Learn a graphical representation of a function as a decision tree. More...

#include <agrum/FMDP/planning/FunctionGraph/iti.h>

Inheritance diagram for gum::ITI< AttributeSelection, isScalar >:

Collaboration diagram for gum::ITI< AttributeSelection, isScalar >:

Public Member Functions
Constructor & destructor.
	ITI (MultiDimFunctionGraph< double > target, double attributeSelectionThreshold, gum::VariableSet attributeListe, const DiscreteVariable learnedValue)
	ITI constructor for functions describing the behaviour of one variable according to a set of other variable such as conditionnal probabilities.
	ITI (MultiDimFunctionGraph< double > *target, double attributeSelectionThreshold, gum::VariableSet attributeListe)
	ITI constructeur for real functions.
	~ITI () override
	Default destructor.
Visit Methods
Size	size ()
NodeId	root () const override
bool	isTerminal (NodeId ni) const override
const DiscreteVariable *	nodeVar (NodeId ni) const override
NodeId	nodeSon (NodeId ni, Idx modality) const override
Idx	nodeNbObservation (NodeId ni) const override
void	insertSetOfVars (MultiDimFunctionGraph< double > *ret) const override

Protected Member Functions
Function Graph Updating methods
void	insertSetOfVars_ (MultiDimFunctionGraph< double > *ret)
	insertSetOfVars_

Protected Attributes
MultiDimFunctionGraph< double > *	target_
	The final diagram we're building.
gum::VariableSet	setOfVars_
const DiscreteVariable *	value_
Sequence< ValueType >	valueAssumed_
bool	needUpdate_
Model handling datastructures
NodeGraphPart	model_
	The source of nodeId.
NodeId	root_
	The root of the ordered tree.
HashTable< NodeId, const DiscreteVariable * >	nodeVarMap_
	Gives for any node its associated variable.
HashTable< NodeId, NodeId * >	nodeSonsMap_
	A table giving for any node a table mapping to its son idx is the modality of associated variable.
HashTable< const DiscreteVariable , LinkedList< NodeId > >	var2Node_
	Associates to any variable the list of all nodes associated to this variable.
HashTable< NodeId, NodeDatabase< AttributeSelection, isScalar > * >	nodeId2Database_
	This hashtable binds every node to an associated NodeDatabase which handles every observation that concerns that node.
HashTable< NodeId, Set< const Observation * > * >	leafDatabase_
	This hashtable binds to every leaf an associated set of all hte observations compatible with it.

Private Types
using	ValueType

Private Member Functions
Constructor & destructor.
void	_clearValue_ ()
	Template function dispatcher.
New Observation insertion methods
void	_assumeValue_ (const Observation *obs)
	Get value assumed by studied variable for current observation.
Idx	_branchObs_ (const Observation obs, const DiscreteVariable var)
	Seek modality assumed in obs for given var.

Private Attributes
HashTable< NodeId, bool >	_staleTable_
	Hashtable indicating if given node has been modified (upon receiving new exemple or through a transpose) The aim is not if we have revise the installed variable on that node.
Idx	_nbTotalObservation_
	The total number of observation added to this tree.
double	_attributeSelectionThreshold_
	The threshold above which we consider variables to be dependant.

New Observation insertion methods
void	addObservation (const Observation *obs) override
	Inserts a new observation.
void	updateNodeWithObservation_ (const Observation *newObs, NodeId currentNodeId) override
	Will update internal graph's NodeDatabase of given node with the new observation.

Graph Structure update methods
void	updateGraph () override
	Updates the internal graph after a new observation has been added.
NodeId	insertNode_ (NodeDatabase< AttributeSelection, isScalar > nDB, const DiscreteVariable boundVar) override
	inserts a new node in internal graph
void	chgNodeBoundVar_ (NodeId chgedNodeId, const DiscreteVariable *desiredVar) override
	Changes the associated variable of a node.
void	removeNode_ (NodeId removedNodeId) override
	Removes a node from the internal graph.

Function Graph Updating methods
void	updateFunctionGraph () override
	Updates target to currently learned graph structure.
NodeId	_insertNodeInFunctionGraph_ (NodeId src)
	Inserts an internal node in the target.
NodeId	_insertTerminalNode_ (NodeId src)
	Insert a terminal node in the target.
NodeId	_insertTerminalNode_ (NodeId src, Int2Type< true >)
	Insert a terminal node in the target.
NodeId	_insertTerminalNode_ (NodeId src, Int2Type< false >)
	Insert a terminal node in the target.

Graph Structure update methods
virtual void	updateVar (const DiscreteVariable *)
	If a new modality appears to exists for given variable, call this method to turn every associated node to this variable into leaf. Graph has then indeed to be revised.
void	updateNode_ (NodeId nody, gum::VariableSet &bestVars)
	From the given sets of node, selects randomly one and installs it on given node.
virtual void	convertNode2Leaf_ (NodeId)
	Turns the given node into a leaf if not already so.
virtual void	transpose_ (NodeId, const DiscreteVariable *)
	Installs given variable to the given node, ensuring that the variable is not present in its subtree.
virtual NodeId	insertInternalNode_ (NodeDatabase< AttributeSelection, isScalar > nDB, const DiscreteVariable boundVar, NodeId *sonsMap)
	inserts a new internal node in internal graph
virtual NodeId	insertLeafNode_ (NodeDatabase< AttributeSelection, isScalar > nDB, const DiscreteVariable boundVar, Set< const Observation * > *obsSet)
	inserts a new leaf node in internal graohs

Detailed Description

template<TESTNAME AttributeSelection, bool isScalar = false>
class gum::ITI< AttributeSelection, isScalar >

Learn a graphical representation of a function as a decision tree.

This learning is done incrementaly. Hence first observation are add to the structure and then the structure is updated. Maintains two graph function : one internal for the learning and a target which is updated on demand.

Definition at line 79 of file iti.h.

Member Typedef Documentation

◆ ValueType

using gum::IncrementalGraphLearner< AttributeSelection, isScalar >::ValueType

privateinherited

Definition at line 84 of file incrementalGraphLearner.h.

Constructor & Destructor Documentation

◆ ITI() [1/2]

template<TESTNAME AttributeSelection, bool isScalar>

gum::ITI< AttributeSelection, isScalar >::ITI	(	MultiDimFunctionGraph< double > *	target,
		double	attributeSelectionThreshold,
		gum::VariableSet	attributeListe,
		const DiscreteVariable *	learnedValue )

ITI constructor for functions describing the behaviour of one variable according to a set of other variable such as conditionnal probabilities.

Parameters

target	: the MultiDimFunctionGraph in which we load the structure
attributeSelectionThreshold	: threshold under which a node is not installed (pe-pruning)
attributeListe	: Set of vars on which we rely to explain the behaviour of learned variable
learnedValue	: the variable from which we try to learn the behaviour
target	: the MultiDimFunctionGraph in which we load the structure
attributeSelectionThreshold	: threshold under which a node is not installed (pe-pruning)
temporaryAPIfix	: Issue in API in regard to IMDDI
attributeListe	: Set of vars on which we rely to explain the behaviour of learned variable
learnedValue	: the variable from which we try to learn the behaviour

Definition at line 85 of file iti_tpl.h.

                                                                                 :
      IncrementalGraphLearner< AttributeSelection, isScalar >(target, attributeListe, learnedValue),
      _nbTotalObservation_(0), _attributeSelectionThreshold_(attributeSelectionThreshold) {
    GUM_CONSTRUCTOR(ITI);
    _staleTable_.insert(this->root_, false);
  }

References gum::IncrementalGraphLearner< AttributeSelection, false >::IncrementalGraphLearner(), ITI(), _attributeSelectionThreshold_, _nbTotalObservation_, _staleTable_, and gum::IncrementalGraphLearner< AttributeSelection, false >::root_.

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

Here is the call graph for this function:

Here is the caller graph for this function:

◆ ITI() [2/2]

template<TESTNAME AttributeSelection, bool isScalar>

gum::ITI< AttributeSelection, isScalar >::ITI	(	MultiDimFunctionGraph< double > *	target,
		double	attributeSelectionThreshold,
		gum::VariableSet	attributeListe )

ITI constructeur for real functions.

We try to predict the output of a function f given a set of variable

Parameters

target	: the MultiDimFunctionGraph in which we load the structure
attributeSelectionThreshold	: threshold under which a node is not installed (pe-pruning)
attributeListe	: Set of vars on which we rely to explain the behaviour of learned function

We try to predict the output of a function f given a set of variable

Parameters

target	: the MultiDimFunctionGraph in which we load the structure
attributeSelectionThreshold	: threshold under which a node is not installed (pe-pruning)
temporaryAPIfix	: Issue in API in regard to IMDDI
attributeListeSet	of vars on which we rely to explain the behaviour of learned function

Definition at line 108 of file iti_tpl.h.

                                                                          :
      IncrementalGraphLearner< AttributeSelection, isScalar >(
          target,
          attributeListe,
          new LabelizedVariable("Reward", "", 2)),
      _nbTotalObservation_(0), _attributeSelectionThreshold_(attributeSelectionThreshold) {
    GUM_CONSTRUCTOR(ITI);
    _staleTable_.insert(this->root_, false);
  }

References gum::IncrementalGraphLearner< AttributeSelection, false >::IncrementalGraphLearner(), ITI(), _attributeSelectionThreshold_, _nbTotalObservation_, _staleTable_, and gum::IncrementalGraphLearner< AttributeSelection, false >::root_.

Here is the call graph for this function:

◆ ~ITI()

template<TESTNAME AttributeSelection, bool isScalar>

gum::ITI< AttributeSelection, isScalar >::~ITI ( )

override

Default destructor.

Definition at line 353 of file iti_tpl.h.

                                            {
    GUM_DESTRUCTOR(ITI);
    ;
  }

References ITI().

Here is the call graph for this function:

Member Function Documentation

◆ _assumeValue_()

void gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_assumeValue_ ( const Observation * obs )

privateinherited

Get value assumed by studied variable for current observation.

Definition at line 153 of file incrementalGraphLearner_tpl.h.

                              {
    _assumeValue_(obs, Int2Type< isScalar >());
  }

References leafDatabase_, and updateNodeWithObservation_().

Here is the call graph for this function:

◆ _branchObs_()

Idx gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_branchObs_	(	const Observation *	obs,
		const DiscreteVariable *	var )

privateinherited

Seek modality assumed in obs for given var.

Definition at line 164 of file incrementalGraphLearner_tpl.h.

                                   {
    return _branchObs_(obs, var, Int2Type< isScalar >());
  }

Referenced by addObservation().

Here is the caller graph for this function:

◆ _clearValue_()

void gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_clearValue_ ( )

privateinherited

Template function dispatcher.

Definition at line 117 of file incrementalGraphLearner_tpl.h.

                                                                             {
    _clearValue_(Int2Type< isScalar >());
  }

References IncrementalGraphLearner(), _clearValue_(), and leafDatabase_.

Referenced by _clearValue_().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ _insertNodeInFunctionGraph_()

template<TESTNAME AttributeSelection, bool isScalar>

NodeId gum::ITI< AttributeSelection, isScalar >::_insertNodeInFunctionGraph_ ( NodeId currentNodeId )

private

Inserts an internal node in the target.

Parameters

src	the source node in internal graph

Returns: the mathcing node id in the target

Parameters

the	source node in internal graph

Returns: the mathcing node id in the target

Definition at line 280 of file iti_tpl.h.

                                                                                              {
    if (this->nodeVarMap_[currentNodeId] == this->value_) {
      NodeId nody = _insertTerminalNode_(currentNodeId);
      return nody;
    }
 
    if (!this->target_->variablesSequence().exists(this->nodeVarMap_[currentNodeId])) {
      this->target_->add(*(this->nodeVarMap_[currentNodeId]));
    }
 
    NodeId nody = this->target_->manager()->addInternalNode(this->nodeVarMap_[currentNodeId]);
    for (Idx moda = 0; moda < this->nodeVarMap_[currentNodeId]->domainSize(); ++moda) {
      NodeId son = this->_insertNodeInFunctionGraph_(this->nodeSonsMap_[currentNodeId][moda]);
      this->target_->manager()->setSon(nody, moda, son);
    }
 
    return nody;
  }

References _insertNodeInFunctionGraph_(), _insertTerminalNode_(), gum::IncrementalGraphLearner< AttributeSelection, false >::nodeSonsMap_, gum::IncrementalGraphLearner< AttributeSelection, false >::nodeVarMap_, gum::IncrementalGraphLearner< AttributeSelection, false >::target_, and gum::IncrementalGraphLearner< AttributeSelection, false >::value_.

Referenced by _insertNodeInFunctionGraph_(), and updateFunctionGraph().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ _insertTerminalNode_() [1/3]

template<TESTNAME AttributeSelection, bool isScalar>

NodeId gum::ITI< AttributeSelection, isScalar >::_insertTerminalNode_ ( NodeId src )

private

Insert a terminal node in the target.

This function is a dispatcher that will call the right function according to the value of the template isScalar

Parameters

src	the source node in the learned graph

Returns: the matching node in the target

Definition at line 359 of file iti_tpl.h.

                                                                             {
    return _insertTerminalNode_(src, Int2Type< isScalar >());
  }

References _insertTerminalNode_().

Referenced by _insertNodeInFunctionGraph_(), and _insertTerminalNode_().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ _insertTerminalNode_() [2/3]

template<TESTNAME AttributeSelection, bool isScalar>

NodeId gum::ITI< AttributeSelection, isScalar >::_insertTerminalNode_	(	NodeId	currentNodeId,
		Int2Type< false >	)

private

Insert a terminal node in the target.

This function is called if we're learning the behaviour of a variable. Inserts then this variable and the relevant value beneath into target.

Parameters

src	the source node in the learned graph

Returns: the matching node in the target

This function is called if we're learning a real value function. Inserts then a single value in target.

Parameters

the	source node in the learned graph

Returns: the matching node in the target

Definition at line 309 of file iti_tpl.h.

                                                                                      {
    if (!this->target_->variablesSequence().exists(this->value_))
      this->target_->add(*(this->value_));
 
    Size tot = this->nodeId2Database_[currentNodeId]->nbObservation();
    if (tot == Size(0)) return this->target_->manager()->addTerminalNode(0.0);
 
    NodeId* sonsMap
        = static_cast< NodeId* >(SOA_ALLOCATE(sizeof(NodeId) * this->value_->domainSize()));
    for (Idx modality = 0; modality < this->value_->domainSize(); ++modality) {
      double newVal = 0.0;
      newVal = (double)this->nodeId2Database_[currentNodeId]->effectif(modality) / (double)tot;
      sonsMap[modality] = this->target_->manager()->addTerminalNode(newVal);
    }
    NodeId nody = this->target_->manager()->addInternalNode(this->value_, sonsMap);
    return nody;
  }

References gum::IncrementalGraphLearner< AttributeSelection, false >::nodeId2Database_, SOA_ALLOCATE, gum::IncrementalGraphLearner< AttributeSelection, false >::target_, and gum::IncrementalGraphLearner< AttributeSelection, false >::value_.

◆ _insertTerminalNode_() [3/3]

template<TESTNAME AttributeSelection, bool isScalar>

NodeId gum::ITI< AttributeSelection, isScalar >::_insertTerminalNode_	(	NodeId	currentNodeId,
		Int2Type< true >	)

private

Insert a terminal node in the target.

This function is called if we're learning a real value function. Inserts then a single value in target.

Parameters

src	the source node in the learned graph

Returns: the matching node in the target

This function is called if we're learning the behaviour of a variable. Inserts then this variable and the relevant value beneath into target.

Parameters

the	source node in the learned graph

Returns: the matching node in the target

Definition at line 338 of file iti_tpl.h.

                                                                                     {
    double value = 0.0;
    for (auto valIter = this->nodeId2Database_[currentNodeId]->cbeginValues();
         valIter != this->nodeId2Database_[currentNodeId]->cendValues();
         ++valIter) {
      value += (double)valIter.key() * valIter.val();
    }
    if (this->nodeId2Database_[currentNodeId]->nbObservation())
      value /= (double)this->nodeId2Database_[currentNodeId]->nbObservation();
    NodeId nody = this->target_->manager()->addTerminalNode(value);
    return nody;
  }

References gum::IncrementalGraphLearner< AttributeSelection, false >::nodeId2Database_, and gum::IncrementalGraphLearner< AttributeSelection, false >::target_.

◆ addObservation()

template<TESTNAME AttributeSelection, bool isScalar>

void gum::ITI< AttributeSelection, isScalar >::addObservation ( const Observation * obs )

overridevirtual

Inserts a new observation.

Parameters

obs	the new observation to learn
the	new observation to learn

Reimplemented from gum::IncrementalGraphLearner< AttributeSelection, false >.

Definition at line 131 of file iti_tpl.h.

                                                                                 {
    _nbTotalObservation_++;
    IncrementalGraphLearner< AttributeSelection, isScalar >::addObservation(obs);
  }

References _nbTotalObservation_, and gum::IncrementalGraphLearner< AttributeSelection, isScalar >::addObservation().

Here is the call graph for this function:

◆ chgNodeBoundVar_()

template<TESTNAME AttributeSelection, bool isScalar>

void gum::ITI< AttributeSelection, isScalar >::chgNodeBoundVar_	(	NodeId	currentNodeId,
		const DiscreteVariable *	desiredVar )

overrideprotectedvirtual

Changes the associated variable of a node.

Parameters

chgedNodeId	: the node to change
desiredVar	: its new associated variable

Reimplemented from gum::IncrementalGraphLearner< AttributeSelection, false >.

Definition at line 238 of file iti_tpl.h.

                                                                                                 {
    if (this->nodeVarMap_[currentNodeId] != desiredVar) {
      _staleTable_[currentNodeId] = true;
      IncrementalGraphLearner< AttributeSelection, isScalar >::chgNodeBoundVar_(currentNodeId,
                                                                                desiredVar);
    }
  }

References _staleTable_, gum::IncrementalGraphLearner< AttributeSelection, isScalar >::chgNodeBoundVar_(), and gum::IncrementalGraphLearner< AttributeSelection, false >::nodeVarMap_.

Here is the call graph for this function:

◆ convertNode2Leaf_()

void gum::IncrementalGraphLearner< AttributeSelection, isScalar >::convertNode2Leaf_ ( NodeId currentNodeId )

protectedvirtualinherited

Turns the given node into a leaf if not already so.

Definition at line 218 of file incrementalGraphLearner_tpl.h.

                            {
    if (nodeVarMap_[currentNodeId] != value_) {
      leafDatabase_.insert(currentNodeId, new Set< const Observation* >());
 
      // Resolving tensor sons issue
      for (Idx modality = 0; modality < nodeVarMap_[currentNodeId]->domainSize(); ++modality) {
        NodeId sonId = nodeSonsMap_[currentNodeId][modality];
        convertNode2Leaf_(sonId);
        (*leafDatabase_[currentNodeId]) = (*leafDatabase_[currentNodeId]) + *(leafDatabase_[sonId]);
        removeNode_(sonId);
      }
 
      SOA_DEALLOCATE(nodeSonsMap_[currentNodeId],
                     sizeof(NodeId) * nodeVarMap_[currentNodeId]->domainSize());
      nodeSonsMap_.erase(currentNodeId);
 
      chgNodeBoundVar_(currentNodeId, value_);
    }
  }

Referenced by transpose_(), and gum::IMDDI< AttributeSelection, isScalar >::updateGraph().

Here is the caller graph for this function:

◆ insertInternalNode_()

NodeId gum::IncrementalGraphLearner< AttributeSelection, isScalar >::insertInternalNode_	(	NodeDatabase< AttributeSelection, isScalar > *	nDB,
		const DiscreteVariable *	boundVar,
		NodeId *	sonsMap )

protectedvirtualinherited

inserts a new internal node in internal graph

Parameters

nDB	: the associated database
boundVar	: the associated variable
sonsMap	: a table giving node's sons node

Returns: the newly created node's id

Definition at line 246 of file incrementalGraphLearner_tpl.h.

                                                             {
    NodeId newNodeId = this->insertNode_(nDB, boundVar);
    nodeSonsMap_.insert(newNodeId, sonsMap);
    return newNodeId;
  }

References nodeVarMap_, and value_.

◆ insertLeafNode_()

NodeId gum::IncrementalGraphLearner< AttributeSelection, isScalar >::insertLeafNode_	(	NodeDatabase< AttributeSelection, isScalar > *	nDB,
		const DiscreteVariable *	boundVar,
		Set< const Observation * > *	obsSet )

protectedvirtualinherited

inserts a new leaf node in internal graohs

Parameters

nDB	: the associated database
boundVar	: the associated variable
obsSet	: the set of observation this leaf retains

Returns: the newly created node's id

Reimplemented in gum::IMDDI< AttributeSelection, isScalar >.

Definition at line 259 of file incrementalGraphLearner_tpl.h.

                                                            {
    NodeId newNodeId = this->insertNode_(nDB, boundVar);
    leafDatabase_.insert(newNodeId, obsSet);
    return newNodeId;
  }

References setOfVars_, and value_.

Referenced by removeNode_().

Here is the caller graph for this function:

◆ insertNode_()

template<TESTNAME AttributeSelection, bool isScalar>

NodeId gum::ITI< AttributeSelection, isScalar >::insertNode_	(	NodeDatabase< AttributeSelection, isScalar > *	nDB,
		const DiscreteVariable *	boundVar )

overrideprotectedvirtual

inserts a new node in internal graph

inserts a new node in internal graohs

Parameters

nDB	: the associated database
boundVar	: the associated variable

Returns: the newly created node's id

Reimplemented from gum::IncrementalGraphLearner< AttributeSelection, false >.

Definition at line 222 of file iti_tpl.h.

                                                              {
    NodeId n = IncrementalGraphLearner< AttributeSelection, isScalar >::insertNode_(nDB, boundVar);
    _staleTable_.insert(n, true);
    return n;
  }

References _staleTable_, and gum::IncrementalGraphLearner< AttributeSelection, isScalar >::insertNode_().

Here is the call graph for this function:

◆ insertSetOfVars()

void gum::IncrementalGraphLearner< AttributeSelection, isScalar >::insertSetOfVars ( MultiDimFunctionGraph< double > * ret ) const

overridevirtualinherited

Implements gum::IVisitableGraphLearner.

Definition at line 315 of file incrementalGraphLearner_tpl.h.

                                                  {
    for (SetIteratorSafe< const DiscreteVariable* > varIter = setOfVars_.beginSafe();
         varIter != setOfVars_.endSafe();
         ++varIter)
      ret->add(**varIter);
  }

References nodeId2Database_, nodeSonsMap_, nodeVarMap_, setOfVars_, and value_.

◆ insertSetOfVars_()

template<TESTNAME AttributeSelection, bool isScalar>

void gum::ITI< AttributeSelection, isScalar >::insertSetOfVars_ ( MultiDimFunctionGraph< double > * ret )

protected

insertSetOfVars_

Parameters

ret

Definition at line 364 of file iti_tpl.h.

                                                                                                 {
    for (SetIteratorSafe< const DiscreteVariable* > varIter = this->setOfVars_.beginSafe();
         varIter != this->setOfVars_.endSafe();
         ++varIter)
      ret->add(**varIter);
  }

References gum::MultiDimFunctionGraph< GUM_ELEMENT, TerminalNodePolicy >::add(), and gum::IncrementalGraphLearner< AttributeSelection, false >::setOfVars_.

Here is the call graph for this function:

◆ isTerminal()

bool gum::IncrementalGraphLearner< AttributeSelection, isScalar >::isTerminal ( NodeId ni ) const

overridevirtualinherited

Implements gum::IVisitableGraphLearner.

Definition at line 307 of file incrementalGraphLearner_tpl.h.

                                                                                          {
    return !this->nodeSonsMap_.exists(ni);
  }

◆ nodeNbObservation()

Idx gum::IncrementalGraphLearner< AttributeSelection, isScalar >::nodeNbObservation ( NodeId ni ) const

overridevirtualinherited

Implements gum::IVisitableGraphLearner.

Definition at line 313 of file incrementalGraphLearner_tpl.h.

                                                                                                {
    return this->nodeId2Database_[ni]->nbObservation();
  }

◆ nodeSon()

NodeId gum::IncrementalGraphLearner< AttributeSelection, isScalar >::nodeSon	(	NodeId	ni,
		Idx	modality ) const

overridevirtualinherited

Implements gum::IVisitableGraphLearner.

Definition at line 311 of file incrementalGraphLearner_tpl.h.

                                                                                                 {
    return this->nodeSonsMap_[ni][modality];
  }

◆ nodeVar()

const DiscreteVariable * gum::IncrementalGraphLearner< AttributeSelection, isScalar >::nodeVar ( NodeId ni ) const

overridevirtualinherited

Implements gum::IVisitableGraphLearner.

Definition at line 309 of file incrementalGraphLearner_tpl.h.

                                                                                      {
    return this->nodeVarMap_[ni];
  }

◆ removeNode_()

template<TESTNAME AttributeSelection, bool isScalar>

void gum::ITI< AttributeSelection, isScalar >::removeNode_ ( NodeId currentNodeId )

overrideprotectedvirtual

Removes a node from the internal graph.

Parameters

removedNodeId : the node to remove

Reimplemented from gum::IncrementalGraphLearner< AttributeSelection, false >.

Definition at line 254 of file iti_tpl.h.

                                                                            {
    IncrementalGraphLearner< AttributeSelection, isScalar >::removeNode_(currentNodeId);
    _staleTable_.erase(currentNodeId);
  }

References _staleTable_, and gum::IncrementalGraphLearner< AttributeSelection, isScalar >::removeNode_().

Here is the call graph for this function:

◆ root()

NodeId gum::IncrementalGraphLearner< AttributeSelection, isScalar >::root ( ) const

overridevirtualinherited

Implements gum::IVisitableGraphLearner.

Definition at line 305 of file incrementalGraphLearner_tpl.h.

                                                                             {
    return this->root_;
  }

◆ size()

Size gum::IncrementalGraphLearner< AttributeSelection, isScalar >::size ( )

inherited

Definition at line 303 of file incrementalGraphLearner_tpl.h.

                                                                     {
    return nodeVarMap_.size();
  }

References nodeSonsMap_, nodeVarMap_, and transpose_().

Here is the call graph for this function:

◆ transpose_()

void gum::IncrementalGraphLearner< AttributeSelection, isScalar >::transpose_	(	NodeId	currentNodeId,
		const DiscreteVariable *	desiredVar )

protectedvirtualinherited

Installs given variable to the given node, ensuring that the variable is not present in its subtree.

Definition at line 224 of file incrementalGraphLearner_tpl.h.

                                          {
    // **************************************************************************************
    // Si le noeud courant contient déjà la variable qu'on souhaite lui amener
    // Il n'y a rien à faire
    if (nodeVarMap_[currentNodeId] == desiredVar) { return; }
 
    // **************************************************************************************
    // Si le noeud courant est terminal,
    // Il faut artificiellement insérer un noeud liant à la variable
    if (nodeVarMap_[currentNodeId] == value_) {
      // We turned this leaf into an internal node.
      // This mean that we'll need to install children leaves for each value of
      // desiredVar
 
      // First We must prepare these new leaves NodeDatabases and Sets<const
      // Observation*>
      NodeDatabase< AttributeSelection, isScalar >** dbMap
          = static_cast< NodeDatabase< AttributeSelection, isScalar >** >(SOA_ALLOCATE(
              sizeof(NodeDatabase< AttributeSelection, isScalar >*) * desiredVar->domainSize()));
      Set< const Observation* >** obsetMap = static_cast< Set< const Observation* >** >(
          SOA_ALLOCATE(sizeof(Set< const Observation* >*) * desiredVar->domainSize()));
      for (Idx modality = 0; modality < desiredVar->domainSize(); ++modality) {
        dbMap[modality]    = new NodeDatabase< AttributeSelection, isScalar >(&setOfVars_, value_);
        obsetMap[modality] = new Set< const Observation* >();
      }
      for (SetIteratorSafe< const Observation* > obsIter
           = leafDatabase_[currentNodeId]->beginSafe();
           leafDatabase_[currentNodeId]->endSafe() != obsIter;
           ++obsIter) {
        dbMap[_branchObs_(*obsIter, desiredVar)]->addObservation(*obsIter);
        obsetMap[_branchObs_(*obsIter, desiredVar)]->insert(*obsIter);
      }
 
      // Then we can install each new leaves (and put in place the sonsMap)
      NodeId* sonsMap
          = static_cast< NodeId* >(SOA_ALLOCATE(sizeof(NodeId) * desiredVar->domainSize()));
      for (Idx modality = 0; modality < desiredVar->domainSize(); ++modality)
        sonsMap[modality] = insertLeafNode_(dbMap[modality], value_, obsetMap[modality]);
 
      // Some necessary clean up
      SOA_DEALLOCATE(dbMap,
                     sizeof(NodeDatabase< AttributeSelection, isScalar >*)
                         * desiredVar->domainSize());
      SOA_DEALLOCATE(obsetMap, sizeof(Set< const Observation* >*) * desiredVar->domainSize());
 
      // And finally we can turn the node into an internal node associated to
      // desiredVar
      chgNodeBoundVar_(currentNodeId, desiredVar);
      nodeSonsMap_.insert(currentNodeId, sonsMap);
 
      return;
    }
 
    // *************************************************************************************
    // Remains the general case where currentNodeId is an internal node.
 
    // First we ensure that children node use desiredVar as variable
    for (Idx modality = 0; modality < nodeVarMap_[currentNodeId]->domainSize(); ++modality)
      transpose_(nodeSonsMap_[currentNodeId][modality], desiredVar);
 
    //        Sequence<NodeDatabase<AttributeSelection, isScalar>*>
    //        sonsNodeDatabase =
    //        nodeId2Database_[currentNodeId]->splitOnVar(desiredVar);
    NodeId* sonsMap
        = static_cast< NodeId* >(SOA_ALLOCATE(sizeof(NodeId) * desiredVar->domainSize()));
 
    // Then we create the new mapping
    for (Idx desiredVarModality = 0; desiredVarModality < desiredVar->domainSize();
         ++desiredVarModality) {
      NodeId* grandSonsMap = static_cast< NodeId* >(
          SOA_ALLOCATE(sizeof(NodeId) * nodeVarMap_[currentNodeId]->domainSize()));
      NodeDatabase< AttributeSelection, isScalar >* sonDB
          = new NodeDatabase< AttributeSelection, isScalar >(&setOfVars_, value_);
      for (Idx currentVarModality = 0;
           currentVarModality < nodeVarMap_[currentNodeId]->domainSize();
           ++currentVarModality) {
        grandSonsMap[currentVarModality]
            = nodeSonsMap_[nodeSonsMap_[currentNodeId][currentVarModality]][desiredVarModality];
        sonDB->operator+=((*nodeId2Database_[grandSonsMap[currentVarModality]]));
      }
 
      sonsMap[desiredVarModality]
          = insertInternalNode_(sonDB, nodeVarMap_[currentNodeId], grandSonsMap);
    }
 
    // Finally we clean the old remaining nodes
    for (Idx currentVarModality = 0; currentVarModality < nodeVarMap_[currentNodeId]->domainSize();
         ++currentVarModality) {
      removeNode_(nodeSonsMap_[currentNodeId][currentVarModality]);
    }
 
    // We suppress the old sons map and remap to the new one
    SOA_DEALLOCATE(nodeSonsMap_[currentNodeId],
                   sizeof(NodeId) * nodeVarMap_[currentNodeId]->domainSize());
    nodeSonsMap_[currentNodeId] = sonsMap;
 
    chgNodeBoundVar_(currentNodeId, desiredVar);
  }

References convertNode2Leaf_(), leafDatabase_, nodeSonsMap_, nodeVarMap_, and removeNode_().

Referenced by gum::IMDDI< AttributeSelection, isScalar >::_updateNodeSet_(), and size().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ updateFunctionGraph()

template<TESTNAME AttributeSelection, bool isScalar>

void gum::ITI< AttributeSelection, isScalar >::updateFunctionGraph ( )

overridevirtual

Updates target to currently learned graph structure.

Implements gum::IncrementalGraphLearner< AttributeSelection, false >.

Definition at line 267 of file iti_tpl.h.

                                                                {
    this->target_->clear();
    this->target_->manager()->setRootNode(this->_insertNodeInFunctionGraph_(this->root_));
  }

References _insertNodeInFunctionGraph_(), gum::IncrementalGraphLearner< AttributeSelection, false >::root_, and gum::IncrementalGraphLearner< AttributeSelection, false >::target_.

Here is the call graph for this function:

◆ updateGraph()

template<TESTNAME AttributeSelection, bool isScalar>

void gum::ITI< AttributeSelection, isScalar >::updateGraph ( )

overridevirtual

Updates the internal graph after a new observation has been added.

Implements gum::IncrementalGraphLearner< AttributeSelection, false >.

Definition at line 161 of file iti_tpl.h.

                                                        {
    std::vector< NodeId > filo;
    filo.push_back(this->root_);
    HashTable< NodeId, gum::VariableSet* > tensorVars;
    tensorVars.insert(this->root_, new gum::VariableSet(this->setOfVars_));
 
 
    while (!filo.empty()) {
      NodeId currentNodeId = filo.back();
      filo.pop_back();
 
      // First we look for the best var to install on the node
      double           bestValue = _attributeSelectionThreshold_;
      gum::VariableSet bestVars;
 
      for (auto varIter = tensorVars[currentNodeId]->cbeginSafe();
           varIter != tensorVars[currentNodeId]->cendSafe();
           ++varIter)
        if (this->nodeId2Database_[currentNodeId]->isTestRelevant(*varIter)) {
          double varValue = this->nodeId2Database_[currentNodeId]->testValue(*varIter);
          if (varValue >= bestValue) {
            if (varValue > bestValue) {
              bestValue = varValue;
              bestVars.clear();
            }
            bestVars.insert(*varIter);
          }
        }
 
      // Then We installed Variable a test on that node
      this->updateNode_(currentNodeId, bestVars);
 
      // The we move on the children if needed
      if (this->nodeVarMap_[currentNodeId] != this->value_) {
        for (Idx moda = 0; moda < this->nodeVarMap_[currentNodeId]->domainSize(); moda++) {
          gum::VariableSet* itsTensorVars = new gum::VariableSet(*tensorVars[currentNodeId]);
          itsTensorVars->erase(this->nodeVarMap_[currentNodeId]);
          NodeId sonId = this->nodeSonsMap_[currentNodeId][moda];
          if (_staleTable_[sonId]) {
            filo.push_back(sonId);
            tensorVars.insert(sonId, itsTensorVars);
          }
        }
      }
    }
 
    for (HashTableIteratorSafe< NodeId, gum::VariableSet* > nodeIter = tensorVars.beginSafe();
         nodeIter != tensorVars.endSafe();
         ++nodeIter)
      delete nodeIter.val();
  }

References _attributeSelectionThreshold_, _staleTable_, gum::HashTable< Key, Val >::beginSafe(), gum::HashTable< Key, Val >::cendSafe(), gum::Set< Key >::clear(), gum::HashTable< Key, Val >::endSafe(), gum::Set< Key >::erase(), gum::HashTable< Key, Val >::insert(), gum::Set< Key >::insert(), gum::IncrementalGraphLearner< AttributeSelection, false >::nodeId2Database_, gum::IncrementalGraphLearner< AttributeSelection, false >::nodeSonsMap_, gum::IncrementalGraphLearner< AttributeSelection, false >::nodeVarMap_, gum::IncrementalGraphLearner< AttributeSelection, false >::root_, gum::IncrementalGraphLearner< AttributeSelection, false >::setOfVars_, gum::IncrementalGraphLearner< AttributeSelection, false >::updateNode_(), and gum::IncrementalGraphLearner< AttributeSelection, false >::value_.

Here is the call graph for this function:

◆ updateNode_()

void gum::IncrementalGraphLearner< AttributeSelection, isScalar >::updateNode_	(	NodeId	updatedNode,
		gum::VariableSet &	varsOfInterest )

protectedinherited

From the given sets of node, selects randomly one and installs it on given node.

Chechks of course if node's current variable is not in that set first.

Parameters

nody	: the node we update
bestVars	: the set of interessting vars to be installed here

Chechks of course if node's current variable is not in that set first.

Parameters

nody	: the node we update
bestVar	: the set of interessting vars to be installed here

Definition at line 213 of file incrementalGraphLearner_tpl.h.

                                      {
    // If this node has no interesting variable, we turn it into a leaf
    if (varsOfInterest.empty()) {
      convertNode2Leaf_(updatedNode);
      return;
    }
 
    // If this node has already one of the best variable intalled as test, we
    // move on
    if (nodeVarMap_.exists(updatedNode) && varsOfInterest.exists(nodeVarMap_[updatedNode])) {
      return;
    }
 
    // In any other case we have to install variable as best test
    Idx randy = randomValue(varsOfInterest.size()), basc = 0;
    SetConstIteratorSafe< const DiscreteVariable* > varIter;
    for (varIter = varsOfInterest.cbeginSafe(), basc = 0;
         varIter != varsOfInterest.cendSafe() && basc < randy;
         ++varIter, basc++)
      ;
 
    transpose_(updatedNode, *varIter);
  }

Referenced by gum::ITI< AttributeSelection, isScalar >::updateGraph().

Here is the caller graph for this function:

◆ updateNodeWithObservation_()

template<TESTNAME AttributeSelection, bool isScalar>

void gum::ITI< AttributeSelection, isScalar >::updateNodeWithObservation_	(	const Observation *	newObs,
		NodeId	currentNodeId )

overrideprotectedvirtual

Will update internal graph's NodeDatabase of given node with the new observation.

Parameters

newObs
currentNodeId

Reimplemented from gum::IncrementalGraphLearner< AttributeSelection, false >.

Definition at line 145 of file iti_tpl.h.

                                                                                             {
    IncrementalGraphLearner< AttributeSelection, isScalar >::updateNodeWithObservation_(
        newObs,
        currentNodeId);
    _staleTable_[currentNodeId] = true;
  }

References _staleTable_, and gum::IncrementalGraphLearner< AttributeSelection, isScalar >::updateNodeWithObservation_().

Here is the call graph for this function:

◆ updateVar()

void gum::IncrementalGraphLearner< AttributeSelection, isScalar >::updateVar ( const DiscreteVariable * var )

virtualinherited

If a new modality appears to exists for given variable, call this method to turn every associated node to this variable into leaf. Graph has then indeed to be revised.

Definition at line 195 of file incrementalGraphLearner_tpl.h.

                                   {
    Link< NodeId >* nodIter = var2Node_[var]->list();
    Link< NodeId >* nni     = nullptr;
    while (nodIter) {
      nni = nodIter->nextLink();
      convertNode2Leaf_(nodIter->element());
      nodIter = nni;
    }
  }

Member Data Documentation

◆ _attributeSelectionThreshold_

template<TESTNAME AttributeSelection, bool isScalar = false>

double gum::ITI< AttributeSelection, isScalar >::_attributeSelectionThreshold_

private

The threshold above which we consider variables to be dependant.

Definition at line 272 of file iti.h.

Referenced by ITI(), ITI(), and updateGraph().

◆ _nbTotalObservation_

template<TESTNAME AttributeSelection, bool isScalar = false>

Idx gum::ITI< AttributeSelection, isScalar >::_nbTotalObservation_

private

The total number of observation added to this tree.

Definition at line 269 of file iti.h.

Referenced by ITI(), ITI(), and addObservation().

◆ _staleTable_

template<TESTNAME AttributeSelection, bool isScalar = false>

HashTable< NodeId, bool > gum::ITI< AttributeSelection, isScalar >::_staleTable_

private

Hashtable indicating if given node has been modified (upon receiving new exemple or through a transpose) The aim is not if we have revise the installed variable on that node.

Definition at line 266 of file iti.h.

Referenced by ITI(), ITI(), chgNodeBoundVar_(), insertNode_(), removeNode_(), updateGraph(), and updateNodeWithObservation_().

◆ leafDatabase_

HashTable< NodeId, Set< const Observation* >* > gum::IncrementalGraphLearner< AttributeSelection, isScalar >::leafDatabase_

protectedinherited

This hashtable binds to every leaf an associated set of all hte observations compatible with it.

Definition at line 364 of file incrementalGraphLearner.h.

Referenced by _assumeValue_(), _clearValue_(), and transpose_().

◆ model_

NodeGraphPart gum::IncrementalGraphLearner< AttributeSelection, isScalar >::model_

protectedinherited

The source of nodeId.

Definition at line 330 of file incrementalGraphLearner.h.

Referenced by gum::IMDDI< AttributeSelection, isScalar >::IMDDI(), and gum::IMDDI< AttributeSelection, isScalar >::IMDDI().

◆ needUpdate_

bool gum::IncrementalGraphLearner< AttributeSelection, isScalar >::needUpdate_

protectedinherited

Definition at line 377 of file incrementalGraphLearner.h.

Referenced by gum::IMDDI< AttributeSelection, isScalar >::updateFunctionGraph().

◆ nodeId2Database_

HashTable< NodeId, NodeDatabase< AttributeSelection, isScalar >* > gum::IncrementalGraphLearner< AttributeSelection, isScalar >::nodeId2Database_

protectedinherited

This hashtable binds every node to an associated NodeDatabase which handles every observation that concerns that node.

Definition at line 358 of file incrementalGraphLearner.h.

Referenced by gum::IMDDI< AttributeSelection, isScalar >::_addLeaf_(), gum::IMDDI< AttributeSelection, isScalar >::_downdateScore_(), gum::ITI< AttributeSelection, isScalar >::_insertTerminalNode_(), gum::ITI< AttributeSelection, isScalar >::_insertTerminalNode_(), gum::IMDDI< AttributeSelection, isScalar >::_updateNodeSet_(), gum::IMDDI< AttributeSelection, isScalar >::_updateScore_(), insertSetOfVars(), and gum::ITI< AttributeSelection, isScalar >::updateGraph().

◆ nodeSonsMap_

HashTable< NodeId, NodeId* > gum::IncrementalGraphLearner< AttributeSelection, isScalar >::nodeSonsMap_

protectedinherited

A table giving for any node a table mapping to its son idx is the modality of associated variable.

Definition at line 346 of file incrementalGraphLearner.h.

Referenced by ~IncrementalGraphLearner(), gum::ITI< AttributeSelection, isScalar >::_insertNodeInFunctionGraph_(), gum::IMDDI< AttributeSelection, isScalar >::_rebuildFunctionGraph_(), gum::IMDDI< AttributeSelection, isScalar >::_updateNodeSet_(), addObservation(), insertSetOfVars(), removeNode_(), size(), transpose_(), and gum::ITI< AttributeSelection, isScalar >::updateGraph().

◆ nodeVarMap_

HashTable< NodeId, const DiscreteVariable* > gum::IncrementalGraphLearner< AttributeSelection, isScalar >::nodeVarMap_

protectedinherited

Gives for any node its associated variable.

Definition at line 340 of file incrementalGraphLearner.h.

Referenced by ~IncrementalGraphLearner(), gum::ITI< AttributeSelection, isScalar >::_insertNodeInFunctionGraph_(), gum::IMDDI< AttributeSelection, isScalar >::_updateNodeSet_(), addObservation(), gum::IMDDI< AttributeSelection, isScalar >::chgNodeBoundVar_(), gum::ITI< AttributeSelection, isScalar >::chgNodeBoundVar_(), insertInternalNode_(), insertSetOfVars(), gum::IMDDI< AttributeSelection, isScalar >::removeNode_(), size(), transpose_(), gum::ITI< AttributeSelection, isScalar >::updateGraph(), and gum::IMDDI< AttributeSelection, isScalar >::updateNodeWithObservation_().

◆ root_

NodeId gum::IncrementalGraphLearner< AttributeSelection, isScalar >::root_

protectedinherited

The root of the ordered tree.

Definition at line 335 of file incrementalGraphLearner.h.

Referenced by gum::IMDDI< AttributeSelection, isScalar >::IMDDI(), gum::IMDDI< AttributeSelection, isScalar >::IMDDI(), gum::ITI< AttributeSelection, isScalar >::ITI(), gum::ITI< AttributeSelection, isScalar >::ITI(), gum::IMDDI< AttributeSelection, isScalar >::_rebuildFunctionGraph_(), gum::ITI< AttributeSelection, isScalar >::updateFunctionGraph(), gum::IMDDI< AttributeSelection, isScalar >::updateGraph(), and gum::ITI< AttributeSelection, isScalar >::updateGraph().

◆ setOfVars_

gum::VariableSet gum::IncrementalGraphLearner< AttributeSelection, isScalar >::setOfVars_

protectedinherited

Definition at line 372 of file incrementalGraphLearner.h.

Referenced by insertLeafNode_(), insertSetOfVars(), gum::ITI< AttributeSelection, isScalar >::insertSetOfVars_(), gum::IMDDI< AttributeSelection, isScalar >::updateGraph(), and gum::ITI< AttributeSelection, isScalar >::updateGraph().

◆ target_

MultiDimFunctionGraph< double >* gum::IncrementalGraphLearner< AttributeSelection, isScalar >::target_

protectedinherited

The final diagram we're building.

Definition at line 370 of file incrementalGraphLearner.h.

Referenced by gum::IMDDI< AttributeSelection, isScalar >::_insertLeafInFunctionGraph_(), gum::IMDDI< AttributeSelection, isScalar >::_insertLeafInFunctionGraph_(), gum::ITI< AttributeSelection, isScalar >::_insertNodeInFunctionGraph_(), gum::ITI< AttributeSelection, isScalar >::_insertTerminalNode_(), gum::ITI< AttributeSelection, isScalar >::_insertTerminalNode_(), gum::IMDDI< AttributeSelection, isScalar >::_rebuildFunctionGraph_(), and gum::ITI< AttributeSelection, isScalar >::updateFunctionGraph().

◆ value_

const DiscreteVariable* gum::IncrementalGraphLearner< AttributeSelection, isScalar >::value_

protectedinherited

Definition at line 374 of file incrementalGraphLearner.h.

◆ valueAssumed_

Sequence< ValueType > gum::IncrementalGraphLearner< AttributeSelection, isScalar >::valueAssumed_

protectedinherited

Definition at line 375 of file incrementalGraphLearner.h.

Referenced by gum::IMDDI< AttributeSelection, isScalar >::_addLeaf_(), and gum::IMDDI< AttributeSelection, isScalar >::_insertLeafInFunctionGraph_().

◆ var2Node_

HashTable< const DiscreteVariable*, LinkedList< NodeId >* > gum::IncrementalGraphLearner< AttributeSelection, isScalar >::var2Node_

protectedinherited

Associates to any variable the list of all nodes associated to this variable.

Definition at line 352 of file incrementalGraphLearner.h.

Referenced by ~IncrementalGraphLearner(), and gum::IMDDI< AttributeSelection, isScalar >::_rebuildFunctionGraph_().

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

agrum/FMDP/learning/datastructure/iti.h
agrum/FMDP/learning/datastructure/iti_tpl.h

Public Member Functions

Protected Member Functions

Protected Attributes

Private Types

Private Member Functions

Private Attributes

New Observation insertion methods

Graph Structure update methods

Function Graph Updating methods

Graph Structure update methods

Detailed Description

Member Typedef Documentation

◆ ValueType

Constructor & Destructor Documentation

◆ ITI() [1/2]

◆ ITI() [2/2]

◆ ~ITI()

Member Function Documentation

◆ _assumeValue_()

◆ _branchObs_()

◆ _clearValue_()

◆ _insertNodeInFunctionGraph_()

◆ _insertTerminalNode_() [1/3]

◆ _insertTerminalNode_() [2/3]

◆ _insertTerminalNode_() [3/3]

◆ addObservation()

◆ chgNodeBoundVar_()

◆ convertNode2Leaf_()

◆ insertInternalNode_()

◆ insertLeafNode_()

◆ insertNode_()

◆ insertSetOfVars()

◆ insertSetOfVars_()

◆ isTerminal()

◆ nodeNbObservation()

◆ nodeSon()

◆ nodeVar()

◆ removeNode_()

◆ root()

◆ size()

◆ transpose_()

◆ updateFunctionGraph()

◆ updateGraph()

◆ updateNode_()

◆ updateNodeWithObservation_()

◆ updateVar()

Member Data Documentation

◆ _attributeSelectionThreshold_

◆ _nbTotalObservation_

◆ _staleTable_

◆ leafDatabase_

◆ model_

◆ needUpdate_

◆ nodeId2Database_

◆ nodeSonsMap_

◆ nodeVarMap_

◆ root_

◆ setOfVars_

◆ target_

◆ value_

◆ valueAssumed_

◆ var2Node_