gum::learning Namespace Reference

include the inlined functions if necessary More...

Namespaces
namespace	DBTranslators

Classes
class	CSVParser
	Class for fast parsing of CSV file (never more than one line in application memory). More...
class	DatabaseTable
	The class representing a tabular database as used by learning tasks. More...
class	DBCell
	The class representing the original values of the cells of databases. More...
class	DBHandler
	The base class for all database handlers. More...
class	DBInitializerFromCSV
	The class for initializing DatabaseTable and RawDatabaseTable instances from CSV files. More...
class	DBInitializerFromSQL
	The class for initializing DatabaseTable and RawDatabaseTable instances from SQL databases. More...
class	DBRow
	The class for storing a record in a database. More...
class	DBRowGenerator
	The base class for all DBRow generators. More...
class	DBRowGenerator4CompleteRows
	A DBRowGenerator class that returns the rows that are complete (fully observed) w.r.t. More...
class	DBRowGeneratorEM
	A DBRowGenerator class that returns incomplete rows as EM would do. More...
class	DBRowGeneratorIdentity
	A DBRowGenerator class that returns exactly the rows it gets in input. More...
class	DBRowGeneratorParser
	the class used to read a row in the database and to transform it into a set of DBRow instances that can be used for learning. More...
class	DBRowGeneratorSet
	The class used to pack sets of generators. More...
class	DBRowGeneratorWithBN
	Base class for DBRowGenerator classes that use a BN for computing their outputs. More...
union	DBTranslatedValue
	The union class for storing the translated values in learning databases. More...
class	DBTranslator
	The base class for all the tabular database cell translators. More...
class	DBTranslator4ContinuousVariable
	The databases' cell translators for continuous variables. More...
class	DBTranslator4DiscretizedVariable
	The databases' cell translators for discretized variables. More...
class	DBTranslator4IntegerVariable
	The databases' cell translators for integer variables. More...
class	DBTranslator4LabelizedVariable
	The databases' cell translators for labelized variables. More...
class	DBTranslator4NumericalDiscreteVariable
	The databases' cell translators for numerical Discrete variables. More...
class	DBTranslator4RangeVariable
	The databases' cell translators for range variables. More...
class	DBTranslatorSet
	the class for packing together the translators used to preprocess the datasets More...
struct	IDatabaseTableInsert4DBCell
struct	IDatabaseTableInsert4DBCell< true >
struct	IDatabaseTableInsert4DBCell< false >
class	IDatabaseTable
	The common class for the tabular database tables. More...
class	IDBInitializer
	The base class for initializing DatabaseTable and RawDatabaseTable instances from CSV files or SQL databases. More...
class	NanodbcParser
	Class for parsing SQL results using Nanodbc. More...
class	RawDatabaseTable
	The table containing the raw/original data of a database. More...
class	IdCondSetIterator
	The iterators for IdSets. More...
class	IdCondSet
	A class for storing a pair of sets of NodeIds, the second one corresponding to a conditional set. More...
class	IndependenceTest
	The base class for all the independence tests used for learning. More...
class	IndepTestChi2
	the class for computing Chi2 independence test scores More...
class	IndepTestG2
	the class for computing G2 independence test scores More...
class	KNML
	the class for computing the NML penalty used by MIIC More...
class	PseudoCount
	The class for giving access to pseudo count : count in the database + prior. More...
class	RecordCounter
	The class that computes counting of observations from the database. More...
class	ScoringCache
	a cache for caching scores and independence tests results More...
class	BNDatabaseGenerator
class	BNLearner
	A pack of learning algorithms that can easily be used. More...
class	BNLearnerListener
	A class that redirects gum_signal from algorithms to the listeners of BNLearn. More...
class	IBNLearner
	A pack of learning algorithms that can easily be used. More...
class	StructuralConstraintEmpty
	the base class for all structural constraints More...
class	StructuralConstraintDAG
	The base class for structural constraints imposed by DAGs. More...
class	StructuralConstraintDiGraph
	The base class for structural constraints used by learning algorithms that learn a directed graph structure. More...
class	StructuralConstraintForbiddenArcs
	the structural constraint for forbidding the creation of some arcs during structure learning More...
class	StructuralConstraintIndegree
	the class for structural constraints limiting the number of parents of nodes in a directed graph More...
class	StructuralConstraintMandatoryArcs
	the structural constraint indicating that some arcs shall never be removed or reversed More...
class	StructuralConstraintNoChildrenNodes
	the structural constraint for forbidding children for some nodes More...
class	StructuralConstraintNoParentNodes
	the structural constraint for forbidding parents for some nodes More...
class	StructuralConstraintPossibleEdges
	the structural constraint for forbidding the creation of some arcs except those defined in the class during structure learning More...
class	StructuralConstraintSetStatic
	the "meta-programming" class for storing structural constraints More...
class	StructuralConstraintSliceOrder
	the structural constraint imposing a partial order over nodes More...
class	StructuralConstraintTabuList
	The class imposing a N-sized tabu list as a structural constraints for learning algorithms. More...
class	StructuralConstraintUndiGraph
	The base class for structural constraints used by learning algorithms that learn an undirected graph structure. More...
class	CorrectedMutualInformation
	The class computing n times the corrected mutual information, as used in the MIIC algorithm. More...
class	GreedyHillClimbing
	The greedy hill climbing learning algorithm (for directed graphs). More...
class	K2
	The K2 algorithm. More...
class	LocalSearchWithTabuList
	The local search with tabu list learning algorithm (for directed graphs). More...
class	GreaterPairOn2nd
class	GreaterAbsPairOn2nd
class	GreaterTupleOnLast
class	Miic
	The Miic learning algorithm. More...
class	DAG2BNLearner
	A class that, given a structure and a parameter estimator returns a full Bayes net. More...
class	EMApproximationScheme
	A class for parameterizing EM's parameter learning approximations. More...
class	ParamEstimator
	The base class for estimating parameters of CPTs. More...
class	ParamEstimatorML
	The class for estimating parameters of CPTs using Maximum Likelihood. More...
class	BDeuPrior
	the internal prior for the BDeu score (N' / (r_i * q_i) More...
class	DirichletPriorFromBN
	A dirichlet priori: computes its N'_ijk from a database. More...
class	DirichletPriorFromDatabase
	A dirichlet priori: computes its N'_ijk from a database. More...
class	K2Prior
	the internal prior for the K2 score = Laplace Prior More...
class	NoPrior
	the no a priorclass: corresponds to 0 weight-sample More...
class	Prior
	the base class for all a priori More...
class	SmoothingPrior
	the smooth a priori: adds a weight w to all the counts More...
class	Score
	The base class for all the scores used for learning (BIC, BDeu, etc). More...
class	ScoreAIC
	the class for computing AIC scores More...
class	ScoreBD
	the class for computing Bayesian Dirichlet (BD) log2 scores More...
class	ScoreBDeu
	the class for computing BDeu scores More...
class	ScoreBIC
	the class for computing BIC scores More...
class	ScorefNML
	the class for computing fNML scores More...
class	ScoreK2
	the class for computing K2 scores (actually their log2 value) More...
class	ScoreLog2Likelihood
	the class for computing Log2-likelihood scores More...
class	SimpleMiic
	The miic learning algorithm. More...
class	GraphChange
class	ArcAddition
	The class for notifying learning algorithms of new arc additions. More...
class	ArcDeletion
	The class for notifying learning algorithms of arc removals. More...
class	ArcReversal
	The class for notifying learning algorithms of arc reversals. More...
class	EdgeAddition
	The class for notifying learning algorithms of new edge additions. More...
class	EdgeDeletion
	The class for notifying learning algorithms of edge removals. More...
class	GraphChangesGenerator4DiGraph
	The basic class for computing the next graph changes possible in a structure learning algorithm. More...
struct	_GraphChangesGenerator4K2_
class	GraphChangesGenerator4K2
	The basic class for computing the next graph changes possible in a structure learning algorithm. More...
class	GraphChangesGenerator4UndiGraph
	The basic class for computing the next graph changes possible in an undirected structure learning algorithm. More...
class	GraphChangesGeneratorOnSubDiGraph
	The basic class for computing the next graph changes possible in a structure learning algorithm. More...
class	GraphChangesSelector4DiGraph
	The mecanism to compute the next available graph changes for directed structure learning search algorithms. More...
class	IGraphChangesGenerator4DiGraph
class	IGraphChangesGenerator4UndiGraph

Typedefs
using	CondThreePoints = std::tuple< NodeId, NodeId, NodeId, std::vector< NodeId > >
using	CondRanking = std::pair< CondThreePoints*, double >
using	ThreePoints = std::tuple< NodeId, NodeId, NodeId >
using	Ranking = std::pair< ThreePoints*, double >
using	ProbabilisticRanking = std::tuple< ThreePoints*, double, double, double >
using	ScoreMDL = ScoreBIC

Enumerations
enum class	DBRowGeneratorGoal : char { DBRowGeneratorGoal::ONLY_REMOVE_MISSING_VALUES , DBRowGeneratorGoal::OTHER_THINGS_THAN_REMOVE_MISSING_VALUES }
	the type of things that a DBRowGenerator is designed for More...
enum class	DBTranslatedValueType : char { DBTranslatedValueType::DISCRETE , DBTranslatedValueType::CONTINUOUS }
	The nature of the elements handled by translators (discrete, continuous). More...
enum class	PriorType { DirichletPriorType , SmoothingPriorType , NoPriorType , BDeuPriorType }
enum	GraphChangeType {   ARC_ADDITION , ARC_DELETION , ARC_REVERSAL , EDGE_ADDITION ,   EDGE_DELETION }
	the type of modification that can be applied to the graph More...

Functions
std::ostream &	operator<< (std::ostream &stream, const IdCondSet &idset)
	the display operator
template<typename GUM_SCALAR>
std::ostream &	operator<< (std::ostream &output, const BNLearner< GUM_SCALAR > &learner)
	Prints BNLearner's current features.
DatabaseTable	readFile (const std::string &filename)
constexpr const char *	priorTypeToString (PriorType e) noexcept
std::ostream &	operator<< (std::ostream &stream, const GraphChange &change)
	a << operator for GraphChanges
std::ostream &	operator<< (std::ostream &stream, const ArcAddition &change)
	a << operator for ArcAddition
std::ostream &	operator<< (std::ostream &stream, const ArcDeletion &change)
	a << operator for ArcDeletion
std::ostream &	operator<< (std::ostream &stream, const ArcReversal &change)
	a << operator for ArcReversal
std::ostream &	operator<< (std::ostream &stream, const EdgeAddition &change)
	a << operator for EdgeAddition
std::ostream &	operator<< (std::ostream &stream, const EdgeDeletion &change)
	a << operator for EdgeDeletion

Detailed Description

include the inlined functions if necessary

Typedef Documentation

CondRanking

using gum::learning::CondRanking = std::pair< CondThreePoints*, double >

Definition at line 87 of file Miic.h.

CondThreePoints

using gum::learning::CondThreePoints = std::tuple< NodeId, NodeId, NodeId, std::vector< NodeId > >

Definition at line 86 of file Miic.h.

ProbabilisticRanking

using gum::learning::ProbabilisticRanking = std::tuple< ThreePoints*, double, double, double >

Definition at line 92 of file Miic.h.

Ranking

using gum::learning::Ranking = std::pair< ThreePoints*, double >

Definition at line 90 of file Miic.h.

ScoreMDL

using gum::learning::ScoreMDL = ScoreBIC

Definition at line 67 of file scoreMDL.h.

ThreePoints

using gum::learning::ThreePoints = std::tuple< NodeId, NodeId, NodeId >

Definition at line 89 of file Miic.h.

Enumeration Type Documentation

GraphChangeType

enum gum::learning::GraphChangeType

the type of modification that can be applied to the graph

Enumerator
ARC_ADDITION
ARC_DELETION
ARC_REVERSAL
EDGE_ADDITION
EDGE_DELETION

Definition at line 66 of file graphChange.h.

{ ARC_ADDITION, ARC_DELETION, ARC_REVERSAL, EDGE_ADDITION, EDGE_DELETION };

PriorType

enum class gum::learning::PriorType

strong

Enumerator
DirichletPriorType
SmoothingPriorType
NoPriorType
BDeuPriorType

Definition at line 61 of file prior.h.

                         {
      DirichletPriorType,
      SmoothingPriorType,
      NoPriorType,
      BDeuPriorType,
    };

Function Documentation

operator<<() [1/8]

template<typename GUM_SCALAR>

std::ostream & gum::learning::operator<<	(	std::ostream &	output,
		const BNLearner< GUM_SCALAR > &	learner )

Prints BNLearner's current features.

operator<<() [2/8]

std::ostream & gum::learning::operator<<	(	std::ostream &	stream,
		const ArcAddition &	change )

a << operator for ArcAddition

Definition at line 130 of file graphChange.cpp.

                                                                          {
      return stream << change.toString();
    }

References gum::learning::ArcAddition::toString().

Here is the call graph for this function:

operator<<() [3/8]

std::ostream & gum::learning::operator<<	(	std::ostream &	stream,
		const ArcDeletion &	change )

a << operator for ArcDeletion

Definition at line 135 of file graphChange.cpp.

                                                                          {
      return stream << change.toString();
    }

References gum::learning::ArcDeletion::toString().

Here is the call graph for this function:

operator<<() [4/8]

std::ostream & gum::learning::operator<<	(	std::ostream &	stream,
		const ArcReversal &	change )

a << operator for ArcReversal

Definition at line 140 of file graphChange.cpp.

                                                                          {
      return stream << change.toString();
    }

References gum::learning::ArcReversal::toString().

Here is the call graph for this function:

operator<<() [5/8]

std::ostream & gum::learning::operator<<	(	std::ostream &	stream,
		const EdgeAddition &	change )

a << operator for EdgeAddition

Definition at line 145 of file graphChange.cpp.

                                                                           {
      return stream << change.toString();
    }

References gum::learning::EdgeAddition::toString().

Here is the call graph for this function:

operator<<() [6/8]

std::ostream & gum::learning::operator<<	(	std::ostream &	stream,
		const EdgeDeletion &	change )

a << operator for EdgeDeletion

Definition at line 150 of file graphChange.cpp.

                                                                           {
      return stream << change.toString();
    }

References gum::learning::EdgeDeletion::toString().

Here is the call graph for this function:

operator<<() [7/8]

std::ostream & gum::learning::operator<<	(	std::ostream &	stream,
		const GraphChange &	change )

a << operator for GraphChanges

Definition at line 125 of file graphChange.cpp.

                                                                          {
      return stream << change.toString();
    }

References gum::learning::GraphChange::toString().

Here is the call graph for this function:

operator<<() [8/8]

std::ostream & gum::learning::operator<<	(	std::ostream &	stream,
		const IdCondSet &	idset )

the display operator

priorTypeToString()

const char * gum::learning::priorTypeToString ( PriorType e )

constexprnoexcept

Definition at line 68 of file prior.h.

                                                                  {
      switch (e) {
        case PriorType::NoPriorType : return "No prior";
        case PriorType::DirichletPriorType : return "Dirichlet prior";
        case PriorType::SmoothingPriorType : return "TriSmoothing prior";
        case PriorType::BDeuPriorType : return "BDeu prior";
      }
      return "Error in prior";
    }

References BDeuPriorType, DirichletPriorType, NoPriorType, and SmoothingPriorType.

readFile()

DatabaseTable gum::learning::readFile

(

const std::string &

filename

)

Definition at line 389 of file IBNLearner.cpp.

                                                    {
    // get the extension of the file
    if (auto filename_size = Size(filename.size()); filename_size < 4) {
      GUM_ERROR(FormatNotFound,
                "IBNLearner could not determine the "
                "file type of the database '"
                    << filename << "'")
    }
 
    std::string extension = filename.substr(filename.size() - 4);
    std::transform(extension.begin(), extension.end(), extension.begin(), ::tolower);
 
    if (extension != ".csv") {
      GUM_ERROR(OperationNotAllowed,
                "IBNLearner does not support yet this type ('" << extension
                                                               << "')"
                                                                  "of database file")
    }
 
    DBInitializerFromCSV initializer(filename);
 
    const auto&       var_names = initializer.variableNames();
    const std::size_t nb_vars   = var_names.size();
 
    DBTranslatorSet                translator_set;
    DBTranslator4LabelizedVariable translator;
    for (std::size_t i = 0; i < nb_vars; ++i) {
      translator_set.insertTranslator(translator, i);
    }
 
    DatabaseTable database(translator_set);
    database.setVariableNames(initializer.variableNames());
    initializer.fillDatabase(database);
 
    return database;
  }

References gum::learning::IDBInitializer::fillDatabase(), GUM_ERROR, gum::learning::DBTranslatorSet::insertTranslator(), gum::learning::DatabaseTable::setVariableNames(), and gum::learning::IDBInitializer::variableNames().

Here is the call graph for this function: