Class AdequacyReport

java.lang.Object
edu.cmu.tetrad.sem.AdequacyReport
All Implemented Interfaces:
TetradSerializable, Serializable

public final class AdequacyReport extends Object implements TetradSerializable
Represents a report evaluating the adequacy of a model's performance relative to a baseline. This class captures key metrics, including the overall mean improvement of nodes, the fraction of nodes that demonstrated improvement, global data alignment metrics, and per-node performance summaries. It is designed to provide a comprehensive assessment of model adequacy. Instances of this class are immutable, ensuring their contents cannot be modified after creation.
See Also:
  • Field Summary

    Fields
    Modifier and Type
    Field
    Description
    final double
    Represents the fraction
    final double
    Represents the mean improvement of nodes over a baseline metric in the context of an adequacy report.
    final double
    Represents the global Maximum Mean Discrepancy squared (MMD²) value between the real and simulated data distributions in the context of an adequacy report.
    Represents a collection of adequacy summaries for individual nodes within a model.
  • Method Summary

    Modifier and Type
    Method
    Description
    double
    Retrieves the fraction of nodes that demonstrated performance improvement relative to the baseline.
    double
    Retrieves the mean improvement of nodes over a baseline metric.
    double
    Retrieves the squared Maximum Mean Discrepancy (MMD²) value, which quantifies the alignment between real and simulated data distributions.
    Converts the adequacy report data into a human-readable textual representation.

    Methods inherited from class Object

    equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
  • Field Details

    • meanImprovement

      public final double meanImprovement
      Represents the mean improvement of nodes over a baseline metric in the context of an adequacy report.

      This value indicates the average magnitude of improvement across all nodes, providing a summary measure for assessing overall performance enhancement.

    • fracNodesImproved

      public final double fracNodesImproved
      Represents the fraction
    • mmd2

      public final double mmd2
      Represents the global Maximum Mean Discrepancy squared (MMD²) value between the real and simulated data distributions in the context of an adequacy report.

      This metric provides a quantitative measure of how well the simulated data aligns with the real data, with smaller values indicating closer alignment. It is useful in evaluating the overall representational quality of the simulation process.

    • nodeSummaries

      public final List<NodeAdequacySummary> nodeSummaries
      Represents a collection of adequacy summaries for individual nodes within a model.

      Each element in this list provides detailed metrics and performance insights for a specific node, including its name, parents, and performance improvements relative to a baseline. This field is used as part of the adequacy report to evaluate the performance and improvements of nodes in the context of the overall model.

      This list is immutable and provides a comprehensive snapshot of per-node evaluation results.

  • Method Details

    • toText

      public String toText()
      Converts the adequacy report data into a human-readable textual representation.

      The resulting text includes global metrics, such as mean improvement over the baseline, the fraction of nodes that showed improvement, and the global MMD^2 value. It also provides a per-node breakdown of holdout improvements, listing each node along with its specific improvement value.

      Returns:
      A string representation of the adequacy report, including global and per-node metrics formatted for readability.
    • getMmd2

      public double getMmd2()
      Retrieves the squared Maximum Mean Discrepancy (MMD²) value, which quantifies the alignment between real and simulated data distributions.
      Returns:
      The global MMD² value as a double, representing the measure of distributional similarity.
    • getMeanImprovement

      public double getMeanImprovement()
      Retrieves the mean improvement of nodes over a baseline metric.
      Returns:
      The mean improvement value as a double, indicating the average improvement across nodes.
    • getFracImproved

      public double getFracImproved()
      Retrieves the fraction of nodes that demonstrated performance improvement relative to the baseline.
      Returns:
      The fraction of nodes improved as a double, representing the proportion of nodes with performance enhancement.