<agrum/BN/generator/simpleBayesNetGenerator.h> More...

#include <simpleBayesNetGenerator.h>

Inheritance diagram for gum::SimpleBayesNetGenerator< GUM_SCALAR, ICPTGenerator >:

Collaboration diagram for gum::SimpleBayesNetGenerator< GUM_SCALAR, ICPTGenerator >:

Public Member Functions
Constructors / Destructor
	SimpleBayesNetGenerator (Size nbrNodes, Size maxArcs, Size maxModality=2)
	Constructor.
	~SimpleBayesNetGenerator () override
	Destructor.
BN generation methods
void	generateBN (BayesNet< GUM_SCALAR > &bayesNet) override
	function that generates a Bayesian networks.
BN generation methods
void	fromDAG (BayesNet< GUM_SCALAR > &bayesNet)
	Virtual function that Generates a Bayesian networks.
void	fillCPT (BayesNet< GUM_SCALAR > &bn) const
	function that insert random values in the CPT of each nodes according to the CPTGenerator.
Getters
Size	nbrNodes () const
	Return a constant reference to the number of nodes imposed on the IBayesNetGenerator.
Size	maxArcs () const
	Return a constant reference to the maximum number of arcs imposed on the IBayesNetGenerator.
Size	maxModality () const
	Return a constant reference to the maximum modality imposed on the IBayesNetGenerator.
Setters
void	setNbrNodes (Size nbrNodes)
	Modifies the value of the number of nodes imposed on the BayesGenerator.
void	setMaxArcs (Size maxArcs)
	Modifies the value of the number of nodes imposed on the BayesGenerator.
void	setMaxModality (Size maxModality)
	Modifies the value of the number of nodes imposed on the BayesGenerator.
CPT generation methods
void	generateCPT (const Idx &varId, const Tensor< GUM_SCALAR > &cpt) const override
	Generates a CPT using floats.

Protected Attributes
Size	nbrNodes_
Size	maxArcs_
Size	maxModality_
DAG	dag_

Detailed Description

template<typename GUM_SCALAR, template< typename > class ICPTGenerator = SimpleCPTGenerator>
class gum::SimpleBayesNetGenerator< GUM_SCALAR, ICPTGenerator >

Class for generating Bayesian networks.

This class randomly generates a Bayesian network given three parameters: the number of nodes and the max number of arcs and the number of maximum modality for each nodes.

Warning: Be Careful when entering the parameters, high Values may cause the density of the Bayesian network to be too high resulting in the failure of most of the inference Methods.

Definition at line 78 of file simpleBayesNetGenerator.h.

Constructor & Destructor Documentation

◆ SimpleBayesNetGenerator()

template<typename GUM_SCALAR, template< typename > class ICPTGenerator>

INLINE gum::SimpleBayesNetGenerator< GUM_SCALAR, ICPTGenerator >::SimpleBayesNetGenerator	(	Size	nbrNodes,
		Size	maxArcs,
		Size	maxModality = 2 )

Constructor.

Use by default the SimpleCPTGenerator for generating the BNs CPT.

Parameters

nbrNodes	The number of nodes imposed on the generator.
maxArcs	The number of maximum arcs imposed on the generator.
maxModality	Each DRV has from 2 to maxModality modalities

Definition at line 57 of file simpleBayesNetGenerator_tpl.h.

                        :
      IBayesNetGenerator< GUM_SCALAR, ICPTGenerator >(nbrNodes, maxArcs, maxModality) {
    GUM_CONSTRUCTOR(SimpleBayesNetGenerator);
  }

References gum::IBayesNetGenerator< GUM_SCALAR, SimpleCPTGenerator >::IBayesNetGenerator(), SimpleBayesNetGenerator(), gum::IBayesNetGenerator< GUM_SCALAR, SimpleCPTGenerator >::maxArcs(), gum::IBayesNetGenerator< GUM_SCALAR, SimpleCPTGenerator >::maxModality(), and gum::IBayesNetGenerator< GUM_SCALAR, SimpleCPTGenerator >::nbrNodes().

Referenced by SimpleBayesNetGenerator(), and ~SimpleBayesNetGenerator().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ ~SimpleBayesNetGenerator()

template<typename GUM_SCALAR, template< typename > class ICPTGenerator>

INLINE gum::SimpleBayesNetGenerator< GUM_SCALAR, ICPTGenerator >::~SimpleBayesNetGenerator ( )

override

Destructor.

Definition at line 79 of file simpleBayesNetGenerator_tpl.h.

                                                                                        {
    GUM_DESTRUCTOR(SimpleBayesNetGenerator);
  }

References SimpleBayesNetGenerator().

Here is the call graph for this function:

Member Function Documentation

◆ fillCPT()

void gum::IBayesNetGenerator< GUM_SCALAR, SimpleCPTGenerator >::fillCPT ( BayesNet< GUM_SCALAR > & bn ) const

inherited

function that insert random values in the CPT of each nodes according to the CPTGenerator.

Returns: null but modify inputed empty Bayesian network

Definition at line 130 of file IBayesNetGenerator_tpl.h.

                                                                                                {
    for (auto node: bn.nodes())
      this->generateCPT(bn.cpt(node).pos(bn.variable(node)), bn.cpt(node));
  }

References dag_, maxModality_, and gum::randomValue().

Referenced by gum::MCBayesNetGenerator< GUM_SCALAR, ICPTGenerator, ICPTDisturber >::generateBN(), and gum::SimpleBayesNetGenerator< GUM_SCALAR, ICPTGenerator >::generateBN().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ fromDAG()

INLINE void gum::IBayesNetGenerator< GUM_SCALAR, SimpleCPTGenerator >::fromDAG ( BayesNet< GUM_SCALAR > & bayesNet )

inherited

Virtual function that Generates a Bayesian networks.

Parameters

bayesNet Bayesian network to be completed after initialisation

Returns: null but modify inputed empty Bayesian network

Definition at line 122 of file IBayesNetGenerator_tpl.h.

                                                                                                 {
    bn.clear();
 
    const auto  width = (this->dag_.size() >= 100) ? 3 : 2;
    int         n     = 0;
    const auto& topo  = this->dag_.topologicalOrder();
    for (const auto node: topo) {
      std::stringstream strBuff;
      strBuff << "X" << std::setfill('0') << std::setw(width) << n++;
      bn.add(RangeVariable(strBuff.str(), "", 0, long(1 + randomValue(this->maxModality_ - 1))),
             node);
    }
    bn.beginTopologyTransformation();
    for (auto arc: this->dag_.arcs()) {
      bn.addArc(arc.tail(), arc.head());
    }
    bn.endTopologyTransformation();
  }

References maxArcs(), and maxArcs_.

Referenced by gum::MCBayesNetGenerator< GUM_SCALAR, ICPTGenerator, ICPTDisturber >::generateBN(), and gum::SimpleBayesNetGenerator< GUM_SCALAR, ICPTGenerator >::generateBN().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ generateBN()

template<typename GUM_SCALAR, template< typename > class ICPTGenerator>

void gum::SimpleBayesNetGenerator< GUM_SCALAR, ICPTGenerator >::generateBN ( BayesNet< GUM_SCALAR > & bayesNet )

overridevirtual

function that generates a Bayesian networks.

Parameters

bayesNet Bayesian network to be completed after initialisation

Returns: null but modify inputed Bayesian network

Implements gum::IBayesNetGenerator< GUM_SCALAR, SimpleCPTGenerator >.

Definition at line 89 of file simpleBayesNetGenerator_tpl.h.

                                        {
    this->dag_.clear();
    this->dag_.addNodes(this->nbrNodes_);
 
    // We add arcs
    float density = (float)(this->maxArcs_ * 2) / (float)(this->nbrNodes_ * (this->nbrNodes_ - 1));
    for (Size i = 0; i < this->nbrNodes_; ++i)
      for (Size j = i + 1; j < this->nbrNodes_; ++j)
        if (randomProba() < density) this->dag_.addArc(i, j);
 
    // Adding arcs until we reach the maxArcs_ number
    while (this->dag_.sizeArcs() < this->maxArcs_) {
      Size i = randomValue(this->nbrNodes_);
      Size j = randomValue(this->nbrNodes_);
      if (i != j) {
        if (i > j) std::swap(i, j);
 
        if (!this->dag_.existsArc(i, j)) this->dag_.addArc(i, j);
      }
    }
 
    this->fromDAG(bayesNet);
    this->fillCPT(bayesNet);
  }

References gum::IBayesNetGenerator< GUM_SCALAR, SimpleCPTGenerator >::dag_, gum::IBayesNetGenerator< GUM_SCALAR, SimpleCPTGenerator >::fillCPT(), gum::IBayesNetGenerator< GUM_SCALAR, SimpleCPTGenerator >::fromDAG(), gum::IBayesNetGenerator< GUM_SCALAR, SimpleCPTGenerator >::maxArcs_, gum::IBayesNetGenerator< GUM_SCALAR, SimpleCPTGenerator >::nbrNodes_, gum::randomProba(), and gum::randomValue().

Here is the call graph for this function:

◆ generateCPT()

template<typename GUM_SCALAR>

void gum::SimpleCPTGenerator< GUM_SCALAR >::generateCPT	(	const Idx &	varId,
		const Tensor< GUM_SCALAR > &	cpt ) const

overrideinherited

Generates a CPT using floats.

Parameters

varId	The variable id of the CPT owner.
cpt	A reference on the CPT to fill.

Definition at line 72 of file simpleCPTGenerator_tpl.h.

                                                                                            {
    cpt.random().normalizeAsCPT(varId);
  }

◆ maxArcs()

INLINE Size gum::IBayesNetGenerator< GUM_SCALAR, SimpleCPTGenerator >::maxArcs ( ) const

inherited

Return a constant reference to the maximum number of arcs imposed on the IBayesNetGenerator.

Definition at line 151 of file IBayesNetGenerator_tpl.h.

                                                                             {
    return maxArcs_;
  }

Referenced by gum::MCBayesNetGenerator< GUM_SCALAR, ICPTGenerator, ICPTDisturber >::MCBayesNetGenerator(), gum::SimpleBayesNetGenerator< GUM_SCALAR, ICPTGenerator >::SimpleBayesNetGenerator(), and fromDAG().

Here is the caller graph for this function:

◆ maxModality()

INLINE Size gum::IBayesNetGenerator< GUM_SCALAR, SimpleCPTGenerator >::maxModality ( ) const

inherited

Return a constant reference to the maximum modality imposed on the IBayesNetGenerator.

Definition at line 157 of file IBayesNetGenerator_tpl.h.

                                                                                 {
    return maxModality_;
  }

Referenced by gum::MCBayesNetGenerator< GUM_SCALAR, ICPTGenerator, ICPTDisturber >::MCBayesNetGenerator(), gum::SimpleBayesNetGenerator< GUM_SCALAR, ICPTGenerator >::SimpleBayesNetGenerator(), and ~IBayesNetGenerator().

Here is the caller graph for this function:

◆ nbrNodes()

INLINE Size gum::IBayesNetGenerator< GUM_SCALAR, SimpleCPTGenerator >::nbrNodes ( ) const

inherited

Return a constant reference to the number of nodes imposed on the IBayesNetGenerator.

Definition at line 145 of file IBayesNetGenerator_tpl.h.

                                                                              {
    return nbrNodes_;
  }

Referenced by gum::MCBayesNetGenerator< GUM_SCALAR, ICPTGenerator, ICPTDisturber >::MCBayesNetGenerator(), and gum::SimpleBayesNetGenerator< GUM_SCALAR, ICPTGenerator >::SimpleBayesNetGenerator().

Here is the caller graph for this function:

◆ setMaxArcs()

INLINE void gum::IBayesNetGenerator< GUM_SCALAR, SimpleCPTGenerator >::setMaxArcs ( Size maxArcs )

inherited

Modifies the value of the number of nodes imposed on the BayesGenerator.

Definition at line 172 of file IBayesNetGenerator_tpl.h.

                                                                                      {
    if (maxArcs < nbrNodes_ - 1 || maxArcs > (nbrNodes_ * (nbrNodes_ - 1)) / 2)
      GUM_ERROR(OperationNotAllowed, " maxArcs value not possible ")
 
    maxArcs_ = maxArcs;
  }

◆ setMaxModality()

INLINE void gum::IBayesNetGenerator< GUM_SCALAR, SimpleCPTGenerator >::setMaxModality ( Size maxModality )

inherited

Modifies the value of the number of nodes imposed on the BayesGenerator.

Definition at line 177 of file IBayesNetGenerator_tpl.h.

                                                                                              {
    if (maxModality < 2)
      GUM_ERROR(OperationNotAllowed, " maxModality must be at least equal to two ")
 
    maxModality_ = maxModality;
  }

◆ setNbrNodes()

INLINE void gum::IBayesNetGenerator< GUM_SCALAR, SimpleCPTGenerator >::setNbrNodes ( Size nbrNodes )

inherited

Modifies the value of the number of nodes imposed on the BayesGenerator.

Definition at line 167 of file IBayesNetGenerator_tpl.h.

                                                                                        {
    if ((maxArcs_ < nbrNodes - 1) || (maxArcs_ > (nbrNodes * (nbrNodes - 1)) / 2))
      GUM_ERROR(OperationNotAllowed, " nbrNodes value not possible ")
 
    nbrNodes_ = nbrNodes;
  }