drasys.or.linear.blas
Class MatrixBLAS2

java.lang.Object
  |
  +--drasys.or.linear.blas.MatrixBLAS2

public class MatrixBLAS2

extends java.lang.Object

BLAS2 layer that operates on contiguous vectors and matrices. For a traditional and more feature rich implementation of the BLAS see BLAS1, BLAS2 and BLAS3.

References:

Linear Algebra and Its Applications
    Gilbert Strang / Hardcover / Published 1988
Matrix Computations (Johns Hopkins Studies in the Mathematical Sciences)
    Gene H. Golub, Charles F. Van Loan (Contributor) / Paperback / Published 1996
Numerical Recipes in C : The Art of Scientific Computing
    William H. Press, et al / Hardcover / Published 1993

Constructor Summary

MatrixBLAS2()           Create a matrix BLAS with the default BLAS implementation and a base index of '0'.

MatrixBLAS2(int base)           Create a matrix BLAS with a default BLAS implementation and an explicit base index.

MatrixBLAS2(int base, BLAS2I blas2)           Create a matrix BLAS with an explicit BLAS implementation and an explicit base index.

 

Method Summary
void	dgemv(boolean transpose, double alpha, ContiguousMatrixI A, ContiguousVectorI x, double beta, ContiguousVectorI y)           Computes y = alpha*A*x + beta*y if transpose is 'false'.
void	dger(double alpha, ContiguousVectorI x, ContiguousVectorI y, ContiguousMatrixI A)           Computes A = alpha*x*y' + A
void	zgemv(boolean transpose, ComplexI alpha, ContiguousMatrixI A, ContiguousVectorI x, ComplexI beta, ContiguousVectorI y)           Computes y = alpha*A*x + beta*y if transpose is 'false'.
void	zgerc(ComplexI alpha, ContiguousVectorI x, ContiguousVectorI y, ContiguousMatrixI A)           Computes A = alpha*x*congujate(y') + A
void	zgeru(ComplexI alpha, ContiguousVectorI x, ContiguousVectorI y, ContiguousMatrixI A)           Computes A = alpha*x*y' + A

 

Methods inherited from class java.lang.Object

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

 

Constructor Detail

MatrixBLAS2

public MatrixBLAS2()

Create a matrix BLAS with the default BLAS implementation and a base index of '0'. All computations will use elements A[0...(nrow-1)][0...(ncol-1)].

MatrixBLAS2

public MatrixBLAS2(int base)

Create a matrix BLAS with a default BLAS implementation and an explicit base index. Typically 'base' will be '0' or '1'. All computations will use elements A[base...(nrow-1)][base...(ncol-1)].

MatrixBLAS2

public MatrixBLAS2(int base,
                   BLAS2I blas2)

Create a matrix BLAS with an explicit BLAS implementation and an explicit base index. Typically 'base' will be '0' or '1'. All computations will use elements A[base...(nrow-1)][base...(ncol-1)].

dgemv

public void dgemv(boolean transpose,
                  double alpha,
                  ContiguousMatrixI A,
                  ContiguousVectorI x,
                  double beta,
                  ContiguousVectorI y)
           throws BlasException

Computes y = alpha*A*x + beta*y if transpose is 'false'. Computes y = alpha*A'*x + beta*y if transpose is 'true'.

dger

public void dger(double alpha,
                 ContiguousVectorI x,
                 ContiguousVectorI y,
                 ContiguousMatrixI A)
          throws BlasException

Computes A = alpha*x*y' + A

zgemv

public void zgemv(boolean transpose,
                  ComplexI alpha,
                  ContiguousMatrixI A,
                  ContiguousVectorI x,
                  ComplexI beta,
                  ContiguousVectorI y)
           throws BlasException

Computes y = alpha*A*x + beta*y if transpose is 'false'. Computes y = alpha*A'*x + beta*y if transpose is 'true'.

zgeru

public void zgeru(ComplexI alpha,
                  ContiguousVectorI x,
                  ContiguousVectorI y,
                  ContiguousMatrixI A)
           throws BlasException

Computes A = alpha*x*y' + A

zgerc

public void zgerc(ComplexI alpha,
                  ContiguousVectorI x,
                  ContiguousVectorI y,
                  ContiguousMatrixI A)
           throws BlasException

Computes A = alpha*x*congujate(y') + A