A pack of learning algorithms that can easily be used. More...

#include <BNLearner.h>

Inheritance diagram for gum::learning::BNLearner< GUM_SCALAR >:

Collaboration diagram for gum::learning::BNLearner< GUM_SCALAR >:

Public Attributes
Signaler3< Size, double, double >	onProgress
	Progression, error and time. More...

Signaler1< std::string >	onStop
	Criteria messageApproximationScheme. More...

Public Member Functions
BayesNet< GUM_SCALAR >	learnBN ()
	learn a Bayes Net from a file (must have read the db before) More...

BayesNet< GUM_SCALAR >	learnParameters (const DAG &dag, bool take_into_account_score=true)
	learns a BN (its parameters) when its structure is known More...

BayesNet< GUM_SCALAR >	learnParameters (bool take_into_account_score=true)

void	setAprioriWeight__ (double weight)
	sets the apriori weight More...

void	setMandatoryArcs (const ArcSet &set)
	assign a set of forbidden arcs More...

Constructors / Destructors
	BNLearner (const std::string &filename, const std::vector< std::string > &missing_symbols={"?"})
	default constructor More...

	BNLearner (const DatabaseTable<> &db)
	default constructor More...

	BNLearner (const std::string &filename, const gum::BayesNet< GUM_SCALAR > &src, const std::vector< std::string > &missing_symbols={"?"})
	Read the database file for the score / parameter estimation and var names. More...

	BNLearner (const BNLearner &)
	copy constructor More...

	BNLearner (BNLearner &&)
	move constructor More...

virtual	~BNLearner ()
	destructor More...

Operators
BNLearner &	operator= (const BNLearner &)
	copy operator More...

BNLearner &	operator= (BNLearner &&)
	move operator More...

Accessors / Modifiers
DAG	learnDAG ()
	learn a structure from a file (must have read the db before) More...

MixedGraph	learnMixedStructure ()
	learn a partial structure from a file (must have read the db before and must have selected miic or 3off2) More...

void	setInitialDAG (const DAG &)
	sets an initial DAG structure More...

const std::vector< std::string > &	names () const
	returns the names of the variables in the database More...

const std::vector< std::size_t > &	domainSizes () const
	returns the domain sizes of the variables in the database More...

Size	domainSize (NodeId var) const
	learn a structure from a file (must have read the db before) More...

Size	domainSize (const std::string &var) const
	learn a structure from a file (must have read the db before) More...

NodeId	idFromName (const std::string &var_name) const
	returns the node id corresponding to a variable name More...

const DatabaseTable &	database () const
	returns the database used by the BNLearner More...

void	setDatabaseWeight (const double new_weight)
	assign a weight to all the rows of the learning database so that the sum of their weights is equal to new_weight More...

void	setRecordWeight (const std::size_t i, const double weight)
	sets the weight of the ith record of the database More...

double	recordWeight (const std::size_t i) const
	returns the weight of the ith record More...

double	databaseWeight () const
	returns the weight of the whole database More...

const std::string &	nameFromId (NodeId id) const
	returns the variable name corresponding to a given node id More...

template<template< typename > class XALLOC>
void	useDatabaseRanges (const std::vector< std::pair< std::size_t, std::size_t >, XALLOC< std::pair< std::size_t, std::size_t > > > &new_ranges)
	use a new set of database rows' ranges to perform learning More...

void	clearDatabaseRanges ()
	reset the ranges to the one range corresponding to the whole database More...

const std::vector< std::pair< std::size_t, std::size_t > > &	databaseRanges () const
	returns the current database rows' ranges used for learning More...

std::pair< std::size_t, std::size_t >	useCrossValidationFold (const std::size_t learning_fold, const std::size_t k_fold)
	sets the ranges of rows to be used for cross-validation learning More...

std::pair< double, double >	chi2 (const NodeId id1, const NodeId id2, const std::vector< NodeId > &knowing={})
	Return the <statistic,pvalue> pair for chi2 test in the database. More...

std::pair< double, double >	chi2 (const std::string &name1, const std::string &name2, const std::vector< std::string > &knowing={})
	Return the <statistic,pvalue> pair for the BNLearner. More...

std::pair< double, double >	G2 (const NodeId id1, const NodeId id2, const std::vector< NodeId > &knowing={})
	Return the <statistic,pvalue> pair for for G2 test in the database. More...

std::pair< double, double >	G2 (const std::string &name1, const std::string &name2, const std::vector< std::string > &knowing={})
	Return the <statistic,pvalue> pair for for G2 test in the database. More...

double	logLikelihood (const std::vector< NodeId > &vars, const std::vector< NodeId > &knowing={})
	Return the loglikelihood of vars in the base, conditioned by knowing for the BNLearner. More...

double	logLikelihood (const std::vector< std::string > &vars, const std::vector< std::string > &knowing={})
	Return the loglikelihood of vars in the base, conditioned by knowing for the BNLearner. More...

std::vector< double >	rawPseudoCount (const std::vector< NodeId > &vars)
	Return the pseudoconts ofNodeIds vars in the base in a raw array. More...

std::vector< double >	rawPseudoCount (const std::vector< std::string > &vars)
	Return the pseudoconts of vars in the base in a raw array. More...

Size	nbCols () const

Size	nbRows () const

void	useEM (const double epsilon)
	use The EM algorithm to learn paramters More...

bool	hasMissingValues () const
	returns true if the learner's database has missing values More...

Score selection
void	useScoreAIC ()
	indicate that we wish to use an AIC score More...

void	useScoreBD ()
	indicate that we wish to use a BD score More...

void	useScoreBDeu ()
	indicate that we wish to use a BDeu score More...

void	useScoreBIC ()
	indicate that we wish to use a BIC score More...

void	useScoreK2 ()
	indicate that we wish to use a K2 score More...

void	useScoreLog2Likelihood ()
	indicate that we wish to use a Log2Likelihood score More...

A priori selection / parameterization
void	useNoApriori ()
	use no apriori More...

void	useAprioriBDeu (double weight=1)
	use the BDeu apriori More...

void	useAprioriSmoothing (double weight=1)
	use the apriori smoothing More...

void	useAprioriDirichlet (const std::string &filename, double weight=1)
	use the Dirichlet apriori More...

std::string	checkScoreAprioriCompatibility ()
	checks whether the current score and apriori are compatible More...

Learning algorithm selection
void	useGreedyHillClimbing ()
	indicate that we wish to use a greedy hill climbing algorithm More...

void	useLocalSearchWithTabuList (Size tabu_size=100, Size nb_decrease=2)
	indicate that we wish to use a local search with tabu list More...

void	useK2 (const Sequence< NodeId > &order)
	indicate that we wish to use K2 More...

void	useK2 (const std::vector< NodeId > &order)
	indicate that we wish to use K2 More...

void	use3off2 ()
	indicate that we wish to use 3off2 More...

void	useMIIC ()
	indicate that we wish to use MIIC More...

3off2/MIIC parameterization and specific results
void	useNML ()
	indicate that we wish to use the NML correction for 3off2 More...

void	useMDL ()
	indicate that we wish to use the MDL correction for 3off2 More...

void	useNoCorr ()
	indicate that we wish to use the NoCorr correction for 3off2 More...

const std::vector< Arc >	latentVariables () const
	get the list of arcs hiding latent variables More...

Accessors / Modifiers for adding constraints on learning
void	setMaxIndegree (Size max_indegree)
	sets the max indegree More...

void	setSliceOrder (const NodeProperty< NodeId > &slice_order)
	sets a partial order on the nodes More...

void	setSliceOrder (const std::vector< std::vector< std::string > > &slices)
	sets a partial order on the nodes More...

void	setForbiddenArcs (const ArcSet &set)
	assign a set of forbidden arcs More...

assign a new forbidden arc
void	addForbiddenArc (const Arc &arc)

void	addForbiddenArc (const NodeId tail, const NodeId head)

void	addForbiddenArc (const std::string &tail, const std::string &head)

void	addMandatoryArc (const Arc &arc)

void	addMandatoryArc (const NodeId tail, const NodeId head)

void	addMandatoryArc (const std::string &tail, const std::string &head)

remove a forbidden arc
void	eraseForbiddenArc (const Arc &arc)

void	eraseForbiddenArc (const NodeId tail, const NodeId head)

void	eraseForbiddenArc (const std::string &tail, const std::string &head)

void	eraseMandatoryArc (const Arc &arc)

void	eraseMandatoryArc (const NodeId tail, const NodeId head)

void	eraseMandatoryArc (const std::string &tail, const std::string &head)


void	setPossibleEdges (const EdgeSet &set)
	assign a set of forbidden edges More...

void	setPossibleSkeleton (const UndiGraph &skeleton)
	assign a set of forbidden edges More...

assign a new possible edge
Warning Once at least one possible edge is defined, all other edges are not possible anymore
void	addPossibleEdge (const Edge &edge)

void	addPossibleEdge (const NodeId tail, const NodeId head)

void	addPossibleEdge (const std::string &tail, const std::string &head)

remove a possible edge
void	erasePossibleEdge (const Edge &edge)

void	erasePossibleEdge (const NodeId tail, const NodeId head)

void	erasePossibleEdge (const std::string &tail, const std::string &head)

redistribute signals AND implemenation of interface
IApproximationSchemeConfiguration
INLINE void	setCurrentApproximationScheme (const ApproximationScheme *approximationScheme)
	{@ /// distribute signals More...

INLINE void	distributeProgress (const ApproximationScheme *approximationScheme, Size pourcent, double error, double time)
	{@ /// distribute signals More...

INLINE void	distributeStop (const ApproximationScheme *approximationScheme, std::string message)
	distribute signals More...


void	setEpsilon (double eps)
	Given that we approximate f(t), stopping criterion on \|f(t+1)-f(t)\| If the criterion was disabled it will be enabled. More...

double	epsilon () const
	Get the value of epsilon. More...

void	disableEpsilon ()
	Disable stopping criterion on epsilon. More...

void	enableEpsilon ()
	Enable stopping criterion on epsilon. More...

bool	isEnabledEpsilon () const


void	setMinEpsilonRate (double rate)
	Given that we approximate f(t), stopping criterion on d/dt(\|f(t+1)-f(t)\|) If the criterion was disabled it will be enabled. More...

double	minEpsilonRate () const
	Get the value of the minimal epsilon rate. More...

void	disableMinEpsilonRate ()
	Disable stopping criterion on epsilon rate. More...

void	enableMinEpsilonRate ()
	Enable stopping criterion on epsilon rate. More...

bool	isEnabledMinEpsilonRate () const


void	setMaxIter (Size max)
	stopping criterion on number of iterationsIf the criterion was disabled it will be enabled More...

Size	maxIter () const

void	disableMaxIter ()
	Disable stopping criterion on max iterations. More...

void	enableMaxIter ()
	Enable stopping criterion on max iterations. More...

bool	isEnabledMaxIter () const


void	setMaxTime (double timeout)
	stopping criterion on timeout If the criterion was disabled it will be enabled More...

double	maxTime () const
	returns the timeout (in seconds) More...

double	currentTime () const
	get the current running time in second (double) More...

void	disableMaxTime ()
	Disable stopping criterion on timeout. More...

void	enableMaxTime ()
	stopping criterion on timeout If the criterion was disabled it will be enabled More...

bool	isEnabledMaxTime () const


void	setPeriodSize (Size p)
	how many samples between 2 stopping isEnableds More...

Size	periodSize () const
	how many samples between 2 stopping isEnableds More...


void	setVerbosity (bool v)
	verbosity More...

bool	verbosity () const
	verbosity More...


ApproximationSchemeSTATE	stateApproximationScheme () const
	history More...

Size	nbrIterations () const

const std::vector< double > &	history () const

Getters and setters
std::string	messageApproximationScheme () const
	Returns the approximation scheme message. More...

Public Types
enum	ScoreType { ScoreType::AIC, ScoreType::BD, ScoreType::BDeu, ScoreType::BIC, ScoreType::K2, ScoreType::LOG2LIKELIHOOD }
	an enumeration enabling to select easily the score we wish to use More...

enum	ParamEstimatorType { ParamEstimatorType::ML }
	an enumeration to select the type of parameter estimation we shall apply More...

enum	AprioriType { AprioriType::NO_APRIORI, AprioriType::SMOOTHING, AprioriType::DIRICHLET_FROM_DATABASE, AprioriType::BDEU }
	an enumeration to select the apriori More...

enum	AlgoType { AlgoType::K2, AlgoType::GREEDY_HILL_CLIMBING, AlgoType::LOCAL_SEARCH_WITH_TABU_LIST, AlgoType::MIIC_THREE_OFF_TWO }
	an enumeration to select easily the learning algorithm to use More...

enum	ApproximationSchemeSTATE : char { ApproximationSchemeSTATE::Undefined, ApproximationSchemeSTATE::Continue, ApproximationSchemeSTATE::Epsilon, ApproximationSchemeSTATE::Rate, ApproximationSchemeSTATE::Limit, ApproximationSchemeSTATE::TimeLimit, ApproximationSchemeSTATE::Stopped }
	The different state of an approximation scheme. More...

Protected Attributes
ScoreType	score_type__ {ScoreType::BDeu}
	the score selected for learning More...

Score *	score__ {nullptr}
	the score used More...

ParamEstimatorType	param_estimator_type__ {ParamEstimatorType::ML}
	the type of the parameter estimator More...

double	EMepsilon__ {0.0}
	epsilon for EM. if espilon=0.0 : no EM More...

CorrectedMutualInformation *	mutual_info__ {nullptr}
	the selected correction for 3off2 and miic More...

AprioriType	apriori_type__ {AprioriType::NO_APRIORI}
	the a priori selected for the score and parameters More...

Apriori *	apriori__ {nullptr}
	the apriori used More...

AprioriNoApriori *	no_apriori__ {nullptr}

double	apriori_weight__ {1.0f}
	the weight of the apriori More...

StructuralConstraintSliceOrder	constraint_SliceOrder__
	the constraint for 2TBNs More...

StructuralConstraintIndegree	constraint_Indegree__
	the constraint for indegrees More...

StructuralConstraintTabuList	constraint_TabuList__
	the constraint for tabu lists More...

StructuralConstraintForbiddenArcs	constraint_ForbiddenArcs__
	the constraint on forbidden arcs More...

StructuralConstraintPossibleEdges	constraint_PossibleEdges__
	the constraint on possible Edges More...

StructuralConstraintMandatoryArcs	constraint_MandatoryArcs__
	the constraint on forbidden arcs More...

AlgoType	selected_algo__ {AlgoType::GREEDY_HILL_CLIMBING}
	the selected learning algorithm More...

K2	K2__
	the K2 algorithm More...

Miic	miic_3off2__
	the 3off2 algorithm More...

CorrectedMutualInformation ::KModeTypes	kmode_3off2__
	the penalty used in 3off2 More...

DAG2BNLearner	Dag2BN__
	the parametric EM More...

GreedyHillClimbing	greedy_hill_climbing__
	the greedy hill climbing algorithm More...

LocalSearchWithTabuList	local_search_with_tabu_list__
	the local search with tabu list algorithm More...

Database	score_database__
	the database to be used by the scores and parameter estimators More...

std::vector< std::pair< std::size_t, std::size_t > >	ranges__
	the set of rows' ranges within the database in which learning is done More...

Database *	apriori_database__ {nullptr}
	the database used by the Dirichlet a priori More...

std::string	apriori_dbname__
	the filename for the Dirichlet a priori, if any More...

DAG	initial_dag__
	an initial DAG given to learners More...

const ApproximationScheme *	current_algorithm__ {nullptr}

Protected Member Functions
void	createApriori__ ()
	create the apriori used for learning More...

void	createScore__ ()
	create the score used for learning More...

ParamEstimator *	createParamEstimator__ (DBRowGeneratorParser<> &parser, bool take_into_account_score=true)
	create the parameter estimator used for learning More...

DAG	learnDAG__ ()
	returns the DAG learnt More...

MixedGraph	prepare_miic_3off2__ ()
	prepares the initial graph for 3off2 or miic More...

const std::string &	getAprioriType__ () const
	returns the type (as a string) of a given apriori More...

void	createCorrectedMutualInformation__ ()
	create the Corrected Mutual Information instance for Miic/3off2 More...

Static Protected Member Functions
static DatabaseTable	readFile__ (const std::string &filename, const std::vector< std::string > &missing_symbols)
	reads a file and returns a databaseVectInRam More...

static void	checkFileName__ (const std::string &filename)
	checks whether the extension of a CSV filename is correct More...

Detailed Description

template<typename GUM_SCALAR>
class gum::learning::BNLearner< GUM_SCALAR >

A pack of learning algorithms that can easily be used.

The pack currently contains K2, GreedyHillClimbing and LocalSearchWithTabuList

Definition at line 59 of file BNLearner.h.

Member Enumeration Documentation

◆ AlgoType

enum gum::learning::genericBNLearner::AlgoType

stronginherited

an enumeration to select easily the learning algorithm to use

Enumerator
K2
GREEDY_HILL_CLIMBING
LOCAL_SEARCH_WITH_TABU_LIST
MIIC_THREE_OFF_TWO

Definition at line 135 of file genericBNLearner.h.

       {
         K2,
         GREEDY_HILL_CLIMBING,
         LOCAL_SEARCH_WITH_TABU_LIST,
         MIIC_THREE_OFF_TWO
       };

◆ ApproximationSchemeSTATE

enum gum::IApproximationSchemeConfiguration::ApproximationSchemeSTATE : char

stronginherited

The different state of an approximation scheme.

Enumerator
Undefined
Continue
Epsilon
Rate
Limit
TimeLimit
Stopped

Definition at line 64 of file IApproximationSchemeConfiguration.h.

                                         : char
     {
       Undefined,
       Continue,
       Epsilon,
       Rate,
       Limit,
       TimeLimit,
       Stopped
     };

◆ AprioriType

enum gum::learning::genericBNLearner::AprioriType

stronginherited

an enumeration to select the apriori

Enumerator
NO_APRIORI
SMOOTHING
DIRICHLET_FROM_DATABASE
BDEU

Definition at line 126 of file genericBNLearner.h.

       {
         NO_APRIORI,
         SMOOTHING,
         DIRICHLET_FROM_DATABASE,
         BDEU
       };

◆ ParamEstimatorType

enum gum::learning::genericBNLearner::ParamEstimatorType

stronginherited

an enumeration to select the type of parameter estimation we shall apply

Enumerator
ML

Definition at line 122 of file genericBNLearner.h.

123 { ML };

◆ ScoreType

enum gum::learning::genericBNLearner::ScoreType

stronginherited

an enumeration enabling to select easily the score we wish to use

Enumerator
AIC
BD
BDeu
BIC
K2
LOG2LIKELIHOOD

Definition at line 110 of file genericBNLearner.h.

       {
         AIC,
         BD,
         BDeu,
         BIC,
         K2,
         LOG2LIKELIHOOD
       };

Constructor & Destructor Documentation

◆ BNLearner() [1/5]

template<typename GUM_SCALAR >

gum::learning::BNLearner< GUM_SCALAR >::BNLearner	(	const std::string &	filename,
		const std::vector< std::string > &	missing_symbols = `{"?"}`
	)

default constructor

read the database file for the score / parameter estimation and var names

◆ BNLearner() [2/5]

template<typename GUM_SCALAR >

gum::learning::BNLearner< GUM_SCALAR >::BNLearner ( const DatabaseTable<> & db )

default constructor

read the database file for the score / parameter estimation and var names

◆ BNLearner() [3/5]

template<typename GUM_SCALAR >

gum::learning::BNLearner< GUM_SCALAR >::BNLearner	(	const std::string &	filename,
		const gum::BayesNet< GUM_SCALAR > &	src,
		const std::vector< std::string > &	missing_symbols = `{"?"}`
	)

Read the database file for the score / parameter estimation and var names.

If modalities = { 1 -> {True, False, Big} }, then the node of id 1 in the BN will have 3 modalities, the first one being True, the second one being False, and the third bein Big.

A parsing of the database will allow to determine which ones are really necessary and will keep them in the order specified by the user (NodeProperty modalities). If parse_database is set to false (the default), then the modalities specified by the user will be considered as being exactly those of the variables of the BN (as a consequence, if we find other values in the database, an exception will be raised during learning).

Parameters

filename	The file to learn from.
modalities	indicate for some nodes (not necessarily all the nodes of the BN) which modalities they should have and in which order these modalities should be stored into the nodes.
parse_database	if true, the modalities specified by the user will be considered as a superset of the modalities of the variables. Wrapper for BNLearner (filename,modalities,parse_database) using a bn to find those modalities and nodeids.

◆ BNLearner() [4/5]

template<typename GUM_SCALAR >

gum::learning::BNLearner< GUM_SCALAR >::BNLearner ( const BNLearner< GUM_SCALAR > & )

copy constructor

◆ BNLearner() [5/5]

template<typename GUM_SCALAR >

gum::learning::BNLearner< GUM_SCALAR >::BNLearner ( BNLearner< GUM_SCALAR > && )

move constructor

◆ ~BNLearner()

template<typename GUM_SCALAR >

virtual gum::learning::BNLearner< GUM_SCALAR >::~BNLearner ( )

virtual

destructor

Member Function Documentation

◆ addForbiddenArc() [1/3]

INLINE void gum::learning::genericBNLearner::addForbiddenArc ( const Arc & arc )

inherited

Definition at line 358 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                 {
       constraint_ForbiddenArcs__.addArc(arc);
     }

Here is the call graph for this function:

◆ addForbiddenArc() [2/3]

INLINE void gum::learning::genericBNLearner::addForbiddenArc	(	const NodeId	tail,
		const NodeId	head
	)

inherited

Definition at line 368 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                      {
       addForbiddenArc(Arc(tail, head));
     }

Here is the call graph for this function:

◆ addForbiddenArc() [3/3]

INLINE void gum::learning::genericBNLearner::addForbiddenArc	(	const std::string &	tail,
		const std::string &	head
	)

inherited

Definition at line 380 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                          {
       addForbiddenArc(Arc(idFromName(tail), idFromName(head)));
     }

Here is the call graph for this function:

◆ addMandatoryArc() [1/3]

INLINE void gum::learning::genericBNLearner::addMandatoryArc ( const Arc & arc )

inherited

Definition at line 397 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                 {
       constraint_MandatoryArcs__.addArc(arc);
     }

Here is the call graph for this function:

◆ addMandatoryArc() [2/3]

INLINE void gum::learning::genericBNLearner::addMandatoryArc	(	const NodeId	tail,
		const NodeId	head
	)

inherited

Definition at line 419 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                      {
       addMandatoryArc(Arc(tail, head));
     }

Here is the call graph for this function:

◆ addMandatoryArc() [3/3]

INLINE void gum::learning::genericBNLearner::addMandatoryArc	(	const std::string &	tail,
		const std::string &	head
	)

inherited

Definition at line 407 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                          {
       addMandatoryArc(Arc(idFromName(tail), idFromName(head)));
     }

Here is the call graph for this function:

◆ addPossibleEdge() [1/3]

INLINE void gum::learning::genericBNLearner::addPossibleEdge ( const Edge & edge )

inherited

Definition at line 319 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                   {
       constraint_PossibleEdges__.addEdge(edge);
     }

Here is the call graph for this function:

◆ addPossibleEdge() [2/3]

INLINE void gum::learning::genericBNLearner::addPossibleEdge	(	const NodeId	tail,
		const NodeId	head
	)

inherited

Definition at line 329 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                      {
       addPossibleEdge(Edge(tail, head));
     }

Here is the call graph for this function:

◆ addPossibleEdge() [3/3]

INLINE void gum::learning::genericBNLearner::addPossibleEdge	(	const std::string &	tail,
		const std::string &	head
	)

inherited

Definition at line 341 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                          {
       addPossibleEdge(Edge(idFromName(tail), idFromName(head)));
     }

Here is the call graph for this function:

◆ checkFileName__()

void gum::learning::genericBNLearner::checkFileName__ ( const std::string & filename )

staticprotectedinherited

checks whether the extension of a CSV filename is correct

Definition at line 414 of file genericBNLearner.cpp.

                                                                     {
       // get the extension of the file
       Size filename_size = Size(filename.size());
 
       if (filename_size < 4) {
         GUM_ERROR(FormatNotFound,
                   "genericBNLearner could not determine the "
                   "file type of the database");
       }
 
       std::string extension = filename.substr(filename.size() - 4);
       std::transform(extension.begin(),
                      extension.end(),
                      extension.begin(),
                      ::tolower);
 
       if (extension != ".csv") {
         GUM_ERROR(
            OperationNotAllowed,
            "genericBNLearner does not support yet this type of database file");
       }
     }

◆ checkScoreAprioriCompatibility()

std::string gum::learning::genericBNLearner::checkScoreAprioriCompatibility ( )

inherited

checks whether the current score and apriori are compatible

Returns: a non empty string if the apriori is somehow compatible with the score.

Definition at line 871 of file genericBNLearner.cpp.

                                                                {
       const std::string& apriori = getAprioriType__();
 
       switch (score_type__) {
         case ScoreType::AIC:
           return ScoreAIC<>::isAprioriCompatible(apriori, apriori_weight__);
 
         case ScoreType::BD:
           return ScoreBD<>::isAprioriCompatible(apriori, apriori_weight__);
 
         case ScoreType::BDeu:
           return ScoreBDeu<>::isAprioriCompatible(apriori, apriori_weight__);
 
         case ScoreType::BIC:
           return ScoreBIC<>::isAprioriCompatible(apriori, apriori_weight__);
 
         case ScoreType::K2:
           return ScoreK2<>::isAprioriCompatible(apriori, apriori_weight__);
 
         case ScoreType::LOG2LIKELIHOOD:
           return ScoreLog2Likelihood<>::isAprioriCompatible(apriori,
                                                             apriori_weight__);
 
         default:
           return "genericBNLearner does not support yet this score";
       }
     }

◆ chi2() [1/2]

std::pair< double, double > gum::learning::genericBNLearner::chi2	(	const NodeId	id1,
		const NodeId	id2,
		const std::vector< NodeId > &	knowing = `{}`
	)

inherited

Return the <statistic,pvalue> pair for chi2 test in the database.

Parameters

id1	first variable
id2	second variable
knowing	list of observed variables

Returns: a std::pair<double,double>

Definition at line 948 of file genericBNLearner.cpp.

                                                                   {
       createApriori__();
       gum::learning::IndepTestChi2<> chi2score(score_database__.parser(),
                                                *apriori__,
                                                databaseRanges());
 
       return chi2score.statistics(id1, id2, knowing);
     }

◆ chi2() [2/2]

std::pair< double, double > gum::learning::genericBNLearner::chi2	(	const std::string &	name1,
		const std::string &	name2,
		const std::vector< std::string > &	knowing = `{}`
	)

inherited

Return the <statistic,pvalue> pair for the BNLearner.

Parameters

id1	first variable
id2	second variable
knowing	list of observed variables

Returns: a std::pair<double,double>

Definition at line 960 of file genericBNLearner.cpp.

                                                                      {
       std::vector< NodeId > knowingIds;
       std::transform(
          knowing.begin(),
          knowing.end(),
          std::back_inserter(knowingIds),
          [this](const std::string& c) -> NodeId { return this->idFromName(c); });
       return chi2(idFromName(name1), idFromName(name2), knowingIds);
     }

◆ clearDatabaseRanges()

INLINE void gum::learning::genericBNLearner::clearDatabaseRanges ( )

inherited

reset the ranges to the one range corresponding to the whole database

Definition at line 554 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

554 { ranges__.clear(); }

gum::learning::genericBNLearner::ranges__

std::vector< std::pair< std::size_t, std::size_t > > ranges__

the set of rows' ranges within the database in which learning is done

Definition: genericBNLearner.h:824

Here is the call graph for this function:

◆ createApriori__()

void gum::learning::genericBNLearner::createApriori__ ( )

protectedinherited

create the apriori used for learning

Definition at line 465 of file genericBNLearner.cpp.

                                            {
       // first, save the old apriori, to be delete if everything is ok
       Apriori<>* old_apriori = apriori__;
 
       // create the new apriori
       switch (apriori_type__) {
         case AprioriType::NO_APRIORI:
           apriori__ = new AprioriNoApriori<>(score_database__.databaseTable(),
                                              score_database__.nodeId2Columns());
           break;
 
         case AprioriType::SMOOTHING:
           apriori__ = new AprioriSmoothing<>(score_database__.databaseTable(),
                                              score_database__.nodeId2Columns());
           break;
 
         case AprioriType::DIRICHLET_FROM_DATABASE:
           if (apriori_database__ != nullptr) {
             delete apriori_database__;
             apriori_database__ = nullptr;
           }
 
           apriori_database__ = new Database(apriori_dbname__,
                                             score_database__,
                                             score_database__.missingSymbols());
 
           apriori__ = new AprioriDirichletFromDatabase<>(
              score_database__.databaseTable(),
              apriori_database__->parser(),
              apriori_database__->nodeId2Columns());
           break;
 
         case AprioriType::BDEU:
           apriori__ = new AprioriBDeu<>(score_database__.databaseTable(),
                                         score_database__.nodeId2Columns());
           break;
 
         default:
           GUM_ERROR(OperationNotAllowed,
                     "The BNLearner does not support yet this apriori");
       }
 
       // do not forget to assign a weight to the apriori
       apriori__->setWeight(apriori_weight__);
 
       // remove the old apriori, if any
       if (old_apriori != nullptr) delete old_apriori;
     }

◆ createCorrectedMutualInformation__()

void gum::learning::genericBNLearner::createCorrectedMutualInformation__ ( )

protectedinherited

create the Corrected Mutual Information instance for Miic/3off2

Definition at line 666 of file genericBNLearner.cpp.

                                                               {
       if (mutual_info__ != nullptr) delete mutual_info__;
 
       mutual_info__
          = new CorrectedMutualInformation<>(score_database__.parser(),
                                             *no_apriori__,
                                             ranges__,
                                             score_database__.nodeId2Columns());
       switch (kmode_3off2__) {
         case CorrectedMutualInformation<>::KModeTypes::MDL:
           mutual_info__->useMDL();
           break;
 
         case CorrectedMutualInformation<>::KModeTypes::NML:
           mutual_info__->useNML();
           break;
 
         case CorrectedMutualInformation<>::KModeTypes::NoCorr:
           mutual_info__->useNoCorr();
           break;
 
         default:
           GUM_ERROR(NotImplementedYet,
                     "The BNLearner's corrected mutual information class does "
                        << "not support yet penalty mode " << int(kmode_3off2__));
       }
     }

◆ createParamEstimator__()

ParamEstimator * gum::learning::genericBNLearner::createParamEstimator__	(	DBRowGeneratorParser<> &	parser,
		bool	take_into_account_score = `true`
	)

protectedinherited

create the parameter estimator used for learning

Definition at line 572 of file genericBNLearner.cpp.

                                                                               {
       ParamEstimator<>* param_estimator = nullptr;
 
       // create the new estimator
       switch (param_estimator_type__) {
         case ParamEstimatorType::ML:
           if (take_into_account_score && (score__ != nullptr)) {
             param_estimator
                = new ParamEstimatorML<>(parser,
                                         *apriori__,
                                         score__->internalApriori(),
                                         ranges__,
                                         score_database__.nodeId2Columns());
           } else {
             param_estimator
                = new ParamEstimatorML<>(parser,
                                         *apriori__,
                                         *no_apriori__,
                                         ranges__,
                                         score_database__.nodeId2Columns());
           }
 
           break;
 
         default:
           GUM_ERROR(OperationNotAllowed,
                     "genericBNLearner does not support "
                        << "yet this parameter estimator");
       }
 
       // assign the set of ranges
       param_estimator->setRanges(ranges__);
 
       return param_estimator;
     }

◆ createScore__()

void gum::learning::genericBNLearner::createScore__ ( )

protectedinherited

create the score used for learning

Definition at line 514 of file genericBNLearner.cpp.

                                          {
       // first, save the old score, to be delete if everything is ok
       Score<>* old_score = score__;
 
       // create the new scoring function
       switch (score_type__) {
         case ScoreType::AIC:
           score__ = new ScoreAIC<>(score_database__.parser(),
                                    *apriori__,
                                    ranges__,
                                    score_database__.nodeId2Columns());
           break;
 
         case ScoreType::BD:
           score__ = new ScoreBD<>(score_database__.parser(),
                                   *apriori__,
                                   ranges__,
                                   score_database__.nodeId2Columns());
           break;
 
         case ScoreType::BDeu:
           score__ = new ScoreBDeu<>(score_database__.parser(),
                                     *apriori__,
                                     ranges__,
                                     score_database__.nodeId2Columns());
           break;
 
         case ScoreType::BIC:
           score__ = new ScoreBIC<>(score_database__.parser(),
                                    *apriori__,
                                    ranges__,
                                    score_database__.nodeId2Columns());
           break;
 
         case ScoreType::K2:
           score__ = new ScoreK2<>(score_database__.parser(),
                                   *apriori__,
                                   ranges__,
                                   score_database__.nodeId2Columns());
           break;
 
         case ScoreType::LOG2LIKELIHOOD:
           score__ = new ScoreLog2Likelihood<>(score_database__.parser(),
                                               *apriori__,
                                               ranges__,
                                               score_database__.nodeId2Columns());
           break;
 
         default:
           GUM_ERROR(OperationNotAllowed,
                     "genericBNLearner does not support yet this score");
       }
 
       // remove the old score, if any
       if (old_score != nullptr) delete old_score;
     }

◆ currentTime()

double gum::learning::genericBNLearner::currentTime ( ) const

inlinevirtualinherited

get the current running time in second (double)

Implements gum::IApproximationSchemeConfiguration.

Definition at line 1059 of file genericBNLearner.h.

                                  {
         if (current_algorithm__ != nullptr)
           return current_algorithm__->currentTime();
         else
           GUM_ERROR(FatalError, "No chosen algorithm for learning");
       };

◆ database()

INLINE const DatabaseTable & gum::learning::genericBNLearner::database ( ) const

inherited

returns the database used by the BNLearner

Definition at line 557 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                    {
       return score_database__.databaseTable();
     }

Here is the call graph for this function:

◆ databaseRanges()

INLINE const std::vector< std::pair< std::size_t, std::size_t > > & gum::learning::genericBNLearner::databaseRanges ( ) const

inherited

returns the current database rows' ranges used for learning

Returns: The method returns a vector of pairs [Xi,Yi) of indices of rows in the database. The learning is performed on these set of rows.

Warning: an empty set of ranges means the whole database.

Definition at line 549 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                         {
       return ranges__;
     }

Here is the call graph for this function:

◆ databaseWeight()

INLINE double gum::learning::genericBNLearner::databaseWeight ( ) const

inherited

returns the weight of the whole database

Definition at line 172 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                          {
       return score_database__.weight();
     }

Here is the call graph for this function:

◆ disableEpsilon()

void gum::learning::genericBNLearner::disableEpsilon ( )

inlinevirtualinherited

Disable stopping criterion on epsilon.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 924 of file genericBNLearner.h.

                             {
         K2__.approximationScheme().disableEpsilon();
         greedy_hill_climbing__.disableEpsilon();
         local_search_with_tabu_list__.disableEpsilon();
         Dag2BN__.disableEpsilon();
       };

◆ disableMaxIter()

void gum::learning::genericBNLearner::disableMaxIter ( )

inlinevirtualinherited

Disable stopping criterion on max iterations.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 1014 of file genericBNLearner.h.

                             {
         K2__.approximationScheme().disableMaxIter();
         greedy_hill_climbing__.disableMaxIter();
         local_search_with_tabu_list__.disableMaxIter();
         Dag2BN__.disableMaxIter();
       };

◆ disableMaxTime()

void gum::learning::genericBNLearner::disableMaxTime ( )

inlinevirtualinherited

Disable stopping criterion on timeout.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 1067 of file genericBNLearner.h.

                             {
         K2__.approximationScheme().disableMaxTime();
         greedy_hill_climbing__.disableMaxTime();
         local_search_with_tabu_list__.disableMaxTime();
         Dag2BN__.disableMaxTime();
       };

◆ disableMinEpsilonRate()

void gum::learning::genericBNLearner::disableMinEpsilonRate ( )

inlinevirtualinherited

Disable stopping criterion on epsilon rate.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 970 of file genericBNLearner.h.

                                    {
         K2__.approximationScheme().disableMinEpsilonRate();
         greedy_hill_climbing__.disableMinEpsilonRate();
         local_search_with_tabu_list__.disableMinEpsilonRate();
         Dag2BN__.disableMinEpsilonRate();
       };

◆ distributeProgress()

INLINE void gum::learning::genericBNLearner::distributeProgress	(	const ApproximationScheme *	approximationScheme,
		Size	pourcent,
		double	error,
		double	time
	)

inlineinherited

{@ /// distribute signals

Definition at line 886 of file genericBNLearner.h.

                                                              {
         setCurrentApproximationScheme(approximationScheme);
 
         if (onProgress.hasListener()) GUM_EMIT3(onProgress, pourcent, error, time);
       };

◆ distributeStop()

INLINE void gum::learning::genericBNLearner::distributeStop	(	const ApproximationScheme *	approximationScheme,
		std::string	message
	)

inlineinherited

distribute signals

Definition at line 896 of file genericBNLearner.h.

                                                                    {
         setCurrentApproximationScheme(approximationScheme);
 
         if (onStop.hasListener()) GUM_EMIT1(onStop, message);
       };

◆ domainSize() [1/2]

INLINE Size gum::learning::genericBNLearner::domainSize ( NodeId var ) const

inherited

learn a structure from a file (must have read the db before)

Definition at line 539 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                              {
       return score_database__.domainSizes()[var];
     }

Here is the call graph for this function:

◆ domainSize() [2/2]

INLINE Size gum::learning::genericBNLearner::domainSize ( const std::string & var ) const

inherited

learn a structure from a file (must have read the db before)

Definition at line 543 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                        {
       return score_database__.domainSizes()[idFromName(var)];
     }

Here is the call graph for this function:

◆ domainSizes()

INLINE const std::vector< std::size_t > & gum::learning::genericBNLearner::domainSizes ( ) const

inherited

returns the domain sizes of the variables in the database

Definition at line 534 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                      {
       return score_database__.domainSizes();
     }

Here is the call graph for this function:

◆ enableEpsilon()

void gum::learning::genericBNLearner::enableEpsilon ( )

inlinevirtualinherited

Enable stopping criterion on epsilon.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 932 of file genericBNLearner.h.

                            {
         K2__.approximationScheme().enableEpsilon();
         greedy_hill_climbing__.enableEpsilon();
         local_search_with_tabu_list__.enableEpsilon();
         Dag2BN__.enableEpsilon();
       };

◆ enableMaxIter()

void gum::learning::genericBNLearner::enableMaxIter ( )

inlinevirtualinherited

Enable stopping criterion on max iterations.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 1021 of file genericBNLearner.h.

                            {
         K2__.approximationScheme().enableMaxIter();
         greedy_hill_climbing__.enableMaxIter();
         local_search_with_tabu_list__.enableMaxIter();
         Dag2BN__.enableMaxIter();
       };

◆ enableMaxTime()

void gum::learning::genericBNLearner::enableMaxTime ( )

inlinevirtualinherited

stopping criterion on timeout If the criterion was disabled it will be enabled

Exceptions

OutOfLowerBound if timeout<=0.0 timeout is time in second (double).

Implements gum::IApproximationSchemeConfiguration.

Definition at line 1073 of file genericBNLearner.h.

                            {
         K2__.approximationScheme().enableMaxTime();
         greedy_hill_climbing__.enableMaxTime();
         local_search_with_tabu_list__.enableMaxTime();
         Dag2BN__.enableMaxTime();
       };

◆ enableMinEpsilonRate()

void gum::learning::genericBNLearner::enableMinEpsilonRate ( )

inlinevirtualinherited

Enable stopping criterion on epsilon rate.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 977 of file genericBNLearner.h.

                                   {
         K2__.approximationScheme().enableMinEpsilonRate();
         greedy_hill_climbing__.enableMinEpsilonRate();
         local_search_with_tabu_list__.enableMinEpsilonRate();
         Dag2BN__.enableMinEpsilonRate();
       };

◆ epsilon()

double gum::learning::genericBNLearner::epsilon ( ) const

inlinevirtualinherited

Get the value of epsilon.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 916 of file genericBNLearner.h.

                              {
         if (current_algorithm__ != nullptr)
           return current_algorithm__->epsilon();
         else
           GUM_ERROR(FatalError, "No chosen algorithm for learning");
       };

◆ eraseForbiddenArc() [1/3]

INLINE void gum::learning::genericBNLearner::eraseForbiddenArc ( const Arc & arc )

inherited

Definition at line 363 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                   {
       constraint_ForbiddenArcs__.eraseArc(arc);
     }

Here is the call graph for this function:

◆ eraseForbiddenArc() [2/3]

INLINE void gum::learning::genericBNLearner::eraseForbiddenArc	(	const NodeId	tail,
		const NodeId	head
	)

inherited

Definition at line 374 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                        {
       eraseForbiddenArc(Arc(tail, head));
     }

Here is the call graph for this function:

◆ eraseForbiddenArc() [3/3]

INLINE void gum::learning::genericBNLearner::eraseForbiddenArc	(	const std::string &	tail,
		const std::string &	head
	)

inherited

Definition at line 386 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                            {
       eraseForbiddenArc(Arc(idFromName(tail), idFromName(head)));
     }

Here is the call graph for this function:

◆ eraseMandatoryArc() [1/3]

INLINE void gum::learning::genericBNLearner::eraseMandatoryArc ( const Arc & arc )

inherited

Definition at line 402 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                   {
       constraint_MandatoryArcs__.eraseArc(arc);
     }

Here is the call graph for this function:

◆ eraseMandatoryArc() [2/3]

INLINE void gum::learning::genericBNLearner::eraseMandatoryArc	(	const NodeId	tail,
		const NodeId	head
	)

inherited

Definition at line 425 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                        {
       eraseMandatoryArc(Arc(tail, head));
     }

Here is the call graph for this function:

◆ eraseMandatoryArc() [3/3]

INLINE void gum::learning::genericBNLearner::eraseMandatoryArc	(	const std::string &	tail,
		const std::string &	head
	)

inherited

Definition at line 413 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                            {
       eraseMandatoryArc(Arc(idFromName(tail), idFromName(head)));
     }

Here is the call graph for this function:

◆ erasePossibleEdge() [1/3]

INLINE void gum::learning::genericBNLearner::erasePossibleEdge ( const Edge & edge )

inherited

Definition at line 324 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                     {
       constraint_PossibleEdges__.eraseEdge(edge);
     }

Here is the call graph for this function:

◆ erasePossibleEdge() [2/3]

INLINE void gum::learning::genericBNLearner::erasePossibleEdge	(	const NodeId	tail,
		const NodeId	head
	)

inherited

Definition at line 335 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                        {
       erasePossibleEdge(Edge(tail, head));
     }

Here is the call graph for this function:

◆ erasePossibleEdge() [3/3]

INLINE void gum::learning::genericBNLearner::erasePossibleEdge	(	const std::string &	tail,
		const std::string &	head
	)

inherited

Definition at line 347 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                            {
       erasePossibleEdge(Edge(idFromName(tail), idFromName(head)));
     }

Here is the call graph for this function:

◆ G2() [1/2]

std::pair< double, double > gum::learning::genericBNLearner::G2	(	const NodeId	id1,
		const NodeId	id2,
		const std::vector< NodeId > &	knowing = `{}`
	)

inherited

Return the <statistic,pvalue> pair for for G2 test in the database.

Parameters

id1	first variable
id2	second variable
knowing	list of observed variables

Returns: a std::pair<double,double>

Definition at line 973 of file genericBNLearner.cpp.

                                                                 {
       createApriori__();
       gum::learning::IndepTestG2<> g2score(score_database__.parser(),
                                            *apriori__,
                                            databaseRanges());
       return g2score.statistics(id1, id2, knowing);
     }

◆ G2() [2/2]

std::pair< double, double > gum::learning::genericBNLearner::G2	(	const std::string &	name1,
		const std::string &	name2,
		const std::vector< std::string > &	knowing = `{}`
	)

inherited

Return the <statistic,pvalue> pair for for G2 test in the database.

Parameters

id1	first variable
id2	second variable
knowing	list of observed variables

Returns: a std::pair<double,double>

Definition at line 984 of file genericBNLearner.cpp.

                                                                    {
       std::vector< NodeId > knowingIds;
       std::transform(
          knowing.begin(),
          knowing.end(),
          std::back_inserter(knowingIds),
          [this](const std::string& c) -> NodeId { return this->idFromName(c); });
       return G2(idFromName(name1), idFromName(name2), knowingIds);
     }

◆ getAprioriType__()

INLINE const std::string & gum::learning::genericBNLearner::getAprioriType__ ( ) const

protectedinherited

returns the type (as a string) of a given apriori

Definition at line 506 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                      {
       switch (apriori_type__) {
         case AprioriType::NO_APRIORI:
           return AprioriNoApriori<>::type::type;
 
         case AprioriType::SMOOTHING:
           return AprioriSmoothing<>::type::type;
 
         case AprioriType::DIRICHLET_FROM_DATABASE:
           return AprioriDirichletFromDatabase<>::type::type;
 
         case AprioriType::BDEU:
           return AprioriBDeu<>::type::type;
 
         default:
           GUM_ERROR(OperationNotAllowed,
                     "genericBNLearner getAprioriType does "
                     "not support yet this apriori");
       }
     }

Here is the call graph for this function:

◆ hasMissingValues()

INLINE bool gum::learning::genericBNLearner::hasMissingValues ( ) const

inherited

returns true if the learner's database has missing values

Definition at line 305 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                          {
       return score_database__.databaseTable().hasMissingValues();
     }

Here is the call graph for this function:

◆ history()

const std::vector< double >& gum::learning::genericBNLearner::history ( ) const

inlinevirtualinherited

Exceptions

OperationNotAllowed if scheme not performed or verbosity=false

Implements gum::IApproximationSchemeConfiguration.

Definition at line 1143 of file genericBNLearner.h.

                                                  {
         if (current_algorithm__ != nullptr)
           return current_algorithm__->history();
         else
           GUM_ERROR(FatalError, "No chosen algorithm for learning");
       };

◆ idFromName()

INLINE NodeId gum::learning::genericBNLearner::idFromName ( const std::string & var_name ) const

inherited

returns the node id corresponding to a variable name

Exceptions

MissingVariableInDatabase if a variable of the BN is not found in the database.

Definition at line 146 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                               {
       return score_database__.idFromName(var_name);
     }

Here is the call graph for this function:

◆ isEnabledEpsilon()

bool gum::learning::genericBNLearner::isEnabledEpsilon ( ) const

inlinevirtualinherited

Returns: true if stopping criterion on epsilon is enabled, false otherwise

Implements gum::IApproximationSchemeConfiguration.

Definition at line 941 of file genericBNLearner.h.

                                     {
         if (current_algorithm__ != nullptr)
           return current_algorithm__->isEnabledEpsilon();
         else
           GUM_ERROR(FatalError, "No chosen algorithm for learning");
       };

◆ isEnabledMaxIter()

bool gum::learning::genericBNLearner::isEnabledMaxIter ( ) const

inlinevirtualinherited

Returns: true if stopping criterion on max iterations is enabled, false otherwise

Implements gum::IApproximationSchemeConfiguration.

Definition at line 1029 of file genericBNLearner.h.

                                     {
         if (current_algorithm__ != nullptr)
           return current_algorithm__->isEnabledMaxIter();
         else
           GUM_ERROR(FatalError, "No chosen algorithm for learning");
       };

◆ isEnabledMaxTime()

bool gum::learning::genericBNLearner::isEnabledMaxTime ( ) const

inlinevirtualinherited

Returns: true if stopping criterion on timeout is enabled, false otherwise

Implements gum::IApproximationSchemeConfiguration.

Definition at line 1081 of file genericBNLearner.h.

                                     {
         if (current_algorithm__ != nullptr)
           return current_algorithm__->isEnabledMaxTime();
         else
           GUM_ERROR(FatalError, "No chosen algorithm for learning");
       };

◆ isEnabledMinEpsilonRate()

bool gum::learning::genericBNLearner::isEnabledMinEpsilonRate ( ) const

inlinevirtualinherited

Returns: true if stopping criterion on epsilon rate is enabled, false otherwise

Implements gum::IApproximationSchemeConfiguration.

Definition at line 985 of file genericBNLearner.h.

                                            {
         if (current_algorithm__ != nullptr)
           return current_algorithm__->isEnabledMinEpsilonRate();
         else
           GUM_ERROR(FatalError, "No chosen algorithm for learning");
       };

◆ labelsFromBN__()

template<typename GUM_SCALAR >

NodeProperty< Sequence< std::string > > gum::learning::BNLearner< GUM_SCALAR >::labelsFromBN__	(	const std::string &	filename,
		const BayesNet< GUM_SCALAR > &	src
	)

private

read the first line of a file to find column names

◆ latentVariables()

INLINE const std::vector< Arc > gum::learning::genericBNLearner::latentVariables ( ) const

inherited

get the list of arcs hiding latent variables

Exceptions

OperationNotAllowed when 3off2 is not the selected algorithm

Definition at line 265 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                           {
       if (selected_algo__ != AlgoType::MIIC_THREE_OFF_TWO) {
         GUM_ERROR(OperationNotAllowed,
                   "You must use the 3off2 algorithm before selecting "
                      << "the latentVariables method");
       }
       return miic_3off2__.latentVariables();
     }

Here is the call graph for this function:

◆ learnBN()

template<typename GUM_SCALAR >

BayesNet< GUM_SCALAR > gum::learning::BNLearner< GUM_SCALAR >::learnBN ( )

learn a Bayes Net from a file (must have read the db before)

◆ learnDAG()

DAG gum::learning::genericBNLearner::learnDAG ( )

inherited

learn a structure from a file (must have read the db before)

Definition at line 658 of file genericBNLearner.cpp.

                                    {
       // create the score and the apriori
       createApriori__();
       createScore__();
 
       return learnDAG__();
     }

◆ learnDAG__()

DAG gum::learning::genericBNLearner::learnDAG__ ( )

protectedinherited

returns the DAG learnt

Definition at line 694 of file genericBNLearner.cpp.

                                      {
       // check that the database does not contain any missing value
       if (score_database__.databaseTable().hasMissingValues()
           || ((apriori_database__ != nullptr)
               && (apriori_type__ == AprioriType::DIRICHLET_FROM_DATABASE)
               && apriori_database__->databaseTable().hasMissingValues())) {
         GUM_ERROR(MissingValueInDatabase,
                   "For the moment, the BNLearner is unable to cope "
                   "with missing values in databases");
       }
       // add the mandatory arcs to the initial dag and remove the forbidden ones
       // from the initial graph
       DAG init_graph = initial_dag__;
 
       const ArcSet& mandatory_arcs = constraint_MandatoryArcs__.arcs();
 
       for (const auto& arc: mandatory_arcs) {
         if (!init_graph.exists(arc.tail())) init_graph.addNodeWithId(arc.tail());
 
         if (!init_graph.exists(arc.head())) init_graph.addNodeWithId(arc.head());
 
         init_graph.addArc(arc.tail(), arc.head());
       }
 
       const ArcSet& forbidden_arcs = constraint_ForbiddenArcs__.arcs();
 
       for (const auto& arc: forbidden_arcs) {
         init_graph.eraseArc(arc);
       }
 
       switch (selected_algo__) {
         // ========================================================================
         case AlgoType::MIIC_THREE_OFF_TWO: {
           BNLearnerListener listener(this, miic_3off2__);
           // create the mixedGraph and the corrected mutual information
           MixedGraph mgraph = this->prepare_miic_3off2__();
 
           return miic_3off2__.learnStructure(*mutual_info__, mgraph);
         }
 
         // ========================================================================
         case AlgoType::GREEDY_HILL_CLIMBING: {
           BNLearnerListener listener(this, greedy_hill_climbing__);
           StructuralConstraintSetStatic< StructuralConstraintMandatoryArcs,
                                          StructuralConstraintForbiddenArcs,
                                          StructuralConstraintPossibleEdges,
                                          StructuralConstraintSliceOrder >
              gen_constraint;
           static_cast< StructuralConstraintMandatoryArcs& >(gen_constraint)
              = constraint_MandatoryArcs__;
           static_cast< StructuralConstraintForbiddenArcs& >(gen_constraint)
              = constraint_ForbiddenArcs__;
           static_cast< StructuralConstraintPossibleEdges& >(gen_constraint)
              = constraint_PossibleEdges__;
           static_cast< StructuralConstraintSliceOrder& >(gen_constraint)
              = constraint_SliceOrder__;
 
           GraphChangesGenerator4DiGraph< decltype(gen_constraint) > op_set(
              gen_constraint);
 
           StructuralConstraintSetStatic< StructuralConstraintIndegree,
                                          StructuralConstraintDAG >
              sel_constraint;
           static_cast< StructuralConstraintIndegree& >(sel_constraint)
              = constraint_Indegree__;
 
           GraphChangesSelector4DiGraph< decltype(sel_constraint),
                                         decltype(op_set) >
              selector(*score__, sel_constraint, op_set);
 
           return greedy_hill_climbing__.learnStructure(selector, init_graph);
         }
 
         // ========================================================================
         case AlgoType::LOCAL_SEARCH_WITH_TABU_LIST: {
           BNLearnerListener listener(this, local_search_with_tabu_list__);
           StructuralConstraintSetStatic< StructuralConstraintMandatoryArcs,
                                          StructuralConstraintForbiddenArcs,
                                          StructuralConstraintPossibleEdges,
                                          StructuralConstraintSliceOrder >
              gen_constraint;
           static_cast< StructuralConstraintMandatoryArcs& >(gen_constraint)
              = constraint_MandatoryArcs__;
           static_cast< StructuralConstraintForbiddenArcs& >(gen_constraint)
              = constraint_ForbiddenArcs__;
           static_cast< StructuralConstraintPossibleEdges& >(gen_constraint)
              = constraint_PossibleEdges__;
           static_cast< StructuralConstraintSliceOrder& >(gen_constraint)
              = constraint_SliceOrder__;
 
           GraphChangesGenerator4DiGraph< decltype(gen_constraint) > op_set(
              gen_constraint);
 
           StructuralConstraintSetStatic< StructuralConstraintTabuList,
                                          StructuralConstraintIndegree,
                                          StructuralConstraintDAG >
              sel_constraint;
           static_cast< StructuralConstraintTabuList& >(sel_constraint)
              = constraint_TabuList__;
           static_cast< StructuralConstraintIndegree& >(sel_constraint)
              = constraint_Indegree__;
 
           GraphChangesSelector4DiGraph< decltype(sel_constraint),
                                         decltype(op_set) >
              selector(*score__, sel_constraint, op_set);
 
           return local_search_with_tabu_list__.learnStructure(selector,
                                                               init_graph);
         }
 
         // ========================================================================
         case AlgoType::K2: {
           BNLearnerListener listener(this, K2__.approximationScheme());
           StructuralConstraintSetStatic< StructuralConstraintMandatoryArcs,
                                          StructuralConstraintForbiddenArcs,
                                          StructuralConstraintPossibleEdges >
              gen_constraint;
           static_cast< StructuralConstraintMandatoryArcs& >(gen_constraint)
              = constraint_MandatoryArcs__;
           static_cast< StructuralConstraintForbiddenArcs& >(gen_constraint)
              = constraint_ForbiddenArcs__;
           static_cast< StructuralConstraintPossibleEdges& >(gen_constraint)
              = constraint_PossibleEdges__;
 
           GraphChangesGenerator4K2< decltype(gen_constraint) > op_set(
              gen_constraint);
 
           // if some mandatory arcs are incompatible with the order, use a DAG
           // constraint instead of a DiGraph constraint to avoid cycles
           const ArcSet& mandatory_arcs
              = static_cast< StructuralConstraintMandatoryArcs& >(gen_constraint)
                   .arcs();
           const Sequence< NodeId >& order            = K2__.order();
           bool                      order_compatible = true;
 
           for (const auto& arc: mandatory_arcs) {
             if (order.pos(arc.tail()) >= order.pos(arc.head())) {
               order_compatible = false;
               break;
             }
           }
 
           if (order_compatible) {
             StructuralConstraintSetStatic< StructuralConstraintIndegree,
                                            StructuralConstraintDiGraph >
                sel_constraint;
             static_cast< StructuralConstraintIndegree& >(sel_constraint)
                = constraint_Indegree__;
 
             GraphChangesSelector4DiGraph< decltype(sel_constraint),
                                           decltype(op_set) >
                selector(*score__, sel_constraint, op_set);
 
             return K2__.learnStructure(selector, init_graph);
           } else {
             StructuralConstraintSetStatic< StructuralConstraintIndegree,
                                            StructuralConstraintDAG >
                sel_constraint;
             static_cast< StructuralConstraintIndegree& >(sel_constraint)
                = constraint_Indegree__;
 
             GraphChangesSelector4DiGraph< decltype(sel_constraint),
                                           decltype(op_set) >
                selector(*score__, sel_constraint, op_set);
 
             return K2__.learnStructure(selector, init_graph);
           }
         }
 
         // ========================================================================
         default:
           GUM_ERROR(OperationNotAllowed,
                     "the learnDAG method has not been implemented for this "
                     "learning algorithm");
       }
     }

◆ learnMixedStructure()

MixedGraph gum::learning::genericBNLearner::learnMixedStructure ( )

inherited

learn a partial structure from a file (must have read the db before and must have selected miic or 3off2)

Definition at line 640 of file genericBNLearner.cpp.

                                                      {
       if (selected_algo__ != AlgoType::MIIC_THREE_OFF_TWO) {
         GUM_ERROR(OperationNotAllowed, "Must be using the miic/3off2 algorithm");
       }
       // check that the database does not contain any missing value
       if (score_database__.databaseTable().hasMissingValues()) {
         GUM_ERROR(MissingValueInDatabase,
                   "For the moment, the BNLearner is unable to learn "
                      << "structures with missing values in databases");
       }
       BNLearnerListener listener(this, miic_3off2__);
 
       // create the mixedGraph_constraint_MandatoryArcs.arcs();
       MixedGraph mgraph = this->prepare_miic_3off2__();
 
       return miic_3off2__.learnMixedStructure(*mutual_info__, mgraph);
     }

◆ learnParameters() [1/2]

template<typename GUM_SCALAR >

BayesNet< GUM_SCALAR > gum::learning::BNLearner< GUM_SCALAR >::learnParameters	(	const DAG &	dag,
		bool	take_into_account_score = `true`
	)

learns a BN (its parameters) when its structure is known

Parameters

dag the structure of the Bayesian network

take_into_account_score The dag passed in argument may have been learnt from a structure learning. In this case, if the score used to learn the structure has an implicit apriori (like K2 which has a 1-smoothing apriori), it is important to also take into account this implicit apriori for parameter learning. By default, if a score exists, we will learn parameters by taking into account the apriori specified by methods useAprioriXXX () + the implicit apriori of the score, else we just take into account the apriori specified by useAprioriXXX ()

◆ learnParameters() [2/2]

template<typename GUM_SCALAR >

BayesNet< GUM_SCALAR > gum::learning::BNLearner< GUM_SCALAR >::learnParameters ( bool take_into_account_score = true )

Parameters

take_into_account_score The dag of the BN which was passed in argument to the BNLearner may have been learnt from a structure learning. In this case, if the score used to learn the structure has an implicit apriori (like K2 which has a 1-smoothing apriori), it is important to also take into account this implicit apriori for parameter learning. By default, if a score exists, we will learn parameters by taking into account the apriori specified by methods useAprioriXXX () + the implicit apriori of the score, else we just take into account the apriori specified by useAprioriXXX ()

Exceptions

MissingVariableInDatabase	if a variable of the BN is not found in the database.
UnknownLabelInDatabase	if a label is found in the databast that do not correpond to the variable.

◆ logLikelihood() [1/2]

double gum::learning::genericBNLearner::logLikelihood	(	const std::vector< NodeId > &	vars,
		const std::vector< NodeId > &	knowing = `{}`
	)

inherited

Return the loglikelihood of vars in the base, conditioned by knowing for the BNLearner.

Parameters

vars	a vector of NodeIds
knowing	an optional vector of conditioning NodeIds

Returns: a std::pair<double,double>

Definition at line 996 of file genericBNLearner.cpp.

                                                                                {
       createApriori__();
       gum::learning::ScoreLog2Likelihood<> ll2score(score_database__.parser(),
                                                     *apriori__,
                                                     databaseRanges());
 
       std::vector< NodeId > total(vars);
       total.insert(total.end(), knowing.begin(), knowing.end());
       double LLtotal = ll2score.score(IdCondSet<>(total, false, true));
       if (knowing.size() == (Size)0) {
         return LLtotal;
       } else {
         double LLknw = ll2score.score(IdCondSet<>(knowing, false, true));
         return LLtotal - LLknw;
       }
     }

◆ logLikelihood() [2/2]

double gum::learning::genericBNLearner::logLikelihood	(	const std::vector< std::string > &	vars,
		const std::vector< std::string > &	knowing = `{}`
	)

inherited

Return the loglikelihood of vars in the base, conditioned by knowing for the BNLearner.

Parameters

vars	a vector of name of rows
knowing	an optional vector of conditioning rows

Returns: a std::pair<double,double>

Definition at line 1015 of file genericBNLearner.cpp.

                                                                               {
       std::vector< NodeId > ids;
       std::vector< NodeId > knowingIds;
 
       auto mapper = [this](const std::string& c) -> NodeId {
         return this->idFromName(c);
       };
 
       std::transform(vars.begin(), vars.end(), std::back_inserter(ids), mapper);
       std::transform(knowing.begin(),
                      knowing.end(),
                      std::back_inserter(knowingIds),
                      mapper);
 
       return logLikelihood(ids, knowingIds);
     }

◆ maxIter()

Size gum::learning::genericBNLearner::maxIter ( ) const

inlinevirtualinherited

Returns: the criterion on number of iterations

Implements gum::IApproximationSchemeConfiguration.

Definition at line 1006 of file genericBNLearner.h.

                            {
         if (current_algorithm__ != nullptr)
           return current_algorithm__->maxIter();
         else
           GUM_ERROR(FatalError, "No chosen algorithm for learning");
       };

◆ maxTime()

double gum::learning::genericBNLearner::maxTime ( ) const

inlinevirtualinherited

returns the timeout (in seconds)

Implements gum::IApproximationSchemeConfiguration.

Definition at line 1051 of file genericBNLearner.h.

                              {
         if (current_algorithm__ != nullptr)
           return current_algorithm__->maxTime();
         else
           GUM_ERROR(FatalError, "No chosen algorithm for learning");
       };

◆ messageApproximationScheme()

INLINE std::string gum::IApproximationSchemeConfiguration::messageApproximationScheme ( ) const

inherited

Returns the approximation scheme message.

Returns: Returns the approximation scheme message.

Definition at line 39 of file IApproximationSchemeConfiguration_inl.h.

References gum::Set< Key, Alloc >::emplace().

                                                                          {
     std::stringstream s;
 
     switch (stateApproximationScheme()) {
       case ApproximationSchemeSTATE::Continue:
         s << "in progress";
         break;
 
       case ApproximationSchemeSTATE::Epsilon:
         s << "stopped with epsilon=" << epsilon();
         break;
 
       case ApproximationSchemeSTATE::Rate:
         s << "stopped with rate=" << minEpsilonRate();
         break;
 
       case ApproximationSchemeSTATE::Limit:
         s << "stopped with max iteration=" << maxIter();
         break;
 
       case ApproximationSchemeSTATE::TimeLimit:
         s << "stopped with timeout=" << maxTime();
         break;
 
       case ApproximationSchemeSTATE::Stopped:
         s << "stopped on request";
         break;
 
       case ApproximationSchemeSTATE::Undefined:
         s << "undefined state";
         break;
     };
 
     return s.str();
   }

Here is the call graph for this function:

◆ minEpsilonRate()

double gum::learning::genericBNLearner::minEpsilonRate ( ) const

inlinevirtualinherited

Get the value of the minimal epsilon rate.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 962 of file genericBNLearner.h.

                                     {
         if (current_algorithm__ != nullptr)
           return current_algorithm__->minEpsilonRate();
         else
           GUM_ERROR(FatalError, "No chosen algorithm for learning");
       };

◆ nameFromId()

INLINE const std::string & gum::learning::genericBNLearner::nameFromId ( NodeId id ) const

inherited

returns the variable name corresponding to a given node id

Definition at line 151 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                         {
       return score_database__.nameFromId(id);
     }

Here is the call graph for this function:

◆ names()

INLINE const std::vector< std::string > & gum::learning::genericBNLearner::names ( ) const

inherited

returns the names of the variables in the database

Definition at line 528 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                        {
       return score_database__.names();
     }

Here is the call graph for this function:

◆ nbCols()

INLINE Size gum::learning::genericBNLearner::nbCols ( ) const

inherited

Returns: the number of cols in the database

Definition at line 561 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                {
       return score_database__.domainSizes().size();
     }

Here is the call graph for this function:

◆ nbrIterations()

Size gum::learning::genericBNLearner::nbrIterations ( ) const

inlinevirtualinherited

Exceptions

OperationNotAllowed if scheme not performed

Implements gum::IApproximationSchemeConfiguration.

Definition at line 1135 of file genericBNLearner.h.

                                  {
         if (current_algorithm__ != nullptr)
           return current_algorithm__->nbrIterations();
         else
           GUM_ERROR(FatalError, "No chosen algorithm for learning");
       };

◆ nbRows()

INLINE Size gum::learning::genericBNLearner::nbRows ( ) const

inherited

Returns: the number of rows in the database

Definition at line 565 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                {
       return score_database__.databaseTable().size();
     }

Here is the call graph for this function:

◆ operator=() [1/2]

template<typename GUM_SCALAR >

BNLearner& gum::learning::BNLearner< GUM_SCALAR >::operator= ( const BNLearner< GUM_SCALAR > & )

copy operator

◆ operator=() [2/2]

template<typename GUM_SCALAR >

BNLearner& gum::learning::BNLearner< GUM_SCALAR >::operator= ( BNLearner< GUM_SCALAR > && )

move operator

◆ periodSize()

Size gum::learning::genericBNLearner::periodSize ( ) const

inlinevirtualinherited

how many samples between 2 stopping isEnableds

Exceptions

OutOfLowerBound if p<1

Implements gum::IApproximationSchemeConfiguration.

Definition at line 1099 of file genericBNLearner.h.

                               {
         if (current_algorithm__ != nullptr)
           return current_algorithm__->periodSize();
         else
           GUM_ERROR(FatalError, "No chosen algorithm for learning");
       };

◆ prepare_miic_3off2__()

MixedGraph gum::learning::genericBNLearner::prepare_miic_3off2__ ( )

protectedinherited

prepares the initial graph for 3off2 or miic

Definition at line 610 of file genericBNLearner.cpp.

                                                       {
       // Initialize the mixed graph to the fully connected graph
       MixedGraph mgraph;
       for (Size i = 0; i < score_database__.databaseTable().nbVariables(); ++i) {
         mgraph.addNodeWithId(i);
         for (Size j = 0; j < i; ++j) {
           mgraph.addEdge(j, i);
         }
       }
 
       // translating the constraints for 3off2 or miic
       HashTable< std::pair< NodeId, NodeId >, char > initial_marks;
       const ArcSet& mandatory_arcs = constraint_MandatoryArcs__.arcs();
       for (const auto& arc: mandatory_arcs) {
         initial_marks.insert({arc.tail(), arc.head()}, '>');
       }
 
       const ArcSet& forbidden_arcs = constraint_ForbiddenArcs__.arcs();
       for (const auto& arc: forbidden_arcs) {
         initial_marks.insert({arc.tail(), arc.head()}, '-');
       }
       miic_3off2__.addConstraints(initial_marks);
 
       // create the mutual entropy object
       // if (mutual_info__ == nullptr) { this->useNML(); }
       createCorrectedMutualInformation__();
 
       return mgraph;
     }

◆ rawPseudoCount() [1/2]

std::vector< double > gum::learning::genericBNLearner::rawPseudoCount ( const std::vector< NodeId > & vars )

inherited

Return the pseudoconts ofNodeIds vars in the base in a raw array.

Parameters

vars	a vector of

Returns: a a std::vector<double> containing the contingency table

Definition at line 1034 of file genericBNLearner.cpp.

                                                                        {
       Potential< double > res;
 
       createApriori__();
       gum::learning::PseudoCount<> count(score_database__.parser(),
                                          *apriori__,
                                          databaseRanges());
       return count.get(vars);
     }

◆ rawPseudoCount() [2/2]

std::vector< double > gum::learning::genericBNLearner::rawPseudoCount ( const std::vector< std::string > & vars )

inherited

Return the pseudoconts of vars in the base in a raw array.

Parameters

vars	a vector of name

Returns: a std::vector<double> containing the contingency table

Definition at line 1046 of file genericBNLearner.cpp.

                                                                           {
       std::vector< NodeId > ids;
 
       auto mapper = [this](const std::string& c) -> NodeId {
         return this->idFromName(c);
       };
 
       std::transform(vars.begin(), vars.end(), std::back_inserter(ids), mapper);
 
       return rawPseudoCount(ids);
     }

◆ readFile__()

DatabaseTable gum::learning::genericBNLearner::readFile__	(	const std::string &	filename,
		const std::vector< std::string > &	missing_symbols
	)

staticprotectedinherited

reads a file and returns a databaseVectInRam

Definition at line 438 of file genericBNLearner.cpp.

                                                       {
       // get the extension of the file
       checkFileName__(filename);
 
       DBInitializerFromCSV<> initializer(filename);
 
       const auto&       var_names = initializer.variableNames();
       const std::size_t nb_vars   = var_names.size();
 
       DBTranslatorSet<>                translator_set;
       DBTranslator4LabelizedVariable<> translator(missing_symbols);
       for (std::size_t i = 0; i < nb_vars; ++i) {
         translator_set.insertTranslator(translator, i);
       }
 
       DatabaseTable<> database(missing_symbols, translator_set);
       database.setVariableNames(initializer.variableNames());
       initializer.fillDatabase(database);
 
       database.reorder();
 
       return database;
     }

◆ recordWeight()

INLINE double gum::learning::genericBNLearner::recordWeight ( const std::size_t i ) const

inherited

returns the weight of the ith record

Exceptions

OutOfBounds if i is outside the set of indices of the records

Definition at line 167 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                         {
       return score_database__.weight(i);
     }

Here is the call graph for this function:

◆ setAprioriWeight__()

INLINE void gum::learning::genericBNLearner::setAprioriWeight__ ( double weight )

inherited

sets the apriori weight

Definition at line 450 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                   {
       if (weight < 0) {
         GUM_ERROR(OutOfBounds, "the weight of the apriori must be positive");
       }
 
       apriori_weight__ = weight;
       checkScoreAprioriCompatibility();
     }

Here is the call graph for this function:

◆ setCurrentApproximationScheme()

INLINE void gum::learning::genericBNLearner::setCurrentApproximationScheme ( const ApproximationScheme * approximationScheme )

inlineinherited

{@ /// distribute signals

Definition at line 880 of file genericBNLearner.h.

                                                          {
         current_algorithm__ = approximationScheme;
       }

◆ setDatabaseWeight()

INLINE void gum::learning::genericBNLearner::setDatabaseWeight ( const double new_weight )

inherited

assign a weight to all the rows of the learning database so that the sum of their weights is equal to new_weight

assign new weight to the rows of the learning database

Definition at line 156 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                            {
       score_database__.setDatabaseWeight(new_weight);
     }

Here is the call graph for this function:

◆ setEpsilon()

void gum::learning::genericBNLearner::setEpsilon ( double eps )

inlinevirtualinherited

Given that we approximate f(t), stopping criterion on |f(t+1)-f(t)| If the criterion was disabled it will be enabled.

Exceptions

OutOfLowerBound if eps<0

Implements gum::IApproximationSchemeConfiguration.

Definition at line 908 of file genericBNLearner.h.

                                   {
         K2__.approximationScheme().setEpsilon(eps);
         greedy_hill_climbing__.setEpsilon(eps);
         local_search_with_tabu_list__.setEpsilon(eps);
         Dag2BN__.setEpsilon(eps);
       };

◆ setForbiddenArcs()

INLINE void gum::learning::genericBNLearner::setForbiddenArcs ( const ArcSet & set )

inherited

assign a set of forbidden arcs

Definition at line 353 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                     {
       constraint_ForbiddenArcs__.setArcs(set);
     }

Here is the call graph for this function:

◆ setInitialDAG()

INLINE void gum::learning::genericBNLearner::setInitialDAG ( const DAG & dag )

inherited

sets an initial DAG structure

Definition at line 177 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                               {
       initial_dag__ = dag;
     }

Here is the call graph for this function:

◆ setMandatoryArcs()

INLINE void gum::learning::genericBNLearner::setMandatoryArcs ( const ArcSet & set )

inherited

assign a set of forbidden arcs

Definition at line 392 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                     {
       constraint_MandatoryArcs__.setArcs(set);
     }

Here is the call graph for this function:

◆ setMaxIndegree()

INLINE void gum::learning::genericBNLearner::setMaxIndegree ( Size max_indegree )

inherited

sets the max indegree

Definition at line 218 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                   {
       constraint_Indegree__.setMaxIndegree(max_indegree);
     }

Here is the call graph for this function:

◆ setMaxIter()

void gum::learning::genericBNLearner::setMaxIter ( Size max )

inlinevirtualinherited

stopping criterion on number of iterationsIf the criterion was disabled it will be enabled

Parameters

max	The maximum number of iterations

Exceptions

OutOfLowerBound if max<=1

Implements gum::IApproximationSchemeConfiguration.

Definition at line 998 of file genericBNLearner.h.

                                 {
         K2__.approximationScheme().setMaxIter(max);
         greedy_hill_climbing__.setMaxIter(max);
         local_search_with_tabu_list__.setMaxIter(max);
         Dag2BN__.setMaxIter(max);
       };

◆ setMaxTime()

void gum::learning::genericBNLearner::setMaxTime ( double timeout )

inlinevirtualinherited

stopping criterion on timeout If the criterion was disabled it will be enabled

Exceptions

OutOfLowerBound if timeout<=0.0 timeout is time in second (double).

Implements gum::IApproximationSchemeConfiguration.

Definition at line 1043 of file genericBNLearner.h.

                                       {
         K2__.approximationScheme().setMaxTime(timeout);
         greedy_hill_climbing__.setMaxTime(timeout);
         local_search_with_tabu_list__.setMaxTime(timeout);
         Dag2BN__.setMaxTime(timeout);
       }

◆ setMinEpsilonRate()

void gum::learning::genericBNLearner::setMinEpsilonRate ( double rate )

inlinevirtualinherited

Given that we approximate f(t), stopping criterion on d/dt(|f(t+1)-f(t)|) If the criterion was disabled it will be enabled.

Exceptions

OutOfLowerBound if rate<0

Implements gum::IApproximationSchemeConfiguration.

Definition at line 954 of file genericBNLearner.h.

                                           {
         K2__.approximationScheme().setMinEpsilonRate(rate);
         greedy_hill_climbing__.setMinEpsilonRate(rate);
         local_search_with_tabu_list__.setMinEpsilonRate(rate);
         Dag2BN__.setMinEpsilonRate(rate);
       };

◆ setPeriodSize()

void gum::learning::genericBNLearner::setPeriodSize ( Size p )

inlinevirtualinherited

how many samples between 2 stopping isEnableds

Exceptions

OutOfLowerBound if p<1

Implements gum::IApproximationSchemeConfiguration.

Definition at line 1092 of file genericBNLearner.h.

                                  {
         K2__.approximationScheme().setPeriodSize(p);
         greedy_hill_climbing__.setPeriodSize(p);
         local_search_with_tabu_list__.setPeriodSize(p);
         Dag2BN__.setPeriodSize(p);
       };

◆ setPossibleEdges()

INLINE void gum::learning::genericBNLearner::setPossibleEdges ( const EdgeSet & set )

inherited

assign a set of forbidden edges

Warning: Once at least one possible edge is defined, all other edges are not possible anymore

Definition at line 310 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                      {
       constraint_PossibleEdges__.setEdges(set);
     }

Here is the call graph for this function:

◆ setPossibleSkeleton()

INLINE void gum::learning::genericBNLearner::setPossibleSkeleton ( const UndiGraph & skeleton )

inherited

assign a set of forbidden edges

Warning: Once at least one possible edge is defined, all other edges are not possible anymore

Definition at line 314 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                            {
       setPossibleEdges(g.edges());
     }

Here is the call graph for this function:

◆ setRecordWeight()

INLINE void gum::learning::genericBNLearner::setRecordWeight	(	const std::size_t	i,
		const double	weight
	)

inherited

sets the weight of the ith record of the database

assign new weight to the ith row of the learning database

Exceptions

OutOfBounds if i is outside the set of indices of the records or if the weight is negative

Definition at line 161 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                                 {
       score_database__.setWeight(i, new_weight);
     }

Here is the call graph for this function:

◆ setSliceOrder() [1/2]

INLINE void gum::learning::genericBNLearner::setSliceOrder ( const NodeProperty< NodeId > & slice_order )

inherited

sets a partial order on the nodes

Parameters

slice_order a NodeProperty given the rank (priority) of nodes in the partial order

Definition at line 432 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                                 {
       constraint_SliceOrder__ = StructuralConstraintSliceOrder(slice_order);
     }

Here is the call graph for this function:

◆ setSliceOrder() [2/2]

INLINE void gum::learning::genericBNLearner::setSliceOrder ( const std::vector< std::vector< std::string > > & slices )

inherited

sets a partial order on the nodes

Parameters

slices the list of list of variable names

Definition at line 436 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                           {
       NodeProperty< NodeId > slice_order;
       NodeId                 rank = 0;
       for (const auto& slice: slices) {
         for (const auto& name: slice) {
           slice_order.insert(idFromName(name), rank);
         }
         rank++;
       }
       setSliceOrder(slice_order);
     }

Here is the call graph for this function:

◆ setVerbosity()

void gum::learning::genericBNLearner::setVerbosity ( bool v )

inlinevirtualinherited

verbosity

Implements gum::IApproximationSchemeConfiguration.

Definition at line 1109 of file genericBNLearner.h.

                                 {
         K2__.approximationScheme().setVerbosity(v);
         greedy_hill_climbing__.setVerbosity(v);
         local_search_with_tabu_list__.setVerbosity(v);
         Dag2BN__.setVerbosity(v);
       };

◆ stateApproximationScheme()

ApproximationSchemeSTATE gum::learning::genericBNLearner::stateApproximationScheme ( ) const

inlinevirtualinherited

history

Implements gum::IApproximationSchemeConfiguration.

Definition at line 1127 of file genericBNLearner.h.

                                                                 {
         if (current_algorithm__ != nullptr)
           return current_algorithm__->stateApproximationScheme();
         else
           GUM_ERROR(FatalError, "No chosen algorithm for learning");
       };

◆ use3off2()

INLINE void gum::learning::genericBNLearner::use3off2 ( )

inherited

indicate that we wish to use 3off2

Definition at line 223 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                            {
       selected_algo__ = AlgoType::MIIC_THREE_OFF_TWO;
       miic_3off2__.set3off2Behaviour();
     }

Here is the call graph for this function:

◆ useAprioriBDeu()

INLINE void gum::learning::genericBNLearner::useAprioriBDeu ( double weight = 1 )

inherited

use the BDeu apriori

The BDeu apriori adds weight to all the cells of the countings tables. In other words, it adds weight rows in the database with equally probable values.

Definition at line 493 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                               {
       if (weight < 0) {
         GUM_ERROR(OutOfBounds, "the weight of the apriori must be positive");
       }
 
       apriori_type__ = AprioriType::BDEU;
       setAprioriWeight__(weight);
 
       checkScoreAprioriCompatibility();
     }

Here is the call graph for this function:

◆ useAprioriDirichlet()

INLINE void gum::learning::genericBNLearner::useAprioriDirichlet	(	const std::string &	filename,
		double	weight = `1`
	)

inherited

use the Dirichlet apriori

Definition at line 478 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                                  {
       if (weight < 0) {
         GUM_ERROR(OutOfBounds, "the weight of the apriori must be positive");
       }
 
       apriori_dbname__ = filename;
       apriori_type__   = AprioriType::DIRICHLET_FROM_DATABASE;
       setAprioriWeight__(weight);
 
       checkScoreAprioriCompatibility();
     }

Here is the call graph for this function:

◆ useAprioriSmoothing()

INLINE void gum::learning::genericBNLearner::useAprioriSmoothing ( double weight = 1 )

inherited

use the apriori smoothing

Parameters

weight pass in argument a weight if you wish to assign a weight to the smoothing, else the current weight of the genericBNLearner will be used.

Definition at line 466 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                    {
       if (weight < 0) {
         GUM_ERROR(OutOfBounds, "the weight of the apriori must be positive");
       }
 
       apriori_type__ = AprioriType::SMOOTHING;
       setAprioriWeight__(weight);
 
       checkScoreAprioriCompatibility();
     }

Here is the call graph for this function:

◆ useCrossValidationFold()

std::pair< std::size_t, std::size_t > gum::learning::genericBNLearner::useCrossValidationFold	(	const std::size_t	learning_fold,
		const std::size_t	k_fold
	)

inherited

sets the ranges of rows to be used for cross-validation learning

When applied on (x,k), the method indicates to the subsequent learnings that they should be performed on the xth fold in a k-fold cross-validation context. For instance, if a database has 1000 rows, and if we perform a 10-fold cross-validation, then, the first learning fold (learning_fold=0) corresponds to rows interval [100,1000) and the test dataset corresponds to [0,100). The second learning fold (learning_fold=1) is [0,100) U [200,1000) and the corresponding test dataset is [100,200).

Parameters

learning_fold	a number indicating the set of rows used for learning. If N denotes the size of the database, and k_fold represents the number of folds in the cross validation, then the set of rows used for testing is [learning_fold * N / k_fold, (learning_fold+1) * N / k_fold) and the learning database is the complement in the database
k_fold	the value of "k" in k-fold cross validation

Returns: a pair [x,y) of rows' indices that corresponds to the indices of rows in the original database that constitute the test dataset

Exceptions

OutOfBounds is raised if k_fold is equal to 0 or learning_fold is greater than or eqal to k_fold, or if k_fold is greater than or equal to the size of the database.

Definition at line 902 of file genericBNLearner.cpp.

                                                                         {
       if (k_fold == 0) {
         GUM_ERROR(OutOfBounds, "K-fold cross validation with k=0 is forbidden");
       }
 
       if (learning_fold >= k_fold) {
         GUM_ERROR(OutOfBounds,
                   "In " << k_fold << "-fold cross validation, the learning "
                         << "fold should be strictly lower than " << k_fold
                         << " but, here, it is equal to " << learning_fold);
       }
 
       const std::size_t db_size = score_database__.databaseTable().nbRows();
       if (k_fold >= db_size) {
         GUM_ERROR(OutOfBounds,
                   "In " << k_fold << "-fold cross validation, the database's "
                         << "size should be strictly greater than " << k_fold
                         << " but, here, the database has only " << db_size
                         << "rows");
       }
 
       // create the ranges of rows of the test database
       const std::size_t foldSize   = db_size / k_fold;
       const std::size_t unfold_deb = learning_fold * foldSize;
       const std::size_t unfold_end = unfold_deb + foldSize;
 
       ranges__.clear();
       if (learning_fold == std::size_t(0)) {
         ranges__.push_back(
            std::pair< std::size_t, std::size_t >(unfold_end, db_size));
       } else {
         ranges__.push_back(
            std::pair< std::size_t, std::size_t >(std::size_t(0), unfold_deb));
 
         if (learning_fold != k_fold - 1) {
           ranges__.push_back(
              std::pair< std::size_t, std::size_t >(unfold_end, db_size));
         }
       }
 
       return std::pair< std::size_t, std::size_t >(unfold_deb, unfold_end);
     }

◆ useDatabaseRanges()

template<template< typename > class XALLOC>

void gum::learning::genericBNLearner::useDatabaseRanges ( const std::vector< std::pair< std::size_t, std::size_t >, XALLOC< std::pair< std::size_t, std::size_t > > > & new_ranges )

inherited

use a new set of database rows' ranges to perform learning

Parameters

ranges a set of pairs {(X1,Y1),...,(Xn,Yn)} of database's rows indices. The subsequent learnings are then performed only on the union of the rows [Xi,Yi), i in {1,...,n}. This is useful, e.g, when performing cross validation tasks, in which part of the database should be ignored. An empty set of ranges is equivalent to an interval [X,Y) ranging over the whole database.

Definition at line 101 of file genericBNLearner_tpl.h.

References gum::learning::genericBNLearner::Database::Database().

                       {
       // use a score to detect whether the ranges are ok
       ScoreLog2Likelihood<> score(score_database__.parser(), *no_apriori__);
       score.setRanges(new_ranges);
       ranges__ = score.ranges();
     }

Here is the call graph for this function:

◆ useEM()

INLINE void gum::learning::genericBNLearner::useEM ( const double epsilon )

inherited

use The EM algorithm to learn paramters

if epsilon=0, EM is not used

Definition at line 300 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                             {
       EMepsilon__ = epsilon;
     }

Here is the call graph for this function:

◆ useGreedyHillClimbing()

INLINE void gum::learning::genericBNLearner::useGreedyHillClimbing ( )

inherited

indicate that we wish to use a greedy hill climbing algorithm

Definition at line 287 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                         {
       selected_algo__ = AlgoType::GREEDY_HILL_CLIMBING;
     }

Here is the call graph for this function:

◆ useK2() [1/2]

INLINE void gum::learning::genericBNLearner::useK2 ( const Sequence< NodeId > & order )

inherited

indicate that we wish to use K2

Definition at line 275 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                        {
       selected_algo__ = AlgoType::K2;
       K2__.setOrder(order);
     }

Here is the call graph for this function:

◆ useK2() [2/2]

INLINE void gum::learning::genericBNLearner::useK2 ( const std::vector< NodeId > & order )

inherited

indicate that we wish to use K2

Definition at line 281 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                         {
       selected_algo__ = AlgoType::K2;
       K2__.setOrder(order);
     }

Here is the call graph for this function:

◆ useLocalSearchWithTabuList()

INLINE void gum::learning::genericBNLearner::useLocalSearchWithTabuList	(	Size	tabu_size = `100`,
		Size	nb_decrease = `2`
	)

inherited

indicate that we wish to use a local search with tabu list

Parameters

tabu_size	indicate the size of the tabu list
nb_decrease	indicate the max number of changes decreasing the score consecutively that we allow to apply

Definition at line 292 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                                                {
       selected_algo__ = AlgoType::LOCAL_SEARCH_WITH_TABU_LIST;
       constraint_TabuList__.setTabuListSize(tabu_size);
       local_search_with_tabu_list__.setMaxNbDecreasingChanges(nb_decrease);
     }

Here is the call graph for this function:

◆ useMDL()

INLINE void gum::learning::genericBNLearner::useMDL ( )

inherited

indicate that we wish to use the MDL correction for 3off2

Exceptions

OperationNotAllowed when 3off2 is not the selected algorithm

Definition at line 245 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                          {
       if (selected_algo__ != AlgoType::MIIC_THREE_OFF_TWO) {
         GUM_ERROR(OperationNotAllowed,
                   "You must use the 3off2 algorithm before selecting "
                      << "the MDL score");
       }
       kmode_3off2__ = CorrectedMutualInformation<>::KModeTypes::MDL;
     }

Here is the call graph for this function:

◆ useMIIC()

INLINE void gum::learning::genericBNLearner::useMIIC ( )

inherited

indicate that we wish to use MIIC

Definition at line 229 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                           {
       selected_algo__ = AlgoType::MIIC_THREE_OFF_TWO;
       miic_3off2__.setMiicBehaviour();
     }

Here is the call graph for this function:

◆ useNML()

INLINE void gum::learning::genericBNLearner::useNML ( )

inherited

indicate that we wish to use the NML correction for 3off2

Exceptions

OperationNotAllowed when 3off2 is not the selected algorithm

Definition at line 235 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                          {
       if (selected_algo__ != AlgoType::MIIC_THREE_OFF_TWO) {
         GUM_ERROR(OperationNotAllowed,
                   "You must use the 3off2 algorithm before selecting "
                      << "the NML score");
       }
       kmode_3off2__ = CorrectedMutualInformation<>::KModeTypes::NML;
     }

Here is the call graph for this function:

◆ useNoApriori()

INLINE void gum::learning::genericBNLearner::useNoApriori ( )

inherited

use no apriori

Definition at line 460 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                {
       apriori_type__ = AprioriType::NO_APRIORI;
       checkScoreAprioriCompatibility();
     }

Here is the call graph for this function:

◆ useNoCorr()

INLINE void gum::learning::genericBNLearner::useNoCorr ( )

inherited

indicate that we wish to use the NoCorr correction for 3off2

Exceptions

OperationNotAllowed when 3off2 is not the selected algorithm

Definition at line 255 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                             {
       if (selected_algo__ != AlgoType::MIIC_THREE_OFF_TWO) {
         GUM_ERROR(OperationNotAllowed,
                   "You must use the 3off2 algorithm before selecting "
                      << "the NoCorr score");
       }
       kmode_3off2__ = CorrectedMutualInformation<>::KModeTypes::NoCorr;
     }

Here is the call graph for this function:

◆ useScoreAIC()

INLINE void gum::learning::genericBNLearner::useScoreAIC ( )

inherited

indicate that we wish to use an AIC score

Definition at line 182 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                               {
       score_type__ = ScoreType::AIC;
       checkScoreAprioriCompatibility();
     }

Here is the call graph for this function:

◆ useScoreBD()

INLINE void gum::learning::genericBNLearner::useScoreBD ( )

inherited

indicate that we wish to use a BD score

Definition at line 188 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                              {
       score_type__ = ScoreType::BD;
       checkScoreAprioriCompatibility();
     }

Here is the call graph for this function:

◆ useScoreBDeu()

INLINE void gum::learning::genericBNLearner::useScoreBDeu ( )

inherited

indicate that we wish to use a BDeu score

Definition at line 194 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                {
       score_type__ = ScoreType::BDeu;
       checkScoreAprioriCompatibility();
     }

Here is the call graph for this function:

◆ useScoreBIC()

INLINE void gum::learning::genericBNLearner::useScoreBIC ( )

inherited

indicate that we wish to use a BIC score

Definition at line 200 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                               {
       score_type__ = ScoreType::BIC;
       checkScoreAprioriCompatibility();
     }

Here is the call graph for this function:

◆ useScoreK2()

INLINE void gum::learning::genericBNLearner::useScoreK2 ( )

inherited

indicate that we wish to use a K2 score

Definition at line 206 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                              {
       score_type__ = ScoreType::K2;
       checkScoreAprioriCompatibility();
     }

Here is the call graph for this function:

◆ useScoreLog2Likelihood()

INLINE void gum::learning::genericBNLearner::useScoreLog2Likelihood ( )

inherited

indicate that we wish to use a Log2Likelihood score

Definition at line 212 of file genericBNLearner_inl.h.

References gum::learning::genericBNLearner::Database::Database().

                                                          {
       score_type__ = ScoreType::LOG2LIKELIHOOD;
       checkScoreAprioriCompatibility();
     }

Here is the call graph for this function:

◆ verbosity()

bool gum::learning::genericBNLearner::verbosity ( ) const

inlinevirtualinherited

verbosity

Implements gum::IApproximationSchemeConfiguration.

Definition at line 1116 of file genericBNLearner.h.

                              {
         if (current_algorithm__ != nullptr)
           return current_algorithm__->verbosity();
         else
           GUM_ERROR(FatalError, "No chosen algorithm for learning");
       };

Member Data Documentation

◆ apriori__

Apriori* gum::learning::genericBNLearner::apriori__ {nullptr}

protectedinherited

the apriori used

Definition at line 773 of file genericBNLearner.h.

◆ apriori_database__

Database* gum::learning::genericBNLearner::apriori_database__ {nullptr}

protectedinherited

the database used by the Dirichlet a priori

Definition at line 827 of file genericBNLearner.h.

◆ apriori_dbname__

std::string gum::learning::genericBNLearner::apriori_dbname__

protectedinherited

the filename for the Dirichlet a priori, if any

Definition at line 830 of file genericBNLearner.h.

◆ apriori_type__

AprioriType gum::learning::genericBNLearner::apriori_type__ {AprioriType::NO_APRIORI}

protectedinherited

the a priori selected for the score and parameters

Definition at line 770 of file genericBNLearner.h.

◆ apriori_weight__

double gum::learning::genericBNLearner::apriori_weight__ {1.0f}

protectedinherited

the weight of the apriori

Definition at line 778 of file genericBNLearner.h.

◆ constraint_ForbiddenArcs__

StructuralConstraintForbiddenArcs gum::learning::genericBNLearner::constraint_ForbiddenArcs__

protectedinherited

the constraint on forbidden arcs

Definition at line 790 of file genericBNLearner.h.

◆ constraint_Indegree__

StructuralConstraintIndegree gum::learning::genericBNLearner::constraint_Indegree__

protectedinherited

the constraint for indegrees

Definition at line 784 of file genericBNLearner.h.

◆ constraint_MandatoryArcs__

StructuralConstraintMandatoryArcs gum::learning::genericBNLearner::constraint_MandatoryArcs__

protectedinherited

the constraint on forbidden arcs

Definition at line 796 of file genericBNLearner.h.

◆ constraint_PossibleEdges__

StructuralConstraintPossibleEdges gum::learning::genericBNLearner::constraint_PossibleEdges__

protectedinherited

the constraint on possible Edges

Definition at line 793 of file genericBNLearner.h.

◆ constraint_SliceOrder__

StructuralConstraintSliceOrder gum::learning::genericBNLearner::constraint_SliceOrder__

protectedinherited

the constraint for 2TBNs

Definition at line 781 of file genericBNLearner.h.

◆ constraint_TabuList__

StructuralConstraintTabuList gum::learning::genericBNLearner::constraint_TabuList__

protectedinherited

the constraint for tabu lists

Definition at line 787 of file genericBNLearner.h.

◆ current_algorithm__

const ApproximationScheme* gum::learning::genericBNLearner::current_algorithm__ {nullptr}

protectedinherited

Definition at line 836 of file genericBNLearner.h.

◆ Dag2BN__

DAG2BNLearner gum::learning::genericBNLearner::Dag2BN__

protectedinherited

the parametric EM

Definition at line 812 of file genericBNLearner.h.

◆ EMepsilon__

double gum::learning::genericBNLearner::EMepsilon__ {0.0}

protectedinherited

epsilon for EM. if espilon=0.0 : no EM

Definition at line 764 of file genericBNLearner.h.

◆ greedy_hill_climbing__

GreedyHillClimbing gum::learning::genericBNLearner::greedy_hill_climbing__

protectedinherited

the greedy hill climbing algorithm

Definition at line 815 of file genericBNLearner.h.

◆ initial_dag__

DAG gum::learning::genericBNLearner::initial_dag__

protectedinherited

an initial DAG given to learners

Definition at line 833 of file genericBNLearner.h.

◆ K2__

K2 gum::learning::genericBNLearner::K2__

protectedinherited

the K2 algorithm

Definition at line 802 of file genericBNLearner.h.

◆ kmode_3off2__

CorrectedMutualInformation ::KModeTypes gum::learning::genericBNLearner::kmode_3off2__

protectedinherited

Initial value:

{

CorrectedMutualInformation<>::KModeTypes::MDL}

the penalty used in 3off2

Definition at line 808 of file genericBNLearner.h.

◆ local_search_with_tabu_list__

LocalSearchWithTabuList gum::learning::genericBNLearner::local_search_with_tabu_list__

protectedinherited

the local search with tabu list algorithm

Definition at line 818 of file genericBNLearner.h.

◆ miic_3off2__

Miic gum::learning::genericBNLearner::miic_3off2__

protectedinherited

the 3off2 algorithm

Definition at line 805 of file genericBNLearner.h.

◆ mutual_info__

CorrectedMutualInformation* gum::learning::genericBNLearner::mutual_info__ {nullptr}

protectedinherited

the selected correction for 3off2 and miic

Definition at line 767 of file genericBNLearner.h.

◆ no_apriori__

AprioriNoApriori* gum::learning::genericBNLearner::no_apriori__ {nullptr}

protectedinherited

Definition at line 775 of file genericBNLearner.h.

◆ onProgress

Signaler3< Size, double, double > gum::IApproximationSchemeConfiguration::onProgress

inherited

Progression, error and time.

Definition at line 58 of file IApproximationSchemeConfiguration.h.

◆ onStop

Signaler1< std::string > gum::IApproximationSchemeConfiguration::onStop

inherited

Criteria messageApproximationScheme.

Definition at line 61 of file IApproximationSchemeConfiguration.h.

◆ param_estimator_type__

ParamEstimatorType gum::learning::genericBNLearner::param_estimator_type__ {ParamEstimatorType::ML}

protectedinherited

the type of the parameter estimator

Definition at line 761 of file genericBNLearner.h.

◆ ranges__

std::vector< std::pair< std::size_t, std::size_t > > gum::learning::genericBNLearner::ranges__

protectedinherited

the set of rows' ranges within the database in which learning is done

Definition at line 824 of file genericBNLearner.h.

◆ score__

Score* gum::learning::genericBNLearner::score__ {nullptr}

protectedinherited

the score used

Definition at line 758 of file genericBNLearner.h.

◆ score_database__

Database gum::learning::genericBNLearner::score_database__

protectedinherited

the database to be used by the scores and parameter estimators

Definition at line 821 of file genericBNLearner.h.

◆ score_type__

ScoreType gum::learning::genericBNLearner::score_type__ {ScoreType::BDeu}

protectedinherited

the score selected for learning

Definition at line 755 of file genericBNLearner.h.

◆ selected_algo__

AlgoType gum::learning::genericBNLearner::selected_algo__ {AlgoType::GREEDY_HILL_CLIMBING}

protectedinherited

the selected learning algorithm

Definition at line 799 of file genericBNLearner.h.

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

agrum/BN/learning/BNLearner.h

Public Attributes

Public Member Functions

Public Types

Protected Attributes

Protected Member Functions

Static Protected Member Functions

Detailed Description

template<typename GUM_SCALAR> class gum::learning::BNLearner< GUM_SCALAR >

Member Enumeration Documentation

◆ AlgoType

◆ ApproximationSchemeSTATE

◆ AprioriType

◆ ParamEstimatorType

◆ ScoreType

Constructor & Destructor Documentation

◆ BNLearner() [1/5]

◆ BNLearner() [2/5]

◆ BNLearner() [3/5]

◆ BNLearner() [4/5]

◆ BNLearner() [5/5]

◆ ~BNLearner()

Member Function Documentation

◆ addForbiddenArc() [1/3]

◆ addForbiddenArc() [2/3]

◆ addForbiddenArc() [3/3]

◆ addMandatoryArc() [1/3]

◆ addMandatoryArc() [2/3]

◆ addMandatoryArc() [3/3]

◆ addPossibleEdge() [1/3]

◆ addPossibleEdge() [2/3]

◆ addPossibleEdge() [3/3]

◆ checkFileName__()

◆ checkScoreAprioriCompatibility()

◆ chi2() [1/2]

◆ chi2() [2/2]

◆ clearDatabaseRanges()

◆ createApriori__()

◆ createCorrectedMutualInformation__()

◆ createParamEstimator__()

◆ createScore__()

◆ currentTime()

◆ database()

◆ databaseRanges()

◆ databaseWeight()

◆ disableEpsilon()

◆ disableMaxIter()

◆ disableMaxTime()

◆ disableMinEpsilonRate()

◆ distributeProgress()

◆ distributeStop()

◆ domainSize() [1/2]

◆ domainSize() [2/2]

◆ domainSizes()

◆ enableEpsilon()

◆ enableMaxIter()

◆ enableMaxTime()

◆ enableMinEpsilonRate()

◆ epsilon()

◆ eraseForbiddenArc() [1/3]

◆ eraseForbiddenArc() [2/3]

◆ eraseForbiddenArc() [3/3]

◆ eraseMandatoryArc() [1/3]

◆ eraseMandatoryArc() [2/3]

◆ eraseMandatoryArc() [3/3]

◆ erasePossibleEdge() [1/3]

◆ erasePossibleEdge() [2/3]

◆ erasePossibleEdge() [3/3]

◆ G2() [1/2]

◆ G2() [2/2]

◆ getAprioriType__()

◆ hasMissingValues()

◆ history()

◆ idFromName()

◆ isEnabledEpsilon()

◆ isEnabledMaxIter()

◆ isEnabledMaxTime()

◆ isEnabledMinEpsilonRate()

◆ labelsFromBN__()

◆ latentVariables()

◆ learnBN()

template<typename GUM_SCALAR>
class gum::learning::BNLearner< GUM_SCALAR >