Package de.unijena.cheminf.clustering.art2a

Class Art2aClusteringTask

java.lang.Object
de.unijena.cheminf.clustering.art2a.Art2aClusteringTask
All Implemented Interfaces:
Callable<IArt2aClusteringResult>
public class Art2aClusteringTask extends Object implements Callable<IArt2aClusteringResult>
Callable class for clustering input vectors (fingerprints).

  • Field Summary

    Fields
    Modifier and Type
    Field
    Description
    static final double
    DEFAULT_LEARNING_PARAMETER_DOUBLE
    Default value of the learning parameter in double
    static final float
    DEFAULT_LEARNING_PARAMETER_FLOAT
    Default value of the learning parameter in float
    static final double
    REQUIRED_SIMILARITY_DOUBLE
    Default value of the required similarity parameter in double
    static final float
    REQUIRED_SIMILARITY_FLOAT
    Default value of the required similarity parameter in float

  • Constructor Summary

    Constructors
    Constructor
    Description
    Art2aClusteringTask(double aVigilanceParameter, double[][] aDataMatrix, int aMaximumEpochsNumber, boolean anIsClusteringResultExported)
    Double clustering task constructor.
    Art2aClusteringTask(double aVigilanceParameter, double[][] aDataMatrix, int aMaximumEpochsNumber, boolean anIsClusteringResultExported, double aRequiredSimilarity, double aLearningParameter)
    Double clustering task constructor.
    Art2aClusteringTask(float aVigilanceParameter, float[][] aDataMatrix, int aMaximumEpochsNumber, boolean anIsClusteringResultExported)
    Float clustering task constructor.
    Art2aClusteringTask(float aVigilanceParameter, float[][] aDataMatrix, int aMaximumEpochsNumber, boolean anIsClusteringResultExported, float aRequiredSimilarity, float aLearningParameter)
    Float clustering task constructor.

  • Method Summary

    Modifier and Type
    Method
    Description
    IArt2aClusteringResult
    call()
    Executes the clustering.
    int
    setSeed(int aSeed)
    User-defined seed value to randomize input vectors.

    Methods inherited from class java.lang.Object

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

  • Field Details

    • DEFAULT_LEARNING_PARAMETER_FLOAT

      public static final float DEFAULT_LEARNING_PARAMETER_FLOAT
      Default value of the learning parameter in float
      See Also:

    • REQUIRED_SIMILARITY_FLOAT

      public static final float REQUIRED_SIMILARITY_FLOAT
      Default value of the required similarity parameter in float
      See Also:

    • DEFAULT_LEARNING_PARAMETER_DOUBLE

      public static final double DEFAULT_LEARNING_PARAMETER_DOUBLE
      Default value of the learning parameter in double
      See Also:

    • REQUIRED_SIMILARITY_DOUBLE

      public static final double REQUIRED_SIMILARITY_DOUBLE
      Default value of the required similarity parameter in double
      See Also:

  • Constructor Details

    • Art2aClusteringTask

      public Art2aClusteringTask(float aVigilanceParameter, float[][] aDataMatrix, int aMaximumEpochsNumber, boolean anIsClusteringResultExported, float aRequiredSimilarity, float aLearningParameter) throws IllegalArgumentException, NullPointerException
      Float clustering task constructor. Creates a new Art2aClusteringTask instance with the specified parameters.
      Parameters:
      aVigilanceParameter - parameter to influence the number of clusters.
      aDataMatrix - matrix contains all inputs for clustering. Each row of the matrix contains one input. In addition, all inputs must have the same length. Each column of the matrix contains one component of the input.
      aMaximumEpochsNumber - maximum number of epochs that the system may use for convergence.
      anIsClusteringResultExported - if the parameter is set to true, the cluster results are exported to text files.
      aRequiredSimilarity - parameter indicating the minimum similarity between the current cluster vectors and the previous cluster vectors. The parameter is crucial for the convergence of the system. If the parameter is set too high, a much more accurate similarity is expected and the convergence may take longer, while a small parameter expects a lower similarity between the cluster vectors and thus the system may converge faster.
      aLearningParameter - parameter to define the intensity of keeping the old cluster vector in mind before the system adapts it to the new sample vector.
      Throws:
      IllegalArgumentException - is thrown, if the given arguments are invalid. The checking of the arguments is done in the constructor of Art2aFloatClustering.
      NullPointerException - is thrown, if the given aDataMatrix is null. The checking of the data matrix is done in the constructor of the ArtaFloatClustering.

    • Art2aClusteringTask

      public Art2aClusteringTask(float aVigilanceParameter, float[][] aDataMatrix, int aMaximumEpochsNumber, boolean anIsClusteringResultExported) throws IllegalArgumentException, NullPointerException
      Float clustering task constructor. Creates a new Art2aClusteringTask instance with the specified parameters. For the required similarity and learning parameter default values are used.
      Parameters:
      aVigilanceParameter - parameter to influence the number of clusters.
      aDataMatrix - matrix contains all inputs for clustering. Each row of the matrix contains one input. In addition, all inputs must have the same length. Each column of the matrix contains one component of the input.
      aMaximumEpochsNumber - maximum number of epochs that the system may use for convergence.
      anIsClusteringResultExported - if the parameter is set to true, the cluster results are exported to text files.
      Throws:
      IllegalArgumentException - is thrown, if the given arguments are invalid. The checking of the arguments is done in the constructor of Art2aFloatClustering.
      NullPointerException - is thrown, if the given aDataMatrix is null. The checking of the data matrix is done in the constructor of the ArtaFloatClustering.
      See Also:

    • Art2aClusteringTask

      public Art2aClusteringTask(double aVigilanceParameter, double[][] aDataMatrix, int aMaximumEpochsNumber, boolean anIsClusteringResultExported, double aRequiredSimilarity, double aLearningParameter) throws IllegalArgumentException, NullPointerException
      Double clustering task constructor. Creates a new Art2aDoubleClustering instance with the specified parameters.
      Parameters:
      aVigilanceParameter - parameter to influence the number of clusters.
      aDataMatrix - matrix contains all inputs for clustering. Each row of the matrix contains one input. In addition, all inputs must have the same length. Each column of the matrix contains one component of the input.
      aMaximumEpochsNumber - maximum number of epochs that the system may use for convergence.
      anIsClusteringResultExported - if the parameter is set to true, the cluster results are exported to text files.
      aRequiredSimilarity - parameter indicating the minimum similarity between the current cluster vectors and the previous cluster vectors.
      aLearningParameter - parameter to define the intensity of keeping the old cluster vector in mind before the system adapts it to the new sample vector.
      Throws:
      IllegalArgumentException - is thrown, if the given arguments are invalid. The checking of the arguments is done in the constructor of Art2aFloatClustering.
      NullPointerException - is thrown, if the given aDataMatrix is null. The checking of the data matrix is done in the constructor of the ArtaFloatClustering.

    • Art2aClusteringTask

      public Art2aClusteringTask(double aVigilanceParameter, double[][] aDataMatrix, int aMaximumEpochsNumber, boolean anIsClusteringResultExported) throws IllegalArgumentException, NullPointerException
      Double clustering task constructor. Creates a new Art2aDoubleClustering instance with the specified parameters. For the required similarity and learning parameter default values are used.
      Parameters:
      aVigilanceParameter - parameter to influence the number of clusters.
      aDataMatrix - matrix contains all inputs for clustering. Each row of the matrix contains one input. In addition, all inputs must have the same length. Each column of the matrix contains one component of the input.
      aMaximumEpochsNumber - maximum number of epochs that the system may use for convergence.
      anIsClusteringResultExported - if the parameter is set to true, the cluster results are exported to text files.
      Throws:
      IllegalArgumentException - is thrown, if the given arguments are invalid. The checking of the arguments is done in the constructor of Art2aFloatClustering.
      NullPointerException - is thrown, if the given aDataMatrix is null. The checking of the data matrix is done in the constructor of the ArtaFloatClustering.
      See Also:

  • Method Details

    • call

      public IArt2aClusteringResult call()
      Executes the clustering.
      Specified by:
      call in interface Callable<IArt2aClusteringResult>
      Returns:
      clustering result.

    • setSeed

      public int setSeed(int aSeed)
      User-defined seed value to randomize input vectors. Different seed values can lead to different clustering results.
      Parameters:
      aSeed - seed value
      Returns:
      user-defined seed value.