HMLP: High-performance Machine Learning Primitives
hmlp::tree::Setup< SPLITTER, DATATYPE > Class Template Reference

Data and setup that are shared with all nodes. More...

#include <tree.hpp>

Public Types

typedef DATATYPE T
 

Public Member Functions

vector< size_t > ContainAny (vector< size_t > &queries, size_t target)
 Check if this node contain any query using morton. Notice that queries[] contains gids; thus, morton[] needs to be accessed using gids. More...
 

Public Attributes

size_t m = 0
 
size_t max_depth = 15
 
Data< pair< T, size_t > > * NN = NULL
 
vector< size_t > morton
 
SPLITTER splitter
 

Detailed Description

template<typename SPLITTER, typename DATATYPE>
class hmlp::tree::Setup< SPLITTER, DATATYPE >

Data and setup that are shared with all nodes.

end class Node

Member Function Documentation

template<typename SPLITTER, typename DATATYPE>
vector<size_t> hmlp::tree::Setup< SPLITTER, DATATYPE >::ContainAny ( vector< size_t > &  queries,
size_t  target 
)
inline

Check if this node contain any query using morton. Notice that queries[] contains gids; thus, morton[] needs to be accessed using gids.

notice that setup->morton only contains local morton ids

Member Data Documentation

template<typename SPLITTER, typename DATATYPE>
size_t hmlp::tree::Setup< SPLITTER, DATATYPE >::m = 0

maximum leaf node size

template<typename SPLITTER, typename DATATYPE>
size_t hmlp::tree::Setup< SPLITTER, DATATYPE >::max_depth = 15

by default we use 4 bits = 0-15 levels

template<typename SPLITTER, typename DATATYPE>
vector<size_t> hmlp::tree::Setup< SPLITTER, DATATYPE >::morton

MortonIDs of all indices.

template<typename SPLITTER, typename DATATYPE>
Data<pair<T, size_t> >* hmlp::tree::Setup< SPLITTER, DATATYPE >::NN = NULL

Coordinates (accessed with gids) neighbors<distance, gid> (accessed with gids)

template<typename SPLITTER, typename DATATYPE>
SPLITTER hmlp::tree::Setup< SPLITTER, DATATYPE >::splitter

Tree splitter


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