32 #ifndef GUM_LEARNING_LOCAL_SEARCH_WITH_TABU_LIST_H 33 #define GUM_LEARNING_LOCAL_SEARCH_WITH_TABU_LIST_H 110 template <
typename GRAPH_CHANGES_
SELECTOR >
115 template <
typename GUM_SCALAR =
double,
116 typename GRAPH_CHANGES_SELECTOR,
117 typename PARAM_ESTIMATOR >
119 PARAM_ESTIMATOR& estimator,
134 #ifndef GUM_NO_INLINE Class representing a Bayesian Network.
This file contains general scheme for iteratively convergent algorithms.
LocalSearchWithTabuList()
default constructor
LocalSearchWithTabuList & operator=(const LocalSearchWithTabuList &from)
copy operator
Class representing Bayesian networks.
gum is the global namespace for all aGrUM entities
The local search with tabu list learning algorithm (for directed graphs)
virtual ~LocalSearchWithTabuList()
destructor
ApproximationScheme & approximationScheme()
returns the approximation policy of the learning algorithm
BayesNet< GUM_SCALAR > learnBN(GRAPH_CHANGES_SELECTOR &selector, PARAM_ESTIMATOR &estimator, DAG initial_dag=DAG())
learns the structure and the parameters of a BN
Size __MaxNbDecreasing
the max number of changes decreasing the score that we allow to apply
The local search with tabu list learning algorithm (for directed graphs)
The local search learning algorithm (for directed graphs)
DAG learnStructure(GRAPH_CHANGES_SELECTOR &selector, DAG initial_dag=DAG())
learns the structure of a Bayes net
std::size_t Size
In aGrUM, hashed values are unsigned long int.
void setMaxNbDecreasingChanges(Size nb)
set the max number of changes decreasing the score that we allow to apply
Base classes for directed acyclic graphs.