lti::manualCrispDecisionTree Class Reference

This class serves the manual construction of a crisp decision tree with crispNodes that have univariate decision functions. More...

#include <ltiManualCrispDecisionTree.h>

Inheritance diagram for lti::manualCrispDecisionTree:

[legend]

Collaboration diagram for lti::manualCrispDecisionTree:

[legend]

List of all members.

Classes
class	parameters
	the parameters for the class manualCrispDecisionTree More...
Public Member Functions
	manualCrispDecisionTree ()
	manualCrispDecisionTree (const manualCrispDecisionTree &other)
virtual	~manualCrispDecisionTree ()
virtual const char *	getTypeName () const
manualCrispDecisionTree &	copy (const manualCrispDecisionTree &other)
manualCrispDecisionTree &	operator= (const manualCrispDecisionTree &other)
virtual classifier *	clone () const
const parameters &	getParameters () const
virtual bool	train (const dmatrix &input, const ivector &ids)

---

Detailed Description

This class serves the manual construction of a crisp decision tree with crispNodes that have univariate decision functions.

Instead of training data, the train method is given the nodes' conditions and ids in pre-order. Only one element of each row vector can be unequal to zero. This dimension and the value are taken for the dimension and threshold of the condition. Thus, only univariate decisions can be used. The following example shows the data and id values of the train method and the corresponding tree:

   *        +-     -+         +-  -+
   *        | 0 0 2 |         | -1 |
   *        | 1 0 0 |         | -1 |
   *        | 0 2 0 |         | -1 |
   *        | 0 1 0 |         | -1 |
   *        | 0 0 0 |         |  2 |
   * data = | 0 0 1 |   ids = | -1 |
   *        | 0 0 0 |         |  5 |
   *        | 0 0 0 |         |  0 |
   *        | 0 0 0 |         |  1 |
   *        | 0 0 0 |         |  3 |
   *        | 0 0 0 |         |  4 |
   *        +-     -+         +-  -+
   * 

The outputTemplate object of the tree will be initialized with the ids given in the ids vector. If other or multiple ids are desired, the outputTemplate object must be set manually.

---

Constructor & Destructor Documentation

lti::manualCrispDecisionTree::manualCrispDecisionTree

(

)

default constructor

lti::manualCrispDecisionTree::manualCrispDecisionTree

(

const manualCrispDecisionTree &

other

)

copy constructor

Parameters:

other

the object to be copied

virtual lti::manualCrispDecisionTree::~manualCrispDecisionTree

(

)

[virtual]

destructor

---

Member Function Documentation

virtual classifier* lti::manualCrispDecisionTree::clone

(

)

const [virtual]

returns a pointer to a clone of this classifier.

Implements lti::classifier.

manualCrispDecisionTree& lti::manualCrispDecisionTree::copy

(

const manualCrispDecisionTree &

other

)

copy data of "other" classifier.

Parameters:

other

the classifier to be copied

Returns:

a reference to this classifier object

Reimplemented from lti::crispDecisionTree.

const parameters& lti::manualCrispDecisionTree::getParameters

(

)

const

returns used parameters

Reimplemented from lti::crispDecisionTree.

virtual const char* lti::manualCrispDecisionTree::getTypeName

(

)

const [virtual]

returns the name of this type ("manualCrispDecisionTree")

Reimplemented from lti::crispDecisionTree.

manualCrispDecisionTree& lti::manualCrispDecisionTree::operator=

(

const manualCrispDecisionTree &

other

)

alias for copy member

Parameters:

other

the classifier to be copied

Returns:

a reference to this classifier object

Reimplemented from lti::crispDecisionTree.

virtual bool lti::manualCrispDecisionTree::train	(	const dmatrix &	input,
		const ivector &	ids
	)			[virtual]

See description of this class.

Parameters:

	input	the conditions of the nodes
	ids	the ids of the nodes.

Returns:

true if successful, false otherwise. (if false you can check the error message with getStatusString())

Implements lti::supervisedInstanceClassifier.

---

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

• ltiManualCrispDecisionTree.h