Class TRffBicScore

java.lang.Object
edu.cmu.tetrad.algcomparison.score.TRffBicScore
All Implemented Interfaces:
ScoreWrapper, HasParameters, TetradSerializable, Serializable
@Score(name="TRFF BIC Score", command="trff-bic-score", dataType=Mixed) @General @Mixed public class TRffBicScore extends Object implements ScoreWrapper
The TRffBicScore class is a scoring mechanism designed to evaluate the fit and complexity of a data model. This score uses the Student-T Fourier Feature Bayes Information Criterion (BIC), which is particularly suitable for mixed-data settings. The class provides parameterized control over regularization, feature tuning, and penalty adjustments, allowing flexibility during model evaluation.

The scoring mechanism is implemented within the Tetrad framework and adheres to the ScoreWrapper interface. It is annotated with Score, specifying the name, command, and applicable data type.

See Also:

  • Constructor Details

    • TRffBicScore

      public TRffBicScore()
      Constructor for the TRffBicScore class. This class serves as a Student-T Fourier Feature Bayes Information Criterion (BIC) scoring mechanism typically applied in mixed-data settings.

      The score is used in conjunction with algorithms that evaluate the fit and complexity of a model based on penalized likelihood principles.

      This constructor initializes a new instance of the TRffBicScore without requiring parameters.

  • Method Details

    • getScore

      public Score getScore(DataModel dataSet, Parameters parameters)
      Calculates and returns a score for the given dataset based on the specified parameters.
      Specified by:
      getScore in interface ScoreWrapper
      Parameters:
      dataSet - The dataset for which the score is to be calculated. It must be an instance of DataSet.
      parameters - The parameters used to configure the scoring process, including tuning knobs such as ridge regularization, number of random Fourier features, and penalty discount.
      Returns:
      An instance of TRffBicScore configured with the provided dataset and parameters.
      Throws:
      IllegalArgumentException - If the provided dataset is not an instance of DataSet.

    • getDescription

      public String getDescription()
      Provides a textual description of the current scoring mechanism.
      Specified by:
      getDescription in interface ScoreWrapper
      Returns:
      a string representing the description of the T-RFF BIC Score

    • getDataType

      public DataType getDataType()
      Retrieves the data type associated with this scoring mechanism.
      Specified by:
      getDataType in interface ScoreWrapper
      Returns:
      the data type of the score, which corresponds to DataType.Mixed.

    • getParameters

      public List<String> getParameters()
      Retrieves a list of parameter names associated with the Student-T Random Fourier Feature Bayes Information Criterion (BIC) scoring mechanism. These parameters can be used to configure the behavior of the scoring process.
      Specified by:
      getParameters in interface HasParameters
      Specified by:
      getParameters in interface ScoreWrapper
      Returns:
      a list of parameter names as strings. Each parameter represents a configurable aspect of the scoring mechanism, such as ridge regularization, number of Fourier features, penalty discount, and other optional settings.

    • getVariable

      public Node getVariable(String name)
      Retrieves a variable from the dataset based on the given name.
      Specified by:
      getVariable in interface ScoreWrapper
      Parameters:
      name - the name of the variable to retrieve. It must correspond to a variable contained in the dataset.
      Returns:
      the Node representing the variable associated with the specified name. Returns null if the dataset does not contain a variable with the given name.