Examples of LUDecompositionAlt

org.ejml.alg.dense.decomposition.lu.LUDecompositionAlt

An LU decomposition algorithm that originally came from Jama. In general this is faster than what is in NR since it creates a cache of a column, which makes a big difference in larger matrices.
@author Peter Abeles

Examples of org.ejml.alg.dense.decomposition.lu.LUDecompositionAlt


            DenseMatrix64F expected = new DenseMatrix64F(N,N);
            DenseMatrix64F found = new DenseMatrix64F(N,N);


            // first compute inverse by LU
            LUDecompositionAlt alg = new LUDecompositionAlt();
            LinearSolverLu solver = new LinearSolverLu(alg);


            assertTrue( solver.setA(A));
            solver.invert(expected);

View Full Code Here

Examples of org.ejml.alg.dense.decomposition.lu.LUDecompositionAlt

    public void testAll() {
        for( int N = 2; N <= UnrolledDeterminantFromMinor.MAX; N++ ) {
            DenseMatrix64F A = RandomMatrices.createRandom(N,N,rand);


            double unrolled = UnrolledDeterminantFromMinor.det(A);
            LUDecompositionAlt alg = new LUDecompositionAlt();
            assertTrue( alg.decompose(A) );
            double expected = alg.computeDeterminant();


            assertEquals(expected,unrolled,1e-8);
        }
    }

View Full Code Here

Examples of org.ejml.alg.dense.decomposition.lu.LUDecompositionAlt

        Random rand = new Random(0xfff);


        for( int width = 5; width < 12; width++ ) {
            DenseMatrix64F A = RandomMatrices.createRandom(width,width,rand);


            LUDecompositionAlt lu = new LUDecompositionAlt();
            lu.decompose(A);


            double luVal = lu.computeDeterminant();


            DeterminantFromMinor minor = new DeterminantFromMinor(width);
            double minorVal = minor.compute(new DenseMatrix64F(width,width, true, A.data));


            assertEquals(luVal,minorVal,1e-6);

View Full Code Here

Examples of org.ejml.alg.dense.decomposition.lu.LUDecompositionAlt

    @Test
    public void invert() {
        DenseMatrix64F A = new DenseMatrix64F(3,3, true, 0, 1, 2, -2, 4, 9, 0.5, 0, 5);
        DenseMatrix64F A_inv = RandomMatrices.createRandom(3,3,rand);


        LUDecompositionAlt decomp = new LUDecompositionAlt();
        LinearSolver solver = new LinearSolverLu(decomp);


        solver.setA(A);
        InvertUsingSolve.invert(solver,A,A_inv);

View Full Code Here

Examples of org.ejml.alg.dense.decomposition.lu.LUDecompositionAlt

     * Creates a solver for linear systems.  The A matrix will have dimensions (m,m).
     *
     * @return A new linear solver.
     */
    public static LinearSolver<DenseMatrix64F> linear( int matrixSize ) {
        return new LinearSolverLu(new LUDecompositionAlt());
    }

View Full Code Here

Examples of org.ejml.alg.dense.decomposition.lu.LUDecompositionAlt

     *
     * @parm matrixWidth The matrix size that the decomposition should be optimized for.
     * @return LUDecomposition
     */
    public static LUDecomposition<DenseMatrix64F> lu( int numRows , int numCol ) {
        return new LUDecompositionAlt();
    }

View Full Code Here

Examples of org.ejml.alg.dense.decomposition.lu.LUDecompositionAlt

                return UnrolledDeterminantFromMinor.det(mat);
            } else {
                return mat.get(0);
            }
        } else {
            LUDecompositionAlt alg = new LUDecompositionAlt();
            if( !alg.decompose(mat) )
                return 0.0;
            return alg.computeDeterminant();
        }
    }

View Full Code Here

Examples of org.ejml.alg.dense.decomposition.lu.LUDecompositionAlt

                UnrolledInverseFromMinor.inv(mat,mat);
            } else {
                mat.set(0, 1.0/mat.get(0));
            }
        } else {
            LUDecompositionAlt alg = new LUDecompositionAlt();
            LinearSolverLu solver = new LinearSolverLu(alg);
            if( solver.setA(mat) ) {
                solver.invert(mat);
            } else {
                return false;

View Full Code Here

Examples of org.ejml.alg.dense.decomposition.lu.LUDecompositionAlt

                UnrolledInverseFromMinor.inv(mat,result);
            } else {
                result.set(0,  1.0/mat.get(0));
            }
        } else {
            LUDecompositionAlt alg = new LUDecompositionAlt();
            LinearSolverLu solver = new LinearSolverLu(alg);


            if( !solver.setA(mat))
                return false;
            solver.invert(result);

View Full Code Here

Examples of org.ejml.alg.dense.decomposition.lu.LUDecompositionAlt

        DenseMatrix64F A = new DenseMatrix64F(3,3, true, 0, 1, 2, -2, 4, 9, 0.5, 0, 5);
        DenseMatrix64F b = new DenseMatrix64F(3,1, true, 8, 33, 15.5);
        DenseMatrix64F x = RandomMatrices.createRandom(3,1,rand);
        DenseMatrix64F x_improved = new DenseMatrix64F(3,1);


        LUDecompositionAlt alg = new LUDecompositionAlt();


        x_improved.set(x);


        LinearSolverLu solver = new LinearSolverLu(alg);
        assertTrue(solver.setA(A));

View Full Code Here

0 1

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.