CKNN Class Reference

---

Detailed Description

Class KNN, an implementation of the standard k-nearest neigbor classifier.

An example is classified to belong to the class of which the majority of the k closest examples belong to.

To avoid ties, k should be an odd number. To define how close examples are k-NN requires a CDistance object to work with (e.g., CEuclideanDistance ).

Note that k-NN has zero training time but classification times increase dramatically with the number of examples. Also note that k-NN is capable of multi-class-classification.

Definition at line 39 of file KNN.h.

Inheritance diagram for CKNN:

List of all members.

Public Member Functions
	CKNN ()
	CKNN (int32_t k, CDistance *d, CLabels *trainlab)
virtual	~CKNN ()
virtual EClassifierType	get_classifier_type ()
virtual bool	train (CFeatures *data=NULL)
virtual CLabels *	classify ()
virtual CLabels *	classify (CFeatures *data)
virtual float64_t	classify_example (int32_t vec_idx)
	get output for example "vec_idx"
void	classify_for_multiple_k (int32_t **output, int32_t *num_vec, int32_t *k_out)
virtual bool	load (FILE *srcfile)
virtual bool	save (FILE *dstfile)
void	set_k (int32_t p_k)
int32_t	get_k ()
virtual const char *	get_name () const
Protected Attributes
int32_t	k
	the k parameter in KNN
int32_t	num_classes
	number of classes (i.e. number of values labels can take)
int32_t	min_label
	smallest label, i.e. -1
int32_t	num_train_labels
	number of train examples
int32_t *	train_labels
	the actual trainlabels

---

Constructor & Destructor Documentation

CKNN

(

)

default constructor

Definition at line 20 of file KNN.cpp.

CKNN	(	int32_t	k,
		CDistance *	d,
		CLabels *	trainlab
	)

constructor

Parameters:

	k	k
	d	distance
	trainlab	labels for training

Definition at line 25 of file KNN.cpp.

~CKNN

(

)

[virtual]

Definition at line 37 of file KNN.cpp.

---

Member Function Documentation

CLabels * classify

(

)

[virtual]

classify all examples

Returns:

resulting labels

histogram of classes and returned output

Implements CDistanceMachine.

Definition at line 78 of file KNN.cpp.

CLabels * classify

(

CFeatures *

data

)

[virtual]

classify objects

Parameters:

data

(test)data to be classified

Returns:

classified labels

Implements CDistanceMachine.

Definition at line 147 of file KNN.cpp.

virtual float64_t classify_example

(

int32_t

vec_idx

)

[virtual]

get output for example "vec_idx"

Reimplemented from CClassifier.

Definition at line 85 of file KNN.h.

void classify_for_multiple_k	(	int32_t **	output,
		int32_t *	num_vec,
		int32_t *	k_out
	)

classify all examples for 1...k

Parameters:

	output	resulting labels for all k
	k_out	number of columns (k)
	num_vec	number of outputs

histogram of classes and returned output

Definition at line 164 of file KNN.cpp.

virtual EClassifierType get_classifier_type

(

)

[virtual]

get classifier type

Returns:

classifier type KNN

Reimplemented from CClassifier.

Definition at line 58 of file KNN.h.

int32_t get_k

(

)

get k

Returns:

k

Definition at line 127 of file KNN.h.

virtual const char* get_name

(

void

)

const [virtual]

Returns:

object name

Reimplemented from CDistanceMachine.

Definition at line 133 of file KNN.h.

bool load

(

FILE *

srcfile

)

[virtual]

load from file

Parameters:

srcfile

file to load from

Returns:

if loading was successful

Reimplemented from CClassifier.

Definition at line 235 of file KNN.cpp.

bool save

(

FILE *

dstfile

)

[virtual]

save to file

Parameters:

dstfile

file to save to

Returns:

if saving was successful

Reimplemented from CClassifier.

Definition at line 242 of file KNN.cpp.

void set_k

(

int32_t

p_k

)

set k

Parameters:

p_k

new k

Definition at line 117 of file KNN.h.

bool train

(

CFeatures *

data = NULL

)

[virtual]

train k-NN classifier

Parameters:

data

training data (parameter can be avoided if distance or kernel-based classifiers are used and distance/kernels are initialized with train data)

Returns:

whether training was successful

Reimplemented from CClassifier.

Definition at line 42 of file KNN.cpp.

---

Member Data Documentation

int32_t k [protected]

the k parameter in KNN

Definition at line 137 of file KNN.h.

int32_t min_label [protected]

smallest label, i.e. -1

Definition at line 143 of file KNN.h.

int32_t num_classes [protected]

number of classes (i.e. number of values labels can take)

Definition at line 140 of file KNN.h.

int32_t num_train_labels [protected]

number of train examples

Definition at line 146 of file KNN.h.

int32_t* train_labels [protected]

the actual trainlabels

Definition at line 149 of file KNN.h.

---

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

• KNN.h
• KNN.cpp