shogun::CMyRandomForest Class Reference

This class implements the Random Forests algorithm. In Random Forests algorithm, we train a number of randomized CART trees (see class CRandomCARTree) using the supplied training data. The number of trees to be trained is a parameter (called number of bags) controlled by the user. Test feature vectors are classified/regressed by combining the outputs of all these trained candidate trees using a combination rule (see class CCombinationRule). The feature for calculating out-of-box error is also provided to help determine the appropriate number of bags. The evaluatin criteria for calculating this out-of-box error is specified by the user (see class CEvaluation). More...

#include <MyRandomForest.h>

Inheritance diagram for shogun::CMyRandomForest:

Collaboration diagram for shogun::CMyRandomForest:

Public Member Functions
	CMyRandomForest ()
virtual	~CMyRandomForest ()
virtual const char *	get_name () const
void	set_weights (SGVector< float64_t > weights)
SGVector< float64_t >	get_weights () const
void	set_feature_types (SGVector< bool > ft)
SGVector< bool >	get_feature_types () const
virtual EProblemType	get_machine_problem_type () const
void	set_machine_problem_type (EProblemType mode)
void	set_num_random_features (int32_t rand_featsize)
int32_t	get_num_random_features () const
std::vector< double >	feature_importances ()
void	set_probabilities (CLabels *labels, CFeatures *data=NULL)

Protected Member Functions
virtual bool	train_machine (CFeatures *data=NULL)
virtual void	set_machine_parameters (CMachine *m, SGVector< index_t > idx)

Private Attributes
SGVector< float64_t >	m_weights
SGMatrix< float64_t >	m_sorted_transposed_feats
SGMatrix< index_t >	m_sorted_indices

Detailed Description

This class implements the Random Forests algorithm. In Random Forests algorithm, we train a number of randomized CART trees (see class CRandomCARTree) using the supplied training data. The number of trees to be trained is a parameter (called number of bags) controlled by the user. Test feature vectors are classified/regressed by combining the outputs of all these trained candidate trees using a combination rule (see class CCombinationRule). The feature for calculating out-of-box error is also provided to help determine the appropriate number of bags. The evaluatin criteria for calculating this out-of-box error is specified by the user (see class CEvaluation).

Definition at line 48 of file MyRandomForest.h.

Constructor & Destructor Documentation

CMyRandomForest()

CMyRandomForest::CMyRandomForest

(

)

constructor

Definition at line 36 of file MyRandomForest.cc.

~CMyRandomForest()

CMyRandomForest::~CMyRandomForest ( )

virtual

destructor

Definition at line 45 of file MyRandomForest.cc.

Member Function Documentation

feature_importances()

std::vector< double > CMyRandomForest::feature_importances

(

)

WGL: return Gini importance scores for features

Definition at line 139 of file MyRandomForest.cc.

get_feature_types()

SGVector< bool > CMyRandomForest::get_feature_types

(

)

const

get feature types of various features

Returns

bool vector - true for nominal feature false for continuous feature type

Definition at line 65 of file MyRandomForest.cc.

get_machine_problem_type()

EProblemType CMyRandomForest::get_machine_problem_type ( ) const

virtual

get problem type - multiclass classification or regression

Returns

PT_MULTICLASS or PT_REGRESSION

Definition at line 71 of file MyRandomForest.cc.

get_name()

virtual const char* shogun::CMyRandomForest::get_name ( ) const

inlinevirtual

get name

Returns

RandomForest

Definition at line 61 of file MyRandomForest.h.

get_num_random_features()

int32_t CMyRandomForest::get_num_random_features

(

)

const

get number of random features to be chosen during node splits

Returns

number of randomly chosen features during each node split

Definition at line 91 of file MyRandomForest.cc.

get_weights()

SGVector< float64_t > CMyRandomForest::get_weights

(

)

const

get weights

Returns

weights of training feature vectors

Definition at line 54 of file MyRandomForest.cc.

set_feature_types()

void CMyRandomForest::set_feature_types

(

SGVector< bool >

ft

)

set feature types of various features

Parameters

ft	bool vector true for nominal feature false for continuous feature type

Definition at line 59 of file MyRandomForest.cc.

set_machine_parameters()

void CMyRandomForest::set_machine_parameters ( CMachine *  m, SGVector< index_t >  idx  )

protectedvirtual

sets parameters of CARTree - sets machine labels and weights here

Parameters

m	machine
idx	indices of training vectors chosen in current bag

Definition at line 97 of file MyRandomForest.cc.

set_machine_problem_type()

void CMyRandomForest::set_machine_problem_type

(

EProblemType

mode

)

set problem type - multiclass classification or regression

Parameters

mode	EProblemType PT_MULTICLASS or PT_REGRESSION

Definition at line 77 of file MyRandomForest.cc.

set_num_random_features()

void CMyRandomForest::set_num_random_features

(

int32_t

rand_featsize

)

set number of random features to be chosen during node splits

Parameters

rand_featsize

number of randomly chosen features during each node split

Definition at line 83 of file MyRandomForest.cc.

set_probabilities()

void CMyRandomForest::set_probabilities	(	CLabels *	labels,
		CFeatures *	data = NULL
	)

WGL: sets the probabilities on each label according to m_certainty

Definition at line 160 of file MyRandomForest.cc.

set_weights()

void CMyRandomForest::set_weights

(

SGVector< float64_t >

weights

)

set weights

Parameters

weights

of training feature vectors

Definition at line 49 of file MyRandomForest.cc.

train_machine()

bool CMyRandomForest::train_machine ( CFeatures *  data = NULL )

protectedvirtual

Definition at line 123 of file MyRandomForest.cc.

Member Data Documentation

m_sorted_indices

SGMatrix<index_t> shogun::CMyRandomForest::m_sorted_indices

private

Indices of pre-sorted features

Definition at line 137 of file MyRandomForest.h.

m_sorted_transposed_feats

SGMatrix<float64_t> shogun::CMyRandomForest::m_sorted_transposed_feats

private

Pre-sorted features

Definition at line 134 of file MyRandomForest.h.

m_weights

SGVector<float64_t> shogun::CMyRandomForest::m_weights

private

weights

Definition at line 131 of file MyRandomForest.h.

---

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

• MyRandomForest.h
• MyRandomForest.cc