jMEF.HierarchicalMixtureModel Class Reference

List of all members.

Public Member Functions
	HierarchicalMixtureModel ()
	Class constructor.
MixtureModel	getResolution (int resolution)
	Extracts a mixture model for a given resolution from the hierarchical mixture model.
MixtureModel	getOptimalMixtureModel (double t)
	Computes the "optimal" mixture model given a threshold on the KLD (estimated using a Monte-Carlo method).
MixtureModel	getOptimalMixtureModel (double t, int n)
	Computes the "optimal" mixture model given a threshold on the KLD (estimated using a Monte-Carlo method).
Public Attributes
ExponentialFamily	EF
	Exponential family member.
double	weight
	Weight of the tree.
MixtureModel	node
	Node containing a mixture model.
HierarchicalMixtureModel	parent
	Parent of the node.
HierarchicalMixtureModel	leftChild
	Left child of the node.
HierarchicalMixtureModel	rightChild
	Right child of the node.
Clustering.CLUSTERING_TYPE	type
	Type of the Bregman divergence.
int	resolutionMax
	Maximum resolution of the hierarchical mixture model.

---

Detailed Description

Author:

Vincent Garcia

Frank Nielsen

Version:

1.0

License

See file LICENSE.txt

Description

A hierarchical mixture model is a hierarchical structure (tree) containing the elements of a mixture model. A HierarchicalMixtureModel object is created by the BregmanHierarchicalClustering class.

---

Constructor & Destructor Documentation

jMEF.HierarchicalMixtureModel.HierarchicalMixtureModel

(

)

Class constructor.

---

Member Function Documentation

MixtureModel jMEF.HierarchicalMixtureModel.getOptimalMixtureModel	(	double	t,
		int	n
	)

Computes the "optimal" mixture model given a threshold on the KLD (estimated using a Monte-Carlo method).

Parameters:

	t	maximum value of the KLD between the initial and the optimal models
	n	number of points drawn from the initial mixture model

Returns:

optimal mixture model

MixtureModel jMEF.HierarchicalMixtureModel.getOptimalMixtureModel

(

double

t

)

Computes the "optimal" mixture model given a threshold on the KLD (estimated using a Monte-Carlo method).

Parameters:

t

maximum value of the KLD between the initial and the optimal models

Returns:

optimal mixture model

MixtureModel jMEF.HierarchicalMixtureModel.getResolution

(

int

resolution

)

Extracts a mixture model for a given resolution from the hierarchical mixture model.

The resolution 1 corresponds to a mixture model with only one model.

Parameters:

resolution

resolution of the mixture model

Returns:

mixture model for a given resolution

---

Member Data Documentation

ExponentialFamily jMEF.HierarchicalMixtureModel.EF

Exponential family member.

HierarchicalMixtureModel jMEF.HierarchicalMixtureModel.leftChild

Left child of the node.

MixtureModel jMEF.HierarchicalMixtureModel.node

Node containing a mixture model.

HierarchicalMixtureModel jMEF.HierarchicalMixtureModel.parent

Parent of the node.

int jMEF.HierarchicalMixtureModel.resolutionMax

Maximum resolution of the hierarchical mixture model.

HierarchicalMixtureModel jMEF.HierarchicalMixtureModel.rightChild

Right child of the node.

Clustering.CLUSTERING_TYPE jMEF.HierarchicalMixtureModel.type

Type of the Bregman divergence.

double jMEF.HierarchicalMixtureModel.weight

Weight of the tree.

---

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

• HierarchicalMixtureModel.java