gum::IMDDI< AttributeSelection, isScalar > Class Template Reference

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

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

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

Public Member Functions
void	insertSetOfVars (MultiDimFunctionGraph< double > *ret) const
Size	size ()
Constructor & destructor.
	IMDDI (MultiDimFunctionGraph< double > *target, double attributeSelectionThreshold, double pairSelectionThreshold, Set< const DiscreteVariable * > attributeListe, const DiscreteVariable *learnedValue)
	Variable Learner constructor. More...
	IMDDI (MultiDimFunctionGraph< double > *target, double attributeSelectionThreshold, double pairSelectionThreshold, Set< const DiscreteVariable * > attributeListe)
	Reward Learner constructor. More...
	~IMDDI ()
	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

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...

Incrementals methods
void	addObservation (const Observation *)
	Adds a new observation to the structure. More...
void	updateGraph ()
	Updates the tree after a new observation has been added. More...
void	updateNodeWithObservation_ (const Observation *newObs, NodeId currentNodeId)
	Adds a new observation to the structure. More...
NodeId	insertLeafNode_ (NodeDatabase< AttributeSelection, isScalar > *nDB, const DiscreteVariable *boundVar, Set< const Observation * > *sonsMap)
	Adds a new observation to the structure. More...
void	chgNodeBoundVar_ (NodeId chgedNodeId, const DiscreteVariable *desiredVar)
	Adds a new observation to the structure. More...
void	removeNode_ (NodeId removedNodeId)
	Adds a new observation to the structure. More...
void	addLeaf__ (NodeId)
	Adds a new observation to the structure. More...
void	removeLeaf__ (NodeId)
	Adds a new observation to the structure. More...

Updating private methods
void	updateFunctionGraph ()
	Computes the score of the given variables for the given node. More...
void	updateScore__ (const DiscreteVariable *, NodeId, VariableSelector &vs)
	Computes the score of the given variables for the given node. More...
void	downdateScore__ (const DiscreteVariable *, NodeId, VariableSelector &vs)
	Computes the score of the given variables for the given node. More...
void	updateNodeSet__ (Set< NodeId > &, const DiscreteVariable *, VariableSelector &)
	For each node in the given set, this methods checks whether or not we should installed the given variable as a test. If so, the node is updated. More...
void	rebuildFunctionGraph__ ()
	Computes the score of the given variables for the given node. More...
NodeId	insertLeafInFunctionGraph__ (AbstractLeaf *, Int2Type< true >)
	Computes the score of the given variables for the given node. More...
NodeId	insertLeafInFunctionGraph__ (AbstractLeaf *, Int2Type< false >)
	Computes the score of the given variables for the given node. 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	insertNode_ (NodeDatabase< AttributeSelection, isScalar > *nDB, const DiscreteVariable *boundVar)
	inserts a new node in internal graph More...
virtual NodeId	insertInternalNode_ (NodeDatabase< AttributeSelection, isScalar > *nDB, const DiscreteVariable *boundVar, NodeId *sonsMap)
	inserts a new internal node in internal graph More...

Detailed Description

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

Definition at line 57 of file imddi.h.

Constructor & Destructor Documentation

IMDDI() [1/2]

template<TESTNAME AttributeSelection, bool isScalar>

gum::IMDDI< AttributeSelection, isScalar >::IMDDI	(	MultiDimFunctionGraph< double > *	target,
		double	attributeSelectionThreshold,
		double	pairSelectionThreshold,
		Set< const DiscreteVariable * >	attributeListe,
		const DiscreteVariable *	learnedValue
	)

Variable Learner constructor.

Definition at line 51 of file imddi_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                                                    :
      IncrementalGraphLearner< AttributeSelection, isScalar >(target,
                                                              attributeListe,
                                                              learnedValue),
      lg__(&(this->model_), pairSelectionThreshold), nbTotalObservation__(0),
      attributeSelectionThreshold__(attributeSelectionThreshold) {
    GUM_CONSTRUCTOR(IMDDI);
    addLeaf__(this->root_);
  }

Here is the call graph for this function:

IMDDI() [2/2]

template<TESTNAME AttributeSelection, bool isScalar>

gum::IMDDI< AttributeSelection, isScalar >::IMDDI	(	MultiDimFunctionGraph< double > *	target,
		double	attributeSelectionThreshold,
		double	pairSelectionThreshold,
		Set< const DiscreteVariable * >	attributeListe
	)

Reward Learner constructor.

Definition at line 70 of file imddi_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                                                      :
      IncrementalGraphLearner< AttributeSelection, isScalar >(
         target,
         attributeListe,
         new LabelizedVariable("Reward", "", 2)),
      lg__(&(this->model_), pairSelectionThreshold), nbTotalObservation__(0),
      attributeSelectionThreshold__(attributeSelectionThreshold) {
    GUM_CONSTRUCTOR(IMDDI);
    addLeaf__(this->root_);
  }

Here is the call graph for this function:

~IMDDI()

template<TESTNAME AttributeSelection, bool isScalar>

gum::IMDDI< AttributeSelection, isScalar >::~IMDDI

(

)

Default destructor.

Definition at line 89 of file imddi_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                                                {
    GUM_DESTRUCTOR(IMDDI);
    for (HashTableIteratorSafe< NodeId, AbstractLeaf* > leafIter
         = leafMap__.beginSafe();
         leafIter != leafMap__.endSafe();
         ++leafIter)
      delete leafIter.val();
  }

Here is the call graph for this function:

Member Function Documentation

addLeaf__()

template<TESTNAME AttributeSelection, bool isScalar>

void gum::IMDDI< AttributeSelection, isScalar >::addLeaf__ ( NodeId  currentNodeId )

private

Adds a new observation to the structure.

Definition at line 302 of file imddi_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                                                                            {
    leafMap__.insert(currentNodeId,
                     new ConcreteLeaf< AttributeSelection, isScalar >(
                        currentNodeId,
                        this->nodeId2Database_[currentNodeId],
                        &(this->valueAssumed_)));
    lg__.addLeaf(leafMap__[currentNodeId]);
  }

Here is the call graph for this function:

addObservation()

template<TESTNAME AttributeSelection, bool isScalar>

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

virtual

Adds a new observation to the structure.

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

Definition at line 104 of file imddi_tpl.h.

References gum::Set< Key, Alloc >::emplace().

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

Here is the call graph for this function:

chgNodeBoundVar_()

template<TESTNAME AttributeSelection, bool isScalar>

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

protectedvirtual

Adds a new observation to the structure.

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

Definition at line 272 of file imddi_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                                         {
    if (this->nodeVarMap_[currentNodeId] == this->value_)
      removeLeaf__(currentNodeId);

    IncrementalGraphLearner< AttributeSelection, isScalar >::chgNodeBoundVar_(
       currentNodeId,
       desiredVar);

    if (desiredVar == this->value_) addLeaf__(currentNodeId);
  }

Here is the call 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 212 of file incrementalGraphLearner_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                           {
    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:

downdateScore__()

template<TESTNAME AttributeSelection, bool isScalar>

void gum::IMDDI< AttributeSelection, isScalar >::downdateScore__ ( const DiscreteVariable *  var, NodeId  nody, VariableSelector &  vs  )

private

Computes the score of the given variables for the given node.

Definition at line 189 of file imddi_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                                 {
    if (!this->nodeId2Database_[nody]->isTestRelevant(var)) return;
    double weight = (double)this->nodeId2Database_[nody]->nbObservation()
                  / (double)this->nbTotalObservation__;
    vs.downdateScore(var,
                     weight * this->nodeId2Database_[nody]->testValue(var),
                     weight
                        * this->nodeId2Database_[nody]->testOtherCriterion(var));
  }

Here is the call 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 386 of file incrementalGraphLearner_tpl.h.

References gum::Set< Key, Alloc >::emplace().

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

Here is the call graph for this function:

insertLeafInFunctionGraph__() [1/2]

template<TESTNAME AttributeSelection, bool isScalar>

NodeId gum::IMDDI< AttributeSelection, isScalar >::insertLeafInFunctionGraph__ ( AbstractLeaf *  leaf, Int2Type< true >    )

private

Computes the score of the given variables for the given node.

Definition at line 403 of file imddi_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                       {
    double value = 0.0;
    for (Idx moda = 0; moda < leaf->nbModa(); moda++) {
      value += (double)leaf->effectif(moda) * this->valueAssumed_.atPos(moda);
    }
    if (leaf->total()) value /= (double)leaf->total();
    return this->target_->manager()->addTerminalNode(value);
  }

Here is the call graph for this function:

insertLeafInFunctionGraph__() [2/2]

template<TESTNAME AttributeSelection, bool isScalar>

NodeId gum::IMDDI< AttributeSelection, isScalar >::insertLeafInFunctionGraph__ ( AbstractLeaf *  leaf, Int2Type< false >    )

private

Computes the score of the given variables for the given node.

Definition at line 419 of file imddi_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                        {
    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;
      if (leaf->total())
        newVal = (double)leaf->effectif(modality) / (double)leaf->total();
      sonsMap[modality] = this->target_->manager()->addTerminalNode(newVal);
    }
    return this->target_->manager()->addInternalNode(this->value_, sonsMap);
  }

Here is the call graph for this function:

insertLeafNode_()

template<TESTNAME AttributeSelection, bool isScalar>

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

protectedvirtual

Adds a new observation to the structure.

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

Definition at line 252 of file imddi_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                                                           {
    NodeId currentNodeId
       = IncrementalGraphLearner< AttributeSelection, isScalar >::insertLeafNode_(
          nDB,
          boundVar,
          obsSet);

    addLeaf__(currentNodeId);

    return currentNodeId;
  }

Here is the call graph for this function:

insertNode_()

template<TESTNAME AttributeSelection, bool isScalar>

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

protectedvirtualinherited

inserts a new node in internal graph

Parameters

nDB	: the associated database
boundVar	: the associated variable

Returns

the newly created node's id

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

Definition at line 361 of file incrementalGraphLearner_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                                                             {
    NodeId newNodeId = model_.addNode();
    nodeVarMap_.insert(newNodeId, boundVar);
    nodeId2Database_.insert(newNodeId, nDB);
    var2Node_[boundVar]->addLink(newNodeId);

    needUpdate_ = true;

    return newNodeId;
  }

Here is the call graph for this function:

insertSetOfVars()

template<TESTNAME AttributeSelection, bool isScalar = false>

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

inlinevirtual

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

Definition at line 158 of file imddi.h.

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

isTerminal()

template<TESTNAME AttributeSelection, bool isScalar = false>

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

inlinevirtualinherited

Implements gum::IVisitableGraphLearner.

Definition at line 317 of file incrementalGraphLearner.h.

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

nodeNbObservation()

template<TESTNAME AttributeSelection, bool isScalar = false>

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

inlinevirtualinherited

Implements gum::IVisitableGraphLearner.

Definition at line 336 of file incrementalGraphLearner.h.

                                           {
      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 329 of file incrementalGraphLearner.h.

                                                  {
      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 322 of file incrementalGraphLearner.h.

                                                     {
      return this->nodeVarMap_[ni];
    }

rebuildFunctionGraph__()

template<TESTNAME AttributeSelection, bool isScalar>

void gum::IMDDI< AttributeSelection, isScalar >::rebuildFunctionGraph__ ( )

private

Computes the score of the given variables for the given node.

Definition at line 340 of file imddi_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                                                                     {
    // *******************************************************************************************************
    // Mise à jour de l'aggregateur de feuille
    lg__.update();

    // *******************************************************************************************************
    // Reinitialisation du Graphe de Décision
    this->target_->clear();
    for (auto varIter = varOrder__.beginSafe(); varIter != varOrder__.endSafe();
         ++varIter)
      this->target_->add(**varIter);
    this->target_->add(*this->value_);

    HashTable< NodeId, NodeId > toTarget;

    // *******************************************************************************************************
    // Insertion des feuilles
    HashTable< NodeId, AbstractLeaf* > treeNode2leaf = lg__.leavesMap();
    HashTable< AbstractLeaf*, NodeId > leaf2DGNode;
    for (HashTableConstIteratorSafe< NodeId, AbstractLeaf* > treeNodeIter
         = treeNode2leaf.cbeginSafe();
         treeNodeIter != treeNode2leaf.cendSafe();
         ++treeNodeIter) {
      if (!leaf2DGNode.exists(treeNodeIter.val()))
        leaf2DGNode.insert(treeNodeIter.val(),
                           insertLeafInFunctionGraph__(treeNodeIter.val(),
                                                       Int2Type< isScalar >()));

      toTarget.insert(treeNodeIter.key(), leaf2DGNode[treeNodeIter.val()]);
    }

    // *******************************************************************************************************
    // Insertion des noeuds internes (avec vérification des possibilités de
    // fusion)
    for (SequenceIteratorSafe< const DiscreteVariable* > varIter
         = varOrder__.rbeginSafe();
         varIter != varOrder__.rendSafe();
         --varIter) {
      for (Link< NodeId >* curNodeIter = this->var2Node_[*varIter]->list();
           curNodeIter;
           curNodeIter = curNodeIter->nextLink()) {
        NodeId* sonsMap = static_cast< NodeId* >(
           SOA_ALLOCATE(sizeof(NodeId) * (*varIter)->domainSize()));
        for (Idx modality = 0; modality < (*varIter)->domainSize(); ++modality)
          sonsMap[modality]
             = toTarget[this->nodeSonsMap_[curNodeIter->element()][modality]];
        toTarget.insert(
           curNodeIter->element(),
           this->target_->manager()->addInternalNode(*varIter, sonsMap));
      }
    }

    // *******************************************************************************************************
    // Polish
    this->target_->manager()->setRootNode(toTarget[this->root_]);
    this->target_->manager()->clean();
  }

Here is the call graph for this function:

removeLeaf__()

template<TESTNAME AttributeSelection, bool isScalar>

void gum::IMDDI< AttributeSelection, isScalar >::removeLeaf__ ( NodeId  currentNodeId )

private

Adds a new observation to the structure.

Definition at line 316 of file imddi_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                                                                               {
    lg__.removeLeaf(leafMap__[currentNodeId]);
    delete leafMap__[currentNodeId];
    leafMap__.erase(currentNodeId);
  }

Here is the call graph for this function:

removeNode_()

template<TESTNAME AttributeSelection, bool isScalar>

void gum::IMDDI< AttributeSelection, isScalar >::removeNode_ ( NodeId  removedNodeId )

protectedvirtual

Adds a new observation to the structure.

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

Definition at line 290 of file imddi_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                                                                              {
    if (this->nodeVarMap_[currentNodeId] == this->value_)
      removeLeaf__(currentNodeId);
    IncrementalGraphLearner< AttributeSelection, isScalar >::removeNode_(
       currentNodeId);
  }

Here is the call graph for this function:

root()

template<TESTNAME AttributeSelection, bool isScalar = false>

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

inlinevirtualinherited

Implements gum::IVisitableGraphLearner.

Definition at line 312 of file incrementalGraphLearner.h.

{ return this->root_; }

size()

template<TESTNAME AttributeSelection, bool isScalar = false>

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

inlineinherited

Definition at line 301 of file incrementalGraphLearner.h.

{ return nodeVarMap_.size(); }

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 241 of file incrementalGraphLearner_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                                         {
    // **************************************************************************************
    // 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:

updateFunctionGraph()

template<TESTNAME AttributeSelection, bool isScalar>

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

virtual

Computes the score of the given variables for the given node.

Implements gum::IncrementalGraphLearner< AttributeSelection, isScalar >.

Definition at line 328 of file imddi_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                                                                  {
    //      if( lg__.needsUpdate() || this->needUpdate_ ){
    rebuildFunctionGraph__();
    this->needUpdate_ = false;
    //      }
  }

Here is the call graph for this function:

updateGraph()

template<TESTNAME AttributeSelection, bool isScalar>

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

virtual

Updates the tree after a new observation has been added.

Implements gum::IncrementalGraphLearner< AttributeSelection, isScalar >.

Definition at line 126 of file imddi_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                                                          {
    varOrder__.clear();

    // First xe initialize the node set which will give us the scores
    Set< NodeId > currentNodeSet;
    currentNodeSet.insert(this->root_);

    // Then we initialize the pool of variables to consider
    VariableSelector vs(this->setOfVars_);
    for (vs.begin(); vs.hasNext(); vs.next()) {
      updateScore__(vs.current(), this->root_, vs);
    }

    // Then, until there's no node remaining
    while (!vs.isEmpty()) {
      // We select the best var
      const DiscreteVariable* selectedVar = vs.select();
      varOrder__.insert(selectedVar);

      // Then we decide if we update each node according to this var
      updateNodeSet__(currentNodeSet, selectedVar, vs);
    }

    // If there are remaining node that are not leaves after we establish the
    // var order
    // these nodes are turned into leaf.
    for (SetIteratorSafe< NodeId > nodeIter = currentNodeSet.beginSafe();
         nodeIter != currentNodeSet.endSafe();
         ++nodeIter)
      this->convertNode2Leaf_(*nodeIter);


    if (lg__.needsUpdate()) lg__.update();
  }

Here is the call 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::Set< Key, Alloc >::emplace().

                                                     {
    // 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:

updateNodeSet__()

template<TESTNAME AttributeSelection, bool isScalar>

void gum::IMDDI< AttributeSelection, isScalar >::updateNodeSet__ ( Set< NodeId > &  nodeSet, const DiscreteVariable *  selectedVar, VariableSelector &  vs  )

private

For each node in the given set, this methods checks whether or not we should installed the given variable as a test. If so, the node is updated.

Definition at line 209 of file imddi_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                                 {
    Set< NodeId > oldNodeSet(nodeSet);
    nodeSet.clear();
    for (SetIteratorSafe< NodeId > nodeIter = oldNodeSet.beginSafe();
         nodeIter != oldNodeSet.endSafe();
         ++nodeIter) {
      if (this->nodeId2Database_[*nodeIter]->isTestRelevant(selectedVar)
          && this->nodeId2Database_[*nodeIter]->testValue(selectedVar)
                > attributeSelectionThreshold__) {
        this->transpose_(*nodeIter, selectedVar);

        // Then we subtract the from the score given to each variables the
        // quantity given by this node
        for (vs.begin(); vs.hasNext(); vs.next()) {
          downdateScore__(vs.current(), *nodeIter, vs);
        }

        // And finally we add all its child to the new set of nodes
        // and updates the remaining var's score
        for (Idx modality = 0;
             modality < this->nodeVarMap_[*nodeIter]->domainSize();
             ++modality) {
          NodeId sonId = this->nodeSonsMap_[*nodeIter][modality];
          nodeSet << sonId;

          for (vs.begin(); vs.hasNext(); vs.next()) {
            updateScore__(vs.current(), sonId, vs);
          }
        }
      } else {
        nodeSet << *nodeIter;
      }
    }
  }

Here is the call graph for this function:

updateNodeWithObservation_()

template<TESTNAME AttributeSelection, bool isScalar>

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

protectedvirtual

Adds a new observation to the structure.

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

Definition at line 111 of file imddi_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                                       {
    IncrementalGraphLearner< AttributeSelection,
                             isScalar >::updateNodeWithObservation_(newObs,
                                                                    currentNodeId);
    if (this->nodeVarMap_[currentNodeId] == this->value_)
      lg__.updateLeaf(leafMap__[currentNodeId]);
  }

Here is the call graph for this function:

updateScore__()

template<TESTNAME AttributeSelection, bool isScalar>

void gum::IMDDI< AttributeSelection, isScalar >::updateScore__ ( const DiscreteVariable *  var, NodeId  nody, VariableSelector &  vs  )

private

Computes the score of the given variables for the given node.

Definition at line 176 of file imddi_tpl.h.

References gum::Set< Key, Alloc >::emplace().

                                 {
    if (!this->nodeId2Database_[nody]->isTestRelevant(var)) return;
    double weight = (double)this->nodeId2Database_[nody]->nbObservation()
                  / (double)this->nbTotalObservation__;
    vs.updateScore(var,
                   weight * this->nodeId2Database_[nody]->testValue(var),
                   weight * this->nodeId2Database_[nody]->testOtherCriterion(var));
  }

Here is the call 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::Set< Key, Alloc >::emplace().

                                  {
    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:

Member Data Documentation

attributeSelectionThreshold__

template<TESTNAME AttributeSelection, bool isScalar = false>

double gum::IMDDI< AttributeSelection, isScalar >::attributeSelectionThreshold__

private

The threshold above which we consider variables to be dependant.

Definition at line 177 of file imddi.h.

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 398 of file incrementalGraphLearner.h.

leafMap__

template<TESTNAME AttributeSelection, bool isScalar = false>

HashTable< NodeId, AbstractLeaf* > gum::IMDDI< AttributeSelection, isScalar >::leafMap__

private

Definition at line 171 of file imddi.h.

lg__

template<TESTNAME AttributeSelection, bool isScalar = false>

LeafAggregator gum::IMDDI< AttributeSelection, isScalar >::lg__

private

Definition at line 169 of file imddi.h.

model_

template<TESTNAME AttributeSelection, bool isScalar = false>

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

protectedinherited

The source of nodeId.

Definition at line 363 of file incrementalGraphLearner.h.

nbTotalObservation__

template<TESTNAME AttributeSelection, bool isScalar = false>

Idx gum::IMDDI< AttributeSelection, isScalar >::nbTotalObservation__

private

The total number of observation added to this tree.

Definition at line 174 of file imddi.h.

needUpdate_

template<TESTNAME AttributeSelection, bool isScalar = false>

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

protectedinherited

Definition at line 411 of file incrementalGraphLearner.h.

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 392 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 379 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 373 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 368 of file incrementalGraphLearner.h.

setOfVars_

template<TESTNAME AttributeSelection, bool isScalar = false>

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

protectedinherited

Definition at line 406 of file incrementalGraphLearner.h.

target_

template<TESTNAME AttributeSelection, bool isScalar = false>

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

protectedinherited

The final diagram we're building.

Definition at line 404 of file incrementalGraphLearner.h.

value_

template<TESTNAME AttributeSelection, bool isScalar = false>

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

protectedinherited

Definition at line 408 of file incrementalGraphLearner.h.

valueAssumed_

template<TESTNAME AttributeSelection, bool isScalar = false>

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

protectedinherited

Definition at line 409 of file incrementalGraphLearner.h.

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 385 of file incrementalGraphLearner.h.

varOrder__

template<TESTNAME AttributeSelection, bool isScalar = false>

Sequence< const DiscreteVariable* > gum::IMDDI< AttributeSelection, isScalar >::varOrder__

private

Definition at line 167 of file imddi.h.

---

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

• agrum/FMDP/learning/datastructure/imddi.h
• agrum/FMDP/learning/datastructure/imddi_tpl.h