knnClassification< T > Class Template Reference

#include <knnClassification.h>

Inheritance diagram for knnClassification< T >:

Collaboration diagram for knnClassification< T >:

Public Member Functions
	knnClassification (const int &num_inputs, const std::vector< int > &which_inputs, const std::vector< trainingExampleTemplate< T > > &trainingSet, const int k)
	~knnClassification ()
void	addNeighbour (const int &classNum, const std::vector< T > &features)
T	run (const std::vector< T > &inputVector)
void	train (const std::vector< trainingExampleTemplate< T > > &trainingSet)
void	reset ()
int	getNumInputs () const
std::vector< int >	getWhichInputs () const
int	getK () const
void	setK (int newK)
void	getJSONDescription (Json::Value &currentModel)
Public Member Functions inherited from baseModel< T >
virtual	~baseModel ()

Additional Inherited Members
Protected Member Functions inherited from baseModel< T >
template<typename TT >
Json::Value	vector2json (TT vec)

Detailed Description

template<typename T>
class knnClassification< T >

Class for implementing a knn classifier

Constructor & Destructor Documentation

knnClassification()

template<typename T >

knnClassification< T >::knnClassification	(	const int &	num_inputs,
		const std::vector< int > &	which_inputs,
		const std::vector< trainingExampleTemplate< T > > &	trainingSet,
		const int	k
	)

Constructor that takes training examples in

Parameters

number	of inputs expected in the training and input vectors
vector	of input numbers to be fed into the classifer.
vector	of training examples
how	many near neighbours to evaluate

~knnClassification()

template<typename T >

knnClassification< T >::~knnClassification

(

)

Member Function Documentation

addNeighbour()

template<typename T >

void knnClassification< T >::addNeighbour	(	const int &	classNum,
		const std::vector< T > &	features
	)

add another example to the existing training set

Parameters

class	number of example
feature	vector of example

getJSONDescription()

template<typename T >

void knnClassification< T >::getJSONDescription ( Json::Value &  currentModel )

virtual

Populate a JSON value with a description of the current model

Parameters

A	JSON value to be populated

Implements baseModel< T >.

getK()

template<typename T >

int knnClassification< T >::getK

(

)

const

Get the number of nearest neighbours used by the kNN algorithm.

getNumInputs()

template<typename T >

int knnClassification< T >::getNumInputs ( ) const

virtual

Find out how many inputs the model expects

Returns

Integer number of intpus

Implements baseModel< T >.

getWhichInputs()

template<typename T >

std::vector< int > knnClassification< T >::getWhichInputs ( ) const

virtual

Find out which inputs in a vector will be used

Returns

Vector of ints, specifying input indices.

Implements baseModel< T >.

reset()

template<typename T >

void knnClassification< T >::reset ( )

virtual

Reset the model to its empty state.

Implements baseModel< T >.

run()

template<typename T >

T knnClassification< T >::run ( const std::vector< T > &  inputVector )

virtual

Generate an output value from a single input vector.

Parameters

A	standard vector of type T to be evaluated.

Returns

A single value of type T: the nearest class as determined by k-nearest neighbor.

Implements baseModel< T >.

setK()

template<typename T >

void knnClassification< T >::setK

(

int

newK

)

Change the number of nearest neighbours used by the kNN algorithm.

Parameters

new	value for k

train()

template<typename T >

void knnClassification< T >::train ( const std::vector< trainingExampleTemplate< T > > &  trainingSet )

virtual

Fill the model with a vector of examples.

Parameters

The	training set is a vector of training examples that contain both a vector of input values and a value specifying desired output class.

Implements baseModel< T >.

---

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

• src/knnClassification.h
• src/knnClassification.cpp