Package org.ejml.simple

Class SimpleSVD<T extends SimpleBase>

  • public class SimpleSVD<T extends SimpleBase>
extends java.lang.Object

    Wrapper around SVD for simple matrix. See SingularValueDecomposition for more details.

    SVD is defined as the following decomposition:
    A = U * W * V T
    where A is m by n, and U and V are orthogonal matrices, and W is a diagonal matrix

    Tolerance for singular values is Math.max(mat.numRows,mat.numCols) * W.get(0,0) * UtilEjml.EPS; where W.get(0,0) is the largest singular value.

    • Constructor Summary

      Constructors 
      Constructor Description
      SimpleSVD​(org.ejml.data.Matrix mat, boolean compact)  

    • Method Summary

      All Methods Instance Methods Concrete Methods 
      Modifier and Type Method Description
      double getSingleValue​(int index)
      Returns the specified singular value.
      double[] getSingularValues()
      Returns an array of all the singular values
      org.ejml.interfaces.decomposition.SingularValueDecomposition getSVD()
      Returns the underlying decomposition that this is a wrapper around.
      T getU()
      Returns the orthogonal 'U' matrix.
      T getV()
      Returns the orthogonal 'V' matrix.
      T getW()
      Returns a diagonal matrix with the singular values.
      int nullity()
      The nullity of the decomposed matrix.
      SimpleMatrix nullSpace()
      Computes the null space from an SVD.
      double quality()
      Computes the quality of the computed decomposition.
      int rank()
      Returns the rank of the decomposed matrix.

      • Methods inherited from class java.lang.Object

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

    • Constructor Detail

      • SimpleSVD

        public SimpleSVD​(org.ejml.data.Matrix mat,
                 boolean compact)

    • Method Detail

      • getU

        public T getU()

        Returns the orthogonal 'U' matrix.

        Returns:
        An orthogonal m by m matrix.

      • getW

        public T getW()
        Returns a diagonal matrix with the singular values. The singular values are ordered from largest to smallest.
        Returns:
        Diagonal matrix with singular values along the diagonal.

      • getV

        public T getV()

        Returns the orthogonal 'V' matrix.

        Returns:
        An orthogonal n by n matrix.

      • quality

        public double quality()

        Computes the quality of the computed decomposition. A value close to or less than 1e-15 is considered to be within machine precision.

        This function must be called before the original matrix has been modified or else it will produce meaningless results.

        Returns:
        Quality of the decomposition.

      • nullSpace

        public SimpleMatrix nullSpace()
        Computes the null space from an SVD. For more information see SingularOps_DDRM.nullSpace(org.ejml.interfaces.decomposition.SingularValueDecomposition_F64<org.ejml.data.DMatrixRMaj>, org.ejml.data.DMatrixRMaj, double).
        Returns:
        Null space vector.

      • getSingleValue

        public double getSingleValue​(int index)
        Returns the specified singular value.
        Parameters:
        index - Which singular value is to be returned.
        Returns:
        A singular value.

      • getSingularValues

        public double[] getSingularValues()
        Returns an array of all the singular values

      • rank

        public int rank()
        Returns the rank of the decomposed matrix.
        Returns:
        The matrix's rank
        See Also:
        SingularOps_DDRM.rank(SingularValueDecomposition_F64, double)

      • nullity

        public int nullity()
        The nullity of the decomposed matrix.
        Returns:
        The matrix's nullity
        See Also:
        SingularOps_DDRM.nullity(SingularValueDecomposition_F64, double)

      • getSVD

        public org.ejml.interfaces.decomposition.SingularValueDecomposition getSVD()
        Returns the underlying decomposition that this is a wrapper around.
        Returns:
        SingularValueDecomposition