rehashing
Public Member Functions | Public Attributes | Protected Member Functions | List of all members
AdaptiveHBE Class Reference

#include <AdaptiveHBE.h>

Inheritance diagram for AdaptiveHBE:
AdaptiveEstimator

Public Member Functions

 AdaptiveHBE (shared_ptr< MatrixXd > data, shared_ptr< Kernel > k, double lb, double eps, bool sketch)
 
- Public Member Functions inherited from AdaptiveEstimator
std::vector< double > query (VectorXd q)
 
void setMedians (int l)
 

Public Attributes

vector< UniformHBEu_levels
 
vector< SketchHBEs_levels
 
- Public Attributes inherited from AdaptiveEstimator
double totalTime = 0
 

Protected Member Functions

std::vector< double > evaluateQuery (VectorXd q, int level)
 

Additional Inherited Members

- Protected Attributes inherited from AdaptiveEstimator
int numPoints
 
double gamma = 0.5
 
int I
 
int L = 3
 
std::vector< double > mui
 
std::vector< int > Mi
 
std::vector< double > ti
 
std::vector< int > ki
 
std::vector< double > wi
 
double r
 
std::string EXP_STR = "exp"
 

Detailed Description

Adaptive sampling via HBE.

Constructor & Destructor Documentation

◆ AdaptiveHBE()

AdaptiveHBE::AdaptiveHBE ( shared_ptr< MatrixXd >  data,
shared_ptr< Kernel >  k,
double  lb,
double  eps,
bool  sketch 
)
Parameters
datadataset
kkernel
lbtau (minimum density)
epsrelative error
sketchif true, use HBS as a sketch; otherwise use uniform sampling as a sketch

Member Function Documentation

◆ evaluateQuery()

std::vector< double > AdaptiveHBE::evaluateQuery ( VectorXd  q,
int  level 
)
protectedvirtual

For subclasses to implement: evaluate density of query q at the given level.

Implements AdaptiveEstimator.

Member Data Documentation

◆ s_levels

vector<SketchHBE> AdaptiveHBE::s_levels

Collection of estimators at all levels. At each level, we have a collection of hashing tables created by evaluating i.i.d hash function with a partiuclar hashing schemes on samples drawn from HBS.

◆ u_levels

vector<UniformHBE> AdaptiveHBE::u_levels

Collection of estimators at all levels. At each level, we have a collection of hashing tables created by evaluating i.i.d hash function with a partiuclar hashing schemes on random samples from the dataset.


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