lti::crispDecisionTree::crispNode Class Reference

A crisp decision tree node. More...

#include <ltiCrispDecisionTree.h>

Inheritance diagram for lti::crispDecisionTree::crispNode:

[legend]

Collaboration diagram for lti::crispDecisionTree::crispNode:

[legend]

List of all members.

Public Member Functions
	crispNode ()
	crispNode (const int &theId)
	crispNode (const int &dim, const double &thresh, const int &theId=-1)
	crispNode (const dvector &w, const double &thresh, const int &theId=-1)
	crispNode (const crispNode &other)
	~crispNode ()
const char *	getTypeName () const
crispNode &	copy (const crispNode &other)
crispNode &	operator= (const crispNode &other)
virtual decisionTree::dtNode *	clone () const
void	setDecisionFunction (const crispDecisionFunction *cdf)
const crispDecisionFunction *	getDecisionFunction (const crispDecisionFunction *cdf)
const crispNode *	propagate (const dvector &data) const
virtual bool	write (ioHandler &handler, const bool complete=true) const
virtual bool	read (ioHandler &handler, const bool complete=true)
Protected Attributes
crispDecisionFunction *	decisionF
Static Protected Attributes
static const objectFactory < crispDecisionFunction >	crispDecisionFunctionFactory
static const crispDecisionFunction *	cdfArray []

---

Detailed Description

A crisp decision tree node.

Its main difference to the dtNode is that it contains a crispDecisionFunction. The propagate method uses this function to determine which child node to return: left node if the condition is true, right node if it is false.

---

Constructor & Destructor Documentation

lti::crispDecisionTree::crispNode::crispNode

(

)

default constructor

lti::crispDecisionTree::crispNode::crispNode

(

const int &

theId

)

Constructor.

Sets the nodes id.

Parameters:

theId

the nodes id

lti::crispDecisionTree::crispNode::crispNode	(	const int &	dim,
		const double &	thresh,
		const int &	theId = -1
	)

Conveniance Constructor.

New node with univariateCrispDecisionFunction with the given parameters.

Parameters:

	dim	the dimension to be tested.
	thresh	the threshold in that dimension
	theId	the nodes id

lti::crispDecisionTree::crispNode::crispNode	(	const dvector &	w,
		const double &	thresh,
		const int &	theId = -1
	)

Conveniance Constructor.

New node with multivariateCrispDecisionFunction with the given parameters.

Parameters:

	w	the weighting vector for the data.
	thresh	the threshold.
	theId	the nodes id

lti::crispDecisionTree::crispNode::crispNode

(

const crispNode &

other

)

copy constructor

Parameters:

other

the crispNode object to be copied

lti::crispDecisionTree::crispNode::~crispNode

(

)

destructor

---

Member Function Documentation

virtual decisionTree::dtNode* lti::crispDecisionTree::crispNode::clone

(

)

const [virtual]

returns a pointer to a clone of the crispNode

Reimplemented from lti::decisionTree::dtNode.

crispNode& lti::crispDecisionTree::crispNode::copy

(

const crispNode &

other

)

copy the contents of a crispNode object

Parameters:

other

the crispNode object to be copied

Returns:

a reference to this crispNode object

Reimplemented from lti::decisionTree::dtNode.

const crispDecisionFunction* lti::crispDecisionTree::crispNode::getDecisionFunction

(

const crispDecisionFunction *

cdf

)

Gets the crispDecisionFunction this node uses.

Returns:

const pointer to the crispDecisionFunction

const char* lti::crispDecisionTree::crispNode::getTypeName

(

)

const [virtual]

returns name of this type

Reimplemented from lti::decisionTree::dtNode.

crispNode& lti::crispDecisionTree::crispNode::operator=

(

const crispNode &

other

)

copy the contents of a crispNode object

Parameters:

other

the crispNode object to be copied

Returns:

a reference to this crispNode object

Reimplemented from lti::decisionTree::dtNode.

const crispNode* lti::crispDecisionTree::crispNode::propagate

(

const dvector &

data

)

const

Propagates the given value through the node.

If the crispDecisionFunction is true for this value, the left child node is returned, else the right.

Parameters:

data

value to be propagated

Returns:

left or right child node

virtual bool lti::crispDecisionTree::crispNode::read	(	ioHandler &	handler,
		const bool	complete = true
	)			[virtual]

read the crispNode from the given ioHandler

Parameters:

	handler	the ioHandler to be used
	complete	if true (the default) the enclosing begin/end will be also written, otherwise only the data block will be written.

Returns:

true if write was successful

Reimplemented from lti::decisionTree::dtNode.

void lti::crispDecisionTree::crispNode::setDecisionFunction

(

const crispDecisionFunction *

cdf

)

Sets the crispDecisionFunction for this node.

Parameters:

cdf

the new crispDecisionFunction

virtual bool lti::crispDecisionTree::crispNode::write	(	ioHandler &	handler,
		const bool	complete = true
	)			const [virtual]

write the crispNode in the given ioHandler

Parameters:

	handler	the ioHandler to be used
	complete	if true (the default) the enclosing begin/end will be also written, otherwise only the data block will be written.

Returns:

true if write was successful

Reimplemented from lti::decisionTree::dtNode.

---

Member Data Documentation

const crispDecisionFunction* lti::crispDecisionTree::crispNode::cdfArray[] [static, protected]

The array of all crispDecisionFunctions.

const objectFactory<crispDecisionFunction> lti::crispDecisionTree::crispNode::crispDecisionFunctionFactory [static, protected]

The objectFactory for all crispDecisionFunctions.

crispDecisionFunction* lti::crispDecisionTree::crispNode::decisionF [protected]

The crispDecisionFunction used for this node.

---

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

• ltiCrispDecisionTree.h