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
Size	size ()

Constructor & destructor.
	ITI (MultiDimFunctionGraph< double > target, double attributeSelectionThreshold, Set< const DiscreteVariable > 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. More...

	ITI (MultiDimFunctionGraph< double > target, double attributeSelectionThreshold, Set< const DiscreteVariable > attributeListe)
	ITI constructeur for real functions. More...

	~ITI ()
	Default destructor. More...

Visit Methods
NodeId	root () const

bool	isTerminal (NodeId ni) const

const DiscreteVariable *	nodeVar (NodeId ni) const

NodeId	nodeSon (NodeId ni, Idx modality) const

Idx	nodeNbObservation (NodeId ni) const

virtual void	insertSetOfVars (MultiDimFunctionGraph< double > *ret) const

Protected Attributes
MultiDimFunctionGraph< double > *	_target
	The final diagram we're building. More...

Set< const DiscreteVariable *>	_setOfVars

const DiscreteVariable *	_value

Sequence< ValueType >	_valueAssumed

bool	_needUpdate

Model handling datastructures
NodeGraphPart	_model
	The source of nodeId. More...

NodeId	_root
	The root of the ordered tree. More...

HashTable< NodeId, const DiscreteVariable *>	_nodeVarMap
	Gives for any node its associated variable. More...

HashTable< NodeId, NodeId *>	_nodeSonsMap
	A table giving for any node a table mapping to its son idx is the modality of associated variable. More...

HashTable< const DiscreteVariable , LinkedList< NodeId > >	_var2Node
	Associates to any variable the list of all nodes associated to this variable. More...

HashTable< NodeId, NodeDatabase< AttributeSelection, isScalar > *>	_nodeId2Database
	This hashtable binds every node to an associated NodeDatabase which handles every observation that concerns that node. More...

HashTable< NodeId, Set< const Observation > >	_leafDatabase
	This hashtable binds to every leaf an associated set of all hte observations compatible with it. More...

Protected Member Functions
void	_insertSetOfVars (MultiDimFunctionGraph< double > *ret)
	_insertSetOfVars More...

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. More...

void	_updateNode (NodeId nody, Set< const DiscreteVariable * > &bestVars)
	From the given sets of node, selects randomly one and installs it on given node. More...

virtual void	_convertNode2Leaf (NodeId)
	Turns the given node into a leaf if not already so. More...

virtual void	_transpose (NodeId, const DiscreteVariable *)
	Installs given variable to the given node, ensuring that the variable is not present in its subtree. More...

virtual NodeId	_insertInternalNode (NodeDatabase< AttributeSelection, isScalar > nDB, const DiscreteVariable boundVar, NodeId *sonsMap)
	inserts a new internal node in internal graph More...

virtual NodeId	_insertLeafNode (NodeDatabase< AttributeSelection, isScalar > nDB, const DiscreteVariable boundVar, Set< const Observation * > *obsSet)
	inserts a new leaf node in internal graohs More...

New Observation insertion methods
void	addObservation (const Observation *obs)
	Inserts a new observation. More...

void	_updateNodeWithObservation (const Observation *newObs, NodeId currentNodeId)
	Will update internal graph's NodeDatabase of given node with the new observation. More...

Graph Structure update methods
void	updateGraph ()
	Updates the internal graph after a new observation has been added. More...

NodeId	_insertNode (NodeDatabase< AttributeSelection, isScalar > nDB, const DiscreteVariable boundVar)
	inserts a new node in internal graph More...

void	_chgNodeBoundVar (NodeId chgedNodeId, const DiscreteVariable *desiredVar)
	Changes the associated variable of a node. More...

void	_removeNode (NodeId removedNodeId)
	Removes a node from the internal graph. More...

Function Graph Updating methods
void	updateFunctionGraph ()
	Updates target to currently learned graph structure. More...

NodeId	__insertNodeInFunctionGraph (NodeId src)
	Inserts an internal node in the target. More...

NodeId	__insertTerminalNode (NodeId src)
	Insert a terminal node in the target. More...

NodeId	__insertTerminalNode (NodeId src, Int2Type< true >)
	Insert a terminal node in the target. More...

NodeId	__insertTerminalNode (NodeId src, Int2Type< false >)
	Insert a terminal node in the target. More...

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 62 of file iti.h.

Constructor & Destructor Documentation

◆ ITI() [1/2]

template<TESTNAME AttributeSelection, bool isScalar>

gum::ITI< AttributeSelection, isScalar >::ITI	(	MultiDimFunctionGraph< double > *	target,
		double	attributeSelectionThreshold,
		Set< const DiscreteVariable * >	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 61 of file iti_tpl.h.

References gum::ITI< AttributeSelection, isScalar >::__staleTable, gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_root, and gum::HashTable< Key, Val, Alloc >::insert().

                                                     :
       IncrementalGraphLearner< AttributeSelection, isScalar >(
          target, attributeListe, learnedValue),
       __nbTotalObservation(0),
       __attributeSelectionThreshold(attributeSelectionThreshold) {
     GUM_CONSTRUCTOR(ITI);
     __staleTable.insert(this->_root, false);
   }

Here is the call graph for this function:

◆ ITI() [2/2]

template<TESTNAME AttributeSelection, bool isScalar>

gum::ITI< AttributeSelection, isScalar >::ITI	(	MultiDimFunctionGraph< double > *	target,
		double	attributeSelectionThreshold,
		Set< const DiscreteVariable * >	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 87 of file iti_tpl.h.

References gum::ITI< AttributeSelection, isScalar >::__staleTable, gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_root, and gum::HashTable< Key, Val, Alloc >::insert().

                                                       :
       IncrementalGraphLearner< AttributeSelection, isScalar >(
          target, attributeListe, new LabelizedVariable("Reward", "", 2)),
       __nbTotalObservation(0),
       __attributeSelectionThreshold(attributeSelectionThreshold) {
     GUM_CONSTRUCTOR(ITI);
     __staleTable.insert(this->_root, false);
   }

Here is the call graph for this function:

◆ ~ITI()

template<TESTNAME AttributeSelection, bool isScalar = false>

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

inline

Default destructor.

Definition at line 105 of file iti.h.

105 { GUM_DESTRUCTOR(ITI); }

gum::ITI::ITI

ITI(MultiDimFunctionGraph< double > *target, double attributeSelectionThreshold, Set< const DiscreteVariable * > attributeListe, const DiscreteVariable *learnedValue)

ITI constructor for functions describing the behaviour of one variable according to a set of other va...

Definition: iti_tpl.h:61

Here is the call graph for this function:

Member Function Documentation

◆ __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 276 of file iti_tpl.h.

References gum::ITI< AttributeSelection, isScalar >::__insertTerminalNode(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_nodeSonsMap, gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_nodeVarMap, gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_target, gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_value, gum::MultiDimFunctionGraph< GUM_SCALAR, TerminalNodePolicy >::add(), gum::MultiDimFunctionGraphManager< GUM_SCALAR, TerminalNodePolicy >::addInternalNode(), gum::MultiDimFunctionGraph< GUM_SCALAR, TerminalNodePolicy >::manager(), gum::MultiDimFunctionGraphManager< GUM_SCALAR, TerminalNodePolicy >::setSon(), and gum::MultiDimImplementation< GUM_SCALAR >::variablesSequence().

Referenced by gum::ITI< AttributeSelection, isScalar >::updateFunctionGraph(), and gum::ITI< AttributeSelection, isScalar >::~ITI().

                            {
     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;
   }

Here is the call graph for this function:

Here is the caller graph for this function:

◆ __insertTerminalNode() [1/3]

template<TESTNAME AttributeSelection, bool isScalar = false>

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

inlineprivate

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 208 of file iti.h.

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

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

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< 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 342 of file iti_tpl.h.

References gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_nodeId2Database, gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_target, gum::MultiDimFunctionGraphManager< GUM_SCALAR, TerminalNodePolicy >::addTerminalNode(), and gum::MultiDimFunctionGraph< GUM_SCALAR, TerminalNodePolicy >::manager().

                                              {
     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;
   }

Here is the call graph for this function:

◆ __insertTerminalNode() [3/3]

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 311 of file iti_tpl.h.

References gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_nodeId2Database, gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_target, gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_value, gum::MultiDimFunctionGraph< GUM_SCALAR, TerminalNodePolicy >::add(), gum::MultiDimFunctionGraphManager< GUM_SCALAR, TerminalNodePolicy >::addInternalNode(), gum::MultiDimFunctionGraphManager< GUM_SCALAR, TerminalNodePolicy >::addTerminalNode(), gum::DiscreteVariable::domainSize(), gum::MultiDimFunctionGraph< GUM_SCALAR, TerminalNodePolicy >::manager(), SOA_ALLOCATE, and gum::MultiDimImplementation< GUM_SCALAR >::variablesSequence().

                                               {
     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;
   }

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
	)

protectedvirtual

Changes the associated variable of a node.

Parameters

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

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

Definition at line 229 of file iti_tpl.h.

References gum::ITI< AttributeSelection, isScalar >::__staleTable, gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_chgNodeBoundVar(), and gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_nodeVarMap.

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

                                                                {
     if (this->_nodeVarMap[currentNodeId] != desiredVar) {
       __staleTable[currentNodeId] = true;
       IncrementalGraphLearner< AttributeSelection, isScalar >::_chgNodeBoundVar(
          currentNodeId, desiredVar);
     }
   }

Here is the call graph for this function:

Here is the caller graph for this function:

◆ _convertNode2Leaf()

template<TESTNAME AttributeSelection, bool isScalar>

void gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_convertNode2Leaf ( NodeId currentNodeId )

protectedvirtualinherited

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

Definition at line 211 of file incrementalGraphLearner_tpl.h.

References gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_chgNodeBoundVar(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_leafDatabase, gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_nodeSonsMap, gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_nodeVarMap, gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_removeNode(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_value, gum::HashTable< Key, Val, Alloc >::erase(), and SOA_DEALLOCATE.

Referenced by gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_updateNode(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_updateNodeWithObservation(), gum::IMDDI< AttributeSelection, isScalar >::updateGraph(), and gum::IncrementalGraphLearner< AttributeSelection, isScalar >::updateVar().

                            {
     if (_nodeVarMap[currentNodeId] != _value) {
       _leafDatabase.insert(currentNodeId, new Set< const Observation* >());
 
       // Resolving potential 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);
     }
   }

Here is the call graph for this function:

Here is the caller graph for this function:

◆ _insertInternalNode()

template<TESTNAME AttributeSelection, bool isScalar>

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 383 of file incrementalGraphLearner_tpl.h.

References gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_insertNode(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_nodeSonsMap, and gum::HashTable< Key, Val, Alloc >::insert().

Referenced by gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_transpose(), and gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_updateNodeWithObservation().

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

Here is the call graph for this function:

Here is the caller graph for this function:

◆ _insertLeafNode()

template<TESTNAME AttributeSelection, bool isScalar>

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 403 of file incrementalGraphLearner_tpl.h.

References gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_insertNode(), and gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_leafDatabase.

Referenced by gum::IMDDI< AttributeSelection, isScalar >::_insertLeafNode(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_transpose(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_updateNodeWithObservation(), and gum::IncrementalGraphLearner< AttributeSelection, isScalar >::IncrementalGraphLearner().

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

Here is the call graph for this function:

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
	)

protectedvirtual

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, isScalar >.

Definition at line 210 of file iti_tpl.h.

References gum::ITI< AttributeSelection, isScalar >::__staleTable, gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_insertNode(), and gum::HashTable< Key, Val, Alloc >::insert().

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

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

Here is the call graph for this function:

Here is the caller graph for this function:

◆ _insertSetOfVars()

template<TESTNAME AttributeSelection, bool isScalar = false>

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

inlineprotected

_insertSetOfVars

Parameters

ret

Definition at line 244 of file iti.h.

References gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_setOfVars, and gum::MultiDimFunctionGraph< GUM_SCALAR, TerminalNodePolicy >::add().

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

Here is the call graph for this function:

◆ _removeNode()

template<TESTNAME AttributeSelection, bool isScalar>

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

protectedvirtual

Removes a node from the internal graph.

Parameters

removedNodeId : the node to remove

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

Definition at line 246 of file iti_tpl.h.

References gum::ITI< AttributeSelection, isScalar >::__staleTable, gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_removeNode(), and gum::HashTable< Key, Val, Alloc >::erase().

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

                                                                             {
     IncrementalGraphLearner< AttributeSelection, isScalar >::_removeNode(
        currentNodeId);
     __staleTable.erase(currentNodeId);
   }

Here is the call graph for this function:

Here is the caller graph for this function:

◆ _transpose()

template<TESTNAME AttributeSelection, bool isScalar>

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 240 of file incrementalGraphLearner_tpl.h.

Referenced by gum::IMDDI< AttributeSelection, isScalar >::__updateNodeSet(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_updateNode(), and gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_updateNodeWithObservation().

                                                                {
     // **************************************************************************************
     // 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);
   }

Here is the call graph for this function:

Here is the caller graph for this function:

◆ _updateNode()

template<TESTNAME AttributeSelection, bool isScalar>

void gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_updateNode	(	NodeId	updatedNode,
		Set< const DiscreteVariable * > &	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 180 of file incrementalGraphLearner_tpl.h.

References gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_convertNode2Leaf(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_nodeVarMap, gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_transpose(), gum::Set< Key, Alloc >::cbeginSafe(), gum::Set< Key, Alloc >::cendSafe(), gum::Set< Key, Alloc >::empty(), gum::Set< Key, Alloc >::exists(), and gum::Set< Key, Alloc >::size().

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

                                                                          {
     // 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 = (Idx)(std::rand() / RAND_MAX) * varsOfInterest.size(), basc = 0;
     SetConstIteratorSafe< const DiscreteVariable* > varIter;
     for (varIter = varsOfInterest.cbeginSafe(), basc = 0;
          varIter != varsOfInterest.cendSafe() && basc < randy;
          ++varIter, basc++)
       ;
 
     _transpose(updatedNode, *varIter);
   }

Here is the call graph for this function:

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
	)

protectedvirtual

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

Parameters

newObs
currentNodeId

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

Definition at line 126 of file iti_tpl.h.

References gum::ITI< AttributeSelection, isScalar >::__staleTable.

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

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

Here is the caller graph for this function:

◆ addObservation()

template<TESTNAME AttributeSelection, bool isScalar>

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

virtual

Inserts a new observation.

Parameters

obs	the new observation to learn
the	new observation to learn

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

Definition at line 112 of file iti_tpl.h.

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

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

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

Here is the call graph for this function:

Here is the caller graph for this function:

◆ insertSetOfVars()

template<TESTNAME AttributeSelection, bool isScalar = false>

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

inlinevirtualinherited

Implements gum::IVisitableGraphLearner.

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

Definition at line 344 of file incrementalGraphLearner.h.

References gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_setOfVars, and gum::MultiDimFunctionGraph< GUM_SCALAR, TerminalNodePolicy >::add().

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

Here is the call graph for this function:

◆ isTerminal()

template<TESTNAME AttributeSelection, bool isScalar = false>

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

inlinevirtualinherited

Implements gum::IVisitableGraphLearner.

Definition at line 318 of file incrementalGraphLearner.h.

References gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_nodeSonsMap, and gum::HashTable< Key, Val, Alloc >::exists().

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

gum::IncrementalGraphLearner::_nodeSonsMap

HashTable< NodeId, NodeId *> _nodeSonsMap

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

Definition: incrementalGraphLearner.h:380

gum::HashTable::exists

bool exists(const Key &key) const

Checks whether there exists an element with a given key in the hashtable.

Definition: hashTable_tpl.h:746

Here is the call graph for this function:

◆ nodeNbObservation()

template<TESTNAME AttributeSelection, bool isScalar = false>

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

inlinevirtualinherited

Implements gum::IVisitableGraphLearner.

Definition at line 337 of file incrementalGraphLearner.h.

References gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_nodeId2Database.

                                            {
       return this->_nodeId2Database[ni]->nbObservation();
     }

◆ nodeSon()

template<TESTNAME AttributeSelection, bool isScalar = false>

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

inlinevirtualinherited

Implements gum::IVisitableGraphLearner.

Definition at line 330 of file incrementalGraphLearner.h.

References gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_nodeSonsMap.

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

◆ nodeVar()

template<TESTNAME AttributeSelection, bool isScalar = false>

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

inlinevirtualinherited

Implements gum::IVisitableGraphLearner.

Definition at line 323 of file incrementalGraphLearner.h.

References gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_nodeVarMap.

                                                      {
       return this->_nodeVarMap[ni];
     }

◆ root()

template<TESTNAME AttributeSelection, bool isScalar = false>

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

inlinevirtualinherited

Implements gum::IVisitableGraphLearner.

Definition at line 313 of file incrementalGraphLearner.h.

References gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_root.

313 { return this->_root; }

gum::IncrementalGraphLearner::_root

NodeId _root

The root of the ordered tree.

Definition: incrementalGraphLearner.h:369

◆ size()

template<TESTNAME AttributeSelection, bool isScalar = false>

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

inlineinherited

Definition at line 302 of file incrementalGraphLearner.h.

References gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_nodeVarMap.

302 { return _nodeVarMap.size(); }

gum::IncrementalGraphLearner::_nodeVarMap

HashTable< NodeId, const DiscreteVariable *> _nodeVarMap

Gives for any node its associated variable.

Definition: incrementalGraphLearner.h:374

◆ updateFunctionGraph()

template<TESTNAME AttributeSelection, bool isScalar>

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

virtual

Updates target to currently learned graph structure.

Implements gum::IncrementalGraphLearner< AttributeSelection, isScalar >.

Definition at line 261 of file iti_tpl.h.

References gum::ITI< AttributeSelection, isScalar >::__insertNodeInFunctionGraph(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_root, gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_target, gum::MultiDimFunctionGraph< GUM_SCALAR, TerminalNodePolicy >::clear(), gum::MultiDimFunctionGraph< GUM_SCALAR, TerminalNodePolicy >::manager(), and gum::MultiDimFunctionGraphManager< GUM_SCALAR, TerminalNodePolicy >::setRootNode().

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

                                                                 {
     this->_target->clear();
     this->_target->manager()->setRootNode(
        this->__insertNodeInFunctionGraph(this->_root));
   }

Here is the call graph for this function:

Here is the caller graph for this function:

◆ updateGraph()

template<TESTNAME AttributeSelection, bool isScalar>

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

virtual

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

Implements gum::IncrementalGraphLearner< AttributeSelection, isScalar >.

Definition at line 143 of file iti_tpl.h.

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

                                                         {
     std::vector< NodeId > filo;
     filo.push_back(this->_root);
     HashTable< NodeId, Set< const DiscreteVariable* >* > potentialVars;
     potentialVars.insert(this->_root,
                          new Set< const DiscreteVariable* >(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;
       Set< const DiscreteVariable* > bestVars;
 
       for (auto varIter = potentialVars[currentNodeId]->cbeginSafe();
            varIter != potentialVars[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++) {
           Set< const DiscreteVariable* >* itsPotentialVars =
              new Set< const DiscreteVariable* >(*potentialVars[currentNodeId]);
           itsPotentialVars->erase(this->_nodeVarMap[currentNodeId]);
           NodeId sonId = this->_nodeSonsMap[currentNodeId][moda];
           if (__staleTable[sonId]) {
             filo.push_back(sonId);
             potentialVars.insert(sonId, itsPotentialVars);
           }
         }
       }
     }
 
     for (HashTableIteratorSafe< NodeId, Set< const DiscreteVariable* >* >
             nodeIter = potentialVars.beginSafe();
          nodeIter != potentialVars.endSafe();
          ++nodeIter)
       delete nodeIter.val();
   }

Here is the call graph for this function:

Here is the caller graph for this function:

◆ updateVar()

template<TESTNAME AttributeSelection, bool isScalar>

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 158 of file incrementalGraphLearner_tpl.h.

References gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_convertNode2Leaf(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_var2Node, gum::Link< T >::element(), and gum::Link< T >::nextLink().

Referenced by gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_updateNodeWithObservation().

                                   {
     Link< NodeId >* nodIter = _var2Node[var]->list();
     Link< NodeId >* nni = nullptr;
     while (nodIter) {
       nni = nodIter->nextLink();
       _convertNode2Leaf(nodIter->element());
       nodIter = nni;
     }
   }

Here is the call graph for this function:

Here is the caller graph for this function:

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 262 of file iti.h.

Referenced by gum::ITI< AttributeSelection, isScalar >::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 259 of file iti.h.

Referenced by gum::ITI< AttributeSelection, isScalar >::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 256 of file iti.h.

Referenced by gum::ITI< AttributeSelection, isScalar >::_chgNodeBoundVar(), gum::ITI< AttributeSelection, isScalar >::_insertNode(), gum::ITI< AttributeSelection, isScalar >::_removeNode(), gum::ITI< AttributeSelection, isScalar >::_updateNodeWithObservation(), gum::ITI< AttributeSelection, isScalar >::ITI(), and gum::ITI< AttributeSelection, isScalar >::updateGraph().

◆ _leafDatabase

template<TESTNAME AttributeSelection, bool isScalar = false>

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 399 of file incrementalGraphLearner.h.

Referenced by gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_chgNodeBoundVar(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_convertNode2Leaf(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_insertLeafNode(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_removeNode(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_transpose(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::addObservation(), and gum::IncrementalGraphLearner< AttributeSelection, isScalar >::~IncrementalGraphLearner().

◆ _model

template<TESTNAME AttributeSelection, bool isScalar = false>

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

protectedinherited

The source of nodeId.

Definition at line 364 of file incrementalGraphLearner.h.

Referenced by gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_insertNode(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_removeNode(), and gum::IncrementalGraphLearner< AttributeSelection, isScalar >::IncrementalGraphLearner().

◆ _needUpdate

template<TESTNAME AttributeSelection, bool isScalar = false>

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

protectedinherited

Definition at line 412 of file incrementalGraphLearner.h.

Referenced by gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_chgNodeBoundVar(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_insertNode(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_removeNode(), and gum::IMDDI< AttributeSelection, isScalar >::updateFunctionGraph().

◆ _nodeId2Database

template<TESTNAME AttributeSelection, bool isScalar = false>

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 393 of file incrementalGraphLearner.h.

◆ _nodeSonsMap

template<TESTNAME AttributeSelection, bool isScalar = false>

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 380 of file incrementalGraphLearner.h.

◆ _nodeVarMap

template<TESTNAME AttributeSelection, bool isScalar = false>

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

protectedinherited

Gives for any node its associated variable.

Definition at line 374 of file incrementalGraphLearner.h.

◆ _root

template<TESTNAME AttributeSelection, bool isScalar = false>

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

protectedinherited

The root of the ordered tree.

Definition at line 369 of file incrementalGraphLearner.h.

◆ _setOfVars

template<TESTNAME AttributeSelection, bool isScalar = false>

Set< const DiscreteVariable* > gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_setOfVars

protectedinherited

Definition at line 407 of file incrementalGraphLearner.h.

Referenced by gum::ITI< AttributeSelection, isScalar >::_insertSetOfVars(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_transpose(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::IncrementalGraphLearner(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::insertSetOfVars(), gum::IMDDI< AttributeSelection, isScalar >::updateGraph(), and gum::ITI< AttributeSelection, isScalar >::updateGraph().

◆ _target

template<TESTNAME AttributeSelection, bool isScalar = false>

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

protectedinherited

The final diagram we're building.

Definition at line 405 of file incrementalGraphLearner.h.

Referenced by gum::IMDDI< AttributeSelection, isScalar >::__insertLeafInFunctionGraph(), gum::ITI< AttributeSelection, isScalar >::__insertNodeInFunctionGraph(), gum::ITI< AttributeSelection, isScalar >::__insertTerminalNode(), gum::IMDDI< AttributeSelection, isScalar >::__rebuildFunctionGraph(), and gum::ITI< AttributeSelection, isScalar >::updateFunctionGraph().

◆ _value

template<TESTNAME AttributeSelection, bool isScalar = false>

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

protectedinherited

Definition at line 409 of file incrementalGraphLearner.h.

◆ _valueAssumed

template<TESTNAME AttributeSelection, bool isScalar = false>

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

protectedinherited

Definition at line 410 of file incrementalGraphLearner.h.

Referenced by gum::IMDDI< AttributeSelection, isScalar >::__addLeaf(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::__assumeValue(), and gum::IMDDI< AttributeSelection, isScalar >::__insertLeafInFunctionGraph().

◆ _var2Node

template<TESTNAME AttributeSelection, bool isScalar = false>

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 386 of file incrementalGraphLearner.h.

Referenced by gum::IMDDI< AttributeSelection, isScalar >::__rebuildFunctionGraph(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_chgNodeBoundVar(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_insertNode(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::_removeNode(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::IncrementalGraphLearner(), gum::IncrementalGraphLearner< AttributeSelection, isScalar >::updateVar(), and gum::IncrementalGraphLearner< AttributeSelection, isScalar >::~IncrementalGraphLearner().

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 Attributes

Protected Member Functions

Graph Structure update methods

New Observation insertion methods

Graph Structure update methods

Function Graph Updating methods

Detailed Description

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

Constructor & Destructor Documentation

◆ ITI() [1/2]

◆ ITI() [2/2]

◆ ~ITI()

Member Function Documentation

◆ __insertNodeInFunctionGraph()

◆ __insertTerminalNode() [1/3]

◆ __insertTerminalNode() [2/3]

◆ __insertTerminalNode() [3/3]

◆ _chgNodeBoundVar()

◆ _convertNode2Leaf()

◆ _insertInternalNode()

◆ _insertLeafNode()

◆ _insertNode()

◆ _insertSetOfVars()

◆ _removeNode()

◆ _transpose()

◆ _updateNode()

◆ _updateNodeWithObservation()

◆ addObservation()

◆ insertSetOfVars()

◆ isTerminal()

◆ nodeNbObservation()

◆ nodeSon()

◆ nodeVar()

◆ root()

◆ size()

◆ updateFunctionGraph()

◆ updateGraph()

◆ updateVar()

Member Data Documentation

◆ __attributeSelectionThreshold

◆ __nbTotalObservation

◆ __staleTable

◆ _leafDatabase

◆ _model

◆ _needUpdate

◆ _nodeId2Database

◆ _nodeSonsMap

◆ _nodeVarMap

◆ _root

◆ _setOfVars

◆ _target

◆ _value

◆ _valueAssumed

◆ _var2Node

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