Examples of SingularValueDecomposition

Jama.SingularValueDecomposition
Singular Value Decomposition.
For an m-by-n matrix A with m >= n, the singular value decomposition is an m-by-n orthogonal matrix U, an n-by-n diagonal matrix S, and an n-by-n orthogonal matrix V so that A = U*S*V'.
The singular values, sigma[k] = S[k][k], are ordered so that sigma[0] >= sigma[1] >= ... >= sigma[n-1].
The singular value decompostion always exists, so the constructor will never fail. The matrix condition number and the effective numerical rank can be computed from this decomposition.
cern.colt.matrix.linalg.SingularValueDecomposition
For an m x n matrix A with m >= n, the singular value decomposition is an m x n orthogonal matrix U, an n x n diagonal matrix S, and an n x n orthogonal matrix V so that A = U*S*V'.
The singular values, sigma[k] = S[k][k], are ordered so that sigma[0] >= sigma[1] >= ... >= sigma[n-1].
The singular value decomposition always exists, so the constructor will never fail. The matrix condition number and the effective numerical rank can be computed from this decomposition.
edu.ucla.sspace.matrix.factorization.SingularValueDecomposition
A refinement of the {@link MatrixFactorization} interface for algorithms thatcompute the Singular Value Decomposition. This interface exposes the three matrices that are generated as a result of the decomposition. @author David Jurgens
org.apache.commons.math.linear.SingularValueDecomposition
nist.gov/javanumerics/jama/">JAMA library, with the following changes:
- the norm2 method which has been renamed as {@link #getNorm() getNorm},
- the cond method which has been renamed as {@link #getConditionNumber() getConditionNumber},
- the rank method which has been renamed as {@link #getRank() getRank},
- a {@link #getUT() getUT} method has been added,
- a {@link #getVT() getVT} method has been added,
- a {@link #getSolver() getSolver} method has been added,
- a {@link #getCovariance(double) getCovariance} method has been added.
@see MathWorld @see Wikipedia @version $Revision: 928081 $ $Date: 2010-03-26 23:36:38 +0100 (ven. 26 mars 2010) $ @since 2.0
org.apache.mahout.math.SingularValueDecomposition
org.ejml.factory.SingularValueDecomposition

This is an abstract class for computing the singular value decomposition (SVD) of a matrix, which is defined as:

A = U * W * V ^T

where A is m by n, and U and V are orthogonal matrices, and W is a diagonal matrix.

The dimension of U,W,V depends if it is a compact SVD or not. If not compact then U is m by m, W is m by n, V is n by n. If compact then let s be the number of singular values, U is m by s, W is s by s, and V is n by s.

Accessor functions for decomposed matrices can return an internally constructed matrix if null is passed in for the optional storage parameter. The exact behavior is implementation specific. If an internally maintained matrix is returned then on the next call to decompose the matrix will be modified. The advantage of this approach is reduced memory overhead.

To create a new instance of SingularValueDecomposition see {@link DecompositionFactory#svd(int,int,boolean,boolean,boolean)}and {@link org.ejml.ops.SingularOps} contains additional helpful SVD related functions.

*Note* that the ordering of singular values is not guaranteed, unless done so by a specific implementation. The singular values can be put into descending order while adjusting U and V using {@link org.ejml.ops.SingularOps#descendingOrder(org.ejml.data.DenseMatrix64F,boolean,org.ejml.data.DenseMatrix64F,org.ejml.data.DenseMatrix64F,boolean)} SingularOps.descendingOrder()}.
@author Peter Abeles
org.encog.mathutil.matrices.decomposition.SingularValueDecomposition
nist.gov/javanumerics/jama/
weka.core.matrix.SingularValueDecomposition
nist.gov/javanumerics/jama/" target="_blank">JAMA package. @author The Mathworks and NIST @author Fracpete (fracpete at waikato dot ac dot nz) @version $Revision: 1.5 $

Examples of org.apache.mahout.math.SingularValueDecomposition

      throw new CardinalityException(m.numRows(), m.numCols());
    }
    // See http://www.mlahanas.de/Math/svd.htm for details,
    // which specifically details the case of covariance matrix inversion
    // Complexity: O(min(nm2,mn2))
    SingularValueDecomposition svd = new SingularValueDecomposition(m);
    Matrix sInv = svd.getS();
    // Inverse Diagonal Elems
    for (int i = 0; i < sInv.numRows(); i++) {
      double diagElem = sInv.get(i, i);
      if (diagElem > 0.0) {
        sInv.set(i, i, 1 / diagElem);
      } else {
        throw new IllegalStateException("Eigen Value equals to 0 found.");
      }
    }
    inverseCovarianceMatrix = svd.getU().times(sInv.times(svd.getU().transpose()));
    Preconditions.checkArgument(inverseCovarianceMatrix != null, "inverseCovarianceMatrix not initialized");
  }

View Full Code Here

Examples of org.apache.mahout.math.SingularValueDecomposition

  @Test
  public void testLeftVectors() throws IOException {
    Matrix A = lowRankMatrixInMemory(20, 20);


    SequentialBigSvd s = new SequentialBigSvd(A, 6);
    SingularValueDecomposition svd = new SingularValueDecomposition(A);


    // can only check first few singular vectors
    Matrix u1 = svd.getU().viewPart(0, 20, 0, 3).assign(Functions.ABS);
    Matrix u2 = s.getU().viewPart(0, 20, 0, 3).assign(Functions.ABS);
    assertEquals(u1, u2);
  }

View Full Code Here

Examples of org.apache.mahout.math.SingularValueDecomposition

  @Test
  public void testRightVectors() throws IOException {
    Matrix A = lowRankMatrixInMemory(20, 20);


    SequentialBigSvd s = new SequentialBigSvd(A, 6);
    SingularValueDecomposition svd = new SingularValueDecomposition(A);


    Matrix v1 = svd.getV().viewPart(0, 20, 0, 3).assign(Functions.ABS);
    Matrix v2 = s.getV().viewPart(0, 20, 0, 3).assign(Functions.ABS);
    assertEquals(v1, v2);
  }

View Full Code Here

Examples of org.apache.mahout.math.SingularValueDecomposition

    // subtract pseudo pca mean
    for (int i = 0; i < m; i++)
      for (int j = 0; j < n; j++)
        a[i][j] -= xi.getQuick(j);


    SingularValueDecomposition svd2 =
      new SingularValueDecomposition(new DenseMatrix(a));


    Vector svalues2 = new DenseVector(svd2.getSingularValues());
    LocalSSVDSolverSparseSequentialTest.dumpSv(svalues2);


    for (int i = 0; i < k + p; i++) {
      assertTrue(Math.abs(svalues2.getQuick(i) - stochasticSValues.getQuick(i)) <= s_epsilon);
    }

View Full Code Here

Examples of org.apache.mahout.math.SingularValueDecomposition

    // try to run the same thing without stochastic algo
    double[][] a = SSVDHelper.loadDistributedRowMatrix(fs, aPath, conf);


    // SingularValueDecompositionImpl svd=new SingularValueDecompositionImpl(new
    // Array2DRowRealMatrix(a));
    SingularValueDecomposition svd2 =
      new SingularValueDecomposition(new DenseMatrix(a));


    Vector svalues2 = new DenseVector(svd2.getSingularValues());
    dumpSv(svalues2);


    for (int i = 0; i < k + p; i++) {
      assertTrue(Math.abs(svalues2.getQuick(i) - stochasticSValues.getQuick(i)) <= s_epsilon);
    }

View Full Code Here

Examples of org.apache.mahout.math.SingularValueDecomposition

    }


    public void compute()
    {
        // Use Colt's SVD
        SingularValueDecomposition svd;
        if (A.columns() > A.rows())
        {
            svd = new SingularValueDecomposition(new DenseMatrix(A.viewDice().toArray()));
            V = toColtMatrix(svd.getU());
            U = toColtMatrix(svd.getV());
        }
        else
        {
            svd = new SingularValueDecomposition(new DenseMatrix(A.toArray()));
            U = toColtMatrix(svd.getU());
            V = toColtMatrix(svd.getV());
        }


        S = svd.getSingularValues();


        if (k > 0 && k < S.length)
        {
            U = U.viewPart(0, 0, U.rows(), k);
            V = V.viewPart(0, 0, V.rows(), k);

View Full Code Here

Examples of org.ejml.factory.SingularValueDecomposition

        DenseMatrix64F A = new DenseMatrix64F(3,4);


        // it will need to create a copy in this case
        Dummy dummy = new Dummy(2,true,true,2,3);


        SingularValueDecomposition decomp = new SafeSvd(dummy);
        assertFalse(decomp.inputModified());


        decomp.decompose(A);


        assertTrue(A != dummy.passedInMatrix);


        // now no need to make a copy
        dummy = new Dummy(2,true,false,2,3);
        decomp = new SafeSvd(dummy);
        assertFalse(decomp.inputModified());


        decomp.decompose(A);


        assertTrue(A == dummy.passedInMatrix);
    }

View Full Code Here

Examples of org.ejml.factory.SingularValueDecomposition


    @Test
    public void checkOtherFunctions() {
        Dummy dummy = new Dummy(2,true,true,2,3);


        SingularValueDecomposition decomp = new SafeSvd(dummy);


        assertTrue(decomp.isCompact());
        assertEquals(2, decomp.numberOfSingularValues());


        assertFalse(dummy.getU_called);
        assertFalse(dummy.getV_called);
        assertFalse(dummy.getW_called);


        decomp.getU(null,false);
        assertTrue(dummy.getU_called);
        decomp.getV(null, false);
        assertTrue(dummy.getV_called);
        decomp.getW(null);
        assertTrue(dummy.getW_called);


        assertEquals(2,decomp.numCols());
        assertEquals(3,decomp.numRows());
    }

View Full Code Here

Examples of org.encog.mathutil.matrices.decomposition.SingularValueDecomposition

        {1,2,3,4},
        {5,6,7,8},
        {9,10,11,12},
        {13,14,15,16} };
    Matrix matrix = new Matrix(m);
    SingularValueDecomposition svd = new SingularValueDecomposition(matrix);
    Assert.assertEquals(2147483647, (int)(svd.cond()) );
    double[] d = svd.getSingularValues();
    
    Assert.assertEquals(4, d.length);


  }

View Full Code Here

Examples of weka.core.matrix.SingularValueDecomposition

    // svd requires rows >= columns, so transpose data if necessary
    if (m_numAttributes < m_numInstances) {
      m_transpose = true;
      trainMatrix = trainMatrix.transpose();
    }
    SingularValueDecomposition trainSVD = trainMatrix.svd();
    m_u = trainSVD.getU(); // left singular vectors
    m_s = trainSVD.getS(); // singular values
    m_v = trainSVD.getV(); // right singular vectors
    
    // find actual rank to use
    int maxSingularValues = trainSVD.rank();
    for (int i = 0; i < m_s.getRowDimension(); i++) {
      m_sumSquaredSingularValues += m_s.get(i, i) * m_s.get(i, i);
    }
    if (maxSingularValues == 0) { // no nonzero singular values (shouldn't happen)
      // reset values from computation

View Full Code Here

0 1 2 3 4 5

TOP

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.