Package adams.data.instancesanalysis.pls

Class SparsePLS

  • All Implemented Interfaces:
    Destroyable, GlobalInfoSupporter, LoggingLevelHandler, LoggingSupporter, OptionHandler, SizeOfHandler, TechnicalInformationHandler, Serializable, weka.core.CapabilitiesHandler, GenericPLSMatrixAccess
    public class SparsePLS
extends AbstractSingleClassPLS
    Nonlinear Iterative Partial Least Squares (SparsePLS).
    Automatically standardizes X and Y internally.

    For more information see:
    Chun H, Keles S. (2010). Sparse partial least squares regression for simultaneous dimension reduction and variable selection. Royal Statistical Society Series B, Statistical Methodology. 1:3-25. URL https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2810828/ 

     @article{ChunH2010,
    author = {Chun H, Keles S.},
    journal = {Royal Statistical Society Series B, Statistical Methodology},
    pages = {3-25},
    title = {Sparse partial least squares regression for simultaneous dimension reduction and variable selection},
    volume = {1},
    year = {2010},
    URL = {https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2810828/}
 }
 


    -logging-level <OFF|SEVERE|WARNING|INFO|CONFIG|FINE|FINER|FINEST> (property: loggingLevel)
    The logging level for outputting errors and debugging output.
    default: WARNING
    min-user-mode: Expert
 
    -preprocessing-type <NONE|CENTER|STANDARDIZE> (property: preprocessingType)
    The type of preprocessing to perform.
    default: NONE
 
    -replace-missing <boolean> (property: replaceMissing)
    Whether to replace missing values.
    default: false
 
    -num-components <int> (property: numComponents)
    The number of components to compute.
    default: 20
    minimum: 1
 
    -prediction-type <NONE|ALL|EXCEPT_CLASS> (property: predictionType)
    The type of prediction to perform.
    default: NONE
 
    -tol <double> (property: tol)
    The inner NIPALS loop improvement tolerance.
    default: 1.0E-7
    minimum: 0.0
 
    -max-iter <int> (property: maxIter)
    The inner NIPALS loop maximum number of iterations.
    default: 500
    minimum: 1
 
    -lambda <double> (property: lambda)
    The sparsity parameter; determines sparseness.
    default: 0.5
    minimum: 0.0
    Author:
    FracPete (fracpete at waikato dot ac dot nz)
    See Also:
    Serialized Form

    • Field Detail

      • m_SparsePLS

        protected com.github.waikatodatamining.matrix.algorithm.pls.SparsePLS m_SparsePLS
        the actual algorithm.

      • m_Tol

        protected double m_Tol
        Inner NIPALS loop improvement tolerance

      • m_MaxIter

        protected int m_MaxIter
        Inner NIPALS loop maximum number of iterations

      • m_Lambda

        protected double m_Lambda
        Sparsity parameter. Determines sparseness.

    • Constructor Detail

      • SparsePLS

        public SparsePLS()

    • Method Detail

      • setTol

        public void setTol​(double value)
        Sets the inner NIPALS loop improvement tolerance.
        Parameters:
        value - the tolerance

      • getTol

        public double getTol()
        Returns the inner NIPALS loop improvement tolerance.
        Returns:
        the tolerance

      • tolTipText

        public String tolTipText()
        Returns the tip text for this property
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui

      • setMaxIter

        public void setMaxIter​(int value)
        Sets the inner NIPALS loop maximum number of iterations.
        Parameters:
        value - the maximum

      • getMaxIter

        public int getMaxIter()
        Returns the NIPALS loop maximum number of iterations.
        Returns:
        the maximum

      • maxIterTipText

        public String maxIterTipText()
        Returns the tip text for this property
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui

      • setLambda

        public void setLambda​(double value)
        Sets sparsity parameter; determines sparseness.
        Parameters:
        value - the lambda

      • getLambda

        public double getLambda()
        Returns the sparsity parameter; determines sparseness.
        Returns:
        the lambda

      • lambdaTipText

        public String lambdaTipText()
        Returns the tip text for this property
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui

      • getMatrix

        public weka.core.matrix.Matrix getMatrix​(String name)
        Returns the matrix with the specified name.
        Specified by:
        getMatrix in interface GenericPLSMatrixAccess
        Specified by:
        getMatrix in class AbstractPLS
        Parameters:
        name - the name of the matrix
        Returns:
        the matrix, null if not available

      • doTransform

        protected weka.core.Instances doTransform​(weka.core.Instances data,
                                          Map<String,​Object> params)
                                   throws Exception
        Transforms the data, initializes if necessary.
        Specified by:
        doTransform in class AbstractPLS
        Parameters:
        data - the data to use
        params - additional parameters
        Returns:
        the transformed data
        Throws:
        Exception