HashValueParzenModel

java.lang.Object

smile.neighbor.lsh.HashValueParzenModel

public class HashValueParzenModel extends Object

Hash value Parzen model for multi-probe hash.

Constructor Summary

Constructors

Constructor

Description

HashValueParzenModel(MultiProbeHash hash, MultiProbeSample[] samples, double sigma)

Constructor.
Method Summary

Modifier and Type

Method

Description

void

estimate(int m, double h)

Given a hash value h, estimate the Gaussian model (mean and variance) of neighbors existing in the corresponding bucket.

double

mean()

Returns the mean.

double

sd()

Returns the standard deviation.

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details
- HashValueParzenModel
  
  public HashValueParzenModel(MultiProbeHash hash, MultiProbeSample[] samples, double sigma)
  
  Constructor.
  
  Parameters:
  
  hash - the hash function.
  
  samples - training data for multi-probe LSH.
  
  sigma - the width of Parzen window.
Method Details
- estimate
  
  public void estimate(int m, double h)
  
  Given a hash value h, estimate the Gaussian model (mean and variance) of neighbors existing in the corresponding bucket.
  
  Parameters:
  
  m - the index of hash function.
  
  h - the given hash value.
- mean
  
  public double mean()
  
  Returns the mean.
  
  Returns:
  
  the mean.
- sd
  
  public double sd()
  
  Returns the standard deviation.
  
  Returns:
  
  the conditional standard deviation.