datafu.pig.hash.lsh.interfaces

Class LSHCreator

java.lang.Object

datafu.pig.hash.lsh.interfaces.LSHCreator

public abstract class LSHCreator
extends java.lang.Object

Create a Locality sensitive hash.

Constructor Summary

Constructors

Constructor and Description
LSHCreator(int dim, int numHashes, int numInternalRepetitions, long seed) Create a LSHCreator

Method Summary

Methods

Modifier and Type	Method and Description
LSHFamily	constructFamily(org.apache.commons.math.random.RandomGenerator rg)
protected abstract LSH	constructLSH(org.apache.commons.math.random.RandomGenerator rg)
org.apache.commons.math.random.RandomGenerator	createGenerator()
int	getDim()
int	getNumHashes()
int	getNumInternalRepetitions() Each locality sensitive hash is a composite of numInternalRepetitions LSHes (this is done to increase range of the LSH)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

LSHCreator

public LSHCreator(int dim,
          int numHashes,
          int numInternalRepetitions,
          long seed)

Create a LSHCreator

Parameters:

dim - The dimension of the vectors which the LSH will hash

numHashes - The number of locality sensitive hashes to create

numInternalRepetitions - Each locality sensitive hash is a composite of numInternalRepetitions LSHes (this is done to increase range of the LSH)

seed - The seed to use

Method Detail

constructLSH

protected abstract LSH constructLSH(org.apache.commons.math.random.RandomGenerator rg)
                             throws org.apache.commons.math.MathException

Throws:

org.apache.commons.math.MathException

getDim

public int getDim()

createGenerator

public org.apache.commons.math.random.RandomGenerator createGenerator()

getNumHashes

public int getNumHashes()

Returns:

The number of locality sensitive hashes to create

getNumInternalRepetitions

public int getNumInternalRepetitions()

Each locality sensitive hash is a composite of numInternalRepetitions LSHes (this is done to increase range of the LSH)

Returns:

The number of internal repetitions

constructFamily

public LSHFamily constructFamily(org.apache.commons.math.random.RandomGenerator rg)
                          throws org.apache.commons.math.MathException

Parameters:

rg - The random generator to use when constructing the family

Returns:

The family of locality sensitive hashes

Throws:

org.apache.commons.math.MathException - MathException