DataSetUtils.h File Reference

Utility to operate on DataSet. More...

#include <vector>
#include <_stdint.h>
#include <fstream>
#include <iostream>
#include <map>
#include <cmath>
#include <stdlib.h>
#include "DataSet.h"
#include "Utility/bitset.h"

Go to the source code of this file.

Functions
template<class AggTypeDst , class AggTypeSrc >
void	Split (DataSetHandle< AggTypeDst > &p, DataSetHandle< AggTypeDst > &n, const DataSetHandle< AggTypeSrc > &src)
template<class AggTypeDst , class AggTypeSrc >
void	ExtractWorst (DataSetHandle< AggTypeDst > &dst, const DataSetHandle< AggTypeSrc > &src, int m)
template<class AggTypeDst , class AggTypeSrc >
void	ExtractBest (DataSetHandle< AggTypeDst > &dst, const DataSetHandle< AggTypeSrc > &src, int m)
template<class AggTypeDst , class AggTypeSrc >
void	RandomSampleWithDistribution (DataSetHandle< AggTypeDst > &dst, const DataSetHandle< AggTypeSrc > &src, int m)
template<class AggTypeDst , class AggTypeSrc >
void	ExtractReweighingEnergy (DataSetHandle< AggTypeDst > &dst, const DataSetHandle< AggTypeSrc > &src, float perc)
template<class AggTypeDst , class AggTypeSrc >
void	RandomSample (DataSetHandle< AggTypeDst > &dst, DataSetHandle< AggTypeDst > *validation, const DataSetHandle< AggTypeSrc > &src, int nA, int nB)
template<class AggTypeDst , class AggTypeSrc >
void	RandomUniqueSample (DataSetHandle< AggTypeDst > &dst, const DataSetHandle< AggTypeSrc > &src, int m)

Detailed Description

Utility to operate on DataSet.

Function Documentation

template<class AggTypeDst , class AggTypeSrc >

void ExtractBest	(	DataSetHandle< AggTypeDst > &	dst,
		const DataSetHandle< AggTypeSrc > &	src,
		int	m
	)

Extract m best elements

Note

requires a d field in pattern

template<class AggTypeDst , class AggTypeSrc >

void ExtractReweighingEnergy	(	DataSetHandle< AggTypeDst > &	dst,
		const DataSetHandle< AggTypeSrc > &	src,
		float	perc
	)

Extract perc of energy random field

Note

requires a d field in pattern

template<class AggTypeDst , class AggTypeSrc >

void ExtractWorst	(	DataSetHandle< AggTypeDst > &	dst,
		const DataSetHandle< AggTypeSrc > &	src,
		int	m
	)

Extract worst m elements

Note

requires a d field in pattern

template<class AggTypeDst , class AggTypeSrc >

void RandomSample	(	DataSetHandle< AggTypeDst > &	dst,
		DataSetHandle< AggTypeDst > *	validation,
		const DataSetHandle< AggTypeSrc > &	src,
		int	nA,
		int	nB
	)

Extract random samples from the DataSet, splitting in output and validation

template<class AggTypeDst , class AggTypeSrc >

void RandomSampleWithDistribution	(	DataSetHandle< AggTypeDst > &	dst,
		const DataSetHandle< AggTypeSrc > &	src,
		int	m
	)

Extract random m elements usingg distribution. uses .d field to extract randomply samples from distribution.

Note

requires d field

template<class AggTypeDst , class AggTypeSrc >

void RandomUniqueSample	(	DataSetHandle< AggTypeDst > &	dst,
		const DataSetHandle< AggTypeSrc > &	src,
		int	m
	)

Random Sample, with check uniqueness

template<class AggTypeDst , class AggTypeSrc >

void Split	(	DataSetHandle< AggTypeDst > &	p,
		DataSetHandle< AggTypeDst > &	n,
		const DataSetHandle< AggTypeSrc > &	src
	)

Split the Set in positive and negative samples