<agrum/base/graphicalModels/graphicalModel.h> More...

#include <graphicalModelInference.h>

Inheritance diagram for gum::GraphicalModelInference< GUM_SCALAR >:

Collaboration diagram for gum::GraphicalModelInference< GUM_SCALAR >:

Public Types
enum class	StateOfInference { OutdatedStructure , OutdatedTensors , ReadyForInference , Done }
	current state of the inference More...

Public Member Functions
Constructors / Destructors
	GraphicalModelInference (const GraphicalModel *model)
	default constructor
	GraphicalModelInference ()
	default constructor with a null model (useful for virtual inheritance)
virtual	~GraphicalModelInference ()
	destructor
Accessors / Modifiers
virtual const GraphicalModel &	model () const final
	Returns a constant reference over the IBayesNet referenced by this class.
virtual const NodeProperty< Size > &	domainSizes () const final
	get the domain sizes of the random variables of the model
virtual bool	isInferenceReady () const noexcept final
	returns whether the inference object is in a ready state
virtual bool	isInferenceOutdatedStructure () const noexcept final
	returns whether the inference object is in a OutdatedStructure state
virtual bool	isInferenceOutdatedTensors () const noexcept final
	returns whether the inference object is in a OutdatedTensor state
virtual bool	isInferenceDone () const noexcept final
	returns whether the inference object is in a InferenceDone state
virtual void	prepareInference () final
	prepare the internal inference structures for the next inference
virtual void	makeInference () final
	perform the heavy computations needed to compute the targets' posteriors
virtual void	clear ()
	clears all the data structures allocated for the last inference
virtual StateOfInference	state () const noexcept final
	returns the state of the inference engine
Evidence
virtual void	addEvidence (NodeId id, const Idx val) final
	adds a new hard evidence on node id
virtual void	addEvidence (const std::string &nodeName, const Idx val) final
	adds a new hard evidence on node named nodeName
virtual void	addEvidence (NodeId id, const std::string &label) final
	adds a new hard evidence on node id
virtual void	addEvidence (const std::string &nodeName, const std::string &label) final
	adds a new hard evidence on node named nodeName
virtual void	addEvidence (NodeId id, const std::vector< GUM_SCALAR > &vals) final
	adds a new evidence on node id (might be soft or hard)
virtual void	addEvidence (const std::string &nodeName, const std::vector< GUM_SCALAR > &vals) final
	adds a new evidence on node named nodeName (might be soft or hard)
virtual void	addEvidence (const Tensor< GUM_SCALAR > &pot) final
	adds a new evidence on node id (might be soft or hard)
virtual void	addEvidence (Tensor< GUM_SCALAR > &&pot) final
	adds a new evidence on node id (might be soft or hard)
virtual void	addSetOfEvidence (const Set< const Tensor< GUM_SCALAR > * > &potset) final
	adds a new set of evidence
virtual void	addListOfEvidence (const List< const Tensor< GUM_SCALAR > * > &potlist) final
	adds a new list of evidence
virtual void	chgEvidence (NodeId id, const Idx val) final
	change the value of an already existing hard evidence
virtual void	chgEvidence (const std::string &nodeName, const Idx val) final
	change the value of an already existing hard evidence
virtual void	chgEvidence (NodeId id, const std::string &label) final
	change the value of an already existing hard evidence
virtual void	chgEvidence (const std::string &nodeName, const std::string &label) final
	change the value of an already existing hard evidence
virtual void	chgEvidence (NodeId id, const std::vector< GUM_SCALAR > &vals) final
	change the value of an already existing evidence (might be soft or hard)
virtual void	chgEvidence (const std::string &nodeName, const std::vector< GUM_SCALAR > &vals) final
	change the value of an already existing evidence (might be soft or hard)
virtual void	chgEvidence (const Tensor< GUM_SCALAR > &pot) final
	change the value of an already existing evidence (might be soft or hard)
virtual void	eraseAllEvidence () final
	removes all the evidence entered into the network
virtual void	eraseEvidence (NodeId id) final
	removed the evidence, if any, corresponding to node id
virtual void	eraseEvidence (const std::string &nodeName) final
	removed the evidence, if any, corresponding to node of name nodeName
virtual bool	hasEvidence () const final
	indicates whether some node(s) have received evidence
virtual bool	hasEvidence (NodeId id) const final
	indicates whether node id has received an evidence
virtual bool	hasHardEvidence (NodeId id) const final
	indicates whether node id has received a hard evidence
virtual bool	hasSoftEvidence (NodeId id) const final
	indicates whether node id has received a soft evidence
virtual bool	hasEvidence (const std::string &nodeName) const final
	indicates whether node id has received an evidence
virtual bool	hasHardEvidence (const std::string &nodeName) const final
	indicates whether node id has received a hard evidence
virtual bool	hasSoftEvidence (const std::string &nodeName) const final
	indicates whether node id has received a soft evidence
virtual Size	nbrEvidence () const final
	returns the number of evidence entered into the Bayesian network
virtual Size	nbrHardEvidence () const final
	returns the number of hard evidence entered into the Bayesian network
virtual Size	nbrSoftEvidence () const final
	returns the number of soft evidence entered into the Bayesian network
const NodeProperty< const Tensor< GUM_SCALAR > * > &	evidence () const
	returns the set of evidence
const NodeSet &	softEvidenceNodes () const
	returns the set of nodes with soft evidence
const NodeSet &	hardEvidenceNodes () const
	returns the set of nodes with hard evidence
const NodeProperty< Idx > &	hardEvidence () const
	indicate for each node with hard evidence which value it took

Protected Member Functions
virtual void	onStateChanged_ ()=0
	fired when the stage is changed
virtual void	onEvidenceAdded_ (const NodeId id, bool isHardEvidence)=0
	fired after a new evidence is inserted
virtual void	onEvidenceErased_ (const NodeId id, bool isHardEvidence)=0
	fired before an evidence is removed
virtual void	onAllEvidenceErased_ (bool contains_hard_evidence)=0
	fired before all the evidence are erased
virtual void	onEvidenceChanged_ (const NodeId id, bool hasChangedSoftHard)=0
	fired after an evidence is changed, in particular when its status (soft/hard) changes
virtual void	onModelChanged_ (const GraphicalModel *model)=0
	fired after a new Bayes net has been assigned to the engine
virtual void	updateOutdatedStructure_ ()=0
	prepares inference when the latter is in OutdatedStructure state
virtual void	updateOutdatedTensors_ ()=0
	prepares inference when the latter is in OutdatedTensors state
virtual void	makeInference_ ()=0
	called when the inference has to be performed effectively
void	setOutdatedStructureState_ ()
	put the inference into an outdated model structure state
void	setOutdatedTensorsState_ ()
	puts the inference into an OutdatedTensors state if it is not already in an OutdatedStructure state
virtual void	setState_ (const StateOfInference state) final
	set the state of the inference engine and call the notification onStateChanged_ when necessary (i.e. when the state has effectively changed).
void	setModel_ (const GraphicalModel *model)
void	setModelDuringConstruction_ (const GraphicalModel *model)
	assigns a model during the inference engine construction
bool	hasNoModel_ () const

Private Member Functions
Tensor< GUM_SCALAR >	_createHardEvidence_ (NodeId id, Idx val) const
	create the internal structure for a hard evidence
bool	_isHardEvidence_ (const Tensor< GUM_SCALAR > &pot, Idx &val) const
	checks whether a tensor corresponds to a hard evidence or not
void	_computeDomainSizes_ ()
	computes the domain sizes of the random variables

Private Attributes
StateOfInference	_state_ {StateOfInference::OutdatedStructure}
	the current state of the inference (outdated/ready/done)
const GraphicalModel *	_model_ {nullptr}
	the Bayes net on which we perform inferences
NodeProperty< Size >	_domain_sizes_
	the domain sizes of the random variables
NodeProperty< const Tensor< GUM_SCALAR > * >	_evidence_
	the set of evidence entered into the network
NodeProperty< Idx >	_hard_evidence_
	assign to each node with a hard evidence the index of its observed value
NodeSet	_soft_evidence_nodes_
	the set of nodes that received soft evidence
NodeSet	_hard_evidence_nodes_
	the set of nodes that received hard evidence

Detailed Description

template<typename GUM_SCALAR>
class gum::GraphicalModelInference< GUM_SCALAR >

A generic class for graphical model inference: handles evidence and the current state of the inference

The goal of the graphicalModelInference class is twofold: i) handling the common resources of graphical Model inference (model, soft/hard evidence); ii) propose a general high-level scheme for all the inference methods.

A specialized inference just has to specify how to prepare inference, how to make inference and how to get the posteriors for nodes and set of nodes. The scheme for every inference derived from graphicalModelInference will be the same:

1- ie=SpecificInference(model); // state <- OutdatedStructure 2- set targets and evidence in ie 3- ie.prepareInference(); // state <- Ready4Inference 4.a- change values of evidence in ie // state <- OutdatedTensors 4.b- change some hard evidence or targets // state <- OutdatedStructure 5- ie.makeInference(); // state <- Done 6- get posteriors 7- goto 2 or 4

graphicalModelInference can be in one of 4 different states:

OutdatedStructure: in this state, the inference is fully unprepared to be applied because some events changed the "logical" structure of the model: for instance a node received a hard evidence, which implies that its outgoing arcs can be removed from the model, hence involving a structural change in the model.
OutdatedTensors: in this state, the structure of the model remains unchanged, only some tensors stored in it have changed. Therefore, the inference probably just needs to invalidate some already computed tensors to be ready. Only a light amount of preparation is needed to be able to perform inference.
Ready4Inference: in this state, all the data structures are ready for inference. There just remains to perform the inference computations.
Done: the heavy computations of inference have been done. There might still remain a few light computations to perform to get the posterior tensors we need.

Definition at line 105 of file graphicalModelInference.h.

Member Enumeration Documentation

◆ StateOfInference

template<typename GUM_SCALAR>

enum class gum::GraphicalModelInference::StateOfInference

strong

current state of the inference

graphicalModelInference can be in one of 4 different states:

OutdatedStructure: in this state, the inference is fully unprepared to be applied because some events changed the "logical" structure of the model: for instance a node received a hard evidence, which implies that its outgoing arcs can be removed from the model, hence involving a structural change in the model.
OutdatedTensors: in this state, the structure of the model remains unchanged, only some tensors stored in it have changed. Therefore, the inference probably just needs to invalidate some already computed tensors to be ready. Only a light amount of preparation is needed to be able to perform inference.
Ready4Inference: in this state, all the data structures are ready for inference. There just remains to perform the inference computations.
Done: the heavy computations of inference have been done. There might still remain a few light computations to perform to get the posterior tensors we need.

Enumerator
OutdatedStructure
OutdatedTensors
ReadyForInference
Done

Definition at line 127 of file graphicalModelInference.h.

127{ OutdatedStructure, OutdatedTensors, ReadyForInference, Done };

Constructor & Destructor Documentation

◆ GraphicalModelInference() [1/2]

template<typename GUM_SCALAR>

gum::GraphicalModelInference< GUM_SCALAR >::GraphicalModelInference ( const GraphicalModel * model )

explicit

default constructor

Warning: note that, by aGrUM's rule, the model is not copied but only referenced by the inference algorithm.

Definition at line 56 of file graphicalModelInference_tpl.h.

                                                                                            :
      _model_(model) {
    _computeDomainSizes_();
 
    GUM_CONSTRUCTOR(GraphicalModelInference);
  }

References GraphicalModelInference(), _computeDomainSizes_(), _model_, and model().

Referenced by gum::BayesNetInference< GUM_SCALAR >::BayesNetInference(), GraphicalModelInference(), GraphicalModelInference(), gum::MRFInference< GUM_SCALAR >::MRFInference(), and ~GraphicalModelInference().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ GraphicalModelInference() [2/2]

template<typename GUM_SCALAR>

gum::GraphicalModelInference< GUM_SCALAR >::GraphicalModelInference ( )

default constructor with a null model (useful for virtual inheritance)

Warning: graphicalModelInference is virtually inherited by MarginalTargetedInference. As a result, the lowest descendant of graphicalModelInference will create the latter. To avoid requiring developers to add in the constructors of their inference algorithms a call to graphicalModelInference( model ), we added constructor graphicalModelInference(), which will be called automatically by the lowest descendant.

Definition at line 65 of file graphicalModelInference_tpl.h.

                                                                 {
    GUM_CONSTRUCTOR(GraphicalModelInference);
  }

References GraphicalModelInference().

Here is the call graph for this function:

◆ ~GraphicalModelInference()

template<typename GUM_SCALAR>

gum::GraphicalModelInference< GUM_SCALAR >::~GraphicalModelInference ( )

virtual

destructor

Definition at line 71 of file graphicalModelInference_tpl.h.

                                                                  {
    // clear all evidence.
    // Warning: Do not use method eraseAllEvidence () because it contains a call
    // to pure virtual method onAllEvidenceErased_ which belongs to an inherited
    // instance and, therefore, does not exist anymore when
    // ~GraphicalModelInference () is called
    for (const auto& pair: _evidence_) {
      if (pair.second != nullptr) { delete (pair.second); }
    }
 
    GUM_DESTRUCTOR(GraphicalModelInference);
  }

References GraphicalModelInference(), and _evidence_.

Here is the call graph for this function:

Member Function Documentation

◆ _computeDomainSizes_()

template<typename GUM_SCALAR>

void gum::GraphicalModelInference< GUM_SCALAR >::_computeDomainSizes_ ( )

private

computes the domain sizes of the random variables

Definition at line 162 of file graphicalModelInference_tpl.h.

                                                                   {
    _domain_sizes_.clear();
    if (!hasNoModel_()) {
      for (auto node: _model_->nodes()) {
        _domain_sizes_.insert(node, _model_->variable(node).domainSize());
      }
    }
  }

References _domain_sizes_, _model_, and hasNoModel_().

Referenced by GraphicalModelInference(), setModel_(), and setModelDuringConstruction_().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ _createHardEvidence_()

template<typename GUM_SCALAR>

Tensor< GUM_SCALAR > gum::GraphicalModelInference< GUM_SCALAR >::_createHardEvidence_	(	NodeId	id,
		Idx	val ) const

private

create the internal structure for a hard evidence

Definition at line 184 of file graphicalModelInference_tpl.h.

                                                                                                {
    // check that it is possible to create the evidence
    if (_model_ == nullptr)
      GUM_ERROR(NullElement,
                "No Bayes net has been assigned to the "
                "inference algorithm");
 
    if (!_model_->exists(id)) { GUM_ERROR(UndefinedElement, id << " is not a NodeId in the model") }
 
    if (_model_->variable(id).domainSize() <= val) {
      GUM_ERROR(InvalidArgument,
                "node " << _model_->variable(id) << " has fewer possible values than " << val);
    }
 
    return Tensor< GUM_SCALAR >::deterministicTensor(_model_->variable(id), val);
  }

References _model_, gum::Tensor< GUM_SCALAR >::deterministicTensor(), and GUM_ERROR.

Referenced by addEvidence(), and chgEvidence().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ _isHardEvidence_()

template<typename GUM_SCALAR>

bool gum::GraphicalModelInference< GUM_SCALAR >::_isHardEvidence_	(	const Tensor< GUM_SCALAR > &	pot,
		Idx &	val ) const

private

checks whether a tensor corresponds to a hard evidence or not

Definition at line 203 of file graphicalModelInference_tpl.h.

                                                                               {
    // checking if pot is determininstic
    bool          notZero = false;
    Instantiation I(pot);
 
    for (I.setFirst(); !I.end(); I.inc()) {
      if (pot[I] != 0.0) {
        if (notZero) {   // we already met a non-zero value
          return false;
        } else {
          val     = I.val(0);
          notZero = true;   // this is the first met non-zero value
        }
      }
    }
 
    if (!notZero) {   // we met no non-zero value
      GUM_ERROR(FatalError, "Evidence of impossibility (vector of 0s)")
    }
 
    return true;   // pot is deterministic
  }

References gum::Instantiation::end(), GUM_ERROR, gum::Instantiation::inc(), gum::Instantiation::setFirst(), and gum::Instantiation::val().

Referenced by addEvidence(), and chgEvidence().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ addEvidence() [1/8]

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::addEvidence	(	const std::string &	nodeName,
		const Idx	val )

finalvirtual

adds a new hard evidence on node named nodeName

Exceptions

UndefinedElement	if nodeName does not belong to the Bayesian network
InvalidArgument	if val is not a value for id
InvalidArgument	if nodeName already has an evidence

Definition at line 235 of file graphicalModelInference_tpl.h.

                                                                                         {
    addEvidence(this->model().idFromName(nodeName), val);
  }

References addEvidence(), and model().

Here is the call graph for this function:

◆ addEvidence() [2/8]

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::addEvidence	(	const std::string &	nodeName,
		const std::string &	label )

finalvirtual

adds a new hard evidence on node named nodeName

Exceptions

UndefinedElement	if nodeName does not belong to the Bayesian network
InvalidArgument	if val is not a value for id
InvalidArgument	if nodeName already has an evidence

Definition at line 249 of file graphicalModelInference_tpl.h.

                                                                                         {
    const NodeId id = this->model().idFromName(nodeName);
    addEvidence(id, this->model().variable(id)[label]);
  }

References addEvidence(), and model().

Here is the call graph for this function:

◆ addEvidence() [3/8]

template<typename GUM_SCALAR>

void gum::GraphicalModelInference< GUM_SCALAR >::addEvidence	(	const std::string &	nodeName,
		const std::vector< GUM_SCALAR > &	vals )

finalvirtual

adds a new evidence on node named nodeName (might be soft or hard)

Exceptions

UndefinedElement	if id does not belong to the Bayesian network
InvalidArgument	if nodeName already has an evidence
FatalError	if vals=[0,0,...,0]
InvalidArgument	if the size of vals is different from the domain size of node nodeName

Definition at line 281 of file graphicalModelInference_tpl.h.

                                                                                               {
    addEvidence(this->model().idFromName(nodeName), vals);
  }

References addEvidence(), and model().

Here is the call graph for this function:

◆ addEvidence() [4/8]

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::addEvidence ( const Tensor< GUM_SCALAR > & pot )

finalvirtual

adds a new evidence on node id (might be soft or hard)

Exceptions

UndefinedElement	if the tensor is defined over several nodes
UndefinedElement	if the node on which the tensor is defined does not belong to the Bayesian network
InvalidArgument	if the node of the tensor already has an evidence
FatalError	if pot=[0,0,...,0]

Definition at line 323 of file graphicalModelInference_tpl.h.

                                                                                                {
    Tensor< GUM_SCALAR > new_pot(pot);
    addEvidence(std::move(new_pot));
  }

References addEvidence().

Here is the call graph for this function:

◆ addEvidence() [5/8]

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::addEvidence	(	NodeId	id,
		const Idx	val )

finalvirtual

adds a new hard evidence on node id

Exceptions

UndefinedElement	if id does not belong to the Bayesian network
InvalidArgument	if val is not a value for id
InvalidArgument	if id already has an evidence

Definition at line 229 of file graphicalModelInference_tpl.h.

                                                                                         {
    addEvidence(_createHardEvidence_(id, val));
  }

References _createHardEvidence_(), and addEvidence().

Referenced by addEvidence(), addEvidence(), addEvidence(), addEvidence(), addEvidence(), addEvidence(), addEvidence(), addListOfEvidence(), addSetOfEvidence(), gum::MarginalTargetedInference< GUM_SCALAR >::evidenceImpact(), gum::MarginalTargetedMRFInference< GUM_SCALAR >::evidenceImpact(), gum::JointTargetedInference< GUM_SCALAR >::evidenceJointImpact(), gum::JointTargetedMRFInference< GUM_SCALAR >::evidenceJointImpact(), and gum::LoopySamplingInference< GUM_SCALAR, APPROX >::makeInference_().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ addEvidence() [6/8]

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::addEvidence	(	NodeId	id,
		const std::string &	label )

finalvirtual

adds a new hard evidence on node id

Exceptions

UndefinedElement	if id does not belong to the Bayesian network
InvalidArgument	if val is not a value for id
InvalidArgument	if id already has an evidence

Definition at line 242 of file graphicalModelInference_tpl.h.

                                                                                         {
    addEvidence(id, this->model().variable(id)[label]);
  }

References addEvidence(), and model().

Here is the call graph for this function:

◆ addEvidence() [7/8]

template<typename GUM_SCALAR>

void gum::GraphicalModelInference< GUM_SCALAR >::addEvidence	(	NodeId	id,
		const std::vector< GUM_SCALAR > &	vals )

finalvirtual

adds a new evidence on node id (might be soft or hard)

Exceptions

UndefinedElement	if id does not belong to the Bayesian network
InvalidArgument	if id already has an evidence
FatalError	if vals=[0,0,...,0]
InvalidArgument	if the size of vals is different from the domain size of node id

Definition at line 257 of file graphicalModelInference_tpl.h.

                                                                                               {
    // checks that the evidence is meaningful
    if (_model_ == nullptr)
      GUM_ERROR(NullElement,
                "No Bayes net has been assigned to the "
                "inference algorithm");
 
    if (!_model_->exists(id)) { GUM_ERROR(UndefinedElement, id << " is not a NodeId in the model") }
 
    if (_model_->variable(id).domainSize() != vals.size()) {
      GUM_ERROR(InvalidArgument,
                "node " << _model_->variable(id)
                        << " and its evidence vector have different sizes.");
    }
 
    Tensor< GUM_SCALAR > pot;
    pot.add(_model_->variable(id));
    pot.fillWith(vals);
    addEvidence(std::move(pot));
  }

References _model_, addEvidence(), and GUM_ERROR.

Here is the call graph for this function:

◆ addEvidence() [8/8]

template<typename GUM_SCALAR>

void gum::GraphicalModelInference< GUM_SCALAR >::addEvidence ( Tensor< GUM_SCALAR > && pot )

finalvirtual

adds a new evidence on node id (might be soft or hard)

Exceptions

UndefinedElement	if the tensor is defined over several nodes
UndefinedElement	if the node on which the tensor is defined does not belong to the Bayesian network
InvalidArgument	if the node of the tensor already has an evidence
FatalError	if pot=[0,0,...,0]

Definition at line 288 of file graphicalModelInference_tpl.h.

                                                                                    {
    // check if the tensor corresponds to an evidence
    if (pot.nbrDim() != 1) { GUM_ERROR(InvalidArgument, pot << " is not mono-dimensional.") }
    if (_model_ == nullptr)
      GUM_ERROR(NullElement,
                "No Bayes net has been assigned to the "
                "inference algorithm");
 
    NodeId id = _model_->nodeId(pot.variable(0));
 
    if (hasEvidence(id)) {
      GUM_ERROR(InvalidArgument,
                " node " << id << " already has an evidence. Please use chgEvidence().");
    }
 
    // check whether we have a hard evidence (and also check whether the
    // tensor only contains 0 (in this case, this will automatically raise
    // an exception) )
    Idx  val              = 0;
    bool is_hard_evidence = _isHardEvidence_(pot, val);
 
    // insert the evidence
    _evidence_.insert(id, new Tensor< GUM_SCALAR >(std::forward< Tensor< GUM_SCALAR > >(pot)));
    if (is_hard_evidence) {   // pot is deterministic
      _hard_evidence_.insert(id, val);
      _hard_evidence_nodes_.insert(id);
    } else {
      _soft_evidence_nodes_.insert(id);
    }
    setState_(StateOfInference::OutdatedStructure);
    onEvidenceAdded_(id, is_hard_evidence);
  }

References _evidence_, _hard_evidence_, _hard_evidence_nodes_, _isHardEvidence_(), _model_, _soft_evidence_nodes_, GUM_ERROR, hasEvidence(), onEvidenceAdded_(), OutdatedStructure, and setState_().

Here is the call graph for this function:

◆ addListOfEvidence()

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::addListOfEvidence ( const List< const Tensor< GUM_SCALAR > * > & potlist )

finalvirtual

adds a new list of evidence

Exceptions

UndefinedElement	if some tensor is defined over several nodes
UndefinedElement	if the node on which some tensor is defined does not belong to the Bayesian network
InvalidArgument	if the node of some tensor already has an evidence
FatalError	if pot=[0,0,...,0]

Definition at line 330 of file graphicalModelInference_tpl.h.

                                                          {
    for (const auto pot: potlist)
      addEvidence(*pot);
  }

References addEvidence().

Here is the call graph for this function:

◆ addSetOfEvidence()

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::addSetOfEvidence ( const Set< const Tensor< GUM_SCALAR > * > & potset )

finalvirtual

adds a new set of evidence

Exceptions

UndefinedElement	if some tensor is defined over several nodes
UndefinedElement	if the node on which some tensor is defined does not belong to the Bayesian network
InvalidArgument	if the node of some tensor already has an evidence
FatalError	if pot=[0,0,...,0]

Definition at line 338 of file graphicalModelInference_tpl.h.

                                                        {
    for (const auto pot: potset)
      addEvidence(*pot);
  }

References addEvidence().

Here is the call graph for this function:

◆ chgEvidence() [1/7]

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::chgEvidence	(	const std::string &	nodeName,
		const Idx	val )

finalvirtual

change the value of an already existing hard evidence

Exceptions

UndefinedElement	if nodeName does not belong to the Bayesian network
InvalidArgument	if val is not a value for id
InvalidArgument	if id does not already have an evidence

Definition at line 397 of file graphicalModelInference_tpl.h.

                                                                                         {
    chgEvidence(this->model().idFromName(nodeName), val);
  }

References chgEvidence(), and model().

Here is the call graph for this function:

◆ chgEvidence() [2/7]

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::chgEvidence	(	const std::string &	nodeName,
		const std::string &	label )

finalvirtual

change the value of an already existing hard evidence

Exceptions

UndefinedElement	if nodeName does not belong to the Bayesian network
InvalidArgument	if val is not a value for id
InvalidArgument	if id does not already have an evidence

Definition at line 411 of file graphicalModelInference_tpl.h.

                                                                                         {
    NodeId id = this->model().idFromName(nodeName);
    chgEvidence(id, this->model().variable(id)[label]);
  }

References chgEvidence(), and model().

Here is the call graph for this function:

◆ chgEvidence() [3/7]

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::chgEvidence	(	const std::string &	nodeName,
		const std::vector< GUM_SCALAR > &	vals )

finalvirtual

change the value of an already existing evidence (might be soft or hard)

Exceptions

UndefinedElement	if nodeName does not belong to the Bayesian network
InvalidArgument	if the node does not already have an evidence
FatalError	if vals=[0,0,...,0]
InvalidArgument	if the size of vals is different from the domain size of node id

Definition at line 445 of file graphicalModelInference_tpl.h.

                                                                                              {
    chgEvidence(this->model().idFromName(nodeName), vals);
  }

References chgEvidence(), and model().

Here is the call graph for this function:

◆ chgEvidence() [4/7]

template<typename GUM_SCALAR>

void gum::GraphicalModelInference< GUM_SCALAR >::chgEvidence ( const Tensor< GUM_SCALAR > & pot )

finalvirtual

change the value of an already existing evidence (might be soft or hard)

Exceptions

UndefinedElement	if the tensor is defined over several nodes
UndefinedElement	if the node on which the tensor is defined does not belong to the Bayesian network
InvalidArgument	if the node of the tensor does not already have an evidence
FatalError	if pot=[0,0,...,0]

Definition at line 452 of file graphicalModelInference_tpl.h.

                                                                                         {
    // check if the tensor corresponds to an evidence
    if (pot.nbrDim() != 1) {
      GUM_ERROR(InvalidArgument, pot << " is not a mono-dimensional tensor.")
    }
    if (_model_ == nullptr)
      GUM_ERROR(NullElement,
                "No Bayes net has been assigned to the "
                "inference algorithm");
 
    NodeId id = _model_->nodeId(pot.variable(0));
 
    if (!hasEvidence(id)) {
      GUM_ERROR(InvalidArgument, id << " has no evidence. Please use addEvidence().")
    }
 
    // check whether we have a hard evidence (and also check whether the
    // tensor only contains 0 (in this case, this will automatically raise
    // an exception) )
    Idx  val;
    bool is_hard_evidence = _isHardEvidence_(pot, val);
 
    // modify the evidence already stored
    const Tensor< GUM_SCALAR >* localPot = _evidence_[id];
    Instantiation               I(pot);
    for (I.setFirst(); !I.end(); I.inc()) {
      localPot->set(I, pot[I]);
    }
 
    // the inference state will be different
    // whether evidence change from Hard to Soft or not.
    bool hasChangedSoftHard = false;
 
    if (is_hard_evidence) {
      if (!hasHardEvidence(id)) {
        hasChangedSoftHard = true;
        _hard_evidence_.insert(id, val);
        _hard_evidence_nodes_.insert(id);
        _soft_evidence_nodes_.erase(id);
      } else {
        _hard_evidence_[id] = val;
      }
    } else {
      if (hasHardEvidence(id)) {   // evidence was hard
        _hard_evidence_.erase(id);
        _hard_evidence_nodes_.erase(id);
        _soft_evidence_nodes_.insert(id);
        hasChangedSoftHard = true;
      }
    }
 
    if (hasChangedSoftHard) {
      setState_(StateOfInference::OutdatedStructure);
    } else {
      if (!isInferenceOutdatedStructure()) { setState_(StateOfInference::OutdatedTensors); }
    }
 
    onEvidenceChanged_(id, hasChangedSoftHard);
  }

References _evidence_, _hard_evidence_, _hard_evidence_nodes_, _isHardEvidence_(), _model_, _soft_evidence_nodes_, gum::Instantiation::end(), GUM_ERROR, hasEvidence(), hasHardEvidence(), gum::Instantiation::inc(), isInferenceOutdatedStructure(), onEvidenceChanged_(), OutdatedStructure, OutdatedTensors, gum::Instantiation::setFirst(), and setState_().

Here is the call graph for this function:

◆ chgEvidence() [5/7]

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::chgEvidence	(	NodeId	id,
		const Idx	val )

finalvirtual

change the value of an already existing hard evidence

Exceptions

UndefinedElement	if id does not belong to the Bayesian network
InvalidArgument	if val is not a value for id
InvalidArgument	if id does not already have an evidence

Definition at line 391 of file graphicalModelInference_tpl.h.

                                                                                         {
    chgEvidence(_createHardEvidence_(id, val));
  }

References _createHardEvidence_(), and chgEvidence().

Referenced by chgEvidence(), chgEvidence(), chgEvidence(), chgEvidence(), chgEvidence(), chgEvidence(), gum::MarginalTargetedInference< GUM_SCALAR >::evidenceImpact(), gum::MarginalTargetedMRFInference< GUM_SCALAR >::evidenceImpact(), gum::JointTargetedInference< GUM_SCALAR >::evidenceJointImpact(), and gum::JointTargetedMRFInference< GUM_SCALAR >::evidenceJointImpact().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ chgEvidence() [6/7]

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::chgEvidence	(	NodeId	id,
		const std::string &	label )

finalvirtual

change the value of an already existing hard evidence

Exceptions

UndefinedElement	if id does not belong to the Bayesian network
InvalidArgument	if val is not a value for id
InvalidArgument	if id does not already have an evidence

Definition at line 404 of file graphicalModelInference_tpl.h.

                                                                                         {
    chgEvidence(id, this->model().variable(id)[label]);
  }

References chgEvidence(), and model().

Here is the call graph for this function:

◆ chgEvidence() [7/7]

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::chgEvidence	(	NodeId	id,
		const std::vector< GUM_SCALAR > &	vals )

finalvirtual

change the value of an already existing evidence (might be soft or hard)

Exceptions

UndefinedElement	if id does not belong to the Bayesian network
InvalidArgument	if the node does not already have an evidence
FatalError	if vals=[0,0,...,0]
InvalidArgument	if the size of vals is different from the domain size of node id

Definition at line 420 of file graphicalModelInference_tpl.h.

                                                                                              {
    // check whether this corresponds to an evidence
    if (_model_ == nullptr)
      GUM_ERROR(NullElement,
                "No Bayes net has been assigned to the "
                "inference algorithm");
 
    if (!_model_->exists(id)) { GUM_ERROR(UndefinedElement, id << " is not a NodeId in the model") }
 
    if (_model_->variable(id).domainSize() != vals.size()) {
      GUM_ERROR(InvalidArgument,
                "node " << _model_->variable(id) << " and its evidence have different sizes.");
    }
 
    // create the tensor corresponding to vals
    Tensor< GUM_SCALAR > pot;
    pot.add(_model_->variable(id));
    pot.fillWith(vals);
    chgEvidence(pot);
  }

References _model_, chgEvidence(), and GUM_ERROR.

Here is the call graph for this function:

◆ clear()

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::clear ( )

virtual

clears all the data structures allocated for the last inference

Reimplemented in gum::ShaferShenoyLIMIDInference< GUM_SCALAR >.

Definition at line 155 of file graphicalModelInference_tpl.h.

                                                           {
    eraseAllEvidence();
    setState_(StateOfInference::OutdatedStructure);
  }

References eraseAllEvidence(), OutdatedStructure, and setState_().

Referenced by setModel_().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ domainSizes()

template<typename GUM_SCALAR>

INLINE const NodeProperty< Size > & gum::GraphicalModelInference< GUM_SCALAR >::domainSizes ( ) const

finalvirtual

get the domain sizes of the random variables of the model

Definition at line 173 of file graphicalModelInference_tpl.h.

                                                                                              {
    return _domain_sizes_;
  }

References _domain_sizes_.

◆ eraseAllEvidence()

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::eraseAllEvidence ( )

finalvirtual

removes all the evidence entered into the network

Definition at line 540 of file graphicalModelInference_tpl.h.

                                                                      {
    bool has_hard_evidence = !_hard_evidence_.empty();
    this->onAllEvidenceErased_(has_hard_evidence);
 
    for (const auto& pair: _evidence_) {
      if (pair.second != nullptr) { delete (pair.second); }
    }
 
    _evidence_.clear();
    _hard_evidence_.clear();
    _hard_evidence_nodes_.clear();
    _soft_evidence_nodes_.clear();
 
    if (has_hard_evidence) {
      setState_(StateOfInference::OutdatedStructure);
    } else {
      if (!isInferenceOutdatedStructure()) { setState_(StateOfInference::OutdatedTensors); }
    }
  }

References _evidence_, _hard_evidence_, _hard_evidence_nodes_, _soft_evidence_nodes_, isInferenceOutdatedStructure(), onAllEvidenceErased_(), OutdatedStructure, OutdatedTensors, and setState_().

Referenced by clear(), gum::MarginalTargetedInference< GUM_SCALAR >::evidenceImpact(), gum::MarginalTargetedMRFInference< GUM_SCALAR >::evidenceImpact(), gum::JointTargetedInference< GUM_SCALAR >::evidenceJointImpact(), gum::JointTargetedMRFInference< GUM_SCALAR >::evidenceJointImpact(), gum::JointTargetedInference< GUM_SCALAR >::jointMutualInformation(), and gum::JointTargetedMRFInference< GUM_SCALAR >::jointMutualInformation().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ eraseEvidence() [1/2]

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::eraseEvidence ( const std::string & nodeName )

finalvirtual

removed the evidence, if any, corresponding to node of name nodeName

Definition at line 534 of file graphicalModelInference_tpl.h.

                                                                                            {
    eraseEvidence(this->model().idFromName(nodeName));
  }

References eraseEvidence(), and model().

Here is the call graph for this function:

◆ eraseEvidence() [2/2]

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::eraseEvidence ( NodeId id )

finalvirtual

removed the evidence, if any, corresponding to node id

Definition at line 514 of file graphicalModelInference_tpl.h.

                                                                            {
    if (hasEvidence(id)) {
      if (hasHardEvidence(id)) {
        onEvidenceErased_(id, true);
        _hard_evidence_.erase(id);
        _hard_evidence_nodes_.erase(id);
        setState_(StateOfInference::OutdatedStructure);
      } else {
        onEvidenceErased_(id, false);
        _soft_evidence_nodes_.erase(id);
        if (!isInferenceOutdatedStructure()) { setState_(StateOfInference::OutdatedTensors); }
      }
 
      delete (_evidence_[id]);
      _evidence_.erase(id);
    }
  }

References _evidence_, _hard_evidence_, _hard_evidence_nodes_, _soft_evidence_nodes_, hasEvidence(), hasHardEvidence(), isInferenceOutdatedStructure(), onEvidenceErased_(), OutdatedStructure, OutdatedTensors, and setState_().

Referenced by eraseEvidence().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ evidence()

template<typename GUM_SCALAR>

INLINE const NodeProperty< const Tensor< GUM_SCALAR > * > & gum::GraphicalModelInference< GUM_SCALAR >::evidence ( ) const

returns the set of evidence

Definition at line 587 of file graphicalModelInference_tpl.h.

                                                                     {
    return _evidence_;
  }

References _evidence_.

Referenced by gum::ImportanceSampling< GUM_SCALAR >::onContextualize_(), gum::MarginalTargetedInference< GUM_SCALAR >::posterior(), and gum::MarginalTargetedMRFInference< GUM_SCALAR >::posterior().

Here is the caller graph for this function:

◆ hardEvidence()

template<typename GUM_SCALAR>

INLINE const NodeProperty< Idx > & gum::GraphicalModelInference< GUM_SCALAR >::hardEvidence ( ) const

indicate for each node with hard evidence which value it took

Definition at line 580 of file graphicalModelInference_tpl.h.

                                                                                              {
    return _hard_evidence_;
  }

References _hard_evidence_.

Referenced by gum::GibbsSampling< GUM_SCALAR >::GibbsSampling(), gum::SamplingInference< GUM_SCALAR >::contextualize(), gum::ImportanceSampling< GUM_SCALAR >::draw_(), and gum::WeightedSampling< GUM_SCALAR >::draw_().

Here is the caller graph for this function:

◆ hardEvidenceNodes()

template<typename GUM_SCALAR>

INLINE const NodeSet & gum::GraphicalModelInference< GUM_SCALAR >::hardEvidenceNodes ( ) const

returns the set of nodes with hard evidence

the set of nodes that received hard evidence

Definition at line 599 of file graphicalModelInference_tpl.h.

                                                                                       {
    return _hard_evidence_nodes_;
  }

References _hard_evidence_nodes_.

Referenced by gum::SamplingInference< GUM_SCALAR >::contextualize(), gum::ImportanceSampling< GUM_SCALAR >::draw_(), gum::MonteCarloSampling< GUM_SCALAR >::draw_(), gum::WeightedSampling< GUM_SCALAR >::draw_(), gum::ImportanceSampling< GUM_SCALAR >::onContextualize_(), gum::MarginalTargetedInference< GUM_SCALAR >::posterior(), gum::MarginalTargetedMRFInference< GUM_SCALAR >::posterior(), gum::SamplingInference< GUM_SCALAR >::setEstimatorFromBN_(), and gum::SamplingInference< GUM_SCALAR >::setEstimatorFromLBP_().

Here is the caller graph for this function:

◆ hasEvidence() [1/3]

template<typename GUM_SCALAR>

INLINE bool gum::GraphicalModelInference< GUM_SCALAR >::hasEvidence ( ) const

finalvirtual

indicates whether some node(s) have received evidence

Definition at line 346 of file graphicalModelInference_tpl.h.

                                                                       {
    return !_evidence_.empty();
  }

References _evidence_.

Referenced by addEvidence(), chgEvidence(), eraseEvidence(), and hasEvidence().

Here is the caller graph for this function:

◆ hasEvidence() [2/3]

template<typename GUM_SCALAR>

INLINE bool gum::GraphicalModelInference< GUM_SCALAR >::hasEvidence ( const std::string & nodeName ) const

finalvirtual

indicates whether node id has received an evidence

Definition at line 371 of file graphicalModelInference_tpl.h.

                                                                                        {
    return hasEvidence(this->model().idFromName(nodeName));
  }

References hasEvidence(), and model().

Here is the call graph for this function:

◆ hasEvidence() [3/3]

template<typename GUM_SCALAR>

INLINE bool gum::GraphicalModelInference< GUM_SCALAR >::hasEvidence ( NodeId id ) const

finalvirtual

indicates whether node id has received an evidence

Definition at line 352 of file graphicalModelInference_tpl.h.

                                                                                {
    return _evidence_.exists(id);
  }

References _evidence_.

◆ hasHardEvidence() [1/2]

template<typename GUM_SCALAR>

INLINE bool gum::GraphicalModelInference< GUM_SCALAR >::hasHardEvidence ( const std::string & nodeName ) const

finalvirtual

indicates whether node id has received a hard evidence

Definition at line 378 of file graphicalModelInference_tpl.h.

                                                                                            {
    return hasHardEvidence(this->model().idFromName(nodeName));
  }

References hasHardEvidence(), and model().

Here is the call graph for this function:

◆ hasHardEvidence() [2/2]

template<typename GUM_SCALAR>

INLINE bool gum::GraphicalModelInference< GUM_SCALAR >::hasHardEvidence ( NodeId id ) const

finalvirtual

indicates whether node id has received a hard evidence

Definition at line 358 of file graphicalModelInference_tpl.h.

                                                                                    {
    return _hard_evidence_nodes_.exists(id);
  }

References _hard_evidence_nodes_.

Referenced by chgEvidence(), gum::ImportanceSampling< GUM_SCALAR >::draw_(), eraseEvidence(), hasHardEvidence(), and gum::JointTargetedMRFInference< GUM_SCALAR >::jointPosterior().

Here is the caller graph for this function:

◆ hasNoModel_()

template<typename GUM_SCALAR>

bool gum::GraphicalModelInference< GUM_SCALAR >::hasNoModel_ ( ) const

inlineprotected

Definition at line 542 of file graphicalModelInference.h.

542{ return _model_ == nullptr; };

References _model_.

Here is the caller graph for this function:

◆ hasSoftEvidence() [1/2]

template<typename GUM_SCALAR>

INLINE bool gum::GraphicalModelInference< GUM_SCALAR >::hasSoftEvidence ( const std::string & nodeName ) const

finalvirtual

indicates whether node id has received a soft evidence

Definition at line 385 of file graphicalModelInference_tpl.h.

                                                                                            {
    return hasSoftEvidence(this->model().idFromName(nodeName));
  }

References hasSoftEvidence(), and model().

Here is the call graph for this function:

◆ hasSoftEvidence() [2/2]

template<typename GUM_SCALAR>

INLINE bool gum::GraphicalModelInference< GUM_SCALAR >::hasSoftEvidence ( NodeId id ) const

finalvirtual

indicates whether node id has received a soft evidence

Definition at line 364 of file graphicalModelInference_tpl.h.

                                                                                    {
    return _soft_evidence_nodes_.exists(id);
  }

References _soft_evidence_nodes_.

Referenced by hasSoftEvidence().

Here is the caller graph for this function:

◆ isInferenceDone()

template<typename GUM_SCALAR>

INLINE bool gum::GraphicalModelInference< GUM_SCALAR >::isInferenceDone ( ) const

finalvirtualnoexcept

returns whether the inference object is in a InferenceDone state

The inference object is in a done state when the posteriors can be retrieved without performing a new inference, i.e., all the heavy computations have already been performed. Typically, in a junction tree algorithm, this corresponds to a situation in which all the messages needed in the JT have been computed and sent.

Definition at line 104 of file graphicalModelInference_tpl.h.

                                                                                    {
    return (_state_ == StateOfInference::Done);
  }

References _state_, and Done.

Referenced by gum::JointTargetedInference< GUM_SCALAR >::jointPosterior(), gum::JointTargetedMRFInference< GUM_SCALAR >::jointPosterior(), makeInference(), gum::MarginalTargetedInference< GUM_SCALAR >::posterior(), gum::MarginalTargetedMRFInference< GUM_SCALAR >::posterior(), and prepareInference().

Here is the caller graph for this function:

◆ isInferenceOutdatedStructure()

template<typename GUM_SCALAR>

INLINE bool gum::GraphicalModelInference< GUM_SCALAR >::isInferenceOutdatedStructure ( ) const

finalvirtualnoexcept

returns whether the inference object is in a OutdatedStructure state

Definition at line 92 of file graphicalModelInference_tpl.h.

                                                                                                 {
    return (_state_ == StateOfInference::OutdatedStructure);
  }

References _state_, and OutdatedStructure.

Referenced by chgEvidence(), eraseAllEvidence(), and eraseEvidence().

Here is the caller graph for this function:

◆ isInferenceOutdatedTensors()

template<typename GUM_SCALAR>

INLINE bool gum::GraphicalModelInference< GUM_SCALAR >::isInferenceOutdatedTensors ( ) const

finalvirtualnoexcept

returns whether the inference object is in a OutdatedTensor state

Definition at line 98 of file graphicalModelInference_tpl.h.

                                                                                               {
    return (_state_ == StateOfInference::OutdatedTensors);
  }

References _state_, and OutdatedTensors.

◆ isInferenceReady()

template<typename GUM_SCALAR>

INLINE bool gum::GraphicalModelInference< GUM_SCALAR >::isInferenceReady ( ) const

finalvirtualnoexcept

returns whether the inference object is in a ready state

Definition at line 86 of file graphicalModelInference_tpl.h.

                                                                                     {
    return (_state_ == StateOfInference::ReadyForInference);
  }

References _state_, and ReadyForInference.

Referenced by makeInference(), gum::SamplingInference< GUM_SCALAR >::onStateChanged_(), and prepareInference().

Here is the caller graph for this function:

◆ makeInference()

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::makeInference ( )

finalvirtual

perform the heavy computations needed to compute the targets' posteriors

In a Junction tree propagation scheme, for instance, the heavy computations are those of the messages sent in the JT. This is precisely what makeInference should compute. Later, the computations of the posteriors can be done "lightly" by multiplying and projecting those messages.

Definition at line 638 of file graphicalModelInference_tpl.h.

                                                                   {
    if (isInferenceDone()) { return; }
 
    if (!isInferenceReady()) { prepareInference(); }
 
    makeInference_();
 
    setState_(StateOfInference::Done);
  }

References Done, isInferenceDone(), isInferenceReady(), makeInference_(), prepareInference(), and setState_().

Referenced by gum::MarginalTargetedInference< GUM_SCALAR >::evidenceImpact(), gum::MarginalTargetedMRFInference< GUM_SCALAR >::evidenceImpact(), gum::JointTargetedInference< GUM_SCALAR >::evidenceJointImpact(), gum::JointTargetedMRFInference< GUM_SCALAR >::evidenceJointImpact(), gum::JointTargetedInference< GUM_SCALAR >::jointMutualInformation(), gum::JointTargetedMRFInference< GUM_SCALAR >::jointMutualInformation(), gum::JointTargetedInference< GUM_SCALAR >::jointPosterior(), gum::JointTargetedMRFInference< GUM_SCALAR >::jointPosterior(), gum::LoopySamplingInference< GUM_SCALAR, APPROX >::makeInference_(), gum::MarginalTargetedInference< GUM_SCALAR >::posterior(), and gum::MarginalTargetedMRFInference< GUM_SCALAR >::posterior().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ makeInference_()

template<typename GUM_SCALAR>

virtual void gum::GraphicalModelInference< GUM_SCALAR >::makeInference_ ( )

protectedpure virtual

called when the inference has to be performed effectively

Once the inference is done, fillPosterior_ can be called.

Implemented in gum::LazyPropagation< GUM_SCALAR >, gum::LoopyBeliefPropagation< GUM_SCALAR >, gum::LoopySamplingInference< GUM_SCALAR, GibbsSampling >, gum::LoopySamplingInference< GUM_SCALAR, ImportanceSampling >, gum::LoopySamplingInference< GUM_SCALAR, MonteCarloSampling >, gum::LoopySamplingInference< GUM_SCALAR, WeightedSampling >, gum::SamplingInference< GUM_SCALAR >, gum::ShaferShenoyInference< GUM_SCALAR >, gum::ShaferShenoyLIMIDInference< GUM_SCALAR >, gum::ShaferShenoyMRFInference< GUM_SCALAR >, and gum::VariableElimination< GUM_SCALAR >.

References makeInference_(), setOutdatedStructureState_(), and setOutdatedTensorsState_().

Referenced by makeInference(), and makeInference_().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ model()

template<typename GUM_SCALAR>

INLINE const GraphicalModel & gum::GraphicalModelInference< GUM_SCALAR >::model ( ) const

finalvirtual

Returns a constant reference over the IBayesNet referenced by this class.

Exceptions

UndefinedElement is raised if no Bayes net has been assigned to the inference.

Definition at line 126 of file graphicalModelInference_tpl.h.

                                                                                  {
    if (_model_ == nullptr)
      GUM_ERROR(UndefinedElement,
                "No Bayes net has been assigned to "
                "the inference algorithm.");
    return *_model_;
  }

References _model_, and GUM_ERROR.

Referenced by GraphicalModelInference(), addEvidence(), addEvidence(), addEvidence(), addEvidence(), gum::BayesNetInference< GUM_SCALAR >::BN(), chgEvidence(), chgEvidence(), chgEvidence(), chgEvidence(), eraseEvidence(), hasEvidence(), hasHardEvidence(), hasSoftEvidence(), gum::MRFInference< GUM_SCALAR >::MRF(), onModelChanged_(), gum::ShaferShenoyLIMIDInference< GUM_SCALAR >::onModelChanged_(), setModel_(), and setModelDuringConstruction_().

Here is the caller graph for this function:

◆ nbrEvidence()

template<typename GUM_SCALAR>

INLINE Size gum::GraphicalModelInference< GUM_SCALAR >::nbrEvidence ( ) const

finalvirtual

returns the number of evidence entered into the Bayesian network

Definition at line 562 of file graphicalModelInference_tpl.h.

                                                                       {
    return _evidence_.size();
  }

References _evidence_.

◆ nbrHardEvidence()

template<typename GUM_SCALAR>

INLINE Size gum::GraphicalModelInference< GUM_SCALAR >::nbrHardEvidence ( ) const

finalvirtual

returns the number of hard evidence entered into the Bayesian network

Definition at line 568 of file graphicalModelInference_tpl.h.

                                                                           {
    return _hard_evidence_nodes_.size();
  }

References _hard_evidence_nodes_.

◆ nbrSoftEvidence()

template<typename GUM_SCALAR>

INLINE Size gum::GraphicalModelInference< GUM_SCALAR >::nbrSoftEvidence ( ) const

finalvirtual

returns the number of soft evidence entered into the Bayesian network

Definition at line 574 of file graphicalModelInference_tpl.h.

                                                                           {
    return _soft_evidence_nodes_.size();
  }

References _soft_evidence_nodes_.

◆ onAllEvidenceErased_()

template<typename GUM_SCALAR>

virtual void gum::GraphicalModelInference< GUM_SCALAR >::onAllEvidenceErased_ ( bool contains_hard_evidence )

protectedpure virtual

fired before all the evidence are erased

Implemented in gum::LazyPropagation< GUM_SCALAR >, gum::LoopyBeliefPropagation< GUM_SCALAR >, gum::SamplingInference< GUM_SCALAR >, gum::ShaferShenoyInference< GUM_SCALAR >, gum::ShaferShenoyLIMIDInference< GUM_SCALAR >, gum::ShaferShenoyMRFInference< GUM_SCALAR >, and gum::VariableElimination< GUM_SCALAR >.

References onAllEvidenceErased_().

Referenced by eraseAllEvidence(), and onAllEvidenceErased_().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ onEvidenceAdded_()

template<typename GUM_SCALAR>

virtual void gum::GraphicalModelInference< GUM_SCALAR >::onEvidenceAdded_	(	const NodeId	id,
		bool	isHardEvidence )

protectedpure virtual

fired after a new evidence is inserted

Implemented in gum::LazyPropagation< GUM_SCALAR >, gum::LoopyBeliefPropagation< GUM_SCALAR >, gum::SamplingInference< GUM_SCALAR >, gum::ShaferShenoyInference< GUM_SCALAR >, gum::ShaferShenoyLIMIDInference< GUM_SCALAR >, gum::ShaferShenoyMRFInference< GUM_SCALAR >, and gum::VariableElimination< GUM_SCALAR >.

References onEvidenceAdded_().

Referenced by addEvidence(), and onEvidenceAdded_().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ onEvidenceChanged_()

template<typename GUM_SCALAR>

virtual void gum::GraphicalModelInference< GUM_SCALAR >::onEvidenceChanged_	(	const NodeId	id,
		bool	hasChangedSoftHard )

protectedpure virtual

fired after an evidence is changed, in particular when its status (soft/hard) changes

Parameters

nodeId	the node of the changed evidence
hasChangedSoftHard	true if the evidence has changed from Soft to Hard or from Hard to Soft

Implemented in gum::LazyPropagation< GUM_SCALAR >, gum::LoopyBeliefPropagation< GUM_SCALAR >, gum::SamplingInference< GUM_SCALAR >, gum::ShaferShenoyInference< GUM_SCALAR >, gum::ShaferShenoyLIMIDInference< GUM_SCALAR >, gum::ShaferShenoyMRFInference< GUM_SCALAR >, and gum::VariableElimination< GUM_SCALAR >.

References onEvidenceChanged_().

Referenced by chgEvidence(), and onEvidenceChanged_().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ onEvidenceErased_()

template<typename GUM_SCALAR>

virtual void gum::GraphicalModelInference< GUM_SCALAR >::onEvidenceErased_	(	const NodeId	id,
		bool	isHardEvidence )

protectedpure virtual

fired before an evidence is removed

Implemented in gum::LazyPropagation< GUM_SCALAR >, gum::LoopyBeliefPropagation< GUM_SCALAR >, gum::SamplingInference< GUM_SCALAR >, gum::ShaferShenoyInference< GUM_SCALAR >, gum::ShaferShenoyLIMIDInference< GUM_SCALAR >, gum::ShaferShenoyMRFInference< GUM_SCALAR >, and gum::VariableElimination< GUM_SCALAR >.

References onEvidenceErased_().

Referenced by eraseEvidence(), and onEvidenceErased_().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ onModelChanged_()

template<typename GUM_SCALAR>

virtual void gum::GraphicalModelInference< GUM_SCALAR >::onModelChanged_ ( const GraphicalModel * model )

protectedpure virtual

fired after a new Bayes net has been assigned to the engine

Implemented in gum::JointTargetedInference< GUM_SCALAR >, gum::JointTargetedMRFInference< GUM_SCALAR >, gum::LazyPropagation< GUM_SCALAR >, gum::LoopyBeliefPropagation< GUM_SCALAR >, gum::MarginalTargetedInference< GUM_SCALAR >, gum::MarginalTargetedMRFInference< GUM_SCALAR >, gum::SamplingInference< GUM_SCALAR >, gum::ShaferShenoyInference< GUM_SCALAR >, gum::ShaferShenoyLIMIDInference< GUM_SCALAR >, gum::ShaferShenoyMRFInference< GUM_SCALAR >, and gum::VariableElimination< GUM_SCALAR >.

References model(), and onModelChanged_().

Referenced by onModelChanged_(), and setModel_().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ onStateChanged_()

template<typename GUM_SCALAR>

virtual void gum::GraphicalModelInference< GUM_SCALAR >::onStateChanged_ ( )

protectedpure virtual

fired when the stage is changed

Implemented in gum::LazyPropagation< GUM_SCALAR >, gum::LoopyBeliefPropagation< GUM_SCALAR >, gum::SamplingInference< GUM_SCALAR >, gum::ShaferShenoyInference< GUM_SCALAR >, gum::ShaferShenoyLIMIDInference< GUM_SCALAR >, gum::ShaferShenoyMRFInference< GUM_SCALAR >, and gum::VariableElimination< GUM_SCALAR >.

References onStateChanged_().

Referenced by onStateChanged_(), and setState_().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ prepareInference()

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::prepareInference ( )

finalvirtual

prepare the internal inference structures for the next inference

Definition at line 622 of file graphicalModelInference_tpl.h.

                                                                      {
    if (isInferenceReady() || isInferenceDone()) { return; }
 
    if (_model_ == nullptr)
      GUM_ERROR(NullElement,
                "No model been assigned to the "
                "inference algorithm");
 
    if (_state_ == StateOfInference::OutdatedStructure) updateOutdatedStructure_();
    else updateOutdatedTensors_();
 
    setState_(StateOfInference::ReadyForInference);
  }

References _model_, _state_, GUM_ERROR, isInferenceDone(), isInferenceReady(), OutdatedStructure, ReadyForInference, setState_(), updateOutdatedStructure_(), and updateOutdatedTensors_().

Referenced by makeInference(), and gum::SamplingInference< GUM_SCALAR >::samplingBN().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ setModel_()

template<typename GUM_SCALAR>

void gum::GraphicalModelInference< GUM_SCALAR >::setModel_ ( const GraphicalModel * model )

protected

Definition at line 136 of file graphicalModelInference_tpl.h.

                                                                                   {
    clear();
    _model_ = model;
    _computeDomainSizes_();
    onModelChanged_(model);
    setState_(StateOfInference::OutdatedStructure);
  }

References _computeDomainSizes_(), _model_, clear(), model(), onModelChanged_(), OutdatedStructure, and setState_().

Referenced by gum::BayesNetInference< GUM_SCALAR >::setBN(), and gum::MRFInference< GUM_SCALAR >::setMRF().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ setModelDuringConstruction_()

template<typename GUM_SCALAR>

void gum::GraphicalModelInference< GUM_SCALAR >::setModelDuringConstruction_ ( const GraphicalModel * model )

protected

assigns a model during the inference engine construction

Definition at line 146 of file graphicalModelInference_tpl.h.

                                   {
    _model_ = model;
    _computeDomainSizes_();
    setState_(StateOfInference::OutdatedStructure);
  }

References _computeDomainSizes_(), _model_, model(), OutdatedStructure, and setState_().

Referenced by gum::BayesNetInference< GUM_SCALAR >::_setBayesNetDuringConstruction_(), and gum::MRFInference< GUM_SCALAR >::_setMRFDuringConstruction_().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ setOutdatedStructureState_()

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::setOutdatedStructureState_ ( )

protected

put the inference into an outdated model structure state

OutdatedStructure: in this state, the inference is fully unprepared to be applied because some events changed the "logical" structure of the model: for instance a node received a hard evidence, which implies that its outgoing arcs can be removed from the model, hence involving a structural change in the model. As a consequence, the (incremental) inference (probably) needs a significant amount of preparation to be ready for the next inference. In a Lazy propagation, for instance, this step amounts to compute a new join tree, hence a new structure in which inference will be applied. Note that classes that inherit from graphicalModelInference may be smarter than graphicalModelInference and may, in some situations, find out that their data structures are still ok for inference and, therefore, only resort to perform the actions related to the OutdatedTensors state. As an example, consider a LazyPropagation inference in Bayes Net A->B->C->D->E in which C has received hard evidence e_C and E is the only target. In this case, A and B are not needed for inference, the only tensors that matter are P(D|e_C) and P(E|D). So the smallest join tree needed for inference contains only one clique DE. Now, adding new evidence e_A on A has no impact on E given hard evidence e_C. In this case, LazyPropagation can be smart and not update its join tree.

Definition at line 609 of file graphicalModelInference_tpl.h.

                                                                                {
    setState_(StateOfInference::OutdatedStructure);
  }

References OutdatedStructure, and setState_().

Referenced by makeInference_().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ setOutdatedTensorsState_()

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::setOutdatedTensorsState_ ( )

protected

puts the inference into an OutdatedTensors state if it is not already in an OutdatedStructure state

OutdatedTensors: in this state, the structure of the model remains unchanged, only some tensors stored in it have changed. Therefore, the inference probably just needs to invalidate some already computed tensors to be ready. Only a light amount of preparation is needed to be able to perform inference.

Definition at line 616 of file graphicalModelInference_tpl.h.

                                                                              {
    setState_(StateOfInference::OutdatedTensors);
  }

References OutdatedTensors, and setState_().

Referenced by makeInference_().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ setState_()

template<typename GUM_SCALAR>

INLINE void gum::GraphicalModelInference< GUM_SCALAR >::setState_ ( const StateOfInference state )

finalprotectedvirtual

set the state of the inference engine and call the notification onStateChanged_ when necessary (i.e. when the state has effectively changed).

Definition at line 117 of file graphicalModelInference_tpl.h.

                                                                                           {
    if (_state_ != state) {
      _state_ = state;
      onStateChanged_();
    }
  }

References _state_, onStateChanged_(), and state().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ softEvidenceNodes()

template<typename GUM_SCALAR>

INLINE const NodeSet & gum::GraphicalModelInference< GUM_SCALAR >::softEvidenceNodes ( ) const

returns the set of nodes with soft evidence

the set of nodes that received soft evidence

Definition at line 593 of file graphicalModelInference_tpl.h.

                                                                                       {
    return _soft_evidence_nodes_;
  }

References _soft_evidence_nodes_.

Referenced by gum::SamplingInference< GUM_SCALAR >::contextualize().

Here is the caller graph for this function:

◆ state()

template<typename GUM_SCALAR>

INLINE GraphicalModelInference< GUM_SCALAR >::StateOfInference gum::GraphicalModelInference< GUM_SCALAR >::state ( ) const

finalvirtualnoexcept

returns the state of the inference engine

Definition at line 111 of file graphicalModelInference_tpl.h.

                                                                  {
    return _state_;
  }

References _state_.

Referenced by setState_().

Here is the caller graph for this function:

◆ updateOutdatedStructure_()

template<typename GUM_SCALAR>

virtual void gum::GraphicalModelInference< GUM_SCALAR >::updateOutdatedStructure_ ( )

protectedpure virtual

prepares inference when the latter is in OutdatedStructure state

Note that the values of evidence are not necessarily known and can be changed between updateOutdatedStructure_ and makeInference_.

Implemented in gum::LazyPropagation< GUM_SCALAR >, gum::LoopyBeliefPropagation< GUM_SCALAR >, gum::SamplingInference< GUM_SCALAR >, gum::ShaferShenoyInference< GUM_SCALAR >, gum::ShaferShenoyLIMIDInference< GUM_SCALAR >, gum::ShaferShenoyMRFInference< GUM_SCALAR >, and gum::VariableElimination< GUM_SCALAR >.

References updateOutdatedStructure_().

Referenced by prepareInference(), and updateOutdatedStructure_().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ updateOutdatedTensors_()

template<typename GUM_SCALAR>

virtual void gum::GraphicalModelInference< GUM_SCALAR >::updateOutdatedTensors_ ( )

protectedpure virtual

prepares inference when the latter is in OutdatedTensors state

Note that the values of evidence are not necessarily known and can be changed between updateOutdatedTensors_ and makeInference_.

Implemented in gum::LazyPropagation< GUM_SCALAR >, gum::LoopyBeliefPropagation< GUM_SCALAR >, gum::SamplingInference< GUM_SCALAR >, gum::ShaferShenoyInference< GUM_SCALAR >, gum::ShaferShenoyLIMIDInference< GUM_SCALAR >, gum::ShaferShenoyMRFInference< GUM_SCALAR >, and gum::VariableElimination< GUM_SCALAR >.

References updateOutdatedTensors_().

Referenced by prepareInference(), and updateOutdatedTensors_().

Here is the call graph for this function:

Here is the caller graph for this function:

Member Data Documentation

◆ _domain_sizes_

template<typename GUM_SCALAR>

NodeProperty< Size > gum::GraphicalModelInference< GUM_SCALAR >::_domain_sizes_

private

the domain sizes of the random variables

Definition at line 507 of file graphicalModelInference.h.

Referenced by _computeDomainSizes_(), and domainSizes().

◆ _evidence_

template<typename GUM_SCALAR>

NodeProperty< const Tensor< GUM_SCALAR >* > gum::GraphicalModelInference< GUM_SCALAR >::_evidence_

private

the set of evidence entered into the network

Definition at line 510 of file graphicalModelInference.h.

Referenced by ~GraphicalModelInference(), addEvidence(), chgEvidence(), eraseAllEvidence(), eraseEvidence(), evidence(), hasEvidence(), hasEvidence(), and nbrEvidence().

◆ _hard_evidence_

template<typename GUM_SCALAR>

NodeProperty< Idx > gum::GraphicalModelInference< GUM_SCALAR >::_hard_evidence_

private

assign to each node with a hard evidence the index of its observed value

Definition at line 513 of file graphicalModelInference.h.

Referenced by addEvidence(), chgEvidence(), eraseAllEvidence(), eraseEvidence(), and hardEvidence().

◆ _hard_evidence_nodes_

template<typename GUM_SCALAR>

NodeSet gum::GraphicalModelInference< GUM_SCALAR >::_hard_evidence_nodes_

private

the set of nodes that received hard evidence

Definition at line 519 of file graphicalModelInference.h.

Referenced by addEvidence(), chgEvidence(), eraseAllEvidence(), eraseEvidence(), hardEvidenceNodes(), hasHardEvidence(), and nbrHardEvidence().

◆ _model_

template<typename GUM_SCALAR>

const GraphicalModel* gum::GraphicalModelInference< GUM_SCALAR >::_model_ {nullptr}

private

the Bayes net on which we perform inferences

Definition at line 504 of file graphicalModelInference.h.

504{nullptr};

Referenced by GraphicalModelInference(), _computeDomainSizes_(), _createHardEvidence_(), addEvidence(), addEvidence(), chgEvidence(), chgEvidence(), hasNoModel_(), model(), prepareInference(), setModel_(), and setModelDuringConstruction_().

◆ _soft_evidence_nodes_

template<typename GUM_SCALAR>

NodeSet gum::GraphicalModelInference< GUM_SCALAR >::_soft_evidence_nodes_

private

the set of nodes that received soft evidence

Definition at line 516 of file graphicalModelInference.h.

Referenced by addEvidence(), chgEvidence(), eraseAllEvidence(), eraseEvidence(), hasSoftEvidence(), nbrSoftEvidence(), and softEvidenceNodes().

◆ _state_

template<typename GUM_SCALAR>

StateOfInference gum::GraphicalModelInference< GUM_SCALAR >::_state_ {StateOfInference::OutdatedStructure}

private

the current state of the inference (outdated/ready/done)

Definition at line 501 of file graphicalModelInference.h.

501{StateOfInference::OutdatedStructure};

Referenced by isInferenceDone(), isInferenceOutdatedStructure(), isInferenceOutdatedTensors(), isInferenceReady(), prepareInference(), setState_(), and state().

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

agrum/base/graphicalModels/inference/graphicalModelInference.h
agrum/base/graphicalModels/inference/graphicalModelInference_tpl.h

Public Types

Public Member Functions

Protected Member Functions

Private Member Functions

Private Attributes

Detailed Description

Member Enumeration Documentation

◆ StateOfInference

Constructor & Destructor Documentation

◆ GraphicalModelInference() [1/2]

◆ GraphicalModelInference() [2/2]

◆ ~GraphicalModelInference()

Member Function Documentation

◆ _computeDomainSizes_()

◆ _createHardEvidence_()

◆ _isHardEvidence_()

◆ addEvidence() [1/8]

◆ addEvidence() [2/8]

◆ addEvidence() [3/8]

◆ addEvidence() [4/8]

◆ addEvidence() [5/8]

◆ addEvidence() [6/8]

◆ addEvidence() [7/8]

◆ addEvidence() [8/8]

◆ addListOfEvidence()

◆ addSetOfEvidence()

◆ chgEvidence() [1/7]

◆ chgEvidence() [2/7]

◆ chgEvidence() [3/7]

◆ chgEvidence() [4/7]

◆ chgEvidence() [5/7]

◆ chgEvidence() [6/7]

◆ chgEvidence() [7/7]

◆ clear()

◆ domainSizes()

◆ eraseAllEvidence()

◆ eraseEvidence() [1/2]

◆ eraseEvidence() [2/2]

◆ evidence()

◆ hardEvidence()

◆ hardEvidenceNodes()

◆ hasEvidence() [1/3]

◆ hasEvidence() [2/3]

◆ hasEvidence() [3/3]

◆ hasHardEvidence() [1/2]

◆ hasHardEvidence() [2/2]

◆ hasNoModel_()

◆ hasSoftEvidence() [1/2]

◆ hasSoftEvidence() [2/2]

◆ isInferenceDone()

◆ isInferenceOutdatedStructure()

◆ isInferenceOutdatedTensors()

◆ isInferenceReady()

◆ makeInference()

◆ makeInference_()

◆ model()

◆ nbrEvidence()

◆ nbrHardEvidence()

◆ nbrSoftEvidence()

◆ onAllEvidenceErased_()

◆ onEvidenceAdded_()

◆ onEvidenceChanged_()

◆ onEvidenceErased_()

◆ onModelChanged_()

◆ onStateChanged_()

◆ prepareInference()

◆ setModel_()

◆ setModelDuringConstruction_()

◆ setOutdatedStructureState_()

◆ setOutdatedTensorsState_()

◆ setState_()

◆ softEvidenceNodes()

◆ state()

◆ updateOutdatedStructure_()

◆ updateOutdatedTensors_()

Member Data Documentation

◆ _domain_sizes_

◆ _evidence_

◆ _hard_evidence_

◆ _hard_evidence_nodes_