Package edu.cmu.tetrad.search.unmix

Class UnmixDiagnostics.GraphDiff

java.lang.Object

edu.cmu.tetrad.search.unmix.UnmixDiagnostics.GraphDiff

Enclosing class:

UnmixDiagnostics

public static final class UnmixDiagnostics.GraphDiff
extends Object

Represents the difference between two clustering graphs in terms of structural and adjacency metrics. The metrics include adjacency F1 score, arrow F1 score, and Structural Hamming Distance (SHD).

Field Summary

Fields

Modifier and Type	Field	Description
final double	adjacencyF1	Represents the adjacency F1 score, which measures the similarity between the adjacency structures of two graphs in terms of precision and recall.
final double	arrowF1	Represents the arrow F1 score, which measures the similarity of directed edges (arrows) between two graphs.
final int	shd	Represents the Structural Hamming Distance (SHD) between two graphs.

Constructor Summary

Constructors

Constructor	Description
GraphDiff(double adjacencyF1, double arrowF1, int shd)	Constructs a GraphDiff instance representing the difference between two graphs based on adjacency F1 score, arrow F1 score, and Structural Hamming Distance (SHD).

Method Summary

Methods inherited from class java.lang.Object

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

Field Details

adjacencyF1

public final double adjacencyF1

Represents the adjacency F1 score, which measures the similarity between the adjacency structures of two graphs in terms of precision and recall. The adjacency F1 score is a commonly used metric in graph comparison to evaluate how well nodes and edges align between two graphs.

arrowF1

public final double arrowF1

Represents the arrow F1 score, which measures the similarity of directed edges (arrows) between two graphs. The metric evaluates precision and recall for correctly inferred arrow directions in a structural comparison of graphs.

shd

public final int shd

Represents the Structural Hamming Distance (SHD) between two graphs. SHD quantifies the structural difference between two graphs by counting the number of edits (addition, deletion, or change of edges or directions) required to transform one graph into another.

Constructor Details

GraphDiff

public GraphDiff(double adjacencyF1,
 double arrowF1,
 int shd)

Constructs a GraphDiff instance representing the difference between two graphs based on adjacency F1 score, arrow F1 score, and Structural Hamming Distance (SHD).

Parameters:

adjacencyF1 - The adjacency F1 score, which measures the similarity of the adjacency structures of two graphs.

arrowF1 - The arrow F1 score, which evaluates the similarity of the directed edges in two graphs.

shd - The Structural Hamming Distance (SHD), representing the number of edits required to make the two graphs identical.