RandomPartitioner Class Reference

Inheritance diagram for RandomPartitioner:

[legend]

List of all members.

---

Detailed Description

Partitions a set of weighted samples into two sets along the midpoint of a random dimension.

Author:

Lawrence Murray <lawrence@indii.org>

Version:

Rev

Date:

Date

Definition at line 17 of file RandomPartitioner.hpp.

Public Member Functions
virtual	~RandomPartitioner ()
	Destructor.
virtual bool	init (DiracMixturePdf *p, const std::vector< unsigned int > &is)
	Initialise the partitioner.
virtual Partitioner::Partition	assign (const aux::vector &x)
	Assign a sample to a partition.

---

Constructor & Destructor Documentation

~RandomPartitioner

(

)

[virtual]

Destructor.

Definition at line 5 of file RandomPartitioner.cpp.

---

Member Function Documentation

bool init	(	DiracMixturePdf *	p,
		const std::vector< unsigned int > &	is
	)			[virtual]

Initialise the partitioner.

Parameters:

	p	Weighted sample set.
	is	Indices of components of interest in the weighted sample set.

Returns:

True if the partitioner is successful in finding a partition point, false otherwise. The partitioner may be unsuccessful if, e.g., all points are identical or one point in a pair has negligible small or zero weight.

Initialises the partitioner after construction, optionally using the given weighted sample set as a basis for the partition (e.g. using its bounds or covariance).

Implements Partitioner.

Definition at line 9 of file RandomPartitioner.cpp.

indii::ml::aux::Partitioner::Partition assign

(

const aux::vector &

x

)

[inline, virtual]

Assign a sample to a partition.

Parameters:

x

The sample to assign.

Returns:

The partition to which the sample is assigned.

Implements Partitioner.

Definition at line 47 of file RandomPartitioner.hpp.

---