KNN Class Reference

k-nearest neighbors algorithm More...

#include <KNN.hpp>

Collaboration diagram for KNN:

Public Types
enum	Distance { HAMMING, EUCLIDEAN }

Public Member Functions
int	getK () const
void	setK (int k)
const vector< vector< double > > &	getData () const
int	getYColumn () const
	KNN (vector< vector< double >> data, int yColumn, int k=1, Distance distance=EUCLIDEAN)
	k-nearest neighbors algorithm, able to do regression and classification More...
double	euclidean (vector< double > a, vector< double > b)
	Calculates the Euclidean distance between two vectors. More...
double	hamming (vector< double > a, vector< double > b)
	Calculates the Hamming distance between two vectors. More...
double	regression (const vector< double > &testie)
	Perform regression using. More...
double	classify (const vector< double > &testie)
vector< double >	classify (const vector< vector< double >> &test, bool verbose=false)
vector< double >	regression (const vector< vector< double >> &test, bool verbose=false)
Distance	getDistance () const
void	setDistance (Distance distance)

Private Member Functions
void	getKNN (int *chosen_indices, const vector< double > &testie)
	Finds the k-nearest neighbors of a data element. More...

Private Attributes
vector< vector< double > >	data
int	yColumn
int	k
Distance	distance

Detailed Description

k-nearest neighbors algorithm

Definition at line 22 of file KNN.hpp.

Member Enumeration Documentation

Distance

enum KNN::Distance

Enumerator
HAMMING
EUCLIDEAN

Definition at line 24 of file KNN.hpp.

Constructor & Destructor Documentation

KNN()

KNN::KNN ( vector< vector< double >>  data, int  yColumn, int  k = 1, Distance  distance = EUCLIDEAN  )

inlineexplicit

k-nearest neighbors algorithm, able to do regression and classification

Parameters

data	a dataset, where each vector represents a data element
yColumn	which column of the dataset is the dependent variable
k	number of nearest neighbors

Definition at line 98 of file KNN.hpp.

Member Function Documentation

classify() [1/2]

double KNN::classify ( const vector< double > &  testie )

inline

Definition at line 168 of file KNN.hpp.

classify() [2/2]

vector<double> KNN::classify ( const vector< vector< double >> &  test, bool  verbose = false  )

inline

Definition at line 203 of file KNN.hpp.

euclidean()

double KNN::euclidean ( vector< double >  a, vector< double >  b  )

inline

Calculates the Euclidean distance between two vectors.

Parameters

a	first vector
b	second vector

Returns

Euclidean distance between a and b

Definition at line 120 of file KNN.hpp.

getData()

const vector<vector<double> >& KNN::getData ( ) const

inline

Definition at line 86 of file KNN.hpp.

getDistance()

Distance KNN::getDistance ( ) const

inline

Definition at line 236 of file KNN.hpp.

getK()

int KNN::getK ( ) const

inline

Definition at line 78 of file KNN.hpp.

getKNN()

void KNN::getKNN ( int *  chosen_indices, const vector< double > &  testie  )

inlineprivate

Finds the k-nearest neighbors of a data element.

Parameters

chosen_indices	integer array that will keep the indices of the k-nearest neighbors
testie	a vector of real values representing a data element

Definition at line 34 of file KNN.hpp.

getYColumn()

int KNN::getYColumn ( ) const

inline

Definition at line 90 of file KNN.hpp.

hamming()

double KNN::hamming ( vector< double >  a, vector< double >  b  )

inline

Calculates the Hamming distance between two vectors.

Parameters

a	first vector
b	second vector

Returns

Hamming distance between a and b

Definition at line 136 of file KNN.hpp.

regression() [1/2]

double KNN::regression ( const vector< double > &  testie )

inline

Perform regression using.

Parameters

testie

Returns

Definition at line 151 of file KNN.hpp.

regression() [2/2]

vector<double> KNN::regression ( const vector< vector< double >> &  test, bool  verbose = false  )

inline

Definition at line 223 of file KNN.hpp.

setDistance()

void KNN::setDistance ( Distance  distance )

inline

Definition at line 240 of file KNN.hpp.

setK()

void KNN::setK ( int  k )

inline

Definition at line 82 of file KNN.hpp.

Field Documentation

data

vector<vector<double> > KNN::data

private

Definition at line 26 of file KNN.hpp.

distance

Distance KNN::distance

private

Definition at line 29 of file KNN.hpp.

k

int KNN::k

private

Definition at line 27 of file KNN.hpp.

yColumn

int KNN::yColumn

private

Definition at line 27 of file KNN.hpp.

---

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

• include/KNN.hpp