org.apache.mahout.math.random

Class Multinomial<T>

java.lang.Object

org.apache.mahout.math.random.Multinomial<T>

All Implemented Interfaces:

Iterable<T>, Sampler<T>

public final class Multinomial<T>
extends Object
implements Sampler<T>, Iterable<T>

Multinomial sampler that allows updates to element probabilities. The basic idea is that sampling is done by using a simple balanced tree. Probabilities are kept in the tree so that we can navigate to any leaf in log N time. Updates are simple because we can just propagate them upwards.

In order to facilitate access by value, we maintain an additional map from value to tree node.

Constructor Summary

Constructors

Constructor and Description
Multinomial()
Multinomial(Iterable<WeightedThing<T>> things)
Multinomial(com.google.common.collect.Multiset<T> counts)

Method Summary

Methods

Modifier and Type	Method and Description
void	add(T value, double w)
void	delete(T value)
double	getProbability(T value)
double	getWeight()
double	getWeight(T value)
Iterator<T>	iterator()
T	sample()
T	sample(double u)
void	set(T value, double newP)

Methods inherited from class java.lang.Object

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

Constructor Detail

Multinomial

public Multinomial()

Multinomial

public Multinomial(com.google.common.collect.Multiset<T> counts)

Multinomial

public Multinomial(Iterable<WeightedThing<T>> things)

Method Detail

add

public void add(T value,
       double w)

getWeight

public double getWeight(T value)

getProbability

public double getProbability(T value)

getWeight

public double getWeight()

delete

public void delete(T value)

set

public void set(T value,
       double newP)

sample

public T sample()

Specified by:

sample in interface Sampler<T>

sample

public T sample(double u)

iterator

public Iterator<T> iterator()

Specified by:

iterator in interface Iterable<T>