be initialized. This will order the indices and distances such that the points in the near set make up the first part of the array and the far set makes up the rest: [ nearSet | farSet ].
indices | List of indices; will be reordered. |
distances | List of distances; will be reordered. |
bound | If the distance is less than or equal to this bound, the point is placed into the near set. |
pointSetSize | Size of point set (because we may be sorting a smaller list than the indices vector will hold). |
|
inline |
Get the statistic for this node.
Definition at line 309 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::stat.
|
inline |
Modify the statistic for this node.
Definition at line 311 of file cover_tree.hpp.
References mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::GetFurthestChild(), mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::GetNearestChild(), mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::MaxDistance(), mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::MinDistance(), mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::point, mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::RangeDistance(), and mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::stat.
|
private |
The base used to construct the tree.
Definition at line 430 of file cover_tree.hpp.
Referenced by mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Base().
|
private |
The list of children; the first is the self-child.
Definition at line 426 of file cover_tree.hpp.
Referenced by mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Child(), mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::ChildPtr(), mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Children(), mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::IsLeaf(), and mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::NumChildren().
|
private |
Reference to the matrix which this tree is built on.
Definition at line 422 of file cover_tree.hpp.
Referenced by mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Center(), and mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Dataset().
|
private |
Definition at line 559 of file cover_tree.hpp.
Referenced by mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::DistanceComps().
|
private |
Distance to the furthest descendant.
Definition at line 440 of file cover_tree.hpp.
Referenced by mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::FurthestDescendantDistance(), and mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::MinimumBoundDistance().
|
private |
If true, we own the dataset and need to destroy it in the destructor.
Definition at line 444 of file cover_tree.hpp.
|
private |
Whether or not we need to destroy the metric in the destructor.
Definition at line 442 of file cover_tree.hpp.
|
private |
The metric used for this tree.
Definition at line 446 of file cover_tree.hpp.
Referenced by mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Metric().
|
private |
The number of descendant points.
Definition at line 434 of file cover_tree.hpp.
|
private |
The parent node (NULL if this is the root of the tree).
Definition at line 436 of file cover_tree.hpp.
Referenced by mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Parent().
|
private |
Distance to the parent.
Definition at line 438 of file cover_tree.hpp.
Referenced by mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::ParentDistance().
|
private |
Index of the point in the matrix which this node represents.
Definition at line 424 of file cover_tree.hpp.
Referenced by mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Point(), and mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Stat().
|
private |
Scale level of the node.
Definition at line 428 of file cover_tree.hpp.
Referenced by mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Scale().
|
private |
The instantiated statistic.
Definition at line 432 of file cover_tree.hpp.
Referenced by mlpack::tree::CoverTree< MetricType, StatisticType, MatType, RootPointPolicy >::Stat().