Package edu.cmu.tetrad.hybridcg

Class HybridCgModel

java.lang.Object
edu.cmu.tetrad.hybridcg.HybridCgModel
public final class HybridCgModel extends Object
Mixed Continuous/Discrete model with discretization for discrete children that have continuous parents.

Design goals:

  • Mirror BayesPm/MlBayesIm and SemPm/SemIm ergonomics where sensible.
  • Allow mixed parent sets. For a continuous child: stratify by discrete parents and fit linear-Gaussian per stratum. For a discrete child: build a CPT whose rows are the cross-product of discrete-parent states and discretized bins of continuous-parent values (bin edges kept in the PM).
  • Keep a stable row-indexing contract so tables can be read/written and scored efficiently.

  • Nested Class Summary

    Nested Classes
    Modifier and Type
    Class
    Description
    static final class 
    HybridCgModel.HybridCgIm
    Represents a hybrid causal graph instantiated model (IM) derived from a hybrid causal graph probabilistic model (PM).
    static final class 
    HybridCgModel.HybridCgPm
    The HybridCgPm class represents a structural model for hybrid Bayesian networks, which may include both discrete and continuous variables.

  • Constructor Summary

    Constructors
    Constructor
    Description
    HybridCgModel()
    Constructs a new instance of the HybridCgModel class.

  • Method Summary

    Methods inherited from class java.lang.Object

    equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

  • Constructor Details

    • HybridCgModel

      public HybridCgModel()
      Constructs a new instance of the HybridCgModel class. This class serves as the foundational representation of a mixed graphical model encompassing both continuous and discrete variables, as well as their dependencies. The HybridCgModel offers a comprehensive framework for combining different variable types within a single hybrid representation.