gum::learning::Miic Class Reference

The miic learning algorithm. More...

#include <Miic.h>

Inheritance diagram for gum::learning::Miic:

Collaboration diagram for gum::learning::Miic:

Public Attributes
Signaler3< Size, double, double >	onProgress
	Progression, error and time. More...
Signaler1< std::string >	onStop
	Criteria messageApproximationScheme. More...

Public Member Functions
Miic &	operator= (const Miic &from)
	copy operator More...
Miic &	operator= (Miic &&from)
	move operator More...
Constructors / Destructors
	Miic ()
	default constructor More...
	Miic (int maxLog)
	default constructor with maxLog More...
	Miic (const Miic &from)
	copy constructor More...
	Miic (Miic &&from)
	move constructor More...
	~Miic ()
	destructor More...
Accessors / Modifiers
MixedGraph	learnMixedStructure (CorrectedMutualInformation<> &I, MixedGraph graph)
	learns the structure of an Essential Graph More...
DAG	learnStructure (CorrectedMutualInformation<> &I, MixedGraph graph)
	learns the structure of an Bayesian network, ie a DAG, by first learning an Essential graph and then directing the remaining edges. More...
template<typename GUM_SCALAR = double, typename GRAPH_CHANGES_SELECTOR , typename PARAM_ESTIMATOR >
BayesNet< GUM_SCALAR >	learnBN (GRAPH_CHANGES_SELECTOR &selector, PARAM_ESTIMATOR &estimator, DAG initial_dag=DAG())
	learns the structure and the parameters of a BN More...
const std::vector< Arc >	latentVariables () const
	get the list of arcs hiding latent variables More...
void	setMiicBehaviour ()
	Sets the orientation phase to follow the one of the MIIC algorithm. More...
void	set3off2Behaviour ()
	Sets the orientation phase to follow the one of the 3off2 algorithm. More...
void	addConstraints (HashTable< std::pair< NodeId, NodeId >, char > constraints)
	Set a ensemble of constraints for the orientation phase. More...
Getters and setters
void	setEpsilon (double eps)
	Given that we approximate f(t), stopping criterion on |f(t+1)-f(t)|. More...
double	epsilon () const
	Returns the value of epsilon. More...
void	disableEpsilon ()
	Disable stopping criterion on epsilon. More...
void	enableEpsilon ()
	Enable stopping criterion on epsilon. More...
bool	isEnabledEpsilon () const
	Returns true if stopping criterion on epsilon is enabled, false otherwise. More...
void	setMinEpsilonRate (double rate)
	Given that we approximate f(t), stopping criterion on d/dt(|f(t+1)-f(t)|). More...
double	minEpsilonRate () const
	Returns the value of the minimal epsilon rate. More...
void	disableMinEpsilonRate ()
	Disable stopping criterion on epsilon rate. More...
void	enableMinEpsilonRate ()
	Enable stopping criterion on epsilon rate. More...
bool	isEnabledMinEpsilonRate () const
	Returns true if stopping criterion on epsilon rate is enabled, false otherwise. More...
void	setMaxIter (Size max)
	Stopping criterion on number of iterations. More...
Size	maxIter () const
	Returns the criterion on number of iterations. More...
void	disableMaxIter ()
	Disable stopping criterion on max iterations. More...
void	enableMaxIter ()
	Enable stopping criterion on max iterations. More...
bool	isEnabledMaxIter () const
	Returns true if stopping criterion on max iterations is enabled, false otherwise. More...
void	setMaxTime (double timeout)
	Stopping criterion on timeout. More...
double	maxTime () const
	Returns the timeout (in seconds). More...
double	currentTime () const
	Returns the current running time in second. More...
void	disableMaxTime ()
	Disable stopping criterion on timeout. More...
void	enableMaxTime ()
	Enable stopping criterion on timeout. More...
bool	isEnabledMaxTime () const
	Returns true if stopping criterion on timeout is enabled, false otherwise. More...
void	setPeriodSize (Size p)
	How many samples between two stopping is enable. More...
Size	periodSize () const
	Returns the period size. More...
void	setVerbosity (bool v)
	Set the verbosity on (true) or off (false). More...
bool	verbosity () const
	Returns true if verbosity is enabled. More...
ApproximationSchemeSTATE	stateApproximationScheme () const
	Returns the approximation scheme state. More...
Size	nbrIterations () const
	Returns the number of iterations. More...
const std::vector< double > &	history () const
	Returns the scheme history. More...
void	initApproximationScheme ()
	Initialise the scheme. More...
bool	startOfPeriod ()
	Returns true if we are at the beginning of a period (compute error is mandatory). More...
void	updateApproximationScheme (unsigned int incr=1)
	Update the scheme w.r.t the new error and increment steps. More...
Size	remainingBurnIn ()
	Returns the remaining burn in. More...
void	stopApproximationScheme ()
	Stop the approximation scheme. More...
bool	continueApproximationScheme (double error)
	Update the scheme w.r.t the new error. More...
Getters and setters
std::string	messageApproximationScheme () const
	Returns the approximation scheme message. More...

Public Types
enum	ApproximationSchemeSTATE : char {   ApproximationSchemeSTATE::Undefined, ApproximationSchemeSTATE::Continue, ApproximationSchemeSTATE::Epsilon, ApproximationSchemeSTATE::Rate,   ApproximationSchemeSTATE::Limit, ApproximationSchemeSTATE::TimeLimit, ApproximationSchemeSTATE::Stopped }
	The different state of an approximation scheme. More...

Protected Attributes
double	_current_epsilon
	Current epsilon. More...
double	_last_epsilon
	Last epsilon value. More...
double	_current_rate
	Current rate. More...
Size	_current_step
	The current step. More...
Timer	_timer
	The timer. More...
ApproximationSchemeSTATE	_current_state
	The current state. More...
std::vector< double >	_history
	The scheme history, used only if verbosity == true. More...
double	_eps
	Threshold for convergence. More...
bool	_enabled_eps
	If true, the threshold convergence is enabled. More...
double	_min_rate_eps
	Threshold for the epsilon rate. More...
bool	_enabled_min_rate_eps
	If true, the minimal threshold for epsilon rate is enabled. More...
double	_max_time
	The timeout. More...
bool	_enabled_max_time
	If true, the timeout is enabled. More...
Size	_max_iter
	The maximum iterations. More...
bool	_enabled_max_iter
	If true, the maximum iterations stopping criterion is enabled. More...
Size	_burn_in
	Number of iterations before checking stopping criteria. More...
Size	_period_size
	Checking criteria frequency. More...
bool	_verbosity
	If true, verbosity is enabled. More...

Protected Member Functions
void	_findBestContributor (NodeId x, NodeId y, const std::vector< NodeId > &ui, const MixedGraph &graph, CorrectedMutualInformation<> &I, Heap< std::pair< std::tuple< NodeId, NodeId, NodeId, std::vector< NodeId > > *, double >, GreaterPairOn2nd > &_rank)
	finds the best contributor node for a pair given a conditioning set More...
std::vector< std::pair< std::tuple< NodeId, NodeId, NodeId > *, double > >	_getUnshieldedTriples (const MixedGraph &graph, CorrectedMutualInformation<> &I, const HashTable< std::pair< NodeId, NodeId >, std::vector< NodeId > > &sep_set)
	gets the list of unshielded triples in the graph in decreasing value of |I'(x, y, z|{ui})| More...
std::vector< std::tuple< std::tuple< NodeId, NodeId, NodeId > *, double, double, double > >	_getUnshieldedTriplesMIIC (const MixedGraph &graph, CorrectedMutualInformation<> &I, const HashTable< std::pair< NodeId, NodeId >, std::vector< NodeId > > &sep_set, HashTable< std::pair< NodeId, NodeId >, char > &marks)
	gets the list of unshielded triples in the graph in decreasing value of |I'(x, y, z|{ui})|, prepares the orientation matrix for MIIC More...
std::vector< std::tuple< std::tuple< NodeId, NodeId, NodeId > *, double, double, double > >	_updateProbaTriples (const MixedGraph &graph, std::vector< std::tuple< std::tuple< NodeId, NodeId, NodeId > *, double, double, double > > proba_triples)
	Gets the orientation probabilities like MIIC for the orientation phase. More...
void	_propagatesHead (MixedGraph &graph, NodeId node)
	Propagates the orientation from a node to its neighbours. More...
Main phases
void	_initiation (CorrectedMutualInformation<> &I, MixedGraph &graph, HashTable< std::pair< NodeId, NodeId >, std::vector< NodeId > > &sep_set, Heap< std::pair< std::tuple< NodeId, NodeId, NodeId, std::vector< NodeId > > *, double >, GreaterPairOn2nd > &_rank)
	Initiation phase. More...
void	_iteration (CorrectedMutualInformation<> &I, MixedGraph &graph, HashTable< std::pair< NodeId, NodeId >, std::vector< NodeId > > &sep_set, Heap< std::pair< std::tuple< NodeId, NodeId, NodeId, std::vector< NodeId > > *, double >, GreaterPairOn2nd > &_rank)
	Iteration phase. More...
void	_orientation_3off2 (CorrectedMutualInformation<> &I, MixedGraph &graph, const HashTable< std::pair< NodeId, NodeId >, std::vector< NodeId > > &sep_set)
	Orientation phase from the 3off2 algorithm, returns a CPDAG. More...
void	_orientation_latents (CorrectedMutualInformation<> &I, MixedGraph &graph, const HashTable< std::pair< NodeId, NodeId >, std::vector< NodeId > > &sep_set)
	Modified version of the orientation phase that tries to propagate orientations from both orientations in case of a bidirected arc, not used. More...
void	_orientation_miic (CorrectedMutualInformation<> &I, MixedGraph &graph, const HashTable< std::pair< NodeId, NodeId >, std::vector< NodeId > > &sep_set)
	Orientation phase from the MIIC algorithm, returns a mixed graph that may contain circles. More...

Detailed Description

The miic learning algorithm.

The miic class implements the miic algorithm based on https://doi.org/10.1371/journal.pcbi.1005662. It starts by eliminating edges that correspond to independent variables to build the skeleton of the graph, and then directs the remaining edges to get an essential graph. Latent variables can be detected using bi-directed arcs.

The variant 3off2 is also implemented as proposed by Affeldt and al. in https://doi.org/10.1186/s12859-015-0856-x. Only the orientation phase differs from miic, with a different ranking method and different propagation rules.

Definition at line 106 of file Miic.h.

Member Enumeration Documentation

ApproximationSchemeSTATE

enum gum::IApproximationSchemeConfiguration::ApproximationSchemeSTATE : char

stronginherited

The different state of an approximation scheme.

Enumerator
Undefined
Continue
Epsilon
Rate
Limit
TimeLimit
Stopped

Definition at line 65 of file IApproximationSchemeConfiguration.h.

                                        : char {
      Undefined,
      Continue,
      Epsilon,
      Rate,
      Limit,
      TimeLimit,
      Stopped
    };

Constructor & Destructor Documentation

Miic() [1/4]

gum::learning::Miic::Miic

(

)

default constructor

Definition at line 43 of file Miic.cpp.

{ GUM_CONSTRUCTOR(Miic); }

Miic() [2/4]

gum::learning::Miic::Miic

(

int

maxLog

)

default constructor with maxLog

Definition at line 46 of file Miic.cpp.

: __maxLog(maxLog) { GUM_CONSTRUCTOR(Miic); }

Miic() [3/4]

gum::learning::Miic::Miic

(

const Miic &

from

)

copy constructor

Definition at line 49 of file Miic.cpp.

                               : ApproximationScheme(from) {
      GUM_CONS_CPY(Miic);
    }

Miic() [4/4]

gum::learning::Miic::Miic

(

Miic &&

from

)

move constructor

Definition at line 54 of file Miic.cpp.

                          : ApproximationScheme(std::move(from)) {
      GUM_CONS_MOV(Miic);
    }

~Miic()

gum::learning::Miic::~Miic

(

)

destructor

Definition at line 59 of file Miic.cpp.

{ GUM_DESTRUCTOR(Miic); }

Member Function Documentation

__existsDirectedPath()

const bool gum::learning::Miic::__existsDirectedPath ( const MixedGraph &  graph, const NodeId  n1, const NodeId  n2  ) const

private

checks for directed paths in a graph, consider double arcs like edges

Definition at line 1072 of file Miic.cpp.

References gum::List< Val, Alloc >::empty(), gum::HashTable< Key, Val, Alloc >::exists(), gum::ArcGraphPart::existsArc(), gum::List< Val, Alloc >::front(), gum::HashTable< Key, Val, Alloc >::insert(), gum::ArcGraphPart::parents(), gum::List< Val, Alloc >::popFront(), and gum::List< Val, Alloc >::pushBack().

Referenced by _orientation_3off2(), _orientation_miic(), and _propagatesHead().

                                                                      {
      // not recursive version => use a FIFO for simulating the recursion
      List< NodeId > nodeFIFO;
      nodeFIFO.pushBack(n2);

      // mark[node] = successor if visited, else mark[node] does not exist
      NodeProperty< NodeId > mark;
      mark.insert(n2, n2);

      NodeId current;

      while (!nodeFIFO.empty()) {
        current = nodeFIFO.front();
        nodeFIFO.popFront();

        // check the parents

        for (const auto new_one: graph.parents(current)) {
          if (mark.exists(new_one))   // if this node is already marked, do not
            continue;                 // check it again

          if (graph.existsArc(current,
                              new_one))   // if there is a double arc, pass
            continue;

          mark.insert(new_one, current);

          if (new_one == n1) { return true; }

          nodeFIFO.pushBack(new_one);
        }
      }

      return false;
    }

Here is the call graph for this function:

Here is the caller graph for this function:

_findBestContributor()

void gum::learning::Miic::_findBestContributor ( NodeId  x, NodeId  y, const std::vector< NodeId > &  ui, const MixedGraph &  graph, CorrectedMutualInformation<> &  I, Heap< std::pair< std::tuple< NodeId, NodeId, NodeId, std::vector< NodeId > > *, double >, GreaterPairOn2nd > &  _rank  )

protected

finds the best contributor node for a pair given a conditioning set

Parameters

x	first node
y	second node
ui	conditioning set
I	A mutual information instance that will do the computations and has loaded the database.
graph	containing the assessed nodes
_rank	the heap of ranks of the algorithm

Definition at line 764 of file Miic.cpp.

References __maxLog, M_LN2, and gum::learning::CorrectedMutualInformation< ALLOC >::score().

Referenced by _initiation(), and _iteration().

                                     {
      double maxP = -1.0;
      NodeId maxZ = 0;

      // compute N
      //__N = I.N();
      const double Ixy_ui = I.score(x, y, ui);

      for (const NodeId z: graph) {
        // if z!=x and z!=y and z not in ui
        if (z != x && z != y && std::find(ui.begin(), ui.end(), z) == ui.end()) {
          double Pnv;
          double Pb;

          // Computing Pnv
          const double Ixyz_ui = I.score(x, y, z, ui);
          double       calc_expo1 = -Ixyz_ui * M_LN2;
          // if exponentials are too high or to low, crop them at |__maxLog|
          if (calc_expo1 > __maxLog) {
            Pnv = 0.0;
          } else if (calc_expo1 < -__maxLog) {
            Pnv = 1.0;
          } else {
            Pnv = 1 / (1 + std::exp(calc_expo1));
          }

          // Computing Pb
          const double Ixz_ui = I.score(x, z, ui);
          const double Iyz_ui = I.score(y, z, ui);

          calc_expo1 = -(Ixz_ui - Ixy_ui) * M_LN2;
          double calc_expo2 = -(Iyz_ui - Ixy_ui) * M_LN2;

          // if exponentials are too high or to low, crop them at __maxLog
          if (calc_expo1 > __maxLog || calc_expo2 > __maxLog) {
            Pb = 0.0;
          } else if (calc_expo1 < -__maxLog && calc_expo2 < -__maxLog) {
            Pb = 1.0;
          } else {
            double expo1, expo2;
            if (calc_expo1 < -__maxLog) {
              expo1 = 0.0;
            } else {
              expo1 = std::exp(calc_expo1);
            }
            if (calc_expo2 < -__maxLog) {
              expo2 = 0.0;
            } else {
              expo2 = std::exp(calc_expo2);
            }
            Pb = 1 / (1 + expo1 + expo2);
          }

          // Getting max(min(Pnv, pb))
          const double min_pnv_pb = std::min(Pnv, Pb);
          if (min_pnv_pb > maxP) {
            maxP = min_pnv_pb;
            maxZ = z;
          }
        }   // if z not in (x, y)
      }     // for z in graph.nodes
      // storing best z in _rank
      std::pair< std::tuple< NodeId, NodeId, NodeId, std::vector< NodeId > >*,
                 double >
           final;
      auto tup = new std::tuple< NodeId, NodeId, NodeId, std::vector< NodeId > >{
         x, y, maxZ, ui};
      final.first = tup;
      final.second = maxP;
      _rank.insert(final);
    }

Here is the call graph for this function:

Here is the caller graph for this function:

_getUnshieldedTriples()

std::vector< std::pair< std::tuple< NodeId, NodeId, NodeId > *, double > > gum::learning::Miic::_getUnshieldedTriples ( const MixedGraph &  graph, CorrectedMutualInformation<> &  I, const HashTable< std::pair< NodeId, NodeId >, std::vector< NodeId > > &  sep_set  )

protected

gets the list of unshielded triples in the graph in decreasing value of |I'(x, y, z|{ui})|

Definition at line 848 of file Miic.cpp.

References gum::learning::CorrectedMutualInformation< ALLOC >::score().

Referenced by _orientation_3off2(), and _orientation_latents().

                      {
      std::vector< std::pair< std::tuple< NodeId, NodeId, NodeId >*, double > >
         triples;
      for (NodeId z: graph) {
        for (NodeId x: graph.neighbours(z)) {
          for (NodeId y: graph.neighbours(z)) {
            if (y < x && !graph.existsEdge(x, y)) {
              std::vector< NodeId >       ui;
              std::pair< NodeId, NodeId > key = {x, y};
              std::pair< NodeId, NodeId > rev_key = {y, x};
              if (sep_set.exists(key)) {
                ui = sep_set[key];
              } else if (sep_set.exists(rev_key)) {
                ui = sep_set[rev_key];
              }
              // remove z from ui if it's present
              const auto iter_z_place = std::find(ui.begin(), ui.end(), z);
              if (iter_z_place != ui.end()) { ui.erase(iter_z_place); }

              double Ixyz_ui = I.score(x, y, z, ui);
              std::pair< std::tuple< NodeId, NodeId, NodeId >*, double > triple;
              auto tup = new std::tuple< NodeId, NodeId, NodeId >{x, y, z};
              triple.first = tup;
              triple.second = Ixyz_ui;
              triples.push_back(triple);
            }
          }
        }
      }
      std::sort(triples.begin(), triples.end(), GreaterAbsPairOn2nd());
      return triples;
    }

Here is the call graph for this function:

Here is the caller graph for this function:

_getUnshieldedTriplesMIIC()

std::vector< std::tuple< std::tuple< NodeId, NodeId, NodeId > *, double, double, double > > gum::learning::Miic::_getUnshieldedTriplesMIIC ( const MixedGraph &  graph, CorrectedMutualInformation<> &  I, const HashTable< std::pair< NodeId, NodeId >, std::vector< NodeId > > &  sep_set, HashTable< std::pair< NodeId, NodeId >, char > &  marks  )

protected

gets the list of unshielded triples in the graph in decreasing value of |I'(x, y, z|{ui})|, prepares the orientation matrix for MIIC

Definition at line 890 of file Miic.cpp.

References _updateProbaTriples(), and gum::learning::CorrectedMutualInformation< ALLOC >::score().

Referenced by _orientation_miic().

                                                               {
      std::vector< std::tuple< std::tuple< NodeId, NodeId, NodeId >*,
                               double,
                               double,
                               double > >
         triples;
      for (NodeId z: graph) {
        for (NodeId x: graph.neighbours(z)) {
          for (NodeId y: graph.neighbours(z)) {
            if (y < x && !graph.existsEdge(x, y)) {
              std::vector< NodeId >       ui;
              std::pair< NodeId, NodeId > key = {x, y};
              std::pair< NodeId, NodeId > rev_key = {y, x};
              if (sep_set.exists(key)) {
                ui = sep_set[key];
              } else if (sep_set.exists(rev_key)) {
                ui = sep_set[rev_key];
              }
              // remove z from ui if it's present
              const auto iter_z_place = std::find(ui.begin(), ui.end(), z);
              if (iter_z_place != ui.end()) { ui.erase(iter_z_place); }

              const double Ixyz_ui = I.score(x, y, z, ui);
              auto         tup = new std::tuple< NodeId, NodeId, NodeId >{x, y, z};
              std::tuple< std::tuple< NodeId, NodeId, NodeId >*,
                          double,
                          double,
                          double >
                 triple{tup, Ixyz_ui, 0.5, 0.5};
              triples.push_back(triple);
              if (!marks.exists({x, z})) { marks.insert({x, z}, 'o'); }
              if (!marks.exists({z, x})) { marks.insert({z, x}, 'o'); }
              if (!marks.exists({y, z})) { marks.insert({y, z}, 'o'); }
              if (!marks.exists({z, y})) { marks.insert({z, y}, 'o'); }
            }
          }
        }
      }
      triples = _updateProbaTriples(graph, triples);
      std::sort(triples.begin(), triples.end(), GreaterTupleOnLast());
      return triples;
    }

Here is the call graph for this function:

Here is the caller graph for this function:

_initiation()

void gum::learning::Miic::_initiation ( CorrectedMutualInformation<> &  I, MixedGraph &  graph, HashTable< std::pair< NodeId, NodeId >, std::vector< NodeId > > &  sep_set, Heap< std::pair< std::tuple< NodeId, NodeId, NodeId, std::vector< NodeId > > *, double >, GreaterPairOn2nd > &  _rank  )

protected

Initiation phase.

We go over all edges and test if the variables are marginally independent. If they are, the edge is deleted. If not, the best contributor is found.

Parameters

I	A mutual information instance that will do the computations and has loaded the database.
graph	the MixedGraph we start from for the learning
sep_set	the separation set for independent couples, here set to {}
_rank	the heap of ranks of the algorithm

Definition at line 150 of file Miic.cpp.

References __empty_set, gum::ApproximationScheme::_current_step, _findBestContributor(), gum::ApproximationScheme::_timer, gum::EdgeGraphPart::edges(), gum::EdgeGraphPart::eraseEdge(), GUM_EMIT3, gum::IApproximationSchemeConfiguration::onProgress, gum::learning::CorrectedMutualInformation< ALLOC >::score(), gum::Set< Key, Alloc >::size(), and gum::Timer::step().

Referenced by learnMixedStructure().

                                     {
      NodeId  x, y;
      EdgeSet edges = graph.edges();
      Size    steps_init = edges.size();

      for (const Edge& edge: edges) {
        x = edge.first();
        y = edge.second();
        double Ixy = I.score(x, y);

        if (Ixy <= 0) {   //< K
          graph.eraseEdge(edge);
          sep_set.insert(std::make_pair(x, y), __empty_set);
        } else {
          _findBestContributor(x, y, __empty_set, graph, I, _rank);
        }

        ++_current_step;
        if (onProgress.hasListener()) {
          GUM_EMIT3(
             onProgress, (_current_step * 33) / steps_init, 0., _timer.step());
        }
      }
    }

Here is the call graph for this function:

Here is the caller graph for this function:

_iteration()

void gum::learning::Miic::_iteration ( CorrectedMutualInformation<> &  I, MixedGraph &  graph, HashTable< std::pair< NodeId, NodeId >, std::vector< NodeId > > &  sep_set, Heap< std::pair< std::tuple< NodeId, NodeId, NodeId, std::vector< NodeId > > *, double >, GreaterPairOn2nd > &  _rank  )

protected

Iteration phase.

As long as we find important nodes for edges, we go over them to see if we can assess the conditional independence of the variables.

Parameters

I	A mutual information instance that will do the computations and has loaded the database.
graph	the MixedGraph returned from the previous phase
sep_set	the separation set for independent couples, built during the iterations of the phase
_rank	the heap of ranks of the algorithm

Definition at line 188 of file Miic.cpp.

References gum::ApproximationScheme::_current_step, _findBestContributor(), gum::ApproximationScheme::_timer, gum::EdgeGraphPart::eraseEdge(), GUM_EMIT3, gum::IApproximationSchemeConfiguration::onProgress, gum::learning::CorrectedMutualInformation< ALLOC >::score(), and gum::Timer::step().

Referenced by learnMixedStructure().

                                     {
      // if no triples to further examine pass
      std::pair< std::tuple< NodeId, NodeId, NodeId, std::vector< NodeId > >*,
                 double >
         best;

      Size steps_init = _current_step;
      Size steps_iter = _rank.size();

      try {
        while (_rank.top().second > 0.5) {
          best = _rank.pop();

          const NodeId          x = std::get< 0 >(*(best.first));
          const NodeId          y = std::get< 1 >(*(best.first));
          const NodeId          z = std::get< 2 >(*(best.first));
          std::vector< NodeId > ui = std::move(std::get< 3 >(*(best.first)));

          ui.push_back(z);
          const double Ixy_ui = I.score(x, y, ui);
          if (Ixy_ui < 0) {
            graph.eraseEdge(Edge(x, y));
            sep_set.insert(std::make_pair(x, y), std::move(ui));
          } else {
            _findBestContributor(x, y, ui, graph, I, _rank);
          }

          delete best.first;

          ++_current_step;
          if (onProgress.hasListener()) {
            GUM_EMIT3(onProgress,
                      (_current_step * 66) / (steps_init + steps_iter),
                      0.,
                      _timer.step());
          }
        }
      } catch (...) {}   // here, rank is empty
      _current_step = steps_init + steps_iter;
      if (onProgress.hasListener()) {
        GUM_EMIT3(onProgress, 66, 0., _timer.step());
      }
      _current_step = steps_init + steps_iter;
    }

Here is the call graph for this function:

Here is the caller graph for this function:

_orientation_3off2()

void gum::learning::Miic::_orientation_3off2 ( CorrectedMutualInformation<> &  I, MixedGraph &  graph, const HashTable< std::pair< NodeId, NodeId >, std::vector< NodeId > > &  sep_set  )

protected

Orientation phase from the 3off2 algorithm, returns a CPDAG.

Parameters

I	A mutual information instance that will do the computations and has loaded the database.
graph	the MixedGraph returned from the previous phase
sep_set	the separation set for independent couples, built during the previous phase

Definition at line 245 of file Miic.cpp.

References __existsDirectedPath(), __initial_marks, __latent_couples, gum::ApproximationScheme::_current_step, _getUnshieldedTriples(), gum::ApproximationScheme::_timer, gum::DiGraph::addArc(), gum::HashTable< Key, Val, Alloc >::begin(), gum::HashTable< Key, Val, Alloc >::end(), gum::ArcGraphPart::eraseArc(), gum::EdgeGraphPart::eraseEdge(), gum::HashTable< Key, Val, Alloc >::exists(), gum::ArcGraphPart::existsArc(), gum::EdgeGraphPart::existsEdge(), GUM_EMIT3, gum::IApproximationSchemeConfiguration::onProgress, and gum::Timer::step().

Referenced by learnMixedStructure().

                   {
      std::vector< std::pair< std::tuple< NodeId, NodeId, NodeId >*, double > >
           triples = _getUnshieldedTriples(graph, I, sep_set);
      Size steps_orient = triples.size();
      Size past_steps = _current_step;

      // marks always correspond to the head of the arc/edge. - is for a forbidden
      // arc, > for a mandatory arc
      // we start by adding the mandatory arcs
      for (auto iter = __initial_marks.begin(); iter != __initial_marks.end();
           ++iter) {
        if (graph.existsEdge(iter.key().first, iter.key().second)
            && iter.val() == '>') {
          graph.eraseEdge(Edge(iter.key().first, iter.key().second));
          graph.addArc(iter.key().first, iter.key().second);
        }
      }

      NodeId i = 0;
      // list of elements that we shouldnt read again, ie elements that are
      // eligible to
      // rule 0 after the first time they are tested, and elements on which rule 1
      // has been applied
      while (i < triples.size()) {
        // if i not in do_not_reread
        std::pair< std::tuple< NodeId, NodeId, NodeId >*, double > triple =
           triples[i];
        NodeId x, y, z;
        x = std::get< 0 >(*triple.first);
        y = std::get< 1 >(*triple.first);
        z = std::get< 2 >(*triple.first);

        std::vector< NodeId >       ui;
        std::pair< NodeId, NodeId > key = {x, y};
        std::pair< NodeId, NodeId > rev_key = {y, x};
        if (sep_set.exists(key)) {
          ui = sep_set[key];
        } else if (sep_set.exists(rev_key)) {
          ui = sep_set[rev_key];
        }
        double Ixyz_ui = triple.second;
        bool   reset{false};
        // try Rule 0
        if (Ixyz_ui < 0) {
          // if ( z not in Sep[x,y])
          if (std::find(ui.begin(), ui.end(), z) == ui.end()) {
            if (!graph.existsArc(x, z) && !graph.existsArc(z, x)) {
              // when we try to add an arc to the graph, we always verify if
              // we are allowed to do so, ie it is not a forbidden arc an it
              // does not create a cycle
              if (!__existsDirectedPath(graph, z, x)
                  && !(__initial_marks.exists({x, z})
                       && __initial_marks[{x, z}] == '-')) {
                reset = true;
                graph.eraseEdge(Edge(x, z));
                graph.addArc(x, z);
              } else if (__existsDirectedPath(graph, z, x)
                         && !(__initial_marks.exists({z, x})
                              && __initial_marks[{z, x}] == '-')) {
                reset = true;
                graph.eraseEdge(Edge(x, z));
                // if we find a cycle, we force the competing edge
                graph.addArc(z, x);
                if (std::find(
                       __latent_couples.begin(), __latent_couples.end(), Arc(z, x))
                    == __latent_couples.end()) {
                  __latent_couples.push_back(Arc(z, x));
                }
              }
            } else if (!graph.existsArc(y, z) && !graph.existsArc(z, y)) {
              if (!__existsDirectedPath(graph, z, y)
                  && !(__initial_marks.exists({x, z})
                       && __initial_marks[{x, z}] == '-')) {
                reset = true;
                graph.eraseEdge(Edge(y, z));
                graph.addArc(y, z);
              } else if (__existsDirectedPath(graph, z, y)
                         && !(__initial_marks.exists({z, y})
                              && __initial_marks[{z, y}] == '-')) {
                reset = true;
                graph.eraseEdge(Edge(y, z));
                // if we find a cycle, we force the competing edge
                graph.addArc(z, y);
                if (std::find(
                       __latent_couples.begin(), __latent_couples.end(), Arc(z, y))
                    == __latent_couples.end()) {
                  __latent_couples.push_back(Arc(z, y));
                }
              }
            } else {
              // checking if the anti-directed arc already exists, to register a
              // latent variable
              if (graph.existsArc(z, x)
                  && std::find(__latent_couples.begin(),
                               __latent_couples.end(),
                               Arc(z, x))
                        == __latent_couples.end()
                  && std::find(__latent_couples.begin(),
                               __latent_couples.end(),
                               Arc(x, z))
                        == __latent_couples.end()) {
                __latent_couples.push_back(Arc(z, x));
              }
              if (graph.existsArc(z, y)
                  && std::find(__latent_couples.begin(),
                               __latent_couples.end(),
                               Arc(z, y))
                        == __latent_couples.end()
                  && std::find(__latent_couples.begin(),
                               __latent_couples.end(),
                               Arc(y, z))
                        == __latent_couples.end()) {
                __latent_couples.push_back(Arc(z, y));
              }
            }
          }
        } else {   // try Rule 1
          if (graph.existsArc(x, z) && !graph.existsArc(z, y)
              && !graph.existsArc(y, z)) {
            if (!__existsDirectedPath(graph, y, z)
                && !(__initial_marks.exists({z, y})
                     && __initial_marks[{z, y}] == '-')) {
              reset = true;
              graph.eraseEdge(Edge(z, y));
              graph.addArc(z, y);
            } else if (__existsDirectedPath(graph, y, z)
                       && !(__initial_marks.exists({y, z})
                            && __initial_marks[{y, z}] == '-')) {
              reset = true;
              graph.eraseEdge(Edge(z, y));
              // if we find a cycle, we force the competing edge
              graph.addArc(y, z);
              if (std::find(
                     __latent_couples.begin(), __latent_couples.end(), Arc(y, z))
                  == __latent_couples.end()) {
                __latent_couples.push_back(Arc(y, z));
              }
            }
          }
          if (graph.existsArc(y, z) && !graph.existsArc(z, x)
              && !graph.existsArc(x, z)) {
            if (!__existsDirectedPath(graph, x, z)
                && !(__initial_marks.exists({z, x})
                     && __initial_marks[{z, x}] == '-')) {
              reset = true;
              graph.eraseEdge(Edge(z, x));
              graph.addArc(z, x);
            } else if (__existsDirectedPath(graph, x, z)
                       && !(__initial_marks.exists({x, z})
                            && __initial_marks[{x, z}] == '-')) {
              reset = true;
              graph.eraseEdge(Edge(z, x));
              // if we find a cycle, we force the competing edge
              graph.addArc(x, z);
              if (std::find(
                     __latent_couples.begin(), __latent_couples.end(), Arc(x, z))
                  == __latent_couples.end()) {
                __latent_couples.push_back(Arc(x, z));
              }
            }
          }
        }   // if rule 0 or rule 1

        // if what we want to add already exists : pass to the next triplet
        if (reset) {
          i = 0;
        } else {
          ++i;
        }
        if (onProgress.hasListener()) {
          GUM_EMIT3(onProgress,
                    ((_current_step + i) * 100) / (past_steps + steps_orient),
                    0.,
                    _timer.step());
        }
      }   // while

      // erasing the the double headed arcs
      for (const Arc& arc: __latent_couples) {
        graph.eraseArc(Arc(arc.head(), arc.tail()));
      }
    }

Here is the call graph for this function:

Here is the caller graph for this function:

_orientation_latents()

void gum::learning::Miic::_orientation_latents ( CorrectedMutualInformation<> &  I, MixedGraph &  graph, const HashTable< std::pair< NodeId, NodeId >, std::vector< NodeId > > &  sep_set  )

protected

Modified version of the orientation phase that tries to propagate orientations from both orientations in case of a bidirected arc, not used.

varient trying to propagate both orientations in a bidirected arc

Parameters

I	A mutual information instance that will do the computations and has loaded the database.
graph	the MixedGraph returned from the previous phase
sep_set	the separation set for independent couples, built during the previous phase

Definition at line 433 of file Miic.cpp.

References __latent_couples, gum::ApproximationScheme::_current_step, _getUnshieldedTriples(), gum::ApproximationScheme::_timer, gum::DiGraph::addArc(), gum::ArcGraphPart::directedPath(), gum::ArcGraphPart::eraseArc(), gum::EdgeGraphPart::eraseEdge(), gum::ArcGraphPart::existsArc(), GUM_EMIT3, gum::IApproximationSchemeConfiguration::onProgress, and gum::Timer::step().

                   {
      std::vector< std::pair< std::tuple< NodeId, NodeId, NodeId >*, double > >
           triples = _getUnshieldedTriples(graph, I, sep_set);
      Size steps_orient = triples.size();
      Size past_steps = _current_step;

      NodeId i = 0;
      // list of elements that we shouldnt read again, ie elements that are
      // eligible to
      // rule 0 after the first time they are tested, and elements on which rule 1
      // has been applied
      while (i < triples.size()) {
        // if i not in do_not_reread
        std::pair< std::tuple< NodeId, NodeId, NodeId >*, double > triple =
           triples[i];
        NodeId x, y, z;
        x = std::get< 0 >(*triple.first);
        y = std::get< 1 >(*triple.first);
        z = std::get< 2 >(*triple.first);

        std::vector< NodeId >       ui;
        std::pair< NodeId, NodeId > key = {x, y};
        std::pair< NodeId, NodeId > rev_key = {y, x};
        if (sep_set.exists(key)) {
          ui = sep_set[key];
        } else if (sep_set.exists(rev_key)) {
          ui = sep_set[rev_key];
        }
        double Ixyz_ui = triple.second;
        // try Rule 0
        if (Ixyz_ui < 0) {
          // if ( z not in Sep[x,y])
          if (std::find(ui.begin(), ui.end(), z) == ui.end()) {
            // if what we want to add already exists : pass
            if ((graph.existsArc(x, z) || graph.existsArc(z, x))
                && (graph.existsArc(y, z) || graph.existsArc(z, y))) {
              ++i;
            } else {
              i = 0;
              graph.eraseEdge(Edge(x, z));
              graph.eraseEdge(Edge(y, z));
              // checking for cycles
              if (graph.existsArc(z, x)) {
                graph.eraseArc(Arc(z, x));
                try {
                  std::vector< NodeId > path = graph.directedPath(z, x);
                  // if we find a cycle, we force the competing edge
                  __latent_couples.push_back(Arc(z, x));
                } catch (gum::NotFound) { graph.addArc(x, z); }
                graph.addArc(z, x);
              } else {
                try {
                  std::vector< NodeId > path = graph.directedPath(z, x);
                  // if we find a cycle, we force the competing edge
                  graph.addArc(z, x);
                  __latent_couples.push_back(Arc(z, x));
                } catch (gum::NotFound) { graph.addArc(x, z); }
              }
              if (graph.existsArc(z, y)) {
                graph.eraseArc(Arc(z, y));
                try {
                  std::vector< NodeId > path = graph.directedPath(z, y);
                  // if we find a cycle, we force the competing edge
                  __latent_couples.push_back(Arc(z, y));
                } catch (gum::NotFound) { graph.addArc(y, z); }
                graph.addArc(z, y);
              } else {
                try {
                  std::vector< NodeId > path = graph.directedPath(z, y);
                  // if we find a cycle, we force the competing edge
                  graph.addArc(z, y);
                  __latent_couples.push_back(Arc(z, y));

                } catch (gum::NotFound) { graph.addArc(y, z); }
              }
              if (graph.existsArc(z, x)
                  && std::find(__latent_couples.begin(),
                               __latent_couples.end(),
                               Arc(z, x))
                        == __latent_couples.end()
                  && std::find(__latent_couples.begin(),
                               __latent_couples.end(),
                               Arc(x, z))
                        == __latent_couples.end()) {
                __latent_couples.push_back(Arc(z, x));
              }
              if (graph.existsArc(z, y)
                  && std::find(__latent_couples.begin(),
                               __latent_couples.end(),
                               Arc(z, y))
                        == __latent_couples.end()
                  && std::find(__latent_couples.begin(),
                               __latent_couples.end(),
                               Arc(y, z))
                        == __latent_couples.end()) {
                __latent_couples.push_back(Arc(z, y));
              }
            }
          } else {
            ++i;
          }
        } else {   // try Rule 1
          bool reset{false};
          if (graph.existsArc(x, z) && !graph.existsArc(z, y)
              && !graph.existsArc(y, z)) {
            reset = true;
            graph.eraseEdge(Edge(z, y));
            try {
              std::vector< NodeId > path = graph.directedPath(y, z);
              // if we find a cycle, we force the competing edge
              graph.addArc(y, z);
              __latent_couples.push_back(Arc(y, z));
            } catch (gum::NotFound) { graph.addArc(z, y); }
          }
          if (graph.existsArc(y, z) && !graph.existsArc(z, x)
              && !graph.existsArc(x, z)) {
            reset = true;
            graph.eraseEdge(Edge(z, x));
            try {
              std::vector< NodeId > path = graph.directedPath(x, z);
              // if we find a cycle, we force the competing edge
              graph.addArc(x, z);
              __latent_couples.push_back(Arc(x, z));
            } catch (gum::NotFound) { graph.addArc(z, x); }
          }

          if (reset) {
            i = 0;
          } else {
            ++i;
          }
        }   // if rule 0 or rule 1
        if (onProgress.hasListener()) {
          GUM_EMIT3(onProgress,
                    ((_current_step + i) * 100) / (past_steps + steps_orient),
                    0.,
                    _timer.step());
        }
      }   // while

      // erasing the the double headed arcs
      for (const Arc& arc: __latent_couples) {
        graph.eraseArc(Arc(arc.head(), arc.tail()));
      }
    }

Here is the call graph for this function:

_orientation_miic()

void gum::learning::Miic::_orientation_miic ( CorrectedMutualInformation<> &  I, MixedGraph &  graph, const HashTable< std::pair< NodeId, NodeId >, std::vector< NodeId > > &  sep_set  )

protected

Orientation phase from the MIIC algorithm, returns a mixed graph that may contain circles.

varient using the orientation protocol of MIIC

Parameters

I	A mutual information instance that will do the computations and has loaded the database.
graph	the MixedGraph returned from the previous phase
sep_set	the separation set for independent couples, built during the previous phase

Definition at line 585 of file Miic.cpp.

References __arc_probas, __existsDirectedPath(), __initial_marks, __latent_couples, gum::ApproximationScheme::_current_step, _getUnshieldedTriplesMIIC(), gum::ApproximationScheme::_timer, _updateProbaTriples(), gum::DiGraph::addArc(), gum::HashTable< Key, Val, Alloc >::begin(), gum::Set< Key, Alloc >::empty(), gum::HashTable< Key, Val, Alloc >::end(), gum::ArcGraphPart::eraseArc(), gum::EdgeGraphPart::eraseEdge(), gum::ArcGraphPart::existsArc(), gum::EdgeGraphPart::existsEdge(), GUM_EMIT3, gum::IApproximationSchemeConfiguration::onProgress, gum::ArcGraphPart::parents(), and gum::Timer::step().

Referenced by learnMixedStructure().

                                                                                {
      // structure to store the orientations marks -, o, or >,
      // Considers the head of the arc/edge first node -* second node
      HashTable< std::pair< NodeId, NodeId >, char > marks = __initial_marks;

      // marks always correspond to the head of the arc/edge. - is for a forbidden
      // arc, > for a mandatory arc
      // we start by adding the mandatory arcs
      for (auto iter = marks.begin(); iter != marks.end(); ++iter) {
        if (graph.existsEdge(iter.key().first, iter.key().second)
            && iter.val() == '>') {
          graph.eraseEdge(Edge(iter.key().first, iter.key().second));
          graph.addArc(iter.key().first, iter.key().second);
        }
      }

      std::vector< std::tuple< std::tuple< NodeId, NodeId, NodeId >*,
                               double,
                               double,
                               double > >
         proba_triples = _getUnshieldedTriplesMIIC(graph, I, sep_set, marks);

      Size steps_orient = proba_triples.size();
      Size past_steps = _current_step;

      std::tuple< std::tuple< NodeId, NodeId, NodeId >*, double, double, double >
         best;
      if (steps_orient > 0) { best = proba_triples[0]; }

      while (!proba_triples.empty()
             && std::max(std::get< 2 >(best), std::get< 3 >(best)) > 0.5) {
        NodeId x, y, z;
        x = std::get< 0 >(*std::get< 0 >(best));
        y = std::get< 1 >(*std::get< 0 >(best));
        z = std::get< 2 >(*std::get< 0 >(best));

        const double i3 = std::get< 1 >(best);

        if (i3 <= 0) {
          // v-structure discovery
          if (marks[{x, z}] == 'o' && marks[{y, z}] == 'o') {
            graph.eraseEdge(Edge(x, z));
            graph.eraseEdge(Edge(y, z));
            graph.addArc(x, z);
            graph.addArc(y, z);
            marks[{x, z}] = '>';
            marks[{y, z}] = '>';
            if (graph.existsArc(z, x)
                && std::find(
                      __latent_couples.begin(), __latent_couples.end(), Arc(z, x))
                      == __latent_couples.end()
                && std::find(
                      __latent_couples.begin(), __latent_couples.end(), Arc(x, z))
                      == __latent_couples.end()) {
              __latent_couples.push_back(Arc(z, x));
            }
            if (graph.existsArc(z, y)
                && std::find(
                      __latent_couples.begin(), __latent_couples.end(), Arc(z, y))
                      == __latent_couples.end()
                && std::find(
                      __latent_couples.begin(), __latent_couples.end(), Arc(y, z))
                      == __latent_couples.end()) {
              __latent_couples.push_back(Arc(z, y));
            }
            if (!__arc_probas.exists(Arc(x, z)))
              __arc_probas.insert(Arc(x, z), std::get< 2 >(best));
            if (!__arc_probas.exists(Arc(y, z)))
              __arc_probas.insert(Arc(y, z), std::get< 3 >(best));
          } else if (marks[{x, z}] == '>' && marks[{y, z}] == 'o') {
            graph.eraseEdge(Edge(y, z));
            graph.addArc(y, z);
            marks[{y, z}] = '>';
            if (graph.existsArc(z, y)
                && std::find(
                      __latent_couples.begin(), __latent_couples.end(), Arc(z, y))
                      == __latent_couples.end()
                && std::find(
                      __latent_couples.begin(), __latent_couples.end(), Arc(y, z))
                      == __latent_couples.end()) {
              __latent_couples.push_back(Arc(z, y));
            }
            if (!__arc_probas.exists(Arc(y, z)))
              __arc_probas.insert(Arc(y, z), std::get< 3 >(best));
          } else if (marks[{y, z}] == '>' && marks[{x, z}] == 'o') {
            graph.eraseEdge(Edge(x, z));
            graph.addArc(x, z);
            marks[{x, z}] = '>';
            if (graph.existsArc(z, x)
                && std::find(
                      __latent_couples.begin(), __latent_couples.end(), Arc(z, x))
                      == __latent_couples.end()
                && std::find(
                      __latent_couples.begin(), __latent_couples.end(), Arc(x, z))
                      == __latent_couples.end()) {
              __latent_couples.push_back(Arc(z, x));
            }
            if (!__arc_probas.exists(Arc(x, z)))
              __arc_probas.insert(Arc(x, z), std::get< 2 >(best));
          }

        } else {
          // orientation propagation
          if (marks[{x, z}] == '>' && marks[{y, z}] == 'o'
              && marks[{z, y}] != '-') {
            graph.eraseEdge(Edge(z, y));
            if (!__existsDirectedPath(graph, y, z) && graph.parents(y).empty()) {
              graph.addArc(z, y);
              marks[{z, y}] = '>';
              marks[{y, z}] = '-';
              if (!__arc_probas.exists(Arc(z, y)))
                __arc_probas.insert(Arc(z, y), std::get< 3 >(best));
            } else if (!__existsDirectedPath(graph, z, y)
                       && graph.parents(z).empty()) {
              graph.addArc(y, z);
              marks[{z, y}] = '-';
              marks[{y, z}] = '>';
              __latent_couples.push_back(Arc(y, z));
              if (!__arc_probas.exists(Arc(y, z)))
                __arc_probas.insert(Arc(y, z), std::get< 3 >(best));
            }
          } else if (marks[{y, z}] == '>' && marks[{x, z}] == 'o'
                     && marks[{z, x}] != '-') {
            graph.eraseEdge(Edge(z, x));
            if (!__existsDirectedPath(graph, x, z) && graph.parents(x).empty()) {
              graph.addArc(z, x);
              marks[{z, x}] = '>';
              marks[{x, z}] = '-';
              if (!__arc_probas.exists(Arc(z, x)))
                __arc_probas.insert(Arc(z, x), std::get< 2 >(best));
            } else if (!__existsDirectedPath(graph, z, x)
                       && graph.parents(z).empty()) {
              graph.addArc(x, z);
              marks[{z, x}] = '-';
              marks[{x, z}] = '>';
              __latent_couples.push_back(Arc(x, z));
              if (!__arc_probas.exists(Arc(x, z)))
                __arc_probas.insert(Arc(x, z), std::get< 2 >(best));
            }
          }
        }

        delete std::get< 0 >(best);
        proba_triples.erase(proba_triples.begin());
        // actualisation of the list of triples
        proba_triples = _updateProbaTriples(graph, proba_triples);

        best = proba_triples[0];

        ++_current_step;
        if (onProgress.hasListener()) {
          GUM_EMIT3(onProgress,
                    (_current_step * 100) / (steps_orient + past_steps),
                    0.,
                    _timer.step());
        }
      }   // while

      // erasing the double headed arcs
      for (auto iter = __latent_couples.rbegin(); iter != __latent_couples.rend();
           ++iter) {
        graph.eraseArc(Arc(iter->head(), iter->tail()));
        if (__existsDirectedPath(graph, iter->head(), iter->tail())) {
          // if we find a cycle, we force the competing edge
          graph.addArc(iter->head(), iter->tail());
          graph.eraseArc(Arc(iter->tail(), iter->head()));
          *iter = Arc(iter->head(), iter->tail());
        }
      }

      if (onProgress.hasListener()) {
        GUM_EMIT3(onProgress, 100, 0., _timer.step());
      }
    }

Here is the call graph for this function:

Here is the caller graph for this function:

_propagatesHead()

void gum::learning::Miic::_propagatesHead ( MixedGraph &  graph, NodeId  node  )

protected

Propagates the orientation from a node to its neighbours.

Definition at line 1018 of file Miic.cpp.

References __existsDirectedPath(), __initial_marks, __latent_couples, gum::DiGraph::addArc(), gum::Set< Key, Alloc >::empty(), gum::EdgeGraphPart::eraseEdge(), gum::HashTable< Key, Val, Alloc >::exists(), gum::EdgeGraphPart::neighbours(), and gum::ArcGraphPart::parents().

Referenced by learnStructure().

                                                             {
      const auto neighbours = graph.neighbours(node);
      for (auto& neighbour: neighbours) {
        // only propagate if it doesn't create a circle and isn't forbidden
        // and doesn't create a new v-structure
        if (!__existsDirectedPath(graph, neighbour, node)
            && !(__initial_marks.exists({node, neighbour})
                 && __initial_marks[{node, neighbour}] == '-')
            && graph.parents(neighbour).empty()) {
          graph.eraseEdge(Edge(neighbour, node));
          graph.addArc(node, neighbour);
          _propagatesHead(graph, neighbour);
        } else if (!__existsDirectedPath(graph, node, neighbour)
                   && !(__initial_marks.exists({neighbour, node})
                        && __initial_marks[{neighbour, node}] == '-')
                   && graph.parents(node).empty()) {
          graph.eraseEdge(Edge(neighbour, node));
          graph.addArc(neighbour, node);
        } else if (!graph.parents(neighbour).empty()
                   && !graph.parents(node).empty()) {
          graph.eraseEdge(Edge(neighbour, node));
          graph.addArc(node, neighbour);
          __latent_couples.push_back(Arc(node, neighbour));
        } else {
          graph.eraseEdge(Edge(neighbour, node));
        }
      }
    }

Here is the call graph for this function:

Here is the caller graph for this function:

_updateProbaTriples()

std::vector< std::tuple< std::tuple< NodeId, NodeId, NodeId > *, double, double, double > > gum::learning::Miic::_updateProbaTriples ( const MixedGraph &  graph, std::vector< std::tuple< std::tuple< NodeId, NodeId, NodeId > *, double, double, double > >  proba_triples  )

protected

Gets the orientation probabilities like MIIC for the orientation phase.

Definition at line 942 of file Miic.cpp.

References gum::ArcGraphPart::existsArc().

Referenced by _getUnshieldedTriplesMIIC(), and _orientation_miic().

                                                             {
      for (auto& triple: proba_triples) {
        NodeId x, y, z;
        x = std::get< 0 >(*std::get< 0 >(triple));
        y = std::get< 1 >(*std::get< 0 >(triple));
        z = std::get< 2 >(*std::get< 0 >(triple));
        const double Ixyz = std::get< 1 >(triple);
        double       Pxz = std::get< 2 >(triple);
        double       Pyz = std::get< 3 >(triple);

        if (Ixyz <= 0) {
          const double expo = std::exp(Ixyz);
          const double P0 = (1 + expo) / (1 + 3 * expo);
          // distinguish betweeen the initialization and the update process
          if (Pxz == Pyz && Pyz == 0.5) {
            std::get< 2 >(triple) = P0;
            std::get< 3 >(triple) = P0;
          } else {
            if (graph.existsArc(x, z) && Pxz >= P0) {
              std::get< 3 >(triple) = Pxz * (1 / (1 + expo) - 0.5) + 0.5;
            } else if (graph.existsArc(y, z) && Pyz >= P0) {
              std::get< 2 >(triple) = Pyz * (1 / (1 + expo) - 0.5) + 0.5;
            }
          }
        } else {
          const double expo = std::exp(-Ixyz);
          if (graph.existsArc(x, z) && Pxz >= 0.5) {
            std::get< 3 >(triple) = Pxz * (1 / (1 + expo) - 0.5) + 0.5;
          } else if (graph.existsArc(y, z) && Pyz >= 0.5) {
            std::get< 2 >(triple) = Pyz * (1 / (1 + expo) - 0.5) + 0.5;
          }
        }
      }
      std::sort(proba_triples.begin(), proba_triples.end(), GreaterTupleOnLast());
      return proba_triples;
    }

Here is the call graph for this function:

Here is the caller graph for this function:

addConstraints()

void gum::learning::Miic::addConstraints

(

HashTable< std::pair< NodeId, NodeId >, char >

constraints

)

Set a ensemble of constraints for the orientation phase.

Definition at line 1066 of file Miic.cpp.

References __initial_marks.

Referenced by gum::learning::genericBNLearner::__prepare_miic_3off2().

                                                                 {
      this->__initial_marks = constraints;
    }

Here is the caller graph for this function:

continueApproximationScheme()

INLINE bool gum::ApproximationScheme::continueApproximationScheme ( double  error )

inherited

Update the scheme w.r.t the new error.

Test the stopping criterion that are enabled.

Parameters

error

The new error value.

Returns

false if state become != ApproximationSchemeSTATE::Continue

Exceptions

OperationNotAllowed

Raised if state != ApproximationSchemeSTATE::Continue.

Definition at line 227 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_current_epsilon, gum::ApproximationScheme::_current_rate, gum::ApproximationScheme::_current_state, gum::ApproximationScheme::_current_step, gum::ApproximationScheme::_enabled_eps, gum::ApproximationScheme::_enabled_max_iter, gum::ApproximationScheme::_enabled_max_time, gum::ApproximationScheme::_enabled_min_rate_eps, gum::ApproximationScheme::_eps, gum::ApproximationScheme::_history, gum::ApproximationScheme::_last_epsilon, gum::ApproximationScheme::_max_iter, gum::ApproximationScheme::_max_time, gum::ApproximationScheme::_min_rate_eps, gum::ApproximationScheme::_stopScheme(), gum::ApproximationScheme::_timer, gum::IApproximationSchemeConfiguration::Continue, gum::IApproximationSchemeConfiguration::Epsilon, GUM_EMIT3, GUM_ERROR, gum::IApproximationSchemeConfiguration::Limit, gum::IApproximationSchemeConfiguration::messageApproximationScheme(), gum::IApproximationSchemeConfiguration::onProgress, gum::IApproximationSchemeConfiguration::Rate, gum::ApproximationScheme::startOfPeriod(), gum::ApproximationScheme::stateApproximationScheme(), gum::Timer::step(), gum::IApproximationSchemeConfiguration::TimeLimit, and gum::ApproximationScheme::verbosity().

Referenced by gum::GibbsBNdistance< GUM_SCALAR >::_computeKL(), gum::SamplingInference< GUM_SCALAR >::_loopApproxInference(), gum::learning::DAG2BNLearner< ALLOC >::createBN(), gum::learning::GreedyHillClimbing::learnStructure(), gum::learning::LocalSearchWithTabuList::learnStructure(), and gum::credal::CNMonteCarloSampling< GUM_SCALAR, BNInferenceEngine >::makeInference().

                                                                           {
    // For coherence, we fix the time used in the method

    double timer_step = _timer.step();

    if (_enabled_max_time) {
      if (timer_step > _max_time) {
        _stopScheme(ApproximationSchemeSTATE::TimeLimit);
        return false;
      }
    }

    if (!startOfPeriod()) { return true; }

    if (_current_state != ApproximationSchemeSTATE::Continue) {
      GUM_ERROR(OperationNotAllowed,
                "state of the approximation scheme is not correct : "
                   + messageApproximationScheme());
    }

    if (verbosity()) { _history.push_back(error); }

    if (_enabled_max_iter) {
      if (_current_step > _max_iter) {
        _stopScheme(ApproximationSchemeSTATE::Limit);
        return false;
      }
    }

    _last_epsilon = _current_epsilon;
    _current_epsilon = error;   // eps rate isEnabled needs it so affectation was
    // moved from eps isEnabled below

    if (_enabled_eps) {
      if (_current_epsilon <= _eps) {
        _stopScheme(ApproximationSchemeSTATE::Epsilon);
        return false;
      }
    }

    if (_last_epsilon >= 0.) {
      if (_current_epsilon > .0) {
        // ! _current_epsilon can be 0. AND epsilon
        // isEnabled can be disabled !
        _current_rate =
           std::fabs((_current_epsilon - _last_epsilon) / _current_epsilon);
      }
      // limit with current eps ---> 0 is | 1 - ( last_eps / 0 ) | --->
      // infinity the else means a return false if we isEnabled the rate below,
      // as we would have returned false if epsilon isEnabled was enabled
      else {
        _current_rate = _min_rate_eps;
      }

      if (_enabled_min_rate_eps) {
        if (_current_rate <= _min_rate_eps) {
          _stopScheme(ApproximationSchemeSTATE::Rate);
          return false;
        }
      }
    }

    if (stateApproximationScheme() == ApproximationSchemeSTATE::Continue) {
      if (onProgress.hasListener()) {
        GUM_EMIT3(onProgress, _current_step, _current_epsilon, timer_step);
      }

      return true;
    } else {
      return false;
    }
  }

Here is the call graph for this function:

Here is the caller graph for this function:

currentTime()

INLINE double gum::ApproximationScheme::currentTime ( ) const

virtualinherited

Returns the current running time in second.

Returns

Returns the current running time in second.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 128 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_timer, and gum::Timer::step().

Referenced by gum::learning::genericBNLearner::currentTime().

{ return _timer.step(); }

Here is the call graph for this function:

Here is the caller graph for this function:

disableEpsilon()

INLINE void gum::ApproximationScheme::disableEpsilon ( )

virtualinherited

Disable stopping criterion on epsilon.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 54 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_enabled_eps.

Referenced by gum::learning::genericBNLearner::disableEpsilon().

{ _enabled_eps = false; }

Here is the caller graph for this function:

disableMaxIter()

INLINE void gum::ApproximationScheme::disableMaxIter ( )

virtualinherited

Disable stopping criterion on max iterations.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 105 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_enabled_max_iter.

Referenced by gum::credal::CNMonteCarloSampling< GUM_SCALAR, BNInferenceEngine >::__mcInitApproximationScheme(), gum::learning::genericBNLearner::disableMaxIter(), and gum::learning::GreedyHillClimbing::GreedyHillClimbing().

{ _enabled_max_iter = false; }

Here is the caller graph for this function:

disableMaxTime()

INLINE void gum::ApproximationScheme::disableMaxTime ( )

virtualinherited

Disable stopping criterion on timeout.

Returns

Disable stopping criterion on timeout.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 131 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_enabled_max_time.

Referenced by gum::learning::genericBNLearner::disableMaxTime(), and gum::learning::GreedyHillClimbing::GreedyHillClimbing().

{ _enabled_max_time = false; }

Here is the caller graph for this function:

disableMinEpsilonRate()

INLINE void gum::ApproximationScheme::disableMinEpsilonRate ( )

virtualinherited

Disable stopping criterion on epsilon rate.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 79 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_enabled_min_rate_eps.

Referenced by gum::credal::CNMonteCarloSampling< GUM_SCALAR, BNInferenceEngine >::__mcInitApproximationScheme(), gum::GibbsBNdistance< GUM_SCALAR >::_computeKL(), gum::learning::genericBNLearner::disableMinEpsilonRate(), and gum::learning::GreedyHillClimbing::GreedyHillClimbing().

                                                         {
    _enabled_min_rate_eps = false;
  }

Here is the caller graph for this function:

enableEpsilon()

INLINE void gum::ApproximationScheme::enableEpsilon ( )

virtualinherited

Enable stopping criterion on epsilon.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 57 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_enabled_eps.

Referenced by gum::credal::CNMonteCarloSampling< GUM_SCALAR, BNInferenceEngine >::__mcInitApproximationScheme(), and gum::learning::genericBNLearner::enableEpsilon().

{ _enabled_eps = true; }

Here is the caller graph for this function:

enableMaxIter()

INLINE void gum::ApproximationScheme::enableMaxIter ( )

virtualinherited

Enable stopping criterion on max iterations.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 108 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_enabled_max_iter.

Referenced by gum::learning::genericBNLearner::enableMaxIter().

{ _enabled_max_iter = true; }

Here is the caller graph for this function:

enableMaxTime()

INLINE void gum::ApproximationScheme::enableMaxTime ( )

virtualinherited

Enable stopping criterion on timeout.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 134 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_enabled_max_time.

Referenced by gum::credal::CNMonteCarloSampling< GUM_SCALAR, BNInferenceEngine >::CNMonteCarloSampling(), and gum::learning::genericBNLearner::enableMaxTime().

{ _enabled_max_time = true; }

Here is the caller graph for this function:

enableMinEpsilonRate()

INLINE void gum::ApproximationScheme::enableMinEpsilonRate ( )

virtualinherited

Enable stopping criterion on epsilon rate.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 84 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_enabled_min_rate_eps.

Referenced by gum::GibbsBNdistance< GUM_SCALAR >::_computeKL(), and gum::learning::genericBNLearner::enableMinEpsilonRate().

                                                        {
    _enabled_min_rate_eps = true;
  }

Here is the caller graph for this function:

epsilon()

INLINE double gum::ApproximationScheme::epsilon ( ) const

virtualinherited

Returns the value of epsilon.

Returns

Returns the value of epsilon.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 51 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_eps.

Referenced by gum::ImportanceSampling< GUM_SCALAR >::_onContextualize(), and gum::learning::genericBNLearner::epsilon().

{ return _eps; }

Here is the caller graph for this function:

history()

INLINE const std::vector< double > & gum::ApproximationScheme::history ( ) const

virtualinherited

Returns the scheme history.

Returns

Returns the scheme history.

Exceptions

OperationNotAllowed

Raised if the scheme did not performed or if verbosity is set to false.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 173 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_history, GUM_ERROR, gum::ApproximationScheme::stateApproximationScheme(), gum::IApproximationSchemeConfiguration::Undefined, and gum::ApproximationScheme::verbosity().

Referenced by gum::learning::genericBNLearner::history().

                                                                       {
    if (stateApproximationScheme() == ApproximationSchemeSTATE::Undefined) {
      GUM_ERROR(OperationNotAllowed,
                "state of the approximation scheme is udefined");
    }

    if (verbosity() == false) {
      GUM_ERROR(OperationNotAllowed, "No history when verbosity=false");
    }

    return _history;
  }

Here is the call graph for this function:

Here is the caller graph for this function:

initApproximationScheme()

INLINE void gum::ApproximationScheme::initApproximationScheme ( )

inherited

Initialise the scheme.

Definition at line 187 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_current_epsilon, gum::ApproximationScheme::_current_rate, gum::ApproximationScheme::_current_state, gum::ApproximationScheme::_current_step, gum::ApproximationScheme::_history, gum::ApproximationScheme::_timer, gum::IApproximationSchemeConfiguration::Continue, and gum::Timer::reset().

Referenced by gum::credal::CNMonteCarloSampling< GUM_SCALAR, BNInferenceEngine >::__mcInitApproximationScheme(), gum::GibbsBNdistance< GUM_SCALAR >::_computeKL(), gum::SamplingInference< GUM_SCALAR >::_loopApproxInference(), gum::SamplingInference< GUM_SCALAR >::_onStateChanged(), gum::learning::DAG2BNLearner< ALLOC >::createBN(), gum::learning::GreedyHillClimbing::learnStructure(), and gum::learning::LocalSearchWithTabuList::learnStructure().

                                                           {
    _current_state = ApproximationSchemeSTATE::Continue;
    _current_step = 0;
    _current_epsilon = _current_rate = -1.0;
    _history.clear();
    _timer.reset();
  }

Here is the call graph for this function:

Here is the caller graph for this function:

isEnabledEpsilon()

INLINE bool gum::ApproximationScheme::isEnabledEpsilon ( ) const

virtualinherited

Returns true if stopping criterion on epsilon is enabled, false otherwise.

Returns

Returns true if stopping criterion on epsilon is enabled, false otherwise.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 61 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_enabled_eps.

Referenced by gum::learning::genericBNLearner::isEnabledEpsilon().

                                                          {
    return _enabled_eps;
  }

Here is the caller graph for this function:

isEnabledMaxIter()

INLINE bool gum::ApproximationScheme::isEnabledMaxIter ( ) const

virtualinherited

Returns true if stopping criterion on max iterations is enabled, false otherwise.

Returns

Returns true if stopping criterion on max iterations is enabled, false otherwise.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 112 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_enabled_max_iter.

Referenced by gum::learning::genericBNLearner::isEnabledMaxIter().

                                                          {
    return _enabled_max_iter;
  }

Here is the caller graph for this function:

isEnabledMaxTime()

INLINE bool gum::ApproximationScheme::isEnabledMaxTime ( ) const

virtualinherited

Returns true if stopping criterion on timeout is enabled, false otherwise.

Returns

Returns true if stopping criterion on timeout is enabled, false otherwise.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 138 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_enabled_max_time.

Referenced by gum::learning::genericBNLearner::isEnabledMaxTime().

                                                          {
    return _enabled_max_time;
  }

Here is the caller graph for this function:

isEnabledMinEpsilonRate()

INLINE bool gum::ApproximationScheme::isEnabledMinEpsilonRate ( ) const

virtualinherited

Returns true if stopping criterion on epsilon rate is enabled, false otherwise.

Returns

Returns true if stopping criterion on epsilon rate is enabled, false otherwise.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 90 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_enabled_min_rate_eps.

Referenced by gum::GibbsBNdistance< GUM_SCALAR >::_computeKL(), and gum::learning::genericBNLearner::isEnabledMinEpsilonRate().

                                                                 {
    return _enabled_min_rate_eps;
  }

Here is the caller graph for this function:

latentVariables()

const std::vector< Arc > gum::learning::Miic::latentVariables

(

)

const

get the list of arcs hiding latent variables

Definition at line 1048 of file Miic.cpp.

References __latent_couples.

Referenced by gum::learning::genericBNLearner::latentVariables().

                                                       {
      return __latent_couples;
    }

Here is the caller graph for this function:

learnBN()

template<typename GUM_SCALAR , typename GRAPH_CHANGES_SELECTOR , typename PARAM_ESTIMATOR >

BayesNet< GUM_SCALAR > gum::learning::Miic::learnBN	(	GRAPH_CHANGES_SELECTOR &	selector,
		PARAM_ESTIMATOR &	estimator,
		DAG	initial_dag = DAG()
	)

learns the structure and the parameters of a BN

Parameters

selector	A selector class that computes the best changes that can be applied and that enables the user to get them very easily. Typically, the selector is a GraphChangesSelector4DiGraph<SCORE, STRUCT_CONSTRAINT, GRAPH_CHANGES_GENERATOR>.
estimator	A estimator.
names	The variables names.
modal	the domain sizes of the random variables observed in the database
translator	The cell translator to use.
initial_dag	the DAG we start from for our learning

Definition at line 1056 of file Miic.cpp.

References learnStructure().

                                                                              {
      return DAG2BNLearner<>::createBN< GUM_SCALAR >(
         estimator, learnStructure(selector, initial_dag));
    }

Here is the call graph for this function:

learnMixedStructure()

MixedGraph gum::learning::Miic::learnMixedStructure	(	CorrectedMutualInformation<> &	I,
		MixedGraph	graph
	)

learns the structure of an Essential Graph

learns the structure of a MixedGraph

Parameters

I	A mutual information instance that will do the computations and has loaded the database.
graph	the MixedGraph we start from for the learning

Definition at line 112 of file Miic.cpp.

References __latent_couples, __usemiic, gum::ApproximationScheme::_current_step, _initiation(), _iteration(), _orientation_3off2(), _orientation_miic(), gum::ApproximationScheme::_timer, and gum::Timer::reset().

Referenced by gum::learning::genericBNLearner::learnMixedStructure(), and learnStructure().

                                                                              {
      _timer.reset();
      _current_step = 0;

      // clear the vector of latent arcs to be sure
      __latent_couples.clear();

      Heap<
         std::pair< std::tuple< NodeId, NodeId, NodeId, std::vector< NodeId > >*,
                    double >,
         GreaterPairOn2nd >
         _rank;

      HashTable< std::pair< NodeId, NodeId >, std::vector< NodeId > > sep_set;

      _initiation(I, graph, sep_set, _rank);

      _iteration(I, graph, sep_set, _rank);

      if (__usemiic) {
        _orientation_miic(I, graph, sep_set);
      } else {
        _orientation_3off2(I, graph, sep_set);
      }

      return graph;
    }

Here is the call graph for this function:

Here is the caller graph for this function:

learnStructure()

DAG gum::learning::Miic::learnStructure	(	CorrectedMutualInformation<> &	I,
		MixedGraph	graph
	)

learns the structure of an Bayesian network, ie a DAG, by first learning an Essential graph and then directing the remaining edges.

learns the structure of an Bayesian network, ie a DAG, from an Essential graph.

Parameters

I	A mutual information instance that will do the computations and has loaded the database
graph	the MixedGraph we start from for the learning

Definition at line 986 of file Miic.cpp.

References _propagatesHead(), gum::DAG::addArc(), gum::NodeGraphPart::addNodeWithId(), gum::Set< Key, Alloc >::empty(), learnMixedStructure(), gum::EdgeGraphPart::neighbours(), gum::ArcGraphPart::parents(), and gum::DiGraph::topologicalOrder().

Referenced by gum::learning::genericBNLearner::__learnDAG(), and learnBN().

                                                                         {
      MixedGraph essentialGraph = learnMixedStructure(I, initialGraph);

      // Second, orientate remaining edges
      const Sequence< NodeId > order = essentialGraph.topologicalOrder();
      // first, propagate existing orientations
      for (NodeId x: order) {
        if (!essentialGraph.parents(x).empty()) {
          _propagatesHead(essentialGraph, x);
        }
      }
      // then decide the orientation by the topological order and propagate them
      for (NodeId x: order) {
        if (!essentialGraph.neighbours(x).empty()) {
          _propagatesHead(essentialGraph, x);
        }
      }

      // turn the mixed graph into a dag
      DAG dag;
      for (auto node: essentialGraph) {
        dag.addNodeWithId(node);
      }
      for (const Arc& arc: essentialGraph.arcs()) {
        dag.addArc(arc.tail(), arc.head());
      }

      return dag;
    }

Here is the call graph for this function:

Here is the caller graph for this function:

maxIter()

INLINE Size gum::ApproximationScheme::maxIter ( ) const

virtualinherited

Returns the criterion on number of iterations.

Returns

Returns the criterion on number of iterations.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 102 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_max_iter.

Referenced by gum::learning::genericBNLearner::maxIter().

{ return _max_iter; }

Here is the caller graph for this function:

maxTime()

INLINE double gum::ApproximationScheme::maxTime ( ) const

virtualinherited

Returns the timeout (in seconds).

Returns

Returns the timeout (in seconds).

Implements gum::IApproximationSchemeConfiguration.

Definition at line 125 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_max_time.

Referenced by gum::learning::genericBNLearner::maxTime().

{ return _max_time; }

Here is the caller graph for this function:

messageApproximationScheme()

INLINE std::string gum::IApproximationSchemeConfiguration::messageApproximationScheme ( ) const

inherited

Returns the approximation scheme message.

Returns

Returns the approximation scheme message.

Definition at line 40 of file IApproximationSchemeConfiguration_inl.h.

References gum::IApproximationSchemeConfiguration::Continue, gum::IApproximationSchemeConfiguration::Epsilon, gum::IApproximationSchemeConfiguration::epsilon(), gum::IApproximationSchemeConfiguration::Limit, gum::IApproximationSchemeConfiguration::maxIter(), gum::IApproximationSchemeConfiguration::maxTime(), gum::IApproximationSchemeConfiguration::minEpsilonRate(), gum::IApproximationSchemeConfiguration::Rate, gum::IApproximationSchemeConfiguration::stateApproximationScheme(), gum::IApproximationSchemeConfiguration::Stopped, gum::IApproximationSchemeConfiguration::TimeLimit, and gum::IApproximationSchemeConfiguration::Undefined.

Referenced by gum::ApproximationScheme::_stopScheme(), gum::ApproximationScheme::continueApproximationScheme(), and gum::credal::InferenceEngine< GUM_SCALAR >::getApproximationSchemeMsg().

                                                                         {
    std::stringstream s;

    switch (stateApproximationScheme()) {
      case ApproximationSchemeSTATE::Continue: s << "in progress"; break;

      case ApproximationSchemeSTATE::Epsilon:
        s << "stopped with epsilon=" << epsilon();
        break;

      case ApproximationSchemeSTATE::Rate:
        s << "stopped with rate=" << minEpsilonRate();
        break;

      case ApproximationSchemeSTATE::Limit:
        s << "stopped with max iteration=" << maxIter();
        break;

      case ApproximationSchemeSTATE::TimeLimit:
        s << "stopped with timeout=" << maxTime();
        break;

      case ApproximationSchemeSTATE::Stopped: s << "stopped on request"; break;

      case ApproximationSchemeSTATE::Undefined: s << "undefined state"; break;
    };

    return s.str();
  }

Here is the call graph for this function:

Here is the caller graph for this function:

minEpsilonRate()

INLINE double gum::ApproximationScheme::minEpsilonRate ( ) const

virtualinherited

Returns the value of the minimal epsilon rate.

Returns

Returns the value of the minimal epsilon rate.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 74 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_min_rate_eps.

Referenced by gum::learning::genericBNLearner::minEpsilonRate().

                                                          {
    return _min_rate_eps;
  }

Here is the caller graph for this function:

nbrIterations()

INLINE Size gum::ApproximationScheme::nbrIterations ( ) const

virtualinherited

Returns the number of iterations.

Returns

Returns the number of iterations.

Exceptions

OperationNotAllowed

Raised if the scheme did not perform.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 163 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_current_step, GUM_ERROR, gum::ApproximationScheme::stateApproximationScheme(), and gum::IApproximationSchemeConfiguration::Undefined.

Referenced by gum::GibbsBNdistance< GUM_SCALAR >::_computeKL(), and gum::learning::genericBNLearner::nbrIterations().

                                                       {
    if (stateApproximationScheme() == ApproximationSchemeSTATE::Undefined) {
      GUM_ERROR(OperationNotAllowed,
                "state of the approximation scheme is undefined");
    }

    return _current_step;
  }

Here is the call graph for this function:

Here is the caller graph for this function:

operator=() [1/2]

Miic & gum::learning::Miic::operator=

(

const Miic &

from

)

copy operator

Definition at line 62 of file Miic.cpp.

                                          {
      ApproximationScheme::operator=(from);
      return *this;
    }

operator=() [2/2]

Miic & gum::learning::Miic::operator=

(

Miic &&

from

)

move operator

Definition at line 68 of file Miic.cpp.

                                     {
      ApproximationScheme::operator=(std::move(from));
      return *this;
    }

periodSize()

INLINE Size gum::ApproximationScheme::periodSize ( ) const

virtualinherited

Returns the period size.

Returns

Returns the period size.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 149 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_period_size.

Referenced by gum::credal::CNMonteCarloSampling< GUM_SCALAR, BNInferenceEngine >::makeInference(), and gum::learning::genericBNLearner::periodSize().

{ return _period_size; }

Here is the caller graph for this function:

remainingBurnIn()

INLINE Size gum::ApproximationScheme::remainingBurnIn ( )

inherited

Returns the remaining burn in.

Returns

Returns the remaining burn in.

Definition at line 210 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_burn_in, and gum::ApproximationScheme::_current_step.

                                                   {
    if (_burn_in > _current_step) {
      return _burn_in - _current_step;
    } else {
      return 0;
    }
  }

set3off2Behaviour()

void gum::learning::Miic::set3off2Behaviour

(

)

Sets the orientation phase to follow the one of the 3off2 algorithm.

Definition at line 1064 of file Miic.cpp.

References __usemiic.

Referenced by gum::learning::genericBNLearner::use3off2().

{ this->__usemiic = false; }

Here is the caller graph for this function:

setEpsilon()

INLINE void gum::ApproximationScheme::setEpsilon ( double  eps )

virtualinherited

Given that we approximate f(t), stopping criterion on |f(t+1)-f(t)|.

If the criterion was disabled it will be enabled.

Parameters

eps	The new epsilon value.

Exceptions

OutOfLowerBound

Raised if eps < 0.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 43 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_enabled_eps, gum::ApproximationScheme::_eps, and GUM_ERROR.

Referenced by gum::credal::CNMonteCarloSampling< GUM_SCALAR, BNInferenceEngine >::__mcInitApproximationScheme(), gum::GibbsBNdistance< GUM_SCALAR >::GibbsBNdistance(), gum::GibbsSampling< GUM_SCALAR >::GibbsSampling(), gum::learning::GreedyHillClimbing::GreedyHillClimbing(), gum::SamplingInference< GUM_SCALAR >::SamplingInference(), and gum::learning::genericBNLearner::setEpsilon().

                                                        {
    if (eps < 0.) { GUM_ERROR(OutOfLowerBound, "eps should be >=0"); }

    _eps = eps;
    _enabled_eps = true;
  }

Here is the caller graph for this function:

setMaxIter()

INLINE void gum::ApproximationScheme::setMaxIter ( Size  max )

virtualinherited

Stopping criterion on number of iterations.

If the criterion was disabled it will be enabled.

Parameters

max	The maximum number of iterations.

Exceptions

OutOfLowerBound

Raised if max <= 1.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 95 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_enabled_max_iter, gum::ApproximationScheme::_max_iter, and GUM_ERROR.

Referenced by gum::GibbsBNdistance< GUM_SCALAR >::GibbsBNdistance(), gum::SamplingInference< GUM_SCALAR >::SamplingInference(), and gum::learning::genericBNLearner::setMaxIter().

                                                      {
    if (max < 1) { GUM_ERROR(OutOfLowerBound, "max should be >=1"); }
    _max_iter = max;
    _enabled_max_iter = true;
  }

Here is the caller graph for this function:

setMaxTime()

INLINE void gum::ApproximationScheme::setMaxTime ( double  timeout )

virtualinherited

Stopping criterion on timeout.

If the criterion was disabled it will be enabled.

Parameters

timeout

The timeout value in seconds.

Exceptions

OutOfLowerBound

Raised if timeout <= 0.0.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 118 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_enabled_max_time, gum::ApproximationScheme::_max_time, and GUM_ERROR.

Referenced by gum::credal::CNMonteCarloSampling< GUM_SCALAR, BNInferenceEngine >::CNMonteCarloSampling(), gum::GibbsBNdistance< GUM_SCALAR >::GibbsBNdistance(), gum::SamplingInference< GUM_SCALAR >::SamplingInference(), and gum::learning::genericBNLearner::setMaxTime().

                                                            {
    if (timeout <= 0.) { GUM_ERROR(OutOfLowerBound, "timeout should be >0."); }
    _max_time = timeout;
    _enabled_max_time = true;
  }

Here is the caller graph for this function:

setMiicBehaviour()

void gum::learning::Miic::setMiicBehaviour

(

)

Sets the orientation phase to follow the one of the MIIC algorithm.

Definition at line 1063 of file Miic.cpp.

References __usemiic.

Referenced by gum::learning::genericBNLearner::useMIIC().

{ this->__usemiic = true; }

Here is the caller graph for this function:

setMinEpsilonRate()

INLINE void gum::ApproximationScheme::setMinEpsilonRate ( double  rate )

virtualinherited

Given that we approximate f(t), stopping criterion on d/dt(|f(t+1)-f(t)|).

If the criterion was disabled it will be enabled

Parameters

rate	The minimal epsilon rate.

Exceptions

OutOfLowerBound

if rate<0

Implements gum::IApproximationSchemeConfiguration.

Definition at line 66 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_enabled_min_rate_eps, gum::ApproximationScheme::_min_rate_eps, and GUM_ERROR.

Referenced by gum::GibbsBNdistance< GUM_SCALAR >::GibbsBNdistance(), gum::GibbsSampling< GUM_SCALAR >::GibbsSampling(), gum::SamplingInference< GUM_SCALAR >::SamplingInference(), and gum::learning::genericBNLearner::setMinEpsilonRate().

                                                                {
    if (rate < 0) { GUM_ERROR(OutOfLowerBound, "rate should be >=0"); }

    _min_rate_eps = rate;
    _enabled_min_rate_eps = true;
  }

Here is the caller graph for this function:

setPeriodSize()

INLINE void gum::ApproximationScheme::setPeriodSize ( Size  p )

virtualinherited

How many samples between two stopping is enable.

Parameters

p	The new period value.

Exceptions

OutOfLowerBound

Raised if p < 1.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 143 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_period_size, and GUM_ERROR.

Referenced by gum::credal::CNMonteCarloSampling< GUM_SCALAR, BNInferenceEngine >::CNMonteCarloSampling(), gum::GibbsBNdistance< GUM_SCALAR >::GibbsBNdistance(), gum::SamplingInference< GUM_SCALAR >::SamplingInference(), and gum::learning::genericBNLearner::setPeriodSize().

                                                       {
    if (p < 1) { GUM_ERROR(OutOfLowerBound, "p should be >=1"); }

    _period_size = p;
  }

Here is the caller graph for this function:

setVerbosity()

INLINE void gum::ApproximationScheme::setVerbosity ( bool  v )

virtualinherited

Set the verbosity on (true) or off (false).

Parameters

v	If true, then verbosity is turned on.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 152 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_verbosity.

Referenced by gum::GibbsBNdistance< GUM_SCALAR >::GibbsBNdistance(), gum::SamplingInference< GUM_SCALAR >::SamplingInference(), and gum::learning::genericBNLearner::setVerbosity().

{ _verbosity = v; }

Here is the caller graph for this function:

startOfPeriod()

INLINE bool gum::ApproximationScheme::startOfPeriod ( )

inherited

Returns true if we are at the beginning of a period (compute error is mandatory).

Returns

Returns true if we are at the beginning of a period (compute error is mandatory).

Definition at line 197 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_burn_in, gum::ApproximationScheme::_current_step, and gum::ApproximationScheme::_period_size.

Referenced by gum::ApproximationScheme::continueApproximationScheme().

                                                 {
    if (_current_step < _burn_in) { return false; }

    if (_period_size == 1) { return true; }

    return ((_current_step - _burn_in) % _period_size == 0);
  }

Here is the caller graph for this function:

stateApproximationScheme()

INLINE IApproximationSchemeConfiguration::ApproximationSchemeSTATE gum::ApproximationScheme::stateApproximationScheme ( ) const

virtualinherited

Returns the approximation scheme state.

Returns

Returns the approximation scheme state.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 158 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_current_state.

Referenced by gum::ApproximationScheme::continueApproximationScheme(), gum::ApproximationScheme::history(), gum::ApproximationScheme::nbrIterations(), and gum::learning::genericBNLearner::stateApproximationScheme().

                                                             {
    return _current_state;
  }

Here is the caller graph for this function:

stopApproximationScheme()

INLINE void gum::ApproximationScheme::stopApproximationScheme ( )

inherited

Stop the approximation scheme.

Definition at line 219 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_current_state, gum::ApproximationScheme::_stopScheme(), gum::IApproximationSchemeConfiguration::Continue, and gum::IApproximationSchemeConfiguration::Stopped.

Referenced by gum::learning::DAG2BNLearner< ALLOC >::createBN(), gum::learning::GreedyHillClimbing::learnStructure(), and gum::learning::LocalSearchWithTabuList::learnStructure().

                                                           {
    if (_current_state == ApproximationSchemeSTATE::Continue) {
      _stopScheme(ApproximationSchemeSTATE::Stopped);
    }
  }

Here is the call graph for this function:

Here is the caller graph for this function:

updateApproximationScheme()

INLINE void gum::ApproximationScheme::updateApproximationScheme ( unsigned int  incr = 1 )

inherited

Update the scheme w.r.t the new error and increment steps.

Parameters

incr	The new increment steps.

Definition at line 206 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_current_step.

Referenced by gum::GibbsBNdistance< GUM_SCALAR >::_computeKL(), gum::SamplingInference< GUM_SCALAR >::_loopApproxInference(), gum::learning::DAG2BNLearner< ALLOC >::createBN(), gum::learning::GreedyHillClimbing::learnStructure(), gum::learning::LocalSearchWithTabuList::learnStructure(), and gum::credal::CNMonteCarloSampling< GUM_SCALAR, BNInferenceEngine >::makeInference().

                                                                              {
    _current_step += incr;
  }

Here is the caller graph for this function:

verbosity()

INLINE bool gum::ApproximationScheme::verbosity ( ) const

virtualinherited

Returns true if verbosity is enabled.

Returns

Returns true if verbosity is enabled.

Implements gum::IApproximationSchemeConfiguration.

Definition at line 154 of file approximationScheme_inl.h.

References gum::ApproximationScheme::_verbosity.

Referenced by gum::ApproximationScheme::continueApproximationScheme(), gum::ApproximationScheme::history(), and gum::learning::genericBNLearner::verbosity().

{ return _verbosity; }

Here is the caller graph for this function:

Member Data Documentation

__arc_probas

ArcProperty< double > gum::learning::Miic::__arc_probas

private

Storing the propabilities for each arc set in the graph.

Definition at line 362 of file Miic.h.

Referenced by _orientation_miic().

__empty_set

const std::vector< NodeId > gum::learning::Miic::__empty_set

private

an empty conditioning set

Definition at line 352 of file Miic.h.

Referenced by _initiation().

__initial_marks

HashTable< std::pair< NodeId, NodeId >, char > gum::learning::Miic::__initial_marks

private

Initial marks for the orientation phase, used to convey constraints.

Definition at line 365 of file Miic.h.

Referenced by _orientation_3off2(), _orientation_miic(), _propagatesHead(), and addConstraints().

__latent_couples

std::vector< Arc > gum::learning::Miic::__latent_couples

private

an empty vector of arcs

Definition at line 354 of file Miic.h.

Referenced by _orientation_3off2(), _orientation_latents(), _orientation_miic(), _propagatesHead(), latentVariables(), and learnMixedStructure().

__maxLog

int gum::learning::Miic::__maxLog = 100

private

Fixes the maximum log that we accept in exponential computations.

Definition at line 350 of file Miic.h.

Referenced by _findBestContributor().

__N

Size gum::learning::Miic::__N

private

size of the database

Definition at line 357 of file Miic.h.

__usemiic

bool gum::learning::Miic::__usemiic {false}

private

wether to use the miic algorithm or not

Definition at line 359 of file Miic.h.

Referenced by learnMixedStructure(), set3off2Behaviour(), and setMiicBehaviour().

_burn_in

Size gum::ApproximationScheme::_burn_in

protectedinherited

Number of iterations before checking stopping criteria.

Definition at line 414 of file approximationScheme.h.

Referenced by gum::GibbsSampling< GUM_SCALAR >::burnIn(), gum::GibbsBNdistance< GUM_SCALAR >::burnIn(), gum::ApproximationScheme::remainingBurnIn(), gum::GibbsSampling< GUM_SCALAR >::setBurnIn(), gum::GibbsBNdistance< GUM_SCALAR >::setBurnIn(), and gum::ApproximationScheme::startOfPeriod().

_current_epsilon

double gum::ApproximationScheme::_current_epsilon

protectedinherited

Current epsilon.

Definition at line 369 of file approximationScheme.h.

Referenced by gum::ApproximationScheme::continueApproximationScheme(), and gum::ApproximationScheme::initApproximationScheme().

_current_rate

double gum::ApproximationScheme::_current_rate

protectedinherited

Current rate.

Definition at line 375 of file approximationScheme.h.

Referenced by gum::ApproximationScheme::continueApproximationScheme(), and gum::ApproximationScheme::initApproximationScheme().

_current_state

ApproximationSchemeSTATE gum::ApproximationScheme::_current_state

protectedinherited

The current state.

Definition at line 384 of file approximationScheme.h.

Referenced by gum::ApproximationScheme::_stopScheme(), gum::ApproximationScheme::continueApproximationScheme(), gum::ApproximationScheme::initApproximationScheme(), gum::ApproximationScheme::stateApproximationScheme(), and gum::ApproximationScheme::stopApproximationScheme().

_current_step

Size gum::ApproximationScheme::_current_step

protectedinherited

The current step.

Definition at line 378 of file approximationScheme.h.

Referenced by _initiation(), _iteration(), _orientation_3off2(), _orientation_latents(), _orientation_miic(), gum::ApproximationScheme::continueApproximationScheme(), gum::ApproximationScheme::initApproximationScheme(), learnMixedStructure(), gum::ApproximationScheme::nbrIterations(), gum::ApproximationScheme::remainingBurnIn(), gum::ApproximationScheme::startOfPeriod(), and gum::ApproximationScheme::updateApproximationScheme().

_enabled_eps

bool gum::ApproximationScheme::_enabled_eps

protectedinherited

If true, the threshold convergence is enabled.

Definition at line 393 of file approximationScheme.h.

Referenced by gum::ApproximationScheme::continueApproximationScheme(), gum::ApproximationScheme::disableEpsilon(), gum::ApproximationScheme::enableEpsilon(), gum::ApproximationScheme::isEnabledEpsilon(), and gum::ApproximationScheme::setEpsilon().

_enabled_max_iter

bool gum::ApproximationScheme::_enabled_max_iter

protectedinherited

If true, the maximum iterations stopping criterion is enabled.

Definition at line 411 of file approximationScheme.h.

Referenced by gum::ApproximationScheme::continueApproximationScheme(), gum::ApproximationScheme::disableMaxIter(), gum::ApproximationScheme::enableMaxIter(), gum::ApproximationScheme::isEnabledMaxIter(), and gum::ApproximationScheme::setMaxIter().

_enabled_max_time

bool gum::ApproximationScheme::_enabled_max_time

protectedinherited

If true, the timeout is enabled.

Definition at line 405 of file approximationScheme.h.

Referenced by gum::ApproximationScheme::continueApproximationScheme(), gum::ApproximationScheme::disableMaxTime(), gum::ApproximationScheme::enableMaxTime(), gum::ApproximationScheme::isEnabledMaxTime(), and gum::ApproximationScheme::setMaxTime().

_enabled_min_rate_eps

bool gum::ApproximationScheme::_enabled_min_rate_eps

protectedinherited

If true, the minimal threshold for epsilon rate is enabled.

Definition at line 399 of file approximationScheme.h.

Referenced by gum::ApproximationScheme::continueApproximationScheme(), gum::ApproximationScheme::disableMinEpsilonRate(), gum::ApproximationScheme::enableMinEpsilonRate(), gum::ApproximationScheme::isEnabledMinEpsilonRate(), and gum::ApproximationScheme::setMinEpsilonRate().

_eps

double gum::ApproximationScheme::_eps

protectedinherited

Threshold for convergence.

Definition at line 390 of file approximationScheme.h.

Referenced by gum::ApproximationScheme::continueApproximationScheme(), gum::ApproximationScheme::epsilon(), and gum::ApproximationScheme::setEpsilon().

_history

std::vector< double > gum::ApproximationScheme::_history

protectedinherited

The scheme history, used only if verbosity == true.

Definition at line 387 of file approximationScheme.h.

Referenced by gum::ApproximationScheme::continueApproximationScheme(), gum::ApproximationScheme::history(), and gum::ApproximationScheme::initApproximationScheme().

_last_epsilon

double gum::ApproximationScheme::_last_epsilon

protectedinherited

Last epsilon value.

Definition at line 372 of file approximationScheme.h.

Referenced by gum::ApproximationScheme::continueApproximationScheme().

_max_iter

Size gum::ApproximationScheme::_max_iter

protectedinherited

The maximum iterations.

Definition at line 408 of file approximationScheme.h.

Referenced by gum::ApproximationScheme::continueApproximationScheme(), gum::ApproximationScheme::maxIter(), and gum::ApproximationScheme::setMaxIter().

_max_time

double gum::ApproximationScheme::_max_time

protectedinherited

The timeout.

Definition at line 402 of file approximationScheme.h.

Referenced by gum::ApproximationScheme::continueApproximationScheme(), gum::ApproximationScheme::maxTime(), and gum::ApproximationScheme::setMaxTime().

_min_rate_eps

double gum::ApproximationScheme::_min_rate_eps

protectedinherited

Threshold for the epsilon rate.

Definition at line 396 of file approximationScheme.h.

Referenced by gum::ApproximationScheme::continueApproximationScheme(), gum::ApproximationScheme::minEpsilonRate(), and gum::ApproximationScheme::setMinEpsilonRate().

_period_size

Size gum::ApproximationScheme::_period_size

protectedinherited

Checking criteria frequency.

Definition at line 417 of file approximationScheme.h.

Referenced by gum::ApproximationScheme::periodSize(), gum::ApproximationScheme::setPeriodSize(), and gum::ApproximationScheme::startOfPeriod().

_timer

Timer gum::ApproximationScheme::_timer

protectedinherited

The timer.

Definition at line 381 of file approximationScheme.h.

Referenced by _initiation(), _iteration(), _orientation_3off2(), _orientation_latents(), _orientation_miic(), gum::ApproximationScheme::_stopScheme(), gum::ApproximationScheme::continueApproximationScheme(), gum::ApproximationScheme::currentTime(), gum::ApproximationScheme::initApproximationScheme(), and learnMixedStructure().

_verbosity

bool gum::ApproximationScheme::_verbosity

protectedinherited

If true, verbosity is enabled.

Definition at line 420 of file approximationScheme.h.

Referenced by gum::ApproximationScheme::setVerbosity(), and gum::ApproximationScheme::verbosity().

onProgress

Signaler3< Size, double, double > gum::IApproximationSchemeConfiguration::onProgress

inherited

Progression, error and time.

Definition at line 59 of file IApproximationSchemeConfiguration.h.

Referenced by _initiation(), _iteration(), _orientation_3off2(), _orientation_latents(), _orientation_miic(), gum::ApproximationScheme::continueApproximationScheme(), and gum::learning::genericBNLearner::distributeProgress().

onStop

Signaler1< std::string > gum::IApproximationSchemeConfiguration::onStop

inherited

Criteria messageApproximationScheme.

Definition at line 62 of file IApproximationSchemeConfiguration.h.

Referenced by gum::ApproximationScheme::_stopScheme(), and gum::learning::genericBNLearner::distributeStop().

---

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

• agrum/BN/learning/Miic.h
• agrum/BN/learning/Miic.cpp