gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC > Class Template Reference

A DBRowGenerator class that returns incomplete rows as EM would do. More...

#include <agrum/tools/database/DBRowGeneratorEM.h>

Inheritance diagram for gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC >:

Collaboration diagram for gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC >:

Public Member Functions
Constructors / Destructors
	DBRowGeneratorEM (const std::vector< DBTranslatedValueType, ALLOC< DBTranslatedValueType > > column_types, const BayesNet< GUM_SCALAR > &bn, const Bijection< NodeId, std::size_t, ALLOC< std::size_t > > &nodeId2columns=Bijection< NodeId, std::size_t, ALLOC< std::size_t > >(), const allocator_type &alloc=allocator_type())
	default constructor More...
	DBRowGeneratorEM (const DBRowGeneratorEM< GUM_SCALAR, ALLOC > &from)
	copy constructor More...
	DBRowGeneratorEM (const DBRowGeneratorEM< GUM_SCALAR, ALLOC > &from, const allocator_type &alloc)
	copy constructor with a given allocator More...
	DBRowGeneratorEM (DBRowGeneratorEM< GUM_SCALAR, ALLOC > &&from)
	move constructor More...
	DBRowGeneratorEM (DBRowGeneratorEM< GUM_SCALAR, ALLOC > &&from, const allocator_type &alloc)
	move constructor with a given allocator More...
virtual DBRowGeneratorEM< GUM_SCALAR, ALLOC > *	clone () const override final
	virtual copy constructor More...
virtual DBRowGeneratorEM< GUM_SCALAR, ALLOC > *	clone (const allocator_type &alloc) const override final
	virtual copy constructor with a given allocator More...
	~DBRowGeneratorEM ()
	destructor More...
Operators
DBRowGeneratorEM< GUM_SCALAR, ALLOC > &	operator= (const DBRowGeneratorEM< GUM_SCALAR, ALLOC > &from)
	copy operator More...
DBRowGeneratorEM< GUM_SCALAR, ALLOC > &	operator= (DBRowGeneratorEM< GUM_SCALAR, ALLOC > &&from)
	move operator More...
Accessors / Modifiers
virtual const DBRow< DBTranslatedValue, ALLOC > &	generate () override final
	generates one ouput DBRow for each DBRow passed to method setInputRow More...
virtual void	setBayesNet (const BayesNet< GUM_SCALAR > &new_bn) override final
	assign a new Bayes net to the generator More...
allocator_type	getAllocator () const
	returns the allocator used More...
Accessors / Modifiers
const BayesNet< GUM_SCALAR > &	getBayesNet () const
	returns the Bayes net used by the generator More...
Accessors / Modifiers
bool	hasRows ()
	returns true if there are still rows that can be output by the DBRowGenerator More...
bool	setInputRow (const DBRow< DBTranslatedValue, ALLOC > &row)
	sets the input row from which the generator will create its output rows More...
void	decreaseRemainingRows ()
	decrease the number of remaining output rows More...
virtual void	reset ()
	resets the generator. There are therefore no more ouput row to generate More...
virtual void	setColumnsOfInterest (const std::vector< std::size_t, ALLOC< std::size_t > > &cols_of_interest)
	sets the columns of interest: the output DBRow needs only contain correct values fot these columns More...
virtual void	setColumnsOfInterest (std::vector< std::size_t, ALLOC< std::size_t > > &&cols_of_interest)
	sets the columns of interest: the output DBRow needs only contain correct values fot these columns More...
const std::vector< std::size_t, ALLOC< std::size_t > > &	columnsOfInterest () const
	returns the current set of columns of interest More...
DBRowGeneratorGoal	goal () const
	returns the goal of the DBRowGenerator More...

Public Types
using	allocator_type = ALLOC< DBTranslatedValue >
	type for the allocators passed in arguments of methods More...

Protected Attributes
const BayesNet< GUM_SCALAR > *	bn_
	the Bayesian network used to fill the unobserved values More...
Bijection< NodeId, std::size_t, ALLOC< std::size_t > >	nodeId2columns_
	the mapping betwen the BN's node ids and the database's columns More...
std::size_t	nb_remaining_output_rows_ {std::size_t(0)}
	the number of output rows still to retrieve through the generate method More...
std::vector< DBTranslatedValueType, ALLOC< DBTranslatedValueType > >	column_types_
	the types of the columns in the DatabaseTable More...
std::vector< std::size_t, ALLOC< std::size_t > >	columns_of_interest_
	the set of columns of interest More...
DBRowGeneratorGoal	goal_ {DBRowGeneratorGoal::OTHER_THINGS_THAN_REMOVE_MISSING_VALUES}
	the goal of the DBRowGenerator (just remove missing values or not) More...

Protected Member Functions
virtual std::size_t	computeRows_ (const DBRow< DBTranslatedValue, ALLOC > &row) override final
	computes the rows it will provide as output More...

Detailed Description

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>
class gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC >

A DBRowGenerator class that returns incomplete rows as EM would do.

This class is a DBRowGenerator that fills the unobserved values of the nodes of interest as the EM algorithm does, i.e., by returning all the possible completed rows with a weight corresponding to the probability of the completion. The standard usage of a DBRowGenerator is the following:

// create a DatabaseTable and fill it
gum::learning::DBTranslatorSet<> set;
for ( int i = 0; i < 10; ++i )
  set.insertTranslator(gum::learning::DBTranslator4LabelizedVariable<>(),i);
gum::learning::DatabaseTable<> database ( set );
// fill the database

// keep in a vector the types of the columns in the database
const std::vector<gum::learning::DBTranslatedValueType>
  column_types ( 10, gum::learning::DBTranslatedValueType::DISCRETE );

// create the generator
gum::learning::DBRowGeneratorEM<> generator ( col_types );

// parse the database and produce output rows
for ( auto dbrow : database ) {
  generator.setInputRow ( dbrow );
  while ( generator.hasRows() ) {
    const auto& output_dbrow = generator.generate ();
    // do something with the output dbrow
  }
}

Definition at line 79 of file DBRowGeneratorEM.h.

Member Typedef Documentation

allocator_type

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>

using gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC >::allocator_type = ALLOC< DBTranslatedValue >

type for the allocators passed in arguments of methods

Definition at line 82 of file DBRowGeneratorEM.h.

Constructor & Destructor Documentation

DBRowGeneratorEM() [1/5]

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>

gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC >::DBRowGeneratorEM	(	const std::vector< DBTranslatedValueType, ALLOC< DBTranslatedValueType > >	column_types,
		const BayesNet< GUM_SCALAR > &	bn,
		const Bijection< NodeId, std::size_t, ALLOC< std::size_t > > &	nodeId2columns = Bijection< NodeId, std::size_t, ALLOC< std::size_t > >(),
		const allocator_type &	alloc = allocator_type()
	)

default constructor

DBRowGeneratorEM() [2/5]

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>

gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC >::DBRowGeneratorEM

(

const DBRowGeneratorEM< GUM_SCALAR, ALLOC > &

from

)

copy constructor

DBRowGeneratorEM() [3/5]

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>

gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC >::DBRowGeneratorEM	(	const DBRowGeneratorEM< GUM_SCALAR, ALLOC > &	from,
		const allocator_type &	alloc
	)

copy constructor with a given allocator

DBRowGeneratorEM() [4/5]

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>

gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC >::DBRowGeneratorEM

(

DBRowGeneratorEM< GUM_SCALAR, ALLOC > &&

from

)

move constructor

DBRowGeneratorEM() [5/5]

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>

gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC >::DBRowGeneratorEM	(	DBRowGeneratorEM< GUM_SCALAR, ALLOC > &&	from,
		const allocator_type &	alloc
	)

move constructor with a given allocator

~DBRowGeneratorEM()

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>

gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC >::~DBRowGeneratorEM

(

)

destructor

Member Function Documentation

clone() [1/2]

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>

virtual DBRowGeneratorEM< GUM_SCALAR, ALLOC >* gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC >::clone ( ) const

finaloverridevirtual

virtual copy constructor

Implements gum::learning::DBRowGenerator< ALLOC >.

clone() [2/2]

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>

virtual DBRowGeneratorEM< GUM_SCALAR, ALLOC >* gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC >::clone ( const allocator_type &  alloc ) const

finaloverridevirtual

virtual copy constructor with a given allocator

Implements gum::learning::DBRowGenerator< ALLOC >.

columnsOfInterest()

template<template< typename > class ALLOC = std::allocator>

const std::vector< std::size_t, ALLOC< std::size_t > >& gum::learning::DBRowGenerator< ALLOC >::columnsOfInterest ( ) const

inherited

returns the current set of columns of interest

computeRows_()

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>

virtual std::size_t gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC >::computeRows_ ( const DBRow< DBTranslatedValue, ALLOC > &  row )

finaloverrideprotectedvirtual

computes the rows it will provide as output

Implements gum::learning::DBRowGenerator< ALLOC >.

decreaseRemainingRows()

template<template< typename > class ALLOC = std::allocator>

void gum::learning::DBRowGenerator< ALLOC >::decreaseRemainingRows ( )

inherited

decrease the number of remaining output rows

When method setInputRow is performed, the DBRowGenerator knows how many output rows it will be able to generate. Each time method decreaseRemainingRows is called, we decrement this number. When the number becomes equal to 0, then there remains no new output row to generate.

generate()

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>

virtual const DBRow< DBTranslatedValue, ALLOC >& gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC >::generate ( )

finaloverridevirtual

generates one ouput DBRow for each DBRow passed to method setInputRow

Implements gum::learning::DBRowGenerator< ALLOC >.

getAllocator()

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>

allocator_type gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC >::getAllocator

(

)

const

returns the allocator used

getBayesNet()

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>

const BayesNet< GUM_SCALAR >& gum::learning::DBRowGeneratorWithBN< GUM_SCALAR, ALLOC >::getBayesNet ( ) const

inherited

returns the Bayes net used by the generator

goal()

template<template< typename > class ALLOC = std::allocator>

DBRowGeneratorGoal gum::learning::DBRowGenerator< ALLOC >::goal ( ) const

inherited

returns the goal of the DBRowGenerator

hasRows()

template<template< typename > class ALLOC = std::allocator>

bool gum::learning::DBRowGenerator< ALLOC >::hasRows ( )

inherited

returns true if there are still rows that can be output by the DBRowGenerator

operator=() [1/2]

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>

DBRowGeneratorEM< GUM_SCALAR, ALLOC >& gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC >::operator=

(

const DBRowGeneratorEM< GUM_SCALAR, ALLOC > &

from

)

copy operator

operator=() [2/2]

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>

DBRowGeneratorEM< GUM_SCALAR, ALLOC >& gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC >::operator=

(

DBRowGeneratorEM< GUM_SCALAR, ALLOC > &&

from

)

move operator

reset()

template<template< typename > class ALLOC = std::allocator>

virtual void gum::learning::DBRowGenerator< ALLOC >::reset ( )

virtualinherited

resets the generator. There are therefore no more ouput row to generate

setBayesNet()

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>

virtual void gum::learning::DBRowGeneratorEM< GUM_SCALAR, ALLOC >::setBayesNet ( const BayesNet< GUM_SCALAR > &  new_bn )

finaloverridevirtual

assign a new Bayes net to the generator

Reimplemented from gum::learning::DBRowGeneratorWithBN< GUM_SCALAR, ALLOC >.

setColumnsOfInterest() [1/2]

template<template< typename > class ALLOC = std::allocator>

virtual void gum::learning::DBRowGenerator< ALLOC >::setColumnsOfInterest ( const std::vector< std::size_t, ALLOC< std::size_t > > &  cols_of_interest )

virtualinherited

sets the columns of interest: the output DBRow needs only contain correct values fot these columns

This method is useful, e.g., for EM-like algorithms that need to know which unobserved variables/values need be filled. In this case, the DBRowGenerator still outputs DBRows with the same columns as the DatabaseTable, but only the columns of these DBRows corresponding to those passed in argument to Method setColumnsOfInterest are meaningful. For instance, if a DatabaseTable contains 10 columns and Method setColumnsOfInterest() is applied with vector<> { 0, 3, 4 }, then the DBRowGenerator will output DBRows with 10 columns, in which only columns 0, 3 and 4 are guaranteed to have correct values (columns are always indexed, starting from 0).

setColumnsOfInterest() [2/2]

template<template< typename > class ALLOC = std::allocator>

virtual void gum::learning::DBRowGenerator< ALLOC >::setColumnsOfInterest ( std::vector< std::size_t, ALLOC< std::size_t > > &&  cols_of_interest )

virtualinherited

sets the columns of interest: the output DBRow needs only contain correct values fot these columns

This method is useful, e.g., for EM-like algorithms that need to know which unobserved variables/values need be filled. In this case, the DBRowGenerator still outputs DBRows with the same columns as the DatabaseTable, but only the columns of these DBRows corresponding to those passed in argument to Method setColumnsOfInterest are meaningful. For instance, if a DatabaseTable contains 10 columns and Method setColumnsOfInterest() is applied with vector<> { 0, 3, 4 }, then the DBRowGenerator will output DBRows with 10 columns, in which only columns 0, 3 and 4 are guaranteed to have correct values (columns are always indexed, starting from 0).

setInputRow()

template<template< typename > class ALLOC = std::allocator>

bool gum::learning::DBRowGenerator< ALLOC >::setInputRow ( const DBRow< DBTranslatedValue, ALLOC > &  row )

inherited

sets the input row from which the generator will create its output rows

Returns

a Boolean indicating whether, from this input DBRow, the DBRowGenerator is capable of outputing at least one row or not

Member Data Documentation

bn_

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>

const BayesNet< GUM_SCALAR >* gum::learning::DBRowGeneratorWithBN< GUM_SCALAR, ALLOC >::bn_

protectedinherited

the Bayesian network used to fill the unobserved values

Definition at line 141 of file DBRowGeneratorWithBN.h.

column_types_

template<template< typename > class ALLOC = std::allocator>

std::vector< DBTranslatedValueType, ALLOC< DBTranslatedValueType > > gum::learning::DBRowGenerator< ALLOC >::column_types_

protectedinherited

the types of the columns in the DatabaseTable

This is useful to determine whether we need to use the .discr_val field or the .cont_val field in DBTranslatedValue instances.

Definition at line 361 of file DBRowGenerator.h.

columns_of_interest_

template<template< typename > class ALLOC = std::allocator>

std::vector< std::size_t, ALLOC< std::size_t > > gum::learning::DBRowGenerator< ALLOC >::columns_of_interest_

protectedinherited

the set of columns of interest

Definition at line 364 of file DBRowGenerator.h.

goal_

template<template< typename > class ALLOC = std::allocator>

DBRowGeneratorGoal gum::learning::DBRowGenerator< ALLOC >::goal_ {DBRowGeneratorGoal::OTHER_THINGS_THAN_REMOVE_MISSING_VALUES}

protectedinherited

the goal of the DBRowGenerator (just remove missing values or not)

Definition at line 367 of file DBRowGenerator.h.

nb_remaining_output_rows_

template<template< typename > class ALLOC = std::allocator>

std::size_t gum::learning::DBRowGenerator< ALLOC >::nb_remaining_output_rows_ {std::size_t(0)}

protectedinherited

the number of output rows still to retrieve through the generate method

Definition at line 356 of file DBRowGenerator.h.

nodeId2columns_

template<typename GUM_SCALAR = double, template< typename > class ALLOC = std::allocator>

Bijection< NodeId, std::size_t, ALLOC< std::size_t > > gum::learning::DBRowGeneratorWithBN< GUM_SCALAR, ALLOC >::nodeId2columns_

protectedinherited

the mapping betwen the BN's node ids and the database's columns

Definition at line 144 of file DBRowGeneratorWithBN.h.

---

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

• agrum/tools/database/DBRowGeneratorEM.h