Package Exports
- ndarray-blas-level2
This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (ndarray-blas-level2) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.
Readme
ndarray-blas-level2
BLAS Level 2 operations for ndarrays
Usage
This library implements the basic matrix-vector operations of the Level 2 Basic Linear Algebra Subprograms (BLAS).
Note: It's possible to accomplish the lower triangular functions with the upper triangular version plus flipping and unflipping dimensions, but that's a little convoluted. Instead, the lower triangular versions are suffixed with _lower just to keep it really simple.
gemv(alpha, A, x, beta, y)
Calculate y <- alpha*A*x + beta*y
trmv(A, x, isLower)
Calculate x <- A*x
for the upper triangular matrix A. Data below the diagonal is ignored. If isLower
is true, uses the lower triangular portion of A instead.
trsv(A, x, isLower)
Calculate x <- A^-1 x
for the upper triangular matrix A. Data below the diagonal is ignored. If isLower
is true, uses the lower triangular portion of A instead.
gbmv(A, kl, ku, x, y, alpha, beta)
Calculates y <- alpha*A*x + beta*y
for banded matrices. kl
is the number of subdiagonals and ku
are the number of super diagonals. alpha
defaults to 1.0 and beta
defaults to 0.0 if not specified.
symv(A, x, y, fromLower, alpha, beta)
Calculates y <- alpha*A*x + beta*y
for symmetric matrices. If fromLower = true
, the function uses the lower triangular part of the matrix; for false
it uses the upper triangular part. alpha
defaults to 1.0 and beta
defaults to 0.0 if not specified.
Credits
© 2016 Scijs Authors. MIT License.