![]() |
aGrUM 3.0.0
a C++ library for (probabilistic) graphical models
|
Abstract base class for CI-test-based structure learning algorithms. More...
#include <vector>#include <agrum/base/stattests/independenceTest.h>#include <agrum/BN/learning/ConstraintBasedLearning.h>Go to the source code of this file.
Classes | |
| class | gum::learning::CIBasedLearning |
| Abstract base for CI-test-based causal structure learning (PC, FCI). More... | |
| struct | gum::learning::CIBasedLearning::SepSetEntry_ |
Namespaces | |
| namespace | gum |
| gum is the global namespace for all aGrUM entities | |
| namespace | gum::learning |
| include the inlined functions if necessary | |
Abstract base class for CI-test-based structure learning algorithms.
CIBasedLearning sits between ConstraintBasedLearning and concrete algorithms (PC, FCI). It holds all state shared by CI-test algorithms: the independence test, alpha threshold, stable-PC mode, conditioning-set size, separation sets, and unshielded-collider orientation logic.
Concrete subclasses implement learnMixedStructure().
Definition in file CIBasedLearning.h.